INTRODUCTION
Although it is 103 years since the discovery of Chagas disease, vector-borne transmission continues to be the most common mode of infection. This has led to national health programme efforts in an attempt to eradicate intra- and peri-domiciliary wild triatomines, which can invade human dwellings.
In endemic areas, such as Brazil, Uruguay, Chile and sustainable areas of Argentina, Bolivia and Paraguay (Schofield et al. Reference Schofield, Jannin and Salvatella2006), Triatoma infestans-mediated infections are gradually being controlled though not eradicated. The presence of sylvatic triatomines, such as Rhodnius prolixus, has been found in other regions of Brazil, such as Amazonia. These triatomines have the ability to invade human houses and infect people. Recently, oral transmission has resulted in many outbreaks due to the consumption of food or drink contaminated with triatomine feces containing Trypanosoma cruzi (Dujardin et al. Reference Dujardin, Garcia-Zapata, Jurberg, Roelants, Cardozo, Panzera, Dias and Schofield1991; Coura et al. Reference Coura, Junqueira, Fernandes, Valente and Miles2002; Yoshida, Reference Yoshida2008; Igreja, Reference Igreja2009).
Under this scenario, over 50% of acute Chagas cases are related to oral infection and several micro-outbreaks have occurred between 1968 and 2000 (Coura et al. Reference Coura, Junqueira, Fernandes, Valente and Miles2002).
The differences in clinical and immunological responses between oral and vector-borne infections vary according to the degree of pathogenicity and the biochemical features of each T. cruzi strain (Andrade et al. Reference Andrade, Carvalho and Figueira1970, Andrade, Reference Andrade1985, Andrade and Magalhães, Reference Andrade and Magalhães1997). The Y strain is characterized by a slender morphology and high parasitaemia peaks, triggering distinct pathologies regardless of the infection route (Miles et al. Reference Miles, Lanham and Povoa1980; Souza et al. Reference Souza, Andrade, Barbosa, Santos, Alves and Andrade1996). For the oral route, several authors have demonstrated that the parasite interacts with the gastric mucosa through specific binding of glycoprotein receptors cruzipain, gp82 and gp85 on the parasite surface (Neira et al. Reference Neira, Silva, Cortez and Yoshida2003). The internalization of T. cruzi into host cells occurs during both routes of infection and requires specific molecules that promote cell invasion through Ca2+ mobilization. This ability of the parasite to invade and replicate in the gastric mucosa and within the macrophage cell matrix allows it to infect systemically (Hoft, Reference Hoft1996; Hoft et al. Reference Hoft, Farrar, Kratz-Owens and Shaffer1996).
Upon invasion by metacyclic parasites during both oral and vector inoculations, macrophages activate the innate immune response, thus promoting CD4+ and CD8+ T cell activation, which mediates the cellular immune response against T. cruzi (Tarleton et al. Reference Tarleton, Sun, Zhang and Postan1994; Reed, Reference Reed, Boothroyd and Komuniecki1995; Aliberti et al. Reference Aliberti, Machado and Souto1999). Consequently, acquired immunity of the infected host is initiated by the production of interleukins. For example, enhanced concentrations of induced nitric oxide (iNO) are achieved by the activation of arginase I expression within macrophages (Gazzinelli et al. Reference Gazzinelli, Oswald, Hieny, James and Sher1992; Aliberti et al. Reference Aliberti, Cardoso, Martins, Gazzineli, Vieira and Silva1996). However, exacerbated production of arginase I and NO during the acute phase of infection is linked to myocarditis in susceptible murine models (Garcia et al. Reference Garcia, Paula, Giovanetti, Zenha, Ramalho, Zucoloto, Silva and Cuha1999). Camargo et al. (Reference Camargo, Franco, Garcia, Dutra, Texeira, Chiari and Machado2000) found that, in Holtzman rats infected with strains ABC, Y and Cl-Brener, the production of iNO promoted severe acute myocarditis during the acute phase and significantly impacted the chronic phase of the disease.
