|Home | About | Journals | Submit | Contact Us | Français|
We have previously shown that splenocytes from mice acutely infected with Trypanosoma cruzi exhibit high levels of nitric oxide (NO)-mediated apoptosis. In the present study, we used the gamma interferon (IFN-γ)-knockout (IFN-γ−/−) mice to investigate the role of IFN-γ in modulating apoptosis induction and host protection during T. cruzi infection in mice. IFN-γ−/− mice were highly susceptible to infection and exhibited significant reduction of NO production and apoptosis levels in splenocytes but normal lymphoproliferative response compared to the infected wild-type (WT) mice. Furthermore, IFN-γ modulates an enhancement of Fas and Fas-L expression after infection, since the infected IFN-γ−/− mice showed significantly lower levels of Fas and Fas-L expression. The addition of recombinant murine IFN-γ to spleen cells cultures from infected IFN-γ−/− mice increased apoptosis levels, Fas expression, and NO production. In the presence of IFN-γ and absence of NO, although Fas expression was maintained, apoptosis levels were significantly reduced but still higher than those found in splenocytes from uninfected mice, suggesting that Fas–Fas-L interaction could also play a role in apoptosis induction in T. cruzi-infected mice. Moreover, in vivo, the treatment of infected WT mice with the inducible nitric oxide synthase inhibitor aminoguanidine also led to decreased NO and apoptosis levels but not Fas expression, suggesting that IFN-γ modulates apoptosis induction by two independent and distinct mechanisms: induction of NO production and of Fas and Fas-L expression. We suggest that besides being of crucial importance in mediating resistance to experimental T. cruzi infection, IFN-γ could participate in the immune response control through apoptosis modulation.
The protozoan parasite Trypanosoma cruzi causes a persistent, lifelong infection, which can lead to Chagas’ disease, a major health problem in Latin America. T. cruzi-infected individuals may develop a chronic disease characterized by cardiopathy or nervous dysfunction of the digestive tract (for a review, see reference 37). In mice and humans, T. cruzi infection leads to an intense suppression of the lymphoproliferative response to mitogens and antigens. This impairment of the proliferative response has been ascribed to many mechanisms, including decreased interleukin-2 (IL-2) production and reduced expression of IL-2 receptor (IL-2R) by spleen cells (43). In the acute phase of experimental T. cruzi infection, there is intense parasite replication, which in resistant hosts can be controlled by innate and adaptive immune responses (36).
It has been demonstrated that the cytokines gamma interferon (IFN-γ) and tumor necrosis factor alpha (TNF-α) are involved in mediating a protective response to T. cruzi (8, 9, 25, 47, 52). IFN-γ is synthesized shortly after infection, mainly by IL-12 and TNF-α-activated NK cells (4, 8). Together with TNF-α, IFN-γ leads to activation of inducible nitric oxide synthase (iNOS) (16), the enzyme that catalyzes nitric oxide (NO) synthesis by macrophages (38). NO has been implicated in parasite killing during T. cruzi and in other protozoan (2, 20, 22, 52), bacterial (10, 23, 44), and fungal (53) infections. Despite its importance as a microbicidal agent, NO has been shown to be involved in the establishment and maintenance of lymphocyte unresponsiveness in mice infected with several parasites (1, 7, 23, 42, 50). In addition, NO induces apoptotic cell death in many different cells, in vitro and in vivo (3, 6, 15, 18, 31).
Apoptosis is a naturally occurring mechanism of cell death involved in a large range of physiological as well as pathological events and is characterized by a set of specific alterations in cell morphology. This finely regulated mechanism of cell death has been shown to be of critical importance in immune response control (41). Apoptosis induction in immune cells can be modulated by many factors, including the CD95 receptor-ligand system (Fas-Fas-L) (48) and cytokines such as TNF-α and IFN-γ (17, 30, 34). The CD95 ligand (Fas-L) is a type II transmembrane molecule which is expressed in many tissues, including the spleen, thymus, lung, testis, heart, and small intestine (48). Fas, a type I membrane protein, belongs to the TNF/nerve growth factor receptor family and is constitutively expressed on the surface of activated T and B lymphocytes, hepatocytes, and several other tissues and tumors. Fas-Fas-L-induced apoptosis has been demonstrated to be of crucial importance in regulating the immune response (48). Moreover, IFN-γ-induced Fas expression has been implicated in induction of apoptosis in mice infected by the protozoa Toxoplasma gondii (30). Although apoptosis has been demonstrated to be enhanced in other parasite infections (11, 30, 33), including infection of mice by the protozoa T. cruzi (35), a correlation between Fas expression and apoptosis induction in vivo has not yet been made.
