The Anthelmintic Activity of Praziquantel Analogs Correlates with Structure–Activity Relationships at TRPMPZQ Orthologs

The anthelmintic drug praziquantel remains a key clinical therapy for treating various diseases caused by parasitic flatworms. The parasite target of praziquantel has remained undefined despite longstanding usage in the clinic, although a candidate ion channel target, named TRPMPZQ, has recently been identified. Intriguingly, certain praziquantel derivatives show different activities against different parasites: for example, some praziquantel analogs are considerably more active against cestodes than against schistosomes. Here we interrogate whether the different activities of praziquantel analogs against different parasites are also reflected by unique structure–activity relationships at the TRPMPZQ channels found in these different organisms. To do this, several praziquantel analogs were synthesized and functionally profiled against schistosome and cestode TRPMPZQ channels. Data demonstrate that structure–activity relationships are closely mirrored between parasites and their TRPMPZQ orthologs, providing further support for TRPMPZQ as the therapeutically relevant target of praziquantel.

E xactly 40 years ago, a highly influential review on the anthelmintic activity of praziquantel (PZQ) was published by Peter Andrews and Herbert Thomas (both at Bayer AG) and Rolf Pohlke and Jurgen Seubert (both at E. Merck KG). 1 That work detailed the discovery of the anthelmintic activity of PZQ, derivatization of the scaffold, the drug's pharmacokinetic and safety profile, and the broad efficacy of this new therapeutic agent against a range of parasitic flatworms.The summary of data interrogating the activity of different pyrazino[2,1-a]isoquinoline derivatives against a representative trematode (Schistosoma mansoni) and cestode model (Hymenolepis nana) established both the "tightness" of the pharmacophore that underpins the efficacy of PZQ and a "structure−activity" fingerprint for the action of this drug that has long served as a reference standard for the field.
Following decades of clinical usage of PZQ for treatment of various diseases caused by parasitic flatworms, 2−5 a candidate target was recently identified 6 in Schistosoma mansoni.This target is a transient receptor potential (TRP) ion channel of the melastatin subfamily, named Sm.TRPM PZQ , 6−8 that mirrors the structure−activity relationship (SAR) of PZQ derivatives 7 as described by Andrews et al. 1 TRPM PZQ is a large nonselective cation channel unique to flatworms. 6,8−10 Activation of Sm.TRPM PZQ is thought to elicit excitotoxicity through membrane depolarization, spastic contraction, and surface damage to the parasite, which then catalyzes immunological clearance from infected hosts. 11Additional evidence from genetic association studies, 12 pharmacological screening, 13 and functional profiling of TRPM PZQ orthologs 9,10 add support for TRPM PZQ serving as the therapeutically relevant parasite target of PZQ.

