The AmpB and TCZB structures were used to evaluate their theoretical physicochemical properties, whether PAINS are present, and the drug interaction prediction. The predictions were calculated using the swissADME web tool (http://www.swissadme.ch) considering Lipinski’s rule of five (Ro5) (Lipinski, 2004) followed by the additional rule proposed by Veber (Veber et al., 2002) and PAINS presence (Baell and Holloway, 2010). Drug interaction prediction was performed using the Drug Interaction Checker of DrugBank (https://go.drugbank.com/drug-interaction-checker).

Leishmania amazonensis (strain MHOM/BR/73/M2269) promastigotes were kept at 25°C in M199 medium added with 40 mM HEPES, 26 mM NaOH, 5 µg/mL of hemin supplemented with 10% of fetal bovine serum (FBS, Gibco^®). The parasites were cultivated for three days for logarithm growth stage parasite assays and seven days for stationary growth stage parasites assays.

Cytotoxicity assay was performed as previously described (Ceole et al., 2017). Cell viability was evaluated by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. Optical density (O.D.) was read in an ELISA plate reader (Biotek Model EL-800, VT, USA) at 570 nm. In order to calculate the 50% cytotoxic concentration index (CC₅₀), THP-1 (ATCC TIB-202) cells were seeded in 96-well plates at density of 25x10⁴ cells/mL in RPMI 1640 medium containing 50 ng/mL of PMA (Phorbol 12-myristate 13-acetate) for 48h as previously described (Alcântara et al., 2020). The plates were incubated at 37°C in a humidified incubator with an atmosphere of 5% CO₂ to induce differentiation into adherent macrophages. After this, macrophages were incubated with increasing concentrations of TCBZ or AmpB for 48 or 72h in a CO₂ incubator. Concentrations used for AmpB ranged from 0 to 10 μM (0, 0.009, 0.019, 0.039, 0.078, 0.156, 0.312, 0.625, 1.25, 2.5, 5 and 10 μM), while concentrations used for TCBZ ranged from 0 to 100 μM (0, 0.097, 0.195, 0.39, 0.78, 1.56, 3.125, 6.25, 12.5, 25, 50 and 100 μM). Then, 50 µL of MTT at 5 mg/mL was added per well, and the plates were incubated for 4h at 37°C. The plates had their medium removed, and 25 µL of 10% SDS solution in HCL was added for cell lysis. Subsequently, 50 µl of DMSO was added for the elution of formazan crystals. The colorimetric assay was read using optical density at 570 nm. Cell inhibition (CI) values were obtained from three experiments performed in triplicate according to the equation CI = (O.D treatment x 100/O.D. negative control) (Escobar et al., 2010; Ceole et al., 2017).

Infection assays were performed as previously described (Alcântara et al., 2020). Differentiated macrophages (25x10⁴ cells/mL) were infected with metacyclic promastigotes (5 to 7-days-old metacyclic promastigote cells culture) at a ratio of 50:1 in RPMI media for 24h using 96 well-plates placed in a humidified incubator at 34°C with an atmosphere of 5% CO_2. Previously to cells treatment, infection tests were made in three different time-points (24, 48 and 72h) for determination of higher infection index.

For cells treatment, increasing concentrations of AmpB and TCBZ were added to cell media and incubated for 48h at 34°C and 5% CO₂ after infection. Concentrations used for AmpB ranged from 0 to 20 μM (0, 0.009, 0.019, 0.039, 0.078, 0.156, 0.312, 0.625, 1.25, 2.5, 5 and 10 μM), while concentrations used for TCBZ ranged from 0 to 500 μM (0, 0.97, 1.95, 3.9, 7.8, 15.6, 31.25, 62.5, 125, 250 and 500 μM). Plates were washed twice with PBS, and cells were fixated with 4% PFA for 20 min, washed twice with ultrapure H₂O, stained with Hoechst for 20 min, and read at Operetta High Content Imaging System (Perkin Elmer). The number of intracellular amastigotes was evaluated to calculate half-maximum inhibitory concentration (IC₅₀). Non-treated wells were considered as having 100% viability of amastigotes.