Many studies have been conducted to explain the immunopathological profile of the host in vector-borne infection. However, scarce data are available concerning oral infection regardless of strain virulence. The goal of this work was to conduct a comparative analysis of the immune profiles of Wistar rats infected with the Y strain of T. cruzi by either intraperitoneal or oral infection, using parameters such as the evaluation of systemic parasitaemia, peritoneal macrophage numbers, production of IL-12, IFN-γ, IL-10 and nitric oxide, and CD4+ and CD8+ T cell populations by flow cytometry and histopathology.
Our hypothesis was that the evaluation of host immune profiles could be used to distinguish between intraperitoneal and oral infections.
MATERIALS AND METHODS
Experimental animals
Male (n = 30) Wistar rats weighing 100–120 g were used. The rats were placed in plastic cages in groups of 5 at constant temperatures (23±2 °C) and a 12/12 day/night schedule. Commercial rodent diet and water were available ad libitum. Cage bedding was changed 3 times/week. The experiments were performed in duplicate. The protocol of this study was approved by the local Ethics Committee protocol number 09.1.989.53.5.
Collection and preparation of infected triatomine feces
Five R. prolixus were fed with the blood of mice infected with the Y strain of T. cruzi. After a period of 20–30 days, the insects were collected and an abdominal compression was performed. The contents, a mixture of feces and urine, were sequentially centrifuged in a series of 100 and 1000 g in a suspension of RPMI medium to separate the parasites from the feces. The quantification of the metacyclic trypomastigotes was obtained by counting in a Neubauer chamber.
Experimental infection
The animals were divided into three groups: intraperitoneally infected (I.P.), orally infected (O.I.) (both with strain Y of T. cruzi) and uninfected controls (U.C.) (Silva and Nussenzweig, Reference Silva and Nussenzweig1953).
Wistar rats are normally resistant to most T. cruzi strains, thus we used relatively high inoculums. The I.P. group was infected intraperitoneally with 1×105 metacyclic trypomastigotes and the O.I. group received 8×105 metacyclic trypomastigotes by oral gavage in 0·2 mL of complete RPMI. The various trypomastigotes loads used for each route of infection were chosen after preliminary analysis in rats infected with strain Y of T. cruzi, for example, 1×105, 4×105, 6×105 and 8×105. We chose the parasite load which showed a peak parasitaemia well established over a period of 30 days. The experiments were performed on pre-determined days based on a parasitaemia curve of the strain and the route of infection: I.P. on day 14 and O.I. on day 21. For each day of the experiment, 5 animals were used per group because a difference in the sample number can influence experimental results (Sogayar et al. Reference Sogayar, Kipnis and Curi1993).
Parasitaemia
Blood trypomastigotes were counted at the peak of parasitaemia for both I.P. and O.I. groups by the method of Brener (Reference Brener1969).
Quantification of macrophage
Macrophages were obtained from the peritoneal cavities of I.P., O.I. and U.C. animals with 10 mL of RPMI 1640 (Cultlab, Campinas, Brazil). The control uninfected groups were injected intraperitoneally with 3% thioglycollate 4 days prior to the experiment. The cells were collected and centrifuged for 15 min, and the pellet was re-suspended in RPMI. For counting, the macrophages were stained with Turk solution, and the final cell count was performed in a Neubauer chamber.
Flow cytometry assay
The cells were released from spleen tissues by extrusion through a 70 μm nylon cell strainer and macerated in RPMI 1640 medium to produce a single cell suspension. A total of 2×106 cells from each organ of each experimental group were placed in 96-well round-bottom plates for cytofluorometric analysis. Following Fc receptor blocking, the cells were incubated with the monoclonal antibodies anti-CD3-phycoerythrin (PE), anti-CD4-fluorescein isothiocyanate (FITC) and anti-CD8-peridin chlorophyll protein (PERCP) as well as immunoglobulin isotype matched controls. The stained cells were stored for analysis in PBS containing 0·01 mL sodium azide and 1% paraformaldehyde in sealed tubes in the dark. All the steps were performed at 4 °C. Analysis was performed using a Becton Dickinson FACScan flow cytometer with DIVA-BD software (Becton Dickinson Immunocytometry Systems, San Jose, CA, USA).
