Introduction

Babesia is an apicomplexan tick-transmitted hemoparasite, which not only impacts the livestock economy but also causes an emerging disease in humans (Jalovecka et al., 2019). Babesia is a global pathogen that is more prevalent in certain regions such as Asia, Europe, and North America. However, as climate change brings with it higher temperatures and humidity, ticks and their reservoir hosts are anticipated to expand northward for survival and activity (Drews et al., 2023). The disease it causes is known as babesiosis, commonly characterized by fever and hemolytic anemia, but chronic infections are asymptomatic (Almazán et al., 2022). In acute cases, up to 20% of patients die, particularly individuals with compromised immune systems (Krause, 2019).

Currently, the most common drugs for the treatment of human babesiosis include a combination of atovaquone (ATQ) plus azithromycin (AZM) or clindamycin plus quinine (Holbrook et al., 2023). Still, these recommended treatments often result in various problems. For instance, multiple mutations in the B. microti cytochrome b, which was associated with the atovaquone-target region, were discovered in an atovaquone-treated relapse patient (Holbrook et al., 2023). The combination of clindamycin and quinine is the last resort for patients with severe symptoms. Despite its efficacy, some had adverse drug reactions after taking these drugs (Vannier and Krause, 2012). The 8-aminoquinoline analog, tafenoquine (TAF), was found to be effective in curing the patient’s B. microti infection that was resistant to azithromycin and atovaquone with a 200 mg weekly dose (Marcos et al., 2022). However, the efficacy of TAF treatment may vary between individuals and cases and its contraindication in G6PD-deficient patients limits its clinical use. Due to the aforementioned issues, it is undoubtedly urgent to search for compounds that treat infections caused by Babesia spp. while simultaneously minimizing the detrimental effects of antibabesial drugs.

Spiroindolone cipargamin (CIP), a promising antimalarial, has been found to effectively suppress the growth of all strains of Plasmodium falciparum and P. vivax with potency in the low nanomolar ranges (Rosling et al., 2018; Rottmann et al., 2010). The assessment of the pharmacokinetic characteristics of the drug indicated that oral administration of CIP had good absorption, a prolonged half-life, and exceptional bioavailability (Schmitt et al., 2022). Following oral CIP treatment (30 mg daily for three days) in adults with simple P. falciparum or P. vivax malaria, the parasitemia was rapidly cleared in a phase II trial (Schmitt et al., 2022). Currently, a clinical trial is being conducted for the intravenous administration of CIP as a treatment for severe malaria patients (ClinicalTrials, 2024). CIP was also evaluated for treating toxoplasmosis as a second-line medicine in cases where there are intolerable toxicity issues or allergies to the currently used treatments. Five days post-Toxoplasma gondii infection, CIP-treated mice (100 mg/kg/day on the day of and the day after infection) had 90% fewer parasites than the control mice (Zhou et al., 2014).

Due to its excellent efficacy against other apicomplexan parasites, we hypothesize that CIP may also be effective as an antibabesial drug. Therefore, the objectives of this study were to determine whether CIP could inhibit the growth of Babesia spp., namely B. bovis and B. gibsoni in vitro and B. microti and B. rodhaini in vivo, and identify the inhibitory mechanisms of CIP on Babesia parasites (BgATP4) by monitoring the Na⁺ and H⁺ balance.

Results

Inhibitory efficacy of CIP on B. bovis and B. gibsoni in vitro

The in vitro activity of CIP against B. bovis and B. gibsoni showed a steep growth inhibition curve with half inhibitory concentration (IC₅₀) values of 20.2 ± 1.4 nM (Figure 1A) and 69.4 ± 2.2 nM (Figure 1B), respectively. The CC₅₀ value of CIP on MDCK cells was 38.7 ± 2.0 μM (Figure supplement 1), resulting in a predicted selectivity index of 557.6. Furthermore, at a concentration of 100 μM, CIP exhibited a low erythrocyte hemolysis rate of 0.11 ± 0.03 % (not shown).

Figure 1.

