Tafenoquine Is a Promising Drug Candidate for the Treatment of Babesiosis

ABSTRACT

Due to drug resistance, commonly used anti-Babesia drugs have limited efficacy against babesiosis and inflict severe side effects. Tafenoquine (TAF) was approved by the U.S. Food and Drug Administration in 2018 for the radical cure of Plasmodium vivax infection and for malaria prophylaxis. Here, we evaluated the efficacy of TAF for the treatment of Babesia infection and elucidated the suspected mechanisms of TAF activity against Babesia parasites. Parasitemia and survival rates of Babesia rodhaini-infected BALB/c and SCID mice were used to explore the role of the immune response in Babesia infection after TAF treatment. Parasitemia, survival rates, body weight, vital signs, complete blood count, and blood biochemistry of B. gibsoni-infected splenectomized dogs were determined to evaluate the anti-Babesia activity and side effects of TAF. Then, to understand the mechanism of TAF activity, hydrogen peroxide was used as an oxidizer for short-term B. rodhaini incubation in vitro, and the expression levels of antioxidant enzymes were confirmed using B. microti-infected mice by reverse transcription-quantitative PCR (qRT-PCR). Acute B. rodhaini and B. gibsoni infections were rapidly eliminated with TAF administration. Repeated administration of TAF or a combination therapy with other antibabesial agents is still needed to avoid a potentially fatal recurrence for immunocompromised hosts. Caution about hyperkalemia should be taken during TAF treatment for Babesia infection. TAF possesses a babesicidal effect that may be related to drug-induced oxidative stress. Considering the lower frequency of glucose-6-phosphate dehydrogenase deficiency in animals compared to that in humans, TAF use on Babesia-infected farm animals and pets is eagerly anticipated.

INTRODUCTION

Babesia are tick-transmitted protozoan parasites that infect mammals and are associated with great economic, veterinary, and medical effects worldwide (1). Human babesiosis, also transmitted by blood transfusion, is an increasing global public health concern (2). Clinical symptoms of Babesia infection are characterized by the presence of fever accompanied by fatigue and malaise, followed by anemia, hemoglobinuria, anorexia, emaciation, and vomiting. The acute hemolytic phase of severe infection often results in death (3). Due to the large diversity of Babesia species and their vertebrate hosts, a broad-spectrum-activity drug is ideal in treating babesiosis (4). Babesiosis is usually treated for 7 to 10 days with high doses of a combination of either atovaquone and azithromycin or clindamycin and quinine (5). Increasing reports of side effects, drug resistance, and recurrence are being observed in clinical cases, the severity of which has been shown to be dose and duration dependent (6). Several new drug candidates, such as clofazimine, trans-chalcone, 17-dimethylaminoethylamino-17-demethoxygeldanamyci (17-DMAG), and nitidine chloride, have been reported for the treatment of babesiosis in vivo (7–10). However, the efficacies of these drugs in controlling Babesia infection have only been evaluated during the preclinical phase of babesiosis. Additionally, the administration of multiple consecutive doses is a risk factor for the emergence and spread of drug-resistant parasites, which could limit the useful lifetime of these drugs, as is the case with atovaquone (11). Still, effective drugs that require less-frequent dosing are urgently needed for the treatment of babesiosis, especially in clinical cases.

In 2018, the U.S. Food and Drug Administration (FDA) approved tafenoquine (TAF) for the radical cure of Plasmodium vivax infection and for malaria prophylaxis (12). TAF is an 8-aminoquinoline that inhibits multiple life stages of various Plasmodium species. Moreover, its much longer half-life (14 to 28 days) allows for single-dose treatment. Unfortunately, despite its approval for long-term use, the mechanisms behind TAF’s activity and side effects are still largely unknown, even in malaria (13). In addition, TAF can cause severe hemolytic anemia in people with glucose-6-phosphate dehydrogenase (G6PD) deficiency. Therefore, TAF use in people with G6PD deficiency or an unknown G6PD status is contraindicated (14).

