INTRODUCTION

Trypanosoma cruzi (T. cruzi) is a flagellated protozoan from the Kinetoplastida order (^{ElSayed et al., 2005}). It is the causative agent of American trypanosomiasis, a vector-borne parasitic disease endemic to South America, Central America and Mexico. T. cruzi represents a major public health problem, with approximately 8 million people infected and 25 million at risk of infection (^{Ibáñez-Cervantes et al., 2018}). In 3-5% of acutely affected patients, infection may result in several conditions: 1) myocarditis, which may progress to heart failure and death; 2) destructive lesions of cardiomyocytes and cardiac nerves, which may also progress to heart failure; or 3) meningoencephalitis (^{James et al., 2005}). In chronically infected patients, the disease may go unnoticed for several years, until 20-35% of these patients develop irreversible lesions of the autonomic nervous system. Chagas disease represents the leading cause of cardiac lesions in young, economically productive adults in endemic countries (^{El-Sayed et al., 2005}; ^{James et al., 2005}).

Currently, treatment options are limited and no vaccine is available to prevent or control T. cruzi infection. The drugs approved by WHO for the treatment of Chagas disease, such as nifurtimox (NFMX) and benznidazole (BNZ), have at least four disadvantages (^{Colantonio et al., 2016}). First, due to the harsh side effects of these drugs, about 40% of patients drop out of treatment. Secondly, these drugs work most effectively during the acute phase of the infection; however, as it is difficult to detect, most people start drug treatment until the chronic phase of the disease has been established, with a consequent reduction in the antiparasitic efficacy of the drugs (^{Colantonio et al., 2016}). Third, public access to therapeutic drugs is limited (^{Chaves et al., 2017}). Fourth, some T. cruzi strains are resistant to NFMX and BZN, which makes their efficacy suboptimal (^{Toledo et al., 2004}). These factors emphasize the need to develop better anti‒T. cruzi drugs and/or new treatment strategies for American trypanosomiasis.

Current medical trends include nanomaterials, especially noble metals such as gold and silver. These materials have had biomedical applications since ancient times (^{Ge et al., 2014}). Recent research regarding gold nanoparticles has been oriented to the treatment of cancer, while silver nanoparticles (AgNPs) have been used to fight pathogens (bacteria, fungi, viruses and protozoa) (^{Ge et al., 2014}).

Crystalline silver nanostructures are available in many shapes and sizes (^{Sun et al., 2002}) that can be used as medical treatments (^{Dankovich et al., 2011}; ^{Dubas et al., 2006}; ^{Kokura et al., 2010}; ^{Lee et al., 2006}), but the effect of silver nanoparticles on parasitic protozoa is still incompletely understood (^{Mao et al., 2016}). In this sense, two studies have reported that traditional method to synthesize the AgNPs may have cytotoxic properties against the protozoa pathogens Leishmania tropica and Toxoplasma gondii (^{Allahverdiyev et al., 2011}; ^{Adeyemi et al., 2017}) and although recently ^{Brito et al., 2020} biosynthesized AgNPs from corn cobs containing xylans were able of causing death of the Y strain of the T. cruzi.

The aim of the present study was to evaluate the cytotoxic potential of AgNPs on trypomastigotes of two T. cruzi biotypes in vitro. To our knowledge, this is the first attempt to define how the dose and structural characteristics of AgNPs affect T. cruzi cultures in vitro, in an effort to find a safe alternative for the treatment of Chagas disease.

MATERIALS AND METHODS

Synthesis of silver nanoparticles

Inorganic silver (AgNO3, ≥ 99.0%), and sodium borohydride (NaBH4, ≥ 99.99%) were purchased from Sigma-Aldrich, USA. All reactions were carried out using deionized water (18MΩ; Millipore). The chosen method induces the dispersion of nanoparticles through a stable colloidal solution, generating a concentration gradient. Sodium borohydride was used as a reducing agent under a cold bath with magnetic stirring (^{Mulfinger et al., 2007}). The AgNP solution (Figure 1) was produced using 0.1 mM AgNO₃/50 mL (H₂O).

AgNPs prepared in aqueous solution facilitated the study of the interaction between the nanoparticles and the parasite, without the intervention of any surfactant or dispersion medium other than deionized water. This was achieved after several pilot syntheses with two reducing agents (citrate and borohydride). The borohydride produced a more stable and concentrated synthesis (0.1 mM AgNPs) that served to achieve the doses of 1-12 µg used by ^{Allahverdiyev et al., 2011} against Leishmania tropica. Sterility of AgNPs was tested by incubating the nanoparticles in sterile Luria Bertani broth (200 µl/L) at 37ºC for two weeks.