CIP effect on B. microti and B. rodhaini infections in vivo

Concurrently, CIP showed effective inhibition on B. microti and B. rodhaini in vivo. The parasitemia of B. microti-infected BALB/c mice grew dramatically in the vehicle-treated control group and peaked at 10 DPI (35.88 ± 4.32%) (Figure 1C). On the other hand, the treatment with CIP (20 mg/kg of body weight) or atovaquone (ATO) plus azithromycin (AZI) administered orally resulted in a significantly lower peak parasitemia, 1.06 ± 0.20% and 1.61 ± 0.20%, respectively. Hematocrit (HCT) variations were tracked every 4 days as an indicator of anemia in B. microti-infected mice. The vehicle-treated group showed a drop in HCT levels from 8 DPI to 24 DPI, with statistically significant decreases at 12 and 16 DPI (P < 0.01). No significant reduction in HCT levels was observed in the CIP-treated group or the ATO plus AZI-treated group (Figure 1D). This suggests that the administration of CIP could control B. microti infection and prevent anemia from developing in B. microti-infected mice. BALB/c mice infected with B. rodhaini treated with sesame oil or ATO plus AZI showed high parasitemia, 90.73 ± 1.97%, and 86.23 ± 3.06%, respectively (Figure 1E), and all mice died within 10 DPI (Figure 1F). CIP treatment in B. rodhaini-infected mice precluded the emergence of parasitemia for the following eight days (Figure 1E), which led to 66.67% of mice surviving the challenge infection (Figure 1F). At 14 DPI, parasites had recurred in all CIP-treated B. rodhaini-infected mice (10.32 ± 15.51%), which were eventually cleared as indicated by undetectable parasites at 18 DPI (Figure 1E).

Identification of B. gibsoni ATP4 mutation in CIP-resistant strains

We sequenced the B. gibsoni ATP4 gene from two resistant parasite wells. A single nucleotide variant (SNV) in BgATP4 with a substitution at position 2,761 (from C to G) was found − a nonsynonymous coding change from leucine to valine (L921V) (Figure 2B). In another resistant strain, the mutation showed in the same position. However, the nucleotide substitution was from C to A, and the coding changed from leucine to isoleucine (L921I) (Figure 2C). BgATP4^L921V and BgATP4^L921I lines were tested for their susceptibility to CIP, and had IC₅₀ values of 421.0 ± 15.88 nM and 887.9 ± 61.97 nM, respectively (Figure 2D). These findings demonstrate a 6.1-and 12.8-fold reduction in CIP sensitivity of resistant parasite lines BgATP4^L921V and BgATP4^L921I.

Figure 2.

The effect of CIP on BgATP4^WT, BgATP4^L921V and BgATP4^L921I function

Microscopic observation of thin blood smears was performed to determine the morphological changes of B. gibsoni exposed to CIP. The CIP-treated parasites became swollen after incubation with the drug for 72 hr (Figure 3A). For both the treatment and control groups, one hundred parasites were measured. The mean size of treated parasites was notably bigger than the parasites in the untreated group (P ˂ 0.0001) (Figure 3B). The addition of the ATP4 inhibitor CIP resulted in a time-dependent increase in the concentrations of [Na⁺]_i in wild-type B. gibsoni (Figure 3C), with improved signal-to-noise ratios at the higher drug concentration of 20 nM. We observed that the Na⁺ concentrations in both BgATP4^L921V and BgATP4^L921I lines decreased compared to those of the control BgATP4^WT line after being exposed to 20 nM CIP for 20 min, with a significantly lower Na⁺ concentration in BgATP4^L921I (P = 0.0087) (Figure 3E). We also demonstrated here that the addition of CIP in wild-type B. gibsoni caused an increase in the cytosolic pH (Figure 3D). Specifically, 4 min after the drug was added, the average pH of 20 nM CIP group reached as high as 7.278, while the 1 nM CIP group reached 7.089 and the untreated group reached 7.062. The pH values of the 20 nM CIP group were consistently higher than those of the other two groups, although declining with time. In resistant lines, a 20-min exposure to 20 nM CIP caused small changes in the pH values compared to the wild-type line (Figure 3F), with the BgATP4^L921I line (7.048 ± 0.042) having a notably lower pH value (P = 0.0229).

Figure 3.

