Mitochondria as a Potential Target for the Development of Prophylactic and Therapeutic Drugs against Schistosoma mansoni Infection

ABSTRACT

The emergence of parasites resistant to praziquantel, the only therapeutic agent, and its ineffectiveness as a prophylactic agent (inactive against the migratory/juvenile Schistosoma mansoni), make the development of new antischistosomal drugs urgent. The parasite’s mitochondrion is an attractive target for drug development, because this organelle is essential for survival throughout the parasite’s life cycle. We investigated the effects of 116 compounds against Schistosoma mansoni cercaria motility that have been reported to affect mitochondrion-related processes in other organisms. Next, eight compounds plus two controls (mefloquine and praziquantel) were selected and assayed against the motility of schistosomula (in vitro) and adults (ex vivo). Prophylactic and therapeutic assays were performed using infected mouse models. Inhibition of oxygen consumption rate (OCR) was assayed using Seahorse XFe24 analyzer. All selected compounds showed excellent prophylactic activity, reducing the worm burden in the lungs to less than 15% of that obtained in the vehicle control. Notably, ascofuranone showed the highest activity, with a 98% reduction of the worm burden, suggesting the potential for the development of ascofuranone as a prophylactic agent. The worm burden of infected mice with S. mansoni at the adult stage was reduced by more than 50% in mice treated with mefloquine, nitazoxanide, amiodarone, ascofuranone, pyrvinium pamoate, or plumbagin. Moreover, adult mitochondrial OCR was severely inhibited by ascofuranone, atovaquone, and nitazoxanide, while pyrvinium pamoate inhibited both mitochondrial and nonmitochondrial OCRs. These results demonstrate that the mitochondria of S. mansoni are a feasible target for drug development.

INTRODUCTION

Schistosomiasis, a disease caused primarily by Schistosoma mansoni, S. japonicum, and S. haematobium (1), results in approximately 280,000 deaths per year, making schistosomiasis the second-most-devastating parasitic disease after malaria (2, 3). In contrast to malaria, little effort has been spent on the development of new drugs against schistosomiasis, making the disease one of the 20 neglected tropical diseases, as designated by the World Health Organization (WHO) (4). In acute schistosomiasis, the cercaria, the larval stage of the parasite, actively penetrates mammalian skin and transforms into a distinct juvenile stage named the schistosomulum, which then invades the blood vessel and migrates sequentially through the lungs, heart, and portal vein, subsequently maturing into female and male adults. After mating, the worm pair migrates to the mesenteric veins where the female lays eggs, causing chronic schistosomiasis (5). The eggs are passed into the stool and to the environment, hatching into miracidia, which in turn infect snails and transform into sporocysts, daughter sporocysts, and then cercariae (6). Finally, cercariae leave the snails and infect mammals, completing the schistosome’s life cycle.

Praziquantel (PZQ) is the only drug available for treatment of schistosomiasis (7). However, PZQ does not confer protection against infection (i.e., prophylaxis) and does not completely kill adult parasites (8). Although the mechanism of action of PZQ is not well understood (9), parasites resistant to PZQ can be induced experimentally in infected mice (10), and reduced susceptibility has been reported to occur in various areas of endemicity (11). Given these shortfalls of PZQ, the development of new drugs for the treatment and prevention of schistosomiasis is needed.

Given the complexity of the helminth’s life cycle, these parasites have evolved efficient mechanisms for the smooth transitions among environments of various hosts and free-living stages (12), where their mitochondria are known to play key roles (13, 14). Under normoxic environment conditions (egg and larval stages), these parasites employ a classical oxygen-dependent electron transport chain (ETC) composed of NADH dehydrogenase (complex I), succinate:quinone reductase (complex II), quinol:cytochrome c reductase (complex III), cytochrome c oxidase (complex IV), and a high-redox-potential quinone (ubiquinone, midpoint potential [E_m] = +110 mV) (15), similar to that found in the mammalian host. Complexes I and II receive electrons from NADH and succinate, respectively, transferring the electrons to ubiquinone and then (consecutively) to complex III and complex IV via cytochrome c. Complexes I, III, and IV pump protons into the intermembrane space, generating an electrochemical gradient used by complex V for ATP synthesis (oxidative phosphorylation). However, once parasites mature into adults in the small intestine (a hypoxic environment), the parasite employs fumarate respiration, a pathway that is composed by complex I, a low-redox-potential quinone (rhodoquinone, E_m = −63 mV), and the reverse reaction of complex II (quinol:fumarate reductase) (16).

