The question of how a protein folds is an important open research question. It is most likely that the native folded state of a protein is the global free-energy minimum. However, protein free-energy surfaces are very rough. Over 60,000 potential energy minima were found for a tetra-alanine [1]. The free-energy surface is funneled, leading the protein toward the free-energy global minimum, but in many simulations the protein gets stuck in deep local minima. Also, the folding time can range from nanoseconds for short polypeptides to minutes for more complex proteins. While the shorter times can be achieved through traditional molecular dynamics (MD) simulations, the longer times are prohibitive for MD. Various algorithms such as multicanonical simulations [2-3] basin hopping [4], and potential energy surface mapping[5-6] have been used to examine both the dynamics and the thermodynamics of protein folding. We developed an accelerated MD algorithm to look at the folding of α-helix polypeptides on the molecular level. Since alanine has a large helical propensity, polyalanine chains are known to form α-helices. We examined Ac(ala)_nNHME where n takes values of 8, 12, and 16, and the polypeptide has varying degrees of helicity at equilibrium. Our simulations use the TINKER modeling package [7] with the AMBER95 force field [8] and are performed in vacuo and in implicit solvent. Acceleration is based on the bond-boost method [9] and applied to torsion angles along the backbone of the polypeptide. This acceleration can be applied in a straightforward manner to other polypeptides and proteins. We compare the relative populations of the local and global minimum dihedral angle values to results obtained through studies of the potential-energy surfaces [5,6] to validate our model. Our acceleration method is able to extend the time that can be simulated to the order of microseconds. Configurational space is explored more completely, as shown by the increase in Ramachandran plot space sampled by the accelerated algorithm. We note a ripple effect, in which small conformational changes in one region of the protein can cause large conformational changes in other areas of the protein due to non-bonded interactions.

[1] K.M. Westerberg and C.A. Floudas, J. Chem. Phys. 110, 9259 (1999). [2] U.H.E. Hansmann and Y. Okamoto, J. Chem. Phys. 110, 1267 (1999). [3] N.A. Alves and U.H.E. Hansmann, J. Chem. Phys. 117, 2337 (2002). [4] A. Verma et al, J. Chem. Phys. 124, 044515 (2006). [5] P.N. Mortenson and D.J. Wales, J. Chem. Phys. 114, 6443 (2001). [6] P.N. Mortenson, D.A. Evans, and D.J. Wales, J. Chem. Phys. 117, 1363 (2002). [7] P. Ren and J. W. Ponder, J. Phys. Chem. B, 107, 5933-5947 (2003). [8] W.D. Cornell et al, J. Am. Chem. Soc. 117, 5179 (1995). [9] R.A. Miron and K.A. Fichthorn, J. Chem. Phys. 119, 6210 (2003).