How Do Mouse Strains and Inoculation Routes Influence the Course of Experimental Trypanosoma cruzi Infection?

Abstract

Chagas’ disease outcomes depend on several factors including parasite and host genetics, immune response, and route of infection. In this study, we investigate the influence of inoculation route and host genetic background on the establishment and development of Chagas disease in mice, using an isolate of Trypanosoma cruzi SC2005 strain (TcII), which was obtained from an oral Chagas’ disease outbreak in Santa Catarina, Brazil. Comparative analysis of the immunopathological, histopathological, and hematological profiles of mice was performed demonstrating the influence of the route of infection in disease severity. In outbred mice, intraperitoneal (IP) infection led to higher infection and mortality rates and more severe parasitaemia, when compared with intragastric (IG) infection. Nevertheless, tissue colonization was similar, showing severe damage in the heart, with intense lymphocytic inflammatory infiltrates, regardless of the route of infection. On the other hand, in mice IG-infected, the host genetic background influences the start timing of immune response against Trypanosoma cruzi. The susceptible BALB/c inbred mouse strain presented an earlier development of a cytotoxic cellular profile, when compared with A mice. We hypothesize that the cytotoxic response mounted before the parasitaemia increase allowed for a milder manifestation of Chagas’ disease in intragastrically infected mice.

1. Introduction

Chagas’ disease is a public health problem that affects about 6 to 7 million people worldwide, mainly who reside in the endemic areas of 21 countries of Latin America [1]. Despite the progress made to control the infection, the disease has been expanding to North America, Europe, Australia, and Japan by migration of million people from endemic countries [2, 3, 4]. The etiological agent of disease is the protozoan Trypanosoma cruzi, which shows a wide genetic variation, being classified in seven Discrete Typing Units (DTUs), TcI–TcVI, and Tcbat [5, 6, 7]. These different DTUs are responsible for different disease outcomes, demonstrating the influence of parasite genetics on disease development [6]. Chagas’ disease is classically transmitted by its insect vector, the triatomine bugs [8], but can also be acquired by blood transfusion [9, 10], congenitally [11, 12], by organ transplantation [13, 14], laboratory accidents [15], sexually [16], and by ingestion of contaminated food/juice [17]. After successful actions to interrupt vector transmission, oral contamination has been considered the main route of transmission in several countries [17, 18, 19]. Brazil is the country with the highest incidence of oral acute Chagas’ disease outbreaks, mainly in the Amazon Basin ([19] reviewed by [20]). Oral acute Chagas’ disease displays a higher number of signs and symptoms and higher lethality than those acquired through the vectorial route [17, 21].

In addition to parasite genetics and route of infection, disease outcome is also influenced by other factors such as evolutive forms of the parasite, parasite load, mixed infections [22, 23, 24, 25], and host factors including immune response, concomitant infections, nutrition deficiencies, and host genetics [26]. Although several studies have tried to identify the genetic basis of Chagas disease, our knowledge on the subject is still vague. Many works have studied genetic polymorphisms located in genes associated with immune response [27] or over the whole genome [28, 29], but the influence of these polymorphisms on the pathology of the disease still needs to be further validated.

Considering the complexity of these factors and their influence in Chagas’ disease immunopathogenesis, anti-Trypanosoma cruzi immune response and chemotherapy, there is a need to improve our understanding about these relationships. On this way, this study investigates the influence of the inoculation route and the host genetic background on the establishment and development of Chagas’ disease in inbred and outbred mice, using an isolate of T. cruzi SC2005 strain (TcII) obtained from a human case from an oral Chagas’ disease outbreak in the south region of Brazil [30]. The aim was to conduct a comparative analysis of the immunopathological, histopathological and hematological profiles, using parameters such as parasite load, survival rates, cytokines production, histopathology, and cell populations in the inflammatory sites.

2. Materials and methods

2.1 Ethics statement

All experiments were conducted following the guidelines for experimental procedures of the National Council for the Control of Animal Experimentation (CONCEA) after approval by the Ethics Committee for Animal Research of the Fundação Oswaldo Cruz (CEUA-FIOCRUZ) under licenses number LW16/11 (Swiss Webster mice) and LW 42/14 (A and BALB/c mice).

2.2 Animals

BALB/c, A and Swiss Webster female mice, 4–6 weeks old, were provided by Instituto de Ciência e Tecnologia em Biomodelos (ICTB - FIOCRUZ) and housed under pathogen-free conditions, controlled temperature, and food and water ad libitum.

2.3 Parasites

Trypanosoma cruzi SC2005 (DTU Tc II), isolated from a case of oral acute Chagas’ disease during an outbreak in Santa Catarina, Brazil [30, 31], was used in this study.

In experiments with outbred Swiss mice, epimastigote forms of T. cruzi SC2005 isolate were maintained in LIT medium for 30 days. Metacyclic trypomastigotes forms were quantified in a Neubauer chamber and used to infect VERO cells. The infected culture was maintained in RPMI medium, supplemented with 10% fetal bovine serum. Trypomastigotes derived from cell culture (TCC) were obtained 10 days after, by recovering parasites from the culture supernatant, and quantified in a Neubauer chamber prior to infecting mouse groups.

In experiments with inbred mice (A and BALB/c), epimastigote forms were maintained at 28°C for 21 days in LIT (Liver Infusion, Triptose) (AGM) culture medium to obtained metacyclic forms that were quantified in a Neubauer hemocytometer prior to infection.

2.4 Experimental design

In order to investigate the influence of the route of infection on the course of T. cruzi SC2005 infection, Swiss mice were organized into three groups: group 1 (n = 30) mice intraperitoneally (IP) infected by 10⁷ TCC forms of T. cruzi SC2005 strain/0.2 mL of RPMI medium; group 2 (n = 55) mice intragastrically infected by 10⁷ TCC forms of T. cruzi SC2005 strain/0.1 mL of RPMI medium using a gavage needle (subjected to 4 h of fasting prior to infection), and group 3 (n = 15) normal uninfected animals (control group). Mice from each group (n = 3) were euthanized at 11 and 18 (IP-infected mice) and 26 and 33 (IG-infected mice) days after infection, and the esophagus, stomach, intestines, heart, thymus, liver, spleen, pancreas, kidney, adrenal gland, bladder, uterus, mesenteric lymph nodes, and brain were removed.

