Self and Nonself Stimulatory Molecules Induce Preferential Expansion of CD5 þ B Cells or Activated T Cells of Chagasic Patients, Respectively

Scand. J. Immunol. 51, 91 97, 2000 Self and Nonself Stimulatory Molecules Induce Preferential Expansion of CD5 þ B Cells or Activated T Cells of Chagasic Patients, Respectively W. O. DUTRA*, D. G. COLLEY, J. C. PINTO-DIAS, G. GAZZINELLI, Z. BRENER, M. E. S. PEREIRA, R. L. COFFMAN, R. CORREA-OLIVEIRA & J. F. CARVALHO-PARRA *Departamento de Morfologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627, Pampulha, Belo Horizonte, 31270 911, MG, Brazil, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Public Health Services, U.S., Department of Health and Human Services, 4770 Buford Hwy NE, Atlanta, 30341, GA, USA, Centro de Pesquisas René Rachou FIOCRUZ, Av. Augusto de Lima, 1715, Belo Horizonte, 30190 002, MG, Brazil, DNAX Research Institute of Molecular and Cellular Biology, 901 California Avenue, Palo Alto, 94301, CA, USA, and Instituto Ludwig de Pesquisa sobre o Câncer, Rua Professor Antônio Prudente, 109/4 andar, São Paulo, 01509 010, SP, Brazil (Received 28 May 1999; Accepted in revised form 8 September 1999) Dutra WO, Colley DG, Pinto-Dias JC, Gazzinelli G, Brener Z, Pereira MES, Coffman RL, Correa-Oliveira R, Carvalho-Parra JF. Self and Nonself Stimulatory Molecules Induce Preferential Expansion of CD5 þ B Cells or Activated T Cells of Chagasic Patients, Respectively. Scand J Immunol 2000:51:91 97 It has previously been demonstrated that Trypanosoma cruzi-derived antigens (TRP) and human parasitespecific antibodies (Id) stimulate proliferation of cells from Chagasic patients. More recently, we have shown that activated T cells and CD5 þ B cells are present in elevated levels in the peripheral blood of Chagasic patients. Upon in vitro exposure to these two different types of stimulatory molecules (TRP, Id), we now show that each of these elevated populations respond differentially to TRP or Id. We found that stimulation with TRP led to preferential expansion of activated T cells, while Id preferentially stimulated CD5 þ B cells and CD8 þ T cells. Moreover, this expansion of CD5 þ B cells by Id was even more pronounced in cultures of cells from Chagasic patients with the severe, cardiac form of the disease, as compared to indeterminate patients. CD8 þ T cells comprise approximately 50% of the total T cells in cultures stimulated by Id while in TRPstimulated cultures their frequency is proportionally lower. Since parasite antigens and antiparasite antibodies are always present in the host during the chronic phase of the disease, they may also be involved with differential activation mechanisms of these cell populations in vivo. Walderez Ornelas Dutra, Departamento de Morfologia, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627, Pampulha, Belo Horizonte, 31270 911, MG, Brazil INTRODUCTION In some chronic parasitic diseases, a major contributing factor for establishing and maintaining pathology is the presence of high levels of the causative agent. However, in Chagas disease, an infection caused by the protozoan parasite Trypanosoma cruzi,as the disease progresses from the acute to the chronic phase, either with or without characteristic symptoms and clinical manifestations, the number of parasites in the host decreases dramatically. Although the presence of T. cruzi is critical to trigger the series of reactions that lead to chronic pathology, the parasite may not be solely responsible for its maintenance. The presence of autoreactive antibodies and the identification of cross-reactive antigens between host and parasite has lead to the hypotheses of the involvement of autoimmune-type reactions in the generation and maintenance of Chagasic morbidity [1, 2]. The involvement of cell populations related to autoimmune processes in the development of Chagasic pathology have been studied in both experimental and human T. cruzi infection. We have shown that elevated proportions of activated CD4 þ and CD8 þ T cells and CD5 þ B cells are present in high levels in the peripheral blood of chagasic patients [3]. It has also been shown that those cell populations are increased in T. cruzi-infected mice [4]. Moreover, it has been demonstrated that activated CD8 þ 2000 Blackwell Science Ltd

