Novel Diamidines with Activity against Babesia divergens In Vitro and Babesia microti In Vivo

Abstract

Dicationic diamidines, such as diminazene and pentamidine, are well-studied chemotherapeutic agents with significant activity against parasitic diseases. The in vitro activities of novel diamidine compounds against the Babesia divergens strains 1903B and 4201 were investigated. The most potent compound, a diphenyl furan, had a 50% inhibitory concentration (IC₅₀) of 1.5 ng/ml. In a murine model, several test compounds were effective enough to cure mice infected with Babesia microti at a dose of 12.5 and/or 25 mg/kg of body weight given by the subcutaneous route for 4 days. The best antibabesial properties were exhibited by terphenyls, benzimidazoles, diphenyl furans, pentamidine, and pentamidine analogues.

INTRODUCTION

Babesiosis is a tick-borne infection caused by intraerythrocytic protozoan parasites of the genus Babesia and affects wild and domestic animals worldwide. It is a well-known disease of veterinary importance in cattle, horses, and dogs which causes considerable economic losses in livestock industry and is gaining interest as an emerging zoonosis. Bovine disease is most common in tropical and subtropical regions (Babesia bovis, Babesia bigemina), but infections are also seen in Europe (Babesia divergens, Babesia major). Clinical manifestations like fever, malaise, hemolytic anemia, and hemoglobinuria may be absent, but babesiosis can develop into a severe, rapidly fatal disease. Even though a number of effective antibabesial drugs exist, they are not readily available in all areas due to safety, residue, or marketing issues. The most widely used babesiacides are imidocarb dipropionate and diminazene aceturate; however, only imidocarb is able to consistently clear the host of parasites and has chemoprophylactic properties (29). Thus, concerns about development of resistance are growing, and the need for alternative compounds for the veterinary market is evident.

Human infections with Babesia species have been known since the late 1950s. While in Europe the causative agent in cases of babesiosis in splenectomized individuals was identified as the cattle species B. divergens, the rodent species Babesia microti was reported in most cases of human babesiosis in North America (8, 14, 28), where the disease is endemic and transmitted by the tick Ixodes damiani (also known as I. scapularis). Babesia duncani and B. divergens-like organisms later attracted attention in the west and midwest of the United States. Sporadic cases of human babesiosis have also been identified in Asia, Africa, and South America (28).

Diamidines have a long history as chemotherapeutic agents against protozoan infections, and their synthetic producibility, as well as their low molecular weight, makes them an attractive drug class. The activity is due mostly to the selective accumulation by the parasite rather than the host cell. Aromatic dicationic molecules are thought to act by binding to the minor groove of DNA at AT-rich sites, which are present in many parasitic organisms, and thus they may possibly inhibit DNA-dependent enzymes or cause direct inhibition of transcription (26). In previous studies, the significant antiparasitic activity of novel diamidines against Trypanosoma brucei and Plasmodium spp. has been shown (2, 15, 25, 26).

In the present study, we selected new diamidine compounds according to their in vitro activities against Plasmodium falciparum and demonstrated their in vitro potential against the two B. divergens strains 1903B and 4201. Subsequently, a selection of compounds was evaluated in a B. microti mouse model.

MATERIALS AND METHODS

Parasite strains and cultivation.

The two bovine B. divergens strains 1903B and 4201 were kindly provided by Laurence Malandrin of the Ecole Nationale Vétérinaire de Nantes, Nantes, France. Continuous in vitro cultures were maintained in human red blood cells (RBC) diluted to 5% hematocrit in RPMI 1640 with 25 mM HEPES and 2 mM glutamine (BioConcept, Allschwil, Switzerland) supplemented with 5 g/liter Albumax I (Gibco/BRL Life Technologies, Belgium) and 10 μg/liter gentamicin (Sigma, Steinheim, Germany). All cultures were kept in 25-ml flasks at 37°C in a 4% CO₂, 3% O₂, 93% N₂ gas mix. The medium was changed daily, and subpassages were performed every 2 to 3 days when the parasitemia reached 20%.

