The Oral Antimalarial Drug Tafenoquine Shows Activity against Trypanosoma brucei

Abstract

The protozoan parasite Trypanosoma brucei causes human African trypanosomiasis, or sleeping sickness, a neglected tropical disease that requires new, safer, and more effective treatments. Repurposing oral drugs could reduce both the time and cost involved in sleeping sickness drug discovery. Tafenoquine (TFQ) is an oral antimalarial drug belonging to the 8-aminoquinoline family which is currently in clinical phase III. We show here that TFQ efficiently kills different T. brucei spp. in the submicromolar concentration range. Our results suggest that TFQ accumulates into acidic compartments and induces a necrotic process involving cell membrane disintegration and loss of cytoplasmic content, leading to parasite death. Cell lysis is preceded by a wide and multitarget drug action, affecting the lysosome, mitochondria, and acidocalcisomes and inducing a depolarization of the mitochondrial membrane potential, elevation of intracellular Ca²⁺, and production of reactive oxygen species. This is the first report of an 8-aminoquinoline demonstrating significant in vitro activity against T. brucei.

INTRODUCTION

Human African trypanosomiasis (HAT) is a vector-borne parasitic disease that threatens approximately 70 million people distributed over 36 countries in sub-Saharan Africa. HAT, more commonly known as sleeping sickness, is caused through infection by two subspecies of the protozoan parasite Trypanosoma brucei: T. brucei gambiense, accounting for 95% of all cases, causes a chronic infection and T. brucei rhodesiense, which produces an acute infection (1). The parasite remains entirely extracellular throughout its life cycle, while the course of the disease is divided into two stages: early blood-stage infection and late-stage infection, in which parasites cross the blood-brain barrier and can be found in the cerebrospinal fluid. If untreated, HAT is usually fatal (2).

To date, there is no single drug that is effective against both stages of the disease or against both subspecies, and all therapies require parenteral administration (3). For first-stage disease, pentamidine is used to treat T. brucei gambiense, while suramin is preferred for T. brucei rhodesiense infection. For second-stage disease caused by T. brucei rhodesiense, the highly toxic arsenical compound melarsoprol is the only choice, but its use may result in a posttreatment reactive encephalopathy which kills 5% of patients (4). A nifurtimox/eflornithine combination therapy (NECT) is the first-line treatment for second-stage disease caused by T. brucei gambiense infection, (3), but, unfortunately, it is not effective for T. brucei rhodesiense (5). Several new drugs such as diamidine derivatives, fexinidazole, and oxaborole SCYX-7158 are now in various stages of the development pipeline for treating HAT (4). However, it will be several years before we discover whether any of these new treatments will be successful. An alternative melarsoprol delivery system is also being studied involving the construction of melarsoprol-cyclodextrin inclusion complexes which increase the drug's water solubility (6, 7), but there are concerns about its development (8). Therefore, a continuous effort is required in order to find new, more effective drugs that provide safer HAT treatments, are simpler to administer, and are less expensive than those currently available. Repurposing oral drugs has the potential to significantly reduce both the time and cost of HAT drug discovery (9).

Tafenoquine (TFQ) is an oral 8-aminoquinoline drug being developed by GlaxoSmithKline, in partnership with Medicines for Malaria Venture, for the radical cure of Plasmodium vivax malaria (13). TFQ was designed as a synthetic analog of primaquine but with a longer half-life, allowing shorter courses to be given (11). In the last decade, several trials have shown that TFQ is a potential alternative to primaquine as a radical cure for P. vivax malaria (11–14). TFQ was observed to be as effective as primaquine but with the advantage of requiring a shorter treatment (12), an important consideration for patient compliance and cost. Very promising results have recently been obtained from a phase IIb clinical trial combining a single dose of TFQ with chloroquine to treat and prevent the relapse of P. vivax malaria (13). These results have been underlined as “a single-dose TFQ radical cure with potential to transform P. vivax therapeutics and become a major contributor to malaria elimination” (11). As a consequence, TFQ has entered phase III clinical trials and has been granted Breakthrough Therapy Designation by the U.S. Food and Drug Administration, in order to accelerate its development (14). On the other hand, for a safer deployment of novel 8-aminoquinolines, such as TFQ, several institutions and companies are currently involved in developing reliable and affordable point-of-care tests for glucose-6-phosphate dehydrogenase (G6PD) deficiency to avoid the severe hemolysis that the 8-aminoquinolines can induce in G6PD-deficient individuals (15).

