Activities of Quinoxaline, Nitroquinoxaline, and [1,2,4]Triazolo[4,3-a]quinoxaline Analogs of MMV007204 against Schistosoma mansoni

The reliance on one drug, praziquantel, to treat the parasitic disease schistosomiasis in millions of people a year shows the need to further develop a pipeline of new drugs to treat this disease. Recently, an antimalarial quinoxaline derivative (MMV007204) from the Medicines for Malaria Venture (MMV) Malaria Box demonstrated promise against Schistosoma mansoni.

ABSTRACT

The reliance on one drug, praziquantel, to treat the parasitic disease schistosomiasis in millions of people a year shows the need to further develop a pipeline of new drugs to treat this disease. Recently, an antimalarial quinoxaline derivative (MMV007204) from the Medicines for Malaria Venture (MMV) Malaria Box demonstrated promise against Schistosoma mansoni. In this study, 47 synthesized compounds containing quinoxaline moieties were first assayed against the larval stage of this parasite, newly transformed schistosomula (NTS); of these, 16 killed over 70% NTS at 10 µM. Further testing against NTS and adult S. mansoni yielded three compounds with 50% inhibitory concentrations (IC₅₀s) of ≤0.31 µM against adult S. mansoni and selectivity indices of ≥8.9. Administration of these compounds as a single oral dose of 400 mg/kg of body weight to S. mansoni-infected mice yielded only moderate worm burden reduction (WBR) (9.3% to 46.3%). The discrepancy between these compounds’ good in vitro activities and their poor in vivo activities indicates that optimization of their pharmacokinetic properties may yield compounds with greater bioavailabilities and better antischistosomiasis activities in vivo.

INTRODUCTION

Schistosomiasis is a parasitic disease affecting over 200 million people, mostly in the developing world, caused by parasites of the Schistosoma genus, primarily Schistosoma mansoni, S. haematobium, and S. japonicum. Though this disease is responsible for a considerable health burden (1), its treatment thus far has relied on only one drug, the tetrahydroisoquinoline praziquantel. While this drug has proven safe, inexpensive at scale, and efficacious, the sheer scale of its use as an antiparasitic suggests that drug resistance may eventually become a concern (2–4).

For this reason, the development of new drugs effective against the parasite is a clear and pressing need. One of us has looked to antimalarial drugs for possible new leads (5) and screened the Medicines for Malaria Venture (MMV) Malaria Box (6) of 400 commercially available compounds for antischistosomal activity against both newly transformed schistosomula (NTS) and adult worms (7). In that work, two of the most active compounds, both in vitro and in vivo in a mouse model, were the N,N′-diarylurea 1 (MMV665852) and the 2,3-dianilinoquinoxaline 2 (MMV007204) (Fig. 1). Further exploration of the former led to the development of several analogs, including N-phenylbenzamides and N-arylphenylcarbamates, with excellent in vitro activity against S. mansoni but only moderate effect in a mouse model (8).

FIG 1.

Hit compounds against S. mansoni from the MMV Malaria Box (7).

In this work, we have developed analogs of the latter of those lead compounds, dianilinoquinoxaline 2. Quinoxaline compounds have shown promising anticancer (9–11), antiprotozoal (12), and antimycobacterial (13, 14) activities. Among this set are several 6-nitroquinoxaline analogs; nitroaromatic antiparasitic compounds have shown activity against malaria (15), giardiasis (16, 17), trypanosomiasis (18, 19), amoebiasis (20), and trichomoniasis (21). Nitroquinoxaline compounds in particular have demonstrated potent activity against Gram-positive bacteria (22).

We have also prepared and tested a small set of other compounds that include a quinoxaline moiety within a more complex polyheterocyclic system, including a series of [1,2,4]triazolo[4,3-a]quinoxalines. Similar triazolopyrazines have shown antimalarial potential in work done in the Open Source Malaria program. (https://openwetware.org/wiki/OpenSourceMalaria:Triazolopyrazine_(TP)_Series). Moreover, in previous studies, triazolopyrazines have exhibited a range of biological activities (11, 23, 24), including broad antimicrobial activity (25).

RESULTS

In vitro activity against NTS.