92 W. O. Dutra et al. T lymphocytes are the major component of the inflammatory infiltrate in human Chagasic cardiac lesions [5]. In other systems, activated T cells as well as CD5 þ B cells have been correlated with auto-immune reactions when these lymphocyte populations were shown to be present in high levels in diseases such as rheumatoid arthritis and lupus erythematosus [6 8]. Determining whether these cell populations are preferentially responsive to parasite antigens or to T. cruzi-related self stimuli such as cross-reactive idiotypes [9, 10] would provide insights into the possible involvement of these cells in autoimmune processes in Chagas disease. Although some involvement of anti-host reactions in Chagas disease is generally accepted by many researchers, it is critical to remember that whatever mechanisms participate in pathology, they also depend on the initial infection and exist in the presence of a continued low grade infection [11]. Thus, it is possible to hypothesize that the early high parasitemia detectable during the acute phase might initiate a series of cellular activation events in some patients that would be maintained throughout the chronic phase, despite the subpatent parasitemia. Furthermore, molecular evidences that the DNA of T. cruzi organisms is always present within clinical lesions of those who have died of cardiac complications of Chagas disease, continues to implicate the parasite in fundamental aspects of severe disease [11]. Regardless which process is dominant in the establishment of pathology, and they are clearly not mutually exclusive mechanisms, the existence of mechanisms of maintenance of cell activation is undoubtedly important. Some of the elements involved in the activation of cell populations that may recognise either parasite or host components or both are the parasite itself, albeit in low numbers, and cross reactive antigens. Additionally, we have previously demonstrated that peripheral blood mononuclear cells (PBMC) from chronic Chagasic patients proliferate in vitro in response to the stimuli provided by antiepimastigote antibodies purified from cardiac Chagasic patients sera [12]. These antibody preparations were able to stimulate PBMC from Chagasic patients in a direct manner, not depending on processing, while preparations derived from chronically infected asymptomatic (indeterminate) patients were less effective in stimulating PBMC and the stimulation was dependent on processing by presenting cells [13]. These data showed an important correlation between the severity of disease and the presence, in the host, of stimulatory antibodies that differentially expressed cross-reactive idiotypes (Id). Since Id-expressing antibodies are present in the bloodstream of chronically infected patients and are able to induce cellular reactivity in vitro, they may be involved in selective mechanisms of immune activation during the chronic phase of Chagas disease. Thus, based on two premises: firstly, the antiparasite antibodies and the parasite antigens present in the host are important in the maintenance of cell activation in vivo and secondly, the activated T cells and CD5 þ B cells may be important in the pathology of Chagas disease, we sought to determine whether two different types of stimulatory molecules (TRP, Id) are able to stimulate CD5 þ B cells and/or induce T-cell activation in vitro. PATIENTS, MATERIAL AND METHODS Patients. The patients analyzed in this study were volunteers from Bambuí, Minas Gerais, Brazil, an endemic area for Chagas disease and were under the medical responsibility of Dr João Carlos Pinto-Dias. Complement fixation and indirect immunofluorescence, serological tests indicative of Chagas disease, were positive in all patients studied. The patients were in the chronic stage of the infection and had well defined clinical forms, classified as indeterminate (asymptomatic) or cardiac, based on clinical analysis and electrocardiographic results. Their ages ranged from 26 to 81 years. The age-matched control group (noninfected individuals) had negative serological tests for Chagas disease. The Ethical Committee from FIOCRUZ (Ministry of Health, Brazil) and the Institutional Review Board of the CDC approved this work. Preparation of parasite-derived antigens and patients anti-epimastigote antibodies. Antigens of the trypomastigote form of Trypanosoma cruzi (TRP) were used. Trypomastigotes were obtained from the culture supernatant of VERO cells infected with T. cruzi. Parasites were washed in saline, lyophilized and stored at ¹ 70 C until antigen preparation. Antigenic preparation was performed by freeze-thawing the parasites and then homogenizing on ice, as previously described [14]. The suspension was centrifuged and the resulting supernatant was dialysed, filtered and used in cell cultures at the final concentrations of 25 g/ml. Anti-T. cruziepimastigotes antibodies used in cell cultures were obtained as described by Gazzinelli and co-workers [12, 15]. Briefly, a pool of sera from Chagasic patients with cardiomyopathies was incubated with cyanogen bromide-activated sepharose previously coupled to epimastigote (EPI) antigen. The anti-epi antibodies (Id) were eluted by using 0.1M glycine-hcl (ph2.8), dialyzed, filtered and used in the cultures at a concentration of 40 g/ml. In our study, the pool of sera used for purifying the anti-epi