Structural characterization of AgNPs

The nanoparticles were characterized by UV-Vis spectroscopy, transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) using a 1% dilution of AgNPs suspended in deionized water (v/v). A Genesys 10S VIS (Thermo Fisher Scientific, Waltham, Massachusetts, USA) was used to perform UV-Vis spectroscopy and verify the characteristic surface plasmon resonance wavelength (400450 nm) of AgNPs (^{Mulfinger et al., 2007}). Aliquots (40 μL) of AgNPs dispersed in methanol (1:5) were placed on 400 carbon-coated copper grids, dried in a desiccator at 25 °C for 24 h, and then observed in a JEM-ARM200F-TEM at 120 kV. To determine the crystallographic direction of AgNPs facets by fast Fourier transform (FFT), interatomic distances were observed in a JEM-ARM200F-TEM at 200 kV (HRTEM) and analyzed with Garantia Microscopy Suite Digital Micrograph Software.

Parasite culture

Trypomastigotes from the T. cruzi strains Sylvio X10/4 (TcI) and Esmeraldo (TcII) were cultured in monolayers of Vero cells, which were kept in DMEM medium supplemented with 2% fetal bovine serum [FBS] and 1% penicillin‒streptomycin (Thermo Fisher Scientific, Waltham, Massachusetts, USA) under controlled conditions (37°C, 5% CO₂, and saturated humidity) (^{Aparicio-Burgos et al., 2015}). As trypomastigotes started bursting out of cells, free parasites were collected and prepared for the AgNP toxicity assay.

Trypomastigotes were placed in 15 mL conical tubes and centrifuged at 704 × g for 7 min. Supernatants were discarded and pellets were re-suspended in 1 mL of fresh DMEM medium. The number of parasites was estimated with a hemocytometer and adjusted to 1×10⁶ per well in a 96-well tissue culture dish for the toxicity assay.

AgNPs toxicity assay for T. cruzi in vitro

Trypomastigotes (1×10⁶ per well) were placed in 96 well plates (Sarstedt, USA) in DMEM supplemented medium and cultured at 37°C, 5% CO₂, and saturated humidity, with different AgNP doses (1, 5, 10 or 12 μg/100 μL). The assay was performed by triplicate with untreated (- Ctrl) or NFMX (400 μg/100 μL, Lampit®, Bayer) treated trypomastigotes as negative and positive controls. NFMX was prepared as described by ^{Rolón et al., 2006}. Briefly, one tablet of the commercial presentation of NFMX (120 mg) was ground in a sterile mortar and pestle and resuspended in 1 mL of dimethyl sulfoxide (DMSO). The final concentration of DMSO in the culture media never exceeded 0.3% (v/v).

Plates were incubated for 24 hours under the same controlled conditions. After drug treatment, parasite and cell viability were estimated by MTS (3-[4,5,dimethylthiazol-2-yl]5-[3-carboxymethoxyphenyl]-2-[4-sulfophenyl]-2H-tetrazolium, internal salt) of CellTiter 96 kit® Aqueous One Solution (Promega, USA), following the manufacturer's instructions, by colorimetry at 490 nm wavelength. For this assay, optical density (OD) values correlated positively with cell viability. All experiments were performed in triplicate. The half-maximal inhibitory concentration (IC₅₀) for each T. cruzi biotype was calculated from the dose-response curve (^{Attarde et al., 2020}; ^{Barile & Hopkinson, 1993}).

In vitro cytotoxic assays in a human keratinocyte cell line (HaCaT)

HaCaT cells (an immortal cell line of non-cancerous human keratinocytes) (2×104/well) were cultured in a 96-well plate (Sarstedt, USA) in supplemented DMEM (2% FBS, penicillin 10,000 units/mL and streptomycin 10,000 µg/mL) and AgNPs at 1, 3, 5, 10 or 12 µg/µL. For each experiment, cells from at least three wells were left untreated (control). The plate was incubated for 48 h under controlled conditions (37 °C, 5 % CO2 and saturated humidity). After treatment, cell viability was estimated by MTS (3[4,5,dimethylthiazol-2-yl]-5-[3-carboxymethylthioxy-phenyl]-2-[4-sulfophenyl]-2Htetrazolium, internal salt) of CellTiter 96 kit® Aqueous One Solution (Promega, USA), following the manufacturer's instructions. The metabolic activity of MTS-treated cells was estimated by colorimetry at a wavelength of 490 nm.

Statistical analyses

ANOVA was used to analyze results from the in vitro viability assay. Mean differences for all assays were assessed using a Tukey test. Differences were considered significant at P < 0.05. Statistical analyses were conducted with the GraphPad Prism 5.0 software package (GraphPad Software Inc., USA).