Growth curve was performed using three-day-old promastigote cells culture of L. amazonensis. Parasites were harvested and had their concentration adjusted to 1x10⁶ parasites/mL and seeded in 24 well plates in 1.5 mL of M199 media. Non-treated parasites were used as control. Promastigotes were treated with increasing concentrations of AmpB (0.312, 0.625, 1.25, 2.5 and 5 μM) or TCBZ (3.125, 6.25, 12.5, 25 and 50 μM) and maintained in a BOD incubator at 25°C for 72h. Plates had absorbance measured daily for drug effect analysis at 600 nm (Kar et al., 2021).

For determination of promastigotes IC_50, 3-day-old culture of promastigotes was seeded in 96 well plates at a concentration of 10⁶ parasites/mL in the presence of AmpB (0, 0.019, 0.039, 0.078, 0.156, 0.312, 0.625, 1.25 and 2.5 μM) or TCBZ (0, 0.097, 0.195, 0.39, 0.78, 1.56, 3.125, 6.25, 12.5, 25, 50 and 100 μM) in M199 media. Plates were maintained at 25°C for 72h. After incubation, promastigotes viability was accessed by MTT assay (as described at item 2c).

Infected macrophages were treated with TCBZ and AmpB using a ratio of 10:1, 5:1, 2:1, and 1:1 in five serial dilutions (base 2) for 48h. The molar concentration for each ratio was designed as follows: 40 μM:4μM; 20μM:4 μM; 16μM:4μM and 4μM:4μM for TCBZ and AmpB respectively After the plates were analyzed using the Operetta CLS High Content Analysis System, Fractional Inhibitory Concentrations (FICs) were calculated for the IC₅₀ of the drug alone and combined (Trinconi et al., 2014; Kar et al., 2021). Isobolograms were plotted using the sum of FICs (ΣFICs) of the ratio of each combined drug. The average of FICs sum of all ratios (χΣFICs) was calculated and the combinatory effect was determined according to Odds (2003), where χΣFICs ≤ 0,5 is synergic, χΣFICs > 0,5< 4 is indifferent, and χΣFICs > 4 is antagonist. The equation for the ΣFIC₅₀ calculation is described below:

To the present, there is no gold standard method established to test antimicrobial synergism. Although the method used in this work can be limited to only testing antimicrobials for a fixed incubation time, it was used because it is a well-established method.

Leishmania amazonensis (10⁶ promastigotes/mL) were previously treated with TCBZ (IC₅₀ for 72h). To assess the mitochondrial membrane potential (ΔΨm), the promastigotes were washed with PBS and incubated for 15 min at 26°C with 5 μg/mL of rhodamine 123 (Rh123) and immediately evaluated in a flow cytometer. For cell cycle analysis, promastigotes were resuspended in 500 μL of PBS, to which the same volume of permeabilizing solution and DNA staining solution was added (3.4 mM Tris-HCl, pH 7.4; 0.1% NP-40; 700 U/mL free RNase A-DNase, 10 mM NaCl and 75 μM propidium iodide). A fluorescence reading was taken after an incubation period of 10 min at room temperature. To evaluate cell death, promastigotes were resuspended in annexin binding buffer (140 mM NaCl, 5 mM CaCl_2, and 10 mM HEPES-Na, pH 7.4) and labeled with Annexin V conjugated to Alexa Fluor 488 and 100 μg/mL of propidium iodide (PI). In all cases, events were read in a FACSCanto II flow cytometer (Becton – Dickinson, San Jose, CA, USA), and the data were analyzed using FlowJo v10 (Treestar Software, Ashland, OR, USA).