Multiple sequence alignment of Babesia ATP4 and molecular docking

The whole amino acid sequence of BgATP4 (GenBank: KAK1443404.1) shared identity values of 29.75%, 49.40%, 49.67%, 62.21%, and 52.47% with Homo sapiens ATP4 (GenBank: NM_000704.3), P. falciparum ATP4 (GenBank: PF3D7_1211900), T. gondii ATP4 (GenBank: XP_018635122.1), B. bovis ATP4 (PiroplasmaDB: BBOV_IV010020), and B. microti ATP4 (GenBank: BMR1_03g01005), respectively (Figure supplement 2).

The pLDDT value of BgATP4^WT prediction was 80.7 using Colab-fold. Multiple potential binding sites for CIP were revealed by blind docking throughout the whole protein surface (Figure supplement 3). CIP binds in close proximity to L921, as demonstrated by focused docking on this area (Figure 4A). The contribution of each residue to the predicted binding affinity in either mutant structure was reduced; the precise values of BgATP4^WT (Figure 4B), BgATP4^L921V (Figure 4C) and BgATP4^L921I (Figure 4D) were -6.43, -6.40 and -6.26 kcal/mol, respectively. The interactions of CIP from each docking simulation are shown in Table supplement 2.

Figure 4.

Discussion

The repositioning of antimalarial drugs is critical in developing novel strategies for treating babesiosis. CIP is a novel compound that inhibits Plasmodium development and has been extensively tested in Phase 1 and Phase 2 clinical trials by targeting PfATP4 (Qiu et al., 2022). Therefore, our study aimed to repurpose the antimalarial CIP by assessing its efficacy on Babesia species. The IC₅₀ values of CIP against B. bovis and B. gibsoni in vitro were lower than the previously reported IC₅₀ value of TAF (Ji et al., 2022b) and ATO against B. gibsoni (Matsuu et al., 2004). The present investigation also demonstrated that the inhibitory effects of CIP on B. microti-infected BALB/c mice were comparable to that of ATO plus AZI, the medication prescribed by the U.S. CDC. CIP was also proven to protect mice from the deadly B. rodhaini infection, with protection limited to 66.67% as recorded in the current trial. These results suggest that CIP may be a potential chemotherapy candidate for babesiosis.

In a previous study, the P. falciparum Dd2 parent line initially had a subnanomolar IC₅₀ value, after the CIP-selected cultures to limiting dilution and selected clonal lines, the highly resistant strain reported IC₅₀ values 4200-4280 fold higher than those of wild-type Dd2 parasites with a G358S mutation in the PfATP4 protein, a P. falciparum P-type Na⁺ ATPase (Qiu et al., 2022). The vital protein ATP4 is found in the parasite plasma membrane and is distinctive to the subclass of the apicomplexan parasite (Mohring et al., 2022). To date, in vitro evolution experiments using CIP have produced at least 18 parasite lines with various PfATP4 mutations (Lee and Fidock, 2016; Rottmann et al., 2010). Another finding showed that the ATP4 activity associated with TgATP4^G419S was 34 times less susceptible to CIP than that of TgATP4^WT (Qiu et al., 2022). It follows that there is a significant chance that the resistant Babesia parasites will also emerge from CIP exposure. In this study, we successfully produced two CIP-resistant strains using six independent selections. The newly discovered L921V and L921I mutations in BgATP4 decreased the CIP sensitivity by 6.1 and 12.8 times, respectively. We provided compelling evidence that natural variety mutations L921V and L921I in BgATP4 significantly affected the proteins’ susceptibility to CIP inhibition, albeit the mutational sites were different from those of P. falciparum and T. gondii.

Although ATP4 was initially recognized as a Ca²⁺ transporter (Krishna et al., 2001), current evidence suggests that ATP4 functions as an ATPase for exporting Na⁺ while importing H⁺ (Mohring et al., 2022), as well as causing a variety of other physiological perturbations including an increase in the volume of parasites and infected erythrocytes, due to the osmotic impact of the [Na⁺]_cyt increase (Dennis et al., 2018), a decline in cholesterol extrusion from the parasite plasma membrane as a result of the increase in [Na⁺]_cyt (Das et al., 2016), and an intensified rigidity of erythrocytes infected with ring-stage parasites (Zhang et al., 2016). In this study, the CIP-exposed wild-type B. gibsoni became swollen, which was identical to a prior study on P. falciparum (Dennis et al., 2018). We hypothesize that the swelling of the isolated parasites can be ascribed to the osmotic consequences of Na⁺ uptake and is contingent upon the presence of Na⁺ in the external environment. This hypothesis is supported by the following experimental results.