The most prominent advantage of fumarate respiration is the ability to produce ATP by oxidative phosphorylation independently of oxygen availability. In S. mansoni, fumarate respiration has been reported in the adult and sporocyst stages, while rhodoquinone-10 has been identified in all life cycle stages (17), suggesting that fumarate respiration occurs in all stages. Although the adult-stage parasite lives in mesenteric veins, where the oxygen saturation is approximately 60% to 75% (18), only 2% of the total oxygen is available in its dissolved form (19). In addition, mesenteric veins carry approximately 300 μM hydrogen sulfide (20), a toxic gas produced by gut flora and a potent inhibitor of oxygen respiration (21), thus suggesting that fumarate respiration is active in such an environment.

Drug development targeting the mitochondrial respiratory chain has been explored (22). The anthelminthic pyrvinium pamoate has been shown to inhibit fumarate respiration by the adult stage of Ascaris suum (23). Moreover, several antifungal agents such as siccanin, flutolanil, and fluopyran target complex II (24, 25). Atovaquone, an antimalarial drug, potently inhibits complex III from Plasmodium falciparum (26). Most recently, bedaquiline, a Food and Drug Administration-approved antitubercular drug, has been reported to be a complex V inhibitor (27) and also a mild uncoupler (28). Despite these pieces of evidence representing the proof of concept that ETC enzymes constitute a valuable target space for the development of new drugs to combat infectious diseases, little information is available about the impact caused by disruption of mitochondrion-related processes in prophylaxis and treatment of S. mansoni infection. In the present study, we investigated the in vitro, ex vivo, and in vivo antischistosomal activities of several compounds reported to inhibit mitochondrion-related processes and demonstrated the potential use of these compounds for prevention and treatment of S. mansoni infection.

RESULTS

Motility assay.

(i) Cercariae. The motility of S. mansoni cercariae was assessed in the presence of a panel of 116 compounds (listed in Table S1 in the supplemental material) known or thought to target mitochondrial function. After 41 h of exposure, 48 compounds showed a motility score of 2.0 or less; of these, 37 compounds showed complete inhibition of cercaria motility (with scores of 0.0), and another 11 compounds showed mean scores of 0.1 to 2.0 (Table S1).

As complex I inhibitors, we tested rotenone (29), pyrvinium pamoate (30), fenpyroximate (31), and derivatives of aurachin C and D (32) (Table S1). Rotenone, pyrvinium pamoate, and aurachin derivatives AC-0-12 and AD-9-1 completely inhibited motility of the cercariae (score, 0.0) after 41 h of exposure. However, fenpyroximate at this concentration was less active than the other tested compounds, given that motile cercariae still were observed after 41 h of exposure (mean score, 1.3) (Fig. 1).

FIG 1.

Effects of selected compounds on the relative motility of S. mansoni cercariae. Cercarial motility was assessed in the presence of pyrvinium pamoate, fenpyroximate, plumbagin, atovaquone, praziquantel, and dimethyl sulfoxide (DMSO; vehicle) (a) and ascofuranone, nitazoxanide, amiodarone, flusulfamide, and mefloquine (b). Each compound was assayed at a final concentration of 10 μM and 1% DMSO. Motility was evaluated microscopically in triplicates at each of three different time points (1, 18, and 41 h); the results are plotted as mean motility scores (ranging from 0 to 4) as described in Materials and Methods.

Tested complex II inhibitors included atpenin A5 (33), ferulenol and its derivatives (34), flutolanil and its derivatives (24), 2-heptyl-4-hydroxyquinoline n-oxide (HQNO) (35) and siccanin (36) (Table S1). Ferulenol showed complete inhibition of cercaria motility after 18 h of incubation; however, none of the tested ferulenol derivatives showed inhibition, even after 41 h. Flutolanil showed limited inhibition (mean score, 2.7) after 41 h; on the other hand, flusulfamide, a flutolanil derivative, inhibited motility of cercariae to a mean score of 0.0 after 18 h. Interestingly, siccanin, an inhibitor of fungal complex II, showed a high mean inhibition of score 1.3 within 1 h of exposure; however, cercaria motility recovered upon prolonged incubation with this compound (Table S1).