Previous studies have shown the broad-spectrum activity of TAF against protozoan parasites such as Trypanosoma (15), Toxoplasma gondii (16), and Babesia microti (17–19). In order to comprehensively evaluate the efficacy of TAF activity against Babesia infection, two species of Babesia parasites, namely, B. rodhaini (rodent) and B. gibsoni (canine), were used in this study. B. rodhaini, a species closely related to B. microti, is the most pathogenic among Babesia parasites (20). B. gibsoni has also been considered an important pathogen that affects canine health with a global distribution (21). In this study, we evaluated the efficacy of TAF for the treatment of Babesia infection and elucidated the suspected mechanisms of TAF activity against Babesia parasites. Herein, we demonstrate that TAF is a promising drug candidate for future treatment against babesiosis.

RESULTS

Effects of TAF on B. rodhaini infection in immunocompetent and immunocompromised mice.

A single dose of 20 mg/kg TAF did not halt B. rodhaini recurrence in either BALB/c (Fig. 1A) or CB17/IcrJcl-Prkdc^scid (SCID) mice (Fig. 1C). Recurrent parasites were cleared by the immune response in the immunocompetent BALB/c mice, leading to curtailed parasitemia (Fig. 1A) and complete survival (Fig. 1B). In contrast, recurrent parasitemia in B. rodhaini-infected SCID mice reached 70.5% at 23 days postinfection (Fig. 1C), and all mice eventually succumbed to the recurrent disease within 26 days postinfection (Fig. 1D). A starry sky appearance in the spleen of untreated B. rodhaini-infected mice, which was formed by macrophage infiltration, showed the presence of diffused, punctate calcifications in a high-magnification image (Fig. S1). Conversely, TAF-treated mice had no abnormalities, similar to the uninfected mice (Fig. S1). TAF treatment prevented the collapse of the immune system caused by infection, which then mediated the radical cure of recurrent parasites in BALB/c mice.

FIG 1.

Parasitemia and survival rates of B. rodhaini infection in BALB/c and SCID mice with tafenoquine (TAF) treatment. Course of parasitemia of single-dose TAF treatment at 3 days postinfection during B. rodhaini infection in BALB/c mice (panel A; n = 6) and SCID mice (panel C; n = 5). Control (Ctrl) group is comprised of untreated BALB/c mice (panel A; n = 6) and SCID mice (panel C; n = 5). Survival of B. rodhaini–infected BALB/c mice (B) and SCID mice (D) treated with single-dose TAF in panels A and C, respectively. Survival rates were calculated using the Kaplan-Meier method, with regard to the log-rank test.

Effects of TAF on B. gibsoni infection in immunocompromised dogs.

TAF contributed to a reduction in the level of parasitemia from 14.8% to undetectable within 72 h; although recurrent parasites from one dog were observed on day 48, those parasites were recleared following the third delivery, and all dogs were protected from death until day 78 (Fig. 2A and B). Different from the symptoms alleviated within few days in treated dogs, the symptoms in the untreated group were more severe, lasting clinical symptoms, including anorexia, hemoglobinuria, and vomiting (Table S1), as well as emaciation, tachycardia, dyspnea, anemia, and thrombocytopenia (Fig. 3 and 4). Moreover, fever was not detected from all dogs, which may be related to the absence of immunity in immunocompromised dogs (Fig. 3B). One dog died at 22 days postinfection, while euthanization was performed on the other dogs after 2 weeks of anorexia caused more than 25% body weight loss at 30 days postinfection (Fig. 2B and Fig. 3A). Similar pulse-respiration quotient (PRQ) values with the baseline (the average PRQ value on day 0) indicated the steady state of cardiorespiratory system after curing the disease (Fig. S2). Administration of TAF avoided B. gibsoni infection-induced hepatocellular injury in dogs (Fig. 5A, B, and C and Fig. S3A). No renal function effects due to B. gibsoni infection were recorded, as creatinine (Cre) was not elevated (Fig. 5E); only blood urea nitrogen (BUN) levels were increased, which may be an indication of dehydration and infection caused by the deterioration of the condition (Fig. 5D). Electrolyte imbalance after infection appeared as increased levels of potassium (K⁺) in both treated and untreated dogs (Fig. 5F, G, and H). Hyperkalemia was not confirmed as a side effect of TAF in a healthy dog (Fig. S3B), but attention should be paid when administering TAF for babesiosis, at least for canine babesiosis.