The output of Na⁺ and input of H⁺ diminished upon CIP-induced inhibition of PfATP4, and the continued outflow of H⁺ via V-type H⁺-ATPase led to an alkalinization that ultimately killed the parasites (Spillman et al., 2013). Our capacity to measure a time-dependent increase in the concentration of [Na⁺]_i and pH value for wild-type B. gibsoni facilitated us to gain insight into the basic mechanisms of CIP on BgATP4^WT function. Furthermore, our results validate that internal alkalinization was the main factor in Babesia death and support our hypothesis that the swollen isolated parasites were produced by Na⁺ absorption. To explore further how mutations in BgATP4 are associated with the upregulation of the parasite’s [Na⁺]_i and [H⁺]_i, we tested two BgATP4-mutant lines that were chosen previously with BgATP4 inhibitors (BgATP4^L921V and BgATP4^L921I). Due to the presence of L921V and L921I mutations in drug-resistant strains of BgATP4, the concentration of Na⁺ did not increase as much as it would have in the wild-type strain following the addition of 20 nM CIP for 20 min. As intraerythrocytic alkalinization rises, the same outcome happens. From these results, we deduced how natural mutations in BgATP4 may affect ATP4 inhibitor susceptibility by dysregulating H⁺ and Na⁺ balance, which helped parasites survive in a relatively high concentration of CIP. An illustration of the putative processes for regulating Na⁺ and H⁺ in erythrocytes infected with the Babesia parasite is presented in Figure supplement 4.

According to studies using the PfATP4 model for molecular docking, the G358S mutation results in a steric clash that lowers CIP’s binding affinity (Qiu et al., 2022). The results from the current study were similar to the PfATP4 model. The molecular docking was constructed using a ColabFold model of wild-type BgATP4, which predicted the binding mode and affinity between ATP4 protein and the ligand to provide a possible mechanistic explanation. It suggested that the L921V mutation caused changes at the atomic level, whereas the L921I mutation created a steric clash that reduced the binding affinity of CIP. The predicted affinity score of the L921I mutation was lower than that of the L921V mutation. Thus, it is possible that BgATP4^L921I has a weaker binding to the drug compared to BgATP4^L921V. These findings are consistent with the results obtained from measuring the IC₅₀ of the drug against parasites with the L921V and L921I mutations.

Materials and methods

Parasite culture

The parasites B. gibsoni (Oita strain) and B. bovis Texas strain were in vitro cultured in 24-well plates and maintained in an atmosphere of 5% CO₂ and 5% O₂ at 37 °C (Liu et al., 2018). For the in vivo studies, B. microti Peabody mjr strain-(ATCC PRA-99) and B. rodhaini Australia strain-infected RBCs (iRBCs), which were collected and diluted with phosphate-buffered saline (PBS) when the parasitemia levels in the donor mice reached ∼20% and 50%, respectively, and were intraperitoneally injected into BALB/c mice. Each BALB/c mouse was infected with 1.0 × 10⁷ B. microti or B. rodhaini iRBCs for the in vivo trials (Ji et al., 2022a).

In vitro cytotoxicity of CIP in canine cells and hemolysis rate in canine erythrocytes

Cultures of Madin-Darby canine kidney (MDCK) cells were maintained at 37 °C under an atmosphere of 5% CO₂ and 5% O₂ and the cytotoxic effect of CIP (MedChem Express, Tokyo, Japan) was assessed using a cell viability assay by CCK-8 (Dojindo, Japan) as described previously (Li et al., 2023). The selectivity index is calculated as the ratio between the half maximal inhibitory concentration (IC₅₀) and the 50% cytotoxic concentration (CC₅₀) values. Canine erythrocytes were collected from healthy beagle dogs raised in NRCPD and stocked in Vega y Martinez (VYM) phosphate-buffered saline solution at 4 °C (Vega et al., 1985). A canine erythrocyte hemolysis assay was performed at concentrations of 0.1, 1, 5, 10, 25, 50, and 100 µM as previously described (Ariefta et al., 2022).

