https://orcid.org/0000-0003-4121-143X
Abstract
Human babesiosis is a malaria-like illness caused by tick-borne intraerythrocytic Babesia parasites of the Apicomplexa phylum. Whereas several species of Babesia can cause severe disease in humans, the ability to propagate Babesia duncani both in vitro in human erythrocytes and in mice makes it a unique pathogen to study Babesia biology and pathogenesis. Here we report an optimized B. duncani in culture–in mouse (ICIM) model that combines continuous in vitro culture of the parasite with a precise model of lethal infection in mice. We demonstrate that B. duncani–infected erythrocytes as well as free merozoites can cause lethal infection in C3H/HeJ mice. Highly reproducible parasitemia and survival outcomes could be established using specific parasite loads in different mouse genetic backgrounds. Using the ICIM model, we discovered 2 new endochin-like quinolone prodrugs (ELQ-331 and ELQ-468) that alone or in combination with atovaquone are highly efficacious against B. duncani and Babesia microti.
The Apicomplexa phylum encompasses a diverse group of mostly obligate intracellular parasites that cause a number of important human and animal diseases. A subclass of these parasites includes intraerythrocytic Plasmodium and Babesia parasites, the causative agents of malaria and babesiosis, respectively [1, 2].
Babesia parasites are phylogenetically closely related to Plasmodium species [3] and cause a malaria-like illness, which in susceptible individuals can lead to hemolytic anemia, acute respiratory distress syndrome, hepatosplenomegaly, multiorgan failure, and death [4, 5]. Human babesiosis is an emerging tick-borne disease and is spreading rapidly mostly due to changes in the geographic distribution of the vector, anthropogenic factors, and global warming [4–6]. Several species of Babesia, including Babesia microti, Babesia duncani, and Babesia divergens, are known to cause infection in humans [4]. Babesia microti, the most commonly reported Babesia pathogen, can be propagated in immunocompetent and immunocompromised mice but has not yet been successfully maintained continuously in vitro in human red blood cells (hRBCs) [4]. Conversely, B. divergens can be successfully propagated in vitro in hRBCs but so far no mouse model exists for this parasite, although gerbil and rat models have been described [7, 8]. A continuous in vitro culture of the B. duncani WA-1 isolate [9] in hRBCs was recently reported [10], which made it possible to screen chemical libraries to evaluate the efficacy of new drugs, study drug–drug interactions, and probe the mode of action of active compounds [11, 12]. Interestingly, B. duncani can also infect mice and hamsters, with infection leading to rapid increase in parasitemia and severe pathology [13–15].
Here we describe an optimized B. duncani in culture–in mouse (ICIM) model that combines in vitro cell culture of the parasite in hRBCs with a precise model of lethal infection in mice. We used this model to identify 2 new endochin-like quinolones with potent antibabesial activity in vitro and strong efficacy in mice against B. duncani as well as B. microti. The ease of use and scalability of the B. duncani ICIM dual model make it an ideal system to study Babesia biology and pathogenesis and accelerate the development of innovative therapeutic strategies that could be translated to unculturable Babesia parasites for which an animal model is lacking.
METHODS
In Vitro Culture of B. duncani in hRBCs
In vitro propagation of B. duncani in hRBCs was carried out as previously reported by Abraham et al [10] and Chiu et al [11]. Parasitemia was monitored either by light microscopy examination of Giemsa-stained blood smears or by fluorescence detection of SYBR Green I [10].
Purification of B. duncani Merozoites
Babesia duncani WA-1 parasites were propagated in hRBCs until parasitemia reached 18%–20%. Culture supernatant containing free merozoites was collected by centrifugation at 355g for 5 minutes at 37°C and subsequently subjected to centrifugation at 1750g for 10 minutes at room temperature. The pellet containing free merozoites was mixed with warm Claycomb medium in a 1:5 (w/v) ratio. Purity of the merozoite suspension was assessed by light microscopy and the number of merozoites were quantified using a hemocytometer.
In Vitro Drug Efficacy and Toxicity
Antibabesial drug potency in vitro and drug–drug interactions against B. duncani and drug toxicity against human cell lines were conducted as previously described [11].
Ethics Statement
All animal experiments were approved by the Institutional Animal Care and Use Committee at Yale University (protocol number 2020-07689). Animals were acclimatized for 1 week after arrival before the start of an experiment. Animals that showed signs of distress or appeared moribund were humanly euthanized using approved protocols.
