Chagas disease (CD), caused by the protozoa Trypanosoma cruzi, is a chronic illness in which parasites persist in the host-infected tissues for years. T. cruzi invasion in cardiomyocytes elicits the production of pro-inflammatory mediators [TNF-α, IL-1β, IFN-γ; nitric oxide (·NO)], leading to mitochondrial dysfunction with increased superoxide radical (O2·−), hydrogen peroxide (H2O2) and peroxynitrite generation. We hypothesize that these redox mediators may control parasite proliferation through the induction of intracellular amastigote programmed cell death (PCD). In this work, we show that T. cruzi (CL-Brener strain) infection in primary cardiomyocytes produced an early (24 h post infection) mitochondrial dysfunction with H2O2 generation and the establishment of an oxidative stress evidenced by FoxO3 activation and target host mitochondrial protein expression (MnSOD and peroxiredoxin 3). TNF-α/IL-1β-stimulated cardiomyocytes were able to control intracellular amastigote proliferation compared with unstimulated cardiomyocytes. In this condition leading to oxidant formation, an enhanced number of intracellular apoptotic amastigotes were detected. The ability of H2O2 to induce T. cruzi PCD was further confirmed in the epimastigote stage of the parasite. H2O2 treatment induced parasite mitochondrial dysfunction together with intra-mitochondrial O2·− generation. Importantly, parasites genetically engineered to overexpress mitochondrial Fe-superoxide dismutase (Fe-SODA) were more infective to TNF-α/IL-1β-stimulated cardiomyocytes with less apoptotic amastigotes; this result underscores the role of this enzyme in parasite survival. Our results indicate that cardiomyocyte-derived diffusible mediators are able to control intracellular amastigote proliferation by triggering T. cruzi PCD and that parasite Fe-SODA tilts the process toward survival as part of an antioxidant-based immune evasion mechanism.
Chagas disease (CD), caused by the protozoan parasite Trypanosoma cruzi, remains a major public health concern in Latin America with the disease spreading worldwide as a result of migration of infected mammalian host and insect vector (www.cdc\chagas\factsheet.htlm). Organ and tissue damage during the acute phase of CD is caused by parasite itself and by the host's acute inflammatory response, elicited by the presence of the pathogen . Between 30 and 40% of the patients will develop CD either in its cardiac, digestive or cardiodigestive manifestation 10–30 years after the initial infection. Destruction and dysfunction of mitochondrial myocardial cells together with inflammation, nitration of cardiac tissue, fibrosis and hypertrophy are present in chronic chagasic cardiomyopathy [2–4]. Exacerbation of the chronic stage with increased parasitemia and intracellular parasite proliferation is found in immunosuppressed individuals, indicating the central role of the host immune mediators in the control of parasite proliferation and persistence [5,6]. It was previously shown that T. cruzi infection to cardiomyocytes elicits the production of pro-inflammatory mediators (TNF-α/IL-1β) and the induction of the cardiomyocyte-inducible nitric oxide synthase (iNOS) with the subsequent generation of ·NO [2,7,8]. In this line, T. cruzi invasion together with pro-inflammatory cytokines and ·NO production by cardiac, endothelial and immune cells leads to an increase in nitroxidative stress that may account not only for the host cell and tissue damage but also be responsible, in some circumstances, for the control of T. cruzi amastigote proliferation inside the myocardial cells. In this scenario, where host-derived oxidant mediators are actively generated, the antioxidant armamentarium of T. cruzi may become decisive for parasite survival and persistence [9–11].
T. cruzi consists of a mixed population with genetic and biochemical heterogeneity that is responsible, in part, for the diverse clinical manifestation of CD [12,13]. Using different phylogenetic T. cruzi strains, a positive correlation was found between the levels of antioxidant enzymes (cytosolic and mitochondrial peroxiredoxins, CPX and MPX, respectively, and trypanothione synthase, TS) and the virulence in the mouse model of CD [10,14]. Strains with higher virulence showed higher parasitemias with an important heart inflammatory infiltrate and tissue destruction .
In the non-infective epimastigote stage of T. cruzi, the intracellular imbalance of the parasite redox status can disengage the programmed cell death process (PCD) with the disruption of mitochondrial homeostasis and the generation of intra-mitochondrial O2·− [15–17]. The modulation of mitochondrial O2·− levels by mitochondrial superoxide dismutase (Fe-SODA) influences parasite PCD, highlighting the role of this enzyme for parasite survival . In this line, the ability of ·NO to inhibit parasite mitochondrial respiration with the generation of intra-mitochondrial O2·− production was previously shown [17,18]. In the presence of ·NO, O2·− reacts rapidly to generate the potent oxidant peroxynitrite (ONOO−) at the site of O2·− production, a strong cytotoxic molecule against T. cruzi [19,20]. Parasites overexpressing mitochondrial peroxiredoxin (TcMPX) are more resistant to ·NO fluxes, which is in line with the predicted ·NO-mediated intra-mitochondrial ONOO− formation and detoxification by this peroxiredoxin . This result makes feasible the concept of ·NO-mediated control of parasite proliferation in CD via intra-parasite O2·− and ONOO− formation, but no studies have yet been performed on the intracellular amastigote infective stage in order to test this concept.
