Backbone and side-chain contributions in protein denaturation by urea

Abstract

Urea is a commonly used protein denaturant, and it is of great interest to determine its interaction with various protein groups to elucidate the molecular basis of its effect on protein stability. Using the Trp-cage miniprotein as a model system, we report what we believe to be the first computation of changes in the preferential interaction coefficient of the protein upon urea denaturation from molecular-dynamics simulations and examine the contributions from the backbone and the side-chain groups. The preferential interaction is obtained from reversible folding/unfolding replica exchange molecular-dynamics simulations of Trp-cage in presence of urea, over a wide range of urea concentration. The increase in preferential interaction upon unfolding is dominated by the side-chain contribution, rather than the backbone. Similar trends are observed in simulations using two different force fields, Amber94 and Amber99sb, for the protein. The magnitudes of the side-chain and backbone contributions differ in the two force fields, despite containing identical protein-solvent interaction terms. The differences arise from the unfolded ensembles sampled, with Amber99sb favoring conformations with larger surface area and lower helical content. These results emphasize the importance of the side-chain interactions with urea in protein denaturation, and highlight the dependence of the computed driving forces on the unfolded ensemble sampled.

Figure 1

Effect of urea on the folding equilibrium. (A and B) Fraction of folded states as a function of temperature for the various urea concentrations studied, shown for both Amber94 and Amber99sb systems.

Figure 2

Characterization of the unfolded ensembles. (A and B) Average solvent-accessible surface (SAS) area of the unfolded ensemble as a function of temperature for the various urea concentrations, shown for both Amber94 and Amber99sb systems. (C and D) Helical content as a function of temperature, shown for Amber94 and Amber99sb systems.

Figure 3

Local solvent distribution around the protein for 3.8 M urea system at 300 K. (A and B) Proximal radial distribution for urea and water around the folded and the unfolded ensembles of the protein, shown for both Amber94 and Amber99sb. (C and D) Preferential interaction, as a function of distance from the protein, for the folded and the unfolded ensembles. (E and F) Decomposition of preferential interaction into backbone and side-chain contributions.

Figure 4

Preferential interaction contributions in a given ensemble. (A) Folded ensemble in Amber94. (B) Unfolded ensemble in Amber94. (C) Folded ensemble in Amber99sb. (D) Unfolded ensemble in Amber99sb. (Solid squares and circles) Backbone and side-chain contribution, respectively.

Figure 5

Comparing the preferential interaction between folded and unfolded ensembles. (A) Γ^P in Amber94. (B) Γ^BB in Amber94. (C) Γ^SC in Amber94. (D) Γ^P in Amber99sb. (E) Γ^BB in Amber99sb. (F) Γ^SC in Amber99sb. (Solid squares and circles) Folded and unfolded ensembles, respectively.

Figure 6

Comparison of the direct interaction and hydrogen bonding. Coulomb (squares) and LJ (circles) contributions to ΔE^PU are shown (A) forAmber94 and (B) for Amber99sb. Comparison of Amber94 (squares) and Amber99sb (circles) is shown (C) for backbone-urea hydrogen bonding and (D) for backbone-water hydrogen bonding.