In experiments to investigate the host genetics influence on the course of T. cruzi SC2005 infection, A and BALB/c mice were intragastrically infected by 10⁷ metacyclic trypomastigote forms of T. cruzi SC2005 strain/0.3 mL of LIT medium using a gavage needle. Prior to intragastric injection, the animals were submitted to 4 h of fasting. The animals were divided into four experimental groups: Group 1 (n = 70): A-infected mice; Group 2 (n = 70) BALB/c-infected, and Group 3 (n = 50) and Group 4 (n = 50)—each one composed by uninfected A and BALB/c mice (control groups), respectively. Mice from each group (n = 6) were euthanized at 7-, 14-, 21-, and 40-day post-infection (dpi), and the blood, esophagus, stomach, gut, heart, and liver were removed.

2.5 Parasitemia, mortality and leukometry

Ten mice of each infected group were monitored daily from 5 to 50 dpi. Parasitaemia was determined as described by Pizzi and Prager [32]. Briefly, 5 μL of blood’s tail vein was collected and placed under a cover slip (22 x 22 mm) and the number of parasites/mL of blood was estimated by counting 50 microscopic fields in a 400X magnification. Mice that did not develop parasitaemia were considered non-infected and excluded from the experiment.

Mortality rate was estimated to obtain the survivors percentage. Mean time of death was calculated following Liddell [33].

At the same time of parasitaemia evaluation, another 10 μL of blood’s tail vein was collected and diluted in Turk’s solution (1/20) [34] for white cells counting using a Neubauer chamber. Differential cell count was made in smears, after MayGrünwald–Giemsa staining, by counting 100 leukocytes/slide.

2.6 Spleen index

Spleen index was calculated to investigate reticuloendothelial stimulation. This index was calculated after evaluation of the relative spleen weight (spleen weight/mouse weight) [35].

2.7 Parasite load

Fragments of heart recovered from infected mice were immediately frozen after euthanasia and stored at −70°C. Fragments were digested in 500 μL of lysis buffer (50 mM Tris, 10 mM NaCL, 5 mM EDTA, 0.5% SDS) containing proteinase K (20 mg/ml). DNA was extracted following a standard phenol/chloroform protocol [36]. DNA concentrations and purity were determined by reading A260 and A280 on a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA).

Parasite DNA was detected by SYBR Green qPCR assay using StepOnePlus Real-Time PCR System (Applied Biosystems). Primers targeting T. cruzi genomic DNA sequence (166 bp) Cruzi 1 (5’-ASTCGGCTGATCGTTTTCGA-3′) and Cruzi 2 (5’-AATTCCTCCAAGCAGCGGATA-3′) and Actb-actin, beta (b-actin) mouse gene (138 bp) (Forward 5’-AGAGGGAAATCGTGCGTGAC-3′; reverse 5’CAATAGTGATGACCTGGCCGT3’) were used following previous reports [37, 38]. To monitor DNA integrity, variation in DNA yield, or the presence of potential inhibitors of PCR, Actb reference gene was used as a positive control. The reaction mixtures contained Power SYBR Green PCR Master Mix 2X (Applied Biosystems), 25 ng of DNA template, and 100 nM of b-actin or 300 nM of Cruzi1/Cruzi2 primers in a final volume of 20 μL. PCR conditions were as follows: hold at 95°C for 10 min, 95°C for 15 s, and 58°C for 1 min (40x). Standard curves from axenic epimastigotes T. cruzi DNA (100 ng–1 pg) were generated. A melt curve analysis was performed on all reactions. The results were analyzed with the StepOne software v2.2.2 (Applied Biosystems).

2.8 Histopathology

Following euthanasia, all removed organs were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS), pH 7.45, at 4°C, for 72 h, cleaved and routinely processed paraffin embedded. Tissue sections (5 μm) were stained with Hematoxylin–Eosin (HE) (Sigma-Aldrich, Saint Louis, USA), Lennert’s Giemsa, or Picro-Sirius red (Direct Red 80, Aldrich Milwaukee, WI 53233, USA) techniques. The presence of inflammatory infiltrates was classified as (−) without infiltrates, (+) very mild lesion areas, (++) mild lesion areas, (+++) moderate areas of infiltrates, (+++) severe areas of infiltrates, (++++) very severe areas of infiltrates, following described by Barreto-de-Albuquerque et al. [39], and arbitrary values from 0 to 5 were attributed to it. Tissues were analyzed and photographed under light microscopy (Zeiss, Axioplan 2, with Axiovision LE64 photomicrograph equipment).

2.9 Hemogram

The blood was collected by cardiac puncture, placed in EDTA tubes, and sent to the Laborlife clinical laboratory (RJ, Brazil). Complete blood count (CBC) was analyzed in an automatic cell counter (Beckman Coulter, Brea, CA). The blood parameters evaluated were red blood cell (RBC), hemoglobin (Hgb), hematocrit (Hct), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), white blood cell (WBC), and platelet (PLT).

2.10 Cytokine analysis

Detection of TNF-𝛂, IFN-𝛄, IL-10, IL-6, IL-17A, IL-4, and IL-2 was performed through BD Cytometric Bead Array (CBA) Mouse Th1/Th2/Th17 Cytokine Kit (cat. 560,485; BD Biosciences, San Jose, CA) following manufacturer’s instructions. Samples were acquired in the BD Q13 FACSCallibur flow cytometer (BD Biosciences) and data were analyzed with FCAP Array software (BD Biosciences).