Evaluation of the efficacy of CIP against Babesia parasites in vitro

The efficacy of CIP against B. gibsoni and B. bovis was determined using a fluorescence assay, as previously described (Guswanto et al., 2014). The IC₅₀ values were determined from the fluorescence values and by non-linear regression analysis (curve fit) in GraphPad Prism 9 (GraphPad Software Inc., USA).

Chemotherapeutic effects of CIP against Babesia infections in vivo

CIP was evaluated on B. microti-and B. rodhaini-infected mice, as previously described (Ndayisaba et al., 2021; Tuvshintulga et al., 2022). When all mice had a 1% average parasitemia at 4 days postinfection (DPI), the drug treatments were administered, which continued for seven days. Three groups of B. microti-infected mice were administered different treatments. The CIP group (n = 6) and the ATO plus AZI group (n = 6) were orally treated with 20 mg/kg CIP and 20 mg/kg ATO plus AZI (Sigma, Tokyo, Japan), respectively. The infected mice of the vehicle group (n = 6) orally received 0.2 mL of sesame oil as the control. Eighteen mice infected with B. rodhaini were likewise placed into three groups for treatment, and they received the same care as the mice infected with B. microti. A light microscope (Nikon, Japan) and a hematology analyzer (Celltac α MEK-6450, Nihon Kohden Corporation, Tokyo, Japan) were used to observe the parasitemia and hematocrit levels every two and four days, respectively.

Selection of CIP-resistant B. gibsoni in vitro

Selections were initiated by exposing six independent flasks, each containing 10 μL B. gibsoni iRBCs mixed with 40 μL RBC into a 450 μL culture medium, which contained increasing concentrations of CIP: 5, 10, 20, 30 to 694 nM (10×IC₅₀). The medium containing CIP was replaced daily until parasites treated with 10×IC₅₀ CIP reached multiplication rates that were approximately comparable to those of the untreated controls (Hwang et al., 2010). Then, to evaluate the decreased sensitivity to CIP, IC₅₀ values of resistant strains were determined by nonlinear regression using the GraphPad Prism software.

Detection of B. gibsoni ATP4 gene mutations

The genomic DNA of the mutant parasites was extracted and sequenced (Ji et al., 2022a). The primer sets used for sequencing are listed in Table supplement 1. The single nucleotide variants were confirmed by pairwise alignment to the BgATP4^WT sequence (GenBank: JAVEPI010000002.1).

Morphological changes in CIP-treated in vitro cultured B. gibsoni

A microscopy assay was used to detect the morphological changes of wild-type B. gibsoni after exposure to 50 nM CIP for three consecutive days (Liu et al., 2021; Tayebwa et al., 2018). At 4 days post-treatment, ImageJ software was used to measure the sizes of 100 randomly selected parasites in the CIP-treated group and the control group on Giemsa-stained blood smears.

Parasites [Na⁺]_i and pH_i measurements

For both [Na⁺]_i and pH_i measurements, the wild-type and mutant parasites were initially separated from erythrocytes by treatment with saponin (0.05% [wt/vol]) for 5 min (Saliba and Kirk, 1999). The Na⁺-sensitive fluorescent dye SBFI (Thermo Fisher Scientific; product S1263) was used to quantify intracellular sodium ([Na⁺]_i). Saponin-isolated parasites (at 1×10⁸ parasites/mL) were loaded with SBFI (5 μM; in the presence of 0.02% w/v Pluronic F127) for 1 hr at room temperature (RT) (Spillman et al., 2013). Thereafter, SBFI-loaded parasites were resuspended in physiological saline (120 mM NaCl, 5 mM KCl, 25 mM HEPES, 20 mM D-glucose, and 1 mM MgCl₂ [pH 7.1]) at RT in the presence or absence of CIP. A 96-well microtiter plate was filled with around 200 µL of parasite suspension per well. The dye-loaded cells fluoresced at 515 nm after being stimulated at 340 and 380 nm. Parasites loaded with SBFI were suspended in calibration buffers containing [Na⁺] values ranging from 0 to 140 mM (pH 7.1) to establish calibration curves (Diarra et al., 2001).

