Simplified Reversed Chloroquines To Overcome Malaria Resistance to Quinoline-Based Drugs

ABSTRACT Building on our earlier work of attaching a chemosensitizer (reversal agent) to a known drug pharmacophore, we have now expanded the structure-activity relationship study to include simplified versions of the chemosensitizer. The change from two aromatic rings in this head group to a single ring does not appear to detrimentally affect the antimalarial activity of the compounds. Data from in vitro heme binding and β-hematin inhibition assays suggest that the single aromatic RCQ compounds retain activities against Plasmodium falciparum similar to those of CQ, although other mechanisms of action may be relevant to their activities.

vaccines for use in the future, and the continued movement of animals to and from tsetse fly-infested areas, alternative chemotherapeutic agents are urgently required.
The activities and efficacies of diamidine (dicationic) compounds against a panel of different kinetoplastid parasites have previously been investigated (8)(9)(10), and these compounds have been demonstrated to have good efficacy against a variety of pathogens (11). Diamidines were also investigated for their in vitro activities and efficacies against T. evansi in animal models (12)(13)(14). With such promising data, it seemed appropriate to ascertain the potential activity that related diamidines (and their analogues) could exert against T. congolense and T. vivax. A selection of compounds was made on the basis of the in vitro activities of the compounds against T. bruceirelated species found previously (8,13). Hence, the aim of this study was to evaluate the in vitro, ex vivo, and in vivo (mouse) efficacies of 37 diamidine compounds against both T. congolense and T. vivax strains.

RESULTS
In total, the activities of two standard drugs and 37 diamidine compounds against a susceptible T. congolense strain (IL-3000) were investigated in vitro. The 50% inhibitory concentrations (IC 50 s; in micrograms per milliliter) of each compound were obtained for three separate assay incubation times: 40, 48, and 72 h. These IC 50 s are shown in Table  1, together with those of the two standard drugs, diminazene aceturate and isometamidium chloride. In summary, the in vitro IC 50 s of the standard drugs for IL-3000 for assay incubation times of 40, 48, and 72 h were observed to be 0.278, 0.076, and 0.066 g/ml, respectively, for diminazene and 0.0014, 0.0004, and 0.0003 g/ml, respectively, for isometamidium. The IC 50 s of the 37 diamidine compounds ranged from 0.039 to 2.721 g/ml for the 40-h assay, from 0.010 to 0.875 g/ml for the 48-h assay, and from 0.007 to 0.562 g/ml for the 72-h assay. In general, the IC 50 s consistently decreased as the incubation time increased. A similar trend was seen for the two standard drugs diminazene and isometamidium. The influence of the incubation time on the IC 50 results could clearly be seen across the 40-, 48-, and 72-h in vitro assays with T. congolense. The IC 50 s did not differ greatly between the 48-and 72-h in vitro assays with T. congolense, with 31 of the 37 compounds tested showing less than 2-fold decreases in their IC 50 s. Furthermore, the remaining 6 of the 37 compounds tested showed less than a 3-fold decrease in their IC 50 s. In contrast, the IC 50 s produced in the 40-and 72-h in vitro assays with T. congolense demonstrated a much wider range, with an up to 8-fold decrease in IC 50 s being seen between the 40-and 72-h assay durations.
In comparison, all 37 diamidine compounds and the two standard drugs were investigated ex vivo using T. congolense (STIB 736/IL-1180) and T. vivax (STIB 719/ILRAD 560) strains, neither of which is currently adapted to axenic culture conditions. The ex vivo assay was adapted from the [ 3 H]hypoxanthine incorporation assay (15) and was performed at 40 h for all compounds. The IC 50 s obtained for both strains are shown in Table 1. The ex vivo IC 50 s of the standards, diminazene and isometamidium, against both parasite strains were similar, namely, 0.095 and 0.0004 g/ml, respectively, for T. congolense and 0.076 and 0.0008 g/ml, respectively, for T. vivax. The IC 50 s of the 37 diamidine compounds tested against T. congolense ranged from 0.012 to 1.793 g/ml, whereas the IC 50 s against T. vivax ranged from 0.019 to 0.607 g/ml. The IC 50 s obtained for T. vivax were generally lower than those obtained for the T. congolense strain.
