Target-based design of praziquantel analogs at cestode TRPM_PZQ

Abstract

The drug praziquantel (PZQ) has been used for decades to treat clinical and veterinary infections caused by parasitic flatworms. Although PZQ is efficacious against many different types of flukes and tapeworms, PZQ activity is lower against certain types of parasites, including pseudophyllidean cestodes. The target of PZQ is a parasitic flatworm transient receptor potential ion channel (TRPM_PZQ), and interrogation of this target affords opportunity to understand why PZQ efficacy varies between different parasites, and how target-based design strategies could help deliver new analogs with improved efficacy against currently hard-to-treat diseases. In this study, we consider natural amino acid variation within cestode TRPM_PZQ binding pockets to design thioamide derivatives of PZQ with greater efficacy at pseudophyllidean cestode TRPM_PZQ. Target-based design across parasite TRPM_PZQ orthologs, as well as at other TRPM paralogs in this ion channel family, provides opportunity to expand and improve on the current anthelmintic toolbox.

Graphical Abstract

The anthelmintic drug praziquantel (PZQ) is a World Health Organization (WHO) essential medicine that is used to treat parasitic flatworm infections of clinical and veterinary importance ¹. Following decades of clinical use, a candidate target for PZQ was recently discovered in the blood fluke Schistosoma mansoni ². The target is a transient receptor potential (TRP) ion channel within the melastatin subfamily, named Sm.TRPM_PZQ ². TRPM_PZQ is a flatworm-specific, nonselective, large conductance cation channel ³. Activation of Sm.TRPM_PZQ triggers excitotoxicity via membrane depolarization causing spastic contraction of the worm, and damage to the tegumental surface of the parasite. Exposure of parasite antigens catalyzes immunological clearance of damaged worms from the infected host ⁴. A growing body of evidence from genetic association studies, pharmacological screening, and functional profiling of TRPM_PZQ orthologs supports TRPM_PZQ as the therapeutically relevant target of PZQ ^{5, 6}.

Despite being a cheap, safe, and effective drug, there remain opportunities to improve on the performance of PZQ. One opportunity is the fact that PZQ is not equally effective against all types of parasitic flatworm ⁶. This includes both complete insensitivity (PZQ is ineffective against liver fluke of the genus Fasciola) and reduced efficacy against specific parasites, such as pseudophyllidean cestodes. Developing strategies to overcome these limitations has been challenging as in the absence of a defined molecular target, researchers have necessarily relied on profiling PZQ analogs through phenotypic screening approaches. Almost all PZQ derivatives that have been profiled display a lower efficacy against schistosomes than PZQ itself ^{1, 7}. In retrospect, this was likely a consequence of stringent structure-activity requirements for engagement of the PZQ binding pocket in schistosome TRPM_PZQ ⁸, as well as unappreciated amino acid variation between the binding pockets of different parasitic flatworms ⁹.

However, with the identity of TRPM_PZQ now established, there is an opportunity to leverage target-based approaches to discover new leads. One such approach is high throughput screening (HTS) ¹⁰. HTS represents an unbiased strategy, and recent application of HTS targeting TRPM_PZQ has yielded new chemotypes active against Fasciola spp. ¹¹. However, HTS requires rigorous pre- and post-assay validation and it remains prohibitively expensive to execute large compound library screens against every TRPM_PZQ ortholog. Alternative approaches encompass methods that focus on target-based ligand design, tailoring analogs to the specific architecture of parasite binding pockets that accommodate their unique sequence variation.

In this study, we present a first attempt at rational target-based design at TRPM_PZQ, focusing on the challenge of identifying PZQ analogs with improved properties against pseudophyllidean cestodes. This group of tapeworms includes several that cause rare clinical infections poorly responsive to PZQ, such as diphyllobothriasis ¹² and sparganosis ^{13, 14}. For example, human sparganosis is only effectively treated by surgery to remove infectious material, given the poor PZQ sensitivity of Spirometra spp. ^{13, 14}. The low sensitivity of these infections toward PZQ likely relates to a specific acidic amino acid variant within the pseudophyllidean TRPM_PZQ binding pocket. While an aspartic acid (‘D’) exists in PZQ-sensitive cyclophyllidean cestodes, a glutamic acid (‘E’) is present in PZQ-resistant pseudophyllidean cestodes. The glutamic acid variant has been shown to be deleterious to PZQ occupancy of the voltage sensor-like domain (VSLD) binding pocket in TRPM_PZQ ⁹.

