Computational protocol: New PI(4,5)P2- and membrane proximal integrin–binding motifs in the talin head control β3-integrin clustering

Similar protocols

Protocol publication

[…] The docking prediction was performed using the protein–protein docking program FTDOCK () on a dual-processor computer. The preliminary models were downloaded from the protein database and prepared for analysis. A monomeric Tal-F2/3 model was obtained from the PDB structure 1MK7 () by keeping only chain B. The MP helical domain of the integrin β3 cytoplasmic tail (residue 720–735) was prepared from the PDB structure 1S4X (). In docking calculations, global scan was performed using Tal as the static model and the helix of β3 cytoplasmic tail as the mobile molecule. The docking possibilities were ranked in terms of shape complementarity, electrostatics, and empirical scores of residue level pair potentials. This result was subsequently filtered using knowledge acquired from mutational experiments, in this case, i.e., setting proximity constraints (4.5 Å) between K316 (Tal) and E726 (integrin). Three docking models satisfied the constraints, and the top one with the highest score was chosen as the final model (). Models were produced using PyMOL (DeLano Scientific). [...] Initial protein coordinates for MD analysis were obtained by manually combining W/NPLY motif and MP helical domains using the program VMD () assuming simultaneous binding of both integrin motifs. Tal-F2/F3–integrin peptide complex (PDB ID 1MK7) was used as a starting model, and the docked α-helix (see previous section) was connected to the rest of the integrin sequence by manually building the missing residues 734–738. The protein complex was placed in a 98 × 71 × 73 Å box filled with 13,915 TIP3 explicit water models (overall system size 45,403 atoms) and subjected to energy minimization. Minimization and MD calculations were performed using the CHARMM27 () force field in the program NAMD (). Each minimization phase contained 4,000 steps and involved conjugate gradient and line search algorithms implemented in the NAMD package. In the first minimization phase, all of the protein atoms were fixed, and the water molecules were allowed to move. Then, both binding motifs of the integrin peptide were constrained by fixing all protein atoms except the inserted integrin residues 734–738. The third minimization phase was performed on all protein Cα atoms fixed to allow side chain optimization. Finally, all atoms were released in the fourth minimization step. Then, the system was gradually heated up to 310 K during 31-ps MD simulation using Berendsen barostat (1 atm; ). Constant temperature (310 K) and pressure (1 atm) MD simulation was then continued for 2–3 ns. Calculations were performed by using parallel processing with 64–256 processors. […]

Pipeline specifications

Software tools VMD, CHARMM, NAMD
Application Protein structure analysis
Chemicals Phosphatidylinositols