legend).
While there are caveats in the interpretation of these data, it is evident that the selected analogs displayed appreciable activity in the cestode model (column 4) but lower or minimal activity when tested against schistosomes (columns 2 and 3).In contrast, (±)-PZQ (1) showed equivalent activity in phenotypic grading across all of the bioassays, with preferential stereoselectivity toward the R enantiomer (Table 1).
Finally, we profiled modifications of the cyclohexyl group of PZQ.Compound 7, a 4′-aminocyclohexyl derivative, lacked activity at Sm.TRPM PZQ and was only weakly active at cestode TRPM PZQ orthologs (Figure 1I), corresponding to the weak in vivo activity against H. nana previously reported (Table 1). 1 Finally, cyclopropyl analog 8 activated both cestode TRPM PZQ representatives at concentrations >10 μM (EC 50 = 53 ± 16 μM for Eg.TRPM PZQ , EC 50 = 65 ± 14 μM for Mc.TRPM PZQ ; Figure 1J).Little activity was observed at Sm.TRPM PZQ .These target-based data are again consistent with the phenotypic observations of Andrews et al. (Table 1). 1 Overall, from the profiled PZQ analogs, only a single analog (compound 6) was sufficiently active at Sm.TRPM PZQ to derive an EC 50 value, while five analogs displayed activity at the cestode channels (Table 1).Analog 6 was not one of the eight PZQ analogs selected based on the differential potency between cestodes and schistosomes but was synthesized to explain the in vivo activity of the other analogs.Therefore, the different potencies of these analogs against cestodes and schistosomes, seen in phenotypic the data of Andrews et al. 1 40 years ago, was mirrored by the same differential potency of these analogs in target-based assays at the different TRPM PZQ channels.Some caveats are however appropriate.First, the original data did not report the activity of the PZQ derivatives against cestodes ex vivo (in vitro), so in selecting these analogs, there was no direct comparator for the action of all of these analogs between schistosomes and cestodes.Therefore, two analogs (R)-2 and (S)-2 were tested on Echinococcus multilocularis protoscoleces for comparison with PZQ (Figure 2A−D).When compared with the vehicle control, treatment with (±)-PZQ (1 μM) caused a sustained contraction of the protoscoleces (Figure 2A,B).When protoscoleces were treated with the 3-pyridyl enantiomers (R)-2 (Figure 2C) and (S)-2 (Figure 2D), a similar contraction was observed, with (R)-2 being more potent than (S)-2.To determine IC 50 values, concentration−response curves were obtained at multiple time points (Figure 2E−G).Low concentrations of PZQ and the 3-pyridyl analogs stimulated motility at the early time points (Figure 2E).After 24 h, the IC 50 for (R)-2 was ∼3 μM, and the IC 50 for (S)-2 was ∼30 μM (Figure 2G).Prolonged treatment with (±)-PZQ proved toxic after 24 h, and therefore, a more realistic IC 50 value (∼100 nM) was calculated at 12 h postincubation (Figure 2F).This mirrors the EC 50 at a cestode TRPM PZQ of 100 nM (Table 1).These data for activity against cestodes ex vivo are again consistent with the potencies of the molecules at cestode TRPM PZQ (Table 1).
Second, the cestode TRPM PZQ and motility assays derive from different cyclophyllidean cestodes (E.granulosus, M. corti, and E. multilocularis) than the model (H.nana) used by Andrews et al. 1 However, we note that the amino acid residues lining the PZQ binding pocket of TRPM PZQ are identical across all cyclophyllidean cestode TRPM PZQ orthologs examined to date, including H. nana TRPM PZQ (Figure S1). 10 This is consistent with the similarity of the functional data from Eg.TRPM PZQ and Mc.TRPM PZQ .
Considering the structures of the analogs profiled here, it is evident that the cestode TRPM PZQ binding pocket is more tolerant to substitutions of the cyclohexyl moiety of PZQ than is the schistosome TRPM PZQ binding pocket.Aniline analog 6 and pyridyl analog (R)-2 show submicromolar potency, and even the cyclopropyl analog 8, which displayed no activity at Sm.TRPM PZQ at 100 μM, was clearly active at the cestode TRPM PZQ orthologs, consistent with the differential activity seen by Andrews et al. (Table 1). 1 This increased tolerability to modifications of the cyclohexyl group of PZQ, a key part of the pharmacophore at Sm.TRPM PZQ , 7 may provide opportunity to accommodate other cyclohexane ring modifications�notably, more metabolically stable PZQ derivatives�within the cestode TRPM PZQ binding pocket. 15,16This could potentially enhance the in vivo efficacy of these analogs for treating cestode infections, and these data therefore highlight an opportunity to design drugs that selectively target cestode TRPM PZQ .Whether the absolute potency of such analogs can be further improved over PZQ to yield better treatments for cestode species less sensitive to PZQ (e.g., noncyclophyllidean cestodes 10,17 ) or for cestode life cycle stages that are hard to treat will require further work and a better understanding of the molecular basis by which cestode-selective analogs engage the TRPM PZQ binding pocket.Such understanding will be aided by the recent mapping of the PZQ binding pocket in TRPM PZQ orthologs in different parasitic flatworms and a capacity to model these interactions. 7,10Of likely relevance are two natural amino acid variants�a histidine residue in the S1 transmembrane helix and a serine residue in the S4/S5 linker�that are different between the PZQ binding pocket of trematode and cyclophyllidean cestode TRPM PZQ (compare Figure 3A and Figure 3B). 10This natural variation within the TRPM PZQ binding pocket provides a possible molecular explanation underpinning the differential SAR of the PZQ analogs.Natural variation in the binding pocket has previously been shown to render Fasciola spp.TRPM PZQ insensitive to PZQ. 7,10 The activity of (R)-2 at cestode TRPM PZQ versus Sm.TRPM PZQ is noteworthy in the context of this natural variation.The transmembrane helix 1 (S1) variation occurs at a position in close proximity to the pyridyl nitrogen (Figure 3B), and it is conceivable that there is an electrostatic interaction between the histidine residue in cestode TRPM PZQ (e.g., H1231 in Eg.TRPM PZQ ) and the pyridine ring that is absent with the uncharged asparagine residue in schistosome TRPM PZQ (e.g., N1388 in Sm.TRPM PZQ ).Interactions between PZQ and this S1 residue are important for PZQ activation of TRPM PZQ across species. 7,10n summary, functional profiling of various PZQ derivatives on parasitic flatworms and at their respective TRPM PZQ orthologs shows that "the glove fits": the SAR between different parasites and different parasite TRPM PZQ orthologs matches well.These data provide additional support for TRPM PZQ serving as the relevant therapeutic target of PZQ in parasitic flatworms.
Columns 5−7 tabulate the EC 50 values for analog activation of trematode and cestode TRPM PZQ orthologs in vitro.Data are shown as mean ± SEM for n ≥ 3 independent transfections.Eg.TRPM PZQ = Echinococcus granulosus TRPM PZQ ; Mc.TRPM PZQ = Mesocestoides corti TRPM PZQ .b EC 50 values on the channel are reported in ref 10 and are provided for completeness.

Table 1 .
that scored the activity of PZQ derivatives against Schistosoma mansoni and Hymenolepis nana.Data were collected against S. mansoni in vitro and in vivo using a Chart 1. Structures of the PZQ Analogs Studied in This Work Comparison of Antischistosomal and Anticestodal Activities of PZQ Derivatives with Their Activities at the Corresponding TRPM PZQ Orthologs a a