FIG 2.

Parasitemia and survival rates of B. gibsoni infection in dogs with tafenoquine (TAF) treatment. (A) Course of parasitemia of TAF treatment against B. gibsoni infection in dogs (n = 3). Dosages: 11.3 mg/kg (day 12), 11.5 mg/kg (day 19), and 9.0 mg/kg (day 48). Control (Ctrl) group is comprised of untreated dogs (n = 3). (B) Survival of B. gibsoni-infected dogs in panel A. Survival rates were calculated using the Kaplan-Meier method, with regard to the log-rank test. Arrows: one tablet of Arakoda (100 mg TAF) was administered.

FIG 3.

Body weight and vital signs of dogs. (A) Body weight. (B) Temperature. (C) Pulse. (D) Respiration. Dotted line: Minimum count in A, maximum count in B, C, and D.

FIG 4.

Complete blood count of dogs. (A) Red blood cells (RBCs). (B) Hemoglobin. (C) Hematocrit. (D) Platelets. Dotted line: minimum count.

FIG 5.

Blood biochemistry of dog sera. (A) Alanine transaminase (ALT). (B) Alkaline phosphatase (ALP). (C) Aspartate transaminase (AST). (D) Blood urea nitrogen (BUN). (E) Creatinine (Cre). (F) Sodium (Na⁺). (G) Chloride (Cl⁻). (H) Potassium (K⁺). Dotted line: reference range.

Babesia elimination occurs through TAF-induced oxidative stress.

Similar to that observed in a previous study (18), the appearance of a vacuole-like aberrant parasite phenotype with only faint chromatin staining was visible within 48 h of TAF treatment (Fig. 6A and Fig. S4). Additionally, a similar phenotype can be observed in a short-term incubation with hydrogen peroxide (H₂O₂), a well-known oxidizing agent, but it was unobservable in untreated mice (Fig. 6A). One hundred parasites were measured for each group. The size of vacuole-like parasites revealed no significant differences between the TAF treatment and H₂O₂ incubation groups (Fig. 6B). However, the vacuole-like parasites were significantly larger than the parasites in untreated mice (P < 0.0001) (Fig. 6B). Due to the unavailability of the B. rodhaini genome sequence, the expression level of antioxidant genes was determined using B. microti. Significant TAF-induced upregulation of relative antioxidant enzymes in B. microti, including thioredoxin peroxidases (TPxs: TPx1 and TPx2) and thioredoxin reductase (TrxR), was confirmed by quantitative reverse transcription-PCR (qRT-PCR) (P < 0.0001) (Fig. 6C).

FIG 6.

Babesia elimination occurs through tafenoquine (TAF)-induced oxidative stress. (A) Giemsa-stained blood smears of B. rodhaini-infected mice from vehicle alone (Ctrl), 24 h post single-dose TAF delivery, and after 2-h incubation with 100 μM hydrogen peroxide (H₂O₂) in vitro (bar: 10 μm). (B) Sizes of 100 parasites in each group measured with ImageJ software in panel A. (C) Significant upregulation of thioredoxin peroxidase 1 (TPx1), TPx2, and thioredoxin reductase (TrxR) in B. microti-infected BALB/c mice (n = 6) 24 h post single-dose TAF delivery. Statistically significant differences between the means of variables determined by t test in panels B and C. ns, not significant (P > 0.05); ****, P < 0.0001.