Intracellular apoptotic amastigotes have been found in experimentally infected cardiomyocytes as well as in vivo in heart tissue of infected mice [22,23]. Moreover, it was found that virulent T. cruzi strains have less intracellular amastigote PCD and higher proliferation rates than attenuated strains in cardiomyocyte infections . It has been proposed that apoptosis enables the regulation of parasite densities in distinct host compartments, facilitating a sustained infection in the vertebrate host. Although intracellular apoptotic amastigotes were detected in vivo, the host-derived mediators that trigger parasite PCD are still unidentified. Here, we propose that the balance between diffusible redox mediators, produced in response to parasite infection (i.e. H2O2 and ·NO), and parasite antioxidant responses may be responsible for the control of intracellular parasite proliferation by PCD induction in the amastigote stage. Parasites containing higher antioxidant enzyme contents may be more resistant to the host-derived death stimuli, leading to a severe infection.
Materials and methods
Rat-derived cardiomyocytes (H9c2, ATCC®CRL1446™ were cultured in DMEM supplemented with l-glutamine (2 mM), penicillin (100 units/ml), streptomycin (100 mg/ml) and 10% (v/v) heat-inactivated fetal bovine serum (FBS) at 37°C in a 5% CO2 atmosphere. Primary cardiomyocyte cultures were obtained from 15 neonatal (1–3 days old) BALB/c mice. Animals were killed by decapitation, hearts were removed aseptically and kept on ice in Hanks' balanced salt solution (HBSS) containing NaCl (8 g/l), KCl (0.4 g/l), KH2PO4 (0.06 g/l), K2HPO4 (0.048 g/l), glucose (1 g/l), MgSO4 (0.098 g/l), CaCl2 (0.14 g/l) and NaHCO3 (0.35 g/l), pH 7.4. Hearts were washed several times with fresh ice-cold HBSS and minced into small fragments. The cardiac tissue was dissociated overnight at 4°C in 7 ml of 0.5% (w/v) trypsin–EDTA [0.2% (w/v), Sigma] diluted in HBSS. Then, tissue was submitted to a second dissociation step in 5 ml collagenase II (Sigma), dissolved in Leibovitz medium (Sigma) to a final concentration of 1 mg/ml. The resultant cell suspension was filtered through a 70 µm cell strainer and centrifuged at 300 g for 5 min. The pellet was resuspended in DMEM supplemented with l-glutamine (2 mM), penicillin (100 U/ml), streptomycin (100 mg/ml) and 10% (v/v) heat-inactivated FBS. The cell suspension was pre-plated for 2 h at 37°C in a 5% CO2 incubator in order to remove most fibroblasts and other non-muscle cells. Afterwards, supernatant enriched in cardiomyocytes was collected and seeded at 1.0–5.0 × 105 cells/well, into wells coated with 0.02% (v/v) fibronectin (Sigma) and 0.05% (w/v) gelatin, in Lab-Tek Chamber slide (Nunc) or 24-well plates. Cells were kept at 37°C in a humidified atmosphere containing 5% CO2 for 72 h before experimental procedures to allow adhesion of cardiomyocytes .
Parasite culture, differentiation and cardiomyocyte infections
T. cruzi epimastigotes from the CL-Brener (wild type, wt) were cultured at 28°C in brain–heart infusion (BHI) medium as described previously . Parasites overexpressing mitochondrial iron superoxide dismutase (Fe-SODA) were kindly provided by Dr Shane Wilkinson (Queen Mary University of London, U.K.)  and cultured in the presence of geneticin and hygromycin (250 and 100 µg/ml, respectively) . Wt and Fe-SODA overexpressers were differentiated to the infective metacyclic stage, and metacyclic forms were purified by overnight incubation with fresh human serum as previously described . Metacyclic trypomastigotes were used to infect confluent Vero cells (American Type Culture Collection) at 37°C in a 5% CO2 atmosphere. Culture-derived trypomastigotes were used to infect primary and/or rat-derived cardiomyocytes (H9c2) with or without stimulation with pro-inflammatory cytokines (TNF-α, 25 ng/ml; IL-1β, 25 ng/ml) in Lab-Tek chamber slides (parasite/cell ratio of 5–10 : 1) . After 4 h, no engulfed parasites were removed by washing twice in Dulbecco's PBS at a pH of 7.4 (Sigma), and cells were further incubated for 24–48 h in DMEM at 37°C. Infected cells were fixed in fresh PFA solution [4% (v/v) in PBS] for 10 min at room temperature, washed with PBS containing glycine (100 mM) and permeabilized for 5 min with Triton X-100 (0.1%, v/v) in PBS. The number of parasites per 100 cardiomyocytes was determined by DAPI staining (5 µg/ml). Preparations were analyzed using a microscope (Nikon Eclipse TE-200) at a magnification of 1000×, and digital photographs of infected cells were recorded. At least 1000 cells from three independent experiments were counted. Results are expressed as the number of amastigotes per 100 cells and represent the mean of three independent experiments.