Subsequently, diminazene and isometamidium, together with 33 of the original 37 diamidine compounds, were further investigated for their in vivo efficacies against T. congolense and T. vivax in mouse models of infection. Four diamidine compounds had to be excluded from the in vivo experiments due to the discontinuation of product availability. The in vivo efficacy and the results of dose-response assays in which infected mice were intraperitoneally (i.p.) treated with the compounds on 4 consecutive days are shown in Table 2. For diminazene, 100% of T. congolense-infected mice were cured by doses of 20, 10, and 5 mg/kg of body weight given i.p. on 4 consecutive days, but only 75% (3/4) could be cured by a dose of 2.5 mg/kg. In comparison, 100% of T. vivax-infected mice could be cured only with a diminazene dose of 20 mg/kg given i.p. on 4 consecutive days, while only 2/4 mice (50%) could be cured at doses of 10, 5, and 2.5 mg/kg. The minimal curative doses of isometamidium in T. congolense-and T. vivax-infected mice were observed to be 0.03125 mg/kg and 0.0625 mg/kg, respectively, when the drug was given i.p. on 4 consecutive days.
In summary, 16 of the 33 diamidine compounds investigated were found to provide a full cure (4/4) in T. congolense-infected mice when they given at the 5-mg/kg dose i.p. on 4 consecutive days. Eleven of the 33 diamidine compounds tested were found to provide at least a 75% curative efficacy (3/4) at the 2.5-mg/kg dose, and just 3 of the 33 diamidine compounds tested were found to be at least 75% curative when they were given at the 1.25-mg/kg dose i.p. on 4 consecutive days. In comparison, 15 of the 33 diamidine compounds investigated were found to provide a full cure (4/4) in T. vivax-infected mice when they were given at the 5-mg/kg dose i.p. on 4 consecutive days. Just 7 of the 33 diamidine compounds tested were found to provide at least 75% curative efficacy (3/4) when they were given at the 2.5-mg/kg dose, and just 4 of the 33 diamidine compounds tested were found to be at least 75% curative (3/4) when they   Consequently, the two standard drugs and 15 of the 33 diamidine compounds previously tested in vivo were additionally examined for their curative efficacy when they were given to T. congolense-and T. vivax-infected mice as a single bolus treatment dose i.p. The resulting in vivo efficacy data can be viewed in Table 3. The minimal dose showing a 100% curative efficacy of diminazene against both parasites when it was given i.p. as a single bolus dose was observed to be 10 mg/kg. A single bolus dose of 5 mg/kg given i.p. was found to have insufficient efficacy (1/4 or 0/4) against both trypanosome species. For isometamidium, the minimal curative dose showing a 100% rate of cure for T. congolense-and T. vivax-infected mice was 0.25 mg/kg given i.p. A single bolus dose of 0.125 mg/kg given i.p. cured 3 out of 4 T. vivax-infected mice but only 1 out of 4 T. congolense-infected mice.
In summary, 4 of the 15 diamidine compounds evaluated were found to provide the full cure (4/4) of both T. congolense-and T. vivax-infected mice when given as a single bolus dose of 10 mg/kg i.p. Two of the 15 diamidine compounds tested were found to provide at least a 75% cure (3/4) of both T. congolense-and T. vivax-infected mice when  Diamidine Activity against African Animal Trypanosomes Antimicrobial Agents and Chemotherapy they were given as a single bolus dose of 5 mg/kg i.p. None of the diamidine compounds tested were found to provide a cure when they were given to T. congolense-infected mice as a single bolus dose of 2.5 mg/kg i.p. However, 2 of the 15 diamidine compounds were found to be 100% curative when they were given to T. vivax-infected mice as a single bolus dose of 2.5 mg/kg. The minimal curative doses providing 100% cure of T. congolense-and T. vivax-infected mice were therefore single bolus doses of 5 and 2.5 mg/kg given i.p., respectively.

