Exploring the energy surface of protein folding by structure-reactivity relationships and engineered proteins: Observation of Hammond behavior for the gross structure of the transition state and anti-Hammond behavior for structural elements for unfolding/folding of barnase

Abstract

The structure of alpha-helix 1 (residues 6-18) in the transition state for the unfolding of barnase has been previously characterized by comparing the kinetics and thermodynamics of folding of wild-type protein with those of mutants whose side chains have been cut back, in the main, to that of alanine. The structure of the transition state has now been explored further by comparing the kinetics and thermodynamics of folding of glycine mutants with those of the alanine mutants at solvent-exposed positions in the alpha-helices of barnase. Such "Ala-->Gly scanning" provides a general procedure for examining the structure of solvent-exposed regions in the transition state. A gradual change of structure of the transition state was detected as helix 1 becomes increasingly destabilized on mutation. The extent of change of structure of helix 1 in the transition state for the mutant proteins was probed by a further round of Ala-->Gly scanning of those mutants. Destabilization of the helix 1 was found to cause the overall transition state for unfolding to become closer in structure to that of the folded protein. This is analogous to the conventional Hammond effect in physical-organic chemistry whereby the transition state moves parallel to the reaction coordinate with change in structure. But, paradoxically, the structure of helix 1 itself becomes less folded in the transition state as helix 1 becomes destabilized. This is analogous, however, to the rarer anti-Hammond effect in which there is movement perpendicular to the reaction coordinate. These observations are rationalized by plotting correlation diagrams of degree of formation of individual elements of structure against the degree of formation of overall structure in the transition state. There is a relatively smooth movement of the degree of compactness in the transition state against changes in activation energy on mutation that suggests a smooth movement of the transition state along the energy surface on mutation rather than a switch between two different parallel pathways. The results are consistent with the transition state having closely spaced energy levels. Helix 1, which appears to be an initiation point and forms early in the folding of wild-type protein, may be radically destabilized to the extent that it forms late in the folding of mutants. The order of events in folding may thus not be crucial.

Testing protein-folding simulations by experiment: B domain of protein A

“…Under these conditions, ⌽ values report back on the interactions made by the deleted side chains in the wild-type protein, and there is an approximate relationship between ⌽ and the extent of formation of the hydrophobic contacts made. The B domain is exceptionally well suited for a ⌽-value analysis, having the appropriate hydrophobic residues for making the nondisruptive deletions (17) in its core as well as many positions for Ala 3 Gly scanning of secondary structure (18), and there were adequately large changes of ⌬G D-N (⌬⌬G D-N ) on mutation. Very accurate rate constants could be obtained from temperature-jump kinetics.…”

“…Mutation of Ala 3 Gly at surface-exposed positions in helices provides an exquisite probe for the extent of formation of ␣-helical structure (18). The interactions changed on mutation of Ala 3 Gly are purely intrahelical in the native structure, and their energetics are proportional to the change in solvent-accessible surface area of the native helix on mutation (19,20) although the ⌽-value analysis does not depend on this.…”

“…Hydrophobic side chains from residues 13,14,17,18,20,23,31,32,35,45,46, 49, and 52 form an extended hydrophobic core, most of which could be mutated to give accurate ⌽ values, and most of these ⌽ values were significant (Table 3). I17 and L18 in the C terminus of H1, which interact with I32 and L35 in H2, had ⌽ Ϸ 0.8-1, and I32 and L35 also had high ⌽ values, indicating strong interactions between H1 and H2.…”

Point Mutations in Membrane Proteins Reshape Energy Landscape and Populate Different Unfolding Pathways

“…26 This principle from physical organic chemistry was first shown experimentally by Fersht and coworkers for protein folding transition states in the small ribonuclease barnase. 27,28 However, for membrane proteins this principle has not been shown yet. k u obtained from DFS was used to estimate the free energy difference, ΔG u *, between the folded and the transition states of a structural segment.…”

Kinetic Role of Helix Caps in Protein Folding Is Context-Dependent

“…Similar substitutions in barnase showed Gly C-caps to be more stable than Ala by 1.13 ( 0.04 kcal/mol at the C-cap of the first helix and 3.15 ( 0.04 kcal/mol for the C-cap of the second helix. The latter site of substitution is unusual in having a φ, ψ of +115°, +6°that is much more unfavorable for the Ala than the usual LR values near +60°, +35°found for C-caps in λ 6-85 and elsewhere (52). In the barstar protein, however, the Gly C-cap in the first helix is 1.3 kcal/mol less stable than the wild-type Ala (5); we do not see an explanation for that unusual behavior.…”