Forty-seven compounds (3 to 49) were synthesized and tested at 10 μM for activity against NTS. Tetracycles 35 to 37 and triazoloquinoxalines 38 to 49 showed marginal activity (<35%) at this concentration after 72 h (data in the supplemental material). Among the other quinoxaline test compounds, the best activity was found with dianilinoquinoxalines, with 16 of these showing an activity of more than 70% after 72 h (Table 1). At a lower (1 µM) concentration, 9 of those 16 compounds revealed an activity above 70% against NTS after 72 h. The most active compounds, nitroquinoxalines 29 and 30, affected NTS with an activity of over 70% at 0.1 µM, and they showed low activity (21.9% for each) even at 0.01 µM, the lowest concentration tested. For comparison, praziquantel shows a 50% inhibitory concentration (IC₅₀) of 2.2 µM against NTS (26).

TABLE 1.

In vitro activities of synthesized analogs^a

^aOnly compounds with ≥70% activity against NTS at 10 µM are shown. Numbers in parentheses are the standard deviations of the data. For full results, see the supplemental material. The dagger indicates 49.1% dead at 10 µM.

In vitro activity against adult S. mansoni.

The 16 quinoxaline compounds that showed good in vitro activity against NTS at 10 μM were also generally active against adult S. mansoni at the same concentration, with 15 of the 16 showing an activity of at least 70% against the worms after 72 h (Table 1). Of these, 8 compounds were active (>70%) at 1 µM. The three most active compounds, compounds 27, 29, and 30, showed also moderate activity against adult worms at 0.1 µM (16 to 41% after 72 h). The IC₅₀ values against adult worms for these compounds were all under 0.3 µM and were comparable to that of praziquantel (0.1 µM) (26).

Calculated physicochemical properties and solubility.

Log P and log S values were calculated for all 47 compounds in this study. In this set, the 19 compounds that showed the greatest antiparasitic activity against NTS all had calculated octanol/water partition coefficient (clog P) values over 4.98, essentially the “Lipinski limit” (27), and low calculated aqueous solubilities, ranging from 1.8 to 25.7 µM (clog S, −5.74 to −4.59) (Table 1). Although these clog P values contravene Lipinski’s “rule of five” (27), the use of those heuristics in antiparasitic drug development has been cautioned against (28).

Antischistosomal selectivity.

Nitroquinoxaline compounds 24 to 32 were tested for cytotoxicity against an L6 rat skeletal muscle cell line. Compounds that showed good activity against adult worms (>70% at 10 µM) were moderately cytotoxic to L6 cells (IC₅₀s of 1.7 to 27.5 µM) (Table 2). Among our three most active compounds, compound 29 showed the highest antischistosomal selectivity (44.6), significantly higher than those of compounds 27 and 30 (18.3 and 8.9, respectively), due largely to the compounds’ differential in cytotoxicity.

TABLE 2.

IC₅₀ and worm burden reduction values of synthesized analogs^a

Compound	IC50 (µM)		SI	WBR (%)
	Adult S. mansoni	L6 cells
27	0.31	5.7	18.3	9.3
29	0.28	12.5	44.6	46.4
30	0.19	1.7	8.9	12.5
Praziquantel	0.1b	>96b	>960b	94.1c

^aSI, selectivity index (IC_50,L6/IC_{50,S. mansoni}); WBR, worm burden reduction.

^bReference 26.

^cReference 29.

In vivo studies.

The three most active compounds progressed to in vivo studies, where they were tested in mice harboring adult S. mansoni. Compound 29 was the most active compound, with a worm burden reduction (WBR) of 46.4% (P < 0.05) at 400 mg/kg (Table 2). This is roughly half of the antiparasitic activity of praziquantel at the same dosage (94%) (29). WBR values for compounds 27 and 30 were considerably lower (9.3% and 12.5%, respectively).

DISCUSSION

The overwhelming dependence on praziquantel to treat schistosomiasis worldwide demonstrates the potential danger that praziquantel resistance poses. There is therefore a clear need for the development of antischistosomal drugs with novel pharmacophores and modes of action. The dianilinoquinoxaline compound 2 was identified in previous work as a promising lead for further antischistosomiasis drug development (8).

