New Cysteine Protease Inhibitors: Electrophilic (Het)arenes and Unexpected Prodrug Identification for the Trypanosoma Protease Rhodesain

Electrophilic (het)arenes can undergo reactions with nucleophiles yielding π- or Meisenheimer (σ-) complexes or the products of the SNAr addition/elimination reactions. Such building blocks have only rarely been employed for the design of enzyme inhibitors. Herein, we demonstrate the combination of a peptidic recognition sequence with such electrophilic (het)arenes to generate highly active inhibitors of disease-relevant proteases. We further elucidate an unexpected mode of action for the trypanosomal protease rhodesain using NMR spectroscopy and mass spectrometry, enzyme kinetics and various types of simulations. After hydrolysis of an ester function in the recognition sequence of a weakly active prodrug inhibitor, the liberated carboxylic acid represents a highly potent inhibitor of rhodesain (Ki = 4.0 nM). The simulations indicate that, after the cleavage of the ester, the carboxylic acid leaves the active site and re-binds to the enzyme in an orientation that allows the formation of a very stable π-complex between the catalytic dyad (Cys-25/His-162) of rhodesain and the electrophilic aromatic moiety. The reversible inhibition mode results because the SNAr reaction, which is found in an alkaline solvent containing a low molecular weight thiol, is hindered within the enzyme due to the presence of the positively charged imidazolium ring of His-162. Comparisons between measured and calculated NMR shifts support this interpretation.


General Information
All reagents and solvents were obtained from commercial suppliers (Sigma Aldrich, Alfa Aesar, TCI chemicals, ABCR, Acros Organics and Fischer Scientific) and used without further purification. Flash column chromatography was performed using silica gel type 60 M (230-400 mesh, Macherey Nagel). Analytical thin-layer chromatography (TLC) was done on Merck silica gel plates (60 F254) with defined solvent mixtures and visualized under UV light irradiation and/or TLC staining reagents. Melting points were determined in open capillary tube. IR spectra were measured with a JASCO (FT/IR-4100) with a diamond ATR unit and are reported in terms of frequency of absorption (ν, cm −1 ). NMR experiments were performed on a 300 MHz (300 MHz 1 H and 75 MHz 13 C), a 400 MHz (400 MHz 1 H and 101 MHz 13 C) or a 600 MHz (600 MHz 1 H and 151 MHz 13 C) spectrometer from Bruker using deuterated solvents ((residual) solvent signals: CDCl3 : δH = 7.26 ppm, δC = 77.16 ppm; (CD3)2SO: δH = 2.50 ppm, δC = 39.52 ppm) as internal references and reported in parts per million (ppm, δ) relative to tetramethylsilane (TMS, δ = 0.00 ppm) [1]. Coupling constants (J) are reported in Hz, and the multiplet abbreviations used were: s, singlet; d, doublet; t, triplet; m, multiplet; br, broad; app, apparent; and combinations thereof. Electrospray ionization (ESI-) mass spectra were recorded on an Agilent/Bruker LC/MSD trap XCT spectrometer or on a 1200-series HPLC-system (Agilent-Technologies) with binary pump, integrated diode array detector and Ascentis Express C18 column (2.7 μm, 30x2.1 mm). ESI-HRMS spectra were recorded on a Waters Q-TOF-Ultima 3 instrument or an Agilent 6545 QTOF-MS mass spectrometer. Field desorption (FD-) mass spectra were measured on a Finnigan MAT 95 spectrometer. Preparative reverse phase separations were carried out on an Agilent 1290 Infinity II preparative system with a 1290 Infinity II preparative binary pump, 1260 Infinity II DAD and a 1290 Infinity II fraction collector. Optical rotations were measured with a Perkin-Elmer 241 polarimeter at 546 and 579 nm. Extrapolation to 589 nm was performed according to Thaler et al. [2].

