Understanding of the protein folding pathway is regarded as one of the most interesting, open, and unsolved problems till date. We have proposed a minimalistic model of protein along with a force field based on the hydropathy scale and helix propensities of the amino acids to understand the aforesaid problem.
The Basic construction of the model protein is shown where C-alpha atoms are numbered as 1, 2, 3, etc., whereas the side residues are shown by 1', 2', 3' etc. Note the varying size of the side residues.
Brownian dynamics simulations have been carried out with the initial configurations generated by configurational-bias Monte Carlo technique, by sudden quenching from a fixed high temperature to a low temperature. The folding study has been performed on a globular protein called chicken villin HP-36. The best folded structure of model HP-36 has a root mean square deviation of 4.5 Å from the native NMR structure. Moreover, the structure shows both the helices and bends at the appropriate positions.
The multistage dynamics of protein folding. Red curve shows the variation in energy with time and blue curve shows corresponding variation in the radius of gyration. Inset shows the magnified plot of the change in minimized energy during folding.
The dynamics of folding shows multistage decay in energy, radius of gyration, relative contact order etc. The fast initial collapse is attributed to the hydrophobic collapse, and the subsequent long plateau indicates the presence of a rate determining step and signature of it being entropic in origin, where the formation of the long range contacts takes place. The structure and dynamics of folding of a non-globular protein beta-amyloid and its different fragments are also explored with the help of minimalistic model described. The model folded structure correctly reproduces the well-known amyloid beta-turn at low temperature and bent-rod like structure at high temperature. Similar to HP-36, multistage dynamics is observed in various quantities. Results have been correlated with the earlier theories of protein folding.
A new orientational potential of mean force has been derived from the statistical analysis of the experimental native structures deposited in the protein data bank. In this model amino acid side chains are represented by single ellipsoidal site. The site-site potential of mean force (PMF) is calculated from the statistics of their distance separation obtained from the crystal structures. These site-site potentials are then used to calculate distance and orientation-dependent potential between all amino acid residues. We found that PMF between two hydrophobic residues is attractive in a short distances and it is repulsive between two hydrophilic residues.
Figure caption:The Interaction between the pair of side residues, each of which is modeled as ellipsoid, plotted as a function of distance for different orientation of the ellipsoids: a) Phenylalanine-phenylalanine (hydrophobic residue: left) and b) lysine-lysine (hydrophilic residue: right). Arrow in the figure indicates the direction of the major axis or the orientation of the side chain. Note that hydropathy scale is greatly obeyed here.
The above results are in good agreement with the hydropathy scale proposed earlier. However, we have observed many unexpected pair interactions which defy from the trend given by hydropathy scale. An example is the Arg-Arg pair interaction which is found to be strongly and surprisingly attractive at short separation, even though it is the most hydrophilic residue.
Figure caption:The Interaction between the pair of histidines as a function of distance for different orientations (see also caption of the figure before): a) In presence of metal (left) and b) in absence of metal (right). This result highlights the importance ( somewhat hidden) role of metals in protein structure and folding.
We have also found the strong influence of metal in determining effective interaction among the amino acid residues. An example is the His-His attractive interaction which is found to be significantly enhanced in metalloproteins. The potential is tested for many proteins with multiple decoy states and the Z-score values are found to be comparable to the existing orientation dependent potential of mean force.