bioRxiv2021
The crowded cellular environments provide ample opportunities for proteins to interact with bystander macromolecules, yet direct evidence, let alone residue-specific information, for such nonspecific binding is rare. Here, by combining NMR spectroscopy and atomistic modeling, we investigated how crowders influence the association equilibrium and kinetics of two protein partners, EIN and HPr. Ficoll-70 increases the EIN-HPr binding affinity whereas bovine serum albumin (BSA) decreases the affinity. The opposite effects of the two crowders are quantitatively explained by atomistic modeling, which shows that the stabilizing effect of Ficoll-70 arises from volume exclusion favoring the bound state. In contrast, the destabilizing effect of BSA arises from preferential soft interactions with the free state; notably, BSA has favorable electrostatic interactions with positively charged HPr residues within the EIN-binding site. Some of the residues from this site indeed experience significant chemical shift perturbation when titrated with BSA, while the relaxation rates of HPr backbone amides exhibit overall elevation. Furthermore, relaxation dispersion data indicate that Ficoll-70 and BSA both slow down the EIN-HPr association rate, but change the dissociate rate in opposite directions. The observations on kinetics are accounted for by two effects of the crowders: increasing the solution microviscosity and reshaping the EIN-HPr interaction energy surface. The kind of preferential interactions between BSA and HPr that leads to competition with EIN should be prevalent in cellular environments. Our NMR results and atomistic modeling provide benchmarks, at both qualitative and quantitative levels, for the effects of crowded cellular environments on protein-protein specific interactions. Significance StatementAlthough nonspecific binding of crowder macromolecules with functional proteins is likely prevalent in vivo, direct evidence is rare. Here we present NMR characterizations showing that bovine serum albumin preferentially interacts with a specific binding site on HPr, leading to competition with the latters partner EIN. The preferential interactions result in destabilization of the EIN-HPr native complex and speedup of its dissociation, contrary to expectations from excluded-volume and viscosity effects. Atomistic modeling of macromolecular crowding rationalizes the experimental observations, and provides qualitative and quantitative insight into the influences of the crowded cellular environment on protein-protein specific interactions. Our work also has implications for evolution, regarding how nonspecific binding can be either minimized or exploited for gaining new functions.