DISCUSSION

TAF has been under development since 1980 and is only the second drug of its class to receive U.S. FDA approval. TAF proved to be as effective as primaquine (PQ) but with the advantage of requiring a shorter treatment duration in P. vivax infection, which is an important consideration for patient compliance and cost (22). Recently, TAF was proposed to revolutionize the treatment of human babesiosis (18). However, a limitation is that TAF can induce severe hemolytic anemia in individuals with G6PD deficiency, a genetic disorder obtained through X-linked recessive inheritance in hemizygous males and homozygous females in humans (23). Surprisingly, G6PD deficiency is exceptionally rare in animals (24, 25), since their allele frequency is lower than that in humans (26). As the major impact of babesiosis worldwide is more apparent on animals rather than on humans (3), a single dose of 10 mg/kg TAF on pets may contribute greatly to canine babesiosis and feline babesiosis control. In addition, given the long half-life of TAF, a safe and low dose of TAF delivery, after determining the median lethal dose (LD₅₀), may help prevent bovine babesiosis, ovine babesiosis, and equine babesiosis in farm animals during the tick season. On the other hand, considering the high efficacy of TAF for the treatment of B. microti in the hamster model via intramuscular injection (17), other routes of administration should be explored which may potentially be more effective than oral and could provide more options to support widespread use in veterinary medicine.

The precise mechanism of TAF activity against Babesia and blood-stage malaria parasites is still unknown. There are a number of hypotheses about TAF’s mechanism of action in malaria; a leading hypothesis is that TAF exerts its effect by binding to the hematin μ-oxo dimer and inhibiting hematin polymerization to induce apoptosis-like death of the Plasmodium parasites as chloroquine (CQ) (27). However, this leading hypothesis mechanism may not be applicable for babesiosis. First, there is clearly a huge difference between CQ and TAF in Babesia elimination. While TAF is highly efficient against Babesia, the parasite does not respond to CQ (28). Second, hematin polymerization is a malaria-specific process that enables the detoxification of free heme following its release in the lysosomal digestive vacuole during hemoglobin degradation (29). On the other hand, due to the absence of enzymes of the de novo heme biosynthesis pathway in the Babesia genome, Babesia parasites obtain all heme from the host hemoglobin digestion and crystallize heme to the disposal product by the presence of an unclear alternative route of heme disposal (30). Different from Plasmodium parasites, Babesia parasites lack not only the process for hematin polymerization but also the food vacuole, where the hemozoin crystals form (31). Therefore, the hypothesized TAF inhibition of hematin polymerization may be limited to blood-stage malaria parasites and unlikely explains the observed anti-Babesia activity.

An alternative hypothesis is that TAF is activated in the red blood cells (RBCs) to target the parasite through oxidative stress (27). Babesia lives in an oxygen-rich environment within the mammalian host’s blood. As a result, the parasite is exposed to the toxic effects of reactive oxygen species (ROS), mainly contributed by H₂O₂, which can cause oxidative damage to membrane lipids, nucleic acids, and proteins (32). Babesia evades oxidative damage through antioxidant enzymes, such as superoxide dismutase (SOD), TPxs, and TrxR (33–36). Although two other important antioxidant enzymes, catalase (CAT) and glutathione peroxidase (GPx), were also mentioned (37), no homology to a specific CAT or GPx was identified from the genome database of Babesia, suggesting that Babesia might lack those enzymes. Surprisingly, another important antioxidant enzyme, glutathione reductase (GR), is also lacking in the Babesia genome, which indicates the blocking of glutathione (GSH) regeneration from oxidized glutathione (GSSG) (32). Babesia may not utilize GSH from the host RBCs, since the parasite membrane is permeable to neither host GSH nor γ-glutamylcysteine (γ-GC). The de novo synthesis of GSH in Babesia by consecutive reactions is facilitated by the enzymes glutamate-cysteine ligase catalytic subunit (GCLC) and glutathione synthase (GS), independent of GSH biosynthesis in the host RBCs. Only glutaredoxins (Grxs) and other unknown enzymes might play vital roles in the process of transforming transferase GSH to GSSG. GSH is one of the few known nonenzymatic antioxidant molecules (38). The GSH redox cycle is a key molecule within redox homeostasis to minimize oxidative damage. However, the lack of GPx and GR leads to an incomplete GSH redox cycle. Therefore, the antioxidant capacity of Babesia parasites is probably weaker than that of the host RBCs, as only the thioredoxin (Trx) redox cycle is present (Fig. 7). Finally, intracellular accumulated ROS oxidative damage to membrane lipids, nucleic acids, and organelles is characterized by the appearance of a vacuole-like aberrant parasite phenotype. Further evidence is the expression of essential antioxidant enzymes in the Trx redox cycle, including TPx1, TPx2, and TrxR, being significantly upregulated by the induction with TAF. An intuitive explanation for this hypothesis could be that the production of NADPH by G6PD is essential in RBCs, which helps protect RBCs from damage by ROS (23). Therefore, TAF is contraindicated in persons with G6PD deficiency because it might cause severe hemolytic anemia through the accumulation of ROS-induced NADPH depletion. In summary, we believe that TAF-induced oxidative stress can only kill the Babesia parasites, with no effect to host RBCs with G6PD activity, as the NADPH produced by G6PD in G6PD-sufficient individuals relieves the TAF-induced oxidative stress (Fig. 7).

