28 January 2018 in The Journal of Chemical Physics
In this paper, we study the early stages of aggregation of a model of polyglutamine (polyQ) for different repeat lengths (number of glutamine amino acid groups in the chain). In particular, we use the Large-scale Atomic/Molecular Massively Parallel Simulator to study a generic coarse-grained model proposed by Bereau and Deserno. We focus on the primary nucleation mechanism involved and find that our results for the initial self-assembly process are consistent with the two-dimensional classical nucleation theory of Kashchiev and Auer. More specifically, we find that with decreasing supersaturation, the oligomer fibril (protofibril) transforms from a one-dimensional β sheet to two-, three-, and higher layer β sheets as the critical nucleus size increases. We also show that the results are consistent with several predictions of their theory, including the dependence of the critical nucleus size on the supersaturation. Our results for the time dependence of the mass aggregation are in reasonable agreement with an approximate analytical solution of the filament theory by Knowles and collaborators that corresponds to an additional secondary nucleation arising from filament fragmentation. Finally, we study the dependence of the critical nucleus size on the repeat length of polyQ. We find that for the larger length polyglutamine chain that we study, the critical nucleus is a monomer, in agreement with experiment and in contrast to the case for the smaller chain, for which the smallest critical nucleus size is four.
10 July 2015 in The Journal of Chemical Physics
We study the kinetics of aggregation of a two site model of interacting spherical molecules. A given site on one molecule can interact with one or more sites on other neighboring molecules. The sites represent the result of a simple coarse graining of putative amino acid residues or two specifically designed sites on a colloidal particle. We study the kinetics and equilibrium morphology for a fixed angle between the two sites, for several angles between 30° and 150°. In the model, the sites interact via an attractive Asakura-Oosawa potential and the molecules have the usual hard sphere repulsion interaction. We find a transition from a micelle-like morphology at small angles to a rod-like morphology at intermediate angles and to a gel-like structure at values of the angle greater than about ninety degrees. However, at 150 degrees, after a long induction time during which there is no aggregation, we observe a nucleation and growth process that leads to a final spherical-like aggregate. Our results show that this angle is a control parameter for the kinetics and equilibrium properties of the system.
28 August 2016 in The Journal of Chemical Physics
There has been considerable recent interest in the self-assembly and phase behavior of models of colloidal and protein particles with anisotropic interactions. One example of particular interest is amelogenin, an important protein involved in the formation of dental enamel. Amelogenin is primarily hydrophobic with a 25-residue charged C-terminus tail. This protein undergoes a hierarchical assembly process that is crucial to mineral deposition, and experimental work has demonstrated that the deletion of the C-terminus tail prevents this self-assembly. A simplified model of amelogenin has been proposed in which the protein is treated as a hydrophobic sphere, interacting via the Asakura-Oosawa (AO) potential, with a tethered point charge on its surface. In this paper, we examine the effect of the Coulomb interaction between the point charges in altering the phase diagram of the AO model. For the parameter case specific to amelogenin, we find that the previous in vitro experimental and model conditions correspond to the system being near the low-density edge of the metastable region of the phase diagram. Our study illustrates more generally the importance of understanding the phase diagram for proteins, in that the kinetic pathway for self-assembly and the resulting aggregate morphology depends on the location of the initial state in the phase diagram.