A mechanistic computational model for simulating protein unfolding.

To model the final stages of folding, and the first steps of unfolding, a mechanistic model was developed for breaking the final structure into smaller pieces via simple rotations and tranlations. The mechanistic model (GeoFold), developed in collaboration with the M. Zaki lab (Comp Sci) generates a map of all topologically possible unfolding (and implicitly, folding) pathways. Kinetic simulations can be done on this ensemble, and such simulations have been shown to reproduce observed unfolding kinetics and the effects of point mutations. Unfolding pathways generated by GeoFold are being used to understand the extreme kinetic sability of some protein, especially those studied by the W. Colon lab (RPI Chemistry).

Try the GeoFold server.

The folding pathways for Protein G (2gb1) according to GeoFold. Orange states are intermediates, red states are transitions. The green line show the one pathway with the highest flux.

click for a closer look

The phone cord effect in protein folding

Anyone who has ever coiled a rope knows that you have to twist the ends to get the coil to stay flat. Depending on which way you twist the ends, you get either a right-handed or a left-handed rope coil. We believe that the same can be said for an α-helix during protein folding that the ends of the helix must also twist. Coiling a rope creates torque on the ends because the rope cannot pivot to relieve torsional stress. The polypeptide can pivot around backbone φ and ψ angles, but these do not rotate freely; both must pass through disallowed regions of the Ramachandran plot. Because those barriers are high, a polypeptide chain is expected to behave partially like a rope perhaps like the familiar coiled phone cord that allows but resists torsional stress. It follows that the formation of a helix should create torque on the ends. This torque may be relieved by the formation of helical supersecondary structures, such as βαβ units and 3 α-helix bundles in protein structures. The "phone cord effect" may also explain the formation of knots in proteins, such as the one in the figure to the right.

This work was published in 2009. Cole B & Bystroff C. (2009) Alpha helical crossovers favor right-handed supersecondary structures by a kinetic trapping mechanism. The phone cord effect in protein folding. Protein Science, (in press)

Phone cord effect supplementary data site

Proposed late stage of the folding of knotted protein 1MXI. (a) Native structure, showing C-terminal helix (blue) inserted through helical crossover loop (green). (b) Proposed non-native folding intermediate with no knot, drawn using TOPS conventions. (c) C-terminal helix formation rotates hydrophobic loop against α5 crossover. (d) α7 is pulled through the loop as it forms, creating a knot.

click for a closer look

Back to research interests