The complex III inhibitors licochalcone A (37), atovaquone/ascofuranone, and their derivatives (38, 39) showed mean inhibition scores of 0.0 at 41 h (Table S1). In addition to atovaquone, two of its derivatives, plumbagin and 511-12 {2‐hydroxy‐3‐[(2E,6E)‐3,7,11‐trimethyldodeca‐2,6,10‐trien‐1‐yl]‐1,4‐dihydronaphthalene‐1,4‐dione}, showed excellent anticercarial activity, with mean scores at 41 h of 0.0 and 0.3, respectively. Of 32 variants of ascofuranone, 26 derivatives showed mean inhibition scores of <1.3 after 41 h (Table S1).

Among the anthelmintics tested in this study, ivermectin (40) showed complete inhibition of cercarial motility after 18 h (mean score, 0.0), while nitazoxanide (41) reduced cercarial motility to a mean score of 1.7 after 41 h (Table S1). (Note that pyrvinium pamoate, which is also an anthelmintic, is listed as a complex I inhibitor in Table S1.) Five antimalarials were tested in this study; only mefloquine inhibited cercarial motility, with mean score of 0.0 after 18 h (Table S1). Among compounds with antitrypanosomal activity tested in this study, only amiodarone (42) showed complete inhibition of cercaria motility (mean score of 0.0) after 18 h (Table S1 and Fig. 1).

Evaluation of compounds reported to affect mitochondrion-related processes in other organisms demonstrated that α-, β-, and γ-mangostin (43), along with gambogic acid (44) and shikonin (45), completely inhibited cercarial motility after 18 h (mean scores, 0.0). 3-Nonylphenol (46) also affected cercaria motility, though with lower potency (mean score, 1.3) than the compounds mentioned above.

Cercariae treated with any of the remaining compounds or with 1% (vol/vol) dimethyl sulfoxide (DMSO) survived 41 h of exposure.

(ii) Schistosomula. No difference was observed in the motility of schistosomula after 48 h of incubation with DMSO, PZQ, ascofuranone, fenpyroximate, or flusulfamide (Fig. 2). Complete inhibition of schistosomula motility was observed after incubation for 8 h with atovaquone, after 24 h with amiodarone, nitazoxanide, mefloquine, or plumbagin, and after 48 h with pyrvinium pamoate (Fig. 2).

FIG 2.

Effects of selected compounds on the relative motility of S. mansoni schistosomula. Schistosomula motility was assessed in the presence of praziquantel, ascofuranone, pyrvinium pamoate, amiodarone, atovaquone, and dimethyl sulfoxide (DMSO; vehicle) (a) and fenpyroximate, flusulfamide, nitazoxanide, mefloquine, and plumbagin (b). Each compound was assayed at a final concentration of 10 μM and 1% DMSO. Motility was evaluated microscopically in triplicates at each of four different time points (1, 8, 24, and 48 h); the results are plotted as mean motility scores (ranging from 0 to 4) as described in Materials and Methods.

(iii) Adults. Upon exposure to selected compounds, the pair of S. mansoni adults began to separate, and the effect on fecundity could not be addressed in this study. Therefore, the effect of each compound was evaluated individually for each male and female. Although amiodarone inhibited the motility of adult S. mansoni, inhibition after 20 h of incubation was not complete, providing mean motility scores of 0.3 and 1.0 for the male and female, respectively (Fig. 3b and d). The remaining compounds completely inhibited the motility of male S. mansoni after 20 h of incubation (Fig. 3a and b). In the case of females, similar results were obtained after 20 h of incubation with PZQ, nitazoxanide, mefloquine, pyrvinium pamoate, plumbagin, ascofuranone, and flusulfamide (Fig. 3c and d). However, atovaquone and fenpyroximate were not effective against females of S. mansoni (Fig. 3c).

FIG 3.

