Neutralization of Reactive Oxygen Species at Dinuclear Cu(II) Cores: Tuning the Anti- Oxidant Manifold in Water by Ligand Design

Neutralization of Reactive Oxygen Species at Dinuclear Cu(II) Cores: Tuning the AntiOxidant Manifold in Water by Ligand Design Andrea Squarcina, Alice Santoro, Neal Hickey, Rita De Zorzi, Mauro Carraro, Silvano Geremia, Marco Bortolus , Marilena Di Valentin , Marcella Bonchio aITM-CNR and Department of Chemical Sciences, University of Padova, Via Marzolo 1, 35131 Padova, Italy. bDepartment of Chemical and Pharmaceutical Sciences, University of Trieste, Via Licio Giorgieri 1, 34127 Trieste, Italy. To whom correspondence should be addressed. E-mail: marcella.bonchio@unipd.it


Instrumentation
UV-Vis spectra were recorded with an UV-Vis Varian Cary 100 for the SOD activity determination and an UV-Vis Varian Cary 50 for other measures. FT-IR spectra of the KBr pellets of the compounds were recorded on a Nicolet 5700 FT-IR instrument. Cyclic voltammetry experiments were performed using a BAS Cell C3 EC-epsilon potentiostat. A standard three-electrode electrochemical cell was used. Glassy carbon electrode (3 mm diameter, geometric surface area = 7 mm 2 ) from BAS and a Pt wire were used respectively as working and auxiliary electrode.
Potentials were referred to an Ag/AgCl/(3 M NaCl) reference electrode. Prior to each experiment, the electrode was polished with 1 µm alumina, rinsed with deionised water and wiped with a paper tissue. Potential were then reported vs NHE. ESI-MS measurements were carried out by using an Agilent Technologies MSD SL Trap mass spectrometer with ESI source. Sample solutions in water with 0.1 % v/v HCOOH were injected into the ion source without the addition of any other solvent at a flow rate of 50 µL/min. For electrospray ionization, the drying gas (nitrogen) was heated at 325 °C. Each species is indicated with the m/z value of the first peak of its isotopic cluster. The peaks in ESI-MS spectra were assigned by comparing the experimental isotopic patterns with the corresponding simulated profiles. EPR spectra were recorded on an ELEXSYS E580 spectrometer equipped with a rectangular cavity, ER4102ST, both from Bruker, Germany, and fitted with a cryostat (ESR900) and a variable-temperature controller (ITC503S), both from Oxford Instruments, UK. All experiments were performed at 50 K on samples frozen in dry ice and sealed under vacuum in quartz tubes (i.d. 3 mm, o.d. 4 mm). The experiments were performed using the following parameters: microwave frequency 9.35 GHz, microwave power 6.4 mW (attenuation 15 S3 dB), sweep width 150 mT, center field 300 mT, conversion time 82 ms, time constant 82 ms, modulation amplitude 0.95 mT, 1024 points, 3 averages.

Caution:
Although the salts and the complexes reported do not appear to be mechanically sensitive, perchlorates should be treated with due caution.

