Macromolecular Crowding by Evolved and Designed Proteins

Most protein studies are conducted in solutions of <0.1 g/L macromolecular cosolutes, whereas the cytoplasmic concentration can surpass 300 g/L. The consequent macromolecular crowding affects protein structure, stability, and interactions via both hard (steric) repulsions and chemical interactions. Although classic crowding theory only considers hard repulsions, growing evidence shows the key role of chemical interactions. Polyethylene glycols and sugar polymers have recently been used to produce a model that explains how crowding affects the free energy, enthalpy, and entropy of protein stability and dimer formation. However, cells are crowded with proteins, not these more inert polymers. To improve our understanding of proteins in physiologically relevant environments, I study crowding by natural and designed proteins. In Chapter 1, I review the history and applications of protein hydrogen- deuterium exchange (HDX) nuclear magnetic resonance spectroscopy (NMR), a popular tool for studying protein stability at the residue level. In Chapter 2, I employ HDX NMR to investigate the effects of protein self-crowding using the B1 domain of streptococcal protein G (GB1) as a function of concentration and temperature. I find that self-crowding stabilizes some residues but destabilizes others, contradicting predictions. Temperature-dependence data show that stabilization can arise enthalpically, not just entropically. Concentration-dependence data show that destabilization often increases with increasing concentration. These results show a key role for chemical interactions. In Chapter 3, I utilize 19F-NMR to study the concentration- and temperature-dependence of Src homology 3 domain (SH3) stability when crowded by a series of designed proteins with varying molecular weights and similar surface chemistry. I show that these designed protein crowders (DPCs) are highly soluble, well-folded, and thermostable. Crowding of SH3 by DPCs is stabilizing in a size- and concentration-dependent manner, with a stabilizing enthalpic component and a destabilizing entropic contribution. This DPC system will be used to test the broad applicability of our crowding model.