These screenshots are from a VMD movie of one of the long simulations of villin headpiece done by the UIUC groups of Martin Gruebele and Klaus Schulten, linked here. The native structure is shown as a transparent shadow. This is one of several simulations of villin by this and other research groups, and it happens to be one in which the native state is successfully attained.
|
|
|
The authors note that the simulation gets stuck in a compact state where the first and third helices are packed with the 3rd helix in the reverse orientation and the 2nd helix is not yet fully formed (1st image). They note that this intermediate state moves on to the native state only after the helices fully dissociate from each other (2nd image). We noted, too, that the helices rotate in the extended state, although in the YouTube movie it happens in the blink of an eye, and it is hard to be certain whether the rotation is right-handed or left-handed. Nevertheless, the helices come back together in a different, near-native orientation after re-collapsing (image 3), and simulataneously the second helix is formed. It makes sense to us that this is not a coincidence, that the formation of the second helix created torque on the ends, forcing the first and third helices -- locked in a right-handed contact -- to rotate. At points in the movie you can see hints of a struggle between helices 1 and 3, as the unsatisfied hydrogen bond donors and acceptors in helix 2 search for i->i+4 partners. Unable to pass through each other, the helices rotate only when they are separated by random forces. Once rotated, they collapse in a state with hydogren bonding groups satisfied, and sidechains nicely packed.
In this simulation, we are seeing the effect on the dynamics of high energy barriers to rotation. We would predict faster kinetics and possibly a different outcome if the torional barriers in phi and psi were removed.
For more on the folding of villin headpiece by the UIUC Theoretical and Computational Biophysics Group, go to this web page.
Ten-microsecond MD simulation of a fast-folding WW domain. Peter L. Freddolino, Feng Liu, Martin Gruebele, and Klaus Schulten. Biophysical Journal, 94:L75-L77, 2008.
Force field bias in protein folding simulations. Peter L. Freddolino, Sanghyun Park, Benoit Roux, and Klaus Schulten. Biophysical Journal, 96:3772-3780, 2009.