Screening and Identification of Metacaspase Inhibitors: Evaluation of Inhibition Mechanism and Trypanocidal Activity

Abstract

A common strategy to identify new antiparasitic agents is the targeting of proteases, due to their essential contributions to parasite growth and development. Metacaspases (MCAs) are cysteine proteases present in fungi, protozoa, and plants. These enzymes, which are associated with crucial cellular events in trypanosomes, are absent in the human host, thus arising as attractive drug targets. To find new MCA inhibitors with trypanocidal activity, we adapted a continuous fluorescence enzymatic assay to a medium-throughput format and carried out screening of different compound collections, followed by the construction of dose-response curves for the most promising hits. We used MCA5 from Trypanosoma brucei (TbMCA5) as a model for the identification of inhibitors from the GlaxoSmithKline HAT and CHAGAS chemical boxes. We also assessed a third collection of nine compounds from the Maybridge database that had been identified by virtual screening as potential inhibitors of the cysteine peptidase falcipain-2 (clan CA) from Plasmodium falciparum Compound HTS01959 (from the Maybridge collection) was the most potent inhibitor, with a 50% inhibitory concentration (IC₅₀) of 14.39 µM; it also inhibited other MCAs from T. brucei and Trypanosoma cruzi (TbMCA2, 4.14 µM; TbMCA3, 5.04 µM; TcMCA5, 151 µM). HTS01959 behaved as a reversible, slow-binding, and noncompetitive inhibitor of TbMCA2, with a mechanism of action that included redox components. Importantly, HTS01959 displayed trypanocidal activity against bloodstream forms of T. brucei and trypomastigote forms of T. cruzi, without cytotoxic effects on Vero cells. Thus, HTS01959 is a promising starting point to develop more specific and potent chemical structures to target MCAs.

Keywords: Chagas disease; antiparasitic agent; inhibitors; metacaspases; sleeping sickness; target-based screening.

FIG 1

Continuous fluorogenic assay for recombinant TbMCA5. (A) Kinetic progression curves for different TbMCA5 concentrations at a fixed dose (75 µM) of Z-VRPR-AMC. (B) Selwyn test for different TbMCA5 concentrations. Data from different enzyme concentrations (represented by different symbols in the graph) are well fitted by a single curve. (C) Curve of V₀ versus [TbMCA5]₀. (D) Michaelis-Menten plot. In all cases, data corresponding to a TbMCA5 concentration of 103 nM with a Z-VRPR-AMC concentration of 75 µM, conditions selected for compound screening, are indicated in red. In panels C and D, V₀ is defined as the slope (dF/dt) of the linear region of progress (fluorescence versus time) curves.

FIG 2

Activity plot for the assayed compounds during primary screening against TbMCA5. The solid red line shows the average of enzyme activity controls (C+), and the dashed green line represents the cutoff value for selection of hits, which is 70% residual activity (equivalent to 30% inhibition). Black open circles represent enzyme controls, black closed circles substrate controls, and orange open circles inhibition controls (Z-VRPR-FMK). Blue and red inverted triangles represent inactive compounds and hits, respectively. Highly autofluorescent compounds (17) were discarded from further analysis due to the negative impact they have on reproducibility (i.e., they are able to interfere significantly with fluorescence assay readouts).

FIG 3

Dose-response curves and structures of the identified TbMCA5 inhibitors. (A) Dose-response curves. For each compound, the solid line represents the best fit of the four-parameter Hill equation to experimental data (open symbols). The best fit for the irreversible inhibitor Z-VRPR-FMK is represented as a gray dashed line. Concentrations of inhibitors (logarithmic x axis) are molar. (B) Structures of identified TbMCA5 inhibitors.

FIG 4

Dose-response curves for the inhibition of different MCAs by HTS01959. (A) Dose-response curves for T. brucei MCAs. (B) Dose-response curves for MCAs from other organisms. For each curve, the solid line represents the best fit of the four-parameter Hill equation to experimental data (open symbols). (C) Summary of results of dose-response assays with different MCAs incubated with HTS01959. HTS01959 concentrations are expressed as molar. ND, not determined.

FIG 5

Reversibility and time dependence of the inhibition of TbMCA2 by HTS01959. (A) Product progress curves for dissociation of the enzyme-inhibitor (E-I) complex by rapid dilution (100-fold) of the enzyme-inhibitor mixture into the substrate solution. (B) Product progress curves for formation of the E-I complex by rapid addition of the enzyme to a substrate-inhibitor mixture. (C) Dose-response curves for HTS01959 at increasing substrate concentrations. (D) Effect of substrate concentration on IC₅₀ values. HTS01959 concentrations are expressed as molar.

FIG 6

Effects of the strength and concentration of reducing agents on the inhibitory activity of HTS01959. (A) Dose-response curves for the inhibition of TbMCA2 by HTS01959 at increasing concentrations of DTT. (B) Dose-response curves in the presence of strong (DTT) and weak (β-mercaptoethanol and cysteine) reducing agents at identical concentrations of 10 mM. For each curve, the solid line represents the best fit of the four-parameter Hill equation to experimental data (open symbols). β-ME, β-mercaptoethanol. (C) Summary of results from panels A and B. N. I., no inhibition. HTS01959 concentrations are expressed as molar.

FIG 7

Specific hydrogen peroxide generation by HTS01959 in the presence of low millimolar DTT concentrations. ((A) Enzymatic quantification of H₂O₂ generated at different conditions by HRPO-catalyzed oxidation of 4-aminophenazone. This reaction produces a colored product with strong absorbance at 505 nm. C28 [2-hydroxy-3-(1-propenyl)-1,4-naphthoquinone] was used as a positive control as a quinoid redox-cycling compound. The statistical significance was evaluated by one-way analysis of variance and Tukey's multiple-comparison posttest. ***, P < 0.001. (B) Dose-response curves for the inhibition of TbMCA2 by HTS01959 in the presence (200 µg/ml) and absence of catalase. For each curve, the dashed line represents the best fit of the four-parameter Hill equation to experimental data (open symbols). HTS01959 concentrations are expressed as molar.

FIG 8

Activity of HTS01959 on cultured T. brucei bloodstream form and Vero cells. (A) Dose-response curves for T. brucei bloodstream forms treated with HTS01959 or nifurtimox as a control. Viability was determined in triplicate using the resazurin method. Averages ± standard deviations (SDs) are shown. For each curve, the dashed line represents the best fit of the four-parameter Hill equation to experimental data (open symbols). (B) Drug efficacy against T. cruzi intracellular amastigotes. The number of infected cells (red bars) and the number of amastigotes (Amas) per cell (blue bars) (average ± SD) were determined by DAPI staining after 2 days of treatment with 130 µM HTS01959, 4 µM benznidazole as a reference inhibitor, or 0.5% (vol/vol) DMSO as a control. The statistical significance was evaluated by one-way analysis of variance and Tukey's multiple-comparison posttest. ***, P < 0.001. (C) Dose-response curve for T. cruzi trypomastigotes with HTS01959. Viability was determined in triplicate using the resazurin method. Averages ± SDs are shown. The dashed line represents the best fit of the four-parameter Hill equation to experimental data (open symbols). (D) Cytotoxicity assay on Vero cells treated with HTS01959 and DMSO (0.5% [vol/vol]) as a growth control. Viability was determined in triplicate using a luminescence assay. Averages ± SDs are shown. In all cases, HTS01959 concentrations are expressed as molar.