ROS quantification was performed as described by Bortoleti et al. (Bortoleti BT da et al., 2021). Promastigotes were cultured in 1 mL of M199 in presence of AmpB or TCBZ IC₅₀ for 72h. In positive controls, 0.2 µL of 2 mM H₂O₂ was added to 1 mL of cell solution 60 min before withdrawing the treatment. The parasites were collected through centrifugation (2000 rpm for 7 min), the supernatant was discarded, and the pellet was resuspended in 500 µL of M199. Parasites were plated in black 96-well plates (99 µL per well). An initial reading was performed to assess cellular autofluorescence in a fluorimeter (Ex 495; Em 520). Subsequently, 1 µL of the H₂DCFDA solution was added (final concentration in the well of 20 µM). Plates were incubated for 1h in a 37°C incubator with an atmosphere of 5% CO_2, and, after this time, the fluorescence was read in a fluorimeter.

Neutral lipids were detected as described in Bortoleti et al. (Bortoleti BT da et al., 2021). Promastigote forms previously treated with TCBZ or AmpB (IC₅₀) for 72h were collected and washed twice with PBS. After that, 199.5 µL of M199 and 0.5 µL of Nile Red solution (1 mg/mL) were added and incubated for 30 min in a BOD. Cells were rewashed and resuspended in 500 µL PBS. Then, the cell suspension was plated in black 96-well plates (100 µL per well), and the fluorescence reading was performed in a fluorimeter (Ex 530; Em 635).

Parasites were treated with TCBZ or AmpB (CI₅₀) for 72h. The parasites were then harvested at a concentration of 1x10⁶ parasites/mL, washed in PBS, and fixed with Karnovsky fixation solution (2.5% glutaraldehyde, 4% paraformaldehyde in 0.1 M sodium cacodylate buffer) for 1h and washed with cacodylate buffer after incubation. Untreated parasites were also fixed and used as control. For scanning electron microscopy (SEM), cells were previously seeded in poly-L-lysin-coated coverslips. Cells were post-fixated with 1% osmium tetroxide for 1h in the dark and then dehydrated at growing ethanol concentrations (30 to 100%). Coverslips containing treated or untreated cells were submitted to critical point drying and then coated with a 20-nm-thick gold layer and observed under a JEOL JSM 6010 PLUS-LA (Akishima, Tokyo, Japan) scanning electron microscope. For transmission electron microscopy (TEM), treated and untreated cells were post-fixated with 1% osmium tetroxide for 1h in the dark, dehydrated using growing concentrations of acetone (30 to 100%) and infiltrated with increasing concentrations of Poly/Bed 812 resin and polymerized for 72h at 60°C. Ultrathin sections were contrasted for 30 min with uranyl acetate and 2 min with lead citrate. The ultrathin sections were observed in a JEOL JEM-1400 Plus transmission electron microscope operating at 80 kV.

Statistical analysis was performed in GraphPad Prism 8 (GraphPad Software, Inc. San Diego, CA, USA) using one or two-way ANOVA test followed by Dunnett’s post-test. IC₅₀ assay results were transformed in logarithm values and analyzed by dose-response inhibition (* p-value<0.04; ** p-value<0.009; *** p-value<0.0009; **** p-value<0.00009). Results were plotted, showing the standard error of the mean. All experiments were performed in biological triplicate.

The physicochemical proprieties of AmpB and TCBZ were assessed to compare their predicted oral bioavailability using Lipinski’s and Veber’s criteria (Veber et al., 2002; Lipinski, 2004). As expected, AmpB presented higher violations than TCBZ ( Table 1 ). This is in line with the high level of toxicity that has been described for it, even when using the intravenous route (Laniado-Laborín and Cabrales-Vargas, 2009). AmpB and TCBZ chemical structures are represented in Figures 1A, B , respectively.

For both drugs, pan assay interference compounds (PAINS) were not identified by the swissADME web tool (Daina et al., 2017). PAINS are substructural features of chemical compounds that are often related to false-positive results in high-throughput screens because they tend to react nonspecifically with numerous biological targets rather than specifically affecting one desired target (Baell and Holloway, 2010). No drug interactions were found by the DrugBank tool for AmpB and TCBZ, suggesting that concomitant use of the drugs is safe.