antibodies were from 15 chagasic patients of the cardiac clinical form, since it has been previously shown that Id from cardiac patients are more effective in stimulating PBMC in vitro than Id from indeterminate patients and does so in a different manner [13]. Proliferative assays. In vitro proliferative responses of PBMC from Chagasic patients and non-infected individuals to parasite-related stimuli were performed according to the protocol described by Gazzinelli et al. [16]. Briefly, PBMC obtained by separating blood cells in a Ficoll gradient were washed 3 times in media, counted and cultured in the presence or absence of different stimuli for 6 days at the concentration of 2.5 105 cells/well. After the incubation period, cultures were exposed to 0.5 mci of 3[H]-thymidine for 6 h, harvested and the incorporated radioactivity measured in an automatic scintillation counter. All cultures were performed in triplicate. The degree of proliferation was calculated using the mean average of triplicate cultures with antigenus the mean average of triplicate cultures with medium alone ( cpm) for each individual patient. Analysis of the proportion of activated T cells and CD5 þ B cells in PBMC cultures stimulated with either trypomastigote antigens or human anti-epimastigote antibodies (Id). PBMC from chagasic patients were submitted to bulk cultures, following the same procedure described above. The proportions of cells and antigens were maintained, but the cultures were performed using 24-well plates to obtain larger numbers of cells. We used flow cytometry to determine the frequency of activated T cells and CD5 þ B cells in the cultures, according to a protocol previously used by us with some alterations [3]. Control analysis of total T cells, CD4 þ and CD8 þ T-cell subsets and total B cells, were also performed following the same protocol. PBMC from Chagasic patients cultured for 6 days with TRP or Id were harvested, counted and 5 105

Differential Expansion of Lymphocytes in Chagas Disease 93 cells were incubated with the different antibody solutions for 30 min at 4 C. After incubation, 1 ml of heat-inactivated Fetal Bovine Serum (FBS SIGMA) was added to the tubes to remove cell debris. The samples were centrifuged (10 min, 1200rpm) and the supernatant containing dead cells and debris, aspirated. Cells were washed with phosphatebuffered saline (PBS, ph 7.2) and fixed in a paraformaldehyde-containing solution. Fixed samples were maintained in the dark at 4 C until acquisition and analysis using a FACScan (Becton-Dickinson, Mountain View, CA, USA). Although there were always less cells in Id-stimulated cultures as compared to those cultures stimulated by TRP (as parallels the differential proliferative responses) we always analysed a minimum of 2000 events/staining. The antibodies used for phenotypic analysis were: anti-cd3-fitc, anti-cd4-pe, anti-cd8-pe, anti-hladr-pe, anti-cd25- PE (IL-2R), anti-cd19-pe and anti-cd5-fitc, all purchased from Becton-Dickinson. Analysis of interferon-g (IFN)-g and interleukin-10 (IL-10) production by PBMC from Chagasic patients stimulated with trypomastigote antigen or anti-epimastigote antibody. To measure the production of IL- 10 and IFN- by PBMC from chagasic patients stimulated with either trypomastigote antigen or antiepimastigote antibodies in vitro, bulk cultures were set up, following the same procedure described for proliferation analysis. The proportions of cells and antigens were maintained, but the cultures were performed using 24-well plates to obtain supernatants (for ELISAs). For cytokine measurement, supernatants were harvested after 24 h, 72 h and 6 days of culture and material from each time point came from a different well (triplicate cultures). The determination of cytokines (IL-10 and IFN- ) secreted by the different cultures was performed by ELISA, using a method described by Abrams [17]. Statistical analysis. All data comparisons are of group means SD and were analysed using the Student s t-test. Differences were considered statistically significant when P < 0.05. RESULTS Proliferative response of PBMC from Chagasic patients to trypomastigote antigens or human anti-epimastigote antibodies Analysis of the proliferative response of PBMC from Chagasic patients and non-infected individuals to TRP or Id stimulation showed that both stimuli induce proliferation of cells from Chagasic but not normal individuals, as measured by 3[H]- thymidine incorporation after 6 days of culture. However, the mean cpm observed in the cultures stimulated with Id was lower than the index observed when TRP was used, and Id induced more disparate individual responses (Fig. 1). These data agree with a previously published analysis of in vitro proliferation upon stimulation with parasite antigens or anti-epimastigote antibodies [9, 18]. Analysis of the