Measurement of NO production
Nitric oxide assays were performed using cells obtained from the peritoneal lavage. The cells were centrifuged for 10 min at 1500 g at 4 °C, and the supernatant was discarded. The pellet was re-suspended in 1 mL of RPMI containing 10% fetal bovine serum and antibiotics. An aliquot of 10 μL was diluted in Turk and trypan blue for counting and cell viability. After the cells were resuspended to 5×106 cells mL−1, 100 μL were distributed into 96-well plates, and 10 μg mL−1 of LPS was added in 2 wells for standard control positive. The cells were incubated for 24 h at 37 °C and 5% CO2.
The plate was then centrifuged at 1500 rpm for 4 min. A volume of 100 μL was collected, and the supernatant was transferred to another 96-well plate. An equal volume of Griess reagent was added, enabling the measurement of the reaction at 540 nm. A 2-fold serial dilution standard curve from 200 to 6 μ m NO3 was performed (Terenzi et al. Reference Terenzi, Diaz-Guerra, Casado, Hortelano, Leoni and Boscá1995). The absorbance was read in a microplate reader.
Cytokine assay
The cytokine levels (IL-12, IFN-γ, IL-4 and IL-10) were detected using an immunoassay (Catalogue No. RT1000, KRCO122 and RIF00 R & D Systems) using serum collected from infected and control animals. The method is based on the basic components for a sandwich ELISA to measure natural and recombinant cytokine levels.
Histologic evaluation
An inflammatory infiltrate was considered present when 30 or more leucocytes were detected in each inflammatory focus. The hearts were collected following euthanasia for each experiment. The organs were washed in 0·9% saline solution, fixed in 10% formaldehyde and embedded in paraffin. Histological smears of 6 μm were stained with haematoxylin and eosin. The smears were mounted with a gap of 70 μm to avoid analysing the same inflammatory foci. For the inflammatory infiltrate score, the total number of foci was counted in 50 microscope fields (400× magnification) per cardiac section. Four sections were counted for each animal, and individual data were determined as the mean results of the 4 sections. The mean and standard deviation of 7 animals are presented for each group.
Statistical analysis
The statistical significance between groups was determined by analysis of variance one-way ANOVA Bonferroni's post-test. The results were expressed as the means/s.e.m. A value of P<0·05 was considered statistically significant. All the statistical analyses were performed using Graph-Pad Prism version 5.0.
RESULTS
Parasitaemia
As shown in Fig. 1, I.P. infection reached its peak of parasitaemia at 14 days while O.I. infection occurred at 21 days post-infection. On the respective peak of parasitaemia for both kinds of infection, a significant enhanced number of blood trypomastigotes was found in I.P. infection when compared with O.I. infection (P<0·05), which represents a rise 1·5 times higher than the values found for O.I. infection.
Fig. 1. A, Parasitaemia curve for each kind of infection (O.I.) and (I.P.). B, Number of parasites in the peak of parasitaemia for each kind of infection (O.I.) and (I.P.). Parasitaemia was evaluated in male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
Macrophages count
Both routes of infection displayed significantly enhanced numbers of macrophages as compared with group U.C. (P<0·01) (Fig. 2). When comparing both kinds of infection, during the respective peak of parasitaemia, a rise of 3·4 times in the number of macrophages was found in I.P. infection when compared with U.C., while a rise of 2·4 times higher was observed for O.I. as compared with the U.C. counterpart.
Fig. 2. Quantification of peritoneal macrophages from male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
On the respective peak of parasitaemia of each route of infection, a significant increase in the number of macrophages was found for the I.P. group as compared with the O.I. counterpart (P<0·05).
When comparing both kinds of infection, a rise of 1·4 times in the number of macrophages was found in I.P. infection when compared with O.I. during the respective peak of parasitaemia.
In vivo T cell profile
The evaluation of CD4+ and CD8+ T cells was performed with the cells released from spleen tissues during the peak of parasitaemia for both routes of infection.
Our results demonstrate that for CD4+ T cells, a non-significant increase in the number of cells was found for I.P. infection as compared with O.I. (P>0·05).
Differences were only significant when the intraperitoneal and oral infection groups were compared with the U.C. group (P<0·01) (Fig. 3). For CD4+ T cells a rise of 2·25 times higher for I.P. and 2·0 times higher for O.I. was found when compared with the U.C. group.