Effects of selected compounds on the relative motility of male and female adult S. mansoni. The mean motility score (ranging from 0 to 4) of male (a and b) and female (c and d) adults in the presence of dimethyl sulfoxide (DMSO), pyrvinium pamoate, atovaquone, fenpyroximate, mefloquine, nitazoxanide, and praziquantel (a and c) and flusulfamide, ascofuranone, amiodarone, and plumbagin (b and d). Each compound was assayed at a final concentration of 10 μM and 1% DMSO. Motility was evaluated microscopically in triplicates at each of three different time points (1, 9, and 20 h); the results are plotted as mean motility scores (ranging from 0 to 4) as described in Materials and Methods.

In vivo studies.

(i) Prophylaxis. Consistent with previous reports (47, 48), PZQ did not protect mice against S. mansoni infection, with animals exhibiting a worm burden of 89.6% (P > 0.05) relative to the negative-control treatment (vehicle) (Fig. 4a). In contrast, the worm burden was significantly suppressed (P < 0.05) following prophylaxis with each of the selected compounds, such that hosts exhibited worm burdens ranging between 1.9% and 15.0% (Fig. 4a). Among the selected compounds, ascofuranone, plumbagin, and pyrvinium pamoate exhibited the strongest prophylactic activity, with worm burden reduced to 1.9%, 2.3%, and 2.9%, respectively (P < 0.05) compared to that with mefloquine (14.1%; positive control) and DMSO (100%; negative control) (Fig. 4a). Furthermore, the worm burdens of mice treated with fenpyroximate, atovaquone, flusulfamide, amiodarone, or nitazoxanide also were reduced, in these cases to 8.7%, 10.8%, 12.8%, 14.7%, or 15%, respectively, relative to that with the vehicle (Fig. 4a).

FIG 4.

Prophylactic and therapeutic activity of selected compounds on S. mansoni infection. Mice (n = 6/group) were treated with ascofuranone (100 mg/kg body weight), plumbagin (2 mg/kg), pyrvinium pamoate (2 mg/kg), fenpyroximate (2 mg/kg), atovaquone (100 mg/kg), flusulfamide (5 mg/kg), amiodarone (50 mg/kg), nitazoxanide (50 mg/kg), mefloquine (100 mg/kg), praziquantel (100 mg/kg), or vehicle (containing 1% DMSO). Animals were dosed by 4 days of once-daily intraperitoneal injection at the indicated dosage, starting 1 day prior to infection. Animals were euthanized 7 days after infection; schistosomula then were recovered from lungs of each mouse, counted, and used to calculate the worm burden. (a and b) Mice (n = 6/group) at week 6 postinfection were treated intraperitoneally with selected compounds for 4 days at the dosage mentioned above. Mice were sacrificed 14 days after the last treatment, and adult parasites were recovered and counted. The worm burden was calculated as mentioned in Materials and Methods. For both panels, data are presented as means and standard deviations of values normalized to those in vehicle-treated animals (defined as 100%).

(ii) Therapy. Under the conditions tested in this study, no worms were recovered from the mice treated with PZQ (Fig. 4b). In the groups of mice treated with fenpyroximate, atovaquone, or flusulfamide, the worm burdens were 72.0%, 69.9%, and 86.6%, respectively (Fig. 4b). A reduction of worm burden to <50% was achieved in mice treated with ascofuranone (45.0%), plumbagin (18.3%), pyrvinium pamoate (23.6%), amiodarone (29.2%), nitazoxanide (23.0%), and mefloquine (19.9%; P > 0.05) (Fig. 4b).

Oxygen consumption assay.

Under all the tested conditions, the OCR was increased upon addition of respiratory substrates (Fig. 5 and Fig. S1). After the addition of a mitochondrial uncoupler [carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP)], further increases in the OCR were observed (Fig. 5 and Fig. S1). A significant reduction in OCR was observed after addition of nitazoxanide, atovaquone, ascofuranone, pyrvinium pamoate, mefloquine, amiodarone, or fenpyroximate (Fig. 5 and Fig. S1). A decrease of the OCR to baseline levels (mitochondrial OCR) was achieved following addition of atovaquone to reaction mixtures containing nitazoxanide, atovaquone, ascofuranone, mefloquine, flusulfamide, amiodarone, or PZQ (Fig. 5 and Fig. S1). Reaction mixtures containing plumbagin or plumbagin plus atovaquone showed no change in OCR, which remained at the same level as that observed in the presence of FCCP (Fig. 5 and Fig. S1). In the case of pyrvinium pamoate, both mitochondrial and nonmitochondrial OCRs were completely inhibited (Fig. 5 and Fig. S1).

