Azologization and repurposing of a hetero-stilbene-based kinase inhibitor: towards the design of photoswitchable sirtuin inhibitors

The use of light as an external trigger to change ligand shape and as a result its bioactivity, allows the probing of pharmacologically relevant systems with spatiotemporal resolution. A hetero-stilbene lead resulting from the screening of a compound that was originally designed as kinase inhibitor served as a starting point for the design of photoswitchable sirtuin inhibitors. Because the original stilbenoid structure exerted unfavourable photochemical characteristics it was remodelled to its heteroarylic diazeno analogue. By this intramolecular azologization, the shape of the molecule was left unaltered, whereas the photoswitching ability was improved. As anticipated, the highly analogous compound showed similar activity in its thermodynamically stable stretched-out (E)-form. Irradiation of this isomer triggers isomerisation to the long-lived (Z)-configuration with a bent geometry causing a considerably shorter end‐to‐end distance. The resulting affinity shifts are intended to enable real‐time photomodulation of sirtuins in vitro.


S3 General remarks
All solvents and reagents were obtained from commercial suppliers and were used without purification. Anhydrous solvents were purchased from Acros Organics. Thin layer chromatography (TLC) was executed on silica gel 60 F254 aluminium plates purchased from Merck. Visualization of the compounds was accomplished by  and by staining with iodine, DNPH/H2SO4 (2 g 2,4-dinitrophenylhydrazine and 5 mL H2SO4 in 50 mL EtOH and 16 mL water) or vanillin/sulfuric acid (3 g vanillin and 0.5 mL H2SO4 in 100 mL EtOH) reagent. Chromatographic purification of products was performed by flash chromatography on silica gel (20-45 µm, Carl Roth) applying pressured air up to 0.8 bar. NMR spectra were recorded on a Bruker Avance III instrument ( 1 H NMR: 400 MHz, 13 C NMR: 100.6 MHz). Chemical shifts were referenced to tetramethylsilane (TMS) as internal standard in deuterated solvents and reported in parts per million (ppm). Coupling constants (J) are reported in Hz using the abbreviations: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet and combinations thereof, br = broad. Infrared (IR) spectra were recorded on a Bruker Alpha FT-IR spectrometer equipped with a diamond ATR unit and are indicated in terms of absorbtion frequency [cm −1 ]. Microwave synthesis was conducted in a Monowave 300 microwave synthesis reactor from Anton Paar equipped with appropriate sealed reaction vessels G10 (6 mL) or G30 (20 mL), applying a maximum initial power of 850 W to reach a given temperature (IR sensor) for a given time with stirring at 600 rpm. Melting points were measured in open capillary tubes using a Melting Point M-565 apparatus from Büchi and are uncorrected. High accuracy mass spectra were recorded on a Shimadzu LCMS-IT-TOF using ESI ionization. An Elementar Vario MICRO cube was used for the experimental determination of elemental configurations of final pure products. Preparative and analytical HPLC were performed using Shimadzu devices CBM-20A, LC-20A P, SIL-20A, FRC-10A with SPD 20A UV/Vis detector and an ELSD-LTⅡ. In analytical mode a LiChroCART ® (250 × 4 mm) and in preparative mode a Hibar® RT (250 × 25 mm) column, both containing LiChrospher ® 100 RP-18e (5 µm), were used. UV-vis spectra were obtained using a Thermo Scientific Genesys 10S UV-VIS spectrophotometer.

S13
Synthesis of 2c and 2g

1-Fluoro-4-vinylbenzene
In an inert gas atmosphere 4-fluorobenzaldehyde (1.24 g, 10.00 mmol, 1.00 equiv) and methyltriphenylphosphonium bromide (4.29 g, 12.00 mmol, 1.20 equiv) were dissolved in anhydrous THF (50 mL). At 0 °C sodium hydride (60 % dispersion in mineral oil, 1.80 g, 45.00 mmol, 4.50 equiv) was added in small portions and the mixture was stirred over night at room temperature. The reaction mixture was taken up in EtOAc (100 mL) and washed with aq. sat. NaCl-solution (3 × 20 mL). It was dried over MgSO4 and freed from solvent under reduced pressure. The residue was purified by silica gel column chromatography (n-hexane), which yielded the product as a colourless oil

Synthesis of 8a
Benzo

PSS composition of 11
Isomer compositions of the thermal equilibrium and the PSS (365 nm and 452 nm) of 11 were determined via analytical HPLC (Table S1). Therefore 20 µL probes of a methanolic solution S44 of 11 (1 mM) were injected before and after 5 minutes of 365 nm or 1 minute of 452 nm irradiation. Isocratic elution with 60% MeOH and 40% water (+ 0.1% HCOOH) for 12 minutes at 1 mL•min −1 was applied and gave complete separation of the isomers. Integration of the peak areas at the isosbestic points (λ obs = 244 nm and 270 nm) gave the relative percentage of (E)-11 and (Z)-11, respectively ( Figure S5). Table S1. Isomer composition ((E)-11 / (Z)-11) at the thermal equilibrium (∆) and the PSS after 5 minutes of 365 nm and 1 minute of 452 nm radiation.

PSS thermal half-life of 11
The thermal half-life of 11 after 5 minutes of 365 nm irradiation was determined by UV/Visspectroscopy. Therefore, a 50 µM solution of 11 in 5% DMSO (v/v) in enzyme assay buffer was prepared and an initial spectrum of the thermal equilibrium recorded. After 5 minutes of irradiation with 365 nm the increase of absorbance at the thermal equilibrium maximum (λ = 324 nm) was followed over a period of 96 hours ( Figure S6). Absorbance at 324 nm was plotted as a function of time. A non-linear regression yielded the half-live of the PSS. Due to the slow rate of thermal isomerization only a short period could be measured. Therefore the limit value was set to 0.939 as this was the absorbance at 324 nm of 11 at the thermal equilibrium.

Photochemistry of 2b and 2f
LCMS-IT-TOF analysis of 2b Figure  Due to photoisomeriaztion at daylight (Z)-2f could be observed in the sample of the thermal equilibrium.
S55 1 H NMR analysis of 2b The relative percentage of (Z)-2b was determined by comparison of integrals of single proton signals in (E)-2b and (Z)-2b. Therefore 1 H NMR spectra of 2b (10 mM) in MeOD were recorded after various irradiation periods with 254 nm. The relative percentage of (Z)-2f was determined by comparison of integrals of single proton signals in (E)-2f and (Z)-2f. Therefore 1 H NMR spectra of 2f (10 mM) in MeOD were recorded after various irradiation periods with 254 nm.

S58
Computational details: Structures were preoptimized according to TD-DFT using PBE functional and the SVP basis set. The first ten electronic excitations were computed using the PBE0 hybrid-functional, approximate coupled-cluster (CC2) singles-and-doubles model and second-order algebraic diagrammatic construction ADC(2). Excited states were also computed using the larger def2-TZVP basis set. In order to investigate solvent effects on the absorption spectra a continuous solvation model (COSMO) was implemented with permittivity (ε0) of 62.14 F/m and a refractive index (Dr) of 1.3379, relating to an aqueous solution of 60% methanol (v/v).  Figure S16: Calculated absorptions spectra of (E)-2b-A and experimental absorption spectrum of (E)-2b.