In this follow-up study, we have synthesized 47 analogs of this lead compound by varying both the amine/aniline substituents on the quinoxaline scaffold and the scaffold itself. Eleven of these analogs (compounds 3 to 13) were dianilinoquinoxalines like compound 2, while 10 others (compounds 14 to 23) were synthesized from 2,3-dichloroquinoxaline and nonaniline amines; the aniline-containing quinoxalines showed better antischistosomal activity than the second group did. Nitroquinoxaline analogs (compounds 24 to 32) similar to the first, more active group were also synthesized, as were a series of triazoloquinoxalines (compounds 38 to 49) and a small set of tetracyclic compounds (compounds 35 to 37) incorporating the quinoxaline moiety.

All compounds were first tested in vitro on newly transformed schistosomula (NTS). Test compounds with aniline substituents (compounds 3 to 13 and 24 to 34) generally showed strong antiparasitic activity against both NTS and adult worms at a concentration of 1 µM; the remainder of our set showed only weak activity against NTS at 10 µM. The most active were nitroquinoxalines 27, 29 and 30, which demonstrated submicromolar IC₅₀ values against adult worms, comparable to the published values for our lead compound 2 and for praziquantel itself (7). Notably, compound 30 is similar to the diarylurea lead compound 1 in that they both carry two 3,4-dichloroaniline moieties.

The addition of a nitro group to the quinoxaline scaffold increased the activity against NTS in a few cases (compound 4 versus compound 29, and compound 5 versus compound 30), but this effect was not consistent across our set of analogs. Like for other nitroaromatic antiparasitic compounds, such as nitazoxanide and metronidazole, this added activity upon nitration may be due to the targeting of parasitic redox systems (16, 30).

Unfortunately, the highest WBR achieved here, with nitroquinoxaline 29, was only half of that measured for praziquantel; two similar compounds with very good in vitro activity, compounds 27 and 30, showed very little WBR at the same concentration. The in vitro/in vivo discrepancy found in this study may be due to rapid metabolic reduction of these nitro compound to anilines (31). However, acetamidoquinoxalines 33 and 34, which would also ostensibly be metabolized to anilines within the parasite, showed markedly less activity against both NTS and adult worms than their corresponding nonacetamido analogs in our compound set (compounds 4, 5, 29, and 30). More generally, the poor in vivo activity of these compounds may simply be due to poor pharmacokinetic properties—that is, high lipophilicities (clog P > 5) and low aqueous solubilities. Further structural optimization may be able to improve the bioavailabilities of these compounds and improve their activities in vivo.

In conclusion, several analogs of the antimalarial quinoxaline MMV007204 were synthesized and shown to have high activities against NTS and adult S. mansoni worms in in vitro experiments. While the in vivo activities of these compounds proved to be moderate at best, further development of more hydrophilic derivatives may provide more active compounds.

MATERIALS AND METHODS

Synthesis.

Disubstituted quinoxaline and nitroquinoxaline compounds were synthesized from the nucleophilic aromatic substitution (S_NAr) reactions of 2,3-dichloroquinoxaline (compounds 3 to 23) and 6-nitro-2,3-dichloroquinoxaline (compounds 24 to 32), respectively (Fig. 2). While substitution reactions with aliphatic amines generally proceeded smoothly at moderate temperatures, those involving anilines required more robust heating. When 2-(dimethylamino)ethylamine was used as the amine nucleophile, the major product isolated was 1-methyl-1,2,3,4-tetrahydropyrazino[2,3-b]quinoxaline (compound 18), the unexpected result of an intramolecular S_NAr reaction by the tertiary amine followed by demethylation. Two nitroquinoxaline products, compounds 29 and 30, were subjected to nitro reduction and acetylation to give the analogous acetamides (compounds 33 and 34).

FIG 2.

Quinoxaline analogs synthesized.

Tetracyclic compounds 35 and 36 were synthesized by the condensation of o-phenylenediamine with isatin and ninhydrin, respectively (32); the latter was then subjected to palladium-catalyzed transfer hydrogenolysis (33) to yield 11H-indeno[1,2-b]quinoxaline (compound 37) (Fig. 3). [1,2,4]Triazolo[4,3-a]quinoxalines (compounds 38 to 49) were synthesized from 2-hydrazino-3-chloroquinoxaline by acid-mediated condensation with an aldehyde or orthoester, followed by an S_NAr reaction at the 4-position with a secondary amine heterocycle (24).