Mitochondrial membrane potential in T. cruzi-infected primary cardiomyocytes
Mitochondrial membrane potential (Δψ) was measured using the J-aggregate-forming lipophilic cation JC-1. Primary cardiomyocytes were grown on 24-well plates (glass bottom) and infected or not with wt T. cruzi trypomastigotes for 24 h. JC-1 (0.5 µM) and DAPI (5 µg/ml) were added to the wells; cells were incubated at 37°C for 1 h and rinsed with Krebs buffer. Cells were visualized by a fluorescence microscope (Nikon Eclipse TE 200), equipped with a digital camera, at a fixed exposure time. JC-1 aggregate fluorescence was observed using a filter cube with an excitation filter (wavelength/band pass) of 540/25 nm, an emission filter of 605/55 nm and a beam splitter at 565 nm, whereas the monomer was observed using a filter cube with an excitation filter (wavelength/band pass) of 480/30 nm, an emission filter of 535/40 nm and a beam splitter at 505 nm.
Oxygen consumption in T. cruzi-infected primary cardiomyocytes
Cardiomyocyte mitochondrial oxygen consumption was evaluated using a Seahorse XF24 extracellular flux analyzer (Agilent technology) according to the manufacturer's instructions with the following adaptation. Briefly, primary cardiomyocytes were seeded (5 × 104 cells/well) in XF24-cell culture plates (Seahorse Bioscience) coated with gelatin (0.02%, w/v) and fibronectin (0.05%, v/v) in DMEM containing FBS (10%, v/v) at 37°C for 24 h. Afterwards, cells were infected or not with T. cruzi trypomastigotes from wt and/or Fe-SODA-overexpressers (parasite : cell ratio 5 : 1). After 24 h of infection, culture medium was thrown away and replaced with unbuffered DMEM-modified medium containing glucose (10 mM), pyruvate (1 mM) and l-glutamine (2 mM). Each measuring cycle consists of a mixing time of 3 min and a data acquisition period of 3 min. The O2 consumption rate was measured and recorded before and after the addition of the following compounds: oligomycin (5 µM), carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP, 5 µM) and antimycin A (AA, 5 µM) plus rotenone (ROT, 1 µM). After subtracting the non-mitochondrial respiration (oxygen consumption in the presence of AA-ROT) from each data item, oxygen consumption rates were normalized against cell counts and reported as pmol O2/min/5 × 104 cells. At least three independent experiments with five biological replicates were performed. The RCR was calculated as the oxygen consumption after oligomycin addition divided the oxygen consumption after FCCP (QO2FCCP/QO2 Oligomycin) .
H2O2 production by T. cruzi-infected cardiomyocytes
Primary cardiomyocytes were infected or not with wt trypomastigotes (parasite : cell ratio 5 : 1) and culture supernatants were collected at different times after infection (0–4 days) and stored at −80°C until use. H2O2 was measured in the supernatants in the presence of the Amplex® Red reagent (50 µM, Invitrogen) and horseradish peroxidase (0.1 U/ml, HRP) in Krebs–Ringer phosphate buffer (pH 7.35). The red-fluorescence product, resorufin, was measured in a fluorescence plate reader at 37°C (Varioskan Flash, Thermo Scientifics) at λex = 568 nm and at λem = 581 nm. Fluorescence values obtained were compared with a standard curve of H2O2 (0–250 µM) in the same experimental conditions and results were expressed as nmol H2O2/105 cardiomyocytes.
Antioxidant response in T. cruz-infected primary cardiomyocytes
The activation by oxidative stress of the transcription factor FoxO-3A was evaluated by fluorescence microscopy. Briefly, primary cardiomyocytes were seeded in Nunc® Lab-Tek® Chambers and infected or not with wt T. cruzi trypomastigotes in the presence and absence of pro-inflammatory cytokines as described above for 12 h. Afterwards, cells were fixed with fresh PFA (4%, w/v) at room temperature for 1 h, rinsed in PBS (pH 7.4) and permeabilized with Triton X-100 (0.1%, v/v) in PBS (pH 7.4) containing bovine serum albumin (1%, w/v) for 1 h. Immunodetection of FoxO-3A activation and nuclear translocation was performed using a mouse antibody anti-FoxO-3A (1 : 50) (Abcam) in Tween-20 (0.1%, v/v in PBS) overnight at 4°C. After copious washes, the slides were then incubated for 1 h with a secondary goat anti-mouse antibody conjugated to Alexa Fluor 488 (Invitrogen) diluted 1 : 100 in Tween-20 (0.1%, v/v) in PBS. DNA was visualized with DAPI (5 μg/ml). All slides were examined using an epifluorescence microscope previously described. Image processing was performed with the program ImageJ 1.36B (Wayne Rasband, National Institutes of Health, http://rsb.info.nih.gov/ij/).
Expression of FoxO-3 target genes
The protein expression levels of mitochondrial MnSOD and peroxiredoxin-3 (both under FoxO-3 activation) were evaluated by western blot. Briefly, cardiomyocytes were infected as described above and lysed in buffer Tris–HCl (50 mM, pH 7.4) containing NaCl (150 mM), nonidet P-40 (1%, v/v), sodium deoxycholate (0.25%, w/v) and EDTA (1 mM). For immunoblotting, samples were separated by SDS–PAGE (15%), transferred onto the nitrocellulose membrane and blocked with nonfat dry milk (3%, w/v) in PBS (pH 7.4) for 1 h. The membranes were incubated with either monoclonal anti-Prx3 antibody (Abcam) or monoclonal anti-MnSOD antibody (Santa Cruz) together with a rabbit polyclonal anti-βactin antibody (Sigma). Primary antibodies were diluted 1 : 2000 in Tween-20 (0.1%, v/v) in PBS (pH 7.4) and incubated overnight at 4°C. The membranes were then washed in Tween-20 (0.1%, v/v) in PBS and were incubated for 1 h with anti-mouse-IgG (IR Dye-800 conjugated) and anti-rabbit-IgG (IR Dye-680 conjugated) (LI-COR Biosciences) diluted 1 : 15 000. After washing of the membranes, inmunoreactive proteins were visualized with an infrared fluorescence detection system (Odyssey, LI-COR Biosciences).