The strain of B. microti was kindly donated by Lise Gern (University of Neuchatel, Neuchatel, Switzerland). It was isolated from a bank vole (Myodes glareolus) in Central Switzerland and finally maintained by passages in female Swiss NMRI mice. The animals were sacrificed when the parasitemia reached 20%, and blood was collected by cardiac puncture. The course of infection was investigated using different infective doses. After intravenous inoculation with 2 × 10⁷ B. microti-infected RBCs, mice typically showed a peak parasitemia of 70 to 80% by day 7 and a subsequent decrease to low or undetectable values by day 50.

Antibabesial agents.

The diamidines were synthesized in the laboratories of David Boykin and Richard Tidwell as previously described (5, 11, 13). The syntheses of the unpublished compounds were achieved in an analogous manner. For in vitro studies, stocks of 10 mg/ml were prepared in dimethyl sulfoxide (DMSO) and subsequently diluted in RPMI cultivation medium, whereas for in vivo studies, compounds were dissolved in a 10% DMSO-water solution. Diminazene aceturate (Berenil; Sigma, Steinheim, Germany), imidocarb dipropionate (Carbesia; Schering-Plough, kindly donated by Pierre Bonnemain), and atovaquone (GSK, Muenchenbuchsee, Switzerland) served as standard drugs. Standards were prepared as described above except for imidocarb dipropionate, a sterile solution that was diluted directly in RPMI or sterile deionized water.

Cytotoxicity determination.

Cytotoxicity for L6 rat skeletal myoblasts was determined using the Alamar blue assay as described earlier (7, 21, 23).

In vitro growth-inhibitory assay.

Growth inhibition was determined by measuring the incorporation of radiolabeled [8-³H]hypoxanthine (GE Healthcare, Amersham, United Kingdom) as described before (4). Twofold serial drug dilutions were prepared in 96-well microtiter plates in order to test seven drug concentrations to determine the 50% inhibitory concentrations (IC₅₀s). Duplicate wells received 100 μl of drug dilution and 100 μl of human RBC (2% parasitemia, 2.5% hematocrit). Controls consisted of infected RBCs without drug and noninfected RBCs. Plates were incubated at 37°C in a 4% CO₂, 3% O₂, 93% N₂ atmosphere for 48 h. Then 50 μl [8-³H]hypoxanthine was added (0.5 μCi/well), and plates were incubated for another 24 h. Cells were then harvested on glass fiber filters with a cell harvester (Betaplate; Wallac PerkinElmer, Switzerland), the incorporated radioactivity was counted in a liquid scintillation counter (Betaplate; Wallac PerkinElmer, Switzerland), and IC₅₀s were calculated.

In vivo drug susceptibility test.

Female Swiss NMRI mice (18 to 20 g; RCC, Switzerland) were used for in vivo drug tests. On day 0, groups of three mice each were inoculated intravenously with 2 × 10⁷ B. microti-infected RBCs. Diamidine compounds and standard drugs were administered subcutaneously or orally (atovaquone) in intervals of 4, 24, 48, and 72 h postinfection. Doses ranged from 50 to 12.5 mg/kg of body weight in a volume of 10 ml/kg. All experiments included a group of untreated controls. Tail blood was collected at least twice a week starting on day 7. Blood smears were stained with Giemsa, and parasitemia was determined microscopically with a detection limit of 1 parasite in 10,000 erythrocytes. A compound was defined as curative if no parasites were detectable up to day 60.

RESULTS

Different chemical classes of diamidines were screened for activity against B. divergens in vitro and B. microti in vivo and compared to the standard drugs diminazene aceturate, imidocarb dipropionate, and atovaquone. The results are summarized in Tables 1 and 2.

Table 1.

In vitro antibabesial and cytotoxic activities of diamidines

^aThe IC₅₀s are mean values for at least two replicates. ND, not done.

^bNot published.

Table 2.

Activities of diamidine compounds and standard drugs against B. microti in NMRI mice and B. divergens in vitro

^aAll doses were given subcutaneously for 4 days.

^bResults are given as number of cured animals/number of infected animals.

^cThe IC₅₀s are mean values for at least two replicates.

In vitro activity against B. divergens.