proportion of total T cells (CD3 þ ) and CD4 þ or CD8 þ T cell subpopulations in cultures of PBMC from indeterminate or cardiac patients stimulated with TRP or Id We determined the percentage of the total CD3 þ T cells and of the CD4 þ and CD8 þ T cell subpopulations in the cultures of PBMC from Chagasic patients stimulated with TRP antigens or anti-epimastigote antibodies using flow cytometry, as described in Materials and Methods. After 6 days of culture, the percentage of CD3 þ T cells was the same in cultures of PBMC from indeterminate patients, regardless of the stimulus used (Fig. 2A). However, when cultures were set up using PBMC from cardiac patients, TRP antigens stimulated a significantly higher mean percentage of CD3 þ T cells than did Id (46 27 versus 27 16, respectively; P < 0.02) (Fig. 2A). Similar results were obtained concerning the percentage of CD4 þ T cells in the different cultures. Thus, cultures of PBMC from indeterminate patients had approximately the same mean percentage of CD4 þ T cells, regardless of the stimulus, while cell cultures from cardiac patients had a higher mean percentage of CD4 þ T cells when stimulated with TRP, as compared to Id (45 26 versus 22 12, respectively; P < 0.001) (Fig. 2B). The mean percentage of CD8 þ T cells was virtually the same in all cultures, regardless of the stimulus or the clinical form of the patients from whom the cells were obtained (Fig. 2C). Although the absolute percentage of CD8 þ cells did not vary among cultures, because fewer CD4 þ cells were stimulated with Id, there was a clear increase in the percentage of CD3 þ cells expressing CD8 in cultures Fig. 1. Analysis of the in vitro proliferative response of PBMC from noninfected individuals (n ¼ 12), indeterminate (n ¼ 18) and cardiac (n ¼ 15) Chagasic patients to T. cruzi-derived antigens (Trypomastigote) and anti-epimastigote antibodies. Results were expressed as cpm (experimental cpm - control culture cpm). Means and statistical significance as compared to NI individuals (P < 0.001) are indicated by a horizontal bar and an asterisk, respectively.

94 W. O. Dutra et al. Fig. 2. Percentage of (A) CD3 þ, (B) CD4 þ and (C) CD8 þ T cells in cultures of PBMC from indeterminate (open circles; n ¼ 20) and cardiac (closed circles; n ¼ 19) patients stimulated with trypomastigote (TRP) antigens or antiepimastigote antibodies (Id). Means are indicated in the figure by the horizontal bars. Before stimulation, mean percentages of CD3 þ, CD4 þ and CD8 þ T cells in indeterminate patients were 65, 41 and 33, respectively, and in cardiac patients, 60, 40 and 32, respectively [3]. stimulated with Id. While in cultures stimulated with TRP only 24 38% of the T cells belonged to the CD8 þ subset, Id stimulation yield 43 50% of the T cells that were CD8 þ (Table 1). Interestingly, the sum of CD4 þ and CD8 þ T cells in cultures stimulated with TRP is higher than the observed percentage of CD3 þ T cells (Table 1) while in Id-stimulated cultures those values were closer (Table 1). Since activated T cells down regulate the CD3 molecule from their surface [19], these data suggested that the TRP stimulation lead to a somewhat higher activation of T cells. Moreover, the overall frequency of T cells was higher in cultures stimulated with TRP, suggesting that this antigen leads to T-cell activation. Trypomastigote antigen stimulation lead to higher frequency of activated T cells in cultures of PBMC from Chagasic patients, as compared to anti-epimastigote stimulation Two phenotypic parameters were used to determine the frequency of activated T cells in the cultures of PBMC from Chagasic patients stimulated with TRP or Id. For the first parameter we analysed the coexpression of CD3 and HLA-DR molecules. We performed double-label flow cytometric studies, as described in Materials and Methods, using anti-cd3-fitc in combination with anti-hla-dr-pe monoclonal antibodies. The second parameter was the analysis of the expression of the IL-2 receptor (IL-2R) by the cells in culture, also by flow cytometry. The co-expression of CD3 and HLA-DR molecules as well as the expression of IL-2R was higher in cultures stimulated with TRP as compared to Id, regardless of the clinical forms of the patients from whom cells were obtained (Fig. 3). These differences of T cell activation between cultures stimulated with TRP or Id were statistically significant (P < 0.03 0.001). Analysis of the proportion of CD5 þ B cells in cultures of PBMC from Chagasic patients stimulated with either trypomastigote antigen or anti-epimastigote antibodies (Id) We evaluated the frequency of total B cells and CD5 þ B cells in cultures of PBMC from indeterminate or cardiac patients stimulated with TRP or Id using flow cytometry, as described in Materials and Methods. Comparisons between different stimuli showed that Id