Fig. 3. A, Phenotypical analysis of T CD4+. B, CD8+ lymphocytes population in spleen of male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
For CD8+ T cells, a non-significant increase in the number of cells was observed for O.I. infection as compared with I.P. (P>0·05). But for the I.P. route of infection the CD8+ T cells displayed a reduced number (3·0 times higher) as compared with O.I. (3·33 times higher), when compared with the U.C. group.
Measurement of NO production
NO production in male Wistar rats was measured in macrophages obtained from the peritoneal wash at the respective parasitaemia peak for each route of infection.
Both routes of infection displayed significantly higher NO production compared with the U.C. group (for I.P. P<0·001 and for O.I. P<0·05). For the I.P. infection a rise of 5 times higher was found when compared with the U.C. group, while for the O.I. infection a rise of 3 times higher was observed when compared with the U.C. group (Fig. 4).
Fig. 4. Concentration of nitrite expressed as μ m was measured in cells harvested from the peritoneal cavity of male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
The I.P. group displayed a significantly higher concentration of NO when compared with the O.I. group (P<0·01). In this case, the O.I. group underwent a reduction of 40% in the concentration of NO when compared with the I.P. group in the respective peaks of parasitaemia.
Cytokine release
Pro-inflammatory cytokines
To access the concentrations of the interleukins studied (IL-12 and IFN-γ) serum collected from infected animals at the peak of parasitaemia for both routes of infection (I.P. and O.I.) was used.
Interleukin 12 (IL-12)
Both routes of infection displayed significantly enhanced concentrations of IL-12 when compared with the control group (P<0·001) (Fig. 5A). When comparing the concentrations of IL-12 with I.P. infection and O.I. infection a rise of 5·4 and 3·0 times higher, respectively, was observed when compared with the values obtained from U.C. group.
Fig. 5. A, Production of IL-12 and B, IFN-γ expressed in pg mL−1 of serum from male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
Observing the values obtained by both kinds of infection, a significant increase in the concentrations of IL-12 for the I.P. group (around 1·8 times higher) as compared with the O.I. group (P<0·01) was found. The range of values for IL-12 according to the immunological assay R&D system is 7·8 to 500 pg mL−1.
Interferon gamma (IFN-γ)
Both routes of infection displayed significantly enhanced concentrations of IFN-γ when compared with the control uninfected group (U.C.) (P<0·001 for I.P. and P<0·01 for O.I.) (Fig. 5B), which means that for I.P. and O.I. infections, a rise of 3·85 times higher and 2·6 respectively was found when compared with the values obtained from the U.C. group.
Observing the values obtained by both kinds of infection, a significant increase in the concentrations of IFN-γ for the I.P. group (around 1·5 times higher) as compared with the O.I. group (P<0·05) was found. The range of values for IFN-γ according to the immunological assay R&D system is 9.4–3000 pg mL−1.
Anti-inflammatory cytokines
Interleukin 10 (IL-10)
Both routes of infection displayed significantly enhanced concentrations of IL-10 when compared with the control uninfected group (U.C.) (P<0·01 for I.P. and O.I.) (Fig. 6), which means that for I.P. and O.I. infections, a rise of 10·0 times higher and 14·0 respectively was found when compared with the values obtained from the U.C. group.
Fig. 6. Production of IL-10 expressed in pg mL−1 serum from male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes of the Y strain on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
Observing the values obtained by both kinds of infection, no significant differences were observed between both routes of infection (O.I. vs I.P.) (P>0·05). The range of values for IL-10 according to the immunological assay R&D system is 7.8–500 pg mL−1.
Interleukin 4 (IL-4)
The infection with Y strain of T. cruzi displayed significantly enhanced rises to both routes on concentrations of IL-4 when compared with the U.C. group (P<0·05 for I.P. and P<0·01 for O.I.) (Fig. 7), which means that for I.P. and O.I. infections, a rise of 16·0 times higher and 18·0, respectively, was found when compared with the values obtained from the U.C. group. The range of values for IL-4 according to the immunological assay R&D system is 62.5–4000 pg mL−1.
Fig. 7. Production of IL-4 expressed in pg mL−1 serum from male Wistar rats intraperitoneally infected (I.P.) with 1×105 metacyclic trypomastigotes of the Y strain on day 14 and orally infected (O.I.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. For all groups n=5 and a value of *P<0·05 was considered statistically significant.