DISCUSSION
The aim of this study was to determine the activities of diamidine compounds (and their analogues) against the animal-pathogenic parasites Trypanosoma congolense and T. vivax. By leveraging such chemical classes of molecules, previously found to be efficacious against a variety of similar kinetoplastid organisms, and the in-depth knowledge already gained from such investigations, the pursuit of more effective, alternative chemotherapeutic agents for the treatment of nagana can efficiently be explored. Both T. congolense and T. vivax originate from subgenera different from those of other trypanosome species, such as T. brucei brucei, T. brucei rhodesiense, and T. evansi. The current inability to continuously culture bloodstream forms of T. vivax under full axenic conditions still presents a severe hindrance to the accurate evaluation of new potential chemotherapeutic molecules with activities against these organisms. By establishing an ex vivo hypoxanthine assay for determination of the drug sensitivity of T. vivax with optimized assay duration, trypanosome concentration, and temperature parameters, the first reported set of novel compounds with activities against the bloodstream forms of T. vivax according to IC 50 s indicating susceptibility has been achieved. Work is already under way to improve this ex vivo approach by establishing a stable and reproducible alamarBlue assay for the drug sensitivity of T. vivax, which will enhance the time efficiency and cost-effectiveness of the current ex vivo hypoxanthine test for drug susceptibility.
Diamidines are known to take 48 to 72 h to fully exert their biological and chemotherapeutic potency, so the decrease in the IC 50 s for T. congolense IL-3000 obtained across the 40-, 48-, and 72-h alamarBlue assays was expected. This trend was similarly observed for the standard drugs diminazene (a diamidine) and isometamidium (an amidine). In comparison, the IC 50 s for T. congolense STIB 736/IL-1180 determined in the ex vivo assays were lower than the IC 50 s determined in the in vitro assays for. Both assays used incubation times of 40 h. The [ 3 H]hypoxanthine assay could be run only for Since the target animals for an alternative chemotherapeutic agent for the treatment of nagana are ruminants, in particular, cattle, the desired drug candidate should be able to be administered effectively via the intramuscular (i.m.) route. In mouse models of infection, i.m. administration is rather cumbersome; therefore, an i.p. route of compound administration was used. Two of the standard drugs, diminazene and isometamidium, were assessed separately in established mouse models of T. congolense and T. vivax infection to determine their effectiveness. Once a reference profile for the standard drugs was established, the diamidine molecules were comparatively assessed for their curative potential on the basis of a 4-day consecutive treatment schedule. The ideal target product profile (TPP) of a new drug for the treatment of nagana should have an optimized treatment regimen, preferably with a single application, since a 4-day treatment schedule would be impractical for rural field settings. Consequently, the top 15 most efficacious diamidines identified in the 4-day treatment schedule in the in vivo mouse models were further examined by application of only a single bolus dose.
Special attention has to be given to the problem of cross-resistance to the standard drugs diminazene aceturate (a diamidine) and isometamidium (an amidine). New diamidines have to be able to overcome this cross-resistance. This could be shown by using the knockout line T. brucei AT1 (which is missing the transporter responsible for the uptake of many diamidines), which showed a level of sensitivity to several diamidines comparable to that of a reference T. b. rhodesiense strain and a drug-sensitive T. evansi strain (13). The use of drug-resistant T. congolense and T. vivax isolates should be envisaged for any further studies with diamidine molecules.
In summary, the process described here highlights that the following compounds are potential candidates for evaluation in preclinical studies as treatments for infections caused by (i) both T. congolense and T. vivax trypanosome species (DB 75, DB 820, DB 829, DB 1406, 19 DAP 025, 28 DAP 010, and 13 SAB 089), (ii) T. congolense only (DB 867, DB 1854, DB 1870, DB 1893, 17 SAB 085 and 32 DAP 022), and (iii) T. vivax only (12 SAB 081 and 18 SAB 023). Having identified several lead diamidines in this study, the next step will be to investigate these compounds in a ruminant (e.g., goat) model of infection to assess their viability as candidates for the clinical treatment of T. congolense and T. vivax infections. Cross-resistance should also be investigated by employing drug-resistant isolates of T. congolense and T. vivax.