RESULTS

Characterization of AgNPs

The UV-Vis spectrum of the aqueous silver colloid (Figure 1) showed a strong peak at 396 nm, a typical surface plasmon resonance (SPR) band of AgNPs. The narrow and symmetric absorption peak indicates a homogeneous size distribution of AgNPs. Micrographs of AgNP samples show hemispherical nanoparticles accompanied by some elongated structures of size 52 nm (Figure 2a and 2b). High-resolution transmission electron microscopy (HRTEM) showed that 80% of the particles were decahedrons (Figure 2c) with {111} facets (Figure 2d).

Figure 1 Characterization of AgNPs by UV-Vis Spectroscopy. Absorbance spectra of AgNPs (λmax = 396 nm). The inserted figure displays the photograph of the silver nanoparticles colloid. 

Figure 2 Micrographs show silver nanoparticles produced when borohydride was used as a reducing agent at a concentration of 0.1 mM. (a) Transmission Electron Microscopy (TEM) image of silver nanoparticles with an average diameter of 52 nm; (b) Particles are mostly semi-spherical, and some elongated structures are also observed; (c) High Resolution Transmission Electron Microscopy (HRTEM) showed decahedral particles and (d) Fast Fourier Transform (FFT) displayed {111} and {002} facets.  

Cell viability for the MTS assay

The trypomastigote viability percentages for both T. cruzi strains showed a dosedependent response; this response was most evident for the Sylvio X10/4 strain, which showed a 50% reduction in parasite viability at the 1 µg dose of AgNPs and a 95% reduction in viability at the highest experimental dose of 12 µg after 24 hours of treatment. However, the Esmeraldo strain showed an 83% reduction in viability at the highest dose (12 µg), showing statistical differences between the Esmeraldo strain and the Sylvio X10/4 strain. Statistical analysis of the results showed differences between the negative control and all other treatments, including the different doses of AgNPs and the positive control (NFMX) (Figure 3).

Figure 3  T. cruzi survival after treatment with any of three different doses of AgNPs (1, 5, 10, 12 µg/100 µl), Nifurtimox (NFMX, 400 µg/100 µl) or without treatment (-Ctrl, used as a negative control for the MTS viability assay). Samples were evaluated after 24 hours of treatment. Each bar represents the absorbance mean value ± SD. Differences (P < 0.05) among groups, according to Tukey’s test, are indicated by characters on top of treatment bars.  

IC₅₀ estimation

AgNPs showed no different toxicity between IC₅₀ from the studied strains of T. cruzi. The IC₅₀ for trypomastigotes for Sylvio X10/4 and Esmeraldo strains were 4.05 and 6.16 µg/mL, respectively (Figure 4).

Figure 4 Inhibition of T. cruzi after 24 hours of AgNPs treatment (1, 5, 10, 12 µg/100 µl). Each bar represents the absorbance mean value ± standard deviation (n=6). Within the same time point, statistical differences (P < 0.05) among groups are shown with different characters above the bars acording to Tukey test. 

HaCaT cells treated with AgNPs

The viability of HaCaT cells was assessed 48 h after treatment (Figure 5). The toxicity of AgNPs on HaCaT cells was assessed by MTS at different concentrations: 1, 3, 5, 10 or 12 μg/100 μL, which resulted in an IC₅₀ of 10 μg/mL (P<0.05), indicating that AgNPs were approximately 2-fold less toxic to HaCaT cells than trypomastigotes. The results showed that AgNPs were not cytotoxic at the effective dose (IC₅₀ values = 4.05 and 6.16 µg/mL) against TcI and TcII strains of T. cruzi.

Figure 5 The viability of HaCaT cells was examined by MTS assay. HaCaT cells were stimulated with various doses of AgNPs (1, 3, 5, 10, 12 µg/100 µl) for 24 hours. The asterisks (*) indicate statistical significant differences compared to control (CN) (P < 0.05).  

DISCUSSION

Chagas disease is a vector-borne disease caused by T. cruzi, a hemoflagellate parasite that often causes debilitating and fatal heart disease. BNZ and NFMX are the only available drugs approved for medical treatment (^{Chaves et al., 2017}; ^{Toledo et al., 2004}). However, these drugs induce harsh side effects that prevent patients from concluding treatment. In many cases, T. cruzi is resistant to these drugs, making them unable to eliminate the parasite from infected tissue (^{Colantino et al., 2016}; ^{Toledo et al., 2004}). Therefore, public health programs require the development of more effective and less toxic therapeutic drugs for the treatment of Chagas disease.

Silver nanoparticles (AgNPs) have antimicrobial effects not only against fungi (^{Soliman et al., 2018}; ^{Mousavi et al., 2015}), Gram-positive (^{Adibhesami et al., 2017}) and Gramnegative bacteria (^{Pal et al., 2007}), but also against protozoa such as Apicomplexa (Toxoplasma gondii) (^{Adeyemi et al., 2017}) and Trypanosomatida (Leishmania tropica and Leishmania infantum) (^{Baiocco et al., 2011}). Here we report the in vitro trypanolytic effects of AgNPs against two different strains of T. cruzi.

