Target-based discovery of a broad spectrum flukicide

Abstract

Diseases caused by parasitic flatworms impart a considerable healthcare burden worldwide. Many of these diseases - for example, the parasitic blood fluke infection, schistosomiasis - are treated with the drug praziquantel (PZQ). However, PZQ is ineffective against disease caused by liver flukes from the genus Fasciola. This is due to a single amino acid change within the target of PZQ, a transient receptor potential ion channel (TRPM_PZQ), in Fasciola species. Here we identify benzamidoquinazolinone analogs that are active against Fasciola TRPM_PZQ. Structure-activity studies define an optimized ligand (BZQ) that caused protracted paralysis and damage to the protective tegument of these liver flukes. BZQ also retained activity against Schistosoma mansoni comparable to PZQ and was active against TRPM_PZQ orthologs in all profiled species of parasitic fluke. This broad spectrum activity was manifest as BZQ adopts a pose within the binding pocket of TRPM_PZQ dependent on a ubiquitously conserved residue. BZQ therefore acts as a universal activator of trematode TRPM_PZQ and a first-in-class, broad spectrum flukicide.

Keywords: Ca2+ signaling; parasite; praziquantel; transient receptor potential channel.

Figure 1.. Functional profiling of TRPM_PZQ orthologs.

Representative Ca²⁺ flux traces depicting the effect of (A, B) (±)-PZQ or (D, E) compound 1 in HEK293 cells stably expressing (A,D) Sm.TRPM_PZQ or (B,E) Fh.TRPM_PZQ. Cells were treated with increasing concentrations (0-100 μM) of each drug added after ~20 s of sampling the baseline fluorescence emission. (C&F) Concentration-response curves resulting from activation of Sm.TRPM_PZQ (blue circles) or Fh.TRPM_PZQ (red circles) by (C) (±)-PZQ or (F) compound 1. Control responses in HEK293 cells lacking TRPM_PZQ are shown (grey diamonds). (G) Schematic of modifiable regions on compound 1. Three regions on the N-benzamidoquinazolinone core were targeted for modification. The Northern Hemisphere (green), Southern Hemisphere (orange), and the aromatic core (pink). (H) Chemical structure of the optimized benzamidoquinazolinone, BZQ, after SAR studies. (I) Concentration-response curves for BZQ in HEK293 cells stably expressing Sm.TRPM_PZQ (blue circles) or Fh.TRPM_PZQ (red circles), compared to control responses (grey diamonds). (J) Concentration-response curves for BZQ in cells transiently expressing various TRPM_PZQ orthologs. These were: Schistosoma mansoni (Sm.TRPM_PZQ, closed blue circles), Fasciola hepatica (Fh.TRPM_PZQ, closed red circles), Schistosoma haematobium (Sh.TRPM_PZQ, open purple squares), Schistosoma japonicum (Sj.TRPM_PZQ, open green triangles), Fasciola gigantica (Fg.TRPM_PZQ, open gold hexagons), Echinostoma caproni (Ec.TRPM_PZQ, open grey diamonds), Clonorchis sinensis (Cs.TRPM_PZQ), and Opisthorchis viverrini (Ov.TRPM_PZQ, open black diamonds). Concentration-response curves were normalized to the maximum response at each channel and represent the mean ± SE of n ≥ 3 independent experiments, each comprised of technical duplicates.

Figure 2.. Electrophysiological analysis of BZQ action.