Intra-cardiomyocyte amastigote purification
The H9c2 cardiomyocytes were used for the purification of intracellular amastigotes due to the higher efficiency in the number of amastigotes obtained. Briefly, cardiomyocytes were infected with T. cruzi trypomastigotes from wt and/or Fe-SODA overexpressers (parasite : cell ratio 5 : 1) for 48–72 h in the presence or absence of pro-inflammatory cytokines (25 ng/ml each) and the ·NO-donor NOC-18 (0.5 mM, yielding 0.15 µM ·NO/min). Amastigotes were purified using the Optiprerp density gradient (Sigma). Briefly, cell cultures were washed twice in ice-cold PBS (pH 7.4) and mechanically harvested with a scraper in ice-cold PBS supplemented with a mixture of protease inhibitors (Sigma). Cell suspension was centrifuged at 300×g for 5 min at room temperature, and the pellet was resuspended in ice-cold PBS supplemented with protease inhibitors. Cells were disrupted using a syringe (27Gx1/2″ 0.4 × 13 mm), and amastigote-enriched suspension was poured on top of 1 volume of Optiprep 16% (v/v) in PBS and centrifuged at 800 g × 15 min. Pellets containing purified amastigotes from the different experimental conditions were evaluated by bright field microscopy and immediately used for TUNEL staining using the APO-BrdU TUNEL assay kit (Invitrogen) following the manufacturer's instructions. Parasites positive for TUNEL staining were analyzed by flow cytometry and the results are expressed as the increase in the M1 population compared with non-treated cardiomyocytes (Beckton Dickson). Alternatively, purified amastigotes were evaluated for mitochondrial O2·− generation using MitoSOX (Invitrogen). For this, purified amastigotes were preloaded with MitoSOX (5 μM, 30 min) washed three times in PBS and incubated at 37°C in the presence or absence of NOC-18 (0.5 mM) for 3 h. Intra-parasite MitoSOX oxidation was evaluated by flow cytometry as previously described .
H2O2 induction of PCD in T. cruzi epimastigotes
Epimastigotes (3 × 108 cells) in the logarithmic phase of growth (5 days) from the different T. cruzi strains (wt and Fe-SODA overexpressers) were incubated in sterile PBS (pH 7.4) and exposed to H2O2 (300 μM) at 28°C for 10 min . After incubation, parasites were collected by centrifugation at 800 g for 10 min, washed twice in PBS and resuspended in PBS and/or BHI at a cell density of 1 × 107 cells/ml. Epimastigote-PCD was evaluated analyzing different hallmarks. For exposure of PS and maintenance of plasma membrane integrity after 3–24 h of oxidant treatment, epimastigotes (1 × 106) were incubated with Annexin V conjugated with Alexa Fluor® 488 (Invitrogen) according to the manufacturer's instructions in the presence of propidium iodide (PI, 1 µg/ml) and cells were analyzed by flow cytometry (FACS-Calibur, Becton Dickinson). Mitochondrial membrane potential (Δψm) was evaluated after 3 h of H2O2 treatment using JC-1 dye (Invitrogen). Epimastigotes (1 × 107) were incubated with JC-1 (5 μM) for 30 min at 28°C and washed twice with PBS (pH 7.4). JC-1 green and red fluorescence were evaluated by flow cytometry. Epimastigote DNA fragmentation was evaluated after 24 h of exposure to H2O2 by TUNEL staining using the APO-BrdU™ TUNEL Assay Kit (Invitrogen) according to the manufacturer's instructions.
Oxygen consumption in H2O2-treated T. cruzi epimastigotes
Analysis of mitochondrial oxygen consumption was evaluated using the Seahorse XF24 analyzer (Agilent Technology) after 4 h of H2O2 treatment. For this, XF24 microplates were first coated with gelatin (2%, w/v in PBS) and incubated for an hour at 28°C in a non-CO2 incubator. Parasites treated or not with H2O2 were resuspended at a cell density of 5 × 107 cells/ml in DMEM-modified medium (NaCl, 1.85 g/l, phenol red, 15 mg/l; sodium pyruvate, 1 mM; glucose, 10 mM and l-glutamine, 2 mM). Epimastigotes were seeded at a cell density of 5 × 106 cells/well (100 µl) in gelatin-precoated XF24 microplates, spun at 800 g for 5 min and allowed to stand at room temperature for 15 min. The homogeneity of the parasite monolayer was checked by light microscopy. Each measuring cycle consisted of a mixing time of 30 s with a data acquisition period of 2 min. The O2 consumption rate was measured before and after the addition of oligomycin (5 µM), FCCP (1 µM) and AA plus ROT (1 µM each), and the different parameters were calculated as described above.