stimulation led to a significant higher frequency of total B cells (CD19 þ cells) as compared to TRP, and this was true with both clinical forms (Table 2). The increase observed in the percentage of total B cells in cultures stimulated with Id is probably owing to an expansion of CD5 þ B cells, since they were Table 1 Mean percentages ( SD) of T cell populations in PBMC from chagasic patients of indeterminate and cardiac clinical forms stimulated in culture with trypomastigote antigen (TRP) or anti-epimastigote antibodies (Id) %Total % Total CD4 þþ %CD4 þ cells in %CD8 þ cells in Stimulus Clinical form* CD3 þ T cells CD8 þ T cells CD3 þ T cells CD3 þ T cells TRP Indeterminate 43 20 50 34 17 16 10 TRP Cardiac 46 27 56 45 26 11 7 Id Indeterminate 38 19 42 25 14 17 11 Id Cardiac 27 16 34 22 12 12 6 * n-values: indeterminate ¼ 20; cardiac ¼ 19.

Differential Expansion of Lymphocytes in Chagas Disease 95 Fig. 3. Percentage of CD3 þ HLA-DRþ cells (A) and IL-2receptorþ (IL-2Rþ) cells (B) in cultures of PBMC from indeterminate (open circles; n ¼ 20) and cardiac (closed circles; n ¼ 19) patients stimulated with trypomastigote (TRP) antigens or anti-epimastigote antibodies (Id). Means are indicated by the horizontal bars. Before stimulation, percentage of CD3 þ HLA- DRþ cells was 15 and 13 in indeterminate and cardiac patients, respectively [3]. Expression of IL-2R was not determined before stimulation. also found in increased percentages, accounting for 50% or more of the B cells in Id-stimulated cultures (Table 2). We observed that total B cells as well as CD5 þ B cells expanded in similar ratios in cultures of PBMC from indeterminate patients, despite the stimulus, and that their expansion were greater than those observed in cultures of PBMC from cardiac patients (Table 2). The greatest differential expansion of CD5 þ B cells occurred upon stimulation of PBMC from cardiac patients with Id (60% of the 10% responder cells) (Table 2). Cells from cultures stimulated with parasite antigen produce more IFN- than do those stimulated with Id Our data showed that Id stimulation led to relatively low levels of proliferation of cells in vitro, but within that response Id induced a disproportional increase in CD5 þ B cells. In contrast, the TRP stimulation led to higher proliferation than did the Id, and most cells in these responding cultures were activated T cells. It has previously been demonstrated that CD5 þ B cells are a major source of the cytokine IL-10 [20] that is a strong inhibitor of proliferative responses [21]. Also, it has been shown that T cells, which predominate in cultures stimulated with TRP, can be a major source of IFN- [21, 22]. To determine whether these cellular phenotypic profiles correlated with production of IFN- and IL-10, we measured, by ELISA, the production of these two cytokines by the different cultures. The data show that even though Id stimulated much lower levels of proliferation than did TRP, the levels of IL-10 were similar in cultures stimulated with Id or TRP (Table 3). In contrast, after 6 days of culture, we detected higher levels of IFN- in the supernatants from the more proliferative responsive cultures stimulated with TRP, as compared to those exposed to Id (Table 3). Upon comparison, there were no statistical differences in the production levels of the cytokines (either IL-10 or IFN- ) by PBMC from patients with indeterminate or cardiac clinical forms of the infection. Therefore, the data presented in Table 3 are combined from both types of patients. DISCUSSION We have previously demonstrated that antigens derived from the trypomastigote form of T. cruzi, as well as antiparasite antibodies purified from the plasma of Chagasic patients are able to stimulate PBMC from Chagasic patients in vitro. Subsequently, we demonstrated that individuals with the T. cruzi chronic infection, regardless of the clinical form, present elevated Table 2 Mean percentages ( SD) of total B cells (CD19 þ ) and CD5 þ B cells in cultures of PBMC from chagasic patients of indeterminate and cardiac clinical forms stimulated in cultures with trypomastigote antigens (TRP) or anti-epimastigote antibodies (Id)* % total CD19 þ % total CD5 þ %CD5 þ B cells within Stimulus Clinical form B cells CD19 þ B cells CD19 þ B cells population TRP Indeterminate 13.0 6 5.6 3 43 TRP Cardiac 8.0 42 2.8 2 35 Id Indeterminate 17.0 8 8.5 4 50 Id Cardiac 10.0 5 6.0 4 60 * Mean percentages ex vivo of total CD19 þ cells were 10 4 (indeterminate), 10 3 (cardiac) and for CD5 þ B cells, 3 2 (indeterminate), 3 3 (cardiac). P < 0.05 between indeterminate versus cardiac after TRP stimulation. P < 0.05 between indeterminate versus cardiac after Id stimulation. A statistically significant difference (P < 0.05) was observed upon comparison of TRP versus Id stimulation of cells from cardiac patients.

