Protein folding ‐ seeing is deceiving

Abstract

This Perspective is intended to raise questions about the conventional interpretation of protein folding. According to the conventional interpretation, developed over many decades, a protein population can visit a vast number of conformations under unfolding conditions, but a single dominant native population emerges under folding conditions. Accordingly, folding comes with a substantial loss of conformational entropy. How is this price paid? The conventional answer is that favorable interactions between and among the side chains can compensate for entropy loss, and moreover, these interactions are responsible for the structural particulars of the native conformation. Challenging this interpretation, the Perspective introduces a proposal that high energy (i.e., unfavorable) excluding interactions winnow the accessible population substantially under physical-chemical conditions that favor folding. Both steric clash and unsatisfied hydrogen bond donors and acceptors are classified as excluding interactions, so called because conformers with such disfavored interactions will be largely excluded from the thermodynamic population. Both excluding interactions and solvent factors that induce compactness are somewhat nonspecific, yet together they promote substantial chain organization. Moreover, proteins are built on a backbone scaffold consisting of α-helices and strands of β-sheet, where the number of hydrogen bond donors and acceptors is exactly balanced. These repetitive secondary structural elements are the only two conformers that can be both completely hydrogen-bond satisfied and extended indefinitely without encountering a steric clash. Consequently, the number of fundamental folds is limited to no more than $10,000 for a protein domain. Once excluding interactions are taken into account, the issue of "frustration" is largely eliminated and the Levinthal paradox is resolved. Putting the "bottom line" at the top: it is likely that hydrogen-bond satisfaction represents a largely under-appreciated parameter in protein folding models.

DMSO-Quenched H/D-Exchange 2D NMR Spectroscopy and Its Applications in Protein Science

“…The problem of protein folding has been described with reference to the Levinthal paradox, in which the initial unfolded state is assumed to be a random coil, and hence, there may exist an astronomically large number of conformations, inaccessible in a reasonable time by a random search, at the beginning of the folding reactions [137][138][139]. Solving the Levinthal paradox is a fundamental problem in folding studies [140][141][142][143][144][145][146]. The presence of the residual structure, if any, in the unfolded state thus invalidates the Levinthal paradox, because such residual structure may form a folding initiation site and guide the subsequent folding reactions.…”

Some Clues about Enzymes from Psychrophilic Microorganisms

“…Indeed, both Kanaya and co-workers on ribonuclease HI from the psychrotrophic bacterium Shewanella oneidensis MR-1 [94], and Feller and co-workers on the α-amylase from the Antarctic bacterium P. haloplanktis [95], have shown that, inserting in the cold-active protein some selected residues present in the mesophilic counterpart, thus allowing the formation of additional attractive interactions among side chains, the protein thermal stability increases Apart the two previous entropic contributions, it is necessary to account for an energetic term that differentiates the two macro-states; ∆E a = [E a (D-state) − E a (N-state) + ∆E(intra)] is the difference in energetic interactions among the D-state and the N-state and surrounding water molecules, and the difference in intra-chain energetic interactions between the D-state and the N-state [81]. This energetic contribution should not be large because: (a) the H-bonds that peptide groups make with water molecules in the D-state are largely reformed as intra-protein H-bonds in the secondary structure elements of the Nstate; (b) side chains able to make H-bonds usually form them both with water molecules in the D-state, and intra-molecularly in the N-state; a H-bond satisfaction principle holds [92]; (c) van der Waals attractions between protein groups and water molecules are regained as intra-protein contacts in the close-packed interior of the N-state. In particular, a complete balance between protein-water energetic attractions in the D-state and N-state, respectively, and the intra-protein energetic attractions implies that ∆E a = 0.…”

Reframing the Protein Folding Problem: Entropy as Organizer

“…The surviving conformations for a blocked monopeptide correspond to ϕ,ψ-basins from which the major hydrogen-bonded backbone structures originate: α-helices, β-strands, and β-turns (Figure ). Perhaps counterintuitively, the stiff repulsive force favors structure formation by excluding alternative conformations, with consequent enhancement of both secondary structure and supersecondary structure …”

“…The surviving conformations for a blocked monopeptide correspond to ϕ,ψ-basins from which the major hydrogen-bonded backbone structures originate: αhelices, β-strands, and β-turns (Figure 4). Perhaps counterintuitively, the stiff repulsive force favors structure formation by excluding alternative conformations, 61 with consequent enhancement of both secondary structure and supersecondary structure. 11 A scatter plot of backbone ϕ,ψ-angles from 112 ultra-highresolution (≤1 Å) proteins is shown; 87% of the 18972 backbone angles fall within four major basins: β-stands (top left), polyproline II (top right), α R (bottom left), and α L (bottom right).…”

“…67 Points in the scatter plots cluster around their best-fit lines closely: ∼50% are within 5°of their respective best-fit line, and ∼85% are within 15°. The harsh repulsive force responsible for the Ramachandran plot 35 excludes alternative conformations, 61 with consequent enhancement of these secondary structures and their supersecondary structure assemblies, 11 αα, ββ, and βαβ.…”