FIG 7.

Proposed mechanisms underlying the activity of tafenoquine (TAF) in bloodstream infection. Host antioxidant enzymes minimize TAF-induced oxidative stress to maintain the glucose-6-phosphate dehydrogenase-mediated redox homeostasis in host red blood cells. Due to the lack of essential antioxidant enzymes, TAF-induced oxidative stress damages membrane lipids, nucleic acids, and organelles of Babesia parasites, resulting in death. ROS, reactive oxygen species; SOD, superoxide dismutase; H₂O₂, hydrogen peroxide; GCLC, glutamate-cysteine ligase catalytic subunit; γ-GC, γ-glutamylcysteine; GS, glutathione synthase; GSH, glutathione; GSSG, oxidized glutathione; GPx, glutathione peroxidase; GST, glutathione S-transferase; Grxs, glutaredoxins; GR, glutathione reductase; G6PD, glucose-6-phosphate dehydrogenase; CAT, catalase; TPxs, thioredoxin peroxidases; TrxR, thioredoxin reductase.

Overall, acute B. rodhaini and B. gibsoni infections were rapidly cured with TAF administration. A single dose of TAF treatment prevented the collapse of the immune system caused by infection, which then mediated the radical cure of recurrent parasites in immunocompetent mice. However, repeated administration of TAF or a combination therapy with other antibabesial agents is still needed to avoid a potentially fatal recurrence for immunocompromised hosts. Babesia infection-related hemolysis may have been the primary reason for hyperkalemia, which appeared earlier with the treatment of TAF. Therefore, as an overlooked risk of Babesia infection, caution about hyperkalemia should be heeded during TAF treatment. Additionally, we conclude that the efficient activity of TAF in Babesia may be mediated through an oxidative stress-dependent mechanism. Undoubtedly, the characterization of parasites with resistance to TAF would advance our mechanistic understanding, and it could generate testable hypotheses about molecular targets. Unfortunately, TAF-resistant parasites have not been observed in this study. In addition, due to the broad-spectrum activity of TAF in targeting multiple stages of Plasmodium parasites (13), exploring the ability of TAF to block the transmission of Babesia gametocytes to ticks may broaden our horizons of unidentified molecular targets. A more complete profiling of the TAF metabolism and the enzymes involved would help elucidate its mechanism of action, and describing how Babesia reacts to oxidative stress is imperative.

MATERIALS AND METHODS

Ethics statement.

Euthanization of dogs was performed by an overdose intravenous injection of Somnopentyl (Kyoritsu Seiyaku Corporation, Tokyo, Japan). All procedures were carried out according to the ethical guidelines approved by the Obihiro University of Agriculture and Veterinary Medicine (permit numbers: animal experiment, 20-127; DNA experiment, 1723-4 and 1724-4; pathogen, 201709-6 and 201712-5).

Animals and parasites.

Female BALB/c mice and female SCID mice of 6 to 9 weeks of age (CLEA, Tokyo, Japan) were used. Splenectomized female beagle dogs (14 months of age) were purchased from Kitayama Labes Co., Ltd. (Yamaguchi, Japan). BALB/c mice were divided into 4 groups of 6 each for both B. rodhaini and B. microti infections. SCID mice were divided into 2 groups of 5 each for B. rodhaini infection. Beagle dogs were divided into 2 groups of 3 each for B. gibsoni infection. The B. rodhaini Australian strain (20) was injected intraperitoneally (10⁷ parasites) to BALB/c and SCID mice when parasitemia reached 50% in the donor mice. The B. microti Peabody mjr strain (ATCC PRA-99) was intraperitoneally administered (10⁸ parasites) to BALB/c mice when parasitemia was 20 to 30% in the donor mice. The B. gibsoni NRCPD strain (39) was injected intravenously (8 × 10⁸ parasites) from infected donor dog.