Encouraged by recent observations that cestode TRPM_PZQ is better accommodating of certain PZQ derivatives than schistosome TRPM_PZQ ¹⁵, we pondered whether this would afford opportunity to design analogs with improved activity toward pseudophyllidean cestodes. By screening a library of fifty PZQ derivatives, we discovered that (i) metabolically resistant analogs of PZQ retained equivalent potency to PZQ at cyclophyllidean cestode TRPM_PZQ, and (ii) these analogs could then be rationally modified to target pseudophyllidean TRPM_PZQ and partially mitigate the effects of the glutamic acid residue deleterious to PZQ occupancy of the binding pocket. The improved efficacy of these new analogs at pseudophyllidean TRPM_PZQ supports further exploration of target-based design strategies to combat rare, hard-to-treat parasitic infections.

Profiling cyclophyllidean cestode TRPM_PZQ.

Previous structure-activity studies using cyclohexyl-modified analogs of PZQ demonstrated many of these analogs displayed activity against cyclophyllidean cestodes but not schistosomes ¹. This observation paralleled the activity of the same analogs at Echinococcus granulosus TRPM_PZQ (Eg.TRPM_PZQ) compared with Schistosoma mansoni TRPM_PZQ (Sm.TRPM_PZQ, ¹⁵). Such data imply that the ligand binding pocket of cyclophyllidean cestode TRPM_PZQ accommodates a broader range of PZQ-based structures than schistosome TRPM_PZQ. However, this cannot be stated unequivocally, given the limited number and structural diversity of the analogs profiled in the prior study ¹⁵. Therefore, we screened a more expansive library of 50 PZQ derivatives against cestode and blood fluke TRPM_PZQ (Eg.TRPM_PZQ and Sm.TRPM_PZQ) using a Ca²⁺ reporter assay. Full concentration response curves were generated, and the potency of each analog at Eg.TRPM_PZQ and Sm.TRPM_PZQ were represented by EC₅₀ values plotted in Figure 1A. A description of analogs and associated data is provided in Table S1. Several trends were evident from these data.

Figure 1.

(A) Functional profiling of a library of fifty praziquantel analogs at Sm.TRPM_PZQ and Eg.TRPM_PZQ. Epsiprantel and the two PZQ enantiomers are highlighted. Data presented as mean EC₅₀. (B) Comparison of the difference in potency for each compound at Sm.TRPM_PZQ and Eg.TRPM_PZQ. (C) Concentration-response curve of praziquantel enantiomers at trematode Sm.TRPM_PZQ (blue) and cestode Eg.TRPM_PZQ (red). Data plotted as mean ± SEM of n ≥ 3 biological replicates comprised of technical duplicates.

First, whereas some PZQ analogs displayed no activity at Sm.TRPM_PZQ, all tested PZQ analogs were active at Eg.TRPM_PZQ (Figure 1A). Each of the tested PZQ analogs displayed higher potency at Eg.TRPM_PZQ compared with Sm.TRPM_PZQ (Figure 1A). These observations confirm a broader tolerability of the cestode TRPM_PZQ binding pocket to PZQ analogs with varied structures compared with schistosome TRPM_PZQ.

Second, while the difference in potency between these TRPM_PZQ orthologs was typically ~10-fold (EC₅₀ at Eg.TRPM_PZQ / EC₅₀ at Sm.TRPM_PZQ; Figure 1B), there were outliers. One example involves the enantiomers of PZQ: (R)-PZQ showed little difference in potency between Eg.TRPM_PZQ and Sm.TRPM_PZQ (~4-fold), while (S)-PZQ was far more potent at Eg.TRPM_PZQ that observed at Sm.TRPM_PZQ (~100-fold; Figure 1B), due to higher potency of (S)-PZQ at Eg.TRPM_PZQ (0.49 ± 0.13 μM, Figure 1A) compared with Sm.TRPM_PZQ (EC₅₀ >50 μM). As PZQ is typically administered as a racemic mixture ((±)-PZQ), the smaller eudysmic ratio of PZQ enantiomers at Eg.TRPM_PZQ (~6 fold) versus Sm.TRPM_PZQ (~270 fold, Figure 1C) implies both PZQ enantiomers likely contribute to the in vivo effectiveness of PZQ in the treatment of cestode infections.

