In silico study of a new class of DNA fluorescent probes: docking, molecular dynamics and quantum chemistry calculations.

contextNatural fluorescence in biochemical structures can result in unsatisfactory outcomes in cell imaging. This drawback can be addressed by using probe molecules that fluoresce via Excited State Intramolecular Proton Transfer (ESIPT), thereby improving resolution and the signal-to-noise ratio. A docking study was conducted to estimate the binding free energy of 132 benzazoles and to determine their binding modes (minor groove (MG) or intercalation (INT)) to DNA, comparing them with the commercial probes 4',6-Diamidino-2-phenylindole dihydrochloride (DAPI) and acridine orange (AO). Atomistic molecular dynamics simulations under physiological conditions were performed for the five top-ranked ligands, as well as for AO and DAPI. Three of the investigated ligands exhibited stronger binding energy than both commercial probes, while the other two showed stronger binding energy than AO, but not than DAPI. The proposed benzazoles acted as intercalators and MG binders. Using molecular mechanics Poisson-Boltzmann surface area (MM/PBSA) calculations to estimate the binding free energy of the ligand-receptor complexes, we could confirm the strong interaction of these molecules with DNA. Quantum Chemical Calculations were performed to estimate the emission energies upon excitation of the selected ligands, which showed large Stokes shift values and, for some molecules, favorable ESIPT processes. Benzazoles 1-5 demonstrated strong interactions with DNA, surpassing the commercial probes in binding strength and displaying promising photophysical properties. Consequently, these ligands are promising fluorescent DNA probes, suitable for various diagnostic techniques.

methods132 Benzazole ligands were constructed and optimized using Gabedit and docked to a B-DNA oligomer using AutoDock. Molecular dynamics simulations were run in GROMACS, using AMBER-parmbsc0 and GAFF force fields. Quantum chemical calculations (TD-DFT) in Gaussian 16 provided excited-state properties, optimized at the CAM-B3LYP/cc-pVDZ level of theory.