2.11 Obtaining of mononuclear cells to flow cytometry

In the determined euthanasia points, the blood of each animal was collected by cardiac puncture using citrated saline (0.87% NaCl, 3.8% Na₃C₆H₅O₇) and diluted in the same volume of complete RPMI medium—RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO), supplemented with 10% fetal bovine serum (FBS) (CultiLab, Campinas, Brazil); 200 mM L-glutamine; 100 U/mL penicillin; and 10 mg/mL streptomycin (Sigma-Aldrich, St. Louis, MO). Cells were then added to a Ficoll–Hypaque (Histopaque 1077; Sigma-Aldrich) sedimentation gradient. After centrifugation at 1030 x g for 20 min at 21 C, without brake, the mononuclear cell (MCs) ring was collected. A lobe of the liver, mesenteric lymph nodes, and spleen were macerated in 4 mL of complete RPMI medium until complete disruption using a glass tissue homogenizer (Corning E.U.A.), being kept on ice throughout the process. Cardiac cells were obtained after organ perfusion with PBS (pH 7.2), subsequently cut into small fragments, and subjected to four cycles of dissociation using 0.2% type II collagenase in RPMI medium without FBS at 37°C, stirring for 30 min. The supernatant obtained after each dissociation cycle was collected in a single tube and kept on ice. Cells from blood, liver, spleen, mesenteric lymph nodes, and heart were washed twice (with centrifugation at 720 x g, 4°C, 5 min) in PBS-BSA-(Bovine Serum Albumin) with 10% FBS and resuspended with 3 mL of ACK (Ammonium-Chloride-Potassium) Lysing Buffer, to lyse of red blood cells, incubated for 5 min at room temperature and washed again. After that, cells were incubated in PBS-BSA with 10% horse serum (HS) for 30 min, washed again, resuspended with complete RPMI medium, and adjusted to 1x10⁶cells/well.

2.12 Flow cytometry

Cells were incubated for 20 min in the dark at room temperature with the following monoclonal-antibody panel: anti-CD3-PC7 (Cat 553,064; BD Biosciences—maximum emission (max-em): 785 nm) diluted 1:40; anti-CD4-APC-H7 (Cat 560,181; BD Biosciences—max-em: 785 nm) diluted 1:80; anti-CD8-BB515 (Cat. 564,422; BD Biosciences—max-em: 515 nm) diluted 1:160, and anti-CD19-APC (Cat MCA 1439; Serotec—max-em: 661 nm) diluted 1:20, in PBS-BSA-HS. Cells were washed with PBS to remove unbound antibodies, fixed in 1% paraformaldehyde for 30 min in the dark at 4°C, washed again, and stored at 4°C in the dark until acquisition in flow cytometry. At least 20,000 events from each sample were acquired through CytoFlex flow cytometer (Beckman Coulter). Single-stained controls were used to set compensation parameters, while unstained cells were used to set analysis regions. After acquisition, flow cytometric analysis to evaluate the frequencies of CD8⁺T, CD4⁺T, CD4⁺/CD8⁺ T, and CD19⁺B cells was performed using CytoExpert Software (Beckman Coulter). A gate strategy was performed as follows: to exclude cell aggregates from analyses, cells were gated on Singlets region in FSC-A vs. FSC-H dot-plot; from Singlets gate an FSC-A vs. Side-Scatter-Area (SSCA), dot plot was created and the analyses region (mononuclear cells) was defined to encompass mononuclear cells and exclude dead cells from analyses; from Mono gate, CD4⁺ and CD8⁺ T lymphocytes were determined by CD3 vs. CD4 and CD3 vs. CD8 dot plots, respectively, and CD19⁺B cells by CD3 vs. CD19 dot plot. CD4⁺/CD8⁺ double-positive T cells was determined by plotting CD8 vs. CD4 gated on CD3⁺.

2.13 Statistical analysis

Data are expressed by mean ± standard error of the mean (SEM) and analyzed statistically by two- or one-way ANOVA and Sidak’s multiple comparison post-test using the GraphPad Prism 6 software. Differences were considered significant when p < 0.05.

3. Results and discussion

3.1 Route of infection and host genetic background influence on the parasitaemia and mortality after T. cruzi SC2005 infection

Several studies have been showing that the course and severity of Chagas’ disease depends on varied factors, among them are the route of infection and host genetic background [25].

The most frequent route of Chagas’ transmission is currently by ingestion of contaminated food and beverages, especially in Brazil and some other endemic countries [19, 40, 41, 42]. Infection by oral route usually leads to a more severe acute disease than vector-borne infection. The amount of metacyclic trypomastigotes contained in a triatomine crushed in food or beverage may be more than 100 times higher than what is typically found on its feces [43]. Furthermore, the digestive mucosa represents an extensive gateway, containing different molecules that serve as attachment points to the parasite [44]. Therefore, orally infected patients tend to experience a more severe acute phase, with more rapid progression to long-term cardiac or gastrointestinal dysfunction and higher mortality [43]. In the experimental model, intraperitoneal inoculation is the most common route of infection, but it does not mimic any natural infection. It delivers elevated loads of trypomastigotes directly in the peritoneum, bypassing all natural barriers in the skin and mucosa. When mucosa and systemic T. cruzi infection were compared, distinct disease patterns can be observed. Several studies in mice showed that intraperitoneal infection induces higher parasitemia and mortality than intragastric (IG) or oral infection with the same inoculum [39, 45, 46]. Marsden [47] showed that systemic infections (intraperitoneal or intravenous) promote higher infection rates (67–100%) and mortality than mucosal (oral, intragastric, intrarectal, genitalia, or conjunctival infection) (17–67%) in mice infected by the Peruvian strain. In the present study, similar results were obtained when outbred Swiss mice were infected IG or IP with TCC forms of SC2005 T. cruzi strain. IP infection was able to infect 100% of animals, while just 36% of IG-infected mice developed parasitaemia. Moreover, IP-infected mice showed earlier (10th and 13th days) and higher parasitaemia peaks (2.9 and 4.3 X 10⁶ parasites/mL, respectively) than those observed in the IG-infected animals, which presented peaks on the 13th and 18th dpi with 0.9 and 1.7 X 10⁶ parasites/mL, respectively (Figure 1A). Furthermore, IP-infected animals died earlier (mean time of death of 16.13 ± 0.8 days), and its mortality rate was 80%. On the other hand, IG-infected mice died later (22.67 ± 2.0), and the mortality rate was around 30% (Figure 1B). Different infection routes submit parasites to different barriers in order to infect the host. Crossing of these barriers may explain the differences observed in course and intensity of the parasitaemia. Nevertheless, independent of the route of infection, parasites were able to make their way to the heart, which showed amastigote nests after infection by both routes (Figure 1C and D).