TFQ revealed significant activity against the trypanosomatid parasite Leishmania in both in vitro tests (promastigote and intracellular amastigote forms) and a mouse model of Leishmania infection (16), demonstrating potential as an oral antileishmanial agent. In a previous study about the mechanism of action of TFQ against Leishmania, we described how TFQ targets mitochondria, leading to an apoptosis-like death in Leishmania (17). In contrast, TFQ showed less in vitro activity against intracellular amastigotes of Trypanosoma cruzi, the causative agent of Chagas disease (16).

We describe here experimental in vitro activity of TFQ against T. brucei bloodstream forms. Our results provide the first insight into the mechanism of the lethal pathway of an 8-aminoquinoline in T. brucei, suggesting TFQ has a multitarget drug action which leads to necrotic cell death.

MATERIALS AND METHODS

Chemical compounds.

Tafenoquine (TFQ) was kindly provided by GlaxoSmithKline (Greenford, United Kingdom). A stock solution of 10 mM TFQ was prepared in ethanol. Fluo-4 AM, Pluronic F-127, H₂DCF-DA (2′,7′-dichlorodihydrofluorescein diacetate), BAPTA-AM, and Sytox green were purchased from Invitrogen (Carlsbad, CA). Rhodamine 123 (Rh123), FCCP [carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone], d-[1-¹³C]glucose, deuterium oxide, and Triton X-100 were purchased from Sigma-Aldrich (Madrid, Spain). The CellTiter-Glo luminescent cell viability assay was purchased from Promega (Spain). All other chemicals were of the highest quality available.

Strains and parasite culture conditions.

Bloodstream forms (BSF) of T. brucei rhodesiense STIB900, one of the species responsible for HAT, were grown at 37°C, 5% CO₂ in HMI-9 medium supplemented with 20% heat-inactivated fetal bovine serum (hiFBS; Invitrogen). BSF parasites of the standard laboratory strains T. brucei brucei S427 and T. brucei brucei “single-marker” S427 (S16) were grown at 37°C and 5% CO₂ in HMI-9 medium supplemented with 10% hiFBS. Leishmania major (MHOM/JL/80/Friedlin) promastigotes were grown at 28°C in RPMI 1640 modified medium (Invitrogen) supplemented with 20% hiFBS. T. brucei brucei S16 was used in all experiments described except the drug susceptibility assay, where the three strains were evaluated.

Drug susceptibility assay.

T. brucei rhodesiense, T. brucei brucei S427, and T. brucei brucei S16 (10⁴ BSF per ml) were incubated in 96-well plates with TFQ (0.02 to 10 μM) for 72 h at 37°C, 5% CO₂ in culture medium. L. major (2 × 10⁶ promastigotes per ml) were incubated in 96-well plates with TFQ (1 to 10 μM) for 72 h at 28°C in culture medium and used as a control for TFQ susceptibility. Cell proliferation was determined using the alamarBlue assay (18). Fluorescence was recorded with an Infinite F200 microplate reader (Tecan Austria GmbH, Austria) equipped with 550- and 590-nm filters for excitation and emission wavelengths, respectively.

Uptake determination.

Concentrations of 2 × 10⁷ per ml of Leishmania promastigotes and T. brucei BSF were incubated at 28 and 37°C and 5% CO₂, respectively, with 5 μM TFQ for 5 min in culture medium. Parasites were then washed twice with phosphate-buffered saline (PBS) and lysed by the addition of 10% sodium dodecyl sulfate. TFQ accumulation was determined fluorometrically by recording an emission spectrum in the range of 360 to 460 nm upon excitation at 340 nm using an Aminco-Bowman Series 2 spectrometer as described previously (19).

Microscopy analysis.

TFQ intracellular distribution was ascertained by fluorescence microscopy using the intrinsic fluorescence of TFQ (excitation, 340 nm; emission, 388 nm). BSF parasites (10⁷ per ml) were treated with 0.5 μM TFQ for 3 min at 37°C and 5% CO₂ in trypanosome dilution buffer (TDB; 5 mM KCl, 80 mM NaCl, 1 mM MgSO₄, 20 mM Na₂HPO₄, 2 mM NaH₂PO₄ [pH 7.4]). The parasites were washed with PBS and observed under a Zeiss Axiophot (Germany) epifluorescence microscope; images were captured with a SPOT camera (Diagnostic Instruments, Inc., USA). Parasites labeled with 100 nM LysoTracker green for 15 min at 37°C and 5% CO₂ in TDB were used to visualize the distribution of acidic organelles and analyzed by fluorescence microscopy.