MATERIALS AND METHODS
Trypanosome stocks. The IL-3000 T. congolense strain was originally derived from the Trans Mara I strain, which was isolated from a bovine (within the Trans Mara region of Kenya) in 1966 (16). The IL-3000 derivative grows well as bloodstream forms in axenic culture and was thus used as the T. congolense reference strain in all in vitro drug sensitivity assays in this study. The STIB 736/IL-1180 T. congolense strain is a clone originally derived from the STIB 212 T. congolense strain, which was isolated from a lion in the Serengeti National Park of Tanzania in 1971 (17). The STIB 736/IL-1180 strain was used for all ex vivo and in vivo experiments performed with T. congolense in this study. Both T. congolense strains used in this study belong to the savannah subgenotype family. The STIB 719/ILRAD 560 T. vivax strain originated from the Y486 T. vivax strain, isolated from a naturally infected bovine in 1976 in Zaria, Nigeria (18). The Y486 T. vivax strain could be cultivated as bloodstream forms over a feeder layer (19), which is not appropriate for drug-screening purposes. Axenic cultivation is still not possible today. To our knowledge, strains derived from the T. vivax Y486 strain are the only T. vivax strains that can be successfully propagated in rodent models and are representative of West African T. vivax strains. The STIB 719/ILRAD 560 T. vivax strain was therefore used in all ex vivo and in vivo experiments carried out in this study.
Mice. Female NMRI mice weighing between 19 and 22 g were used for all in vivo experiments. Mice were specific pathogen free (SPF) and were housed in standard Macrolon type II cages at 22°C with a relative humidity of 60 to 70%. The mice received pelleted food and water ad libitum. All in vivo experiments were carried out in compliance with the regulations set out by the Swiss Federal Veterinary Office.
Diamidine test compounds. All the diamidine test compounds investigated had previously been synthesized in the laboratories of David W. Boykin (Georgia State University, Atlanta, GA, USA) and Richard R. Tidwell (University of North Carolina, Chapel Hill, NC, USA) with the aim of obtaining structural diversity, chemical stability, and a low cost of goods. For the in vivo experiments evaluating the activities of the diamidine test compounds against T. congolense and T. vivax, the diamidine test compounds were selected according to their previously demonstrated in vivo efficacies against T. brucei-related species and their absence of acute toxicity in previous experiments (8,12). All selected compounds showed greater than 75% in vivo efficacy against T. b. rhodesiense or T. evansi and no acute in vivo toxicity at cumulative doses of up to 100 mg/kg of body weight given intraperitoneally (i.p.); an exception to this was the parent compound DB 75, where acute toxicity in mice was seen at a cumulative dose of 20 mg/kg of body weight given i.p. Stock solutions and dilutions. A 10-mg/ml stock solution was prepared for each compound (dissolved in 100% dimethyl sulfoxide [DMSO]) and was stored frozen at Ϫ20°C. From these stock solutions, further stock solutions and compound dilutions were made for use in the various in vitro T. congolense cell viability assays and the T. congolense and T. vivax ex vivo incorporation assays using the appropriate culture medium as a solvent. Compound dilutions were prepared fresh on the day of the respective assays. For the in vivo mouse experiments, a 10-mg/ml stock solution was similarly prepared for each diamidine test compound, which was dissolved in sterile distilled water, containing 10% DMSO. Further dilutions depending on the dose being tested were made from these stock solutions. Stock solutions and dilutions of the standard trypanocidal drugs were prepared in sterile distilled water. All stock solutions and dilutions for the in vivo mouse experiments were made fresh on the day of administration and for each individual in vivo experiment.
In vitro antitrypanosomal assay. The IC 50 s of the test compounds for T. congolense (IL-3000) were determined using the alamarBlue assay (20), but with modified incubation times of 40, 48, and 72 h. Trypanosome densities were calculated using a cell counter and analyzer system (CASY; Schärfe System, Reutlingen, Germany), and the trypanosomes were diluted accordingly. Trypanosome seeding densities of 2 ϫ 10 5 /ml, 1 ϫ 10 5 /ml, and 1 ϫ 10 5 /ml in culture medium were used for the 40-, 48-, and 72-h alamarBlue assays, respectively. All assay plates were incubated at 34°C with 5% CO 2 for the time period being tested (24, 44, and 68 h), before the plates were removed from the incubator and 10 l of resazurin dye (12.5 mg in 100 ml phosphate-buffered saline; catalog number 33934; Aldrich/Fluka, Buchs, Switzerland) was added to each well. The plates were then further incubated for 16, 4, and 4 h respectively, under the same conditions described above. Thereafter, the assay plates were read using a fluorescence reader (SpectraMax, Gemini XS; Bucher Biotec, Basel, Switzerland) at excitation and emission wavelengths of 536 and 588 nm, respectively. The data generated were analyzed using SOFTmax Pro software (version 5.2) to determine the IC 50 s. All in vitro experiments were performed in duplicate in three independent assay runs for each compound.