Figure 1.

Influence of the route of infection on parasitaemia, mortality, and heart histopathology. Swiss Webster outbred mice were infected either intraperitoneally (IP) or intragastrically (IG) with 10⁷ TCC of Trypanosoma cruzi SC2005 strain. Parasitaemia (A) shows an early and strong increase of parasites in the blood in IP-infected mice, which leads to an earlier time to death (B). IG-infected mice present a weak rate of infection (B) and a later and lighter parasitaemia (A). Nevertheless, both routes led to a colonization of the heart myocardium, which can be seen after 18 days on IP-infected mice (C) and 26 days on IG-infected animals (D). Hematoxylin and eosin. Arrows show parasite nests. Two-way ANOVA followed by Sidak’s multiple comparison test. *** = p < 0.001.

The success of T. cruzi infection and the level of parasitaemia after oral contamination are mainly determined by the magnitude of the mucosal immune response developed [48], which depends on the interaction between both parasite and host genetics [49, 50]. Studies of genetic susceptibility to Chagas’ disease are scarce and its contribution to disease pathology is still unsolved. Inbred mice from different genetic backgrounds have been used to assess the influence of host genetics to several pathogens. Studies using inbred mouse strains infected by T. cruzi showed different profiles of response to infection and degrees of susceptibility in the hosts, with differences in mortality rate, cytokine production, inflammatory infiltrate, and parasite load [51, 52, 53]. C57BL/10 mice infected with T. cruzi SC2005 strain showed lower parasitaemia and mortality rate, while CBA-infected mice showed high parasitaemia and mortality rate [52]. Other studies using different inbred mouse strains (A/J, BALB/c, C3H/HePas, C57BL/6, and DBA mice) infected with T. cruzi Y also showed differences in mortality rate and parasitaemia. C57BL/6 mice were less susceptible to infection, while A/J was the most susceptible strain, showing the highest parasitemia and mortality rate [51]. In this work, we also observed significant differences in both parasite load and mortality rate between two mouse strains intragastrically infected with T. cruzi SC2005. The A mice were less susceptible to infection, showing lower parasitaemia (Figure 2A), parasite load in the heart (Figure 2B), and mortality rates (Figure 2C). Although these animals presented an earlier mortality (23 dpi), only 10% of the infected mice died. On the other hand, BALB/c mice presented a higher mortality rate (25%), but a later mean time of death (28,4 dpi) (Figure 2C). Since the same protocol and T. cruzi strain were used to infect both mouse lineages, we can suggest that the differences in the parasite load and mortality are influenced by immunological response developed by each mouse strain, which is determined by their genetic backgrounds. Nonetheless, the actual basis for that difference is unknown. BALB/c mice have been described to carry the susceptible genotype for the Scl11c1 gene, a divalent ion transporter present on monocytes, macrophages, NK cells, and ɣδ T cells, which have roles in phagosome maturation, cell activation, and IFN production, rendering hosts susceptible to several intracellular pathogens, such as Leishmania donovani, Salmonella typhimurium, and Mycobacterium sp [54, 55, 56]. The A/J strain, on the other side, carries the resistant genotype [57, 58], being less susceptible to those pathogens. Nevertheless, this locus alone cannot explain host resistance to all pathogens, since BALB/c and A/J mice are both highly susceptible to Staphylococcus aureus infection [56] and BALB/c is resistant to hepatitis caused by Rift Valley Virus [57]. Susceptibility to infection is usually complex and dependent on multiple loci, which cannot be easily identified, but mouse genetics is a powerful tool to study host genetics.

Figure 2.

Influence of the mouse strain on the parasite load. Inbred mice from A and BALB/c genetic backgrounds were infected intragastrically (IG) with 10⁷ metacyclic forms of Trypanosoma cruzi SC2005 strain. BALB/c presented higher parasitaemia (A), higher parasite load on the heart, measured by qPCR (B) and higher mortality rate (C) than A mice. Histopathology from heart of BALB/c animals shows inflammatory infiltration (D) and parasite nests (E; arrows). Hematoxylin and eosin. Two-way ANOVA followed by Sidak’s multiple comparison test. * = p < 0.05; **** = p < 0.0001.

In our study, BALB/c and A mice presented parasites nest on the heart (Figure 2D and E). Cardiac damage in acute Chagas disease is closely related to cases of death [5, 59]. The highest number of deaths observed in BALB/c T. cruzi SC2005-infected mice may be related to the extensive damage of the heart, caused not only by the parasite itself, but also to the inflammatory response against the parasite in this organ, once this mouse strain showed an inflammation on the heart more extensive and intense than A mice. A wide genomic association study carried out with patients from Colombia, Bolivia, and Argentina has identified a QTL in chromosome 11 which is associated with the development of Chagasic cardiomyopathy [28]. The QTL seems to correspond to a methylation site at the CCDC88B gene, which is involved in the inflammatory response [60].

3.2 T. cruzi SC2005 infection induces hematological alterations

In this work, hypochromic anemia was the main hematological alteration found after T. cruzi SC2005 intragastric infection. BALB/c mice showed hypochromic anemia earlier (14 dpi) than A-infected mice (21 dpi) (Figure 3). Anemia is a common hematological alteration observed in acute Chagas’ disease, as described by Chagas in patients infected by T. cruzi [8]. In experimental infections, this hematological alteration also has been shown in different mouse strains [61]. However, the mechanisms responsible for this alteration are still unsolved. In acute T. cruzi infection, the lethality is associated with the reduced number of blood cells and the impaired bone marrow function [62]. All these alterations may be influenced by cytokine secretion and parasite or cell-dependent cytotoxicity in the blood and bone marrow [63].

Figure 3.