Inflammatory foci in the heart
The histopathological analysis of the heart smears showed slight impairment of the heart fibres architecture in both kinds of infection when compared with the U.C. group (Fig. 8A–C). An increased number of inflammatory foci in the I.P. group as compared with the O.I. group (Fig. 8B vs C) was also observed.
Fig. 8. Histopathological aspects from male Wistar rats intraperitoneally infected (I.P.) on day 14 with 1×105 metacyclic trypomastigotes and orally infected (I.O.) with 8×105 metacyclic trypomastigotes of the Y strain of T. cruzi on day 21. H&E, 400×. A, Control uninfected (U.C.) without parasite and focal inflammatory. B, Intraperitoneally infected (I.P.) section of the heart with moderate inflammatory infiltrate with several amastigote nests. C, Orally infected (O.I.) Mild focus inflammatory with cardiac cellular architecture unaltered. D, Each bar represents the mean±SD of an experimental group composed of 5 rats. (I.P.) – Y vs. (O.I.) – Y, on specific day of experiment and value of *P<0·05 was considered statistically significant.
The evaluation of inflammatory foci revealed significant increased numbers for I.P. (1·6 times higher) when compared with the values found for O.I. (P<0·01).
DISCUSSION
Many studies have indicated vector transmission as a primary route of infection for numerous pathogens, such as malaria, leishmaniasis and American trypanosomiasis. Epidemiological, molecular and immunological studies concerning this route of infection have contributed to a better understanding of Chagas disease (Coura et al. Reference Coura, Junqueira, Fernandes, Valente and Miles2002).
There are few pathogenic microorganisms transmitted by oral infection. Among these pathogens are bacteria, such as Helicobacter pylori, Mycobacterium tuberculosis, Treponema pallidum and Histoplasma capsulatum, and viruses, such as cytomegalovirus, which causes gastric granulomas in patients with poor nutrition, alcoholism, HIV and other conditions triggered by immune suppression (Wang and Peura, Reference Wang and Peura2011). Several enteric nematodes can be transmitted orally, such as Anisakis simplex, Pseudoterranova decipiens and Strongyloides stercoralis (Ozturk et al. Reference Ozturk, Aydinli, Celebi and Gursan2011).
In recent years, there has been a significant increase in oral infections by T. cruzi. Often, these infections are related to a lack of sanitary inspections of foods that are consumed by the human population. In 2007 in the north and northeast, approximately 88 cases occurred due to contamination of sugarcane and açaí juice (Dias, Reference Dias, Brener, Andrade and Barral Netto1999; Carlier et al. Reference Carlier, Dias, Luquetti, Hontebeyrie, Torrico and Truyens2002).
To better understand differences in the immune response according to the type of infection, the first step was to evaluate the parasitaemia.
During T. cruzi infection, the level of parasites in an infected host is directly related to the virulence of the strain, parasite load, host immunity and the type of infection. For the vector-borne route, the parasite is directly engulfed by cells of the immune system, while for the oral route, the gastric mucosal surface is the first barrier as well as an attractive point of entry for different pathogens, such as HIV, tuberculosis, helminths and protozoans (Moncada et al. Reference Moncada, Keeller and Chadee2003; Neutra and Kozlowski, Reference Neutra and Kozlowski2006; Mathers et al. Reference Mathers, Ezzati and Lopez2007). In addition to vector transmission, T. cruzi is primarily maintained in nature by oral infections between several animal species, through predatory mechanisms of survival, keeping the parasite circulating within hosts. Collins et al. (Reference Collins, Craft, Bustamante and Tarleton2011) reported that an oral infection could elicit a mucosal immune response when the parasite invades the host, the magnitude of this immune response being the main factor which determines the fate of the parasite and the level of systemic parasitaemia (Round and Mazmanian, Reference Round and Mazmanian2009).
