Phenothiazinium Dyes Are Active against Trypanosoma cruzi In Vitro

Chagas disease is a tropical illness caused by the protozoan Trypanosoma cruzi. The disease affects populations of the Americas and has been spread to other continents due to the migration process. The disease is partially controlled by two drugs, Benznidazole and Nifurtimox. These molecules are active in the acute phase of the infection but are usually ineffective during the symptomatic chronic phase. Several research groups have developed novel candidates to control Chagas disease; however, no novel commercial formulation is available. In this article, we described the anti-T. cruzi effects of phenothiazinium dyes in amastigote and trypomastigote forms of the parasite. Methylene Blue, New Methylene Blue, Toluidine Blue O, and 1,9-Dimethyl Methylene Blue inhibited the parasite proliferation at nanomolar concentrations and also demonstrated low toxicity in host cells. Moreover, combinations of phenothiazinium dyes indicated a synergic pattern against amastigotes compared to the Benznidazole counterparts. Phenothiazinium dyes levels of reactive oxygen species (ROS) and decreased the mitochondrial potential in trypomastigotes, indicating the mechanism of action of the dyes in T. cruzi. Our article offers a basis for future strategies for the control of Chagas disease using low-cost formulations, an important point for endemic underdeveloped regions.


Introduction
Trypanosoma cruzi is the etiologic agent of the Chagas disease, endemic in the Americas. Chagas disease afflicts ∼ 6 million to 7 million people and nowadays has spread to North America, Europe, Asia, and Oceania due to migration. The disease has two distinct phases; the acute phase is asymptomatic or causes unspecific symptoms. The chronic phase, when symptomatic, may lead to cardiac and/or digestive degeneration [1].
Despite the effort of several groups to develop novel therapies [2], there are only two commercial drugs for Chagas disease control, Benznidazole and Nifurtimox [3]. Moreover, several cases of treatment failure have been described [4], demanding new active and safe drugs. In this article, we investigated the anti-T. cruzi effect of phenothiazinium dyes, a low-cost family of drugs with potential against malaria [5]. Methylene Blue (MB), the most used phenothiazinium dye, has been applied against malaria since the XIX century, representing the first report of treatment with a synthetic molecule [6]. Due to the reversible side effects (green sclera and urine), MB was replaced by alternative antimalarial molecules, such as quinine, artemisinin, and chloroquine [7]. Nowadays MB has been revived for control of malaria, cancer, cyanide poisoning, and methaemoglobinaemia and recently in the treatment of patients with Alzheimer's disease [8].
In this study, we evaluated the potential of the phenothiazinium dyes for the control of T. cruzi in vitro. We also determined several synergic formulations composed by phenothiazinium dyes and/or Benznidazole. ROS production and mitochondrial activity in treated cultures were assayed by flow cytometry, in order to investigate some aspects of the mechanism of action of phenothiazinium dyes in T. cruzi. The screening of low-cost compounds is a key factor for the control of the Chagas disease in endemic regions, most of them being in a delicate economic state.

Material and Methods
2.1. T. cruzi Culture. T. cruzi (Tulahuen strain) was maintained in cultures of LLCMK2 cells and cultivated in RMPI supplemented with 10% fetal bovine serum (FBS) at 37 ∘ C, 5% CO 2 . The parasites used in proliferation and flow cytometry assays were genetically modified to express the enzymegalactosidase (T. cruzi-LacZ [9]). To illustrate the inhibition of parasite proliferation, a T. cruzi expressing GFP (G-GFP strain [10]) was used.

Fluorescence
Microscopy. LLCMK2 cells (2 × 10 3 cells per well) were cultivated in 96-well, black, flat-bottomed plates for 6 h, 37 ∘ C, and 5% CO 2 . The plate was washed with PBS and the cells were infected with T. cruzi (G-GFP strain) at a multiplicity of infection (MOI) of 5:1 (1×10 5 cells/ml). The cultures were incubated for 24 h, 37 ∘ C, 5% CO 2, and the wells washed 3 times with PBS in order to remove noninternalized parasites. Subsequently, IC 50 concentrations of MB, NMB, TBO, DMMB, and BZ were added to the cultures and the plates were incubated for 72 h, 37 ∘ C, and 5% CO 2 . Infected and nontreated and noninfected cells were used as negative controls and positive controls, respectively. After the incubation, the plates were fixed with 4% paraformaldehyde for 20 minutes, followed by washing with PBS. The nucleus of host cells was stained with 0.25 M 4 耠 ,6-diamidine-2 耠phenylindole dihydrochloride (DAPI) (Ex/Em = 340/488 nm) and the GFP T. cruzi parasites were detected by the green florescence (Ex/Em = 488/510 nm). The images were acquired in an Image Xpress Micro XLS Widefield High-Content Analysis System from Molecular Devices. The system also calculated the percentage of inhibition compared to the nontreated control. This was done using the number of intracellular amastigotes in twenty-five images obtained for each well.
2.5. Cytotoxicity. The cytotoxic effects of the phenothiazinium dyes in LLCMK2 cells was measured by Thiazolyl Blue Tetrazolium Bromide (MTT) assay, as described in [12]. A suspension of LLCMK2 cells (5 × 10 4 /ml) in RPMI supplemented with 10% FBS was distributed in 96-well plates and cultivated (37 ∘ C, 5% CO 2 ) to the confluence. The media were discarded and serial dilutions of the compounds were added (starting from 100 M) in duplicate. The cultures were incubated for 72 h, 37 ∘ C, and 5% CO 2 . For the measurement of the cellular esterase activity, the media were carefully removed and an MTT solution (500 g/ml MTT in PBS) was added. The reaction was performed for 4 h at 37 ∘ C, 5% CO 2 . After the reaction, the MTT solution was discarded and the formazan crystals solubilized with DMSO. The absorbance was measured in an ELISA reader (Synergy6 H1, Biotek) at 570 nm.

