Computational protocol: A Structural Basis for IκB Kinase 2 Activation Via Oligomerization-Dependent Trans Auto-Phosphorylation

Similar protocols

Protocol publication

[…] Due to data quality and resolution and the presence of multiple molecules in the asymmetric unit, molecular replacement proved challenging. Initially, a brute force molecular replacement strategy was employed in which more than one hundred different IKK2 tetramer, dimer, or monomer models were generated manually based on the xIKK2 structure and used as search probes in various software packages. One of these test models identified two pairs of closely packed protomers (chains A:B and C:D) in space group P4122 using the CCP4 program MOLREP . With the A:B and C:D dimers fixed, one rigid-body refined monomer was then used as search model in MOLREP to identify the position of chain E, which participates in a similar interaction with its own symmetry related partner along a 2-fold axis. Crystal packing and electron density map calculation from this initial model indicated that there were likely additional molecules in the asymmetric unit but molecular replacement failed to locate them. The initial position of the sixth molecule, monomer F, was obtained by rotating monomer E about the 2-fold pseudo-symmetry axis relating the A:B and C:D V-shaped dimers. Refinement and crystal packing confirmed this to be the correct position for the sixth IKK2 subunit in the asymmetric unit. With six molecules in the asymmetric unit, the calculated solvent volume in the crystal is 64%. [...] The orientation and position of the initial model containing six IKK2 chains were refined by rigid-body refinement followed by minimization and simulated annealing refinement with a maximum likelihood target function and a flat bulk-solvent correction using the CNS system version 1.1 . Model rebuilding was carried out using 2F O-F C electron density maps in XTALVIEW . Due to the relatively low resolution data used throughout refinement, upon refinement of the model to R-factors of 27.5% (R-cryst) and 36.6% (R-free), CNS version 1.3 was employed in order to carry out Deformable Elastic Network (DEN)–assisted refinement . Initially, coordinates from a rigid-body refined xIKK2 model were employed as a reference. After a few refinement cycles with application of strict NCS restraints, the hIKK2 exhibited better model geometry than the reference structures and the chain with the best geometry among the six molecules was used to direct building and refinement of the final crystallographic model. DEN refinement dramatically improved the geometry and R-factors of the hIKK2 structure to a final R-cryst of 26.7% and R-free of 29.9%. Detailed refinement statistics are included in . Coordinates and structure factors for the hIKK2(11-669)EE crystal structure have been deposited into the Protein Data Bank as entry code 4E3C . Model figures were made in PyMOL . […]

Pipeline specifications

Software tools CCP4, Molrep, CNS, PyMOL
Application Protein structure analysis
Organisms Dipturus trachyderma, Homo sapiens, Xenopus laevis