96 W. O. Dutra et al. Table 3 IL-10 and IFN-gð production in cultures of PBMC from indeterminate and cardiac chagasic patients stimulated in cultre with trypomastigote antigens (TRP) or anti-epimastigote antibodies (Id) TRP Cytokine Stimulant 24 h* 72 h* 6 days* L-10 TRP 0.5 0.5 0.2 0.1 1.4 2.0 IL-10 Id 0.5 0.4 0.2 0.1 0.5 0.7 IFN- TRP 0.2 1.2 1.3 9.6 8.0 IFN- Id 0.2 0.8 0.9 2.1 1.6 Unit: ng/ml. * No statistically significant differences were found between different time points or different stimuli. levels of circulating activated T cells as well as CD5 þ B cells [3]. We now report that each of these elevated populations respond differently to the two stimulatory molecules, TRP or Id. These data show that the in vitro stimulation of PBMC from Chagasic patients with T. cruzi-derived antigens led to high proportions of activated T cells, while host-derived anti-t.cruzi antibodies preferentially stimulated CD5 þ B cells. The differential expansion of CD5 þ B cells in response to Id was even more pronounced if cultured cells were from patients with the severe, cardiac form of the disease. In fact, CD5 þ B cells were 60% of the total B cell population found in Id-stimulated PBMC from cardiac patients, while this proportion was lower in the other cultures (35 50%). This suggests that, among B cells from cardiac patients, Id induces a biased expansion of CD5 þ B cells. CD5 þ B cells have been correlated with pathology in the experimental Chagas disease [4] and we found that their frequency is elevated in the peripheral blood of patients with Chagas disease [3]. Also, this cell population is increased at the sites of lesion in some autoimmune diseases [6, 7]. Our data show that Id, a stimulus constantly present in chronic Chagasic patients, causes the preferential expansion of a cell population that has been correlated with Chagasic and autoimmune pathology in patients with Chagas disease. In addition, we also found that CD8 þ T cells were significantly, proportionally higher in cultures stimulated with Id, than those exposed to TRP. Considering that both types of stimuli are present in the chronic chagasic patients, they may also be involved with activation of those cell populations in vivo. The response of T cells to the different stimuli was also analysed and two major observations were made: firstly, CD8 þ T cells comprise approximately 50% of the total T cells in cultures stimulated by Id, while in TRP-stimulated cultures of such cells they constitute only 24% of the cells; secondly, stimulation of PBMC from Chagasic patients with TRP led to a preferential expansion of the activated T cells as compared to using Id as the stimulus. CD8 þ T cells have been shown to be the major component of inflammatory infiltrates in Chagasic cardiac lesions [5]. The presence of these cells, which are capable of cytotoxic activity and IFN- production, in the site of the lesions, suggest that they may play a critical role in the damage processes leading to severe inflammation and cardiac pathology. We have previously shown that circulating activated CD4 þ and CD8 þ T cells are present in high levels in Chagasic patients [3]. Other studies have correlated high levels of T cells expressing HLA-DR with autoimmune pathology [8]. The role of IFN- as a factor that augments in vitro proliferation of PBMC has been reported by many groups [21]. Also, it is known that this cytokine is able to induce an increase in the expression of HLA-DR molecule, a class II MHC gene product, that is important in antigen presentation [22]. In our study, we found a positive correlation between the elevated expression of the HLA-DR molecule, IFN- production and proliferation. Cultures stimulated with TRP showed the highest proliferation and also displayed the highest levels of HLA-DR expression and IFN- production (even though not