Third, epsiprantel, a pyrazinobenzazepine PZQ derivative broadly used as a veterinary cestocide was less potent than PZQ at Eg.TRPM_PZQ (EC₅₀ = 1.92 ± 0.39 μM, Figure 1A), as well as a different cestode TRPM_PZQ ortholog, Mesocestoides corti, Mc.TRPM_PZQ (EC₅₀ = 1.02 ± 0.27 μM, Figure S1). These data likely underscore the importance of pharmacokinetic and pharmacodynamic contributions to the in vivo activity of epsiprantel. It is a very insoluble compound that is poorly absorbed from the gastrointestinal tract ^{16, 17}. Oral dosing therefore maximizes exposure of intestinal tapeworms to epsiprantel due to this lack of absorption, even though ‘on-target’ activity at cestode TRPM_PZQ is lower than that of PZQ.

Finally, it was clear that metabolically stable analogs of PZQ were well tolerated in the cestode TRPM_PZQ binding pocket. The cyclohexyl moiety of PZQ is the main site of CYP-mediated oxidative metabolism (half time of PZQ is 2-3hrs in humans ¹⁸), with the -OH metabolite (compound 20) exhibiting considerably lower activity at TRPM_PZQ. This was the case for (compound 20) at both Eg.TRPM_PZQ (EC₅₀ = 2.5 ± 0.48 μM, Figure 2A) and at Sm.TRPM_PZQ ¹⁹. However, PZQ analogs (compound 43 and compound 49) modified at this position, that are more metabolically stable ¹⁹, retained equivalent potency to PZQ within the cestode TRPM_PZQ binding pocket (Figure 2B& 2C). This was not the case at Sm.TRPM_PZQ, where the metabolically resistant difluorocyclohexyl analog was poorly tolerated ¹⁹. Again, these data underscore broader ligand tolerability of the cestode TRPM_PZQ binding pocket, and potential equivalency of these metabolically resistant analogs at cestode TRPM_PZQ.

Figure 2.

Functional profiling of (A) PZQ-metabolite, 20, and metabolically stable analogs (B) 43 and (C) 49 at Eg.TRPM_PZQ, compared with (R)-PZQ in each condition. Data plotted as mean ± SEM of n ≥ 3 biological replicates comprised of technical duplicates.

Profiling pseduophyllidean cestode TRPM_PZQ.

Pseudophyllidean cestodes show poor PZQ sensitivity ^{9, 20, 21}, creating a challenge for drug treatment of clinical infections. A proposed explanation for this lower PZQ sensitivity is the presence of a glutamic acid residue in the TRP helix of pseduophyllidean cestodes (E1525 in Spirometra erinaceieuropaei, Se.TRPM_PZQ, Figure 3A) that adversely impacts PZQ occupancy of the VSLD binding pocket ⁹. This residue is represented by an aspartic acid residue in cyclophyllidean cestode TRPM_PZQ and trematode TRPM_PZQ ⁹. Comparison of the PZQ sensitivity of Eg.TRPM_PZQ and Se.TRPM_PZQ, a pseduophyllidean cestode TRPM_PZQ representative, revealed (R)-PZQ exhibited lower potency (EC₅₀ = 1.0 ± 0.30 μM) and efficacy (B_max = 17 ± 0.8% of (R)-PZQ at Eg.TRPM_PZQ) at Se.TRPM_PZQ (Figure 3B).

Figure 3. The interactions of (R)-PZQ in pseudophyllidean cestode Se.TRPM_PZQ.

(A) Schematic of the PZQ binding pocket of Se.TRPM_PZQ. Key interactions of PZQ within the pocket are highlighted, including E1525 (orange), predicted to be deleterious to the effective binding of PZQ in this receptor. (B) Concentration-response curve of (R)-PZQ at Eg.TRPM_PZQ (closed circles) and Se.TRPM_PZQ (open circles). PZQ is a partial agonist of Se.TRPM_PZQ. (C) MD simulations of (R)-PZQ complexed with Se.TRPM_PZQ. The top pose from each independent run is overlaid (blue), and compared to the pose of (R)-PZQ in Eg.TRPM_PZQ (pink). (D) Most common binding pose of (R)-PZQ in Se.TRPM_PZQ, resulting from the clustering of trajectories in C. (E) A zoomed-in view of the binding pose from D, showing the acute angle that was formed between H1239 and the carbonyl of (R)-PZQ. (F) Schematic of the angle preference of a hydrogen bond donor with a carbonyl (left) vs. thiocarbonyl (right). Data plotted as mean ± SEM of n ≥ 3 biological replicates comprised of technical duplicates.

