Predicting conformational ensembles of the intrinsically disordered protein α-Synuclein via molecular simulation

Intrinsically disordered proteins (IDPs) represent 30% of the human genome. They are frequently involved in a variety of diseases including neurodegenerative disorders, cancer and cardiovascular diseases. Interfering with the function of disease-associated IDPs offers a highly attractive objective for drug development. Unfortunately, the highly dynamic nature of IDPs, the presence of continuous local and global conformational rearrangements, the transient secondary and long-range tertiary structure pose serious limits to experimental approaches aimed at designing rationally new drugs. Computational methods may be of great help to characterize IDPs and their binding to drugs, yet they also faces critical challenges in sampling the wide IDPs’ conformational space, with force field and analysis tools tailored for structured proteins. My thesis addresses IDPs computational issues by adopting apt computational protocols, combined with new analysis tools developed here. I focus on human alpha-synuclein (AS), a presynaptic protein of poorly defined function, whose aggregations constitute the main component of Parkinson disease-associated Lewy bodies. The first aim of this thesis is to characterize the structural determinants of the binding between AS and an anti-aggregation ligand, dopamine (DOP), which has been shown to prevent aggregation. We developed a new conformational analysis tool based on directional statistics. The tool not only is able to detect and quantitatively analyze protein residues’ flexibility and backbone conformational transitions, it is also able to quantitative detect the effects of the ligand on the dynamical spectra of the protein, avoiding problems associated with usual computational analysis tools based on Cartesian coordinates, which are biased by the problematic alignment of extremely flexible structures, besides being affected by the movements of neighboring residues. This tool provided a rationale for experimentally observed changes in chemical shifts on passing from free AS to ASDOP adducts, distinguishing variations arising from conformational rearrangements of residues with respect to the ones due to direct contacts with the drug. The second aim of the thesis is to adapt an enhanced sampling method to investigate the changes of conformational ensemble upon a physiologically relevant chemical modification, the N-term acetylation, recently revealed in human cells. The degree of sampling of our simulations and the quality of our models, in the first and second project, has been carefully validated against experimental data. Structural experimental quantities have been successfully reproduced. The methods proposed in my thesis provide a solid computational framework, which could be of great help for the characterization of the structural determinants of IDPs of pharmacological relevance also in the presence of drugs. Computational and semi-quantitative characterizations methods, like the ones adopted here, may provide crucial insights for future therapeutic strategies in drug design