Superoxide radical production in H2O2-treated epimastigotes
2-OH-ethidium HPLC detection
Mitochondrial O2·− production after H2O2 treatment was evaluated in wt and/or Fe-SODA overexpressers by the HPLC detection of the specific product 2-OH-E+ . Briefly, epimastigotes (3 × 108 cells) were preloaded with DHE (50 µM, Invitrogen) before H2O2 treatment for 30 min at 28°C and washed three times in PBS in order to eliminate all non-incorporated probe. Parasites were then incubated with H2O2 as described above and, after 3 h, cells were lysed in Triton X-100 (0.1%, v/v) in PBS (pH 7.4) by passing 100 times by a syringe (27Gx1/2″ 0.4 × 13 mm). Lysates were then mixed with an equal volume of acetonitrile and vortex-mixed, and proteins precipitated for 2 h at 4 °C. The samples were centrifuged at 16 000 g for an hour at 4°C, and the organic phase was removed and dried with a rotary evaporator (Rapid Vap, LABCONCO). Dried samples were then resuspended in 20 µl of sample buffer [90% water, 10% ACN and 0.1% trifluoroacetic acid (TFA)], and 2 µl was analyzed by HPLC using a Supelco Ascentis Express Phenyl-Hexyl column (5 cm × 4.6 mm, 2.7 µm, Sigma). Samples were eluted isocratically with a mobile phase containing 65% water, 35% ACN and 0.1% TFA. 2-OH-E+ was generated incubating DHE (50 µM) in phosphate buffer (50 mM, pH 7.4) for 1 h with xanthine (200 µM), xanthine-oxidase (50 mU/ml) and catalase (0.2 mg/ml) at room temperature, and the concentration was calculated using the extinction coefficient at 470 nm (1.2 × 104 M−1 cm−1) as previously reported [29,30]. The parasite 2-OH-E+ concentration at the different experimental conditions was calculated using a standard curve for 2-OH-E+ (0–2 µM). The products were analyzed by HPLC with fluorometric detection as above .
Parasites were treated or not as above with H2O2 for 10 min and washed three times, and parasites were resuspended in PBS at the same concentration. After 3 h, the total parasite extract was obtained by three cycles of freeze–thawing in 0.5 ml of Tris–HCl (50 mM, pH 7.5) containing MnCl2 (0.6 mM) and fluorocitrate (2 µM). After cell lysis, debris was removed by centrifugation at 13 000×g for 30 min at 4°C. Aconitase activity was measured in the supernatants following absorbance at 240 nm in the presence of cis-aconitate (0.5 mM). Activity is expressed as a percentage of control conditions (no H2O2 treatment).
All experiments were performed at least three times on independent days. Results are expressed as mean ± standard error of the mean. Student's t-test was performed for comparison between two groups and the ANOVA was used for comparison between more than two groups. For post hoc analysis, the least significant difference method was performed. A value of P < 0.05 was considered significant.
Mitochondrial dysfunction and oxidative stress in T. cruzi-infected cardiomyocytes
We evaluated the mitochondrial function and H2O2 production of T. cruzi-infected cardiomyocytes, extending previous data  under our specific experimental conditions. Cardiomyocyte mitochondrial function and the generation of oxidants (specifically H2O2) were evaluated in primary cardiomyocytes after T. cruzi infection (24–48 h, parasite : cell infection ratio; 5 : 1). By using JC-1, mitochondrial membrane potentials in control and infected cardiomyocytes were assayed by epifluorescence microscopy. CL-Brener-infected cardiomyocytes (24 h) have a drop in the red mitochondrial fluorescence (intra-mitochondrial J-aggregates) when compared with non-infected cultures, increasing its cytoplasmic green fluorescence (J-monomers). This indicates a rapid drop in the mitochondrial membrane potential following T. cruzi infection as was previously shown (Figure 1A) [8,31]. Cardiomyocyte mitochondrial dysfunction was also evaluated 24 h post infection using the SeaHorse analyzer before and after the addition of drugs (Oligomycin, FCCP and AA-Rot) (Figure 1B). The respiratory control ratio (RCR), calculated as the oxygen consumption in the presence of FCCP divided by the oxygen consumption in the presence of oligomycin, was used as an indicator of mitochondrial dysfunction in cardiomyocytes [26,32]. A drop in the RCR was observed for infected cardiomyocytes, being more pronounced for cardiomyocytes infected with Fe-SODA overexpressers with respect to the wt CL-Brener strain (Figure 1C). Overall, the above data indicate that after 24 h of infection, mitochondrial dysfunction in primary cardiomyocytes is observed in our experimental conditions. Cardiomyocyte (H9c2 and/or primary cardiomyocytes) oxidant production was evaluated by measuring H2O2 in the culture supernatant of control and/or infected cultures, stimulated or not with pro-inflammatory cytokines. For both cardiomyocytes, cytokine stimulation enhances H2O2 production and this was further increased by T. cruzi infection (Figure 2A,B, respectively). No differences were observed in the rate of primary cardiomyocyte H2O2 production 48 h post infection with CL-Brener or Fe-SODA overexpressers, or cytokine treatment (Supplementary Figure S1). Finally, the establishment of an oxidative stress in the cardiomyocyte culture may be accompanied by the activation of the transcription factor FoxO3 and the enhanced expression of its target genes. FoxO3 activation and its nuclear translocation were evaluated in T. cruzi-infected cardiomyocytes by immunofluorescence microscopy. Twelve hours post infection, a clear nuclear localization of FoxO3 was evident in T. cruzi-infected and/or -stimulated cardiomyocytes (Figure 3A). In agreement with the FoxO3 translocation to cell nuclei, an enhanced protein expression of the mitochondrial MnSOD and peroxiredoxin-3, typical target genes for FoxO3, were observed in the T. cruzi-infected cultures (Figure 3B).
