Similar protocols

Protocol publication

[…] WT AurA in complex with ADP was simulated with and without Tpx2. All simulations were started from the X-ray structure of WT AurA bound to Tpx2 and ADP in the presence of three magnesium ions (PDB ID: 1OL5 []). From the crystal structure, PDBFixer (https://github.com/pandegroup/pdbfixer) version 1.2 was used to model in Tpx2 residues 23–29 (unresolved in 1OL5), add hydrogens belonging to standard dominant protein residue protonation states at pH 7.4, and remove phosphorylation from threonine residues 287 and 288 (). Crystallographic waters were retained to prevent nonphysical collapse of hydrophilic pockets during minimization. The chain containing Tpx2 was then removed for simulations without Tpx2 (-Tpx2) and retained for simulations with Tpx2 (+Tpx2). Sulfate ions present in the crystal structure were manually removed. The crystallographic ADP (containing only heavy atoms) was extracted from the structure and converted to a protonated Tripos mol2 file using OpenEye toolkit OEChem v2015.June (; ). The protein structure was then loaded as an OpenMM version 7.0.1 Modeller object, and the protonated ADP was reintroduced through conversion from mol2 to OpenMM format via MDTraj 1.4.2 (; ). [...] Because AMBER parameters for spin probes were not widely available, we used the equivalently modern CHARMM36 generation of forcefields for MTSL-labeled AurA, simulated in complex with ADP and with and without both TPX2 and phosphorylation, for a total of four possible combinations per starting structure. Simulations were started from two different starting configurations: DFG-in (PDB ID: 1OL5) and DFG-out (5L8K ()). For the crystal structure of 1OL5, Schrödinger’s PrepWizard () (release 2016–4) was used to model in Tpx2 residues 23–29 (unresolved in 1OL5), and add in hydrogens at pH 7.4 for both protein residues and ADP. The protonation state of ADP was assigned the lowest energy state using Epik at pH 7.4 ± 2. Hydrogen bonding was optimized using PROPKA at pH 7.4 ± 2. The entire structure was minimized using OPLS3 and an RMSD convergence cutoff of 0.3 Å. Because TPX2 is not present in the 5L8K structure, the coordinates and crystal waters of Tpx2 in 1OL5 after preparation were transferred to the unprepared 5L8K after aligning the kinase domain to 1OL5 and deleting the vNAR domain (chain B) from 5L8K. 5L8K was then prepared using the same protocol as above, removing any organic solvent molecules but retaining all crystal waters. All three structures were run through CHARMM-GUI Solvator tool () (http://www.charmm-gui.org/?doc=input/solvator). In the first stage of this tool, all crystal waters and magnesiums in the structures where retained, while sulfates were deleted. Tpx2 was deleted at this stage for the non-Tpx2 conditions. Phosphorylation was either built in or deleted for the threonine at residue 288 in the second stage of the Solvator tool. Also in this stage, residues 225 and 284 were mutated to cysteines and the MTSL spin label (named CYR1) was added to those residues. C290 was mutated to serine in the unphosphorylated conditions and alanine in the phosphorylated conditions, to match the DEER experimental conditions. After the PDB file was generated, a rectangular solvent box was generated using 10 Å edge distance fit to the protein size, with 300 mM NaCl placed using the Monte-Carlo method. [...] Distances and torsions were computed using the compute_distances and compute_dihedrals functions in MDTraj v 1.8.0. For the WT Amber simulations, the first 100 ns of each CLONE were discarded. The first 250 ns were discarded for the CHARMM MTSL labeled AurA simulations to allow sufficient relaxation following the introduction of spin probes. Distance probability plots were generated using Seaborn v0.8.1 (https://doi.org/10.5281/zenodo.883859) distplot using the norm_hist parameter. The distance and dihedral contour plots were generated using the kdeplot function in Seaborn v0.8.1. All analysis scripts have been made publically available (). […]

Pipeline specifications