We have previously shown that spleen cells from acutely T. cruzi-infected mice present high levels of apoptosis, which is partially mediated by IFN-γ and TNF-α-induced NO (35). In this work, we used a genetically manipulated mice lacking IFN-γ to investigated if this cytokine could play a direct role in apoptosis induction during the acute phase of T. cruzi infection. We found that T. cruzi infection in mice leads to an enhancement of Fas and Fas-L expression which is modulated by IFN-γ in a NO-independent manner. Our results demonstrate that besides modulating NO-induced apoptosis, IFN-γ could lead to apoptosis induction by mediating an enhancement in Fas and Fas-L expression, suggesting a potential role of this cytokine in control of the immune response during experimental T. cruzi infection.
Five- to six-week-old female C57BL/6 wild-type (WT) and IFN-γ-deficient (IFN-γ−/−) (13) mice were maintained under specific-pathogen-free conditions. C57BL/6 WT mice were obtained from the animal house of the Division of Immunology, School of Medicine of Ribeirão Preto, University of São Paulo, São Paulo, São Paulo, Brazil. IFN-γ−/− mice were generously provided by R. T. Gazzinelli (Federal University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil).
To investigate whether NO is involved in mediating Fas expression during T. cruzi infection, WT mice received aminoguanidine (AG; 50 mg/kg of body weight; RBI, Natick, Mass.) diluted in phosphate-buffered saline (PBS) intraperitoneally (i.p.) daily. The first inoculation was done 4 h before the infection, and animals were treated for 11 days thereafter. Control mice received PBS only.
Mice were infected i.p. with 103 blood-derived trypomastigote forms of the Y strain. The levels of parasitemia were evaluated in 5 μl of blood drawn from the tail vein.
Suspensions of splenocytes from uninfected and infected WT or IFN-γ−/− mice were washed in Hanks’ balanced salt solution (HBSS) and treated with lysing buffer (9 parts of 0.16 M ammonium chloride and 1 part of 0.17 M Tris-HCl [pH 7.5]) for 4 min. The erythrocyte-free cells were then washed three times in HBSS and adjusted to 3 × 106 cells/ml in RPMI 1640 (Flow Laboratories, Inc., McLean, Va.) supplemented with 5% fetal calf serum (HyClone, Logan, Utah), 2-mercaptoethanol (5 × 10−5 M), l-glutamine (2 mM), and antibiotics (all purchased from Sigma Chemical Co. St. Louis, Mo.). The cell suspension was distributed at 1 ml per well in 24-well tissue culture plates (Corning Glass Works, Corning, N.Y.) and cultured for 48 h at 37°C in a humidified 5% CO2 atmosphere, in the presence or absence of AG (300 μM), l-NMMA (NG-monomethyl-l-arginine; Sigma) (500 μM), recombinant murine IFN-γ (rMuIFN-γ; Genentech Inc., San Francisco, Calif.) (10 U/ml) or IFN-γ plus l-NMMA. The cells were subsequently used to assay apoptosis and Fas or Fas-L expression, and the supernatant was collected to evaluate NO production.
The concanavalin A (ConA)-induced T-cell proliferative response was evaluated in splenocytes harvested from three uninfected and three infected WT or IFN-γ−/− mice on the day 11 of infection. Cells (5 × 105/well) were cultured in the presence of ConA (2 μg/ml; Sigma), in triplicates, in flat-bottom microwell tissue culture plates at a final volume of 200 μl. Cells were maintained at 37°C in a humidified 5% CO2 atmosphere for 3 days. During the last 8 h of culture, [methyl-3H]thymidine (0.5 μCi/well; Amersham, Chicago, Ill.) was added. Cells were collected with a cell harvester (Cambridge Technology, Inc., Watertown, Mass.) onto glass filters; incorporated radioactivity was quantified by liquid scintillation (Beckman Instruments Inc., Fullerton, Calif.).