statistically significant for the IFN- production), as compared to Id-stimulated cultures. It is possible that the parasite-related antigens induce the IFN- secretion in vitro and, as a consequence, the HLA-DR expression is increased, facilitating antigen presentation and leading to optimum proliferation. We have previously reported that the HLA-ABC antigen expression is upregulated on myocardial cells and HLA-DR antigen expression is upregulated on endothelial cells in lesions in the hearts of patients who died of cardiopathies [23]. Since IL-10 can have inhibitory effects on proliferation in vitro [21, 22] we speculate that the lower proliferative responses observed in Id-stimulated cultures might be correlated with a high IL-10 production. However, the same levels of IL-10 were found in each type of culture albeit in the presence of disproportional proliferative responses. Furthermore, we have demonstrated by the PCR analysis of specific mrna that the IL-10 message is elevated early on in cultures stimulated with Id, as compared to parasite antigens [24]. Thus, it remains possible that the lack of difference found in IL-10 production by ELISA during these cultures is owing to a consumption by the cells in Id-stimulated cultures rather than a moderate production of this cytokine. The finding that parasite antigens and human antiepimastigote antibodies, both present in the host during the chronic phase of Chagas disease, stimulate different cell populations that may be related to pathologic processes points out elements and factors that may play a role in the development and maintenance of immunopathology in the human Chagas disease. Furthermore, the preferential stimulation of both CD5 þ B cells and CD8 þ T cells by Id is especially noteworthy as a result of their correlation with severe disease [13]. An additional possible interpretation for our findings concerning activated T cells that appear to be preferentially stimulated by parasite antigens, could be that these cells contribute to the control of parasitemia in chronic patients, as proposed in T. cruzi infected mice [25]. ACKNOWLEDGMENTS The authors are grateful to Smita Mauze and Olindo Martins-Filho for outstanding help with ELISA assays and phenotypic analysis, respectively. Our special thanks to Alexandre José Fernandes

Differential Expansion of Lymphocytes in Chagas Disease 97 (in memorian), coordinator at the time, of Centro de Estudos Emmanuel Dias in Bambuí, MG, where patients were identified and to Dr Ken Gollob for critical review of this manuscript. This work was supported by CAPES, FIOCRUZ-PAPES, ICIDR/NIH (AI-26505) and DNAX Reasearch Institute, supported by The Scherring Plough Corporation. REFERENCES 1 Brener Z. Pathogenesis and immunopathology of chronic Chagas disease. Mem Inst Osw Cruz 1987;82:205. 2 Cossio PM, Damilano G, de la Vega MT et al. In vitro reactions between lymphocytes of chagasic individuals and heart tissue. Medicine (Buenos Aires) 1976;36:287 91. 3 Dutra WO, Martins-Filho OA, Cançado JR et al. Activated T and B lymphocytes in peripheral blood of patients with Chagas disease. Internat Immunol 1994;6:499 504. 4 Minoprio P, Itohara S, HeusSeries C, Tonegawa S, Coutinho A. Preferential expansion of LY-1 B and CD4 CD8-T cells in the polyclonal lymphocyte responses to murine Trypanosoma cruzi infection. Internat Immunol 1989;1:176 83. 5 Reis Dd Jones EM, Tostes S, Gazzinelli G, Colley DG, McCurley T. Characterization of inflammatory infiltrates in chronic chagasic myocardial lesions: presence of TNF-alfa þ cells and dominance of Granzyme Aþ, CD8 þ lymphocytes. Am J Trop Med Hyg 1993;43:637 41. 6 Becker J, Weber C, Storch S, Federlin K. Relationship between CD5 þ B lymphocytes and the activity of lupus erythematosus and rheumatoid arthritis. Clin Immunol Immunopathol 1990;56:219 22. 