Transforming an esterase into an enantioselective catecholase through bioconjugation of a versatile metal-chelating inhibitor

,

. Gradient used for HPLC detection used for the first experiments Mass spectrometry was performed on a hybrid quadrupole time-of-flight (QTOF) analyzer, model QSTAR, Pulsar I, from AB Sciex (Framingham, MA, U.S.A.), as previously described. S3 Samples were analyzed by direct infusion and ionized by electrospray ionization mass spectrometry (ESI-MS) with methanol as the mobile phase in positive reflector mode. High-resolution mass spectrometry (HR-MS) analysis (see Figure S4) was carried out by flow injection analysis combined with electrospray ionization mass spectrometry (FIA-ESI-MS) on a QTOF Agilent G6530A accurate mass QTOF liquid chromatography-mass spectrometry (LCMS) system (Agilent Technologies, Santa Clara, CA, U.S.A.). The sample was directly infused and ionized by ESI in negative reflector mode. Ionization was enhanced by JetStream technology, and the mobile phase was 99.9:0.  Electrochemical measurements. The electrochemical experiments were run, following conditions previously reported, S3 at 22±1 ºC in a 3-electrode electrochemical cell configuration. Rotating gold disc electrodes (Pine, 5mm diameter) were used as working electrodes. The reference electrode selected was BAS Ag/AgCl 3M (+210 mV vs SHE) and the counter electrode was a platinum wire. All redox potentials shown in the work are given vs. Ag/AgCl. The electrochemical experiments were recorded with an Autolab PGSTAT30 controlled with NOVA2 software (EcoChemie, NL). The electrochemical impedance spectroscopy was performed using the same cell configuration in presence 1 mM of the external redox probe [Fe(CN) 6 ] 3-/4-(Merck Life Science S.L.U., Madrid, Spain) at a scanrate of 20 mV s -1 . The EIS measurements were carried out at a for-mal potential of the redox couple at which potential pulses of frequency between 0.1 Hz and 10 kHz were applied.
Before electrochemical measurements, gold electrode surface cleaning and modification was performed as previously described. S3 Briefly, gold electrodes were immersed in "piranha" solution (3 H 2 SO 4 98% (Merck Life Science S.L.U., Madrid, Spain): 1 H 2 O 2 30% (Panreac, Madrid, Spain)) during 10 min. Afterwards the electrodes were rinsed with water and polished successively with alumina suspensions of 1, 0.3, and 0.05 μm in diameter, respectively during a total of 3 min. (CAUTION: Piranha solution is especially dangerous, is corrosive, and may explode if contained in a closed vessel; it should be handled with special care.) After rinsing, the electrodes were taken into an EtOH/H 2 O 2:1 solution and immersed into an ultrasound bath during 15 min. Later the electrodes were taken into an electrochemical cell containing 0.5 NaOH and 20 electrochemical reductive cyclic voltammograms from 0 to -1.5 V using 200 mV·s -1 scan rate were performed to clean the gold surface. Final activation/cleaning step consisted on 25 oxidative cyclic voltammograms from 0 to +1.5 V using 100 mV·s -1 scan rate and H 2 SO 4 0.1 M as electrolyte. Modification of gold electrodes was performed by immersion of the freshly cleaned electrodes into an ethanol solution containing 1 mM of 3-mercaptopropionic acid (MPA) (99%; Merck Life Science S.L.U., Madrid, Spain). Solutions were let to react overnight and clean by immersion in ethanol during 15 min. EH 3 was immobilized by physical adsorption by depositing on the electrode 10 µL of a solution containing the enzyme, 10 mg·mL -1 in 50 mM K 2 HPO 4 buffer pH 6.5, during 1 hour. Then, the electrode was washed with water, afterwards it was taken into the electrochemical cell to determine the electrochemical response. Biomimetic EH 3 , herein referred to as bEH 3 , was prepared by incubating, at 30ºC, a solution containing the enzyme, 10 mg mL -1 (or ca. 0.26 mM) in 50 mM K 2 HPO 4 buffer pH 6.5, with 0.63 mM of 2 (from a 126 mM stock solution in dimethyl sulfoxide), and after 10 min incubation biomimetic was extensively dialysed against 50 mM K 2 HPO 4 buffer pH 6.5 using Pur-A-LyzerTM Maxi 1200 dialysis kit (Merck Life Science S.L.U., Madrid, Spain), overnight at 4 °C. The dialyzed biomimetic solution was recovered and incubated with 0.63 mM Cu(NO 3 ) 2 (from a 20 mM stock solution in the same buffer) for additional 10 minutes, after which the solution was again dialyzed to remove the unbound cation. The dialyzed protein solution was recovered and concentrated by ultra-filtration through low-adsorption hydrophilic 10000 nominal molecular weight limit cutoff membranes (regenerated cellulose, Amicon) to reach a final protein concentration of 0.26 mM (per monomer), and used inmediately.