Scanning electron microscopy (SEM) was performed as follows. Log-phase cultures of BSF parasites were incubated at 37°C and 5% CO₂ in culture medium for the indicated times in the absence or presence of 1 μM TFQ and fixed with glutaraldehyde (2.5%) for 24 h at 4°C. Fixed cells were sedimented onto electron microscopy stubs previously covered with poly-l-lysine at 4°C over 24 h and in the presence of humidity. The samples were postfixed with 1% (wt/vol) osmium tetroxide (OsO₄) and then dehydrated using a graded series of ethanol solutions (50, 70, and 90% and three times at 100%) over a period of 15 min at room temperature. The critical point was reached in a Leica EM CPD300 automated dryer, before the samples were fully dried in the presence of carbon using an EMITECH K975X Turbo Evaporator and then observed using an Auriga (FIB-FESEM) Carl Zeiss SMT scanning electron microscope.

Transmission electron microscopy (TEM) was performed as described previously (20). Briefly, BSF parasites were incubated as described above with 1 μM TFQ and fixed with glutaraldehyde (2.5%) for 4 h at 4°C. After fixation, the cells were washed three times for 20 min at 4°C with 0.1 M cacodylate buffer (pH 7.4). Postfixation was performed in 2% (wt/vol) osmium tetroxide (OsO₄) for 2 h at room temperature. Subsequently, washed cells were dehydrated in ethanol (with steps of 50, 70, and 90% and then two times at 100%) and then embedded in Epon 812. Ultrathin sections of 500 Å were cut on a Ultracent S Leica microtome, counterstained with uranyl acetate and lead citrate and then observed using a Zeiss 902 transmission electron microscope.

Analysis of the mitochondrial membrane potential (ΔΨ_m).

The variation of ΔΨ_m in BSF parasites was monitored using Rh123 accumulation, as described previously (21, 22) with some modifications. BSF parasites (10⁷ per ml) were incubated in culture medium with 0.5, 1, 5, or 10 μM TFQ for 10 min at 37°C and 5% CO₂ and then treated with 250 nM Rh123 for 5 min. Afterward, the parasites were washed twice, resuspended in PBS, and analyzed by flow cytometry in a FACScan flow cytometer (Becton Dickinson, CA) equipped with an argon laser operating at 488 nm. Fluorescence emission between 515 and 545 nm was quantified using CellQuest software. Parasites that were either left untreated or fully depolarized by incubation with 10 μM FCCP for 10 min were used as controls.

ATP measurement.

ATP was measured using a CellTiter-Glo luminescent assay, which generates a luminescent signal proportional to the amount of ATP present, as described previously (19). BSF parasites (10⁷ per ml) were incubated in culture medium with or without 10 μM TFQ for 10, 60, 120, and 180 min. Next, 25-μl aliquots of parasites were then transferred to a 96-well plate, mixed with the same volume of CellTiter-Glo, and incubated in the dark for 10 min, and then the bioluminescence was measured using an Infinite F200 microplate reader (Tecan Austria GmbH, Austria).

Glucose metabolic flux analysis.

The glucose metabolic flux of BSF parasites was analyzed using d-[1-¹³C]glucose. Parasites were washed three times with PBS and resuspended at 10⁸ parasites per ml in separation buffer (44 mM NaCl, 57 mM Na₂HPO₄, 3 mM NaH₂PO₄, 10 mM glucose). The parasites were then incubated at 37°C for 5 min, centrifuged (1,500 × g, 10 min), and resuspended at 4 × 10⁷ parasites per ml in assay buffer (135 mM NaCl, 5 mM KCl, 1 mM CaCl₂, 1 mM MgSO₄, 10 mM sodium phosphate, and 20 mM d-[1-¹³C]glucose). Three 1-ml aliquots were treated with 0, 5, and 10 μM TFQ, respectively, for 10 min at 37°C and then placed in an ice-water bath for 1 min. The metabolite extraction was then carried out as previously described (23) with slight modifications. Parasites were centrifuged (20,800 × g, 0°C, 3 min), the pellet was washed twice with cold PBS, resuspended with 225 μl of cold chloroform-methanol-water (20:60:20 vol/vol/vol) and shaken for 1 h in a Thermomixer (1,400 rpm, 4°C). The samples were centrifuged again (20,800 × g, 0°C, 3 min) to pellet cell debris, and extracted metabolites were recovered in the supernatant. For nuclear magnetic resonance (NMR) experiments, the supernatant from each sample was evaporated to dryness under a nitrogen stream, and the residue reconstituted in 40 μl of buffer (50 mM phosphate and 1 mM acetonitrile [pH 7.0] prepared using deuterated water [D₂O]) before being transferred to 1.7-mm NMR tubes. NMR spectroscopy was performed at 298 K on a liquid-state high-resolution Bruker Avance III 500 MHz spectrometer equipped with a gradient inverse triple-resonance 1.7-mm TCI MicroCryoProbe. Two-dimensional (2D) ¹H-¹³C multiplicity-edited HSQC spectra were acquired with a standard pulse sequence from Bruker library “hsqcedetgpsp.3” with a 90° pulse, 0.2-s acquisition time, 1.2-s relaxation delay, ¹J_C-H of 145 Hz, 100 scans, and acquisition of 1,024 data points (for ¹H) and 256 increments (for ¹³C). The ¹H and ¹³C pulse widths used were p₁ = 10.0 μs and p₃ = 12.5 μs, respectively. The ¹H and ¹³C spectral widths were 9.90 and 170.00 ppm, respectively. The acetonitrile methyl cross peak was used for chemical-shift calibration (2.06 ppm for ¹H and 1.47 for ¹³C). The matrix size 1k×256 was zero filled with 1k×1k by application of a cosine squared function prior to Fourier transformation. The samples were maintained at a temperature of 298 K during data acquisition. Line broadening of 1.0 and 0.3 Hz, respectively, for F2 and F1, were applied before Fourier transformation, and all spectral data sets were processed using TOPSPIN 3.0.

