Figure 4. Probing the dynamics of CK1δ with GaMD simulations.

(A-D) Stability of the activation loop assessed by the RMSD of residues 168–175 with respect to the ‘loop down’ (RMSD_down, blue) or ‘loop up’ conformation (RMSD^up, pink), as observed in the crystal structure. For each system, the RMSDs from all five MD replicas are superimposed. Panel A, WT CK1δ^{loop down}; (B) tau CK1δ^{loop down}; C, WT CK1δ^{loop up}; D, tau CK1δ^{loop up}. (E-H) Dynamics of the EF loop assessed by the RMSD of residues 213–224 with respect to the initial structure. For each system, the RMSD was calculated for individual replica (gray lines, n = 5) and then averaged (green). The molecular representations in panels E-H show the crystallographic structure of CK1δ (gray) superimposed with snapshots of the L-EF loop extracted from the GaMD simulations (green). When present in the crystal structure and the simulation, sulfate anions are represented by spheres. I-L, Alterations in anion and substrate-binding clefts arise from the activation loop switch and tau mutation. Volumes for the binding clefts were extracted and averaged from GaMD simulations in the four states: panel I, WT CK1δ^{loop down}; J, WT CK1δ^{loop up}; K, tau CK1δ^{loop down}; L, tau CK1δ^{loop up}. Water and anions were removed from the analysis. Volumetric maps are contoured at 0.1 and represent regions that were consistently open during the simulations. See also Figure 4—figure supplement 1 and Supplementary file 1d for more information.

Figure 4—figure supplement 1. Additional details for GaMD simulations.

(A) Defining the localization of anion binding sites 1 (green), 2 (magenta) and 3 (yellow) for MD simulations. For reference, the loop down conformation of WT CK1δ is represented as ribbons (cyan). (B) Residue Y225 displays more conformational freedom in the tau mutant (right panel) than in WT CK1δ (left panel). Populations of the gauche+ (g+), trans (t) and gauche- (g-) rotamers for Y225 from MD simulations of the loop up and loop down conformations of WT and tau kinase are depicted. (C-D) Stability of the activation loop assessed by the Root Mean Square Deviation (RMSD) of residues 168–175 with respect to the loop down conformation (RMSD_down, blue) or to the loop up conformation (RMSD^up, magenta) as observed in the crystal structures. For both systems, the Site 2 anion was removed computationally before beginning MD simulations. Panel C, WT CK1δ^{loop down}; D, tau CK1δ^{loop down}; both show RMSDs superimposed from all five MD replicas. (E-F) Dynamics of the L-EF loop obtained from Gaussian Accelerated MD simulations in WT CK1δ (panel E) and tau mutant (panel F) when the activation loop in the ‘down’ conformation and the second anion binding site has been removed computationally. The RMSD was calculated for the backbone atoms of residues 213–224 for each independent replica (gray, n = 5) and then averaged (green). The molecular representations show crystallographic structures of the enzyme (in gray) superimposed with several snapshots of the loop L-EF extracted from the Gaussian Accelerated MD (GaMD) simulations (in green). When present, sulfate anions are represented by van der Waals spheres. G, Volumetric analysis of the Site 1 binding site (gray sphere), substrate binding cleft (golden sphere) and Site 2 anion binding site (rose sphere) during the GaMD simulations. In each system, the volumes were calculated for snapshots extracted from the GaMD trajectories every 2 ns, using POVME 3.0 (Wagner et al., 2017). Water and ions were computationally removed prior to volume calculations.