Different authors have suggested several possibilities as to how the immune system recognizes T. cruzi and the initial behaviour of the parasite upon crossing the gastric mucosa. This event is coordinated by the participation of glycoproteins gp82 and gp30 as well as others that are expressed on the surface of trypomastigotes of almost all T. cruzi strains. These glycoproteins are directly related to the invasion of phagocytes that reside near the gastric mucosa (Neira et al. Reference Neira, Silva, Cortez and Yoshida2003; Yoshida, Reference Yoshida2006; Collins et al. Reference Collins, Craft, Bustamante and Tarleton2011).
Our data revealed that the peak of parasitaemia occurred on different days. For the I.P. route, at 14 days post infection, a statistically significant increase in the number of blood trypomastigotes was found as compared with O.I. at 21 days post infection (Fig. 1).
Some hypotheses may explain the immunological differences between oral and vector-borne infections. The works of Goodrich and McGee (Reference Goodrich and McGee1998) and Yamamoto et al. (Reference Yamamoto, Vancott, Okahashi, Marinaro, Kiyono, Fujihashi, Jackson, Chatfield, Bluethmann and McGhee1996) suggest that the first contact between the mucosa and the parasite promotes a Th-2 type immune response and at the same time, IgA production, which is responsible for protection against the parasite. Other works have reported that a deficiency of gp82 in metacyclic trypomastigotes reduces the degree of infectivity in the gastric mucosa and reduces the ability of host cells to internalize the parasite (Cortez et al. Reference Cortez, Neira, Ferreira, Luquetti, Rassi, Atayde and Yoshida2003; Yoshida, Reference Yoshida2009).
According to Eickhoff et al. (Reference Eickhoff, Giddingsa, Yoshida and Hoft2010), gp82 is crucial for cell invasion, and it is a major target of the immune system. According to Yoshida (Reference Yoshida2006), this protein is preserved in almost all strains of T. cruzi, including Y, CL, F and Tulahuen.
Since the stomach represents the main barrier for T. cruzi insect-derived metacyclic trypomastigotes to invade the gastric mucosal epithelium and to reach the target cells, we can hypothesize, based on the concepts described by Yoshida (Reference Yoshida2008), that since the metacyclic forms are equipped with mucin-like surface molecules, the lower parasitaemia during oral inoculation could be a consequence of reduced gp82 expression and increased local production of antibodies. These could temporarily impair T. cruzi invasion and replication as amastigote forms when compared with intraperitoneal infection in which parasites are directly engulfed by macrophages, easing the replication and dissemination throughout the organism.
Camandaroba et al. (Reference Camandaroba, Pinheiro Lima and Andrade2002), using murine models of oral and intraperitoneal infections with the Peruvian and Colombian strains of T. cruzi, described significant differences in parasitaemia levels and pathological alterations according to the kind of the route of infections and strains. Our results concerning different parasitaemia levels in both routes of infection can indirectly corroborate the findings cited above (Fig. 1).
After the establishment of a T. cruzi infection, several immune cells are activated through specific host glycoproteins, and macrophages, T cells and cytokines work together to control parasite replication during infection (Taliaferro and Pizzi, Reference Taliaferro and Pizzi1955; Costa et al. Reference Costa, Torres, Mendonça, Gresser, Gollob and Abrahamsohn2006; Padilla et al. Reference Padilla, Xu, Martin and Tarleton2007).
Innate immunity is the first line of defence against invading microorganisms. Cells from the monocytic lineage are directly involved in parasite control during the early stages of acute infection with T. cruzi in vivo and in vitro (Brunet, Reference Brunet2001; Reyes et al. Reference Reyes, Terrazas, Espinoza, Cruz-Robles, Soto and Rivera-Montoya2006; Kuehn et al. Reference Kuehn, Oliveira, Santos, Augusto, Toldo and Prado2011). Classically activated macrophages are involved in the signalling of naive CD4+ and CD8+ T cells, which trigger inflammatory molecules, such as IFN-γ, IL-12, TNF-α and IL-10 and increase the production of inducible NO synthase (iNOS) to produce NO (Camussi et al. Reference Camussi, Albano, Tetta and Bussolino1991; Brazão et al. Reference Brazão, Filipin, Santello, Caetano, Abrahão, Toldo and Prado2011).
The complex interactions that occur between cells during T. cruzi infection play a critical role in the course of the disease. Macrophages also present antigens via major histocompatibility complex (MHC) class I and class II on the cell surface and may be involved both through their phagocytic and secretory activities as well as the production of reactive nitrogen and oxygen intermediates, which endow these cells with potent microbicidal activity (MacMicking et al. Reference MacMicking, Xie and Nathan1997).