Mouse Infections
For B. duncani infections, 5- to 6-week old female and male mice (3–6 male and 3–6 female mice were used per group) of the following strains were purchased from Jackson Laboratories: C3H/HeJ, Balb/cJ, C57BL/6J, C3H/HeOuJ, C57BL/6J(Cg) Tlr4 knockout (KO), C3H/HeJ- severe combined immunodeficient (SCID), Balb/cJ-SCID, and C57BL/6J-SCID. Mice were inoculated by the intravenous (IV) or intraperitoneal (IP) route with either blood from an infected stock mouse at indicated doses, in vitro–cultured B. duncani–infected hRBCs (8.5 × 105) or purified B. duncani merozoites (2 × 107) and monitored over time. For B. microti infections, 5- to 6-week-old female C.B-17.SCID (C.B-17/IcrHsd-Prkdcscid) mice were obtained from Envigo. Babesia microti (strain LabS1)–infected mouse red blood cells (mRBCs) were thawed from previously cryopreserved stocks of infected mouse blood and subsequently used to inoculate mice using IV or IP routes as indicated. Animals were bled at specified time points either by retro-orbital or tail vein bleeding, and parasitemia was determined by light microscopy examination of Giemsa-stained thin blood smears.
In Vivo Drug Efficacy in B. duncani– and B. microti– Infected Mice
For drug efficacy following B. duncani infection, C3H/HeJ female mice (n = 5–10 mice/group) were infected IV with 103 B. duncani–infected mRBCs. For drug efficacy following B. microti infection, female C.B-17.SCID mice (n = 5 mice/group) were infected using 104 (IV) or 106 (IP) B. microti–infected mRBCs. All mice were treated by oral gavage (100 μL) for 10 days (days postinfection [DPI] 1–10) with the vehicle (PEG-400), ELQ-331, ELQ-468, or atovaquone alone at 10 mg/kg, or with the following combinations: ELQ-331 + atovaquone or ELQ-468 + atovaquone at 10 + 10 mg/kg. Parasitemia was determined by light microscopy examination of Giemsa-stained thin blood smears.
RESULTS
Establishment of an ICIM Model of B. duncani Infection
Continuous in vitro culture of B. duncani WA-1 isolate in hRBCs was previously reported using commercially available HL-1 and Claycomb media [10] (Supplementary Figure S1). To determine whether in vitro–cultured parasites can initiate infection in mice, female C3H/HeJ mice were injected IV with either 8.5 × 105 B. duncani–infected hRBCs or 2 × 107 cell-free merozoites purified from in vitro B. duncani culture (Figure 1A). As shown in Figure 1B, parasitemia was established in both groups of mice by 8 and 12 DPI following the administration of infected hRBCs and cell-free merozoites, respectively. In both groups, peak parasitemia reached 4%–7% and animals were euthanized at endpoints of DPI 11 and DPI 15, following inoculation with infected hRBCs or cell-free merozoites, respectively (Figure 1C).
Transmission of Babesia duncani (Bd)–infected human red blood cells (hRBCs) from continuous in vitro culture and B. duncani free merozoites into mice. A, Schematic representation of transmission of in vitro–cultured B. duncani parasites in hRBCs or purified merozoites isolated from B. duncani in vitro culture, into C3H/HeJ mice. B, Parasitemia profile over time in female C3H/HeJ mice (n = 3/group) following infection with in vitro–cultured B. duncani parasites (blue), or purified B. duncani merozoites (green). Uninfected mice profile depicted in red. E: euthanized. C, Kaplan–Meier plot of percentage survival of uninfected mice (red), mice infected with in vitro–cultured B. duncani parasites (blue), or mice infected with purified merozoites (green).
To optimize the mouse model of B. duncani infection, C3H/HeJ (n = 6 mice/dose) were inoculated IV with parasite loads ranging from 107 to 102 B. duncani–infected red blood cells (iRBCs) collected from a previously infected mouse. As shown in Figure 2A, consistent infection in both male and female mice was established in a dose-dependent manner, as represented by the parasitemia onset. Interestingly, all infection doses resulted in lethal outcome. Mice inoculated with 107 and 106 B. duncani–iRBCs showed an acute increase in parasitemia (up to 35%) within a short span of time (DPI 3–5), became moribund, and were euthanized by DPI 5–7 (Figure 2B). Animals infected with doses between 105 and 102 iRBCs showed a delayed onset of infection reaching successively lower peak parasitemia compared to those administered with 107/106 iRBCs (Figure 2A). The parasite burden was also a function of the infection dose, with animals infected with 105 iRBCs developing parasitemia levels up to 9%, whereas mice infected with 102 iRBCs reached no more than 2.5% parasitemia (Figure 2A). Nonetheless, all mice in each infection group became moribund and were euthanized (Figure 2B). Interestingly, parasitemia levels appeared consistently higher in female mice than in male mice (Figure 2A).