Crystallization and X-ray structure determination of EH 3 complexed with 2. The complex bEH 3 , was obtained by cocrystallization assays, using the sitting drop vapor diffusion method at 18°C. Initial crystallization conditions were explored by high-throughput techniques with a NanoDrop robot (Innovadyne Technologies, USA), incubating EH 3 (24 mg mL -1 in 40 mM 2-[4-(2hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) pH 7, 100 mM NaCl) with inhibitor 2 (in 100% dimethylsulfoxide (DMSO)) to a final concentration of 10 mM inhibitor, for 2-3 hours before setting the drops. The commercial screens Index (Hampton Research, USA) and PACT++ (Jena Bioscience, Germany) were explored by mixing 250 nL protein solution with 125 nL of the reservoir. Further optimization led to suitable plaque-shaped crystals growing after four days, from 1 µL of the protein mixture and 0.5 µL of the reservoir containing 27% PEG3350, 0.1M Bis-Tris pH 6.5. For data collection, crystals were transferred to cryoprotectant solutions consisting of 29% PEG3350, 0.1M Bis-Tris pH 6.5 and 25% glycerol, before being cryocooled in liquid nitrogen. Diffraction data were collected using synchrotron radiation on the XALOC beamline at ALBA (Cerdanyola del Vallés, Spain). Diffraction images were processed with XDS S4 and merged using AIMLESS from the CCP4 package. S5 The crystals were indexed in the P2 1 space group, with four molecules in the asymmetric unit and 48% solvent content within the unit cell. The data-collection statistics are given in Table S2. The structure of bEH 3 was solved by Molecular Replacement with MOLREP, S6 using the coordinates from the unliganded EH 3 as template (PDB code 6SXP). Crystallographic refinement was performed using the program REFMAC S7 within the CCP4 suite with automatic local noncrystallographic symmetry (NCS) and amplitude-based twin refinement. Free R-factor was calculated using a subset of 5% randomly selected structure-factor amplitudes that were excluded from automated refinement. At the later stages, ligands were manually built into the electron density maps with COOT8 S8 and water molecules were included in the model, which, combined with more rounds of restrained refinement, reached the R factors listed in Table S2. For inhibitor 2, not present in the Protein Data Bank, a model was built using MacPyMOLX11Hybrid (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC). The model was used to automatically generate coordinates and molecular topologies with eLBOW suitable for REFMAC refinement. S9 The figures were generated with PyMOL. The crystallographic statistics of bEH 3 are listed in Table S2. [c] Rwork / Rfree = ∑hkl | Fo -Fc | / ∑hkl | Fo |, where Fc is the calculated and Fo is the observed structure factor amplitude of reflection hkl for the working / free (5%) set, respectively.
Preparation of biomimetic catalysts. Biomimetics, herein referred to as bEH 3 and bEH 1AB1 , were prepared by incubating at 30ºC a solution containing the enzyme, 0.26 mM, in 50 mM K 2 HPO 4 buffer pH 6.5, with 0.63 mM of 2 (from a 126 mM stock solution in DMSO). The reaction was monitored indirectly by measuring the biocatalytic activity using a pH indicator assay with glyceryl tripropionate as substrates, as described previously. S10 In all cases, full inhibition was observed after 10 min, after which biomimetics were extensively dialysed against 50 mM K 2 HPO 4 buffer pH 6.5 using Pur-A-LyzerTM Maxi 1200 dialysis kit (Merck Life Science S.L.U., Madrid, Spain), overnight at 4°C. The dialyzed biomimetic solutions were recovered and incubated with 0.63 mM Cu(NO 3 ) 2 (from a 20 mM stock solution in the same buffer) for additional 10 minutes, after which the solution was again dialyzed to remove the unbound cation. The dialyzed protein solution was recovered and concentrated by ultra-filtration through low-adsorption hydrophilic 10000 nominal molecular weight limit cutoff membranes (regenerated cellulose, Amicon) to reach a final protein concentration of 0.26 mM (per monomer), and used inmediately for activity tests.