FIG 5.

Oxygen consumption rates (OCRs) of adult S. mansoni pairs in the presence of selected compounds. The OCR was determined in the presence of 25 mM glucose, 1 mM pyruvate, and 5 mM l-glutamine (substrates). The first reading after addition of substrates was set as baseline. Carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP) was added to a final concentration of 10 μM to induce maximum respiration. Effects on the OCR were evaluated following addition of ascofuranone, plumbagin, nitazoxanide, pyrvinium pamoate, or atovaquone (a) or mefloquine, praziquantel (PZQ), flusulfamide, amiodarone, or fenpyroximate (b) to final concentrations of 50 μM each. Atovaquone was then added to completely inhibit mitochondrial OCR. The OCR resistant to atovaquone was defined as nonmitochondrial respiration. Means and standard deviations are shown for each time point of OCR measured in triplicates.

DISCUSSION

In contrast to mammalian respiration, which is strictly aerobic, depending on the life cycle stage, helminths are able to perform aerobic (oxygen) and anaerobic (fumarate) respiration (16, 49). Interestingly, fumarate respiration also has been detected in isolated mitochondria from protozoan parasites such as P. falciparum (50) and Eimeria tenella (24), suggesting the use of fumarate respiration among evolutionarily unrelated parasitic organisms. It is important to note that development from S. mansoni cercariae to the adult stage is characterized by a gradual transition from a normoxic to a hypoxic environment, with changes in energy metabolism according to available carbon sources. We hypothesized that, given these transitions in setting, disruption of mitochondrion-related processes could be detrimental for S. mansoni development and survival, a weakness that might be exploited to prevent infection as well as to treat established infection.

While the effects of ascofuranone, flusulfamide, fenpyroximate, and amiodarone on the various life cycle stages of S. mansoni are reported here for the first time, 4 of the 8 selected compounds (atovaquone, pyrvinium pamoate, plumbagin, and nitazoxanide) were previously reported to be active against the adult stage in vivo (atovaquone and nitazoxanide) and ex vivo (nitazoxanide and plumbagin) and against schistosomula (pyrvinium pamoate, nitazoxanide, and plumbagin) and cercariae (nitazoxanide and plumbagin) in vitro, as summarized in Table S2 in the supplemental material. For better comparison, all of these compounds (including the controls) were evaluated in this study.

Most compounds active against cercariae are thought to be inhibitors of complex I, II, and/or III, suggesting that S. mansoni depends on an active aerobic respiratory chain to survive. Interestingly, the majority of the anticercarial compounds identified in the present work are molecules that target complex III and include ascofuranone, atovaquone, and their derivatives (Table S1). These results suggest that modifications to meroterpenoid (51) and naphthoquinone (52) scaffolds may provide candidates with excellent anticercarial activity.

In contrast to cercariae, schistosomula were insensitive to PZQ and to inhibitors of complex I (fenpyroximate) (53), complex II (flusulfamide) (54), and complex III (ascofuranone) (55) and were less sensitive to a complex I and II inhibitor (pyrvinium pamoate) (23) but were sensitive to a complex II and III inhibitor (atovaquone) (39) (Fig. 2). These findings suggest that simultaneous inhibition of oxygen and fumarate respirations is required to cause lethality in this stage, indicating that mechanically transformed schistosomula depend on at least one of this pair of respiratory strategies for survival. Moreover, mefloquine (56), amiodarone (57), and nitazoxanide (58) were active against schistosomula, probably because of these compounds’ ability to induce the depletion of intracellular ATP levels (Fig. 2). Plumbagin’s efficacy against schistosomula may be attributable to its ability to compete with respiratory quinones (rhodoquinones and ubiquinones) for electrons, resulting in the generation of semiquinone radicals and reactive oxygen species (ROS) (59) (Fig. 2).