Cell-free culture medium was obtained by centrifugation and assayed for nitrite content. In addition, serum samples were collected from the retro-orbital plexus of uninfected or infected WT or IFN-γ−/− mice on different days after T. cruzi infection. The nitrate in the above samples was reduced to nitrite with nitrate reductase as described elsewhere (52), and the nitrite concentration was then determined by the Griess method (21). For this assay, 0.1 ml of culture medium or serum was mixed with 0.1 ml of Griess reagent in a multiwell plate, and the absorbance at 550 nm read 10 min later. The NO2 concentration was determined by reference to a NaNO2 standard curve (1 to 200 μM).
Two different methods were used for DNA labeling. The first was based on the use of propidium iodide (PI) (Sigma) as previously described (39). Briefly, 1.5 × 106 spleen cells were washed in HBSS and gently resuspended in 750 μl of hypotonic fluorochrome solution (50 μg of PI/ml in 0.1% sodium citrate plus 0.1% Triton X-100; Sigma) in 12- by 75-mm polypropylene tubes (Becton Dickinson, Mountain View, Calif.). The tubes were incubated overnight at 4°C in the dark before flow cytometry (FCM) analysis. The second method involved the use of 7-amino-actinomycin D (7-AAD; Calbiochem-Novabiochem Corp., La Jolla, Calif.) as previously described (45), with few modifications. Briefly, cells (3 × 106) were washed in PBS, resuspended in 500 μl of 7-AAD (10 μg/ml) in PBS, and incubated for 20 min at 4°C, shielded from light. Fluorescence of individual nuclei labeled with PI (FL-2) or with 7-AAD (FL-3) was measured in a fluorescence-activated cell sorting flow cytometer (Becton Dickinson, San Jose, Calif.) after gating cells to exclude debris and necrotic cells. At least 104 cells of each sample were analyzed. All measurements were made at the same instrument settings.
Expression of Fas and Fas-L was evaluated by incubating splenocytes (106 cells/100 μl) from uninfected or infected WT or IFN-γ−/− mice, for 30 min at 4°C, with 0.5 μg of anti-CD16/CD32 monoclonal antibody (Fc block), followed by the addition of 0.5 μg of fluorescein isothiocyanate (FITC)-labeled anti-Fas or 0.5 μg of phycoerythrin-labeled anti-Fas-L. Background staining was determined by incubating cells with 0.5 μg of FITC-labeled antitrinitrophenol hamster immunoglobulin G1 diluted in 1% bovine serum albumin (Sigma) in PBS for 30 min at 4°C in the dark. Cells were then washed twice, incubated with 500 μl of 7-AAD (10 μg/ml) for 20 min (4°C), and analyzed on a flow cytometer as described elsewhere (45). All antibodies were purchased from Pharmingen (San Diego, Calif.). Multivariate data analysis was performed with the LYSYS II software (Becton Dickinson) by setting a gate on the apoptotic cells or lymphocytes on a FL-3 versus FSC scatter dot plot and determining the expression of Fas in FL-1 histograms.
Results are expressed as the mean ± standard error of the mean (SEM) or standard deviation (SD) of the indicated number of animals or experiments. Statistical analysis was performed using analysis of variance followed by the Student-Newman-Keuls test (INSTAT software; GraphPad, San Diego, Calif.). A P value of <0.05 was considered to indicate significance.
WT and IFN-γ−/− mice were infected with 103 trypomastigote forms of T. cruzi, and parasitemia and mortality were evaluated. In comparison to WT mice, IFN-γ−/− mice exhibited significantly increased parasitemia after day 8 of infection (Fig. (Fig.1A).1A). This increased parasitemia remained uncontrolled until day 11, when the animals presented a sixfold increase in parasitemia. On day 13, all IFN-γ−/− mice had died without showing any control of parasitemia, whereas the WT group survived acute infection (Fig. (Fig.1B).1B).
The lymphoproliferative response to ConA was evaluated in splenocytes from WT and IFN-γ−/− mice on day 11 after infection. As previously described (21), the proliferative response of spleen cells from infected WT mice was dramatically reduced (30% of the response found for noninfected controls). In contrast, splenocytes from the IFN-γ−/− infected mice showed a proliferative response similar to that of the noninfected controls (Fig. (Fig.2).2).
To study the involvement of IFN-γ-induced NO production in mediating apoptosis during the acute phase of T. cruzi infection, splenocytes were harvested from WT and IFN-γ−/− mice infected for 11 days with T. cruzi, and NO production and apoptosis levels were evaluated after 48 h of culture. The results showed high NO production by splenocytes from infected WT mice but not by cells from IFN-γ−/− mice. Nitrite levels found in culture supernatants from infected WT mice were 20 times higher than those in cultures from knockout mice (Fig. (Fig.3A).3A). In addition, after culture, 16.9% of cells from infected IFN-γ−/− mice were apoptotic, whereas the percentage of WT apoptotic cells was 62.8 (3.7 times higher) (Fig. (Fig.3B).3B).