To evaluate the effects of TCBZ on parasite growth and determine which treatment time was the best for determining the effects of the drug on promastigote forms, parasites were seeded in 24-well plates, and their growth was estimated by absorbance for three consecutive days. AmpB was used as a comparative drug, as it is one of the most common drugs used to treat leishmaniasis worldwide. The results showed that AmpB affects L. amazonensis promastigote proliferation after 72h of treatment for all concentrations tested. On the other hand, TCBZ inhibited parasite growth at higher concentrations after 48h of treatment ( Figure 2 ). Thus, promastigote assays were subsequently conducted only at 72h since this time point showed a statistically significant effect for both drugs.

The IC₅₀ of TCBZ and AmpB for promastigotes were determined after 72h of treatment by MTT assay. TCBZ IC₅₀ was 6 µM, while AmpB IC₅₀ was 0.06 µM ( Table 2 ).

Exploratory infection assays were performed for macrophages derived from THP-1 cells infected with intracellular amastigotes. Based on the results obtained ( Supplementary Figure 1 ), it was decided that 48h would be the most accurate time-point for amastigote forms, since it presented a higher infection rate then 72h. Although 24h showed similar results when compared to 48h, based on the lack of statistically significant activity of AmpB and TCBZ in promastigotes proliferation analysis ( Figure 2 ) this time-point was not considered for further evaluations. Infected macrophages were therefore treated for 48h with growing concentrations of AmpB and TCBZ, and the number of infected cells was analyzed using the Operetta High Content Imaging System. AmpB showed significant results at 5 µM and TCBZ at 3.1, 6.2, and 50 µM ( Figure 3 ). The number of total amastigotes was used to determine the IC₅₀ for both AmpB (2.02 µM) and TCBZ (45.67 µM).

Half-minimal cytotoxicity concentrations (CC₅₀) of TCBZ and AmpB were also determined by MTT assay in THP-1 derivate macrophages. Assays were conducted at 48 and 72h since these were the times evaluated in amastigote and promastigotes assays, respectively. TCBZ showed lower cytotoxicity than AmpB for both 48h (276.6 µM and 8.1 µM, respectively) and 72h (39.4 µM and 0.3 µM, respectively) ( Figure 4 ). The selectivity index (SI) was determinate using CC₅₀ and IC₅₀ of amastigote forms being 4 and 6 for AmpB and TCBZ , respectively ( Table 2 ).

Treated parasites (IC₅₀ for 72h) were observed by electron microscopy to analyze ultrastructural ( Figure 5 ) and morphological ( Figure 6 ) alterations induced by TCBZ or AmpB. In L. amazonensis promastigotes treated with AmpB IC₅₀, several cytoplasmatic electron-dense spherical bodies were observed, which probably contain lipids ( Figure 5B ). TCBZ IC₅₀ treated parasites showed the same cytoplasmatic electron-dense spherical bodies observed after AmpB treatment (white arrowheads), and other ultrastructural alterations such as vesicles at the flagellar pocket were also frequently noticed (black arrows, Figures 5C, C’ ). Infected cells were also treated with the highest statistically relevant concentration of TCBZ (50 μM for 48h) to assess the drug’s effects. Although the same electron-dense structures seen in promastigotes were not observed, treated parasites showed intense vesiculation, a feature not seen in untreated parasites ( Figure 6B ). Moreover, structures resembling cell debris inside the parasitophorous vacuole of treated samples were frequently observed in these cells ( Figure 6B-B’’ , white arrowheads).

The morphology of treated promastigotes was observed by SEM. Untreated parasites were used as control and showed normal morphology ( Figure 7 , row 1). AmpB-treated parasites showed morphological alterations such as shrinkage and roundness of the parasite cell body ( Figure 7 , row 2, white arrowheads). Leishmania amazonensis promastigotes treated with TCBZ IC₅₀ for 72h also showed morphological alterations like shrinkage and roundness in the parasite shape cell ( Figure 7 , row 3, white arrowheads). Moreover, dividing cells were frequently observed, indicating a possible alteration in the cytokinesis process ( Figure 7 , row 3, black arrows).