It previously was reported that male schistosomes have a higher mitochondrial respiration rate than females, while the nonmitochondrial respiration rate is higher in females than in males (60). These observations highlight the stronger dependence of males than females on the mitochondrial respiratory chain (60), consistent with the results obtained in the present study. Interestingly, pyrvinium pamoate and ascofuranone, which have been reported to inhibit fumarate respiration in parasitic helminths (23, 61), completely inhibited parasite motility within 20 h of incubation (Fig. 3), suggesting that both females and males employ active fumarate respiration.

From the selected compounds, nitazoxanide (62), atovaquone (39, 63–65), ascofuranone (66), mefloquine (65, 67), and praziquantel (65, 67–70) have been reported to be effective by oral administration for treatment of parasitic infection in models. Since the oral administrations in previous reports were not standardized and pyrvinium pamoate has no absorption by oral route (71), for better comparison, the administration of all selected compounds either as prophylactics or as therapeutics of S. mansoni infection was performed intraperitoneally.

PZQ does not inhibit the motility of schistosomula in vitro, and as expected, this compound failed to prevent the S. mansoni infection in the present study (Fig. 4a) (8). In contrast to the insensitivity to several compounds of schistosomula transformed in vitro (mechanically) (Fig. 2), schistosomula transformed in vivo (subcutaneously) were susceptible to all of the selected compounds (Fig. 4b), suggesting differential dependency on mitochondrial respiration between in vitro- and in vivo-transformed schistosomula. Our results support previous results indicating that subcutaneously transformed schistosomula show higher rates of mitochondrial metabolism than mechanically transformed schistosomula (72).

Although the worm burden was nominally decreased in mice treated with flusulfamide, fenpyroximate, or atovaquone, the effect was not significant (P value > 0.05); similar results (apart from flusulfamide) were obtained ex vivo using females (Fig. 4b). Possible explanations are that (i) flusulfamide is a weak inhibitor of complex II (having a reported 50% inhibitory concentration [IC₅₀]of 76.5 μM in A. suum) (54) and (ii) at the dose of 5 mg/kg body weight, the plasma concentration of this compound did not reach a level sufficient to kill the parasites (73). We previously showed that ascofuranone (61) and pyrvinium pamoate (23) inhibit helminth fumarate respiration. Given the significant (P value < 0.05) reductions in worm burdens (to 45% and 23.6% of vehicle control) in mice treated with ascofuranone or pyrvinium pamoate, respectively (Fig. 4b), these results suggest that the S. mansoni adult stage depends on fumarate respiration. Under our experimental conditions, plumbagin showed the strongest reduction in the worm burden (to 18.3% of vehicle control), a result that may be attributable to the generation of ROS (59, 74), as discussed above. Mefloquine, nitazoxanide, and amiodarone reduced the worm burden to 19.9% to 29.2% of the vehicle control (Fig. 4b); however, the mechanisms of action of these compounds are not completely understood, though these molecules have been suggested to affect mitochondrial membrane potential or ROS generation (75–78). Collectively, our results reinforce the notion that the S. mansoni adult stage relies on active mitochondria, making the related pathways feasible as drug targets.

The significant reduction in OCR observed with atovaquone, ascofuranone, pyrvinium pamoate, mefloquine, amiodarone, and fenpyroximate (Fig. 5 and Fig. S1) indicates that these compounds are, in fact, S. mansoni respiratory chain inhibitors. Unexpectedly, nitazoxanide, which has been reported to act as a mild uncoupler and thereby enhance OCR (77, 78), inhibited the OCR of adult pairs to the same degree as ascofuranone and atovaquone. Based on this finding, it is tempting to speculate that nitazoxanide may be inhibiting complex III, an effect that may be surpassing the compound’s mild uncoupling effect, thereby causing the observed decrease in OCR. However, additional studies will be needed to verify these results and this hypothesis. Interestingly, pyrvinium pamoate completely inhibited both mitochondrial and nonmitochondrial OCRs (Fig. 5), suggesting that this compound might have other targets related to nonmitochondrial respiration. Moreover, plumbagin did not reduce the OCR but instead maintained the OCR at a level similar to that seen with FCCP; in the presence of plumbagin, OCR was insensitive to the effect of atovaquone (Fig. 5). This result supports the hypothesis that plumbagin acts as an electron acceptor for complex I or II while bypassing complex III, thereby maintaining the OCR at a high level.