Hematological analysis. A-E inbred mice from A and BALB/c genetic backgrounds were infected intragastrically (lg) with 10⁷ metacyclic forms of Trypanosoma cruzi SC2005 strain. BALB/c mice shows a reduction in red blood cells count (RBC; A), hematocrit (Hct; B), and hemoglubin (Hgb; C). The mean corpuscular volume (MCV) of BALB/c red blood cells also presented a slight reduction (D). Extracellular hematopoiesis could be observed in the liver of BALB/c mice (E) but also in outbred Swiss mice intraperitoneally infected with 10⁷ TCC of T. cruzi SC2005 strain (F), where immature cells are also observed (G). E: Giemsa; F and G: Hematoxylin and eosin. Two-way ANOVA followed by Sidak’s multiple comparison test. * = p < 0.05; **p = <0.01; *** = p < 0.001; **** = p < 0.0001 in comparison with normal mice from the same background. # = p < 0.05; ## = p < 0.01; ### = p < 0.001; #### = p < 0.0001, comparing infected mice from different genetic backgrounds.

A reduced life span or sequestration of RBC by autoantibodies or other mechanisms have been described as a factor that can contribute to anemia in protozoan and viral infections [64, 65, 66, 67, 68, 69]. The average life span of a red blood cell is 40 days in a normal mouse [70]. In this study, T. cruzi infection reduced the life span of RBC contributing to anemia, once both mouse-infected strains showed an anemia earlier to this period, between 14 and 21 dpi (Figure 3).

Bone marrow suppression is related with the activity of several cytokines, and among them are TNF-α and IFN-γ [71, 72, 73, 74]. It was shown that an excessive production of TNF-α and IFN-γ promotes damage in hematopoiesis in mice infected with lymphocytic choriomeningitis virus (LCMV) [71]. In studies with malaria, TNF-α has been described as an important anemia mediator [64]. In previous studies, inhibitory effects of TNF-α on erythropoiesis have been demonstrated [74, 75]. This inhibitory effect was also observed during acute infection by T. cruzi, in which the production of TNF-α by activated macrophages was correlated with a decrease in erythropoiesis [63]. In our study, the results indicate that the anemia observed after T. cruzi infection can be associated with a depressed bone marrow function induced by TNF-α and IFN-γ, once both infected mouse strains produced high levels of TNF-α and IFN-γ, 14 days after infection (see below), correlating with the decrease of RBC and Hgb (Figure 3A and C).

One consequence of the depressed bone marrow function, characterized by the poor quality or insufficiency of the blood elements production, is extramedullary hematopoiesis (EMH). EMH can occur in adult mouse livers [65, 66] under myelosuppression by various pathological lesions, including hemoglobinopathies, most commonly sickle cell anemia and thalassemia [67, 68]. In our study, we observed the occurrence of extramedullary hematopoiesis in the livers of T. cruzi-infected animals at 14 and 21 dpi, characterized by the presence of megakaryocytes, immature hematopoietic cells, and mitotic cells (Figure 3E,F and G). Altogether, these findings corroborate the hypothesis that T. cruzi SC2005 infection causing an impairment in bone marrow function, induced by TNF-α and IFN-γ, which lead to the occurrence of anemia and extramedullary hematopoiesis in mice livers as a compensatory mechanism.

Leukocytosis is another hematological alteration described after T. cruzi infection. A systematic review analyzing 31 articles up to 2016 showed that half of them described anemia in infected mammals and 68.2% described leukocytosis [76]. According to Tribullatti et al. [77], molecules such as chemokines, cytokines, antibodies, and nitric oxide produced during T. cruzi infection, together with molecules produced by the parasite itself, lead to hematological alterations in infected animals. In this study, a significant leukocytosis was observed in A and BALB/c mouse strains 21 and 40 days after T. cruzi SC2005 intragastric infection (Figure 4A and B). This leukocytosis was associated with the parasitemia levels and characterized by monocytosis and lymphocytosis, as well as by the presence of a lymphocytic atypia. When comparing infection routes, both IP- and IG-infected animals presented an increased number of leukocytes at the same time when parasitaemia increased. Nevertheless, leukocytosis is much higher in IG-infected mice (p < 0.0001, comparing leukocytosis peak, day 12 for IP- and 18 for IG-infected mice), even with a lower parasitaemia. In both cases, leukocytosis is caused mainly by an increase of lymphocytes, although neutrophils also follow the increase (Figure 4C–F). Several works previously reported alterations in leukocyte counts associated with parasitemia levels in different experimental models. Cynomolgus macaque (Macaca fascicularis) naturally infected by T. cruzi [78] and Rhesus monkeys experimentally infected with T. cruzi Colombian strain [79] showed a positive correlation between leukocytosis, lymphocytosis, and parasitemia peaks. The same correlation was found in Beagle dogs infected by different T. cruzi strains [80, 81]. On the other hand, other authors obtained the contradictory results during experimental murine infection with T. cruzi CL strain in C3H mice, showing an exponential growth of parasites accompanied by leukopenia in these animals [82]. Such dissimilarity in experimental data suggests that both the host genetic background and the T. cruzi strain may influence the hematological alterations occurring after infection.

Figure 4.

White blood cell count analysis. Inbred mice from A (A) and BALB/c (B) genetic backgrounds intragastrically infected with 10⁷ metacyclic forms of Trypanosoma cruzi SC2005 strain presented leukocytosis associated with parasitaemia. C-F. Swiss Webster outbred mice were infected either intraperitoneally or intragastrically with 10⁷ TCC of T. cruzi SC2005 strain. Mice infected intragastrically show an increase in the total leukocytes, lymphocytes (D), and neutrophils (F) at the same time of the parasitemic peak.

3.3 Immune response in T. cruzi SC2005-infected mice

Recognition of T. cruzi by macrophages and dendritic cells leads to phagocytosis and elicits a predominately T helper type 1 (Th1) response with the production of pro-inflammatory cytokines (i.e., IFN-γ, IL-2, IL-6, IL-12, and TNF-α) [83, 84, 85]. This promotes an antiparasitic response, with differentiation and proliferation of Th1 CD4+ cells, activation of CD8+ T cells, and macrophages [86]. The recognition of T. cruzi-infected cells and effective control of infection during the acute and chronic phases are mainly related to the action of effector CD8+ T cells [48, 79, 87]. More severe clinical disease is associated with reduced CD8+ T cell responses and the boost of CD8+ T cell response after treatment improves the clinical outcome [88, 89]. However, studies indicate that, in the absence of CD4⁺ T cells, CD8⁺ T lymphocytes fail to restrain the parasite growth [90], because CD4⁺ T lymphocytes are responsible for promoting the macrophages activation and CD8⁺ T and B cells proliferation. Therefore, its deficiency leads to an overall reduction of host immune response and a consequent increase in tissue parasitism [91].

