Oxime@Zirconium-Metal–Organic Framework Hybrid Material as a Potential Antidote for Organophosphate Poisoning

A novel material with dual activity toward organophosphate (OP) poisoning, based on Zr-MOF-808 and neutral oxime RS69N, has been prepared. The hybrid material has a significant drug payload (5.2 ± 0.9 oxime to MOF-808 molar ratio) and shows a sustained oxime release in simulated physiological media, leading to the successful reactivation of OP-inhibited acetylcholinesterase. At the same time, the hybrid system presents an efficient and moderately fast removal rate of a toxic organophosphorus model compound (diisopropylfluorophosphate) from simulated physiological media (t1/2 = 183 min; 95% removal rate after 24 h).


S.1. General Methods of characterization
All chemicals were commercially available at commercial sources and used without further purification. MOF-808 was synthesized using an Anton Paar Monowave 300 reactor. Powder X-ray diffraction data (PXRD) were obtained at room temperature on a D2 PHASER Bruker diffractometer using Cu Kα radiation (λ = 1.5418 Å) and collected in the 5°-35° 2θ range, with steps of 0.02° and time per step of 1.0 sec. Prior to each measurement, the samples were manually grounded in an agate mortar and then deposited in the hollow of a zero-background silicon sample holder. Infrared spectra were collected in a Fourier transform infrared spectrophotometer Bruker Tensor 27 (32 scans, resolution = 2 cm -1 ). 1 H and 31 P Nuclear Magnetic Resonance Spectroscopy data were obtained in a BRUKER Nanobay Avance III HD High Definition 400 MHz (2-channel) NMR spectrometer. Thermogravimetric Analysis (TGA) were carried out by a Thermogravimetric analyser METTLER-TOLEDO mod. TGA STAR system under air flow (20 mL min -1 ) running from room temperature to 950 °C with a heating rate of 10 °C min -1 (CIC, University of Granada). Scanning electron microscopy (SEM) images were obtained on a Zeiss SUPRA40VP (FESEM) system. DIFP adsorption experiments were carried out in an Agilent 8860 GC Chromatograph. This chromatograph has FID detector and a 16 port autosampler. An HP-5 column of 50 m length, 0.320 mm diameter and 1.05 um thickness was used, which allows working from -60 °C to 325 °C. Enzymatic assays were performed in a Tecan Infinite® 200 PRO NanoQuant.
Afterwards, the suspension was heated up to 95 °C (heating ramp of 60 min), then kept at 95 °C for 1 hour and finally, a fast cooling-step of 2 min to reach room temperature.
The white powder was centrifuged (2,486 g / 5 min) and washed three times with H2O and once with acetone. To remove the solvent trapped inside the pores, the material was activated under dynamic vacuum at 100 ºC during 8 h. TGA residue (calc./exp.): 54.21/54.61 % ( Figure S11). Yield: 71.8 %
Triethylamine (1.40 mL, 10 mmol) was then added dropwise and the mixture was stirred for 1 h. Then, the solution was concentrated via rotary evaporation to give a white residue.
The mixture was sonicated and rotary evaporated to an oil. Then, 20 mg MOF-808 (0.01 mmol) and 1 mL of MeOH was added and stirred for 24 h. The white solid was washed with H2O (2 x 1 mL) and recovered by centrifugation (9,168 g / 5 min). To remove the solvent, the material was lyophilized. In order to quantify the amount of RS69N encapsulated into MOF-808, 1 H spectra were recorded at room temperature. In a typical experiment, RS69N@MOF-808 (20 mg) was mixed with 600 µL of deuterated NaOH 10 M and digested for a period of 24 h. The supernatant was collected by centrifugation (9,168 g / 5 min) and analysed by 1 H NMR spectra. TGA residue (calc./exp.): 30.8/35.8 % ( Figure S11).

S.3. RS69N release from RS69N@MOF-808
20 mg of RS69N@MOF-808 and 20 µL of dimethylacetamide (0.02 mmol), as internal reference, were suspended in 480 µL of deuterated PBS (100 mM). The concentration of released RS69N was followed by 1 H NMR spectra, at different times. In addition, the supernatant at 24 h was separated from the solid, mixed with 1 mL of deuterated NaOD 1 M and analysed by 1 H NMR, as well. All the experiments were performed by duplicate.

S.4. Enzymatic assays
The assays of inhibition enzymatic activity were carried out similarly than the reactivation assays. Specifically, a DIFP solution (0.028 M) and a suspension of RS69N@MOF-808 (20 mg) and DIFP (0.028 mM) in PBS were incubated at 37 ºC. After 24 hours of incubation, the supernatants were collected by centrifugation and diluted 560 times.
Afterwards, 100 µL of the diluted-solutions (final DIFP concentration of 5·10 -6 M in negative control) were added to the well containing the enzyme solution (725 μL Tris-HCl (0.1 M, pH 7.5) and 25 μL of AChE aqueous solution (75 U/mL)) and the enzymatic activity was estimated following the indoxyl-acetate colorimetric method explained above (Table S2). All enzymatic assays were performed by triplicate.

S.5.1. Gas chromatography studies
The adsorption of diisopropylfluorophosphate (DIFP) was studied by mixing MOF-808 or RS69N@MOF-808 (20 mg, 0.015 mmol), 2.5 µL of dimethyl acetamide (DMA, internal reference) and DIFP (2.5 µL, 0.015 mmol) in 500 µL of PBS (100 mM) in a closed vial with a septum. The evolution of the concentration of DIFP was followed at room temperature by means of Gas Chromatography taking 0.5 µL aliquots of the supernatant solution at each time.
The experimental data were fitted to a pseudo-second order model (Eq. 1):

S.5.2. 1 H and 31 P NMR studies
The adsorption/degradation of DIFP was also studied by 1 H and 31 P NMR. In a typical experiment, MOF-808 or RS69N@MOF-808 (20 mg, 0.015 mmol) and DIFP (2.5 µL, 0.015 mmol) were mixed in 500 µL of deuterated PBS. After 24 h, the supernatant was separated from the solid by centrifugation (2,486 g / 5 min) and the solid was suspended in DMSO-d6 to extract the compounds trapped inside the cavities. Finally, both fractions (supernatant and extracted solution) were analysed by 1 H and 31 P NMR spectroscopy, using dimethylcetamide as an internal reference.

S.6. Computational modelling (Adsorbate locator)
Computational modelling of the interaction of RS69N oxime and DIFP with MOF-808 host framework was performed with the BIOVIA Materials Studio 2018 Adsorption Locator Module (https://www.3ds.com/products-services/biovia/products/molecularmodeling-simulation/biovia-materials-studio/). Monte Carlo searches of the configurational space of the substrate-adsorbate system were carried out in order to 6 identify the most favourable adsorption configurations of the studied substrates in the MOF pore structure. We have explored oxime loadings of 1 to 52 oxime molecules per crystal cell which corresponds to 0.0625 to 3.25 oxime molecules per MOF formula unit.
For DIFP molecule we have explored 0.0625 to 0.313 molecules per MOF formula unit.