Aggregation of polyglutamine peptides with β-sheet structures is related to some important neurodegenerative diseases such as Huntington's disease. However, it is not clear how polyglutamine peptides form the β-sheets and aggregate. To understand this problem, we performed all-atom replica-exchange molecular dynamics simulations of one and two polyglutamine peptides with 10 glutamine residues in explicit water molecules. Our results show that two polyglutamine peptides mainly formed helix or coil structures when they are separated, as in the system with one-polyglutamine peptide. As the interpeptide distance decreases, the intrapeptide β-sheet structure sometimes appear as an intermediate state, and finally the interpeptide β-sheets are formed. We also find that the polyglutamine dimer tends to form the antiparallel β-sheet conformations rather than the parallel β-sheet, which is consistent with previous experiments and a coarse-grained molecular dynamics simulation.

Recently it has been proposed a model for fibrils of human insulin in which the fibril growth proceeds via stacking LVEALYL (fragment 11–17 from chain B of insulin) into pairs of tightly interdigitated -sheets. The experiments have also shown that LVEALYL has high propensity to self-assembly and binding to insulin. This necessitates study of oligomerization of LVEALYL and its binding affinity to full-length insulin. Using the all-atom simulations with Gromos96 43a1 force field and explicit water it is shown that LVEALYL can aggregate. Theoretical estimation of the binding free energy of LVEALYL to insulin by the molecular mechanic Poisson-Boltzmann surface area method reveals its strong binding affinity to chain B, implying that, in agreement with the experiments, LVEALYL can affect insulin aggregation via binding mechanism. We predict that, similar to LVEALYL, peptide RGFFYT (fragment B22-27) can self-assemble and bind to insulin modulating its fibril growth process. The binding affinity of RGFFYT is shown to be comparable with that of LVEALYL.

A novel direct phase-selection method to select optimized phases from the ambiguous phases of a subset of reflections to replace the corresponding initial SAD phases has been developed. With the improved phases, the completeness of built residues of protein molecules is enhanced for efficient structure determination.