For this study, we produced hemispherical, decahedral AgNPs with a mean diameter of 52 nm (Figures 1 and 2), and used them at 5 different doses (1, 3, 5, 10 or 12 μg/mL) to test toxicity in HaCaT cells and trypanolytic activity in trypomastigotes produced in coculture with Vero cell monolayers. We found that silver nanoparticles significantly reduce cell viability in a dose-dependent manner in HaCaT cells. This observation is in agreement with other studies (^{Chen et al., 2019}; ^{Perde-Schrepler et al., 2019}). Importantly, AgNPs are not cytotoxic at doses effective against T. cruzi strains, and the toxicity of AgNPs on trypomastigotes was indirectly assessed by an MTS proliferation assay, which showed a negative effect of AgNPs on trypomastigote viability. Figure 3 shows that nanoparticles significantly reduced cell viability compared to the negative control, and that AgNPs had similar trypanolytic power to NFMX at different doses. Although no significant differences in T. cruzi viability were observed with different doses of AgNPs, a dose-dependent trend in viability was observed. These data suggest that higher doses of AgNPs may have additional effects on the viability of T. cruzi trypomastigotes. Interestingly, a relatively low dose (12 μg/100 μL) of AgNPs killed up to 96.3% (with an IC₅₀ = 4.05 - 6.16 µg/100 μL) of trypomastigotes in vitro, whereas NFMX (400 μg/100 μL) killed only 70% (Figures 3 and 4). This also suggests that AgNPs could be a more efficient trypanolytic agent in vitro than NFMX, as observed in the present study, but also in comparison to BNZ nanoparticles (with an IC₅₀ = 36 µg/mL) (^{Scalise et al., 2016}).

T. cruzi has already developed some resistance to BNZ and NFMX, the only clinically approved drugs against Chagas disease (^{Mejia et al., 2012}; ^{Molina et al., 2000}). Different T. cruzi biotypes have differential resistance to BNZ. TcI strains of T. cruzi are more resistant to BNZ, whereas TcII strains are more sensitive to this drug (^{Toledo et al., 2004}; ^{Mejia et al., 2012}). Here, we found no differences in sensitivity to AgNPs between the two T. cruzi strains tested (Sylvio X10/4 and Esmeraldo) belonging to different biotypes (TcI and TcII, respectively). Although further studies are needed, our data regarding the broad spectrum of antiparasitic activity of AgNPs suggest that nanoparticles could be an interesting option to solve the problem of parasite resistance to other drugs, either used as an alternative drug or in combination with those currently in use.

Further studies should address the effects of AgNPs as alternative drugs against T. cruzi, especially to understand whether these nanoparticles can exert a detrimental effect on amastigotes, the intracellular developmental stage of the parasite in mammals. Leishmania tropica amastigotes have been reported to be more sensitive to AgNPs than promastigotes at a dose of 10 μg/0.1 mL (^{Allahverdiyev et al., 2011}). If that hypothesis holds for T. cruzi, the use of AgNPs in vivo for therapeutic use against Chagas disease would become an interesting topic of study.

AgNPs possess unique physical and electronic properties, which make them excellent candidates for medical applications. The antimicrobial mechanism of AgNPs seems to be related to the formation of free radicals that damage the microorganisms’ membrane (^{Kim et al., 2007}). These properties depend on size, shape, and nanoparticle coating (^{Dubey et al., 2015}). In the present study, we synthesized AgNPs with an average diameter of 52 nm; most of them were hemispherical (decahedral), had {111} and {110} facets, and had acceptable trypanolytic activity. However, we believe that the trypanolytic activity of AgNPs could be improved, if the shape and size of the nanoparticles were adjusted to triangles with a size of 1-10 nm. This type of nanoparticles shows higher antibacterial activity than larger AgNPs with hemispherical shapes due to the higher density on the facets {111}. This allows triangular nanoparticles to freely permeate the plasma membrane and thus induce more damage to microorganisms. In addition, truncated triangular AgNPs require lower doses (1-10 μg/0.1 μL) to exert bactericidal properties than larger spherical and nanorodal AgNPs (12.5 and 50-100 μg/0.1 μL, respectively) (^{Ge et al., 2014}; ^{Dubey et al., 2015}; ^{Morones et al., 2005}).

CONCLUSIONS

A dose of 12 µg/100 µL dose of AgNPs with an average diameter of 52 nm was cytotoxic for trypomastigotes from TcI and TcII T. cruzi biotypes in vitro. Future studies test triangular shaped AgNPs with {111} facets to improve the antiprotozoal effects of AgNPs on different biotypes of T. cruzi.

Conflict of interest statement

We declare that we have no conflict of interest.