In our study, the frequencies of CD4⁺, CD8⁺, CD4⁺/CD8⁺ T cells and B cells in different organs varied according to the route of infection, and mouse strain. After intraperitoneal infection by T. cruzi SC2005 strain, Swiss mice presented an increase in CD4⁺ T cells frequency in the spleen at 18 dpi. On the other hand, IG-infected mice showed an increase in the frequencies of CD8⁺ T cells 26 and 33 dpi (Figure 5A). The increase of these cells corroborates with the increase in spleen weight (Figure 6A), caused by the intense production of cells, as shown by the hyperplasia of the germinal center induced by both infection routes in this organ (Figure 6B and C). An increase in the cellularity and size of spleen, caused by the amplification of T and B lymphocytes polyclonal activation, has been previously described during T. cruzi infection [92].

Figure 5.

T cells present in the inflammatory infiltrate. Swiss Webster outbred mice were infected either intraperitoneally (IP) or intragastrically (IG) with 10⁷ TCC of Trypanosoma cruzi SC2005 strain. Intragastrically-infected mice show an increase of T CD8⁺ cells in the spleen (A) and blood (B), but not in the draining lymph node (C). IP = intraperitoneally infected; IG = intragastrically infected. Two-way ANOVA followed by Sidak’s multiple comparison test, in comparison to non-infected normal mice. * = p < 0.05; *** = p < 0.001; **** = p < 0.0001.

Figure 6.

Cell proliferation in the spleen. Swiss Webster outbred mice were infected either intraperitoneally or intragastrically with 10⁷ TCC of Trypanosoma cruzi SC2005 strain. All mice present a higher spleen index (A) than normal mice, showing cell proliferation is occurring on the organ, which is corroborated with histopathology image from intraperitoneally (B) and intragastrically (C) infected mice. Hematoxylin and eosin. One-way ANOVA followed by Sidak’s multiple comparison test. * = p < 0.05; **** = p < 0.0001.

In the blood, IG-infected mice showed a reduction of CD4⁺ T cells frequency (26 dpi) and an increase of CD8⁺ T cells 26 and 33 days after infection (Figure 5B). Meanwhile, the analysis of lymph nodes showed a decrease of CD4⁺ T cell frequency on both infected groups: IP-infected mice at 11 dpi and IG-infected mice 26 and 33 dpi (Figure 5C). The variation in the T cells frequency in the mesenteric lymph nodes is caused by depletion of lymphocytes and increased apoptosis rates, which may be related to the production of different cytokines [92].

These results show a fluctuation in the percentage of lymphocytes in the analyzed tissues. This variation demonstrates the differentiated role of each compartment and a different response profile with a specific and coordinated response of these sites against the parasite [93]. Besides, we demonstrated the influence of infection route on the lymphocyte’s frequency and magnitude, once T. cruzi IG-infected mice showed a higher expansion of CD8⁺ T cells in the spleen and blood.

The differences in expansion and distribution of T and B cell frequencies in the blood, liver, and heart of A and BALB/c mice IG infected by T. cruzi SC2005 were also demonstrated in this work. SC2005 T. cruzi IG infection induced an increase of CD8⁺ T and CD4⁺/CD8⁺ T lymphocytes in all analyzed organs on both mouse strains, although in A mice the increase occurred earlier than in BALB/c mice. The infection also induced a decrease in CD4⁺ T cell and B lymphocyte frequencies in the heart and blood of infected animals (Figure 7). It is known that in both acute and chronic T. cruzi infections, the increase of CD8⁺ T lymphocytes is common [69]. Increased levels of these cells have been described as Chagas disease biomarkers in humans and monkeys naturally infected by T. cruzi [70]. CD8⁺ T cell type-1 (Tc1) subsets are the main cause of T. cruzi death through the production of both IFN-γ and TNF-α [94, 95], contributing to the control of parasitaemia levels [91, 96, 97]. In this study, BALB/c and A mice infected by T. cruzi SC2005 presented lymphocytosis (Figure 4A and B) and an increase of CD8⁺ T cells in the blood at the same time of the parasitaemia peak (21 dpi), remaining until the end of the experiment (40 dpi), when the parasite load was very low (Figure 2A). These results reinforce the previous observations about the role of CD8⁺ T cells in the control of parasitaemia levels and its probable cytotoxic activity against the parasite, as reported previously [78].

Figure 7.

Influence of the mouse strain on cell recruitment. Inbred mice from A and BALB/c genetic backgrounds were infected intragastrically (IG) with 10⁷ metacyclic forms of Trypanosoma cruzi SC2005 strain. T. cruzi Infection caused an increase of CD8+ cells on the heart (A), blood (B) and liver (C) of infected mice. However, BALB/c CD8+ cells increased from 21 days after infection, whereas at 14 days after infection A mice already presented higher frequencies of CD8+ cells. The same happened with CD4+/CD8+ in mice blood. (B) CD19+ cell frequency was reduced in the blood from both mouse strains. Lines show normal mice mean frequencies. Two-way ANOVA followed by Sidak’s multiple comparison test in comparison with normal mice from the same background. * = p < 0.05; ** = p < 0.01; *** = p < 0.001; **** = p < 0.0001.

CD4⁺/CD8⁺ double-positive T cells, under normal conditions, are found in the thymus, where they undergo differentiation into mature CD4⁺ and CD8⁺ T cells. During T. cruzi infection, there is a significant impairment of this organ, due to a deregulated cascade of proinflammatory cytokines. This impairment leads to cell maturation in extrathymic organs such as the bone marrow and liver [98, 99]. In this study, the increase of the CD4⁺/CD8⁺ T double-positive cells frequency in the heart, liver, and blood indicates an impairment of the thymus and consequent liberation of immature cells in the circulation or the occurrence of extracellular hematopoiesis in the liver (Figure 3E).

