Competition between protein folding and aggregation: A three-dimensional lattice-model simulation

Aggregation of protein molecules resulting in the loss of biological activi ty and the formation of insoluble deposits represents a serious problem for the biotechnology and pharmaceutical industries and in medicine. Considera ble experimental and theoretical efforts are being made in order to improve our understanding of, and ability to control, the process. In the present work, we describe a Monte Carlo study of a multichain system of coarse-grai ned model proteins akin to lattice models developed for simulations of prot ein folding. The model is designed to examine the competition between intra molecular interactions leading to the native protein structure, and intermo lecular association, resulting in the formation of aggregates of misfolded chains. Interactions between the segments are described by a variation of t he Go potential [N. Go and H. Abe, Biopolymers 20, 1013 (1981)] that extend s the recognition between attracting types of segments to pairs on distinct chains. For the particular model we adopt, the global free energy minimum of a pair of protein molecules corresponds to a dimer of native proteins. W hen three or more molecules interact, clusters of misfolded chains can be m ore stable than aggregates of native folds. A considerable fraction of nati ve structure, however, is preserved in these cases. Rates of conformational changes rapidly decrease with the size of the protein cluster. Within the timescale accessible to computer simulations, the folding-aggregation balan ce is strongly affected by kinetic considerations. Both the native form and aggregates can persist in metastable states, even if conditions such as te mperature or concentration favor a transition to an alternative form. Refol ding yield can be affected by the presence of an additional polymer species mimicking the function of a molecular chaperone; (C) 2001 American Institu te of Physics.