Figure 3

Methods to estimate protein flexibility. A) Ramachandran plots of ψ versus ϕ of all amino acids (yellow dots) for myoglobin (top) and titin I91 domain (bottom). The blue regions of the plot are the allowed pairs of angles since they don't present steric repulsion, while green regions present more repulsion and are termed partially allowed regions. B) Cartoon representation of titin domain I91 coloured by B-factors, going from blue representing low B-factors to red for high B-factors. Loops connecting β-strands and N- and C- terminal domains appear in general more flexible than β-sheets. C) Ribbon representation of the family of structures of titin I91 domain obtained by NMR spectroscopy (PDB file 1TIU). The 24 different models calculated from NMR data represent I91 domain conformations that are compatible with the experimental data. They present good superpositions of the regions in beta-sheet fold, while the loops in between and the N- and C-terminal domains show variable positions. These observations are in agreement with the B-factors calculated from X-ray crystallography (Fig. 3B), which are low for β-sheets and higher for the loops and terminals and thus intrinsically more flexible. D) Neutron scattering experiments on bacteriorhodopsin (cartoon) providing the root-mean square displacement of protein atoms (<x²>) as a function of temperature. The slope of the curves at physiological temperatures (steeper slope) provided a quantification of the force constant of 0.12 N/m for the whole protein (unlabelled) and of 0.33 N/m for the labelled extracellular half of bacteriorhodopsin (labelled). Reproduced with permission from ref. [9]. Zaccai "From [Full Reference Citation]. Reprinted with permission from AAAS." DOI: 10.1126/science.288.5471.1604