Protein folding is difficult to simulate with classical molecular dynamics. Secondary structure motifs such as alpha-helices and beta-hairpins can form in 0.1-10 micros (ref. 1), whereas small proteins have been shown to fold completely in tens of microseconds. The longest folding simulation to date is a single 1- micro s simulation of the villin headpiece; however, such single runs may miss many features of the folding process as it is a heterogeneous reaction involving an ensemble of transition states. Here, we have used a distributed computing implementation to produce tens of thousands of 5-20-ns trajectories (700 micros) to simulate mutants of the designed mini-protein BBA5. The fast relaxation dynamics these predict were compared with the results of laser temperature-jump experiments. Our computational predictions are in excellent agreement with the experimentally determined mean folding times and equilibrium constants. The rapid folding of BBA5 is due to the swift formation of secondary structure. The convergence of experimentally and computationally accessible timescales will allow the comparison of absolute quantities characterizing in vitro and in silico (computed) protein folding.

No abstract

Backbone hydrogen bonds (H-bonds) are prominent features of protein structures; however, their role in protein folding remains controversial because they cannot be selectively perturbed by traditional methods of protein mutagenesis. Here we have assessed the contribution of backbone H-bonds to the folding kinetics and thermodynamics of the PIN WW domain, a small beta-sheet protein, by individually replacing its backbone amides with esters. Amide-to-ester mutations site-specifically perturb backbone H-bonds in two ways: a H-bond donor is eliminated by replacing an amide NH with an ester oxygen, and a H-bond acceptor is weakened by replacing an amide carbonyl with an ester carbonyl. We perturbed the 11 backbone H-bonds of the PIN WW domain by synthesizing 19 amide-to-ester mutants. Thermodynamic studies on these variants show that the protein is most destabilized when H-bonds that are enveloped by a hydrophobic cluster are perturbed. Kinetic studies indicate that native-like secondary structure forms in one of the protein's loops in the folding transition state, but the backbone is less ordered elsewhere in the sequence. Collectively, our results provide an unusually detailed picture of the folding of a beta-sheet protein.

Protein folding barriers result from a combination of factors including unavoidable energetic frustration from nonnative interactions, natural variation and selection of the amino acid sequence for function, and͞or selection pressure against aggregation. The rate-limiting step for human Pin1 WW domain folding is the formation of the loop 1 substructure. The native conformation of this six-residue loop positions side chains that are important for mediating protein-protein interactions through the binding of Pro-rich sequences. Replacement of the wild-type loop 1 primary structure by shorter sequences with a high propensity to fold into a type-I ␤-turn conformation or the statistically preferred type-I G1 bulge conformation accelerates WW domain folding by almost an order of magnitude and increases thermodynamic stability. However, loop engineering to optimize folding energetics has a significant downside: it effectively eliminates WW domain function according to ligand-binding studies. The energetic contribution of loop 1 to ligand binding appears to have evolved at the expense of fast folding and additional protein stability. Thus, the two-state barrier exhibited by the wild-type human Pin1 WW domain principally results from functional requirements, rather than from physical constraints inherent to even the most efficient loop formation process.␤-turn ͉ ligand binding ͉ protein folding ͉ ␤-sheet ͉ protein function G lobular proteins evolve by mutation and selection. Selection criteria include function, and sufficient thermodynamic stability and folding rate to avoid sustained chaperone binding and proteasome degradation. The selection criteria cannot always be optimized independently over the entire sequence of a protein. For the human Pin1 (hPin1) WW domain (Pin WW hereafter), we have shown that residues important for stability and folding rate are segregated in the sequence (1-4). It is likely that functional selection criteria are predominant once minimal energetic criteria are met. Therefore, sequence evolution to enhance function may lead to a decrease in protein stability and folding rate compared with a sequence optimized for folding energetics.The hPin1 cell cycle regulatory proline (Pro) cis͞trans-isomerase is a two-domain protein (5). In its physiological role, the N-terminal WW domain binds Pro-rich ligands of the consensus sequence (pS͞pT)P, whereas the C-terminal domain catalyzes the Pro cis͞ trans-isomerization at the pS͞pT-P peptide bond. NMR solution studies show that the two domains, which are connected by a flexible solvated linker, interact only weakly before ligand binding (6, 7). The structure of the isolated Pin WW domain is virtually superimposable on that of the WW domain in the two-domain hPin1 protein (8). Moreover, Pin WW exhibits sufficient thermodynamic stability for biophysical analysis, folds rapidly, and retains its ligand-binding function (3, 9). These attributes, combined with sequence information on Ͼ150 WW domain family members (10, 11), makes Pin WW an excellent small model p...