ROS Measurement and Evaluation of Mitochondrial
Membrane Potential Status. LLCMK2 cells in 75 cm 2 flasks were cultivated to confluence and were infected with 1 × 10 7 T. cruzi trypomastigotes. The cultures were cultivated for 6 days when free trypomastigotes were observed in the supernatant. The infected cells were centrifuged for 10 min, 3000 g, and washed with PBS and suspended in trypsin to separate cells and parasites. The cultures were distributed in 1.5 ml microtubes (∼ 1 × 10 6 trypomastigotes/tube) and incubated with 4 × IC 50 of compounds at 37 ∘ C, 5% CO 2 for 2 h. Menadione (MN), a classic inducer of cellular oxidative stress [13], was used as the positive control of ROS production as previously described in [14]. After treatment, the cultures were centrifuged for 10 minutes, 3000 g, and washed with PBS and suspended in trypsin. The ROS measurement was performed after incubation with 5 M 2,7-dichlorofluorescin diacetate (DCFDA) for 15 minutes at room temperature in the dark. Similarly, the mitochondrial membrane potential was determined after incubation with 5 M JC-1 under the same conditions. JC-1 is a cationic dye that accumulates in active mitochondria. In low concentrations, the dye exhibits a green fluorescence. In polarized mitochondria, the dye forms J-aggregates, emitting red fluorescence. The cultures were washed with PBS and analyzed in a BD FACS-Canto (BD Biosciences) flow cytometer with the FACSDiva (BD) 6.1.3 software. Due to the size, host cells (infected and noninfected) and trypomastigotes were separately analyzed by flow cytometry [11]. The median of fluorescence of oxidized DCF and the populations with active mitochondria were measured with excitation/emission at 488 nm/529 nm and 488 nm/529 nm, respectively. The percentage of DCF fluorescence or cells with polarized mitochondria was calculated in relation to the respective nontreated controls.

Statistical
Analysis. The percentages of proliferation and cytotoxicity were calculated using the mean absorbance of the compound-free control and the absorbance of each treatment. The IC 50 and the combination index (CI) were achieved using Compusyn software (http://www.combosyn.com/) and the selective index (CC 50 /IC 50 ) was also determined. From CI values, synergistic (CI < 1), and antagonistic (CI > 1) interactions were identified [15]. For oxidative stress and mitochondrial activity, data were analyzed by one-way ANOVA followed by a Dunnett's post hoc test (compared to nontreated groups) on GraphPad 5.0 software.  (Table 1). Moreover, a similar inhibition pattern between BZ and phenothiazinium dyes was observed using high-content screening analysis by amastigotes counting (Figure 1). At IC 50 concentrations (Table 1)  Despite the relative higher host cell toxicity, the selective index (SI) was above 25, which is adequate compared to the index recommended for candidate drugs for Chagas disease control. In drug screening procedures against T. cruzi, an SI > 10 is recommended for candidates with potential for use in in vivo and clinical assays [16,17]. Moreover, MB has been successfully applied against malaria and to treat methemoglobinemia with few cases of toxicity [18,19], reinforcing its potential use as anti-T. cruzi compound in further assays.

Phenothiazinium Combinations Are Synergic against
Amastigotes In Vitro. The combination of compounds has an important potential for the control of the Chagas disease. Several combinations of BZ with Itraconazole, E1224 (ravuconazole prodrug), and ketoconazole have demonstrated improved anti-T. cruzi pattern compared to the single treatments [20][21][22]. The main aim of combinations with BZ is the decrease of dosages allied to the alleviation of side effects, which allow longer-term treatments. However, the combinations of phenothiazinium dyes with BZ resulted in an antagonist effect (CI > 1), despite the elevation of the parasite inhibition, mostly at lower concentrations (Figures 2(a), 2(b), 2(c), 2(d), and Table 2). However, the combinations among phenothiazinium dyes were synergic against amastigotes (except MB + DMMB), improving the parasite inhibition in all concentrations analyzed (Figures 2(e), 2(f), 2(g), 2(h), 2(i), and 2(j)). The CI of MB + NMB, MB + TBO, MB + DMMB, NMB + TBO, NMB + DMMB, and TBO + DMMB were 0.90, 0.89, 1.05, 0.74, 0.80, and 0.42, respectively ( Table 2). The CI values probably indicate some differences in mechanisms of action among the phenothiazinium dyes. For example, the combinations containing TBO demonstrated the lowest CI values, probably due to low structure similarity with other dyes. In contrast, MB and DMMB are structurally similar, which reflected in an antagonist pattern when combined (Table 2). However, TBO demonstrated a high IC 50 (compared to NMB or DMMB), which may affect negatively further in vivo treatments. Therefore, a wide and complex strategy for in vivo procedures is demanded, once there are multiple combinations and treatment regimens (acute, chronic, and pregnant models) to test.