The cytosolic pH of wild-type and mutant strains was measured using the pH-sensitive fluorescent dye BCECF [2’,7’-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein] (Biotium; product 51011). The BCECF was added to saponin-isolated parasites by suspension (1×10⁸ parasites/mL) and incubated for 20 minutes at 37 °C in RPMI-1640 culture medium (Gibco, USA) (Saliba and Kirk, 1999). Thereafter, the parasites loaded with dye were rinsed thrice (12,000 × g, 1 min) in RPMI-1640 culture medium, then resuspended in physiological saline (as previously mentioned) with or without various concentrations of CIP. A 96-well microtiter plate was filled with around 200 µL of parasite suspension per well. The dye-loaded cells fluoresced at 520 nm after being stimulated at 440 and 490 nm. A pH calibration was carried out for each experiment (Mohring et al., 2022).

Multiple ATP4 sequence alignment and molecular docking

B. gibsoni ATP4 (GenBank: KAK1443404.1), B. bovis ATP4 (PiroplasmaDB: BBOV_IV010020), B. microti ATP4 (GenBank: BMR1_03g01005), T. gondii ATP4 (GenBank: XP_018635122.1), and Homo sapiens ATP4 (GenBank: NM_000704.3) sequences were obtained by a homology search using P. falciparum ATP4 (GenBank: PF3D7_1211900). Sequence alignment was analyzed using MUSCLE in Jalview v2.11.3.2 software and BLAST (http://www.ncbi.nlm.nih.gov/BLAST/).

AlphaFold was used to predict the structure of BgATP4^WT (Jumper et al., 2021). Using the GROMACS 2021 Molecular Dynamics package, the energy minimization was carried out following the model’s generation (Lindahl et al., 2021). The mutations were produced by using Charmm-GUI PDB reader (Jo et al., 2014). PyMol (version 2.0 Schrödinger, LLC) was used to confirm the position of the mutation site (center: -20.367, 10.904, 9.435; size: 30×30×30) (Trott and Olson, 2010). The ligand molecules CIP was downloaded from PubChem (CID 44469321 (https://pubchem.ncbi.nlm.nih.gov/compound/ 44469321). The ligand was used to dock with Gnina (Eberhardt et al., 2021). The affinity score and binding pose were chosen only from the highest CNN (convolutional neural network) score results from docking simulations. The models were visualized using the PyMOL Molecular Graphic System and Discovery Studio.

Statistical analysis

Data analysis, namely one-way analysis of variance (ANOVA) and two-tailed unpaired t-tests, was performed using GraphPad Prism (La Jolla, CA, USA) version 9. A P value of <0.05 was considered a statistically significant result.

Competing interest

The authors have declared that no conflict of interest exists.

Acknowledgements

This work was supported by a Grant-in-Aid for Scientific Research (22H02509), the JSPS Core-to-Core program, both from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, and a grant from the Strategic International Collaborative Research Project (JPJ008837) promoted by the Ministry of Agriculture, Forestry, and Fisheries of Japan, and Natural Science Foundation of Hubei Province (2023AFC001).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Author contributions

Hang Li, Investigation, Data curation, Formal analysis, Writing – original draft, Writing – review & editing; Shengwei Ji, Methodology, Investigation, Software, Validation; Nanang R. Ariefta, Methodology, Software, Formal analysis; Eloiza May S. Galon, Visualization, Writing – review & editing; Shimaa AES El-Sayed, Methodology, Visualization; Lijun Jia, Data curation, Visualization; Yoshifumi Nishikawa, Methodology, Resources; Mingming Liu, Conceptualization, Methodology, Investigation, Resources, Data curation, Funding acquisition, Supervision, Writing – review & editing; Xuenan Xuan, Conceptualization, Methodology, Validation, Formal analysis, Supervision, Project administration, Writing – review & editing.

Ethics

All animal experiments were performed according to the Guide for the Care and Use of Laboratory Animals of Obihiro University of Agriculture and Veterinary Medicine, which was also approved by the Committee on the Ethics of Animal Experiments at the Obihiro University of Agriculture and Veterinary Medicine, Japan (permit numbers: animal experiment, 22-145 and 23-132; DNA experiment, 2207 and 2208; pathogen, 202308 and 202306).

Supplementary Material

Figure supplement 1.

Figure supplement 2.

Figure supplement 3.

Figure supplement 4.

Table supplement 1.

Table supplement 2.