Protein denaturation with guanidine hydrochloride or urea provides a different estimate of stability depending on the contributions of electrostatic interactions

The objective of this study was to address the question of whether or not urea and guanidine hydrochloride (GdnHCl) give the same estimates of the stability of a particular protein. We previously suspected that the estimates of protein stability from GdnHCl and urea denaturation data might differ depending on the electrostatic interactions stabilizing the proteins. Therefore, 4 coiled-coil analogs were designed, where the number of intrachain and interchain electrostatic attractions (A) were systematically changed to repulsions (R): 20A, 15A5R, 10A10R, and 20R. The GdnHCl denaturation data showed that the 4 coiled-coil analogs, which had electrostatic interactions ranging from 20 attractions to 20 repulsions, had very similar [GdnHCl]1/2 values (average of congruent to 3.5 M) and, as well, their delta delta Gu values were very close to 0 (0.2 kcal/mol). In contrast, urea denaturation showed that the [urea]1/2 values proportionately decreased with the stepwise change from 20 electrostatic attractions to 20 repulsions (20A, 7.4 M; 15A5R, 5.4 M; 10A10R, 3.2 M; and 20R, 1.4 M), and the delta delta Gu values correspondingly increased with the increasing differences in electrostatic interactions (20A-15A5R, 1.5 kcal/mol; 20A-10A10R, 3.7 kcal/mol; and 20A-20R, 5.8 kcal/mol). These results indicate that the ionic nature of GdnHCl masks electrostatic interactions in these model proteins, a phenomenon that was absent when the unchanged urea was used. Thus, GdnHCl and urea denaturations may give vastly different estimates of protein stability, depending on how important electrostatic interactions are to the protein.