Benzyl-N-(Tert-Butoxycarbonyl)-L-Phenylalanyl-L-Leucinate (A1)
The product was synthesized according to Lawesson et al. [3]. In an inert gas atmosphere, N-Boc-L-phenylalanine (0.67 g, 2.54 mmol), L-leucine benzyl ester p-toluenesulfonate salt (1.00 g, 2.54 mmol) and triethylamine (360 μL, 2.54 mmol) were dissolved in dichloromethane (7 mL). The suspension was cooled to -18 °C and N,N'-dicyclohexylcarbodiimide (0.52 g, 2.54 mmol) was added. The resulting mixture was stirred for 1 h at that temperature and then 20 h at room temperature. The mixture was filtered off and the solvent was removed under reduced pressure. The residue was dissolved in ethyl acetate (10 mL) and washed with hydrochloric acid (0.1 M, 2.5 mL), sodium bicarbonate (0.5 M, 2.5 mL) and brine (2.5 mL). The organic phase was dried over magnesium sulfate and the solvent was removed under reduced pressure.

Benzyl-L-Penylalanyl-L-Leucinate Hydrochloride Salt (A2)
The deprotection was carried out similar to Hruby et al. [4]. In an inert gas atmosphere, the dipeptide A1 (500 mg, 1.07 mmol) was added to HCl/ dioxane (4 M, 5 mL) at 0 °C. The ice bath was removed and the solution was stirred additional 30 min. The solvent was removed under reduced pressure, the product filtered off and washed with diethyl ether to produce a colorless solid (430 mg,

Benzyl-L-Penylalanyl-L-Leucinate Trifluoroacetic Acid Salt (A5)
The product was obtained by deprotection with trifluoroacetic acid. The dipeptide A1 (113 mg, 0.24 mmol) was added to a mixture of TFA (1 mL) and dichloromethane (2 mL) at 0 °C. After 10 min. the ice bath was removed and the solution was stirred for 2 h at room temperature. The solvent was removed under reduced pressure and the product (116 mg, 0.24 mmol, >99 %) was used without further purification.

1,2-Difluoro-4,5-dinitrobenzene (A4)
The product was synthesized according to Nitschke et al. [5]. To a mixture of sulfuric acid (6.4 mL) and nitric acid (15.9 mL) 1,2-difluorobenzene (2.00 g, 17.5 mmol) was slowly added at 0 °C. The resulting mixture was stirred 2 h at room temperature and 18 h at 100 °C. The solution was poured on ice and the precipitate was collected. Recrystallization from ethanol gave the title compound
After stirring for 17 h at room temperature, a significant amount of starting material was still present, so the suspension was heated to 80 °C for 21 h in a closed tube. After that time, all solids were dissolved and the solution had an intense yellow color. After evaporating the solvent, column chromatography (SiO2, DCM:MeOH; 20:1 to 10:1) afforded a product mixture also containing the nitro substituted byproduct. Preparative HPLC (ACE C18PFP, 5 μm, 150 mm × 30 mm, 37.5 mL/min, MeCN:water 1:1) gave the pure product (4 mg, 0.009 mmol, 11 %).

Calculation of the Chemical Shifts / Determination of the Regiochemistry of the Addition Products
Conformational searches were performed using Spartan'10 [17]. DFT calculations were performed using Gaussian 16, Rev. A.03 [18]. All structures were confirmed as local minima by vibrational frequency analysis (Nimag = 0).
Duplicates were removed and conformers within a relative energy range of up to 4.5 kcal mol -1 were selected. NMR shielding tensors were calculated using GIAOs [28] at the mPW1PW91/6-31+G(d,p) [25][26][27]29] level with IEFPCM solvation [30] for chloroform. Boltzmann-weighted average shielding tensors were generated and compared with the experimental 1 H and 13 C NMR data in the DP4+ framework to obtain the DP4+ probabilities [31].

Mass Spectrometry
Lyophilized rhodesain was reconstituted at 4 mg/mL in 50 mM Na acetate, pH 5.5, 200 mM NaCl, 5 mM EDTA. For mass spectrometric analysis, the protein was further diluted in the same buffer (to a final concentration of 850 nM) and reduced with DTT for 1 h at room temperature. After the addition of the inhibitors at a final concentration of 0.1 mM, samples were analyzed by LC-MS using a nanoAcquity UPLC system (Waters Corporation) coupled to a nano-ESI-Q-TOF mass spectrometer (Synapt G2-S HDMS, Waters Corporation). Rhodesain without compound served as control. Protein-drug complexes were loaded onto a 200 μm x 5 cm PepSwift Monolithic PS-DVB column from Dionex (Thermo Scientific) using direct injection mode. For LC separation, two mobile phases were used. Mobile phase A contained 0.1% formic acid (FA) and 3% DMSO in ultrapure water, whereas mobile phase B consisted of 0.1% FA and 3% DMSO in ACN. A gradient of 10-90% mobile phase B was run over 7 minutes at a flow rate of 2000 nL/min. Column temperature was set to 45 °C. After separation, the column was rinsed with 90% of mobile phase B and reequilibrated at initial conditions. All MS analyses were conducted in positive-mode ESI.