Similar results were found when we analyzed serum nitrate concentration and apoptosis levels in ex vivo splenocytes from T. cruzi-infected WT and IFN-γ−/− mice. A 70.9% reduction of NO production was found in sera from infected IFN-γ−/− mice compared with NO in sera from infected WT mice (Fig. (Fig.3C).3C). Similarly, apoptosis was reduced by 52.7% in spleens from infected IFN-γ−/− mice.
To investigate the mechanism of apoptosis induction during acute T. cruzi infection in mice, splenocytes were harvested from WT and IFN-γ−/− mice on day 11 after infection and cultured for 48 h; then expression of Fas was quantified by FCM analysis. The percentages of Fas-positive splenocytes were similar in uninfected WT and IFN-γ−/− mice. However, we found a significant increase of Fas-positive splenocytes in infected WT (around 180%) but not infected IFN-γ−/− mice (Fig. (Fig.4A).4A). Remarkably, analysis of Fas expression in the apoptotic gate of cultured splenocytes revealed no difference between WT and IFN-γ−/− mice, indicating that most cells that die in culture express Fas antigen (Fig. (Fig.4B).4B).
We next investigated if the enhancement of Fas expression was implicated in apoptosis induction and if Fas expression could be related to the enhanced NO production during the infection. Splenocytes harvested from uninfected and infected WT or IFN-γ−/− mice were cultured in the presence of rMuIFN-γ with or without l-NMMA, and the amount of nitrite, level of apoptosis, and percentage of Fas-expressing cells were evaluated. The addition of 10 U of rMu IFN-γ per ml significantly increased NO production (Fig. (Fig.5A),5A), Fas expression (Fig. (Fig.5B),5B), and apoptosis levels (Fig. (Fig.5C)5C) by splenocytes from infected IFN-γ−/− mice to levels similar to those found in splenocytes from infected WT mice. However, when splenocytes from infected mice were cultured in the presence of IFN-γ plus l-NMMA, NO production (Fig. (Fig.5A)5A) and apoptosis levels (Fig. (Fig.5B),5B), but not Fas expression (Fig. (Fig.5C),5C), were significantly reduced. The finding that Fas expression is elevated in the presence of IFN-γ and absence of NO appears to indicate that NO is not necessary for the enhancement of Fas expression in these cultures. The inhibition of NO production by l-NMMA addition leads to reduced apoptosis levels in splenocytes from WT mice, but it was still significantly higher than the levels found in normal splenocytes (Fig. (Fig.5C).5C). Inhibition of NO production did not change the percentage of Fas-positive cells (Fig. (Fig.5B),5B), suggesting again that NO is not crucial for the enhancement of Fas expression.
We next investigated the expression of Fas and Fas-L in vivo during the course of the acute phase of T. cruzi infection. We found that T. cruzi infection leads to an enhancement in the expression of both Fas and Fas-L (Fig. (Fig.6).6). High levels of Fas expression were detected in infected WT and IFN-γ−/− mice at day 5 after infection. However, whereas in the infected WT mice Fas expression increased steadily until day 13 after infection, in the IFN-γ−/− mice Fas expression decreased after day 7 of infection. In the infected WT mice, the percentage of Fas-expressing cells remained elevated throughout the acute phase (Fig. (Fig.6A)6A) but declined to normal levels in the chronic phase (data not shown). Interestingly, in WT mice, the increased percentage of Fas expression by splenocytes was coincident with the presence of circulating parasites. At this time point postinfection (day 11), the percentages of CD95+ CD4+, CD95+ CD8+, CD95-L+ CD4+, and CD95-L+ CD8+ cells were 15.75 ± 0.53, 7.83 ± 0.31, 1.5 ± 0.2, and 3.4 ± 0.2, respectively.
Measurement of Fas-L expression after T. cruzi infection in cells from WT and IFN-γ−/− mice showed that the percentage of Fas-L-expressing splenocytes is normal up to day 9 after infection. By day 11, Fas-L expression was increased in WT and IFN-γ−/− mice, but in the absence of IFN-γ, the percentage of Fas-L-expressing cells was significantly lower. By day 20 after infection, Fas-L expression was still elevated in the WT mice (Fig. (Fig.6B).6B). Since all of the IFN-γ−/− mice had died by day 13 of infection, we were unable to evaluate Fas or Fas-L expression at later time points.