To better understand TCBZ’s mechanism of action, treated parasites were analyzed by flow cytometry for cell death, cell cycle, and mitochondrial membrane potential. Cell death analysis showed that treatment with AmpB increased the number of stained parasites using both Annexin V and PI ( Figure 8A ). Untreated promastigotes showed 99.4% of living cells, 0.29% of apoptosis, 0.14% of necrosis, and 0.17% of late apoptosis. Similarly, TCBZ-treated parasites showed 99.4% of living cells, 0.3% of apoptosis, 0.1% of necrosis, and 0.2% of late apoptosis. On the other hand, AmpB-treated parasites showed 20.4% of living cells, 34.9% of apoptosis, 2.7% of necrosis, and 42% of late apoptosis.

Mitochondrial membrane potential was evaluated in treated promastigotes. Parasites treated with AmpB showed a clear loss of fluorescence when compared to controls (93%), while TCBZ-treated parasites showed a decrease of 50% in fluorescence (p = 0.07, Figure 8B ).

The cell cycle was analyzed with PI staining, and non-treated parasites were used as control. TCBZ-treated L. amazonensis promastigotes showed significant alterations in cell cycle phases, with a decrease in G1 from 74.6% to 54.3% and an increase in S and G2 from 9.7% to 16.6% and 11.9% to 25.4%, respectively ( Figure 8C ).

ROS were quantified to access cell death pathway. For this, promastigote forms were treated with IC₅₀ of both drugs for 72h. Results showed that treated cells present a ROS profile that is similar to non-treated parasites, indicating that the anti-Leishmania effect of TCBZ may not be using the ROS production pathway ( Figure 9A ).

Furthermore, to verify if electron-dense vesicles observed by TEM could be filled with neutral lipids, this lipid class was quantified in promastigotes after 72h of treatment with IC₅₀. We found a 69% tendency to increase the fluorescence intensity of parasites treated with TCZB compared to untreated promastigotes. However, no statistical difference was found ( Figure 9B ).

The next step was to examine whether TCBZ and AmpB have a synergistic effect that lowers the IC₅₀ of AmpB and, consequently, its toxicity, elevating the SI. In order to do this, combined proportions of 10:1, 5:1, 4:1, and 1:1 concentrations for TCBZ and AmpB were prepared, and infected macrophages were treated for 48h. These ratios were chosen based on the previous observations that TCBZ alone appears to have a comparable anti-Leishmania effect in doses 10x higher than AmpB. Also, as the IC₅₀ of AmpB is around 2 μM, the focus of this assay was to evaluate if TCBZ was capable of lowering AmpB’s IC₅₀ through a synergistic effect. The results showed that at all ratios tested, the IC₅₀ of the drug in combination ranged from three to 120 times lower than the drugs alone ( Table 3 ). Cytotoxicity assay was also performed in macrophages derived from THP-1 cells. For 1:1 ratio, CC₅₀ was 29.34 μM for both drugs. For 1:4, 1:5 and 1:10 ratios, AmpB CC₅₀ was 14.73, 27.77 and 34.16 μM, respectively, while TCBZ CC₅₀ was 33.60, 36.14 and 21.55 μM, respectively. Regarding SI, AmpB and TCBZ presented the value of 77.2 for 1:1 ratio. For ratios of 1:4, 1:5 and 1:10, AmpB presented values of 39.8, 74.4 and 69.7, respectively, while TCBZ presented values of 46.6, 15.7 and 5.5, respectively.

Once the IC₅₀ for each drug in each combination was determined, the ΣFICs equation (see item 2e) was used to calculate the average of the sum FICs (χΣFICs). We found that the combination of TCBZ and AmpB resulted in a synergistic effect with a χΣFICs of 0.25. According to Odds (2003), the drug combination is considered synergic when χΣFICs ≤ 0.5. An isobologram plotted with every analyzed ratio also demonstrated FICs below the addictive line that represents theoretical addictive (or non-interactive) χΣFICs 1 value, corroborating with χΣFICs value found in this study ( Figure 10 ).