Lethal Babesia duncani infection model in immunocompetent mice. A, Parasitemia profile of B. duncani infection in female (solid symbols) and male (open symbols) C3H/HeJ mice (n = 6/group; 3 females + 3 males) were infected intravenously using B. duncani–infected mouse blood with doses of 107 (green), 106 (blue), 105 (orange), 104 (purple), 103 (red), and 102 (brown) infected red blood cells. E: euthanized. Uninfected mice profile shown in red. B, Kaplan–Meier plot of percentage survival of female (solid symbols) and male (open symbols) C3H/HeJ mice infected with different doses of B. duncani, as described above. Uninfected mice (red) were used as control.
Influence of Mouse Genetic Background, Route of Administration, and Host Immunity on the Response to B. duncani Infection
To examine the importance of the host genetic background in susceptibility to B. duncani infection, C3H/HeJ, Balb/cJ, and C57BL/6J mice (n = 6 mice/group; 3 females and 3 males) were infected by IV (Figure 3A and 3B) or IP (Supplementary Figure S2) routes with 104 B. duncani–iRBCs. This dose was chosen as it allows sufficient time to evaluate the progression of parasitemia and to collect samples such as sera and plasma over time for subsequent analyses. As shown in Figure 3A and Supplementary Figure S2A, all C3H/HeJ mice, inoculated by either IV or IP route, showed detectable parasitemia by DPI 6 or 7, with a peak parasitemia of 6% for males and 10.7% for females, respectively. Infected mice became moribund by DPI 10 and were euthanized (Figure 3B and Supplementary Figure S2D). In contrast, C57BL/6J mice infected either IV or IP with 104 iRBCs were resistant to B. duncani infection and showed no detectable parasitemia throughout the 28-day monitoring period (Figure 3A and 3B and Supplementary Figure S2C and S2D). Interestingly, Balb/cJ mice showed intermediate range of susceptibility with 40%–60% survival when infected IV or IP. Some of the IV-infected Balb/cJ mice developed parasitemia between 3% and 9% by DPI 9, became moribund, and were euthanized; the rest of the cohort cleared the infection (Figure 3A). The IP-infected Balb/cJ mice showed a similar profile but reached lower levels of peak parasitemia (1%–2%) (Supplementary Figure S2B).
Role of genetic background and immunity status of mice on Babesia duncani infection outcome. Immunocompetent mice (n = 6/group; 3 females + 3 males) from 3 genetic backgrounds were infected with 104 B. duncani–infected red blood cells (iRBCs) by intravenous (IV) route and parasitemia was monitored over time. A, Parasitemia profile over time following IV administration of B. duncani–iRBCs in C3H/HeJ (green), Balb/cJ (blue), and C57BL/6J (orange) mice. Female and male profiles are shown in solid and open symbols, respectively. E: euthanized; WT, wild type. B, Kaplan–Meier plot showing percentage survival of C3H/HeJ, Balb/cJ, and C57BL/6J mice post–IV infection with 104 B. duncani–iRBCs. C, Parasitemia profile in severe combined immunodeficient (SCID) mice from 3 genetic backgrounds (n = 6/group; 3 females + 3 males). C3H/HeJ-SCID (green), Balb/cJ-SCID (blue), C57BL/6J-SCID (orange) mice were infected IV with 104 B. duncani–iRBCs. Female and male mice profiles are shown in solid and open symbols, respectively. D, Kaplan–Meier plot showing percentage survival of the above-mentioned groups of mice post–IV infection with 104 B. duncani–iRBCs.
