No abstract

No abstract

Our understanding of the factors stabilizing alpha-helical structure has been greatly enhanced by the study of model alpha-helical peptides. However, the relationship of these results to the folding of helices in intact proteins is not well characterized. Helix propensities measured in model peptides are not in good agreement with those from proteins. In order to address these questions, we have measured helix propensities in the alpha-helix of ribonuclease T1 and a helical peptide of identical sequence. We have previously demonstrated excellent agreement between peptide and protein for the nonpolar amino acids [Myers, J. K., Pace, C. N., and Scholtz, J. M. (1997) Proc. Natl. Acad. Sci. U.S.A. 94, 2833-2837]. Most other amino acids also show good agreement, although certain polar amino acids are exceptions. Helix propensities measured in the ribonuclease T1 peptide/protein are compared with those measured in other systems. Reasonable agreement is found between most systems; however, our propensities differ substantially from those measured in several model peptide systems. Alanine-based peptides overestimate the propensity differences by a factor of 2, and host/guest experiments underestimate them by a factor of 2-3.

To investigate the pH dependence of the conformational stability of barnase, urea denaturation curves were determined over the pH range 2-10. The maximum conformational stability of barnase is 9 kcal mol-1 and occurs between pH 5 and 6. The dependence of delta G on urea concentration increases from 1850 cal mol-1 M-1 at high pH to about 3000 cal mol-1 M-1 near pH 3. This suggests that the unfolded conformations of barnase become more accessible to urea as the net charge on the molecule increases. Previous studies suggested that in 8 M urea barnase unfolds more completely than ribonuclease T1, even with the disulfide bonds broken [Pace, C.N., Laurents, D. V., & Thomson, J.A. (1990) Biochemistry 29, 2564-2572]. In support of this, solvent perturbation difference spectroscopy showed that in 8 M urea the Trp and Tyr residues in barnase are more accessible to perturbation by dimethyl sulfoxide than in ribonuclease T1 with the disulfide bonds broken.

On the basis of studies of Asn to Ala mutants, the gain in stability from burying amide groups that are hydrogen bonded to peptide groups is 80 cal/(mol A(3)). On the basis of similar studies of Leu to Ala and Ile to Val mutants, the gain in stability from burying -CH(2)- groups is 50 cal/(mol A(3)). Thus, the burial of an amide group contributes more to protein stability than the burial of an equivalent volume of -CH(2)- groups. Applying these results to folded proteins leads to the surprising conclusion that peptide group burial makes a larger contribution to protein stability than nonpolar side chain burial. Several studies have shown that the desolvation penalty for burying peptide groups is considerably smaller than generally thought. This suggests that the hydrogen bonding and van der Waals interactions of peptide groups in the tightly packed interior of folded protein are more favorable than similar interactions with water in the unfolded protein.