To determine whether the enhancement of Fas and Fas-L expression could be modulated by the high levels of NO produced during the acute phase of infection, we treated T. cruzi-infected WT mice with AG and evaluated the levels of apoptosis, Fas, and Fas-L expression. Splenocytes were analyzed freshly after harvesting and after culture in presence or absence of AG. Treatment of infected mice with AG led to a reduction of NO production in vivo (Fig. (Fig.7A)7A) and in vitro (Fig. (Fig.7E)7E) and to a decrease in apoptosis levels in freshly isolated (Fig. (Fig.7B)7B) or cultured (Fig. (Fig.7F)7F) splenocytes. However, the inhibition of NO production did not modify Fas or Fas-L expression (Fig. (Fig.7C7C and D), even after culture with additional AG (Fig. (Fig.7G7G and H). These results demonstrated that although NO induces apoptosis in cells from T. cruzi-infected mice, it is not required for the induction of Fas and Fas-L expression in vivo.
In this present work, we confirmed the previous suggested role of IFN-γ in the modulation of lymphoproliferative response and apoptosis induction (35), and we also investigated if IFN-γ contributes in the apoptosis induction by another pathway than through NO production. Our observations suggest, for the first time, that besides being implicated in protection and apoptosis induction by mediating NO production, IFN-γ also directly modulates Fas and Fas-L expression in vivo during the acute phase of T. cruzi infection. Considering the broad importance of Fas-induced apoptosis in immune response control, our data suggest a crucial role for IFN-γ in controlling the immune response during the acute phase of infection with T. cruzi.
In accordance with previous observation (8, 25), IFN-γ plays a central role in the resistance to infection with T. cruzi, since IFN-γ−/− mice are more susceptible to infection than WT mice (Fig. (Fig.1).1). We also verified that the T-cell unresponsiveness could be due to IFN-γ, through the induction of NO production, since infected IFN-γ−/− mice did not produce NO and presented normal ConA-induced cell proliferation (Fig. (Fig.2).2). Thus, IFN-γ produced early after T. cruzi infection (8) leads to NO production, which in turn mediates host resistance (52) and also immunosuppression, as demonstrated in mice infected with several other parasites (1, 7, 42, 50). Absence of IFN-γ also resulted in a dramatic reduction in apoptosis of splenocytes from infected mice. Although these results suggest that cell unresponsiveness is due to apoptosis, direct evidence to link these phenomena in vivo is still missing. The occurrence of apoptosis in vivo could also be a consequence of the decreased expression of IL-2R induced by the presence of the parasites (27, 28), or of the parasite-induced polyclonal cell activation during this phase of infection (37), since the various enzymes required for cell replication are susceptible to the inhibitory effects of NO (29, 38).
The in vivo inhibition of NO production leads to a significant reduction of apoptosis levels in splenocytes from T. cruzi-infected mice (Fig. (Fig.7).7). However, in vivo inhibition of NO production did not bring apoptosis to the lower levels found for noninfected or IFN-γ−/− mice (Fig. (Fig.3B3B and D). This observation led us to inquire about a direct role for IFN-γ in apoptosis induction during the acute phase of the infection. In this context, IFN-γ has been demonstrated to play a critical role in the regulation of Fas antigen expression and induction of apoptosis in murine and human cells (30, 31). We found that infected IFN-γ−/− mice showed Fas expression and apoptosis levels significantly lower than those in infected WT mice, suggesting that the enhancement in Fas expression is mediated by IFN-γ. Induction of Fas expression could be another pathway of IFN-γ-induced apoptosis, as suggested to occur in mice infected with Toxoplasma gondii (30).