To assess whether the resistance of C57BL/6J mice was the result of rapid clearance of the parasites by the immune system or inability of the parasite to infect RBCs from this genetic background at the selected 104 B. duncani–iRBC dose, we compared the parasitemia profile between immunocompetent and immunocompromised mice following infection with B. duncani. SCID mice from C3H/HeJ, Balb/cJ, and C57BL/6J backgrounds (C3H/HeJ-SCID, Balb/cJ-SCID, and C57BL/6J-SCID) (n = 6 mice/group; 3 females and 3 males) were infected IV with 104 B. duncani–iRBCs and their parasite loads were monitored over time. The C3H/HeJ-SCID mice reached a higher parasitemia (up to 60% at DPI 13) than their wild-type (WT) counterpart (peak at 10% at DPI 7) (Figure 3C and 3A), became moribund, and were euthanized by DPI 13 (Figure 3D). The difference in peak parasitemia between immunocompetent and SCID mice was also observed in the case of Balb/cJ mice (8% vs 60% for Balb/cJ and Balb/cJ-SCID, respectively) (Figure 3C and 3A), and all but 1 Balb/cJ-SCID mouse had to be euthanized by DPI 17 (Figure 3D). Interestingly, unlike C57BL/6J immunocompetent mice (Figure 3A), all C57BL/6J-SCID mice developed high parasitemia (up to 40%) and reached the endpoint by DPI 22 (Figure 3C and 3D).
Susceptibility of immunocompromised C57BL/6J-SCID mice to B. duncani infection at the 104 B. duncani–iRBC dose suggests that immunocompetent C57BL/6J mice are able to clear infection at this dose. Hence, we evaluated the effect of a high-dose infection in C57BL/6J mice. Ten C57BL/6J females were infected IV with 107 B. duncani–iRBCs, and parasitemia and survival profile were monitored. All 10 mice developed parasitemia with peak ranging from 1.3% to 7.2% between DPI 6 and 7 (Supplementary Figure S3A), with all mice succumbing to infection between DPI 7 and DPI 13 (Supplementary Figure S3C). In contrast, a similar high-dose infection (107 B. duncani–iRBCs) in Balb/cJ or C3H/HeJ mice resulted in an acute increase in parasitemia with peak parasitemia reaching up to 30%–35% within DPI 7 (Supplementary Figure S3B and Figure 2A), at which point animals became moribund and were euthanized.
Tlr4 Does Not Play a Role in Mouse Susceptibility to B. duncani Infection
Since the highly susceptible immunocompetent C3H/HeJ mouse carries a missense mutation in the 3rd exon of the Tlr4 gene [16], we investigated whether Tlr4 plays a role in the susceptibility to B. duncani infection. We infected C3H/HeOuJ mice (that possess a WT Tlr4) and C57BL/6J(Cg) Tlr4 KO mice (a Tlr4 KO strain on C57BL/6 background) with 104 B. duncani–iRBCs and monitored parasitemia and mouse survival postinfection. We found that the C3H/HeOuJ mice (n = 3) developed similar levels of parasitemia and were equally susceptible to B. duncani infection as their C3H/HeJ counterparts (Figure 4A and 4C). Similarly, C57BL/6J(Cg) Tlr4 KO mice (n = 3) were found equally resistant to the infection as the C57BL/6J mice (Figure 4B and 4C), indicating that Tlr4 does not play a role in host susceptibility to B. duncani infection.
Role of Tlr4 gene in susceptibility to Babesia duncani infection in vivo. Female C3H/HeOuJ, C3H/HeJ, C57BL/6J(Cg) Tlr4 knockout (KO), or C57BL/6J mice (n = 3/group) were infected intravenously with 104 B. duncani–infected red blood cells (iRBCs). Parasitemia profile in C3H/HeOuJ (red) and C3H/HeJ (green) (A) and C57BL/6J(Cg) Tlr4 KO (orange) and C57BL/6J mice (cyan) (B). E: euthanized. C, Kaplan–Meier plot showing percentage survival of the above-mentioned groups of mice post–intravenous infection with 104 B. duncani–iRBCs.
