Photochemistry of 1,3-dicarbonyl compoundsdna photodamage vs. Photoprotection

Abstract

The main objective of this thesis is to contrast the role of these 1,3-dicarbonyl compounds as DNA damaging agents to their photoprotective potential. Firstly, 5,6-dihydropyrimidines have been derivatized using a tert-butyl ketone photolabile group in order to study the generation of C5-centered radicals in non aqueous media. Then, laser flash photolysis study in acetonitrile of the designed 1,3-dicarbonyl derivatives yields the formation of the purported 5,6-dihydropyrimidin-5-yl radicals. Their characterization shows long lived transient species, which do not decay in the µs range and are centered at 400-420 nm or 350-400 nm for the 5,6-dihydrouridine or 5,6-dihydrothymidine derivatives, respectively. Moreover, radical generation has also been evidenced by steady state fluorescence experiments by using a profluorescent radical trap (AAA-TEMPO). Thus, irradiation of the photolabile nucleic acid derivatives in the presence of AAA-TEMPO results in an increased emission, in agreement with the trapping of C5 radical by the paramagnetic probe. Formation of the resulting adduct has been confirmed by UPLC-HRMS. Experimental data have been corroborated with ab initio CASPT2//CASSCF theoretical calculations. In a second chapter, another 1,3-dicarbonyl derivative of pyrimidine has been investigated. Indeed, 5-formyluracil (ForU) presents interesting features as a potential intrinsic DNA photosensitizing agent. Thus, spectroscopic studies reveal that ForU has not only an absorption in the UVA/UVB range, but also a triplet excited state (3ForU*) with a lifetime of some micros and with an energy high enough to photosensitize the well-known cyclobutane pyrimidine dimers (CPDs) through triplet-triplet energy transfer. This process has been confirmed by means of the synthesis of model Thy-Thy and Cyt-Cyt dyads, which after irradiation in the presence of ForU have been demonstrated to produce CPDs. Finally, the study extended to plasmid DNA allows establishing the ability of ForU to produce single strand breaks and CPDs. Next, the attention has been focused on the development of a new strategy for photoprotection of bioactive molecules taking advantage of the photochemical reactivity of the 1,3-diketo tautomer of the UVA filter avobenzone (AB). The selected bioactive compounds are two photosensitive topical non steroidal anti-inflammatory drugs, (S)-ketoprofen (KP) and diclofenac (DF). In this context, the diketo tautomer of avobenzone contains two phenacyl moieties, which are well-known photoremovable protecting groups. Thus, a judicious design of a pro-drug/pro-filter dyad allows the photorelease of the drug and its protecting shield, avobenzone. The viability of this controlled release of the active ingredients was checked in different solvents of different H donating properties and viscosity to simulate topical formulation.Plus, laser flash photolysis studies in ethanol allow characterization of a transient absorption band at 400-420 nm assigned to the triplet excited state of the dyad by comparison with that of the diketo form of AB. Finally, the photosafety of the photoactivatable AB-KP dyad has been assessed. The transient absorption spectra obtained for AB-KP dyad in cyclohexane showed the triplet excited state of KP and not that of the AB in its diketo form. The impact on the cellular membrane has been addressed by UVA irradiation of linoleic acid solutions in the presence of the dyad. Phototoxic potential of the dyad has been evidenced by UV-Vis spectrophotometry through the formation of the conjugated dienic hydroperoxides derived from linoleic acid. However, AB-KP does not exhibit a photogenotoxic potential as demonstrated by comet assay experiments, where by contrast with KP, the non damaged round shape of the cell is still observed after UVA irradiation.