In this study, we demonstrated for the first time (to our knowledge) that inhibitors of mitochondrion-related processes have potential for use in chemoprophylaxis and merit further development. Although the molecular target (complexes I to IV) of these compounds could not be identified, we show that among the mitochondrion-related processes, mitochondrial respiration is severely inhibited and potentially the target pathway. These compounds, especially those of the ascofuranone and the FDA-approved antimalarial drug atovaquone classes, have great advantages over PZQ, given their high efficacy in reducing the worm burden in the lungs. Thus, these compounds have the potential to meet the requirements of at least two target product profiles of the four proposed by Caffrey et al. (79). We postulate that such chemicals could be used in combination with PZQ for the control and elimination of schistosomiasis. In conclusion, the mitochondrion of S. mansoni is a good drug target space; the results obtained in the present study provide starting points for the development of new drugs for the prevention and treatment of schistosomiasis.

MATERIALS AND METHODS

Ethical statement.

Mouse experiments were approved by Nagasaki University’s Animal Research Committee (number [no.] 1506181240); animals were handled per the relevant protocols of Japanese law specified in the Humane Treatment and Management of Animals (law no. 105, dated 19 October 1973 and subsequently revised as of 2 June 2006).

Compounds.

This study tested a collection of 116 compounds (see Table S1 in the supplemental material). The panel included compounds that have been reported to inhibit mitochondrion-related processes (such as ETC, cellular respiration, and membrane potential), classical antiparasitic agents, and a small number of molecules used for treatment of human disease. This panel (of our laboratory compound library) previously was described as part of a separate study conducted in our laboratory (80). Stock solutions of the compounds were available at 1 mM and assayed at 10 μM, in order to maintain the final concentration of dimethyl sulfoxide (DMSO) at no more than 1%, against S. mansoni in vitro and ex vivo as has been reported (81).

Maintenance of S. mansoni parasites.

A Puerto Rican strain of S. mansoni was maintained essentially as described previously (82).

Motility assay.

The motility and viability assay has been widely used to screen a small subset of compounds. It can also be easily adapted to laboratories because of its low cost, and it was used for the first screening in this study (83). Cercariae were evaluated microscopically by comparing parasites in the presence of 10 μM compounds to those in wells containing vehicle (DMSO) at three different time points (≤1 h, 18 h, and 41 h), as has been reported (84). Motility was scored using a 5-point scale, as described previously (4, normal motility; 3, reduced motility; 2, uncoordinated minimal motility, 1, severe reduction in motility; 0, total absence of mobility) (60).

Schistosomula were obtained through mechanical transformation from cercariae and purified using Percoll gradient as described previously (85). Schistosomula were transferred (at approximately 30 per well) into 96-well plates containing RPMI medium supplemented with 5% (vol/vol) fetal bovine serum (FBS), 10 mM glutamine, and penicillin-streptomycin (10 U/ml and 10 μg/ml, respectively), and the plates were incubated overnight at 37°C in a CO₂ incubator. On the following day, a subset of 8 compounds (selected according to their activity against cercariae as well as their commercial availability and presence in amounts sufficient for in vivo experiments) was selected, including atovaquone, nitazoxanide, flusulfamide, fenpyroximate, plumbagin, amiodarone, pyrvinium pamoate (MP Biomedicals, LLC), and ascofuranone (Institute of Mitochondrial Sciences, Inc.). Compounds of this panel of 8, as well as mefloquine (Tokyo Chemical Industry Co., Ltd.) (67, 81, 86) and PZQ (the positive controls), were added at a final concentration of 10 μM (and 1% [vol/vol] DMSO) to the individual wells of the schistosomula-containing plates; each compound was tested in triplicates. The motility of the schistosomula was scored (as described above) at 4 time points (≤1, 8, 24, and 48 h).