Plasma membrane permeabilization.

Sytox green dye was used to assess plasma membrane integrity. BSF parasites (10⁷ per ml) were incubated in culture media with or without 10 μM TFQ for 10, 60, 120, and 180 min at 37°C and 5% CO₂, before being incubated with 5 nM Sytox green for 10 min at 4°C. Afterward, the parasites were washed twice, resuspended in PBS, and analyzed by flow cytometry in a FACScan flow cytometer equipped with an argon laser operating at 488 nm. Fluorescence emission between 515 and 545 nm was quantified using CellQuest software. The control for maximum permeabilization was obtained by the addition of 0.05% Triton X-100.

Measurement of free intracellular Ca²⁺.

Changes in the cytosolic Ca²⁺ level were monitored using the Ca²⁺-specific fluorescent probe Fluo4-AM. BSF parasites (10⁷ per ml) were incubated with 5 μM Fluo-4 AM for 60 min at 37°C and 5% CO₂ in HMI-9 medium supplemented with 0.02% pluronic acid F127 to improve dispersion of the nonpolar acetyloxymethyl ester in aqueous media with or without a 25 μM BAPTA-AM supplement for 20 min. Next, TFQ (5 and 10 μM) was added, with or without 8 mM EGTA for 10 min at 37°C. Afterward, the cells were washed twice, resuspended in PBS, and analyzed by flow cytometry in a FACScan flow cytometer. Fluorescence emission was quantified using CellQuest software.

Detection of the production of ROS.

The generation of reactive oxygen species (ROS) was measured using the H₂DCF-DA cell-permeating probe, as described previously (24) with some modifications. The increase in fluorescence due to the oxidation of H₂DCF (nonfluorescent) to the fluorogenic compound 2′,7′-dichlorofluorescein (DCF) is commonly used to detect the generation of ROS. BSF parasites (10⁷ per ml) were preincubated in culture media with 10 μM H₂DCF-DA for 20 min at 37°C and 5% CO₂, with or without a 25 μM BAPTA-AM supplement, and then incubated with 10 μM TFQ for 10 min at 37°C and 5% CO₂. The parasites were subsequently washed twice, resuspended in PBS, and DCF fluorescence was evaluated using flow cytometry in a FACScan flow cytometer with fluorescence emission quantified using CellQuest software. Parasites treated with H₂DCF-DA alone were used as a control.

RESULTS AND DISCUSSION

The antimalarial 8-aminoquinoline TFQ has shown significant activity against Leishmania (16, 17), while, in contrast, TFQ was not effective against T. cruzi (16). In the present study, we evaluated the activity and the mechanism of action of TFQ against T. brucei spp.

TFQ shows in vitro activity against T. brucei.

The oral antimalarial drug TFQ efficiently inhibited in vitro proliferation of different species of T. brucei at concentrations in the nanomolar range, with 50% effective concentrations (EC₅₀s) ranging from 0.17 ± 0.02 μM for T. brucei rhodesiense and 0.22 ± 0.03 μM for T. brucei brucei S427 to 0.42 ± 0.02 μM for T. brucei brucei S16. These EC₅₀s were between 13 and 32-fold lower than those observed for L. major promastigotes (EC₅₀ = 5.5 ± 1.1 μM) (Fig. 1) and between 52- and 128-fold lower than those reported by Yardley et al. (16) with respect to T. cruzi (EC₅₀ = 21.9 μM). This is the first evidence of an 8-aminoquinoline's activity against T. brucei.

FIG 1.

TFQ sensitivity in Trypanosoma brucei and Leishmania. Differences in sensitivity to TFQ between T. brucei brucei S16 (●) and L. major (○) were assayed by determination of the percent cell viability using an alamarBlue-based assay, as described in Materials and Methods. Shown are the means ± the standard deviations (SD) from three independent experiments.

