Theoretical and Computational Studies of Biophysical Phenomena: Mechanical Stability of Proteins, and Origin of Biological Homochirality

Abstract

In the first part of this dissertation, we develop a new formalism for computing local mechanical stresses in order to understand the stabilization mechanism of proteins in glassy carbohydrate-water matrices. To our knowledge, this is the first such computation of local mechanical stresses in molecular simulation that accommodates electrostatic lattice sums, many-body interactions and non-planar interfaces. We demonstrate the formalism's usefulness through selected results on ubiquitin and an α-keratin

fragment. We find that protein-level normal stresses increase upon vitrification, and that both proteins experience compressive stresses of the order of 100 bar in the glassy state.

In the second part of this dissertation, extensive protein folding simulations are used to explore the stability of proteins at negative pressure. Although hot, cold and high pressure denaturation are well

characterized, the possibility of negative pressure unfolding has received much less attention. Proteins under negative pressure, however, are important in applications such medical ultrasound, and the survival of biopoloymers in the xylem of vascular plants. We use extensive replica-exchange molecular dynamics (REMD) simulations and thermodynamic analysis to obtain folding/unfolding equilibrium phase diagrams for the miniproteins trp-cage (α-structure, 20-residue), GB1 β-hairpin (β-structure, 16- residue) and AK16 peptide (α-helix, 16 residue). While trp-cage is destabilized by negative pressure, GB1 β-hairpin and AK16 peptide are stabilized by this condition.

Finally, an elementary lattice model is formulated to simulate the kinetics of chiral symmetry breaking via autocatalysis and inhibition in a mixture of non-chiral reactants, chiral products and inert solvent. Starting from a chirally unbiased initial state, spontaneous symmetry breaking occurs in spite of equal a

priori probability for creating either product enantiomer. The processes of reaction and diffusion are kinetically intertwined in a way leading to competition in the creation of a symmetry-broken outcome. The model exhibits two modes of symmetry breaking: in the absence of inhibition, reactions are predominantly autocatalytic under both reaction control (fast diffusion, slow reaction) or diffusion

control (fast reaction, slow diffusion) conditions. In the presence of inhibition, simulations with different system sizes converge to the same transition temperature under reaction control conditions,

and in this limit the reactions are predominantly non-autocatalytic.