Non-native aggregation of alpha-chymotrypsinogen A from a combined experimental and modeling approach

Non-native aggregation is the process in which protein monomers self-assemble into aggregates in which the constituent monomers have significantly non-native structure. These aggregates range in size and morphology, but are commonly rich in beta-sheet content. Non-native aggregates and their upstream intermediates have been implicated in a number of debilitating diseases including Alzheimer's, Huntington's, Parkinson's, and Creutzfield-Jacob disease. In the pharmaceutical industry, protein aggregation is one of the major impediments to rapid product development, affecting product purification, formulation, filling, shipment, storage, and delivery.;The mechanism of protein aggregation has been the focus of extensive research. In spite of this effort, details of the intermediate stages of protein aggregation typically remain elusive due to difficulties in experimentally isolating and characterizing the transient intermediates of the aggregation pathway. Mathematical models of protein aggregation provide a means to infer mechanistic details from experimental aggregation data. Knowledge of details of the aggregation mechanism can serve as a foundation for future approaches to predict aggregation rates and identify ways to rationally control aggregation processes.;A goal of this thesis is to develop a combined modeling and experimental approach to discern details of non-native protein aggregation mechanisms. The Lumry-Eyring nucleated polymerization (LENP) model was adapted from simpler extended Lumry-Eyring models and validated against experimental alpha-chymotrypsinogen aggregation data. An attractive feature of the LENP model is that its outputs are experimentally tractable quantities such as extent of conversion, average aggregate size, and aggregate size distribution as functions of time. Several distinguishing features of this model are: (1) it captures a variety of experimentally observed aggregation behaviors within the same model; (2) it self-consistently and simultaneously treats the contributions from unfolding, self-association, and growth; (3) it and its predecessor are the only models to treat the effects of aggregate solubility on observed kinetics.;The thermodynamics and kinetics of unfolding of the model protein alpha-chymotrypsinogen (aCgn) were monitored with differential scanning calorimetry, equilibrium and stopped-flow fluorescence, and circular dichroism (CD). Size exclusion chromatography, seeding experiments, CD, light scattering, isothermal and temperature-jump monomer loss, and calorimetry were used to probe different stages of aggregation. The data were analyzed self-consistently in terms of the LENP model.;Key results of the analysis are summarized here. Aggregation of aCgn can be described by a nucleated polymerization mechanism in which the irreversible structural conversion of monomers to beta-sheet either is part of or occurs soon after the rate-limiting step for locking monomers into aggregate. The observed aggregation rate coefficient and the nucleus size are highly temperature and protein concentration-dependent. The intrinsic aggregation rate coefficient has a negative activation energy that can be attributed to the nucleation step, while the intrinsic growth timescale is temperature-independent. Aggregate formation is enthalpically favored at elevated temperatures despite the highly repulsive electrostatics, and is inferred to be stabilized by hydrogen bonding and favorable enthalpic contributions involved in forming the beta-sheet aggregate structure.;The LENP model was successfully applied to determine details of the aCgn aggregation mechanism. The combined modeling and experimental approach developed and employed in this thesis is expected to be of general utility in interpreting aggregation behaviors for a broad range of experimental systems beyond alpha-chymotrypsinogen.