The higher activity of TFQ upon T. brucei with respect to Leishmania could be due to a higher accumulation of the drug in the former. We therefore studied the uptake of TFQ in both trypanosomatid parasites at the saturation time point (5 min). Using a spectrofluorometric assay, we found that the accumulation of TFQ in T. brucei BSF was 50% lower than in promastigotes of L. major (P < 0.05), this being despite its higher activity against T. brucei. The absence of correlation between susceptibility and uptake level was also reported for the case of sitamaquine, a related 8-aminoquinoline, against different Leishmania species, (25) and for several analogs of the antitrypanosomal diamidines DB75 and DB820 against T. brucei (26).

TFQ distributes mainly within lysosomes and acidocalcisomes of T. brucei.

We studied TFQ distribution in T. brucei taking advantage of the intrinsic fluorescence of the drug (17). Fluorescence microscopy images showed a single discrete organelle, located in a perinuclear position at the posterior end of the cell, together with a vesiculated pattern throughout the whole-cell body (Fig. 2a). This pattern resembles the single lysosome and the acidocalcisomes stained with LysoTracker green (Fig. 2b), a fluorescent acidotropic probe used to label the acidic organelles in the cell. Since TFQ is a weak base, the drug probably enters the cells by diffusing across the membrane, as described for Leishmania spp. (17), and is subsequently protonated and sequestrated in acidic compartments. This could lead to lysosome alterations related to the TFQ mechanism of action. Indeed, this acidotropism has also been described for the accumulation of chloroquine, another trypanocidal week base, in T. brucei (27). On the other hand, TFQ accumulation in Leishmania acidocalcisomes has also been described (17). These organelles, rich in polyphosphates complexed with calcium, represent the largest Ca²⁺ reserve in many trypanosomatid parasites (28). Furthermore, the diamidines DB75 and DB820, potent compounds against T. brucei, have also been observed to accumulate not only in the DNA-containing nuclei and kinetoplast of trypanosomes but also in the acidocalcisomes (26). The role of acidocalcisomes in the mechanism of action of these diamidines is still unclear, but it has been suggested that when the diamidines accumulate in the acidocalcisomes, they may interfere with trypanosome homeostasis mechanisms (26). We cannot rule out the distribution of TFQ in other parts of the parasite, possibly at lower concentrations which cannot be detected by fluorescence microscopy. Indeed, chloroquine is still accumulated in T. brucei once the finite volume of acidic organelles has been filled (27).

FIG 2.

TFQ intracellular distribution. Intracellular localization of TFQ and LysoTracker green were observed by fluorescence microscopy (upper panels) after incubation of the parasites with 0.5 μM TFQ (a) or 100 nM LysoTracker green (b) as described in Materials and Methods. Lysosome could be identified in a perinuclear position (arrowhead), while the acidocalcisomes were observed as punctate structures in the cytoplasm (arrow). The bottom panels show the Nomarski picture.

TFQ produces necrotic cell death in T. brucei.

It was previously reported that TFQ kills Leishmania by inducing a process of apoptosis-like death (17). However, the grossly abnormal parasites observed with phase-contrast light microscopy (data not shown) that were produced by TFQ acting on BSF parasites of T. brucei suggested death by necrosis. To confirm this point, surface alterations in control and TFQ treated parasites were examined by scanning electron microscopy (SEM). Figure 3 shows the time-dependent membrane damage produced by TFQ compared to untreated parasites. TFQ produced a disintegration of cell membranes with loss of cytoplasmic contents confirming necrotic cell death. Another distinct feature observed in SEM images of these treated cells was the increased number of flagella present. This observation correlates with the observed multiple flagella, sometimes detached from the central cell body, observed through phase-contrast light microscopy (data not shown).

FIG 3.

Effects of TFQ on the cellular morphology of BSF cells observed by SEM. (a) Control BSF parasites showing the normal cell morphology and membrane integrity. (b to e) Parasites treated with 1 μM TFQ during 3 h (b and c), 9 h (d), or 15 h (e) showing altered cell surface and loss of cytoplasmic contents caused by the drug. Scale bar, 1 μm.