The dicationic molecules demonstrated a high selectivity for B. divergens compared to their cytotoxicity for mammalian cells, which was tested using L6 rat myoblast cells in an Alamar blue assay (7, 21, 23). Of the 214 diamidine compounds, 80 showed excellent IC₅₀s below 20 ng/ml, comparable to the in vitro activities of the standard antibabesial agents, but were up to 10 times more active than those agents. The best inhibitory effects on the growth of the cultured parasites were demonstrated by the benzimidiazoles, diphenyl furans, and indoles (Table 1). It is noteworthy that almost exclusively compounds of the class of the benzimidazoles also showed high in vitro activity against both Plasmodium falciparum and Trypanosoma brucei rhodesiense (data not presented).

In vivo activity against B. microti.

NMRI mice infected with 2 × 10⁷ parasitized mouse RBCs typically developed a peak parasitemia of 70 to 80% by day 7 postinfection followed by a second smaller peak of 20 to 25% around day 15. Subsequently, the parasitemia decreased to low and undetectable values (Fig. 1). The standard drugs diminazene at 25 mg/kg and imidocarb at 12.5 mg/kg administered on four consecutive days did not cure the mice but resulted in a delayed parasitemia peak of approximately 50% around day 23 followed by a decrease to 5% or less by day 30 postinfection. Most of the compounds tested in vivo were potent enough to cause rapid suppression and initial clearance of parasites from the blood by day 10 at a dosage of 25 mg/kg subcutaneously for the first 4 days. Failure to cure resulted in recrudescence by day 30, and peak parasitemia was generally seen 5 to 10 days after the relapse.

Fig. 1.

Parasitemia in mice infected with B. microti in an untreated control group and after treatment with the standard drugs diminazene (25 mg/kg) and imidocarb (12.5 mg/kg) and with diamidine 19DAP025 (12.5 mg/kg). The standard drugs led to a delayed parasitemia of shorter duration than that for the untreated control, while the diamidine-treated mice were cured.

The best antibabesial properties were exhibited by the terphenyls, benzimidazoles, diphenyl furans, and pentamidine and its analogues as well as a diaryldiamidine (Table 2).

Three terphenyls provided cures of 2/3 (23DAP055, 19DAP085) or 3/3 (19DAP025) mice at a dosage of 25 mg/kg administered for 4 days. At a lower dose of 12.5 mg/kg, 19DAP025 still showed excellent activity and yielded a 3/3 cure.

The tested benzimidazoles were either very efficient in eliminating the parasites or could not reduce the parasite load significantly. DB818 and 6KXR030 cured 3/3 mice at a dosage of 25 mg/kg subcutaneously for 4 days. Administration of 12.5 mg/kg for 4 days resulted in a 2/3 cure for 6KXR030; however, in the case of DB818, all mice showed recrudescence around day 20 postinfection. DB942 and DB921 provided 100% suppression on day 10 but did not suppress parasitemia after day 18.

The diphenyl furan DB75, also known as furamidine, had excellent antibabesial activity when given subcutaneously. Four doses at 25 mg/kg cured all mice; at 20 mg/kg, 2/3 mice were cured, and at 12.5 mg/kg, 1/3 was cured. Even at lower doses, DB75 suppressed parasitemia completely through day 13 (6.25 mg/kg) or reduced parasitemia by >99% by day 7 (3.125 mg/kg). Mice treated with other diphenyl furans (DB530, DB555, and DB930) showed no detectable parasitemia until 10 days after the last of four injections of 25 mg/kg, but then parasites gradually reemerged.

Pentamidine is well known for its broad-spectrum antimicrobial activity, including antibabesial properties (24). In the series of pentamidine and pentamidine analogues, two compounds provided cures at 25 mg/kg: the pentamidine HCl 3SLT057 (1/3 cure) and the analogue 3KEG083 (3/3 cure).

The isoxazoles and thiophenes did not show very good potency against B. microti. Although most of them caused a decrease in parasitemias to <1% by day 7, only one compound in each series could initially clear the parasites, and all mice had relapsed by day 17.

Further chemical groups tested only in small numbers included guanidino indenes, aza-furans, and biphenyls. All three groups seemed to bear considerable antibabesial potential. When administered subcutaneously on four consecutive days at 25 mg/kg, DB905A, DB820C, and DB986 cured 3/3, 2/3, and 1/3 mice, respectively.