In T. cruzi infection, the clearance of blood trypomastigotes occurs in the liver [100], regardless of the gateway. The liver is the main organ involved in the defense against disseminating blood pathogens, being critical to host immunity and survival [101, 102]. In this work, an increase of both CD8⁺ T cells and CD19⁺ B cells was observed in the hepatic parenchyma (Figure 7C). B cells act as antigen-presenting cells (APC), in the secretion of antibodies, and in the activation of CD8⁺ T cells [103]. Besides, they have been implicated in the mobilization of inflammatory cells to the tissues and are fundamental to the control of parasite growth, by triggering a Th1 response [90, 104, 105]. The observation of these cells in the liver suggests the role of CD19⁺ cells as antigen-present cells (APC). A mice showed an earlier increase of these cell frequencies (14 dpi) than BALB/c-infected mice (21 dpi) (Figure 7C). The early presence of B and CD8⁺ T cells in the liver of A mice indicates the importance of this organ in parasite clearance and may explain why this mouse strain has lower parasite load and mortality.

Redistribution and circulation of lymphocyte subtypes to other sites are modulated by the local and systemic immune system [106]. This regulation is affected by the infection and often related to the severity of the clinical manifestations.

Several murine studies have described the importance of proinflammatory cytokines during T. cruzi infection. C57BL/6 IL-17A knockout (IL-17A −/−) mice infected by T. cruzi Tulahuén strain showed a reduced production of IFN-γ, IL-6, and TNF-α cytokines, which was related to a more severe parasitemia and mortality, than observed in wild-type mice [107]. On the other hand, an improvement of the mice resistance to parasite growth was observed when IL-18 or 5-lipoxygenase was inhibited, generating an increase in the IL-12, IFN-γ, IL-1β, and IL-6 levels during the acute phase of disease [108, 109]. Studies in vitro, using PBMC infected by T. cruzi Tulahuén strain, indicate that IL-6 improves the survival and effector functions of cytotoxic cells [110]. These data reveal the important role of these cytokines in the control of T. cruzi infection and host mortality. In the present work, both infected mouse strains produced a similar pattern of TNF-α, IFN-γ, and IL-6 production, but A-infected mice showed an earlier increase in the production of these cytokines, when compared to BALB/c mice (Figure 8).

Figure 8.

Cytokine production by mice from different strains. Inbred mice from BALB/c (A) and A (B) genetic backgrounds were infected intragastrically (IG) with 10⁷ metacyclic forms of Trypanosoma cruzi SC2005 strain. Both strain presented similar cytokine production, although it starts earlier in A mice. Two-way ANOVA followed by Sidak’s multiple comparison test. ## = p < 0.01, in comparison with the other mouse strain.

3.4 Histopathological alterations caused by SC2005 T. cruzi infection

After infection, T. cruzi can be detected in several organs/tissues and induce an inflammatory response, which can be found even where the parasite is not detected [111, 112]. The preferential tropism, as well as distinct local and systemic immune responses can be influenced by different transmission routes of T. cruzi [39, 47, 112, 113] as well as by the host genetic background [114].

In this study, we observed differences in the parasite load and inflammatory infiltrate intensity depending on the inoculation route used. IP infection showed an higher tissue colonization by SC2005 T. cruzi and a more intense inflammatory infiltrate than in IG infection. IP-infected mice showed a moderate-to-intense diffuse mononuclear inflammatory infiltrate (composed mainly by monocytes and lymphocytes) in the esophagus, stomach, intestine, heart, liver, pancreas, adrenal gland, bladder, uterus, trachea, and adipose tissue (Figure 9). Parasite nests were observed in the stomach (Figure 9A), esophagus (Figure 9B), intestine, heart, pancreas, bladder, and adipose tissue. Mast cells were observed only at 18 dpi in inflammatory infiltrates of the heart, stomach, and adipose tissue (Figure 9D). Severe pancreatitis with focal necrosis were also observed (Figure 9E and F) and the liver of the infected mice showed immature cells, megakaryocytes, and dividing cells (Figure 3F and G). In addition, a redistribution and increase of collagen fibers, associated with inflammatory infiltrates, was observed in the esophagus, heart (Figure 9G), stomach (Figure 9H), bladder, and uterus. Thymus and brain showed no changes in these animals.

Figure 9.

Histopathological alterations in intragastrically-infected mice. Swiss Webster outbred mice were intraperitoneally (IP) infected with 10⁷ TCC of Trypanosoma cruzi SC2005 strain. Inflammatory infiltration and parasite nests (arrows) can be observed in mice stomach (A) and esophagus (B). Different other tissues also present intense inflammation, including supra renal (C) and adipocytes, in which we can observe mast cells (arrow head) (D). IP infected mice presented severe pancreatite (E and F). Collagen fibers can be observed in heart (G) and stomach (H) of mice. Hematoxylin and eosin (A-F) and Picrus Sirius red in polarized light (G-H).

On the other hand, IG infection induced a moderate-to-intense diffuse inflammatory infiltrate essentially lymphomonocytic in the esophagus, stomach, liver, kidney, bladder, uterus, brain, intestine, heart, pancreas, and adipose tissue (Figure 10). Mast cells were observed in the inflammatory infiltrates of the adipose tissue, bladder, and stomach (Figure 10A). Parasite nests were observed in lower numbers than IP-infected mice only in stomach, heart, bladder (Figure 10D), and adipose tissue. Thymus showed no changes in these animals. A common alteration observed in both IP- and IG-infected animals was the germinal centers hyperplasia in spleen and lymph nodes (Figure 6C), as well as omental and mesenteric milky spots were activated (Figure 10C), and myeloid cells are present. In addition, the liver of the infected mice showed immature cells, megakaryocytes, and dividing cells.

Figure 10.

Histopathological alterations in intraperitoneal infected mice. Swiss Webster outbred mice were intragastrically (IG) infected with 10⁷ TCC of Trypanosoma cruzi SC2005 strain. Moderate inflammatory infiltrate are present in different tissues such as stomach (A), esophagus (B), intestine (C), bladder (D), uterus (E), and meninges (F). Several mast cells can be observed in mice stomach (A). Activated milk spots are present in the intestine (C) and small parasite nests were observed in the bladder (arrow) (D). Hematoxylin and eosin.

