Zhou et al. 10.1073/pnas.0701249104. |
Fig. 5. Comparison of the fraction of native contacts for the other four trajectories starting from different initial configurations (total of five trajectories, with run 1 shown in the main article). Runs 2 and 3 have finished the entire 1-ms simulation as run 1, whereas runs 4 and 5 only finish 500 ns each (they seem to be long enough, though, to show the difference we are interested in here). All trajectories show a faster loss of native contacts in the mutant, with a smaller fraction of native contacts at the end. Here, a native contact is defined as a pair of residues i and i + n (nonnearest and non-second-nearest, n ³ 3), where the distance between their backbone Ca carbons are <6.5 Å.
Fig. 6. Comparison of the local contacts of the mutation site (residue 62) for the other four trajectories starting from different initial configurations (total of five trajectories, with run 1 shown in the main article). Runs 2 and 3 have finished the entire 1-ms simulation as run 1, whereas runs 4 and 5 only finish 500 ns each (again, they seem to be long enough to show the difference we are interested in here). All trajectories show a higher contact number for the wild type than the mutant. Essentially, Trp-62 plays a key role in bridging the neighboring positively charged basic residues, which, in turn, help form a nucleation core through long-range electrostatic interactions. Here, a "local contact" (native or not) is defined the same way as a native contact except that the distance between Ca carbons is chosen to be <10 Å, rather than 6.5 Å for a native contact. This will give us a broader view of residues proximate to the mutation site.
Fig. 7. Time evolution of the secondary structure at a 900- to 1,000-ns MD simulation for the wild type (a) and the W62G mutant (b). The secondary structure a-helix is colored blue, 310-helix light blue, b-strand red, and coils and turns black. Even at the end of the 1-m s simulation, some of the a-helical structure still persists, whereas the tertiary structures are all destroyed much earlier.
Fig. 8. Representative RMS fluctuation of the wild type and mutant at the first 100-ns MD simulation in 8 M urea at pH 2. The mutant has shown a much larger fluctuation across all residues than the wild type. A closer look also reveals that the disruption starts in the b -domain region of the protein. When the simulation progresses beyond 100 ns, the wild type also unfolds in this high concentration of denaturant and shows much higher RMS fluctuations as well.
Fig. 9. One representative time evolution of the fraction of native contacts (a) and local contacts of residue 62 (b) at pH 7 and 350 K. Again, the wild type shows more native-like contacts than the mutant during the 100-ns MD simulation, and the Trp-62 residue (wild type) has a lot more local contacts than the Gly-62 residue (mutant). Overall, the pH 7 simulations showed results similar to those of the pH 2 simulations.