NMR Materials and Methods
For 19 F NMR, compounds 7 (800 μM) or 8 (400 μM) were dissolved in rhodesain sample buffer (50 mM NaAcetate, pH 5.5, 200 mM NaCl, 5 mM Na-EDTA, 1 mM TCEP) with 10% d6-DMSO from a 40 mM inhibitor stock in d6-DMSO. Spectra were recorded at 303 K on a Bruker Avance 3 600 MHz spectrometer with a Prodigy TCI cryoprobe (Bruker, Karlsruhe). Because 1 H and 19 F are measured in the same coil, protons were not decoupled during 19 F acquisition. Spectra were recorded with 512 scans (cpd 8) or with 3k scans (cdp 7) due to solubility issues. Unlabeled rhodesain was purified from P. pastoris as described previously and lyophilized until needed (see below). For NMR measurements, the lyophilized protein was resuspended in rhodesain buffer sample. For the time trace experiments, 2.1 nmol rhodesain (4 μM final conc.) was added to a NMR sample containing 800 μM of cpd. 7 and spectra recorded every 2 minutes. For observing the bound inhibitor, spectra of 400 μM cpd. 8 with 150 or 450 μM rhodesain were recorded (3k scans).

Expression of Rhodesain:
Rhodesain with a S172A mutation and without the C-terminal domain beginning at Thr343 was expressed and purified from Pichia pastoris as described previously [7,8].

Hydrolysis Assays:
Rhodesain and rhodesain inactivated with K11777 were dissolved in 99 μL assay buffer (50 mM NaOAc, pH 5.5, 200 mM, 5 mM EDTA, 5 mM DTT) to a final concentration of 0.4 mg/mL. Afterwards, 1 μL of 10 mg/mL cpd. 7 dissolved in DMSO were added and the mixture was incubated at room temperature for 24 h. As a negative control, 1 μL cpd. 7 was added to 99 μL assay buffer without any protease. The reaction was quenched by addition of 100 μL ACN and heated up to 95 °C for 10 min. The solution was centrifuged for 10 min at 17,000 g and the supernatant was filtrated through a 0.22 μm syringe filter prior to LC-MS analysis.
The LC-MS analysis of the reaction mixtures was performed at an HP Agilent 1100 system coupled to an Agilent LC/MSD ion trap. An Agilent Poroshell 120 EC-C18 (150 x 2.10 mm) column at 40°C was used for separation with an isocratic flow of 30% ACN and 70% H2O with 0.1% formic acid. A control proved that the ester 7 was not hydrolyzed by the reaction buffer. Furthermore, no conversion was observed for rhodesain irreversibly inactivated by K11777. For the latter experiment, 40.2 μM rhodesain was incubated in a mixture of 8.7 mM K11777 with 5% DMSO for 2 h at room temperature until no activity was observed in the substrate assay. Insoluble inhibitor was removed by centrifugation and the supernatant was used for the hydrolysis assay as described above.

Enzyme assays / Rhodesain and Cathepsins
The assays were performed as described previously [7,8] by incubating rhodesain in 190 μL assay buffer (50 mM sodium acetate pH 5.5, 5 mM EDTA, 200 mM NaCl, 0.005% Brij) with 5 μL of an inhibitor DMSO stock solution of a given concentration for 10 min at room temperature. The reaction was started by adding 5 μL of 400 μM Z-Phe-Arg-7-amino-4-methylcumarin (Z-Phe-Arg-AMC) in DMSO. The released AMC was excited at 380 nm and the upcoming fluorescence measured at 460 nm at 25 °C over time. The final substrate concentration was 10 μM.
Quenching of the fluorescence of the released AMC by the inhibitors was accounted for as described previously [9].
The cathepsin B (Calbiochem) assays were performed accordingly in 50 mM Tris buffer pH 6.5, 5 mM EDTA, 200 mM NaCl and 0.005% Brij leading to final enzyme and substrate concentrations of 484 nM and 100 μM.

