Early hydrogen-bonding events in the folding reaction of ubiquitin.

Abstract

The formation of hydrogen-bonded structure in the folding reaction of ubiquitin, a small cytoplasmic protein with an extended (3-sheet and an a-helix surrounding a pronounced hydrophobic core, has been investigated by hydrogendeuterium exchange labeling in conjunction with rapid mixing methods and two-dimensional NMR analysis. The time course of protection from exchange has been measured for 26 backbone amide protons that form stable hydrogen bonds upon refolding and exchange slowly under native conditions. Amide protons in the (3-sheet and the a-helix, as well as protons involved in hydrogen bonds at the helix/sheet interface, become 80% protected in an initial 8-ms folding phase, indicating that the two elements of secondary structure form and associate in a common cooperative folding event. Somewhat slower protection rates for residues 59, 61, and 69 provide evidence for the subsequent stabilization of a surface loop. Most probes also exhibit two minor phases with time constants of about 100 ms and 10 s. Only two of the observed residues, Gln-41 and Arg-42, display significant slow folding phases, with amplitudes of 37% and 22%, respectively, which can be attributed to native-like folding intermediates containing cis peptide bonds for Pro-37 and/or Pro-38. Compared with other proteins studied by pulse labeling, including cytochrome c, ribonuclease, and barnase, the initial formation of hydrogen-bonded structure in ubiquitin occurs at a more rapid rate and slowfolding species are less prominent.A central goal of experimental protein folding studies is to obtain structural and kinetic information on intermediate states between the disordered polypeptide chain of a denatured protein and the folded structure of a native protein. The existence of protein folding pathways with well-defined intermediates is now widely accepted (1, 2), but we know little about the nature of folding intermediates in precise structural terms. Because of the cooperative nature of foldingunfolding transitions, partially folded forms are rarely observed at equilibrium, and kinetic intermediates tend to be too short-lived for the direct application of high-resolution structural tools such as two-dimensional NMR.Our approach to this problem makes use of pulsed hydrogen-deuterium (H-D) exchange in conjunction with rapid mixing methods and two-dimensional NMR analysis (3, 4). In a typical pulse-labeling experiment, the protein is initially unfolded in a D20 denaturant solution so that the labile backbone amide protons (NH) are replaced by deuterons (ND). Refolding, initiated by rapid dilution of the denaturant, is allowed to proceed in D20 (or in H20 at mildly acidic pH, where NH exchange is slow) for various times before the partially refolded protein is briefly exposed to H20 at basic pH. Amide hydrogens that are still exposed at this time become protonated during the labeling pulse, while those that are already protected from exchange (e.g., by forming hydrogen bonds) remain deuterated. After the labeling pulse, the protein is allo...

A folding pathway for βpep‐4 peptide 33mer: From unfolded monomers and β‐sheet sandwich dimers to well‐structured tetramers

“…This is as fast as the folding of monomeric proteins where initial stages of folding occur on a time scale of less than about 5 ms and the final folding step to a well-structured state is rate limiting and occurs on a slower time. For example, native structure is formed in a few hundred ms in barstar (Nolting et al, 1997) and hen lysozyme (Radford et al, 1992), and in less than 20 ms in chymotrypsin inhibitor-2 (Jackson & Fersht, 1991) and less than 10 ms in ubiquitin (Briggs & Roder, 1992). With ppep-4, however, the folding process includes both dimer-dimer inter-molecular association and conformational re-shuffling to acheive the final folded state.…”

Protein secondary structural types are differentially coded on messenger RNA

“…In the partially folded state of hen egg white lysozyme in trifluoroethanol (Buck et al, 1993), only little protection from deuterium/hydrogen exchanges could be observed for the amides located in a long loop while a majority of the observed protection occurred in helical regions. Experiments on the refolding reaction of ubiquitin (Briggs and Roder, 1992) indicated that while the amide protons from the beta sheet and the alpha helix as well as the protons involved in hydrogen bonding at the helixkheet interface become 80% protected in an initial 8-ms folding phase, somewhat slower protection rates were observed for the residues from a surface loop. In the partially folded species of lactalbumin, native secondary structures are largely formed while the loop regions remain disordered (Feng et al, 1994;Redfield et al, 1994).…”

“…However, recent refolding experiments from denatured polypeptides of ribonuclease A (Udgaonkar and Baldwin, 1990), ribonuclease TI (Mul- lins et al, 1993). ubiquitin (Briggs and Roder, 1992), lysozyme , and staphylococcal nuclease (Jacobs and Fox, 1994) suggested that beta sheets can be formed at a rate comparable to that of alpha helices, and sometimes the residues in beta sheet are protected from hydrogeddeuterium exchange even before protection is observed in helices (Jacobs and Fox, 1994). However, the nuclease structures depend on several disulfide bridges and the helical domain of lysozyme as well as lactalbumin (Chyan et al, 1993) folds first.…”

Conformer-specific characterization of nonnative protein states using hydrogen exchange and top-down mass spectrometry

“…Ub populates a partially structured state (the socalled A-state) in the presence of alcohol in acidic solution at low salt (30), and this state was postulated to exist under a variety of other conditions (31), including those that favor correlated exchange regimes, where a distinction between this conformer and the native protein can be made based on markedly different levels of deuterium incorporation in HDX MS measurements (32). The A-state of Ub, the structure of which has been extensively characterized using NMR (33)(34)(35)(36)(37)(38) and other spectroscopic techniques (39), largely maintains the native secondary structure within the N-terminal part, whereas significant conformational changes within the C-terminal part eliminate the native fold and transform it into a transient helix. Top-down HDX MS characterization of the nonnative conformer of Ub reported in this work generates backbone-amide protection patterns fully consistent with the A-state of the protein and demonstrates that that top-down HDX MS/MS is capable of selective characterization of structural features of nonnative protein states without interference from other conformers coexisting in solution at equilibrium.…”