ELQ Compounds Demonstrate Potent In Vitro Activity and In Vivo Efficacy Against B. duncani
Recent structure-activity relationship studies in B. duncani and B. microti identified the endochin-like quinolone prodrug ELQ-502 as a potent antiparasitic drug [11]. In addition to ELQ-502, 2 additional prodrugs, ELQ-331 and ELQ-468, were identified as potent inhibitors of B. duncani growth in vitro with half maximal inhibitory concentration (IC50) values of 15 nM and 141 nM, respectively [11]. Here we used the ICIM model to evaluate the potency of the active drugs of ELQ-300 and ELQ-446 (ELQ-331 and ELQ-468 prodrugs, respectively) in vitro, the possible synergy of the activity of the prodrugs with atovaquone, and their in vivo efficacy in mice. ELQ-300 and ELQ-446 showed strong potency against B. duncani with IC50 values of 36 nM and 34 nM, respectively (Figure 5B). Furthermore, in vitro combination studies of ELQ-331 and ELQ-468 with atovaquone showed synergistic interactions with mean fractional inhibitory concentrations (ΣFIC50) of 0.6 and 0.8 for ELQ-331 + atovaquone and ELQ-468 + atovaquone, respectively (Supplementary Figure S4A and S4B and Table 1). Toxicity studies using 4 human cell lines (HeLa, HepG2, HEK293, and hTERT) showed no activity at concentrations to 10 µM (Table 2). The favorable therapeutic indices of ELQ-331 and ELQ-468 led us to evaluate their efficacy in C3H/HeJ mice alone or in combination with atovaquone. Whereas vehicle-treated B. duncani–infected mice succumbed to infection by DPI 11 (Figure 5C), animals treated with ELQ-331 and ELQ-468, either as monotherapies or in combination with atovaquone, remained mostly clear of parasites throughout the study (Supplementary Figure S4C) and survived (Figure 5C). Only a few sporadic events of recrudescence were noted (Supplementary Figure S4C). Whereas mice treated with ELQ-331 alone showed no parasitemia during the 45-day monitoring period (Supplementary Figure S4C), mice treated with atovaquone or ELQ-468 monotherapies were only partially protected; 2 of 10 mice and 2 of 5 mice in the respective treatment groups developed parasitemia starting around DPI 20 (Supplementary Figure S4C) and were euthanized by DPI 26 (Figure 5C). In the case of the combination therapies, 2 of 10 mice in the ELQ-331 + atovaquone treatment group developed parasitemia from DPI 24 (Supplementary Figure S4D) and were euthanized by DPI 26 (Figure 5C), and 1 of 10 mice in the ELQ-468 + atovaquone–treated group showed parasitemia from DPI 23 (Supplementary Figure S4D) and were euthanized by DPI 24 (Figure 5C).
![In vitro and in vivo evaluation of ELQs against Babesia duncani and Babesia microti infections. A, Chemical structures of atovaquone, ELQ-331 and its active drug ELQ-300, and ELQ-468 and its active drug ELQ-446. The prodrugs are cleaved by host esterase to release the active compound. B, Dose-response curves of ELQ-300 (blue), ELQ-331 (brown), ELQ-446 (orange), and ELQ-468 (black) with corresponding half maximal inhibitory concentration (IC50) values indicated. Data presented as mean ± standard deviation of 2 independent experiments performed in biological triplicates. C, Kaplan–Meier plot showing percentage survival of 104 B. duncani–infected red blood cell (iRBC) C3H/HeJ mice (n = 5–10/group) following treatment with monotherapies of ELQ331 (blue closed square), ELQ-468 (orange closed triangle), or atovaquone (green closed inverted triangle) at 10 mg/kg or combination therapies with ELQ-331 + atovaquone (blue open square) or ELQ-468 + atovaquone (orange triangle) at 10 + 10 mg/kg. Vehicle (PEG-400)–treated mice (black) were used as controls. Treatment was administered daily for 10 days (days postinfection [DPI] 1–10) by oral gavage. D, Parasitemia profiles of female C.B-17.SCID mice (n = 5/group) infected with 104 (intravenous) or 106 (intraperitoneal) Babesia microti–iRBCs and treated with monotherapies of ELQ-331 (orange open triangle), ELQ-468 (blue inverted open triangle), or atovaquone (green square) at 10 mg/kg or combination therapies of ELQ-331 + atovaquone (red open hexagon) or