Tafenoquine and Arakoda.

Tafenoquine succinate (TAF, Sigma-Aldrich, Tokyo, Japan) was dissolved in dimethyl sulfoxide (DMSO) to prepare a 40 mg/ml stock solution and diluted with Milli-Q water for B. rodhaini treatment. Arakoda (60 Degrees Pharmaceuticals, Washington, DC, USA), a high-dose (100 mg of active ingredient) pharmaceutical tablet for human use, was orally administered for B. gibsoni treatment.

Evaluation of anti-Babesia effects.

The experiments against B. rodhaini infection were conducted via oral delivery of a single dose of 20 mg/kg TAF when the parasitemia of infected BALB/c and SCID mice exceeded 10%. For the treatment of B. gibsoni infection, the first tablet of Arakoda (100 mg TAF) was delivered at 12 days postinfection, the first manifestation of hemoglobinuria. Then, the second delivery was done with a 1-week interval at 19 days postinfection to avoid the toxic accumulation of repeated administration in a short period. Due to the observation of recurrent parasites (parasitemia was 0.22%) from one dog on day 48, the third delivery was conducted for all three dogs at the same day. The presence of parasites in mice and dogs was monitored by examining 5,000 RBCs using Giemsa-stained thin smears.

Histopathology of BALB/c mouse spleen.

The 10% formalin-fixed mouse spleen was embedded in paraffin using standard procedures. The specimens were cut frontally into sections 4 μm thick, deparaffinized, stained with hematoxylin-eosin, and assessed using a Microphot-FX (Nikon, Tokyo, Japan) equipped with a Digital Sight DS-5M camera (Nikon).

Body weight, vital signs, complete blood count, and blood biochemistry in dogs.

The monitoring of body weight, temperature, pulse, and respiration of each dog was performed every 3 days. RBCs, hemoglobin, hematocrit, and platelets were measured using an automated veterinary hematology analyzer (MEK-6550 Celltac α, Nihon Kohden, Tokyo, Japan). An automated clinical chemistry analyzer (DRI-CHEM 7000 V, Fujifilm, Tokyo, Japan) was used to assess levels of alanine transaminase (ALT), alkaline phosphatase (ALP), aspartate transaminase (AST), BUN, Cre, sodium (Na⁺), K⁺, and chloride (Cl⁻) in dog sera.

Hydrogen peroxide incubation of B. rodhaini.

Blood was collected from B. rodhaini-infected mice by cardiac puncture after diethyl ether (Wako, Osaka, Japan) anesthesia. The infected RBCs (iRBCs) were subjected to consecutive centrifugations and washed three times with phosphate-buffered saline (PBS). A total of 50 μl (5 × 10⁸) of B. rodhaini iRBCs was subjected to short-term incubation with 100 μM H₂O₂ (Wako) for 2 h in a 24-well plate supplemented with RPMI 1640 culture medium at 37°C in a humidified atmosphere (5% CO₂ and 5% O₂).

qRT-PCR assay of antioxidant enzymes.

To normalize the expression levels of relative antioxidant enzymes in B. microti, β-tubulin was used as an internal control. The specific primers used to measure the expression levels of antioxidant enzymes and β-tubulin are listed in Table S2. The qRT-PCR was performed with a PrimeScript RT master mix (TaKaRa Bio, Shiga, Japan) and PowerUp SYBR Green master mix (Thermo Fisher Scientific, Waltham, MA, USA) using a 7900HT Fast real-time PCR system (Applied Biosystems, Foster City, CA, USA) with total RNA from untreated mice or mice treated with TAF for 24 h.

Statistical analysis.

Data analysis was performed using GraphPad Prism (La Jolla, CA, USA) version 7. Statistically significant differences between the means of variables were determined by t test. Survival rates were calculated using the Kaplan-Meier method, with regard to the log-rank test. A difference was declared significant when the P value was less than 0.05.