X-Ray Diffraction Analysis
Diffraction data from single crystals of dinuclear complexes [Cu2L 1 2], [Cu2L 2 2] and [Cu2L 3 2] were collected by the rotating crystal method using synchrotron radiation at the XRD1 beam-line of the Elettra Synchrotron, Trieste, Italy. In all cases, a single crystal of the complex was dipped in the paratone cryoprotectant and flash-frozen to 100 K in a stream of N2 vapour. Diffraction images were indexed and integrated using the XDS package and the resulting data sets were scaled using XSCALE. [S3,S4] The structure of [Cu2L 1 2] was solved in space group P-1 with SHELXS (Direct Methods); while the structures of complexes [Cu2L 2 2] and [Cu2L 3 2] were solved in space groups P-1 and P21/a, respectively, with SIR-2014 (VLD and Patterson Phasing, respectively). Refinements were implemented with SHELXL-14, operating through the WinGX GUI, by full-matrix least-squares (FMLS) methods on F 2 . [S5-S7] Thermal parameters of all non-hydrogen atoms were refined anisotropically. Hydrogen atoms were placed at the geometrically calculated positions and refined using the riding model. Crystal data and final refinement details for the three complexes are reported in Table S1. The asymmetric unit of [Cu2L 1 2] contains one half of the dinuclear dimer, a perchlorate ion and two acetonitrile solvent molecules. Extinction correction was applied using the card EXTI in SHELXL-14. [S6] The asymmetric unit of [Cu2L 2 2] contains one half of the dinuclear dimer, which includes a DMF molecule to complete the coordination sphere, a perchlorate anion and half an ethanol molecule. All atoms of the perchlorate anion are disordered in two positions and the occupancies were refined freely to 0.73/0.27. DFIX and DANG restraints were applied, using idealised tetrahedral bond lengths and angles as targets, to maintain the tetrahedral S5 geometry. The ethanol molecule lays on an inversion centre and was refined at half occupancy.
DFIX and DANG instructions were used to restrain distances and angles. EADP restraints were used to restrain the thermal parameters of selected atoms of the perchlorate ions and DMF molecule, while thermal parameters of the ethanol molecule were restrained with RIGU, SIMU and ISOR instructions. A FLAT instruction was used to maintain all the non-hydrogen atoms of the DMF molecule on a single plane. Analogously to [Cu2L 2 2], the asymmetric unit of [Cu2L 3 2] contains one half of the dinuclear dimer, which includes a dimethylformamide (DMF) molecule to complete the coordination sphere, a perchlorate anion and half an ethanol molecule. During refinement restrains were applied to disordered and/or partially occupied atoms, as outlined below. The DMF molecule in the coordination sphere is disordered in two positions, with the coordinating oxygen atom common to both disordered fragments at full occupancy. The occupancies of the disordered atoms were freely refined to 0.54/0.46. Bond lengths and angles were restrained with DFIX and DANG instructions using the idealised values as targets. The perchlorate anion is disordered in two positions around a 3-fold common axis and the occupancies of the disordered atoms were freely refined to 0.74/0.26. Bond lengths and angles were restrained with DFIX instructions using the idealised values as targets. Finally, the ethanol molecule lays on an inversion centre and was refined at half occupancy. Distances and angles were restrained with DFIX and DANG instructions using idealised values as targets. EADP instructions were used to restrain thermal parameters of selected disordered atoms of the perchlorate and DMF fragments.   Figure S1.

Cyclic Voltammograms
CV experiments were performed using a three-electrode setup system in a small volume cell.
Glassy carbon electrode (3 mm diameter, geometric surface area = 7 mm 2 ) from BAS and a Pt wire were used respectively as working and auxiliary electrode. Potentials were referred to an Ag/AgCl/(3 M NaCl) reference electrode; scan rate of 50 mV/s. Prior to each experiment, the electrode was polished with 1 µm alumina, rinsed with deionised water and wiped with a paper tissue. In a typical experiment, a 0.5-1 mM solution of the copper compound in 0.05 M phosphate buffer, pH = 7.8, with 0.1 M NaCl as support electrolyte was purged with nitrogen and then analysed. Potential were then reported vs NHE.                          Table S3 and Figure S51.  Table S3 and Figure S51.