FIG 3.

Tetracyclic compounds 35 to 37 and triazoloquinoxalines 38 to 49.

Experimental details and ¹H nuclear magnetic resonance (NMR) spectral characterization data for all synthesized compounds can be found in the supplemental material.

Drugs and culture media.

Compounds were prepared as 10 mM stock solutions in dimethyl sulfoxide (DMSO) (Sigma-Aldrich). The culture media were prepared from medium 199 (NTS testing) or RPMI 1640 (adult testing) (Life Technologies) with l-glutamine (Sigma-Aldrich), 5% heat-inactivated fetal calf serum (FCS), and 1% penicillin-streptomycin mix, which were purchased from LubioScience.

Mice and parasites.

Animal studies were carried out following Swiss national and cantonal regulations on animal welfare at the Swiss Tropical and Public Health Institute (Basel, Switzerland [Swiss TPH]; permission no. 2070). The S. mansoni life cycle (Liberian strain) is maintained at Swiss TPH. For the in vitro and in vivo studies, female mice (NMRI strain; age, 3 weeks; weight, ca. 20 to 22 g) were purchased from Charles River, Germany. Mice were kept under environmentally controlled conditions (temperature, ∼25°C; humidity, ∼70%; 12-h light and 12-h dark cycle) with free access to water and rodent diet and were acclimatized for 1 week before infection.

NTS drug assay.

S. mansoni cercariae were gathered from infected snails and mechanically transformed to newly transformed schistosomula (NTS). A total of 30 to 40 NTS/well were incubated with 0.01 to 10 µM concentrations of the drugs for 72 h at 37°C and 5% CO₂ in a final well volume of 200 to 250 µl. Compounds were tested in triplicate, and the highest concentration of DMSO (<1%) served as a control. Evaluation was done by microscopic readout (Carl Zeiss, Germany; magnification, ×80) using a viability scale as described recently (3 = motile, no changes to morphology or transparency; 2 = reduced motility and/or some damage to tegument noted, as well as reduced transparency and granularity; 1 = severe reduction of motility and/or damage to tegument observed, with high opacity and high granularity; 0 = dead) (34).

Adult S. mansoni drug assay.

Adult schistosomes were collected by mechanical picking from the hepatic portal system and mesenteric veins 49 days after infection with 100 S. mansoni cercariae. Worms were incubated with 0.1 and 1 µM concentrations of the compounds for 72 h. Wells with 1% DMSO served as negative controls. Phenotypes were monitored under an inverted microscope, and viability scores were calculated (34). Each compound was tested twice in duplicate. To calculate IC₅₀ values, viability scores were converted into effect scores using CompuSyn2 (ComboSyn Inc., 2007).

Rat skeletal L6 myoblast cytotoxicity.

Rat skeletal L6 myoblasts were seeded in 96-well plates (2 × 10³ cells/well) using supplemented RPMI 1640 medium as described above. Following adhesion of the cells for 24 h at 37°C and 5% CO₂, the IC₅₀ of the compounds was determined using concentrations of 0.12, 0.37, 1.11, 3.33, 10, 30, and 90 µM. Podophyllotoxin served as a positive control. After 70 h postincubation, 10 µl of resazurin dye (Sigma) was added and the plates were incubated for another 2 h. Analysis was done at 72 h using a SpectraMax M2 (Molecular Devices) plate reader with an excitation wavelength of 530 nm and emission wavelength of 590 nm.

Calculation of physicochemical properties.

The in silico prediction tool ALOGPS 2.1 (http://www.vcclab.org) was used to calculate clog P and log S values for all compounds (35).

In vivo studies.

Mice were infected with 100 S. mansoni cercariae subcutaneously. Single oral doses of 400 mg/kg of the three lead compounds were administered to groups of four mice 49 days (adult infection) postinfection, respectively. A 400-mg/kg dose is often used as a starting dose in S. mansoni in vivo experiments (26), as it is the efficacious dose of praziquantel in S. mansoni-infected mice. Untreated mice (36) served as controls. Mice were euthanized using CO₂ 3 weeks posttreatment, worms were picked, sexed, and counted, and the worm burden reduction was calculated. The Kruskal-Wallis test was employed to determine statistical significance.