ELQ-468 + atovaquone (olive open square) at 10 + 10 mg/kg. Control animals (black) received the vehicle alone (PEG-400). Treatment was administered daily for 10 days (DPI 1–10) by oral gavage.](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/jid/226/7/10.1093_infdis_jiac181/4/m_jiac181f5.jpeg?Expires=1749687591&Signature=lDrzHKCBwwLKL4DT7H~e6GwvgoZskXsx2GLHH1Yfe3NmCKFwTITXjZ4I6t0809khxCXRB23Xk4u0VM6VZqaZF~LjBBeWWVqAIBU7OSxIJO1wiRmeiUAbLQreYo0hZegLJgg3JwBdOKsRxIy5ZEvJUvKEY0UdD3pf5IoaIFClZRhxEQXlWkLVz4vzbPW34uq~pPzc6~W1CBdbAwk3dcncFKJ0JgDBtCJZtVnpIhPqUtll9vBOER6TZw1GBwWFhY2YN2AsFdiM1QP96gQ1EL7SXA-olL3Ox4L4E3hBYX3r0JOgbNuwg4wbnDYMhH--7JUMYBgO4yuEMSgd-g4sflGt6A__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
In vitro and in vivo evaluation of ELQs against Babesia duncani and Babesia microti infections. A, Chemical structures of atovaquone, ELQ-331 and its active drug ELQ-300, and ELQ-468 and its active drug ELQ-446. The prodrugs are cleaved by host esterase to release the active compound. B, Dose-response curves of ELQ-300 (blue), ELQ-331 (brown), ELQ-446 (orange), and ELQ-468 (black) with corresponding half maximal inhibitory concentration (IC50) values indicated. Data presented as mean ± standard deviation of 2 independent experiments performed in biological triplicates. C, Kaplan–Meier plot showing percentage survival of 104 B. duncani–infected red blood cell (iRBC) C3H/HeJ mice (n = 5–10/group) following treatment with monotherapies of ELQ331 (blue closed square), ELQ-468 (orange closed triangle), or atovaquone (green closed inverted triangle) at 10 mg/kg or combination therapies with ELQ-331 + atovaquone (blue open square) or ELQ-468 + atovaquone (orange triangle) at 10 + 10 mg/kg. Vehicle (PEG-400)–treated mice (black) were used as controls. Treatment was administered daily for 10 days (days postinfection [DPI] 1–10) by oral gavage. D, Parasitemia profiles of female C.B-17.SCID mice (n = 5/group) infected with 104 (intravenous) or 106 (intraperitoneal) Babesia microti–iRBCs and treated with monotherapies of ELQ-331 (orange open triangle), ELQ-468 (blue inverted open triangle), or atovaquone (green square) at 10 mg/kg or combination therapies of ELQ-331 + atovaquone (red open hexagon) or ELQ-468 + atovaquone (olive open square) at 10 + 10 mg/kg. Control animals (black) received the vehicle alone (PEG-400). Treatment was administered daily for 10 days (DPI 1–10) by oral gavage.
Evidence for Synergistic Activity of ELQ-331 or ELQ-468 Plus Atovaquone Against Babesia duncani In Vitro
| Drug Combination | ΣFIC50 | Type of Interaction |
|---|---|---|
| ELQ-331 + atovaquone | 0.6 | Synergistic |
| ELQ-468 + atovaquone | 0.8 | Synergistic |
| Drug Combination | ΣFIC50 | Type of Interaction |
|---|---|---|
| ELQ-331 + atovaquone | 0.6 | Synergistic |
| ELQ-468 + atovaquone | 0.8 | Synergistic |
Abbreviation: ∑FIC50, mean fractional inhibitory concentration.
Evidence for Synergistic Activity of ELQ-331 or ELQ-468 Plus Atovaquone Against Babesia duncani In Vitro
| Drug Combination | ΣFIC50 | Type of Interaction |
|---|---|---|
| ELQ-331 + atovaquone | 0.6 | Synergistic |
| ELQ-468 + atovaquone | 0.8 | Synergistic |
| Drug Combination | ΣFIC50 | Type of Interaction |
|---|---|---|
| ELQ-331 + atovaquone | 0.6 | Synergistic |
| ELQ-468 + atovaquone | 0.8 | Synergistic |
Abbreviation: ∑FIC50, mean fractional inhibitory concentration.
Activity of ELQ-331 and ELQ-468 in Babesia duncani–Infected Human Erythrocytes, HeLa, HepG2, HEK293, and hTERT Cells
| IC50 | ||||||
|---|---|---|---|---|---|---|
| Prodrug | Babesia duncani | HeLa | HepG2 | HEK293 | hTERT | Therapeutic Index |
| ELQ-331 | 141 ± 22 nM | >10 µM | >10 µM | >10 µM | >10 µM | >71 |
| ELQ-468 | 15 ± 1 nM | >10 µM | >10 µM | >10 µM | >10 µM | > 667 |
| IC50 | ||||||
|---|---|---|---|---|---|---|
| Prodrug | Babesia duncani | HeLa | HepG2 | HEK293 | hTERT | Therapeutic Index |
| ELQ-331 | 141 ± 22 nM | >10 µM | >10 µM | >10 µM | >10 µM | >71 |
| ELQ-468 | 15 ± 1 nM | >10 µM | >10 µM | >10 µM | >10 µM | > 667 |
Abbreviation: IC50, half maximal inhibitory concentration.
