Folding of a predominantly .beta.-structure protein: rat intestinal fatty acid binding protein

Abstract
The equilibrium and kinetic properties of the unfolding-refolding transitions of Escherichia coli derived rat intestinal fatty acid binding protein have been examined using several different denaturants. This protein, which contains 2 tryptophans but no prolines or cysteines, has a predominantly .beta.-structure: its 10 antiparallel .beta.-strands are organized into 2 orthogonal sheets surrounding a large solvent-filled internal cavity. For urea and guanidine hydrochloride, the completely reversible transition was monitored by circular dichroism, absorbance, and fluorescence spectroscopy. Each of these data sets was best fit by a simple, two-state model involving only native and unfolded forms. However, linear extrapolation to determine the free energy of folding in the absence of denaturant resulted in different values for the free energy of folding depending upon which denaturant was used. When fluorescence was used to monitor the transition, the extrapolated free energy estimates for the two denaturants were markedly different: 10.03 .+-. 0.24 kcal mol-1 for urea versus 5.22 .+-. 0.33 kcal mol-1 for guanidine hydrochloride. The midpoints of these transitions were 5.51 and 1.36 M, respectively. The transition caused by either denaturant as monitored by circular dichroism and absorbance spectroscopy was virtually coincident with that monitored by fluorescence, further supporting the assignment of a two-state model for the equilibrium results. The addition of a 2-fold molar excess of ligand (oleate) increased the extrapolated estimates approximately 2.5 kcal mol-1 for both denaturants. Stop-flow kinetic studies of the guanidine hydrochloride induced transitions showed that the unfolding and folding processes are rapid (complete in less than a minute under the slowest conditions) and complex, indicating the presence of intermediates on both pathways. The unfolding process was biphasic: the two phases accounted for the entire amplitude of the transition, suggesting that some type of sequential unfolding of the protein occurs. Two phases were also observed for the folding reaction, but at least one additional phase must occur in the dead time of the instrument (< 10 ms) to account for the expected amplitude change. The changes in the amplitudes of the folding phases suggest the existence of multiple pathways for the folding of rat intestinal fatty acid binding protein and that the proportion of protein molecules following any particular pathway is dependent on the final denaturant concentration.