Discovery of Membrane-Bound Pyrophosphatase Inhibitors Derived from an Isoxazole Fragment

Membrane-bound pyrophosphatases (mPPases) regulate energy homeostasis in pathogenic protozoan parasites and lack human homologues, which makes them promising targets in e.g. malaria. Yet only few nonphosphorus inhibitors have been reported so far. Here, we explore an isoxazole fragment hit, leading to the discovery of small mPPase inhibitors with 6–10 μM IC50 values in the Thermotoga maritima test system. Promisingly, the compounds retained activity against Plasmodium falciparum mPPase in membranes and inhibited parasite growth.

Ambinter LE: 0.20 Purity: 87% a LE computed using the pIC50 and the "Heavy Atom Count" (LEHA) normalization method with Discovery Studio. S16 Aggregation data for 5, 12k and 12l ( Figure S1) Potential colloidal aggregation in the TmPPase model system was studied at five concentrations (50 µM, 25 µM, 10 µM, 5 µM, and 1 µM). The original assay conditions were simulated, and the light scattering of potential aggregates in the mixture was studied by nephelometric methods using Nepheloskan Ascent ® (Labsystems, Finland). Aggregation of the blank and compounds 5, 12k and 12l were measured as triplicates from one independent experiment at three different voltages (300 V, 400 V and 500 V) at room temperature. For compound 5 there was no detectable aggregation at all. Compounds 12k and 12l showed aggregate formation at both 50 µM and 25 µM. However, at 10 µM, 5 µM and 1 µM concentrations the obtained values were close to those observed by the blank, thereby indicating only minor/no detectable aggregation.  Figure S1. Aggregation data for compounds 5, 12k and 12l. The data, measured in relative nephelometric units (RNU), is shown as the mean ± SD and normalized to the blank. S17

IC50 plots from the inhibition assay
The blue and black curves refer to compounds presented in the main article and only shown in the Supporting Information, respectively. IC50, half maximal inhibitory concentration; CI95%, half maximal inhibitory concentration expressed as a 95% confidence interval (given in square brackets); NA, IC50 not assigned.

Computational methods
Commercial analogues were retrieved by screening the ZINC12 database 1 (clean drug-like subset; downloaded on 27.04.2016) using a KNIME 2 workflow connected to the RDKit 3 nodes. The SAR analyses and docking studies (data not shown) underlying this work were conducted using the Schrödinger molecular modeling platform (Canvas and Glide modules, respectively) 4 .

Chemistry
Compounds 1-7, 13-20 and 24 were obtained from commercial suppliers. Supporting Information, Table S1 summarizes the compounds tested. Compound purity was >95% as determined by LC-MS and characterized by HRMS. For all commercial compounds after the initial screen, purity was tested independently from the vendor specification (a few compounds of lower purity are reported only in Supporting Information). To evaluate potential aggregation of samples in the TmPPase based assay, aggregate formation in the assay conditions was studied at five concentrations (50 μM, 25 μM, 10 μM, 5 μM, and 1 μM) for compounds 5, 12k and 12l ( Figure S1).
All chemicals used were commercially available. Reactions using anhydrous conditions were conducted in oven-dried (130 °C, >24 h) glassware, purged with argon prior to use. Microwave reactions were performed using a Biotage Initiator + instrument (Uppsala, Sweden). The progress of the reactions was monitored by thin-layer chromatography on silica gel 60-F254 aluminum plates and visualized by using a dual short/long wave (254/366 nm) UV lamp. The combined organic solutions from extractions were dried over anhydrous Na2SO4, filtered and concentrated with a rotary evaporator at reduced pressure. Flash SiO2 column chromatography was performed with automated high performance flash chromatography, Biotage Isolera ™ Spektra Systems with ACI ™ and Assist (ISO-1SW Isolera One) equipped with a variable UV-VIS (200-800 nm) photodiode array (Uppsala, Sweden) using SNAP KP-SIL/Ultra 10 g or 25 g cartridges and the indicated mobile phase gradient. The reactions were not optimized and all the yields are given for purified products.