SOD-like Activity
SOD-like activity of the complexes was measured by using the cyt c assay, described by McCord and Fridovich, with some modifications. [S9] Screening of catalysts as SOD-mimicking systems was performed by spectrophotometric analysis of the inhibition of the superoxide-dependent reduction of the cyt c chromophore to ferricythocrome ( Figure S52). Figure S52. Scheme of cytchrome c assay for SOD activity. The green box highlight the function of the SOD active catalyst and in the red box is highlighted the chomophore monitored during the reaction.
The superoxide radical anions was generated by the xanthine/xanthine oxidase system. In all experiments, the reaction mixture was prepared with 40 µM xanthine, 10 µM cyt c, catalase 15 µg/ml, 50 mM phosphate buffer (pH=7.80) and xanthine oxidase 0.0053 U/ml. Catalyst solutions (64-550 nM) were prepared upon dilution of a stock solution in water. Possible interference through inhibition of the xanthine/xanthine oxidase reaction by test compounds was examined by following the rate of urate accumulation at 295 nm in the absence of cyt c. [S9] The concentration of the stock catalyst solutions was chosen in order to obtain approximately an inhibition interval of the superoxide-dependent reduction of the cyt c between 40-60 %. Among standard conditions the apparent rate constant, kMcF can be determined from the IC50 value as kMcF = kcyt[cyt]/IC50 (where kcyt is 2.6×10 5 M −1 s −1 ). Each experiment was performed in triplicate. Figure S53. Plot of inhibition of cyt c reduction versus [Cu2L 1 2], in 50 mM phosphate buffer at pH 7.8 25° C, 40 M xanthine, 2 nM xanthine oxidase.

CAT-like activity
CAT-like activity was determined by adding to a solution of H2O2 (30 mM) in BBS 50 mM (pH=7.8) or Krebs-Henseleit (KH) buffer (pH=7.4) (12 ml), 300 µL of a 8 mM acetonitrile solution of the Cu2L x 2 complex to start the dismutation reaction. The reactor was maintained at 25 °C by a circulating thermostat, and the progress of reaction was determined by monitoring the pressure developed by molecular oxygen generated from dismutation of hydrogen peroxide into a closed vessel. The amount of O2 was determined by continuous detection of pressure variation, through a pressure transducer. Initial rates were calculated by linear regression of data within 10 % H2O2 conversion. Kinetic runs were performed in triplicate. Control experiments performed without the Cu2-based catalyst, confirmed that no oxygen evolution is detected from the buffer solution in the presence of H2O2. [S10] S37 Figure S56. O2 evolution kinetics by Cu2L 1 2 (200 µM) upon incubation with H2O2 (30 mM) at 25°C in borate buffer (50 mM, pH=7.8).           Figure S63).

EPR investigation
Sample preparation. EPR spectra of all three complexes were recorded before and after addition of EPR Spectra of Cu2L 2 2 and Cu2L 3 2. In Figure S76, the evolution of the EPR spectra of Cu2L 2 2 and Cu2L 3 2 are reported. Before the addition of hydrogen peroxide, both complexes are diamagnetic S48 and decompose to the monomeric form as reported for the parent Cu2L 1 2 showing an analogous behaviour. In the presence of bromide, the breakdown to paramagnetic monomers is remarkably slowed down. Figure S76. EPR spectra of Cu2L 2 2 (left) and Cu2L 3 2 (right) in the absence (top) or in the presence (bottom) of NaBr (50 mM).

Morin test
Pro-oxidant activity of the complexes was measured by following the oxidative degradation of morin in the presence of hydrogen peroxide by UV-Vis spectroscopy. [S11] Scheme S1. Morin bleaching with H2O2 catalysed by Cu2L 1 2.
The reactions were carried out in borate buffer (50 mM, pH 7.8) at 25°C over a period of 3 hours. A freshly prepared morin solution in DMSO was diluted in borate buffer to obtain a morin solution of 0.12 mM in all the experiments. To this mixture, a stock catalyst solution (1 mM) was added in order to obtain the desired catalyst concentration (50 μM). Finally, a commercial H2O2 stock S49 solution (30% w/w) was added to get a concentration between 10-30 mM. The change in the absorption maximum of morin at 390 nm was monitored at intervals of 10 min, to evaluate the peroxidase-like activity of the complexes. Considering the excess of hydrogen peroxide employed in the experiments we can extrapolate a pseudo-first order kinetic constant kb, where v = kobs•[morin]. The value of kb was obtained by linear regression of the initials rate of morin degradation at different hydrogen peroxide concentration (10-30 mM). Second-order rate constants were determined from linear regression of the observed pseudo-first-order rate constants vs [H2O2].     Table S5. Pseudo-first order rate constant kobs and rate constant kb for oxidative degradation of morin with Cu2L 1 2.