Determination of hydrolytic activity. Before bionjugation at large, esterase activity inhibition was measured over time by following the specific activity (units/mg protein) using glyceryl tripropionate (Merck Life Science S.L.U., Madrid, Spain) using the Phenol Red pH indicator assay. S1 The activity was calculated by determining the absorbance per minute from the slopes generated, and by applying the following formula (Equation 1 All values, in triplicate, were corrected for nonenzymatic transformation. The absence of activity was defined as at least a twofold background signal as described. S1 Catecholase activity determination and High Performance Liquid Chromatography (HPLC) analysis. Catecholase activity was evaluated (at 30ºC) by following the conversion of 3,5-di-tert-butylcatechol to 3,5-di-tert-butyl-O-benzoquinone. Briefly, 2 L of 3,5-ditert-butylcatechol (from a stock solution of 225 mM in dimethyl sulfoxide) were added to 88 L of 50 mM K 2 HPO 4 buffer at pH 6.5. Then, 10 L of biomimetic enzyme solution (from a stock solution of 2.6 mM in 40 mM HEPES buffer, pH 7.0) were added. After 60 min, reactions (in triplicates) were stopped by adding 900 L HPLC-grade methanol and the reaction products analysed by HPLC, compared to a control reaction without enzyme. For the HPLC analysis, the samples were filtered employing a 0.45 µm nylon filter and the presence of the compounds were quantified by HPLC analysis, performed using a quaternary pump (model 600, Waters) coupled to an autosampler (Varian ProStar, model 420). The injection volume was 10 µL. The column was a Zorbax Eclipse Plus C-18 (4.6 x 100 mm, 3.5 μm, Agilent Technologies) with a constant temperature of 40 °C. The mobile phase consisted of an isocratic mixture of acetonitrile/H 2 O 58:42 (v/v) acidified with a 0.1% (v/v) of formic acid and degassed with helium. The flow rate was 0.8 mg mL -1 during 14 min of analysis. The detection of peaks was carried out using a photodiode array detector (ProStar, Varian). Quantification was performed at 275 nm and integration was carried out using the software Varian Star LC workstation 6.41. The compound concentration of each sample was quantified from the peak area extracted from the chromatograms. Calibration curves for 3,5-di-tert-butylcatechol and 3,5-di-tert-butyl-O-benzoquinone (both provided by Merck Life Science S.L.U., Madrid, Spain), between 0 and 5 mM, were performed to extract exact concentrations in reaction mixtures.
Oxidation of catechin: spectrophotometer and HPLC. The reaction conditions for the oxidation of ()-catechin or ()-catechin (Merck Life Science S.L.U., Madrid, Spain) were set up in 96 well plate, as follows.
 200 µL 50 mM K 2 HPO 4 buffer at pH of 6.5 are added to each well.  50 µL of (+) or ()-catechin solution (10.4 mg mL -1 stock solution in acetonitrile) are added (final concentration, 2 mg mL -1 , or 6.9 mM).  Then, 10 µL biomimetic solution (from a 0.26 mM stock solution, or 10 mg mL -1 ) are added (final concentration, 0.01 mM or 0.38 mg mL -1 ).  Measure the decrease of absorbance at 290 nm (catechin) and increase of absorbance at 330 nm (catechin quinone) in a spectrophotometer at 30 ˚C every one min. All values, in triplicate, were corrected for nonenzymatic transformation.
Specific activity (unit/mg protein) is calculated by determining the absorbance per minute from the slopes generated, and by applying the following formula (Equation 2): Assuming the following parameters:  Extinction coefficient () catechin: 10230 M -1 cm -1  Total assay volume: 260 μL  Total amount protein: 0.0026 µmol or 0.1 mg For HPLC analysis, the samples were diluted 1 : 4 in methanol and filtered employing a 0.45 µm nylon filter and the presence of the compounds were analyzed by HPLC, performed using a quaternary pump (model 600, Waters) coupled to an autosampler (Varian ProStar, model 420). The injection volume was 5 µL. The column was a Zorbax Eclipse Plus C-18 (4.6 x 100 mm, 3.5 μm, Agilent Technologies) with a constant temperature of 40 °C. The mobile phase consisted of a gradient with acetonitrile and water, both acidified with a 0.1% (v/v) of formic acid and degassed with helium. The mobile phase was fixed at 5 % of acetonitrile during 1 minute and, after that, a gradient from 5 % to 50 % of acetonitrile in 3 minutes was carried out, keeping the acetonitrile fixed at 50 % during 3 minutes more. After that, the mobile phase return to initial conditions in 3 minutes and the column was equilibrated during 10 minutes before the next injection. The flow rate was 0.8 mL min -1 during the 20 minutes of analysis. The detection of peaks was carried out using a photodiode array detector at 280 and 350 nm (ProStar, Varian). Relative amount of catechin (see Figure S5) was obtained from the peak area at 280 nm (see Figure S6) employing the software Varian Star LC workstation 6.41. We considered 100 % the area of the total amount of catechin added.