To better understand the partial agonism of (R)-PZQ at Se.TRPM_PZQ, molecular dynamics experiments were performed. Eight replicates of 500 ns simulations of (R)-PZQ within Se.TRPM_PZQ yielded a reproducible (R)-PZQ binding pose (Figure 3C, blue) that diverged from the predicted binding pose of (R)-PZQ in Eg.TRPM_PZQ (Figure 3C, pink). The eight simulations were clustered into a common binding pose (Figure 3D). In this pose, PZQ adopts a vertical orientation, with a critically-positioned histidine residue in the S1 helix adopting a conformation that positions the hydrogen responsible for the key hydrogen-bond to PZQ at ~89° angle (Figure 3E). Favorable interactions for an NH-O hydrogen bond to a carbonyl position the dihedral angle at ~116° (Figure 3F). Therefore, we posit that when E1525 dislodges PZQ from its preferred binding orientation, this S1 histidine (H1239) is no longer able to interact optimally with PZQ.

It has been described by Lampkin and VanVeller ²² that based on orbital overlap and electrostatic interactions, thioamide hydrogen bond acceptors prefer a more acute dihedral angle compared to their oxygen-containing carbonyl counterparts (Figure 3F). Therefore, we hypothesized that an analog of PZQ possessing a thioamide in place of the internal carbonyl may restore activity at the channel by recovering an interaction between H1239 and the ligand. Based on this hypothesis, we synthesized this internal thioamide in the context of the metabolically stable cyclohexyl derivatives profiled in Figure 2. This resulted in compounds 51 and 52 (Figure 4A). When molecular dynamics were performed with 52 in the pocket of Se.TRPM_PZQ, simulations predicted that the molecule adopts an orientation more reminiscent of the proposed binding pose of PZQ in Eg.TRPM_PZQ (Figure 4B, green vs pink); this is also consistent with the predicted binding pose of PZQ in other trematode TRPM_PZQ binding pockets ⁹. After clustering the 8 replicates into the most common binding pose in the binding pocket of Se.TRPM_PZQ, the new thiocarbonyl in 52 was predicted to interact with H1239 (Figure 4C). During MD simulations, this angle of the hydrogen bond between H1239 and 52 oscillated between ~95° and 115°, and this newly formed interaction is predicted to anchor 53 within the VSLD pocket. Likewise, there was simultaneous hydrogen bonding to R1362, and π-π stacking with Y1365 as is observed in previous simulations with PZQ in the Sm.TRPM_PZQ binding pocket ⁹.

Figure 4. Thiocarbonyl derivatives of PZQ restore efficacy at Se.TRPM_PZQ.

(A) Structures of compounds 51 and 52. (B) MD simulations of 52 complexed with Se.TRPM_PZQ. The top pose from each independent run is overlaid (green) and compared to the pose of (R)-PZQ in Eg.TRPM_PZQ (pink). (C) The most common binding pose of 52 in Se.TRPM_PZQ, resulting from the clustering of the trajectories in B. The thiocarbonyl is predicted to interact with H1239, unlike the carbonyl-containing molecule. (D) Functional profiling of thiocarbonyl-containing analogs at Se.TRPM_PZQ. The concentration-response curves demonstrate that the thiocarbonyl increases the pharmacological efficacy of these compounds at Se.TRPM_PZQ when compared to the oxygen-containing counterpart.

The activity of these new thioamides, and their oxygen-retaining (e.g. amide) counterparts, were then functionally profiled at Se.TRPM_PZQ. Both 52 (EC₅₀ = 0.38 ± 0.08 μM) and 51 (EC₅₀ = 14.1 ± 2.9 μM) were active at Se.TRPM_PZQ, and the efficacy of both analogs was greater than (R)-PZQ (B_max of 52 and 51 were 154% and 192% of (R)-PZQ, respectively (Figure 4D)). The oxygen-containing analogs retained the lower efficacy seen with PZQ (Figure 4D). Thus, while not having a consistent effect on potency, both thioamides had a much higher efficacy (measured as B_max) compared with their amide congeners.

If improved interactions of the thiocarbonyl moiety with the histidine in S1 of Se.TRPM_PZQ underpins the increased efficacy of these analogs, then this improvement should depend on the presence of the S1 histidine residue. This was evaluated by functionally profiling 51 and 52 at a Sm.TRPM_PZQ[D1677E] mutant. This construct lacks the cestode-specific S1 histidine, possessing an asparagine residue at the equivalent position. Sm.TRPM_PZQ[D1677E] is unable to be activated by PZQ (Figure S2A) (9), and no enhancement of activity was seen with either 51 (Figure S2B) or 52 (Figure S2C). Therefore, restoration of activity depends upon the key S1 histidine residue, which likely permits exploitation of thioamide electrostatics to deliver the improved efficacy of compounds 51 and 52 at pseudophyllidean cestode TRPM_PZQ.

