Reliable Protein Folding on Complex Energy Landscapes: The Free Energy Reaction Path

From Soft-Matter
Revision as of 01:17, 31 October 2011 by Peter (Talk | contribs)

Jump to: navigation, search

Original Entry: Peter Foster, AP 225, Fall 2011 In Progress.....

Figure 1, taken from [1].
Figure 2, taken from [1].
Figure 3, taken from [1].

General Information

Authors: Gregg Lois, Jerzy Blawzdziewicz, and Corey S. O’Hern

Publication: Lois and et al.. Reliable Protein Folding on Complex Energy Landscapes: The Free Energy Reaction Path. Biophysical journal (2008)


This paper deals with a computational approach to the protein folding problem. Exactly how a protein folds into its three dimensional native configuration is an ongoing unsolved problem that people have worked on for quite some time. Early on, it was shown that protein's cannot do a random search in order to find their native conformation because this would take an amount of time that's longer than the average human lifetime (Levinthal's Paradox). Instead, the protein's conformational change is thought to be driven by changing to conformations that lower the protein's free energy. Many people think that the energy landscape is like a funnel, with the lowest energy state being the native conformation.

This paper considers how proteins fold when the temperature is lowered from higher to lower temperatures at a set rate. If this rate is higher than the folding rate of the protein, it is possible that the protein can become trapped in a metastable conformation. It is possible to escape these misfolded metastable states. There is a certain rate (for the exact expression check out eqn 7 in the paper) that the proteins will escape from the metastable state. However it the rate at which one lowers temperature (defined simply as r) is greater than the rate that proteins can escape from metastable states, then the protein will be trapped in the metastable state, leading to misfolding in the final protein.

A Brownian Dynamics simulation was used to simulate protein folding in two and three dimensions. Brownian Dynamics simulations are a coarse grained approach that divides up the protein into hard spheres with an added potential between some of the spheres. For their protein model they choose a chain of 13 spheres. Four of these spheres (labeled as green in the picture) have an attractive Lennard-Jones potential. An addition potential is added between each sphere and the two adjacent spheres in order to keep the linear chain intact. There is also a repulsive Lennard-Jones potential between white-white pairs and green-white pairs. Figure 1 shows the free energy landscape for this protein, along with the conformations for the native state (lower left corner) and two populated metastable states. The vertical axis is the radius of gyration and the horizontal axis is the end to end distance. Even such a simple model as this leads to the result that metastable states are possible.

Figure 2 uses data from the same simulations and is a plot of the end to end distance vs the energy divided by Ec (Ec is the well depth for the attraction between the green spheres) for different values of c (c is the magnitude of Ec divided by the temperature). When c is small (small attractive well depth) the chain acts like a random coil polymer (Fig 2 left panel). However, as the well depth gets lower and lower (higher c) the protein folds more and more until the phase spaces collapses into the lower left hand corner. When they allow the value of c to change with time (c=rt)



[1] Lois and et al.. Reliable Protein Folding on Complex Energy Landscapes: The Free Energy Reaction Path. Biophysical journal (2008)