The aggregation of the amyloid-beta peptide (A-beta) into fibrils and successively in plaques is the main event in Alzheimer's disease. The A-beta originates from a membrane protein called amyloid-precursor protein. In healthy brain A-beta is present as soluble form and the predominant chains are formed by 40 or 42 amino acids. The oligomers of A-beta, rather then fully formed fibrils, are suspected to have nerotoxic properties. The transition metal zinc, copper, iron are present in amyloid plaques in abnormally elevated concentrations. It has been experimentally observed that Zn(II) and Cu(II) ions bind at A-beta(1-16) N-terminal sequence and influence aggregation behaviour: in particular Zn triggers and accelerates aggregation and Cu slows down or accelerates aggregation, dependently on conditions. Moreover A-beta bonded to Cu or Fe is able, probably by reducing oxygen, to produce reactive-oxygen species that are widely accepted to play a key role in most of neurodegenerative diseases. Afterwards there are several open questions about the metals/A-beta interaction: coordination chemistry, reactivity of these complexes, different behaviour of metals, peptide structure prone to aggregation, mechanisms of metal-induced aggregation. We propose to use computer simulations (both semiempirical and first principles) to understand the metal/A-beta system. Quantum mechanics methods (DFT) will be used to investigate the nature of metal bond and reactivity but statistical mechanics methods will be also necessary to obtain accessible peptide conformations in water. The Car-Parrinello method, that was successfully applied to similar systems in applicant's previous work, will be used. The project includes the acquisition of new relevant experiments on the kinetics of ROS production catalysed by Cu and Fe complexes of Abeta. Experiments will be based on the expertise of the host institution and specific training of the applicant is planned.