Enzyme Assays / DENV PR
The assays were performed as described previously [32].

Dilution Assays:
Dilution assays were performed by adding 50 μL of cpd. 8 (0.1 mM in DMSO) to rhodesain (0.37 mM final concentration) in 950 μL assay buffer (50 mM sodium acetate pH 5.5, 5 mM EDTA, 200 mM NaCl) followed by 45 min incubation at room temperature. In a control, the cpd. 8 solution was replaced by 50 μL pure DMSO and treated equally. 5 μL of 400 mM Z-Phe-Arg-AMC were added to 195 μL of rhodesain-8 or rhodesain-DMSO stock solutions and the upcoming fluorescence was measured over time to assure complete inhibition (see Figure SI1)). For the dilution assay, 2 mL of the rhodesain-8 stock solution were diluted with 193 mL 50 mM sodium acetate buffer (pH 5.5 with 5 mM EDTA, 200 mM NaCl 5 mM DTT) and the recovery of the enzymatic activity was measured immediately after adding 5 mL of 400 μM Z-Phe-Arg-AMC (see Figure SI2). Figure S2: Progress curve for substrate hydrolysis by rhodesain, completely inhibited with inhibitor 8, after dilution. The recovery of the enzymatic activity proves the reversibility of the inhibition.

T. b. brucei Cell Survival Assay
Toxicity of 7 and 8 against trypanosomes (T. b. brucei 449 cell line) were determined via an ATPlite assay as described previously [10][11][12] using the cellular ATP levels as a proxy for cell viability. 7 ( 5 mM stock solution in DMSO) and 8 (50 mM stock solution in DMSO) were first diluted 1:3 in medium, followed by a 1:10 dilution step in a microplate and ten subsequent 1:2 dilution steps. 90 μL HMI-9 medium containing 2500 cells/mL were distributed in 96-well microplates (PerkinElmer). 10 μL of the 1:2 dilution step preparations of the tested compounds were added to the 90 μL cell suspension leading to final concentrations from 16.67 μM to 32.55 nM for 7 and 166.67 μM to 325.52 nM for 8 in the microplates. As a negative control, addition of 0.3% of DMSO corresponding to the highest DMSO concentration added by compound application was used. 10% DMSO was used as a positive control, since at this DMSO concentration, all cells die. Measurements were carried out as two sets of triplicates incubated at 37 °C for 24 h and 48 h. 50 μL of ATPlite 1 step solution (PerkinElmer) was added to each well of the microplate and luminescence measured at room temperature with an Infinite ® M200 PRO plate reader (Tecan Trading AG). The measured values were plotted against the compound concentrations top yield the dose-response curve. The EC50 values were calculated using GraFit version 5.013 (Erithacus Software Ltd.).

Docking Procedures
Docking experiments were performed using the crystal structure of rhodesain bound to the covalent inhibitor K11777 (pdb 2p7u) [13]. To define the binding site, all residues within a 6.5 Å shell around K11777 were selected. All water molecules present in the crystal structure were omitted. Generation of 3D-coordinates and energy minimization of the ligands were accomplished with the Molecular Operating Environment (MOE) using the MMFF94x force field [14]. Docking calculations were run with FlexX (version 2.3.2) and ranked with the built-in empirical scoring function. [15] Selected docking solutions were visualized with PyMOL (version 2.3.0) [16]. Figure S3. Highest ranked docking solution of ester 7 (score -20.56). The substituted aromatic ring binds to the S1' region and is not located close to the nucleophilic cysteine (distance 6.2 Å). Light grey: solvent accessible surface of rhodesain, grey: carbon atoms of rhodesain amino acid residues, orange: carbon atoms of ester 7, blue: nitrogen, red: oxygen, yellow: sulfur, cyan: fluorine. The substituted aromatic ring is located in close proximity to the nucleophilic cysteine (2.5 Å). One nitro group forms hydrogen bonds to Gln-19 and Trp-184. The phenylalanine side chain is placed in the S2 pocket and the benzyl ester group occupies the S3 pocket. Light grey: solvent accessible surface of rhodesain, grey: carbon atoms of rhodesain amino acid residues, orange: carbon atoms of ester 7, blue: nitrogen, red: oxygen, yellow: sulfur, cyan: fluorine.