Beyond the current progress, the next step will be MD simulations. We will use the General Atomic Force Field (GAFF) as implemented in Amber. Preliminary calculations on representative dimers shows that this force field describes our molecules correctly except for the rotation along the inter-monomer bond. Therefore, will be the re-parametrization of the GAFF parameters for rotation along these bonds, using corrected BLYP/6-311G** as the benchmark and a genetic algorithm for the parametrization. We will use our modified version of GAFF to simulate two levels of aggregation: protomolecules made of 5 oligomers (A5 aggregates), and subunits made of 30 oligomers (A30 aggregates), to reach a size of ~4 nm. For practical reasons, we will limit ourselves to homo-aggregates where every A5 or A30 system is made of a single oligomer component. We will start the periodic MD simulations with a high temperature phase to provide enough flexibility to reach the most stable aggregate conformation; after that we will lower the temperature to do the simulation under physiological temperature and pressure conditions. For each aggregate we will do two simulations, one in water solution and the other one of the (periodic) aggregate without solvent. The periodic structures aim at modelling the properties of large eumelanosomes. To create the final set of library structures, a cluster analysis of the MD simulations will be done to see if structural averaging is sufficient or more sophisticated sampling is needed. Based on this library, we will calculate the aggregate spectra to identify the contribution of the different aggregates to the spectra and establish structure-property relationships. Finally, we will study the decay mechanisms of the different oligomers of our library embedded in some of the A30 aggregates.
The expected result is in line with one of the main melanin-related challenges identified by the EuMelaNet special interest group of the European society for Pigment Cell Research (ESPCR), 21 the "need to expand the present set of structure-property-function relationships (...) to tailor melanins for specific applications". Our contribution on the role of different aggregates in the photoprotecting function and our derived structure property relationships will be important to understand melanin biological function and broaden the technological use of its synthetic analogues. Although we will not model chemiexcitation itself, our results will be also highly valuable to assess the possible pathogenic role of melanin chemiexcitation in future studies, because excited states must play a key role in this process. Our research will also be a starting point for future studies on other relevant properties such as antioxidant and free radical scavenger activity, generation of reactive oxygen species through photo- or chemiexcitation, bio-adhesion and coating properties, metal ion sequestration, or paramagnetic and semiconducting properties.