Ex vivo [ 3 H]hypoxanthine incorporation assay.
The exact procedure for the ex vivo [ 3 H]hypoxanthine incorporation assay has been described previously (15) but was slightly modified for use in this study. Briefly, 50 l of culture medium containing no hypoxanthine was added to each well of a 96-well microtiter plate, except for the first two and last two wells of the last column (to which 100 l was added instead to act as a negative control) and all the wells in the first column. The drugs were applied at 75-l volumes (containing two times the highest drug concentration) into the empty wells of the first column, corresponding to the required starting concentration of each drug being tested. Thereafter, 25-l volumes were removed from the first column using a multichannel pipette and mixed with the contents in the wells in the next column. Again, 25 l was removed from the second column and placed into the next column, and the contents were mixed several times. This step was repeated until the 11th column was reached. The final 25 l from this 11th column was then discarded. This process created a 3-fold serial drug dilution across the microtiter plate.
Cardiac puncture of a highly parasitemic NMRI (female) mouse that had previously been infected with the corresponding T. congolense or T. vivax strain was performed. The blood collected was then mixed with phosphate-buffered saline with glucose (PSG; 6:4) in a 1:2 ratio, and the mixture was centrifuged for 12 min at 70 ϫ g to separate the blood cells from the trypanosomes. After centrifugation, the supernatant containing the trypanosomes was carefully transferred to a fresh tube, and the trypanosome concentration was determined using a Neubauer chamber. The trypanosome density was adjusted to provide starting concentrations of 2 ϫ 10 6 /ml and 2 ϫ 10 5 /ml for T. congolense and T. vivax, respectively. A 50-l volume of this trypanosome suspension was then added to all 96 wells, with the exception of the 4 negative-control wells in the last column. The plates were then incubated in a humidified atmosphere containing 5% CO 2 at 34°C for T. congolense or 37°C for T. vivax. After 24 h of incubation, the plates were removed, and a solution of 1 Ci of radioactive hypoxanthine in 20 l of culture medium was placed into each well. The plates were returned to the incubator for a further 16-h incubation period under the same conditions described above. After a complete incubation time of 40 h, the plates were removed and the contents of the wells were harvested on glass fiber filters using a 96-well harvester (model 1290-004 Betaplate; Berthold Technologies GmbH, Regensdorf, Switzerland). Thereafter, the radioactivity counts were measured using a liquid scintillation counter (model 1205 Betaplate; Berthold Technologies GmbH, Regensdorf, Switzerland). The data obtained were further analyzed by transferring them into a standard operating protocol template in a graphics program (Microsoft Excel) for determination of the IC 50 s. All ex vivo experiments were performed in duplicate in three independent assay runs for each compound.
In vivo mouse efficacy experiments. NMRI (female) mice were arranged into groups of four before being independently infected with either 10 5 or 10 4 parasites in 0.25 ml of PSG in a ratio of 6:4 for T. congolense (STIB 736/IL-1180) or T. vivax (STIB 719/ILRAD 560), respectively. Infection in all experiments was performed from stabilated blood, stored frozen in liquid nitrogen, using the i.p. route. For all experiments of the efficacies of the compounds against T. congolense, a parasitemia of 10 6 per ml blood was allowed to develop over 168 h (7 days), before treatment was administered i.p. on days 7 to 10 postinfection. Comparatively, for all experiments of the efficacies of the compounds against T. vivax, a parasitemia of 10 6 per ml blood was allowed to develop over 72 h (3 days), before treatment was administered i.p. on days 3 to 6 postinfection. Thereafter, the level of parasitemia in the mice was monitored using a tail blood examination technique until day 60 posttreatment. This lengthy follow-up period posttreatment was carried out to account for any possible relapses during the experiments. Parasitemia was checked twice a week for the first month and then once per week for the remaining month. Thereafter, any surviving and aparasitemic mice were considered cured. Untreated (control) mice infected with T. congolense and T. vivax survived, on average, for 11 or 6 days postinfection, respectively.