TFQ-treated parasites were also examined by transmission electron microscopy (TEM) to analyze their ultrastructural changes (Fig. 4 and Fig. S1 in the supplemental material). The drug produced lysosome swelling and lysis, an effect that can be a main cause of TFQ-mediated cell death. This effect agrees with the lysosome TFQ accumulation observed by fluorescence microscopy (Fig. 2a). Due to its acidotropism, drug concentration in this acidic compartment could reach the millimolar range (27), altering lysosome pH and producing the observed lysosome swelling and lysis as a result of the increase in osmotic pressure (29). Parasites treated with TFQ also showed an enlarged flagellar pocket, probably corresponding to the “big eye” phenotype observed during phase-contrast light microscopy (data not shown). This could be due to a defect in vesicular trafficking. The changed number and position of flagella detected by SEM was also clearly observed through TEM. An increased number of nuclei was also detected in many cells, suggesting a defect in cytokinesis. Other alterations observed included: nuclear envelope dilation, mitochondrial disorganization, lysis of internal vesicles, the presence of cytoplasm-free areas, altered cell shape, and the presence of invaginations and evaginations of the plasma membrane. Finally, TEM images also confirmed the necrotic process incorporating cell membrane disintegration and loss of intracellular components that led to parasite death (Fig. 4 and Fig. S1 in the supplemental material).

FIG 4.

Ultrastructural effects of TFQ on BSF cells observed by TEM. Cells were grown in the absence (a) or the presence of 1 μM TFQ for 1 h (b), 3 h (c), 9 h (d), or 15 h (e) and prepared for TEM analysis. Ultrastructural alterations observed included lysosome swelling and lysis, enlarged flagellar pocket and cell membrane disintegration, and loss of cytoplasmic contents. f, flagellum; fp, flagellar pocket; l, lysosome; nc, necrotic cell. Scale bar, 1 μm.

In addition to disruption of the plasma membrane, necrotic cell death involves previous intracellular events such as calcium homeostasis alteration, ΔΨm dissipation with ATP depletion and ROS generation, perinuclear clustering of organelles, activation of proteases, and swelling and lysis of lysosomes. (30–32). We therefore analyzed the occurrence of some of these events in TFQ-treated parasites.

TFQ induces ΔΨm depolarization in T. brucei.

In BSF parasites, ΔΨm is generated and maintained exclusively by electrogenic proton translocation, catalyzed by mitochondrial F₁F₀-ATPase at the expense of ATP, and is insensitive to the respiratory chain inhibitor antimycin A that acts upon cytochrome c reductase (33, 34). The presence of ΔΨm appears to be an absolute requirement for mitochondrial function and biogenesis, as well as for the growth and development of the parasite (34, 35). Furthermore, the action mechanisms of the aromatic diamidine DB75 and the commonly used drug pentamidine against T. brucei BSF have been shown to involve the collapse of ΔΨm (36). We studied the effect of TFQ on the ΔΨm in BSF parasites by flow cytometry using the accumulation of the mitochondrial dye Rh123 as an indicator. We observed that exposure to 10 μM TFQ for 10 min produced a similar decrease in Rh123 fluorescence to that obtained after treatment with the uncoupling reagent FCCP (10 μM), a mitochondrial depolarization control (Fig. 5). A 5 μM concentration of TFQ produced approximately half of the effect of FCCP, whereas 0.5 and 1 μM concentrations produced less, but still significant (P < 0.05), ΔΨm depolarization. The concentration of TFQ required to depolarize the mitochondrial membrane is ∼10-fold higher than the EC₅₀, probably because the incubation times are much shorter (10 min versus 72 h) and the number of cells used are higher (10⁷ per ml versus 10⁴ per ml) in this experiment. Therefore, we selected these concentrations (5 and 10 μM) for further experiments into TFQ activity against T. brucei.

FIG 5.

TFQ induces ΔΨm depolarization. BSF were treated without (C) or with 0.5, 1, 5, and 10 μM TFQ for 10 min, stained with 250 nM Rh123, and analyzed for fluorescence by flow cytometry. Parasites treated with 10 μM FCCP for 10 min were used as a depolarization control. (a) Histograms from a representative experiment of three independent experiments. (b) TFQ-induced ΔΨm depolarization is represented as the percentage of TFQ treated compared to FCCP-treated parasites. The results are means ± the SD from three independent experiments. The experimental values were significantly different from control values (untreated BSF) by using the Student t test (P < 0.05).

Effect of TFQ on the intracellular ATP levels of T. brucei.

In T. brucei BSF, the ΔΨm is generated at the expense of ATP (34). Therefore, one possible explanation for the observed TFQ-induced ΔΨm depolarization is that TFQ induces a significant decrease in the intracellular level of ATP. We treated BSF parasites with 10 μM TFQ and determined the intracellular ATP content at different time points. We observed that TFQ induced a slow decrease of ATP levels (Fig. 6) over periods which were longer than were necessary to induce ΔΨm depolarization (Fig. 5), indicating that the TFQ-induced decrease in ATP levels is not responsible for the observed ΔΨm depolarization. This observed decrease in ATP levels may be due to the inhibitory action of TFQ over ATP synthesis or because ATP is released into the medium after losing plasma membrane integrity during the necrotic process described above.