Five-week-old female ICR mice (Japan SLC Inc.) were kept in the environmentally controlled animal facility at Nagasaki University (25°C, 70% humidity, 12-h light and dark cycle) with availability of water and food. Mice were kept for a week to acclimatize before treatment and/or infection. Thirty-five mice were infected with approximately 150 cercariae per animal. After 8 weeks, mice were euthanized, and adult worms were recovered through perfusion of the hepatic portal system and mesenteric veins (85). Schistosomes were washed using RPMI medium supplemented with 5% (vol/vol) FBS plus penicillin-streptomycin (10 U/ml and 10 μg/ml) and incubated overnight at 37°C in a CO₂ incubator. Adult schistosomes (10 pairs/well) were transferred to 24-well plates containing selected compounds at 10 μM; each compound was tested in triplicates. Motility was evaluated microscopically and scored as described above.

In vivo prophylactic assay.

Because there are no validated drugs or vaccines for prophylaxis of schistosomiasis, we tested whether or not the selected compounds have potential prophylactic activity. Groups of 6 ICR mice (maintained as described above) each were used to test the effectiveness of selected compounds: atovaquone, 100 mg/kg body weight; nitazoxanide, 50 mg/kg; ascofuranone, 100 mg/kg; flusulfamide, 5 mg/kg; fenpyroximate, 2 mg/kg; plumbagin, 2 mg/kg; pyrvinium pamoate, 2 mg/kg; amiodarone, 50 mg/kg; mefloquine, 100 mg/kg; PZQ, 100 mg/kg; and vehicle (1× phosphate-buffered saline [PBS] containing 3% [vol/vol] ethanol and 7% [vol/vol] Tween 80). The compounds were administered intraperitoneally (on the left side of the abdomen) 1 day before infection (day −1). The respective compounds were administered again on day 0, and 3 h later, the mice were infected subcutaneously (on the right side of the abdomen) with approximately 500 S. mansoni cercariae/mouse. Administration of the respective compounds was repeated once daily for 2 additional days postinfection (days 1 and 2) (i.e., for a total of 4 doses). Mefloquine (67, 81, 86) and vehicle containing 1% (vol/vol) DMSO were used as positive and negative controls, respectively. Six days postinfection, mice were euthanized, lungs were collected, and schistosomula were recovered (85, 87) and counted. The worm burden was calculated as described previously using the following formula (68):

$$\text{Worm burden}\hspace{0.17em}\left(\text{\%}\right)=\hspace{0.17em}\frac{(\mathit{NW}\text{neg}\hspace{0.17em}-\hspace{0.17em}\mathit{NW}\text{tre})}{\mathit{NW}\text{neg}}\hspace{0.17em}\times 100,$$

where NW_neg and NW_tre represent the mean numbers of worms in the negative-control and treated groups, respectively.

In vivo therapeutic assay.

ICR mice were infected, as described above, with approximately 150 cercariae/mouse. At week 6, mice were treated with selected compounds by 4 days of once-daily intraperitoneal injection using the same dosage as for the prophylaxis assay. PZQ and mefloquine were used as positive controls (67, 81, 86), and the vehicle was used as a negative control. At 14 days after the final dose administration, worms were collected via perfusion, mesenteric veins were examined to count any trapped adult worms (88), and the worm burden was calculated as described above.

Determination of oxygen consumption rates.

The oxygen consumption rate (OCR) was determined using a Seahorse XFe24 extracellular flux analyzer (Agilent Technologies) as described previously (89, 90). A pair of adults was placed in each well, an islet capture screen was inserted into the wells, and the plate was loaded into the XFe24 analyzer. OCR was determined at 37°C in the presence of 15 mM glucose, 1 mM pyruvate, and 5 mM l-glutamine (port A) as respiratory substrates and 10 μM carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP; port B) as the uncoupler. Reproducible detection of OCR changes in a short period of time (25 min) was achieved using 50 μM of each of the selected compounds (port C). To completely quench mitochondrial OCR, atovaquone was added at 50 μM (port D). The experiments were performed using a programmed protocol consisting of 2 min of mixing, 3 min of waiting, and 3 min of measuring time per cycle for five cycles between injections.

Data analysis.

All data were analyzed using Prism version 8 (GraphPad). The t test was performed for all groups, and a P value of <0.05 was considered significant.