7 Smith HR, Olson RR. B lymphocytes in systemic lupus erythematosus and rheumatoid arthritis. J Rheumatol 1990;17:833 9. 8 Jahn B, Burmester GR, Stock P, Rohwer P, Kalden JR. Functional and phenotypic characterization of activated T cells from intraarticular sites in inflammatory joint diseases. Possible modulation of the CD3 antigen. Scand J Immunol 1987;26:45 50. 9 Colley DG. Occurrence, roles and uses of iditypes and anti-idiotypes in paraitic diseases. In: Cerny J, Hiernauz J., eds. Idiotypic Network and Diseases. Washington: ASM, 1990:71 105. 10 Colley DG, Howard MJ. The fundamental nature of Chagas disease: a view provided by immune responses and idiotypes. Jap J Trop Med Hyg 1991;19:175 90. 11 Jones EM, Colley DG, Tostes S Jr, Lopes ER, Vneucek-Jones CL, McCurley TL. Amplification of a Trypanosoma cruzi DNA sequence from inflammatory lesions of human chagasic cardiomiopathy. Am J Trop Med Hyg 1993;48:348 57. 12 Gazzinelli RT, Gazzinelli G, Cançado JR, Cardoso JE, Brener Z, Colley DG. Two models of idiotypic stimulation of T lymphocytes from patients with Chagas disease: correlation with clinical forms of infection. Res Immunol 1990;140:757 61. 13 Reis Dd Gazzinelli R, Gazzinelli G, Colley DG. Antibodies to T. cruzi express idiotypic patterns that can differentiate between patients with asymptomatic and severe forms of Chagas disease. J Immunol 1993;150:611 8. 14 Morato MJF, Brener Z, Cançado JR, Nunes RMB, Chiari E, Gazzinelli G. Cellular Immune responses of chagasic patients to antigens derived from different strains and clones. Am J Trop Med Hyg 1986;35:505 11. 15 Gazzinelli RT, Parra JFC, Correa-Oliveira R et al. Idiotypic/anti idiotypic interactions in schistosomiasis and Chagas disease. Am J Trop Med Hyg 1988;39:288 94. 16 Gazzinelli G, Katz N, Rocha RS, Colley DG. Immune response during human schistosomiasis mansoni. X. Production and standartization of an antigen induced mitogenic activity by peripheral blood mononuclear cells from treated, but not active cases of schistosomiasis. J Immunol 1983;130:2981 7. 17 Abrams JS. Immunoenzymatic assays of mouse and human cytokines using NIP-linked anti-cytokine antibodies. In: Colligan JE, Kruisbeek AM, Margulies DH, Shevach EM, Strober W., eds. Current Protocols Immunology. New York: John Wiley and Sons, 1995: 6.20.1 6.20.15. 18 Dutra WO, Luz ZMP, Cançado JR et al. Influence of parasite presence on the immunologic profile of peripheral blood mononuclear cells from chagasic patients after specific drug therapy. Parasite Immunol 1996;18:579 85. 19 Pantaleo G, Zocchi MR, Ferrini S et al. Human cytolytic cell clones lacking surface expression of T cell receptor alpha-beta or gammadelta. Evidence that surface structures other than CD3 and CD2 molecules are required for signal transduction. J Exp Med 1988;168:345 51. 20 O Garra A, Chang R, Go N, Hastings R, Haughton G, Howard M. Ly-1 B (B1) cells are the main source of B-cell derived interleukin 10. Eur J Immunol 1992;22:711 5. 21 O Garra A, Murphy K. Role of cytokines in determining T-lymphocyte function. Curr Opin Immunol 1994;6:458 66. 22 Howard MC, Myiajima A, Coffman RL. T-cell derived cytokines and their receptors. In: Paul W. eds. Fundamental Immunology. Bethesda: Raven Press, 1993: 763 800. 23 Reis Dd Jones EM, Tostes S, Lopes ER et al. Expression of major histocompatibility complex antigens and adhesion molecules in hearts of patients with chronic Chagas disease. Am J Trop Med Hyg 1993;49:192 200. 24 Dutra WO, Gollob KJ, Pinto-Dias JC et al. Cytokine mrna profile of peripheral blood mononuclear cells isolated from individuals with Trypanosoma cruzi chronic infection. Scand J Immunol 1997;45: 74 80. 25 Tarleton RL, Sun J, Zhang C, Postan M. Depletion of T-cell subpopulations result in exacerbation of myocarditis and parasitism in experimental Chagas disease. Infec Immunity 1994:62:1820 9.