(A) Whole-cell current (pA) versus time (s) plot of Fh.TRPM_PZQ expressing HEK293 cell perfused with different concentrations of compound 1 (1 nM to 100 μM) prior to addition of LaCl₃ (1 nM to 10 mM). Extracellular solution: 140 mM NaCl, 5 mM glucose, 4 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, pH 7.4 with NaOH. Intracellular solution: 130 mM CsF, 10 mM CsCl, 10 mM NaCl, 10 mM EGTA, 10 mM HEPES, pH 7.4 with NaOH. (B) Concentration response curves for BZQ and 1 from experiments such as shown in (A) recorded from Fh.TRPM_PZQ (red) or Sm.TRPM_PZQ (blue) expressing HEK293 cells. Data are shown as mean ± SE, n≥6. (C) Representative cell-attached recordings from Fh.TRPM_PZQ or Sm.TRPM_PZQ expressing HEK293 cells in the presence of (±)-PZQ (10 μM) or BZQ (10 μM) in the bath solution. Bath solution: 145 mM NaCl, 10 mM HEPES, 1 mM EGTA, pH 7.4. Pipette solution: 145 mM NaCl, 10 mM HEPES, 1 mM EGTA, pH 7.4. c, closed state. Holding voltage, 60mV. (D) Current-voltage plot for Sm.TRPM_PZQ (blue) activated by BZQ (closed circle, 10 μM) or (±)-PZQ (open circle, 10 μM)) and Fh.TRPM_PZQ activated by BZQ (red, 10 μM). Data are shown as mean ± SE, n≥3. (E) Single channel open probability (P_open) of Fh.TRPM_PZQ (red) or Sm.TRPM_PZQ (blue) activated by BZQ or (±)-PZQ (each at 10 μM) in the bath solution. Data are shown as mean ± SE, n≥6.

Figure 3.. Effects of BZQ and (±)-PZQ on parasitic flukes.

(A&B) Exposure of S. mansoni and F. hepatica to (±)-PZQ or BZQ compared with DMSO (1-1.25%, control). A rapid contraction of schistosomes to (±)-PZQ (0.5 μM) or BZQ (0.5 μM) was apparent. BZQ (6.25 μM), but not (±)-PZQ (50 μM), caused spastic paralysis of adult liver flukes. (C-H) Studies of the ultrastructure of BZQ-treated flukes. Transmission electron microscopy of drug-induced damage to S. mansoni tegument (C) without treatment or (D) after treatment with BZQ (1 μM). Scanning electron microscopy of drug-induced damage to immature F. hepatica tegument after treatment with: (E&F) DMSO (1.25%, control) or (G&H) BZQ (6.25 μM, 24 h exposure). BZQ caused blebs to occur on the fluke surface (arrows). (I-K) Motility of (I) adult, (J) triclabendazole (TCBZ)-sensitive immature, and (K) TCBZ-resistant immature F. hepatica after treatment with BZQ (blue triangles) or triclabendazole (black/green circles) compared with application of DMSO (1.25%, control, grey squares). Motility scores are reported as the mean ± SE of n ≥ 3 independent experiments. (L) Dose-response curve for motility of adult (triangle) and immature (circle) F. hepatica treated with BZQ. (M) BZQ activity in a murine model of schistosomiasis. Mice, infected with schistosomes, were treated at 7 weeks post-infection with either BZQ or (±)-PZQ. Mice were dosed once daily with each drug for three sequential days (50 mg/kg, intraperitoneally), and worm burden was evaluated on the fourth day after initiation of treatment. Worm burden was reduced by treatment with BZQ or (±)-PZQ compared with either untreated or vehicle-treated mice as described in the methods. N = 13 mice per group; data are shown as mean ± SD and analyzed using the Mann-Whitney test. **** = p ≤ 0.0001, ns = not significant. Scale bars for (A) = 250 μm, (B) = 1 mm; (C&D) = 1 μM, (E&G) = 500 μm, (F&H) = 10 μm.

Figure 4.. TRPM_PZQ engagement by BZQ.

In silico binding pose for (A&B) (R)-PZQ and (C&D) BZQ in Sm.TRPM_PZQ. Concentration-response curves for (E) (±)-PZQ and (F) BZQ in specified Sm.TRPM_PZQ mutants. WT = blue circles, Sm.TRPM[N1388]A_PZQ = orange circles, Sm.TRPM[T1389A]_PZQ = green circles, Sm.TRPM[R1514A]_PZQ = open purple circles, Sm.TRPM[Y1678A]_PZQ = open pink circles. Data are presented as mean ± SEM of biological triplicates performed in technical duplicate.