Simultaneous Determination of Structural and Thermodynamic Effects of Carbohydrate Solutes on the Thermal Stability of Ribonuclease A Thomas F. O’Connor, Pablo G. Debenedetti, and Jeffrey D. Carbeck* Department of Chemical Engineering, Princeton UniVersity, Princeton, New Jersey 08544 Received March 30, 2004; E-mail: jcarbeck@princeton.edu The production, processing, and utilization of proteins, in nature and in biotechnology, occur under nonideal solution conditions. Nature uses high concentrations of organic molecules to protect organisms from dehydration. 1 In vivo, biomolecular interactions occur in crowded intracellular environments. 2,3 In vitro, the addition of high concentrations of sugars increases the stability of the native state of proteins over denatured states. 4 This enhancement of stability typically increases with concentration and molecular weight of the sugar. The mechanisms by which sugars increase stability are not completely understood. This communication describes a study of the effects of fructose and sucrose on the thermal stability of ribonuclease A (RNase) using capillary electrophoresis (CE) and protein charge ladders, collections of proteins that differ incremen- tally in number of chemically modified charged groups. 5 Because this approach provides information on both the thermodynamics (i.e., the free energy, ΔG N-D , of denaturation) and structural changes (i.e., the effective hydrodynamic radius, R H , of proteins in both the native and denatured states) associated with stability, it allows a simple microscopic interpretation of the effects of sugars on the stability of RNase. We used CE to measure ΔG N-D of RNase at 25 °C and pH 8.4 in the presence of different concentrations of sucrose and fructose. 6-8 Specifically, the fraction of protein in the native and denatured states was determined as a function of temperature by measuring shifts in electrophoretic mobility. We used a two-state model of dena- turation, a reasonable approximation for RNase, 9 to determine ΔG N-D . Values of ΔG N-D and melting temperature, T m , are shown in Table 1. Values of T m increase modestly in the presence of the sugars: 4.5 °C for the addition of 0.8 M sucrose and 5.4 °C for the addition of 1.52 M fructose. Effects of sugar on values of ΔG N-D are significant. At the largest concentrations of sucrose (0.8 M) and fructose (1.52 M), ΔG N-D is 3 kcal/mol greater than in the absence of sugar, an increase of ∼36%. To further quantify the effects of sugars on stability, we calculated values of ΔΔG by subtracting values of ΔG N-D measured in buffer from values obtained in the presence of sugar; values of ΔΔG are plotted as a function of molar concentration in Figure 1. The results are qualitatively consistent with previous studies: 4 (i) values of ΔΔG increase approximately linearly with concentration of sugar; (ii) the larger sugar, sucrose, has a greater stabilizing effect on RNase than fructose at a given molar concentration. This second effect can be quantified by the gradient of ΔG N-D with concentra- tion of sugar, which yields values of 3.6 kcal/mol per mole of sucrose and 2.0 kcal/mol per mole of fructose. We used the combination of CE and charge ladders to measure values of R H of RNase in the native and denatured states. 7 Values of electrophoretic mobility of the rungs of the charge ladder were fit to Henry’s model of electrophoresis. The results are presented in Table 1. The value of R H for the native state in the absence of sugars was confirmed independently using pulse-field gradient NMR. 10 Thermal denaturation results in a change in R H of ∼6 Å, independent of the concentration of sugars. To rationalize the effects of sugars on the stability of RNase we used the simplest possible model of the free energy of solvation of proteins: contributions of sugars to the enthalpy of solvation are ignored; only effects of entropy are considered. This situation is described by scaled particle theory (SPT), 11 where the solution is modeled as a mixture of hard spheres of different size. Values of ΔΔG were calculated using SPT (solid lines in Figure 1). SPT predicts the free energy of solvation of the protein in terms of the work of forming a cavity in the solution large enough to * To whom correspondence should be addressed. Table 1. Effects of Carbohydrates on the Stability and Hydrodynamic Radius of RNase a Tm (°C) ΔGN-D (kcal/mol) RH,N (Å) RH,D (Å) buffer 62.6 (0.1) 8.2 (0.1) 21.2 (0.3) 27.5 (0.3) 0.15 M sucrose 63.6 (0.1) 9.1 (0.2) 0.31 M sucrose 64.7 (0.2) 9.5 (0.2) 21.9 (0.3) 28.2 (0.3) 0.47 M sucrose 65.9 (0.1) 10.2 (0.2) 0.63 M sucrose 66.5 (0.2) 10.6 (0.2) 22.2 (0.3) 28.0 (0.3) 0.80 M sucrose 67.1 (0.2) 11.1 (0.2) 0.29 M fructose 63.9 (0.1) 9.1 (0.2) 0.57 M fructose 64.5 (0.2) 9.5 (0.2) 21.7 (0.3) 28.0 (0.3) 0.89M fructose 66.0 (0.2) 10.1 (0.2) 1.21 M fructose 66.5 (0.1) 10.5 (0.2) 22.0 (0.3) 28.0 (0.3) 1.52 M fructose 68.0 (0.2) 11.2 (0.2) a Tm is the melting temperature, ΔGN-D is the free energy of denaturation of RNase at 25 °C, RH,N is the hydrodynamic radius of the protein in the native state and RH,D in the denatured state. Values in parentheses are estimates of standard error of the data. Figure 1. Values of ΔΔG, the difference in ΔGN-D measured in buffer and in the presence of sugars, for RNase at 25 °C as a function of molar concentration of sugars. Lines are predictions using scaled particle theory with radii of 1.38 Å for water, 3.25 Å for fructose, and 4.30 Å for sucrose. RH values of 21 Å for native RNase and 28 Å for denatured RNase were used for the hard sphere radii of the protein. All experiments were conducted at pH 8.4 in 25 mM Tris, 192 mM Gly, and 30 mM NaCl. Published on Web 09/04/2004 11794 9 J. AM. CHEM. SOC. 2004, 126, 11794-11795 10.1021/ja0481777 CCC: $27.50 © 2004 American Chemical Society