FIG 6.

Effect of TFQ on ATP levels. Changes in the intracellular ATP levels were determined by incubating BSF (10⁷ per ml) without or with 10 μM TFQ for 10, 60, 120, and 180 min in HMI-9 at 37°C. Afterward, 25-μl aliquots of the parasites were transferred to a 96-well plate, mixed with the same volume of CellTiter-Glo (Promega), and incubated for 10 min in the dark, and the sample bioluminescence was measured. ATP levels were calculated relative to 100% of the untreated control. Similar results were obtained in three independent experiments. The experimental values were significantly different from control values by using the Student t test (P < 0.05).

TFQ does not affect T. brucei glycolysis.

ATP synthesis in the BSF of T. brucei is achieved solely via glycolysis (37). Glycolysis has been proposed as a likely target of suramin, the drug used against first-stage HAT (38). In order to evaluate the effect of TFQ on glycolysis, we analyzed the intracellular glucose metabolic flux in BSF parasites by NMR and with the use of d-[1-¹³C]glucose for cell culturing. Metabolic fluxomics by NMR using ¹³C-labeled substrates is a well-established technique (39). ¹³C-NMR has been widely used to follow glycolytic flux, and it has already been applied to the study of T. brucei glucose and proline metabolism (40, 41). Taking advantage of [¹³C]glucose labeling, we decided to use 2D multiplicity-edited ¹H-¹³C HSQC NMR spectra due to its reasonable sensitivity and power for metabolite identification in complex mixtures (42). The analysis of the obtained spectra revealed negligible differences in the intracellular amounts of glycolytic intermediates between parasites treated for 10 min with 5 μM TFQ and the untreated control parasites (Fig. 7a), and a similar result was observed after treatment with 10 μM TFQ (data not shown). These results indicated that TFQ does not target any enzyme involved in T. brucei BSF glycolysis, so the observed decrease in ATP levels produced after TFQ treatment is not due to any effect of TFQ on ATP synthesis.

FIG 7.

Effect of TFQ on glycolysis and plasma membrane integrity. (a) Comparison of 2D ¹H-¹³C HSQC spectra obtained for the untreated control sample (upper panel) and the TFQ-treated sample (lower panel), showing the similar patterns observed. (b) Histogram of a representative experiment. The effect of TFQ on the plasma membrane permeability was determined by incubating BSF (10⁷ per ml) without (control [C]) or with 10 μM TFQ for 10, 60, 120, and 180 min in HMI-9 at 37°C and then treating the samples with 25 nM Sytox green for 10 min at 4°C. We used 0.05% Triton X-100 (T) as 100% permeabilization. The percentages of permeabilized cells are shown.

Effect of TFQ on plasma membrane integrity of T. brucei.

We next evaluated whether TFQ could induce ATP liberation to the medium as a result of an effect on plasma membrane integrity. To evaluate the effect of TFQ at this level, we monitored the entrance of the vital dye Sytox green (molecular mass = 600 g/mol) into the cytoplasm of BSF parasites, while considering that this probe does not penetrate into intact cells. As for the ATP assay, BSF parasites were treated with 10 μM TFQ at different time points. Flow cytometry showed a time-dependent increase of Sytox green fluorescence (Fig. 7b), which after 120 min reached values close to those obtained with the control for maximal permeabilization (obtained using 0.05% Triton X-100); this result also supports the observations of necrotic cell death produced by TFQ. In addition, after 10 min of TFQ treatment, only 6.82% of the parasites presented lesions on their plasma membranes, ruling out the possibility that ATP release after the permeabilization of the plasma membrane was responsible for ΔΨm depolarization. Thus, results concerning ATP levels and plasma membrane integrity suggest that the decreases in ATP levels are due to the release of ATP by necrotic cells, but not the cause of mitochondria depolarization.

TFQ increases free cytosolic Ca²⁺ levels in T. brucei.