Mitochondrial function in T. cruzi-infected primary cardiomyocytes.
H2O2 generation by T. cruzi-infected and pro-inflammatory cytokine- stimulated cardiomyocytes.
FoxO3 activation and expression of target genes in infected primary cardiomyocytes.
Cardiomyocyte control of parasite proliferation through the induction of amastigote PCD
The above data indicate that, in our experimental conditions, oxidative stress was established in cardiomyocytes infected with T. cruzi with the generation of H2O2 that could reach the intracellular amastigote, controlling parasite proliferation. To evaluate this hypothesis, we first conducted infection studies using primary cardiomyocytes in the absence and presence of pro-inflammatory cytokines. After 48 h, stimulated cardiomyocytes were able to partially control wt CL-Brener infection unlike unstimulated cardiomyocytes, but failed to do so for T. cruzi Fe-SODA overexpresser-infected cultures (Figure 4A). The increase in parasite survival and, thus, in the number of intracellular amastigotes, may be the cause of the decreased RCR found in the cardiomyocyte cultures infected with Fe-SODA overexpressers with respect to CL-Brener (Figure 1C).
Cardiomyocyte control of parasite proliferation through induction of intracellular amastigote PCD.
To test if ·NO could be a candidate for the induction of amastigote PCD controlling parasite proliferation, we exposed infected cardiomyocytes (H9c2) to physiologically relevant fluxes of ·NO using the NO-donor NOC-18 (0.5 mM), which produce 0.15 µM ·NO/min. Similar to what was observed for pro-inflammatory cytokine-treated primary cardiomyocyte cultures, H9c2 exposed to ·NO was able to control parasite proliferation of wt CL-Brener but failed to control proliferation of T. cruzi Fe-SODA overexpressers (Figure 4B).
We have previously shown, using fresh human serum as the death stimuli, that overexpression of mitochondrial Fe-SODA renders T. cruzi resistant to the induction of PCD in the non-infective epimastigote stage of the parasite [15,16]. Since the above results also show that these overexpressers are more resistant than wt parasites in cardiomyocyte infections, we searched for intracellular parasite PCD. For this purpose, we purified intracellular amastigotes to homogeneity from infected cardiomyocyte cultures (Figure 4C) and subjected them to TUNEL staining as a hallmark of PCD. Purified wt CL-Brener amastigotes from pro-inflammatory stimulated cardiomyocyte cultures showed an increase in the number of TUNEL-positive amastigotes (15 ± 5%) compared with unstimulated conditions (Figure 4D), in agreement with its lower proliferation rate. Moreover, a similar increase in TUNEL-positive parasites was also observed in intra-cardiomyocyte amastigotes exposed to ·NO when compared with control conditions (Figure 4D), indicating the ability of ·NO to induce intracellular amastigote PCD. We have previously shown, in the non-infective epimastigote stage, that ·NO can inhibit parasite oxygen consumption due to the inhibition of the mitochondrial electron transport chain . To evaluate if this was also the case in the infective amastigote stage, we exposed purified amastigotes to ·NO (0.5 mM NOC-18; 3 h) and evaluated MitoSOX oxidation by flow cytometry . An increase in MitoSOX oxidation was observed, indicating that, as was previously shown for epimastigotes, ·NO can enhance intra-mitochondrial O2·− generation in the amastigote stage (Figure 4E). Finally, parasites overexpressing Fe-SODA were more resistant to intra-cardiomyocyte PCD induction, highlighting the role of mitochondrial-derived O2·− and/or peroxynitrite in the signaling of the T. cruzi PCD process (Figure 5).
T. cruzi Fe-SODA overexpressers are more resistant to ·NO-induced PCD.
H2O2 induction T. cruzi PCD: parasite mitochondrial superoxide radical as the proximal redox mediator
The results presented above show the ability of stimulated cardiomyocytes to control parasite proliferation. To evaluate the ability of H2O2 to induce PCD in T. cruzi, we conducted experiments in the epimastigote stage of the parasite. Wt CL-Brener parasites (3 × 108 cells ml−1) were exposed to H2O2 (300 μM for 10 min at 28°C) in PBS (pH 7.4). Immediately after oxidant treatment, parasites were washed and incubated in complete medium (supplemented BHI) for different time periods (3–24 h). As early as 3 h after H2O2 addition, a drop in the mitochondrial membrane potential (ΔΨ) together with phosphatidylserine (PS) externalization to the outer leaflet of the plasma membrane was evident with the maintenance of parasite plasma membrane integrity (Figure 6A–C). After 24 h, 30% of the wt population was positive for TUNEL (Figure 6D). These results are in agreement with previous data showing PARP activation and apoptotic death in T. cruzi epimastigotes exposed to H2O2 in the same experimental conditions . Parasites overexpressing Fe-SODA were resistant to the H2O2 death stimuli as assayed by TUNEL staining (Figure 6E,F). Parasite mitochondrial dysfunction was further evaluated using SeaHorse XF-24 analyzer, 4 h after oxidant treatment . A representative experiment is shown for wt parasites after the additions of the different compounds (Figure 7A). Four hours after H2O2 treatment, the RCR value for wt CL-Brener epimastigotes dropped significantly, while for parasites overexpressing mitochondrial Fe-SODA, there was no change in the RCR, further supporting the resistance of these overexpressers to mitochondrial damage (Figure 7B) .