Despite the fact that macrophages display microbicidal and immune-activating properties, several protozoan parasites successfully invade, survive and replicate within these cells. Our findings revealed that both infections promoted the stimulation of mononuclear cells in the attempt to control parasitaemia when compared with the U.C. group. The I.P. infection triggered a significant increase in the number of peritoneal macrophages compared with O.I. infection. This difference is probably due to the distinct mechanisms of invasion and recognition of glycoproteins during the oral infection, which triggered an unsuccessful Th-1 immune response.
Zhang and Tarleton (Reference Zhang and Tarleton1999) found that the host immune response activates distinct T lymphocyte subpopulations during acute infection. According to Perez et al. (Reference Perez, Morrot, Berbert, Terra-Granado and Savino2012), T cells expressing both CD4+ and CD8+ co-receptors have been described in healthy individuals as well as in patients suffering pathological conditions, including infectious and autoimmune diseases and chronic inflammatory disorders. Most CD4+ and CD8+ T cells undergo differentiation to mature T cells in the thymus. However, during T. cruzi infection, the thymus is severely affected by a deregulated cascade of pro-inflammatory cytokines and hypothalamus, pituitary and adrenal-related hormones (Correa-de-Santana et al. Reference Correa-de-Santana, Paez-Pereda, Theodoropoulou, Kenji Nihei, Gruebler, Bozza, Arzt, Villa-Verde, Renner, Stalla, Stalla and Savino2006; Perez et al. Reference Perez, Roggero, Nicora, Palazzi, Besedovsky, Del Rey and Bottasso2007). Therefore, an excessive activation of the immune system may contribute to the pathological immune dysfunction in Chagas’ disease (Fiuza et al. Reference Fiuza, Fujiwara, Gomes, Rocha, Chaves and de Araújo2009).
The involvement of T cell populations in oral infection may induce distinct parasitological and/or immunological outcomes compared with other routes of infection. The stomach mucosal tissue plays the role of an interface between the host's body and the parasite, constituting a wide surface area constantly in contact with potential pathogens. So, the immune response associated with GI mucosa has evolved the ability to determine when an aggressive response is appropriate, through balancing activation and regulation (Round and Mazmanian, Reference Round and Mazmanian2009).
In our experiment, any significant alterations were observed between the levels of CD4+ T cells for I.P. vs O.I. as well as for CD8+ T cells for I.P vs O.I. infection (Fig. 4). Our data partially corroborate the findings of Collins et al. (Reference Collins, Craft, Bustamante and Tarleton2011) in which the authors describe that the CD8+ T cell response to T. cruzi infection develops similarly, regardless of inYfection route. T. cruzi-specific CD8+ T cells were found in all assayed lymphoid and non-lymphoid (including gut) tissues following either subcutaneous or oral infection.
Hoft and Eickhoff (Reference Hoft and Eickhoff2005) demonstrated that recombinant key cytokines such as IL-12 and anti-IL-4 induced type 1-biased responses in vivo that were highly protective against normally lethal T. cruzi systemic oral challenges. Thus, increased concentrations of Th-1 cytokines, such as IL-12, IFN-γ and NO, are of vital importance to control parasite replication during the acute phase. Consequently, NK cells are activated to produce IFN-γ, which in turn stimulates macrophages to produce NO with antiviral, antimicrobial and antiparasitic activity (Clark and Rockett, Reference Clark and Rockett1996; Pinsky et al. Reference Pinsky, Walif, Szabolcs, Athan, Liu, Yang, Kline, Olson and Cannon1999; Oliveira et al. Reference Oliveira, Santiago and Lisboa2000).