Biological activities
The inhibition assay was measured in 96 well plate assay as previously reported. 5,6 Briefly, 1 µL of purified TmPPase (13 mg/mL) was reactivated in 99 µL of reactivation buffer (20 mM 2-(Nmorpholino) ethanesulfonic acid (MES) pH 6.5, 3.5% (v/v) glycerol, 2 mM dithiothreitol (DTT), 4.5% dodecyl maltoside (DDM) and 12 mg/mL L-α-phosphatidylcholine from soybean) and incubated at 55 C for 15 min. For the assay, to each tube strips, 0.2 µL of the reactivated enzyme was added to 14.8 µL of reaction buffer (200 mM Tris-Cl pH 8.0, 8.0 mM MgCl2, 333 mM KCl, and 67 mM NaCl) and 25 µL of inhibitor. The reaction mixtures were sealed with an adhesive sealing sheet (Thermo Scientific) and incubated at 71 C for 5 min. The reactions were started with the addition of 10 µL of 2 mM sodium pyrophosphate, further incubation at 71 C for 5 min and stopped by placing the reaction mixtures on ice for 10 min. The color development was done with the addition of ammonium heptamolybdate + ascorbic acid solution on ice for 10 min and then 90 µL of arsenic solution, at room temperature, for a minimum of 30 min. 180 µL of each reaction mixtures was then transferred into a clear 96 well polystyrene microplate and the absorbance was measured at 860 nm using a microplate spectrophotometer (MultiSkan Go, Thermo Scientific). The pilot inhibitor screening of 16 phosphate-mimicking fragments was done at 100 μM inhibitor concentration. Further screening was done with three different concentrations; 1 μM, 5 μM and 50 μM for soluble compounds, and 1 μM, 5 μM, and 20 μM for sparingly soluble compounds. 6 The best hits were validated using 8 concentrations in triplicate in TmPPase.
The potent hits 12k and 12l compounds were then tested on the P. falciparum (PfPPase-VP1) membrane fractions expressed in baculovirus-infected insect cells (the expression method and further enzyme analysis will be described in more detail elsewhere). Briefly, to each tube strips, 2.4 µL of PfPPase-VP1 membrane fraction (5 mg/mL) was added into 12.6 µL of reaction buffer (200 mM Tris-Cl pH 8.0, 8.0 mM MgCl2, 333 mM KCl, 67 mM NaCl, 3.3 mM NaF, and 17 μM Gramicidin D) and 25 µL of inhibitor. The reaction mixtures were incubated at 30 C for 5 min and the assay was started by the addition of 10 µL of 2 mM sodium pyrophosphate and further incubation at 30 C for 10 min. The reaction termination and color development were done as before.
The in vitro anti-plasmodial activity assay strain 3D7 was maintained in culture as described previously. 7 Briefly, parasites were cultured in O+ human erythrocytes at 5% haematocrit in RPMI-1640 medium supplemented with 0.5% Albumax II (Gibco, Carlsbad, CA, USA), 200 μM hypoxanthine (Sigma, St. Louis, MO, USA) and 20 μg/mL gentamycin (Gibco). Parasites were synchronized with the sorbitol method as described. 8 Test compound was dissolved in DMSO to make a stock solution of 10 mM. Two-fold serial dilutions of the test compound were made in the culture medium to cover the range of 400 µM -400 nM. Fifty microliters of each dilution were mixed in 96 well plate with 150 µL of 2% haematocrit of 0.5% parasitemia (ring stage). DMSO concentration in the highest compound concentration was 0.1%. After 72 h incubation at 37 °C, the parasite growth was quantified using fluorescent SYBR Green I ® -based assay as described. 9 The half-maximal inhibitory concentration (IC50) of the test compound was assessed using the non-linear regression fit model in Prism 7 (GraphPad Software). The potent anti-malarial drug, artemisinin, was use in parallel as a positive control. Each concentration of the test compound was tested in duplicate and the assay was repeated three times. The mean and standard deviation of the three repeats were used to estimate the IC50.

S36
IC50 determinations were done using R 3.6.1 10 and the n-parameter logistic regression (nplr) package 11 version 0.1-7. In this package, the weighted nplr, given x (compound concentrations) and y values (activity) is calculated using Richard's equation (eq 1) where B and T are the bottom and top asymptotes, respectively; b and xmid are the Hill slope and the x coordinate at inflexion point, respectively; and s is an asymmetric coefficient. The model parameters were simultaneously optimized using non-linear minimization, using a sum of squared errors. 12 The standard error of the model, defined as the squared error on the fitted values, is used to estimate a confidence interval at 95% on the predicted IC50 values.

General procedures A-D
The isoxazole core was mainly constructed by two different synthetic approaches (General procedures A or B), followed by hydrolysis (General procedure C) to give isoxazole-3-carboxylic acids. Some compounds were subsequently esterified to their corresponding 2-bromophenyl isoxazole-3-carboxylates (General procedure D).
Crossed condensation reaction (compounds 10e-j, 21 and 22). Freshly prepared sodium ethoxide (1 equiv) in EtOH (0.3 mL/mmol) was diluted with anhydrous Et2O (3 mL/mmol). To this solution was added subsequently diethyl oxalate (1 equiv) and arylketone 9, which were allowed to react at room temperature overnight. If a precipitate was formed, it was filtered off and then dissolved in H2O; if not the resulting mixture was extracted multiple times with H2O. The combined aqueous phases were acidified with glacial acetic acid and the formed precipitate was filtered off and dried under reduced pressure. The arylbutanoate product was used in the next step without further purification.
To a solution of arylbutanoate in EtOH (6 mL/mmol) was added glacial acetic acid (1 equiv) and hydrazine hydrate (1 equiv). The solution was stirred at room temperature for 5 h. The reaction mixture was diluted with an aqueous saturated solution of NaHCO3, extracted with EtOAc, dried with Na2SO4, filtered, evaporated and the residue was purified by flash chromatography with nhexane/EtOAc (1:0 → 0:1) as the eluent.

General procedure C -Hydrolysis of ethyl ester (compounds 11a-n, 21 and 22)
The ethyl ester and lithium hydroxide (2 equiv, except for 10b-d and 10n that required 4 equiv) were dissolved in a 2:1 mixture of EtOH and H2O (20 mL/mmol). The mixture was stirred at room temperature for one hour to two days until completion. The mixture was diluted with H2O. Depending on the aqueous solubility of the salts, the easily soluble ones were washed with EtOAc, whereas the sparingly soluble salts were not, before acidification using 1 M hydrochloric acid. The product was extracted with EtoAc, dried with Na2SO4, evaporated and dried in vacuo overnight.

General procedure D -Esterification of carboxylic acids (compounds 12h-l and 25)
To the carboxylic acid was carefully added a solution of oxalyl chloride (1.2 equiv) in DCM (20 mL/mmol) and a catalytic amount of anhydrous DMF. The resulting mixture was allowed to stir for 30-60 min before it was bubbled with argon for 5 min. The formed acid chloride was cooled to 0 °C in an ice bath. To this solution was added 2-bromophenol (1.2 equiv) in DCM (15 mL/mmol) during 10 min, followed by triethylamine (1.2 equiv) in DCM (15 mL/mmol). The mixture was stirred for 20 min in the ice bath and another 30 min at room temperature (in total about one hour since the phenol addition). The solvent was evaporated and the residue was purified by flash chromatography with n-hexane/EtOAc (1:0 → 0:1) as the eluent.