Finally, the activity of these new molecules on cestodes was examined. Since there is no established ex vivo model for drug screening on pseudophyllidean cestodes, 51 and 52 were profiled on cyclophyllidean (E. multilocularis) protoscoleces (23) as proof-of-principle for biological activity. Unlike DMSO-treated controls (Figure 5A), both 51 (Figure 5B) and 52 (Figure 5C) caused sustained contraction and blebbing of the tegument, phenocopying PZQ (Figure 5D). The EC₅₀ for 51 and 52 were 29 ± 6 nM, and 4.3 ± 0.5 nM respectively, compared with 7.7 ± 1.3 nM for (R)-PZQ (Figure 5E). These data parallel the observed activity trend, and rank order, seen with these compounds at Se.TRPM_PZQ (compare with Figure 4D). Thus, the new compounds were active on the Se.TRPM_PZQ channel in vitro and ex vivo on protoscoleces (E. multilocularis).

Figure 5. Biological activity of new praziquantel derivatives.

E. multilocularis protoscoleces after treatment with (A) DMSO, (B) 51 (1 μM), (C) 52 (1 μM), or (D) (R)-PZQ (1 μM). (E) Concentration-response motility graph of E. multilocularis protoscoleces after treatment with (R)-PZQ (blue), 51 (pink), or 52 (green).

Conclusions.

This study investigated the challenge of treating pseudophyllidean cestode infections, clinically recalcitrant to therapy with PZQ. Capitalizing on the discovery of TRPM_PZQ and realization that cestode TRPM_PZQ binding pockets display a broader analog binding profile we profiled PZQ analogs to identify derivatives that target cyclophyllidean (metabolically resistant analogs) and pseudophyllidean TRPM_PZQ (thioamide derivatives 51 and 52). Target-based design at TRPM_PZQ therefore provides new opportunity to tailor treatments for specific infections caused by parasitic flatworms.

Synthetic Chemistry.

Materials used for chemical synthesis, synthetic procedures, and small molecule characterization data are presented in the supplementary information.

Cell Culture and transfection.

Cell culture and transfection were performed as previously described.^9,15 Briefly, activation of the three TRPM_PZQ orthologs (Sm.TRPM_PZQ, Eg.TRPM_PZQ and Mc.TRPM_PZQ) was measured using a Ca²⁺ reporter assay, following transient transfection of codon-optimized constructs (Genscript) into HEK293 cells. The HEK293 cell line (ATCC CRL-1573.3) was authenticated by STR profiling (ATCC), and cells were evaluated for mycoplasma contamination by monthly scheduled testing (LookOut^® Mycoplasma PCR Detection Kit, Sigma). HEK293 cells were cultured at 37 °C in a humidified incubator in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), penicillin (100 units/mL), streptomycin (100 μg/mL), and L-glutamine (290 μg/mL). Cells were transfected using Lipofectamine-2000 at a density of 3x10⁶ cells per petri dish (100 mm diameter) 24 h prior to replating into 384 well plates for functional assays.

Ca²⁺ reporter assays.

Functional profiling of Schistosoma mansoni TRPM_PZQ (Sm.TRPM_PZQ), Echinococcus granulosus TRPM_PZQ (Eg.TRPM_PZQ) and Mesocestoides corti TRPM_PZQ (Mc.TRPM_PZQ) in response to PZQ analogs was performed using a Ca²⁺ reporter assay in HEK293 cells as previously described.^9,15 Ca²⁺ imaging assays were performed using a Fluorescence Imaging Plate Reader (FLIPR^TETRA, Molecular Devices). HEK293 cells (naïve or transfected) were seeded (20,000 cells/well) in a black-walled clear-bottomed poly-d-lysine coated 384-well plate (Greiner Bio-One) in DMEM growth media supplemented with 10% FBS. After 24 h, growth medium was removed, and cells were loaded with a fluorescent Ca²⁺ indicator (Fluo-4 NW dye, Invitrogen) by incubating (20 μL per well, 1 h at 37°C) in Hanks' balanced salt solution (HBSS) assay buffer containing probenecid (2.5 mM) and HEPES (20 mM). Drug dilutions were prepared in assay buffer, without probenecid and dye, in ‘V’-shape 384-well plates (Greiner Bio-one, Germany). After indicator loading, the Ca²⁺ assay was performed at room temperature. Basal fluorescence was monitored for 20 s, then 5 μL of each drug added, and the signal (raw fluorescence units) was monitored over an additional 250 s. Changes in fluorescence were calculated by subtracting the average basal fluorescence (averaged values between 0 to 20 s) from the maximum fluorescence value. These values were normalized to the response of (±)-PZQ at the corresponding TRPM_PZQ channel and plotted. Concentration-response curves were generated in GraphPad Prism (v. 9.5.1) using the default three-parameter curve fitting and constraining the bottom of the curve to 0%. The reported EC₅₀ values represent the mean ± SEM of n ≥ 3 independent transfections.