H2O2 induction of PCD in T. cruzi epimastigotes.
T. cruzi epimastigote mitochondrial function after H2O2 treatment.
Induction of PCD in T. cruzi epimastigotes by fresh human serum leads to the generation of mitochondrial O2·− production . To evaluate if mitochondrial dysfunction in H2O2-treated epimastigotes is accompanied by O2·− generation, we evaluated the specific oxidation product of dihydroethidium (DHE), 2-OH-ethidium+ (2-OH-E+), by analytical HPLC-detection techniques as recently described (Figure 8A) . A representative chromatogram is shown for wt CL-Brener parasites before and after H2O2 exposure. As was previously shown for fresh human serum , an increase in 2-OH-E+ detection was observed 4 h after H2O2 treatment, indicating the enhanced generation of mitochondrial O2·− (Figure 8A,B). Importantly, in parasites overexpressing Fe-SODA, there was no significant increase in 2-OH-E+ levels after H2O2 treatment, with basal levels lower than that observed for wt CL-Brener, indicating the ability of this parasite to detoxify O2·− before PCD induction (Figure 8B). Finally, the activity of the O2·−-sensitive enzyme aconitase was measured, as a parasite target for O2·− (Figure 8C). After 3 h of H2O2 treatment, a significant inactivation of total aconitase activity (∼20 ± 2%) was observed for wt parasites, while the activity was not affected in the Fe-SODA overexpressers (Figure 8C). Mitochondrial aconitase activity represents 60% of total parasite aconitase activity and, thus, inactivation of the mitochondrial aconitase under these experimental conditions must be larger than 20%, possibly >40% . The above results support the concept that mitochondrial O2·− generation may be a common feature in the redox signaling of T. cruzi PCD .
Superoxide radical generation by T. cruzi H2O2-treated epimastigotes.
During the chronic stage of CD, T. cruzi persists in the infected tissues at low numbers for years. Host immune response controls parasite replication as evidenced by the reactivation of parasite proliferation in immunosuppressed individuals . The control of parasite densities in CD is essential for maintaining parasite population at low quantities in the infected tissue, safeguarding host survival and thus parasite persistence. The mechanisms by which cardiomyocytes control parasite proliferation are not fully understood but may involve the secretion of immune mediators together with ·NO generation, either by cardiac, endothelial or immune cells infiltrated in the infected tissue. T. cruzi infection in cardiomyocytes can elicit the generation of pro-inflammatory cytokines that can control intracellular amastigote proliferation [7,35]. Enhanced cytokine synthesis by the infected cardiomyocyte may be responsible for the increase in oxidant generation due to cardiomyocyte mitochondrial dysfunction and probably also by NOX-4 activation [7,8,32,36]. In agreement with previous reports, we show that T. cruzi (CL-Brener strain) infection in primary cardiomyocytes elicits an early mitochondrial dysfunction (Figure 1) with increased generation of H2O2 (Figure 2). This was corroborated by our results showing the activation of the transcription factor FoxO3, known to be activated upon cardiomyocyte induction of oxidative stress [37,38], as well as the increased expression of its target mitochondrial antioxidant enzymes, MnSOD and Prx-3 (Figure 3).
Depending on their levels, oxidants (specifically H2O2) can promote cell redox signaling or cytotoxicity . In T. cruzi, H2O2 has been shown to modulate intracellular development, either boosting or compromising amastigote proliferation [16,32,40,41]. In our experimental conditions, pro-inflammatory stimulated cardiomyocytes were able to partially control wt (CL-Brener) intracellular amastigote proliferation, but failed to control proliferation of T. cruzi mitochondrial Fe-SODA overexpressers (Figure 4A). These results indicate that in a pro-inflammatory milieu, the enhanced generation of oxidants will be able to induce parasite PCD. The participation of mitochondrial calcium overload and O2·− generation in the induction of T. cruzi PCD was previously shown in the non-infective epimastigote stage [15,16]. To evaluate if this could also be the case for intracellular amastigotes, we searched for parasite PCD in stimulated cardiomyocyte cultures. Intracellular amastigotes were purified (Figure 4C) and PCD evaluated by TUNEL staining. In fact, control of wt parasite proliferation in pro-inflammatory stimulated cardiomyocytes was accompanied with an increase in the number of intracellular apoptotic amastigotes (Figure 4D). To explore the possibility that ·NO could also trigger intracellular amastigote PCD, we exposed infected cardiomyocyte cultures to controlled physiological fluxes of ·NO (0.15 µM/min), detecting an increase in the number of apoptotic wt CL-Brener amastigotes (Figure 4D). The above data indicate that both mediators can diffuse and reach the intracellular amastigote, leading to the induction of parasite PCD. External fluxes of ·NO were used due to the low cardiomyocyte iNOS induction observed under our experimental conditions  and considering that, in the infected cardiac tissue, there will be other ·NO sources. Interestingly, T. cruzi overexpressing mitochondrial Fe-SODA were more resistant to the induction of PCD even in the presence of extracellular ·NO