Activity of ELQ-331 and ELQ-468 in Babesia duncani–Infected Human Erythrocytes, HeLa, HepG2, HEK293, and hTERT Cells
| IC50 | ||||||
|---|---|---|---|---|---|---|
| Prodrug | Babesia duncani | HeLa | HepG2 | HEK293 | hTERT | Therapeutic Index |
| ELQ-331 | 141 ± 22 nM | >10 µM | >10 µM | >10 µM | >10 µM | >71 |
| ELQ-468 | 15 ± 1 nM | >10 µM | >10 µM | >10 µM | >10 µM | > 667 |
| IC50 | ||||||
|---|---|---|---|---|---|---|
| Prodrug | Babesia duncani | HeLa | HepG2 | HEK293 | hTERT | Therapeutic Index |
| ELQ-331 | 141 ± 22 nM | >10 µM | >10 µM | >10 µM | >10 µM | >71 |
| ELQ-468 | 15 ± 1 nM | >10 µM | >10 µM | >10 µM | >10 µM | > 667 |
Abbreviation: IC50, half maximal inhibitory concentration.
In Vivo Efficacy of ELQ-331 and ELQ-468 in B. microti–Infected Mice
Based on the studies from the B. duncani ICIM model, we evaluated the efficacy of the combinations ELQ-331 + atovaquone and ELQ-468 + atovaquone in B. microti–infected mice. As shown in (Figure 5D), whereas vehicle-administered mice developed high levels of parasitemia by DPI 20, followed by a plateau at 60%–80% parasitemia until the end of the study (DPI 45), mice that received monotherapies of ELQ-331 or ELQ-468 remained free of parasites till DPI 45 (end of study). Groups receiving drug combinations of ELQ-331 + atovaquone and ELQ-468 + atovaquone also showed no signs of infection during the entire duration of the study. Mice treated with atovaquone alone showed recrudescence by DPI 24 (Figure 5D).
DISCUSSION
In this study, we describe the ICIM model of B. duncani infection that allows analysis of the intraerythrocytic development of the parasite in hRBCs in vitro and in a mouse model of lethal infection. We show that C3H/HeJ mice inoculated intravenously with either infected hRBCs or cell-free merozoites isolated from an in vitro culture of B. duncani develop parasitemia over time and ultimately succumb to infection. The onset of infection in C3H/HeJ mice following inoculation with B. duncani–infected hRBCs was significantly delayed compared to that with infected mRBCs. This difference can be attributed to the robust depletion of hRBCs by murine phagocytic cells [17]. Despite the higher infection dose of in vitro–cultured parasites needed to establish infection in mice, the availability of this model makes it possible for the first time to examine the importance of specific genes in Babesia growth and virulence in vivo using transgenic parasites generated in vitro.
Our studies showed that the onset of parasitemia in C3H/HeJ mice was rapid following inoculation with 107 and 106 B. duncani–infected mRBCs with parasitemia levels reaching as high as 35% by DPI 5. This was followed by a rapid development of pathology requiring animal euthanasia by DPI 7. Inoculation of C3H/HeJ mice with doses between 102 and 105 resulted in a slow onset of infection with parasitemia levels not exceeding 9% (105). However, even with low parasite burdens, all animals succumbed to infection. Consistent with previous observations [18], we also found that male mice develop lower parasitemia than female mice, despite the same fatal endpoint.
Unlike C3H/HeJ mice, which are highly susceptible to B. duncani infection (100% mortality rate), C57BL/6J mice inoculated with 104 B. duncani–infected mRBCs do not succumb to infection (100% survival), nor do they develop parasitemia. Interestingly, at this inoculation dose, Balb/cJ mice showed an intermediate profile with 40%–60% survival rates. However, unlike C57BL/6J, all mice develop parasitemia over time, but survival was dependent on the ability of the individual mouse to clear the infection. Interestingly, independent of the genetic background, all mice carrying the scid mutation became infected following inoculation with B. duncani and showed significantly higher parasitemia levels than their immunocompetent counterparts (66% vs 10% for C3H/HeJ, 60% vs 8% for Balb/cJ, and 40% vs 0% for C57BL/6J) and were euthanized between DPI 11–22. Taken together, these studies indicate that susceptibility to B. duncani infection is strongly linked to the host immune response. Our studies also showed that following a high-dose IV inoculation (107 iRBCs) of C57BL/6J with B. duncani–infected RBCs, infected mice develop parasitemia and succumb to infection. This is in contrast to a previous study, where B. duncani infection of C57BL/6 mice using 108 iRBCs by IP route resulted in 95% survival [19]. One possible explanation for this difference is that, unlike IV inoculation, the IP injection of B. duncani results in a lower parasite load in blood, leading to a more successful control of the infection.