Phenothiazinium Dyes Increase ROS in Treated Cells and
Trypomastigotes. The literature is contradictory regarding the role of ROS produced by the host cell during T. cruzi infection. Several studies have implicated the host cell respiratory burst in controlling the infection [23,24]. Others have reported a positive correlation between ROS production by the host and parasite proliferation [25,26]. Also, when a sublethal hydrogen peroxide concentration was added to a T. cruzi infected culture, parasite proliferation increased [27]. Our results show increased ROS production in T. cruzi infected cells (untreated) when compared to noninfected cells (Figure 3(a)), corroborating an effect also observed for proliferating bacteria [28], fungi [29], viruses [30,31], and Leishmania [32]. Despite the efficient antioxidant machinery [33], we observed increased ROS production in T. cruzi trypomastigotes after treatment with phenothiazinium dyes (Figure 3(b)). This effect was similar to the one observed for MN a classic inducer of oxidative stress [13]. On the other hand, no ROS production was detected in trypomastigotes treated with BZ. Although BZ induces the production of free radicals and electrophilic metabolites after the reduction of its nitro group, no ROS is produced in treated parasites [34]. We have observed that treatment with phenothiazinium dyes (and to a lesser extent BZ and MN) increased ROS production in noninfected cells (Figure 3(b)). In contrast, treating T. cruzi infected cells with phenothiazinium dyes decreased ROS production (Figure 3(b)). Indeed, Methylene Blue scavenges ROS in several biological models such as skin and HT-22 cells, elevating oxygen consumption and mitochondrial activity [35][36][37]. Because T. cruzi proliferation is stimulated by ROS production [27], we speculate that scavenging of such ROS by phenothiazinium dyes impairs parasite proliferation. However, for the complete elucidation of the roles of phenothiazinium dyes in ROS scavenging, complementary assays will be required in further studies. As an example, the ability of phenothiazinium dyes to scavenge   , and the proliferation measured after CPRG assay. As references, isolated compounds of each combination were evaluated concomitantly in the same plate. The inhibition percentage was calculated in comparison to the nontreated group and the IC 50 and CI values were achieved using Compusyn software.

Mitochondria of Trypomastigotes
Are Sensitive to Phenothiazinium Dyes. All noninfected and infected LLCMK2 cells are positive to JC-1 accumulation, indicating the presence of active mitochondria. However, 22% of trypomastigotes were negative to mitochondrial polarization (Figure 4(a)). Moreover, phenothiazinium dyes decreased mitochondrial membrane potential in trypomastigotes, whereas the infected and noninfected cells were unaffected by incubation with the compounds (Figure 4(b)). Also, BZ and MN displayed distinct effects on the mitochondrial membrane potential of trypomastigotes. The former did not affect mitochondria whereas the latter reduced the membrane potential (Figure 4(b)). It was previously observed that BZ elevates free radical production and causes DNA damage without affecting mitochondria [40]. The effect of phenothiazinium dyes on trypomastigotes was similar to that observed for MN (Figure 4(b)), a classic mitochondrial inhibitor [41]. Thus, the increment of ROS observed in trypomastigotes under treatment with dyes may be a consequence of mitochondrial depolarization. Moreover, decreased mitochondrial activity may be also related to the ROS impairment production in infected cells treated with phenothiazinium dyes. Drugs that alter the mitochondrial activity usually lead to loss of parasite ATP production and induction of apoptosis [42]. For example, MB improves mitochondrial respiration in cells [35,43], decreasing ROS in cardiac tissue from diabetic rats [44].
We have observed that active and synergic combinations of phenothiazinium dyes are effective in killing T. cruzi. Our results also indicate the potential of phenothiazinium dyes as  candidates for the control of T. cruzi. We speculate that phenothiazinium dyes scavenge ROS during T. cruzi infection thus reducing parasite proliferation. However, future assays will be necessary to elucidate the mechanism. Furthermore, the induction of apoptosis by disruption of mitochondrial polarity should also be considered as a potential mechanism operating in conjunction with the ROS scavenging by phenothiazinium dyes. The present article opens the perspective for the development of low-cost and active anti-T. cruzi molecules, pivotal characteristics for the control of Chagas disease in low-middle-income endemic regions.

Data Availability
The data used to support the findings of this study are included within the article.

Conflicts of Interest
The authors declare that they have no conflicts of interest.