In comparison with the diamidine compounds, the standard drugs were not particularly effective (Fig. 1). Even at a dosage of 50 mg/kg for 4 days, diminazene treatment resulted in a pronounced relapse. Atovaquone only reduced the parasite burden and slightly delayed the peak parasitemia. Only imidocarb dipropionate provided a 2/3 cure at a dosage of 25 mg/kg for 4 days.

Correlation of in vitro activity and in vivo efficacy.

Of the 25 compounds with low IC₅₀s (<20 ng/ml) tested in vivo against B. microti, 6 caused clearance of the parasites in one or more mice. A comparable number (7/30) of compounds with IC₅₀s in the middle range (20 to 85 ng/ml) could also eliminate B. microti from the blood.

DISCUSSION

The results presented in this study demonstrate that a variety of diamidines possess in vitro and in vivo antibabesial activities equal or superior to those of the existing drugs (diminazene aceturate, imidocarb dipropionate, and atovaquone).

The in vitro data with IC₅₀s of less than 10 ng/ml reflect the great antiprotozoal potential of the dications that had already been found against other protozoan parasites. A clear correlation between in vitro activities of the diamidines against B. divergens, P. falciparum, and T. brucei rhodesiense, however, was not evident. A fourth of the tested compounds showed IC₅₀s that were very similar to those of P. falciparum, and another fourth showed IC₅₀s that were very similar to those of T. brucei rhodesiense. Nevertheless, only a very limited number demonstrated comparable activity against all three parasites (data not shown). Interestingly, most of those compounds belonged to the benzimidazoles. This result could suggest that these parasites have common transporters which recognize the amidino-benzimidazole motif.

In the B. microti mouse model, the test compounds were considerably more active than the standard drugs. Diminazene aceturate and imidocarb dipropionate are used for treatment and prophylaxis of bovine, equine, canine, and ovine babesiosis in the field, usually as an intramuscular or subcutaneous injection of 3 to 5 mg/kg. In our B. microti mouse model, diminazene could not provide a cure at a 10-fold-higher dosage. In previous studies, where different antiprotozoal drugs had been screened against B. microti in Mongolian gerbils (24) and hamsters (20), diminazene was one of the most active drugs tested. Imidocarb, though more effective than diminazene, yielded only 2/3 cures at 25 mg/kg for 4 days, while several of the diamidines provided 3/3 cures. To our knowledge, there are no similar studies published where the efficacy of imidocarb was tested in B. microti mouse models, which is probably due to the fact that imidocarb is not licensed for human use. With atovaquone we expected to see a higher efficacy, since the drug had been reported to be very effective in B. microti-infected hamsters (10, 31) as well as in Mongolian gerbils infected with B. microti (9) or B. divergens (22). It was suggested, though, that B. divergens might be more sensitive than B. microti. In comparing our results with previous ones, it has to be considered that we used a different animal species and different treatment regimens. While in our study diminazene was administered at 50 mg/kg on four consecutive days, the Mongolian gerbils were treated at 20 mg/kg over a period of 14 days to achieve a complete eradication of the infection (24). Hughes and Oz administered atovaquone at doses up to 300 mg/kg over 2 weeks to hamsters and reported survival at day 24, but parasites were not cleared at all doses. In a further study in hamsters where atovaquone was used as monotherapy, recrudescences with the emergence of resistant organisms were described. Resistance could be prevented by the addition of azithromycin (31). The combination of atovaquone and azithromycin is also used for the treatment of human babesiosis (17, 30). However, unlike the situation with P. falciparum (16), there has been no evidence of treatment failure due to resistant strains of B. microti (18) in humans. Nevertheless, for further in vivo experiments in our B. microti mouse model, combination treatments could provide valuable information.

The IC₅₀ values of the in vivo active diamidines varied from 3.5 ng/ml to >60 ng/ml. This finding suggests that a low IC₅₀ should not be the only selection criterion for in vivo testing, as pharmacological parameters are as important as the antiparasitic activity. The ability of IC₅₀s to predict in vivo activity is problematic, and thresholds should not be set too low when compounds are being selected for in vivo trials.

In conclusion, the data presented in this report indicate the potential of dicationic aromatic molecules as antibabesial agents. It is evident that further studies of structure-activity relationships, mode of action, toxicity, and in vivo efficacy of these compounds are needed before a clinical candidate can be selected.