Our studies also showed that Toll-like receptor 4 (TLR4) deficiency does not play a role in mouse susceptibility to B. duncani infection as C3H/HeOuJ (WT Tlr4) were equally susceptible to infection as C3H/HeJ, which are deficient for TLR4, and C57BL/6J(Cg) Tlr4 KO mice remained clear of parasitemia throughout the duration of the study. These results are consistent with previous studies that showed C3H/HeN mice (TLR4 WT) to be susceptible to B. duncani infection [14]. Similar to infection in C3H/HeN, our studies showed that in susceptible mice, B. duncani infection causes spleen enlargement and fluid accumulation in their lungs.
In a recent study, we identified ELQ-502 as a safe and effective therapy against Babesia infection in mice [11]. To expand the panel of late leads for preclinical evaluation, 2 additional prodrugs, ELQ-331 and ELQ-468, were also identified as potent antibabesial drugs, with IC50 values of 141 nM and 15 nM, respectively [11]. In vitro drug efficacy studies using ELQ-300 and ELQ-446, the active compounds of ELQ-331 and ELQ-468, respectively, showed similar high potency against B. duncani with IC50 values of 36 nM and 34 nM, respectively. Consistent with previous reports investigating other ELQs [11, 20], drug–drug interaction studies showed that ELQ-331 and ELQ-468 are synergistic with atovaquone against B. duncani in vitro. In vivo evaluation of these derivatives further revealed the high potency of ELQ-331 and ELQ-468 against B. duncani as well as B. microti infections, both as monotherapies and in combination with atovaquone. Altogether, our study identified 2 new ELQ derivatives as promising candidates for the treatment of human babesiosis, adding these late leads to the pipeline of antibabesial therapeutic drugs.
The continuous in vitro culture system of B. duncani in human erythrocytes offers a suitable platform to conduct a variety of assays to (1) identify novel and potent antiparasitic drugs (high-throughput screening of chemical libraries, drug efficacy studies, drug–drug interaction studies); (2) understand the mode of action and mechanism of resistance of antiparasitic drugs (selection of drug-resistant parasites, propagation of recrudescent parasites from in vivo experiments); and (3) unravel the metabolic requirements of parasites for their survival within host RBCs. Similarly, the in vivo model of B. duncani infection in mice allows the evaluation of drug efficacy and relapse, the study of virulence and pathology, the investigation of tissue distribution and persistence/dormancy mechanisms, the understanding of host–pathogen interactions, and the identification of promising vaccine candidates as well as the evaluation of their protective properties. Finally, the use of B. duncani as a model organism could help advance our understanding of the biology of other Babesia species, as B. duncani could be used as a surrogate system to study cellular and metabolic machineries that are conserved between different Babesia species, to express antigens from different parasites to evaluate their immunogenicity and vaccine potential, and to unravel the mode of action of antiparasitic drugs with a broad scope of activity against different parasites. While the B. duncani ICIM model may provide useful insights into the biology and pathogenesis of Babesia species, differences between species exist and the study of unique aspects of the biology of other Babesia parasites may require future efforts to develop suitable in vitro or in vivo models for each of these species.
Supplementary Data
Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.
Notes
Acknowledgments. The authors thank Heather Wallace from the Yale Animal Resources Center for performing oral gavage in in vivo studies. The authors also thank Dr Peter Krause for constructive criticism of the manuscript.
Financial support. This work was supported by the National Institutes of Health (NIH) (grant number R01AI123321 to C. B. M., J. S. D., and M. K. R.). Research in the Ben Mamoun Laboratory is also supported by NIH grants AI138139, AI152220, and AI136118; the Steven and Alexandra Cohen Foundation (Lyme 62 2020 to C. B. M.); and the Global Lyme Alliance. M. K. R. is also supported by the NIH (grant numbers R01AI100569 and R01AI141412); the United States Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development Program (i01 BX003312); and the United States Department of Defense Peer Reviewed Medical Research Program (log number PR181134). M. K. R. is a recipient of a Veterans Affairs Research Career Scientist Award (14S-RCS001). J. S. D. is a recipient of a Veterans Affairs Merit Review Award (BX004522, US Department of Veterans Affairs Biomedical Laboratory Research and Development).
References
Author notes
Potential conflicts of interest. All authors: No reported conflicts of interest.
All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