Both infected mice showed a redistribution and an increase of collagen fibers, associated with inflammatory infiltrates and a decrease of these infiltrates with the course of infection, with exception of the heart, where there is an increase of these infiltrates in the later times (IP-18 dpi and IG-33 dpi). These inflammatory infiltrates were preferentially located in the muscle layer of the organs. In the heart, they were preferentially located in the atria, while parasites were found in the heart ventricles. IP-infected mice showed an increase in heart parasitism at 18 dpi. On the contrary, IG-infected mice showed scarce parasite nests in this organ at 33 dpi and a much larger inflammatory infiltrate (essentially lymphoid) than observed in IP-infected mice (Figure 1C and D).

Similar to the observed Swiss mice infected by IG or IP routes, A and BALB/c mice IG infected by SC2005 T. cruzi strain showed immature cells and megakaryocytes in the liver (Figure 3E); and mononuclear inflammatory infiltrates associated with the increase of collagen fibers in several tissues (Figure 11). Both mouse strains presented inflammatory infiltrates in several organs, but in BALB/c the stomach (21 and 40 dpi), heart (14 and 21 dpi) (Figure 2D), and liver (7 dpi) were more inflamed. On the other hand, the esophagus of A mice presented an intense inflammatory infiltrate, whereas BALB/c’s did not show any alterations. Neither mouse strains present histopathological changes in the gut (Figure 11A).

Figure 11.

Histopathological alterations in intragastrically-infected BALB/c mice. Inbred BALB/c mice were infected intragastrically (IG) with 10⁷ metacyclic forms of Trypanosoma cruzi SC2005 strain. Inflammatory infiltration was quantified in various organs (A) inflammation can be observed in mice stomach (B) and liver (C). Alteration of collagen fibers are also observed in heart (D), and stomach (E). Hematoxylin and eosin (B-C) and Picrus Sirius red (D-E).

As it can be observed, the SC2005 T. cruzi infection induced similar alterations independent of the route of infection (IP or IG) and the mouse strain (BALB/c or A). The histopathological analysis showed a mononuclear infiltrate mainly located in the muscular layers, which was associated with neoformation and remodeling of collagen fibers in different organs (Figure 11D and E). Studies with T. cruzi strains belonging to the biodemes type II and III showed the same histopathological patterns of localization of inflammatory infiltration [5]. The deposition of collagen and the tissue remodeling during T. cruzi infection have already been described in several studies [115, 116, 117]. In the heart, amastigote nests were present more frequently in the ventricles than in the atria. However, the atria presented a more intense inflammatory infiltration than ventricles. These findings suggest that the inflammatory response against the parasite first occurs in the auricular tissue and only later reaches the ventricular tissue. It is interesting to note that Quijano-Hernández et al. in 2012 [118] observed the same histopathological patterns in infected dogs. Cardiac damages are closely related to cases of death in acute Chagas disease [5, 59]. In this study, a larger cardiac involvement and a higher parasite load were observed in the hearts of BALB/c-infected mice. The extensive damage of the heart plus the large number of parasite DNA and a late CD8⁺ T cell response, corroborate to the highest number of deaths in this group.

The presence of immature cells and megakaryocytes in the liver was also a common finding in all infected mice. This finding suggests that extramedullary hematopoiesis was occurring in this organ. During the acute phase of T. cruzi infection, Marcondes et al. in 2000 [82] demonstrated alterations in blood cell counts associated with bone marrow suppression and anemia, which explain the occurrence of extramedullary hematopoiesis.

4. Conclusion

Altogether, the findings of this study point that T. cruzi SC2005 strain can spread through the bloodstream and colonize different organs and tissues, producing a systemic response, which is variable depending on the inoculation route and the genetic background of the host. Heart and stomach were the most intensely parasitized and inflamed organs in all models. T. cruzi SC2005 strain infects preferentially the muscular layers of the organs, where inflammatory infiltrates are also observed with higher intensity, being associated with an increase and redistribution of collagen fibers. In the heart, inflammatory infiltrates are preferentially located in the atria, while parasites are mostly found in the ventricles.

T. cruzi SC2005 intragastric infection induces a hypochromic anemia, and an extramedullary hematopoiesis in mouse livers, characterized by the presence of immature cells and megakaryocytes, as well as an increase of CD4⁺/CD8⁺ frequencies, corroborating the hypothesis that T. cruzi SC2005 infection causes an impairment in bone marrow function. Besides, the infection also induces a leukocytosis, characterized by an increase of CD8⁺ T lymphocytes, as well as an increase of proinflammatory cytokines production. The increase of leukocytes was correlated with the increase of parasitaemia.

The intraperitoneal infection proved to be more infective and severe than the intragastric route, leading to a higher parasitaemia, mortality, tissue colonization, and tissue inflammatory response.

The mouse strain influences the immune response, pointing to a role for host genetics in the susceptibility to infection. In this study, although T. cruzi SC2005 intragastric infection has been induced a similar profile of changes in A and BALB/c mice, the earlier development of a proinflammatory cytotoxic cellular profile of A mice led to a less severe disease outcome, with lower parasite load and mortality. Infected A mice also exhibited an early induction of CD8⁺ T cells and proinflammatory cytokine production. On the other hand, for BALB/c-infected mice the response to infection occurred later, after a considerable increase in parasitemia, favoring the parasite multiplication and spread, and consequent higher mortality rate and tissue inflammation.

This study adds highlights on the factors that influence the pathology of Chagas disease, helping in the understanding of its different outcomes.

Acknowledgments

The authors would like to thank the Flow Cytometry Core Facilities of the Instituto Oswaldo Cruz/FIOCRUZ-RJ. We also thank Sandy Santos Pereira for providing technical support, and Thaize Q. Chometon for operating the Cytoflex Flow Cytometer.

This work was supported by CAPES by student scholarship and Oswaldo Cruz Institute.

Conflict of interest

The authors declare no conflict of interest.