Protein folding by zipping and assembly

Abstract

How do proteins fold so quickly? Some denatured proteins fold to their native structures in only microseconds, on average, implying that there is a folding "mechanism," i.e., a particular set of events by which the protein short-circuits a broader conformational search. Predicting protein structures using atomically detailed physical models is currently challenging. The most definitive proof of a putative folding mechanism would be whether it speeds up protein structure prediction in physical models. In the zipping and assembly (ZA) mechanism, local structuring happens first at independent sites along the chain, then those structures either grow (zip) or coalescence (assemble) with other structures. Here, we apply the ZA search mechanism to protein native structure prediction by using the AMBER96 force field with a generalized Born/surface area implicit solvent model and sampling by replica exchange molecular dynamics. Starting from open denatured conformations, our algorithm, called the ZA method, converges to an average of 2.2 A from the Protein Data Bank native structures of eight of nine proteins that we tested, which ranged from 25 to 73 aa in length. In addition, experimental Phi values, where available on these proteins, are consistent with the predicted routes. We conclude that ZA is a viable model for how proteins physically fold. The present work also shows that physics-based force fields are quite good and that physics-based protein structure prediction may be practical, at least for some small proteins.

Fig. 1.

Ribbon diagrams of the predicted protein structures using the ZAM (purple) vs. PDB structures (orange). The backbone C^α rmsds with respect to PDB structures are: protein A, 1.9 Å; albumin-binding domain protein, 2.4 Å; α3D, 2.85 Å (excluding the residues in the loops) or 4.6 Å; 1–35 residue fragment of ubiquitin, 2.0 Å; protein G, 1.6 Å; FBP26 and YJQ8 WW domains, 2.2 Å and 2.0 Å; and α-spectrin SH3, 2.2 Å. Our method fails to find the src-SH3 structure. Shown here is a conformation that is 6 Å from native. The problem in this case appears to be in the GB/SA implicit solvation model.

Fig. 2.

Experimental average Φ values (black bars) and estimated kinetic impact values (gray bars) based on ZA folding routes for protein G, WW domain (Pin), protein A, and α-spectrin SH3. The kinetic impact value ranges are high (0.6–0.8), medium (0.3–0.5), and low (0–0.2).

Fig. 3.

ZA process for protein G. The chain is parsed into fragments of 8–12 residues. For each fragment, REMD simulation is performed for 5 ns per replica. PMFs are computed to determine whether a fragment is structured or unstructured. For protein G, the PMFs reveal that the C-terminal hairpin and the N-terminal β hairpin each form favorable hydrophobic contacts independently. For each segment that is structured, a spring is added to enforce that structure. Then new residues are added to the fragment ends (“growth”) for another round of REMD simulation. For protein G, this results in the complete formation of both hairpins, with a helix packing onto the C-terminal β hairpin (b). When growth is no longer possible, as in protein G, the two folded units attempt to assemble, which, in this case, successfully leads to the native structure (a).