Computational Modeling.

The Schrodinger Computational Modeling Suite, in conjunction with the Maestro GUI were utilized for all computational modeling. The binding pose of (R)-PZQ in Sm.TRPM_PZQ was previously reported.^8,9 Homology models of Eg.TRPM_PZQ and Se.TRPM_PZQ were created as described in ref. 8 for Sm.TRPM_PZQ.

To prepare ligands for modeling, the required ligand was drawn in ChemDraw Professional, imported into the Maestro GUI, and prepared using the LigPrep tool with default settings in the OPLS4 force field at pH = 7.4. The output structure was used for subsequent modeling.

Unbiased MD simulations were performed using Desmond within the Schrodinger Computational Suite using default settings as previously described.⁹ Eight independent runs of 500 ns were completed. The simulations were run in the NPγT ensemble using both the Langevin thermostat (300 K) and semiisotropic barostat (1.01325 bar). The system was relaxed before simulation and brought to temperature per the default series of Desmond simulations. Each simulation began from a random seed, the velocities were randomized, and frames were recorded at an interval of 50 ps, allowing for the collection of 10,000 frames in each simulation.

Each of the eight runs for each channel were independently clustered using the ‘Trajectory Clustering Tool’ within the Maestro GUI. The poses of the most populated cluster from each run were then superimposed to provide, e.g., Figure 3D or Figure 4C.

Cestode motility assays.

Protoscoleces of E. multilocularis strain MB17 were isolated from metacestode material grown in experimentally infected gerbils (kindly provided by Prof. Dr. K. Brehm, University of Würzburg, Germany). Extraction and activation of protoscoleces, as well as the subsequent motility assays were performed as described previously.^15,23 For isolation of cestode material, all animals were treated in compliance with the Swiss Federal Protection of Animals Act (TSchV, SR455), and experiments were approved by the Animal Welfare Committee of the Canton of Bern under license numbers BE30/19 and BE2/2022. Protoscoleces were activated by 10% DMSO (3 h at 37°C in a humid atmosphere with 5% CO₂) and left to recover overnight in DMEM (Biochrom, Berlin, Germany) supplemented with 10% FBS (Biochrom) at 37°C. The following day, 25 protoscoleces were distributed into individual wells of a white, flat-bottom 384 well plate (Huberlab, Aesch, Switzerland) for drug screening. Each assay plate was sealed with a clear view seal foil (Huberlab), and motility was measured in a live cell imaging system (Nikon TE2000E, Hamatsu ORCA ER camera) with the software NIS-Elements AR V4.51 and the JOBS module (Nikon) at 37°C. Motility of protoscoleces was calculated relative to the DMSO control at individual time points with the highest and lowest values were excluded. Results are shown as mean ± SEM.

Supplementary Material

The Supporting Information is available free of charge on the ACS Publications website.

Table S1 (compounds referenced in Figure 1), Figure S1 (epsiprantel concentration-response curves), Figure S2 (concentration-response curves on Sm.TRPM_PZQ), synthetic methods, and small molecule characterization data (PDF)

ABBREVIATIONS

TRPM_PZQ

transient receptor potential melastatin praziquantel

PZQ

praziquantel

WHO

World Health Organization

TRP

transient receptor potential

Sm.TRPM_PZQ

Schistosoma mansoni transient receptor potential melastatin praziquantel

HTS

high-throughput screening

VSLD

voltage sensor-like domain

Eg.TRPM_PZQ

Echinococcus granulosus transient receptor potential melastatin praziquantel

EC₅₀

half-maximal effective concentration

Mc.TRPM_PZQ

Mesocestoides corti transient receptor potential melastatin praziquantel

Se.TRPM_PZQ

Spirometra erinaceieuropaei transient receptor potential melastatin praziquantel

MD

molecular dynamics

E. multilocularis

Echinococcus multilocularis.