Computational protocol: Structure of Bradavidin – C Terminal Residues Act as Intrinsic Ligands

Similar protocols

Protocol publication

[…] Suitable conditions for crystallization of bradavidin were found using the Classics™ (Nextal Biotechnology) screen, the vapour diffusion method and sitting drops (1–2 µl) on 96 well plates (Corning Inc.). The protein solution (∼0.4 mg/ml) contained 50 mM sodium acetate (pH 4). Saturated solution of an azo dye HABA (<10 mg/ml) was added to the protein solution in 1∶10 (v/v) ratio, respectively, before crystallization. Bar-like crystals of a typical size of 0.2×0.05×0.05 mm were used for data collection. The crystals were formed at 22°C using 0.7 µl of well solution containing 25% (w/v) PEG 4000, 0.17 M ammonium acetate and 0.08 M sodium acetate (pH 4.6), and 0.8 µl of the protein solution.X-ray diffraction data were collected at the MAX-lab beam line I911-2 (Lund, Sweden) equipped with a MarCCD detector. The crystal was cryoprotected by adding 0.7 µl of 4 M sodium formate to the crystallization drop just prior to flash-freezing in a 100 K liquid nitrogen stream (Oxford Cryosystem). The collected data was originally processed with Mosflm (7.0.3) using the iMosflm (0.6.1) GUI and scaled with Scala of the CCP4 program suite using the CCP4i GUI , and later reprocessed with XDS (see for X-ray structure determination statistics).The initial phase information for structure factors was obtained using the molecular replacement program Phaser within the CCP4i GUI . Multiple search models and ensembles were tested before a solution could be found, finally using homology models as a search ensemble. Shortly, three tetrameric homology models were produced using Modeller of Discovery Studio 2.1 (Accelrys Software Inc.). A structural alignment (data not shown) of the core sequence of bradavidin (Uniprot Q89IH6) and the sequences from X-ray structures of avidin [PDB: 1AVD], streptavidin [PDB: 1MK5] and xenavidin [PDB: 2UYW], respectively, was used for modelling. The models (one model per template structure) were structurally aligned using Pymol and their N- and C-termini were trimmed based on visual checking of the models. Monomers of the aligned structures were then used as a search ensemble in Phaser (two monomers were searched) giving an initial solution with TFZ = 8.5 and LGG = 80 in space group P212121. The initial, incomplete dimeric model of bradavidin was refined using Refmac5 resulting in an Rfactor = 0.513, Rfree = 0.536 and FOM = 0.297 and then used as a search model for the next round of Phaser, where two dimers were searched giving a solution with TFZ = 13.5 and LGG = 246. Refinement of the solution, i.e. a tetrameric model of bradavidin, resulted in an Rfactor = 0.449, Rfree = 0.482 and FOM = 0.477. This model was used then, again, as a search model in Phaser now yielding a solution with TFZ = 20.7 and LGG = 320. The model was manually edited/rebuilt using Coot and refined with Refmac5 (Rfactor = 0.420, Rfree = 0.456 and FOM = 0.546) before rebuilding the whole structure with ARP/wARP (starting from an existing model; v. 6.1.1) –, finally giving a model of bradavidin that could be finished through further cycles of refinement with Refmac5 and modification/rebuilding with Coot. Solvent atoms and other non-protein atoms were added to the model either with the automatic procedure of ARP/wARP or Coot, or manually in Coot. Few cycles of the refinement in the middle of the structure building was done also with the software suite Phenix . We could not solve the structure without the step-by-step procedure used for molecular replacement described above.The final structure of wt bradavidin was validated using the inbuilt tools of Coot , and using MolProbity of the Phenix software suite , before deposition to the Protein Data Bank , with PDB entry code 2Y32. The data collection and structure determination statistics are summarized in . [...] The structure-based sequence alignment was done using Malign of the Bodil software, a modular, multi-platform software package for biomolecular visualization and modeling , . ESPript was used for visualization of the sequence alignment. PyMOL (The PyMOL Molecular Graphics System, Version 1.3, Schrödinger, LLC) was used to create all the figures relating to structural representations. ABPS plugin of PyMOL (MG Lerner and HA Carlson. APBS plugin for PyMOL, 2006, University of Michigan, Ann Arbor) was used for electropotential calculations – alternative rotamers were excluded from the calculations. Inkscape 0.47 was used to edit the figures related to structural representations. […]

Pipeline specifications

Software tools iMosflm, CCP4, MODELLER, PyMOL, REFMAC5, Coot, ARP/wARP, PHENIX, MolProbity, Bodil, ESPript, Inkscape
Organisms Bradyrhizobium japonicum, Dipturus trachyderma, Glycine max
Chemicals Biotin, Nitrogen, Sepharose