fluxes (Figure 5). This highlights the role of parasite intra-mitochondrial O2·− generation in the initiation phase of the death process as was previously shown for the epimastigote stage [9,16]. To further sustain this idea, an enhancement in MitoSOX oxidation was also observed in ·NO-treated wt amastigotes, strongly suggesting intra-mitochondrial O2·− (and possibly also peroxynitrite) generation (Figure 4E). The presence of intracellular T. cruzi apoptotic amastigotes was previously reported both in vitro and in vivo experimental T. cruzi infections, but the host mediators responsible for the induction of the parasite PCD were not identified [22,23,43]. Herein, and using primary cardiomyocytes, we show the ability of two host-derived diffusible redox mediators, H2O2 and ·NO, to trigger PCD in the intracellular amastigote stage of the parasite (CL-Brener strain). H2O2 also triggered PCD in the epimastigote stage of the parasite with early loss (3 h) of the mitochondrial membrane potential, which was accompanied by phosphatidylserine externalization to the outer leaflet of the plasma membrane (Figure 6A,B). Additionally, 24 h later, TUNEL-positive parasites were observed with the maintenance of plasma membrane integrity (Figure 6D,C) all of them hallmarks of the PCD process. Mitochondrial function was evaluated in H2O2-treated epimastigotes by the SeaHorse analyzer and a significant drop in the RCR value was observed after 4 h of oxidant treatment, indicating parasite mitochondrial dysfunction (Figure 7A,B). In these experimental conditions, an increase in O2·− generation was observed by the specific analytical detection of 2-OH-Et+ (Figure 8A,B). Importantly, Fe-SODA overexpressers were resistant to H2O2 PCD induction with no significant detection of O2·− generation (Figure 8B). The fact that Fe-SODA overexpressers were more resistant clearly shows the participation of parasite mitochondrial O2·− generation in the induction of the death process. Moreover, O2·− detoxification by Fe-SODA will also limit peroxynitrite generation in the presence of ·NO and thus prevent mitochondrial damage.
The cellular targets and signaling events provoked by O2·−, either in mammalian or parasite cells, are still unknown but their contribution in the initiation phase of the cell death process is well recognized [15,44,45]. In biological systems, O2·− can act as a univalent oxidant, with the family of the [4Fe-4S] clusters containing dehydratases being its preferential targets . Oxidation of the cluster leads to Fe release with concomitant enzyme inactivation and the plausible reaction with H2O2 with the potent oxidant hydroxyl radical (·OH) capable of producing site-specific oxidative damage to parasite mitochondrial DNA [47,48]. Moreover, in the presence of ·NO, peroxynitrite will be formed in the parasite mitochondria, producing oxidative modifications to proteins, lipids and DNA [49,50]. We have previously shown that PCD in T. cruzi induced by FHS leads to an O2·−-dependent aconitase inactivation and that Fe-SODA overexpressers attenuated enzyme inactivation . Herein, and using H2O2 as the PCD inducer, aconitase inactivation was also observed (Figure 8C), in a Fe-SODA-resistant manner, underscoring the role of mitochondrial O2·− in the early signaling of parasite PCD.
Redox responses and outcome during cardiomyocyte infection may differ with different T. cruzi strains, which are largely variable in infectivity and tissue tropism [32,51]. The proposed sequence of events after CL-Brener T. cruzi invasion/infection to cardiomyocytes and the impact of parasite mitochondrial Fe-SODA contents in the balance between cell death and proliferation are illustrated in Figure 9.
Cardiomyocyte control of parasite proliferation.
Overall, the results present here shed light in some of the possible mechanisms that may control intra-cardiomyocyte T. cruzi amastigote proliferation, identifying H2O2 and/or ·NO as redox mediators of parasite PCD. The fact that T. cruzi Fe-SODA overexpressers were more resistant to PCD induction underscores the importance of the Fe-SOD as part of an antioxidant-based immune evasion mechanism either by O2·−detoxification and/or limiting peroxynitrite formation in this organelle, therefore probably contributing to parasite survival and virulence during the chronic stage of CD.
fetal bovine serum
Hanks' balanced salt solution
inducible nitric oxide synthase
programmed cell death
programmed cell death
respiratory control ratio
D.E., R.R. and L.P. designed and wrote the manuscript. D.E., G.S., A.M., P.P.S., B.H. and L.P. performed this study. D.E., L.O.A., R.R. and L.P. analyzed the study.
This work was supported by grants from Universidad de la República (Comisión Sectorial de Investigación Científica and Espacio Interdisciplinario), Uruguay, Universal FAPEMIG (Fundação de Amparo à Pesquisa do Estado de Minas Gerais) [fAPQ-02269-14], Programa de Apoio a Núcleos de Excelência FAPEMIG [APQ-01419-14], Brazil to L.P. and R.R. Additional supported was obtained from Programa de Desarrollo de las Ciencias Básicas (Uruguay). D.E., G.S. and A.M. were partially supported by fellowships of Agencia Nacional de Investigación e Innovación, Comisión Académica de Posgrado and Comisión Sectorial de Investigación Científica (Uruguay).
The Authors declare that there are no competing interests associated with the manuscript.