• AppA BLUF protein
• excited-state SCF
• extrapolation technique
• hybrid QM/MM molecular dynamics
• molecular dynamics
• multiscale polarizable QM/MM
• photoactivation
• photoreceptor
• polarizable embedding
• proton-coupled electron transfer

Blue-Light Using Flavin (BLUF) domains are a family of flavin-based photoreceptors, found in some bacteria. Their activation is induced by blue-light absorption and a sequence of conformational changes bring the photoreceptor to its active state, which is able to transmit signals over long distances. The proposed mechanism of activation of BLUF proteins involves the flavin and two amino acids, which are tyrosine and glutamine, and they are thought to undergo two Proton-Coupled Electron (PCET) transfer processes. Blue-light absorption brings the flavin to its locally-excited state (LE). Then, one electron moves from tyrosine to flavin, which passes to a charge-transfer (CT) state, and this is the driving force for a double proton transfer from tyrosine to flavin, mediated by glutamine (forward PCET). The next step is the rotation of the glutamine residue (which exists as an imidic acid group after the double proton transfer). In the end, the photoreceptor reaches its active state through the so-called reverse PCET: an electron transfer occurs from flavin to tyrosine, followed by a proton transfer from flavin to tyrosine, mediated by the imidic acid residue.
In this thesis, we present the investigation of the photoactivation mechanism of the AppA BLUF protein through hybrid QM/MM excited state molecular dynamics. Since the well-established linear response methods to calculate excited-state energy require a high computational cost, this study is also combined to the development of a new hybrid simulation technique, which is expected to be faster without a great loss of accuracy. This new methodology consists of the coupling of excited-state SCF approaches to a polarizable description of the environment and to molecular dynamics. We also adopt an extrapolation technique to further reduce the computational cost of simulations.
The new simulation method has been applied to study the photoactivation of the AppA photoreceptor.