In our experiment, two cytokines of the TH-1 immune response were analysed. For both IL-12 (Fig. 5A) and IFN-γ (Fig. 5B), a statistically enhanced concentrations of these cytokines were observed for I.P. infection when compared with O.I. The observation that IFN-γ was enhanced in I.P. infection can be explained by the premise that mucosal and systemic protection requires different immune responses. When T cells produce interleukin-4 (IL-4), IL-5 and IL-10 (type 2 phenotype) or high levels of transforming growth factor β (TGF-β; type 3 phenotype), immunoglobulin A secretion is induced, which protects the mucosa against infection (Yamamoto et al. Reference Yamamoto, Vancott, Okahashi, Marinaro, Kiyono, Fujihashi, Jackson, Chatfield, Bluethmann and McGhee1996; Goodrich and McGee 1998). Contrarily, T cells producing IFN-γ, TNF-α and IL-2 (type 1 phenotype) play a protective role against systemic intracellular replication of many human pathogens (O'Garra, Reference O'Garra1998).
Our data concerning the profile of TH-1 cytokines displayed enhanced levels for IL-12 and IFN-γ, independently of the route of infection, although the highest concentrations of both cytokines were seen in I.P. infection.
Since it is well documented that type 1 immune responses are critical for protection against both mucosal and systemic T. cruzi infection (Hoft et al. Reference Hoft, Schnapp, Eickhoff and Roodman2000; Hoft and Eickhoff, Reference Hoft and Eickhoff2005) this hypothesis can be considered the most logical explanation for our data.
Concerning the profile of TH-2 cytokines, our for O.I. infection showed enhanced levels of IL-10 when compared with I.P. (Fig. 6). According to Mestecky et al. (Reference Mestecky, Russell and Elson2007) and Elson (Reference Elson2007) antigen (Ag) encounter at the GI mucosa or in gut-associated lymphoid tissue often results in tolerance, particularly for T cell responses, a process normally mediated by the TH-2 cytokines such as transforming growth factor β (TGF-β) and interleukin-10 (IL-10) and a possible involvement of IL-4 in the oral infection.
After T. cruzi infection, the progression of the disease is closely controlled by cytokines involved in macrophage activation. The importance of IFN-γ in mediating resistance during the acute phase of infection was demonstrated by several authors who reported an exacerbation of infections by in vivo treatment with anti-IFN-γ. Thus the production of cytokines that are able of antagonizing macrophage activators, such as IL-10 and Il-4, are of particular interest in T. cruzi infection because they have the ability to interfere with both the production and use of IFN-γ. According to Reed (Reference Reed, Boothroyd and Komuniecki1995), an essential role for IL-10 in mediating susceptibility to acute T. cruzi infection was demonstrated by in vivo treatment with neutralizing anti-IL-10 which triggered reduced blood parasitaemias and prevented death in a strain of susceptible mice. In our experiment, we observed enhanced levels of IL-10 for O.I. as compared with I.P. and a rise, although not statistically significant, of IL-4 for O.I. Based on the discussion above, our results indicate that these cytokines are associated with the control of exacerbation of the disease. Through our data we can say that IL-10 and IL-4 secretion also may be fundamental in determining the pattern of resistance or susceptibility in both kinds of infection, O.I. or I.P.
The histopathological analysis of the heart tissue revealed a distinct pattern of parasite colonization, in which the I.P. infection displayed a moderate inflammatory infiltrate with several amastigote nests. For the O.I. infection, rare amastigote nests and inflammatory cells were observed, although the cardiac cellular architecture was unaltered (Fig. 8). The precise mechanisms promoted by the immune system to control the dissemination of orally inoculated trypomastigotes are not well defined, although we believe that the trypomastigote load was much heavier in the I.P. infection despite a more potent type 1 immune response. Since it is well established in the literature that the TH-2 immune response reduces the inflammatory process in the cardiac tissue (Hoft et al. Reference Hoft, Farrar, Kratz-Owens and Shaffer1996, Reference Hoft, Schnapp, Eickhoff and Roodman2000; Schnapp et al. Reference Schnapp, Eickhoff, Sizemore, Curtiss and Hoft2000), these data indirectly corroborate ours since the enhanced levels of IL-4 found in O.I. infection induced a reduced number of inflammatory foci among heart fibres as compared with I.P. infection. Importantly, both types of infection had the ability to promote cardiac damage.
ACKNOWLEDGEMENTS
We are grateful to Professor Dr João Aristeu da Rosa for donating the triatomines Rhodnius prolixus.
FINANCIAL SUPPORT
This study was supported by fellowships from CNPq.