TFQ distribution in the calcium-storage organelles of BSF parasites (Fig. 2) could alter Ca²⁺ homeostasis. Indeed, we have previously shown that TFQ accumulation in Leishmania acidocalcisomes causes Ca²⁺ to be released into the cytosol (17). Furthermore, changes in the ΔΨm of T. brucei BSF are associated with alterations in intracellular Ca²⁺ homeostasis (43). We therefore analyzed the effect of TFQ on Ca²⁺ homeostasis in T. brucei BSF. Cytosolic Ca²⁺ levels were measured using the fluorescent probe Fluo-4. BSF parasites treated for 10 min with 5 and 10 μM TFQ showed an increase of cytosolic Ca²⁺ levels compared to untreated control parasites (P < 0.05) (Fig. 8). This experiment was repeated in the presence of EGTA, a chelator of extracellular Ca²⁺. After treatment with TFQ, we only observed a slight difference in the increase of cytosolic Ca²⁺ concentrations between the samples pretreated with EGTA and those which were not. In contrast, the effect of TFQ on cytosolic Ca²⁺ levels was abolished in the presence of BAPTA-AM, a cell-permeant chelator which is highly selective for Ca²⁺ (Fig. 8). These results strongly suggest that the mobilization of Ca²⁺ from intracellular reserves was the main reason for the observed increase of cytosolic Ca²⁺ concentrations. Calcium ions are potent signal molecules that have the ability to regulate a wide range of cellular activities in eukaryotic cells (44). Cell survival has been described as depending on the ability to carefully control cytosolic free Ca²⁺ concentrations. Therefore, the transient disruption of Ca²⁺ homeostasis can be used as a signal, while prolonged exposure to elevated cytosolic free Ca²⁺ concentrations can result in cell death (44–46). TFQ accumulation within acidocalcisomes of T. brucei BSF (Fig. 2a) probably induces alkalinization of the organelle, as described for Leishmania (17), with the subsequent release of Ca²⁺ into the cytosol. The trypanosome mitochondrion maintains a low Ca²⁺ resting level but transiently accumulates large quantities of Ca²⁺ from the cytosol after influx across the plasma membrane or after release from the acidocalcisomes (47). Ca²⁺ uptake is mediated by the ATPase-dependent energization of the inner mitochondrial membrane, and is associated with the depolarization of the ΔΨm (43). It appears that a TFQ-induced ΔΨm depolarization blocks calcium from entering into the mitochondria and therefore prevents restoration of calcium homeostasis.

FIG 8.

TFQ increases cytosolic Ca²⁺ levels. Fluo4-preloaded BSF were treated without (C) or with 5 and 10 μM TFQ for 10 min at 37°C. The experiments were assessed with or without the presence of the Ca²⁺ chelator EGTA or BAPTA-AM. The fluorescence intensity was determined by flow cytometry analysis, as described in Materials and Methods. The geometric mean (Geo Mean) channel fluorescence values ± the SD from three experiments are shown. Significant differences were determined by using the Student t test (*, P < 0.05 [versus control parasites]; §, P < 0.05 [versus parasites treated with 5 μM TFQ]; †, P < 0.05 [versus parasites treated with 10 μM TFQ]).

TFQ increases production of ROS in T. brucei.

Calcium and ΔΨm alterations are usually associated with the production of ROS in programmed cell death processes affecting trypanosomatids (48). ROS production has also been described as a hallmark for cell death by necrosis (30). We used the ROS-sensitive probe H₂DCF-DA, a probe not highly specific but extensively used in the parasites Leishmania (49, 50) and T. brucei (51, 52), in order to determine TFQ's potential as an inductor of ROS production in parasites. for 10 min with 1, 5, and 10 μM TFQ. Flow cytometry revealed that 1, 5, and 10 μM TFQ induced significant (P < 0.05) ROS production (Fig. 9). We also used BAPTA-AM to evaluate whether cytosolic calcium levels play a role in ROS production. We observed that a pretreatment with BAPTA-AM prevented TFQ-induced ROS production (Fig. 9). These results suggest that ROS generation is triggered by the TFQ-induced increase of cytosolic calcium levels.

FIG 9.

TFQ induces ROS generation. ROS levels were measured using the specific fluorescent dye H₂DCF-DA. BSF preloaded with 10 μM H₂DCF-DA (control [C]) were pretreated with or without the Ca²⁺ chelator BAPTA-AM and with 1, 5, and 10 μM TFQ for 10 min at 37°C. The fluorescence intensity was determined by flow cytometry analysis as described in Materials and Methods. (a) Histograms from a representative experiment of three independent ones. (b) Geometric mean channel fluorescence values ± the SD from three experiments. Significant differences were determined by using the Student t test (*, P < 0.05 [ versus control parasites]; †, P < 0.05 [versus parasites treated with 10 μM TFQ]).

Conclusions.

Tafenoquine kills different T. brucei spp. at concentrations in the nanomolar range. This is the first report of an 8-aminoquinoline with significant in vitro activity against T. brucei, the causative agent of HAT. Since TFQ is a promising oral drug currently in phase III clinical trials against malaria, these results support further in vivo studies using both TFQ alone and in combination with other drugs to evaluate the potential of repositioning this drug as a treatment for HAT (and nagana). In addition, the information gathered about TFQ's mechanism of action in T. brucei has provided insights into a wide and multitarget drug action that induces a necrotic process with cell membrane disintegration and loss of cytoplasmic contents which results in cell death.