Direct evidence for a role of IFN-γ in mediating apoptosis induction during the acute phase of T. cruzi infection was obtained when splenocytes from IFN-γ−/− mice were cultured in the presence of rMuIFN-γ, which simultaneously restored NO production and apoptosis levels (Fig. (Fig.5).5). In addition, IFN-γ also restored Fas expression. Although NO has been found to upregulate Fas expression (19, 49), inhibition of NO production by addition of l-NMMA did not change significantly the amount of Fas-positive cells (Fig. (Fig.4),4), indicating that in these experiments, Fas expression is modulated by IFN-γ but not by NO. Moreover, spleen cells from T. cruzi-infected WT mice treated in vivo with AG (Fig. (Fig.7)7) or from iNOS knockout mice (unpublished data) displayed reduced apoptosis levels but not decreased Fas expression, despite the significant inhibition (Fig. (Fig.7)7) or abrogation of inducible NO production. These observations support the hypothesis that NO mediates apoptosis induction but is not critical for the enhancement in Fas expression. Moreover, in contrast to the activation-induced apoptosis showed to occur in vitro in cells from T. cruzi-infected mice (40) and to be modulated by Fas and Fas-L interaction, the spontaneous apoptosis that occurs in vivo seems to be mainly due to NO (Fig. (Fig.7B).7B). Taken together, these data suggest that IFN-γ mediates induction of apoptosis in T. cruzi-infected mice by inducing NO production and Fas and Fas-L expression. As suggested previously, such a mechanism of apoptosis induction could contribute to control of the immune response and possibly to limit the host tissue damage during infection (24, 31, 46, 51). It is noteworthy that both IFN-γ−/− and IFN-γR−/− mice showed an enhanced inflammatory response in the liver after infection (25, 43a).
WT mice exhibits increased Fas and Fas-L expression in spleen cells in the acute phase of infection. Surprisingly, during the early phase of infection, Fas but not Fas-L expression in splenic cells was increased also in the IFN-γ−/− mice. Although we cannot explain this result, the induction of Fas in the absence of IFN-γ could be due to the presence early in infection of other cytokines such as TNF-α, IL-1, and IL-12 (4, 54), which have been shown to upregulate Fas expression (14, 32, 34). These results suggest that IFN-γ is not essential to induce Fas expression early after infection but is crucial for its maintenance during the acute phase. IFN-γ was also required for the enhanced Fas-L expression after day 11 of infection. This increased Fas and Fas-L expression is coincident with the highest levels of apoptosis found during this phase of infection (35). Since the interaction of Fas and Fas-L results in apoptosis induction, it is possible that this interaction mediates induction of apoptosis in the early phase of infection. As discussed, a role for IFN-γ in maintenance of Fas expression and induction of Fas-L expression in the acute phase of infection, independently of NO production, seems to be a feasible hypothesis, since inhibition of NO production did not completely block apoptosis (Fig. (Fig.7).7). Another tempting possibility is that the high levels of NO produced during the infection enhance cell susceptibility to Fas-mediated apoptosis, as previously demonstrated to occur in pancreatic beta cells (32). The fact that the majority of apoptotic cells express Fas, even in IFN-γ−/− mice (Fig. (Fig.4),4), may favor this possibility.
Fas-mediated apoptosis of CD4+ T cells was shown to promote the exacerbation of T. cruzi growth in infected macrophages in vitro, suggesting that the occurrence of apoptosis during experimental T. cruzi infection may have a deleterious role (40). However, in vivo, apoptosis could be important to mediate deletion of potentially autoreactive T cells (12). Therefore, apoptosis induction may contribute to control of the immune response and possibly to limit host tissue damage during the infection, as suggested to occur in autoimmune interstitial nephritis (46) and in viral infection (24), in which the absence of Fas-mediated apoptosis results in severe inflammation of host tissues. In this context, Fas-mediated apoptosis is required for the resolution of lesions in mice infected with Leishmania major (11). Strikingly, Fas-defective mice are highly susceptible to infection with this parasite (26). Similarly, lpr mutant mice are more susceptible than WT mice to T. cruzi infection (5), suggesting that somehow Fas (and probably Fas-induced apoptosis) participate in the modulation of an efficient response against T. cruzi. Experiments to test this possibility are under way.
In summary, results of this study suggest that in addition to playing a crucial role in parasite killing, IFN-γ may be implicated in control of the immune response, by inducing NO production and Fas and Fas-L expression, which in turn lead to apoptosis induction during experimental T. cruzi infection. NO-mediated parasite killing and control of inflammatory response through apoptosis induction could be involved in limiting the damage to host tissues and promoting the establishment of chronic disease in T. cruzi-infected mice.
This study was supported by grants 96/04304-7 and 96/4118-9 from FAPESP and by fellowships from CNPq (F.Q.C. and J.S.S.).
We thank R. N. Kitsis for critical review of the manuscript.