Computational protocol: Integrative Approach for Computationally Inferring Interactions between the Alpha and Beta Subunits of the Calcium-Activated Potassium Channel (BK): a Docking Study

Similar protocols

Protocol publication

[…] Nine secondary-structure prediction programs were used to predict the length of each helical stretch of the subunits: DAS, ALOM2, PHDtm, TMAP, TMHMM2, TMPRED, TOPPRED2, SPLIT4, and HMMTOP2. Based on the results, a consensus prediction was calculated according to a simple majority vote type procedure. Alom2 and DAS are methods that focus on local properties of amino acid sequences to decide which sub-sequences are most likely to span the membrane, usually in sliding Window approach. PHD uses a neuronal network and should still be regarded as a local approach. Global approaches are HMMTOP and TMHMM, which both implement circular Hidden Markov Models. These approaches determine the statistically most probable topology for the whole protein according to the underlying model. TMAP, Toppred2 and SPLIT4 represent combined forms, in which results on a local level are evaluated by global heuristics such as the positive-inside rule or other differences in the distribution of amino acids. [...] The sequence of the human BK potassium channel was obtained from the Swiss-Prot database at http://expasy.org/sprot. BLAST (basic local alignment search tool) sequence analysis was performed against the whole protein data bank to detect homologous/analogous protein templates by matching the whole-chain sequences of target subunits to solved protein structures. Since conserved-structure pieces excised from the different protein structures were directly used to assemble the new protein-structure models, the threading modeling approach was chosen together with the HHpred program., The prediction was made by constructing a structure model by placing the backbone atoms of the target sequence at their aligned backbone positions of the selected structural templates (PDB IDs: 3naf, 1k4c and 1f6g).HHpred allows the user to test several alternative alignments and to evaluate the quality of the resulting models in order to achieve an optimal result. Results files contain the superposed template structures, and the alignment between each subunit file and template file are generated inside the HHpred program. After following this approach, the generated alpha and beta1 subunits were validated using the PROCHECK program and compared with the previous secondary structure predictions for the helix of each subunit at the level of the length of the secondary structure. The aim of PROCHECK is to assess how normal or, conversely, how unusual the geometry of the residues in a given protein structure is when compared to the stereochemical parameters of well-refined and high-resolution structures. [...] The alpha and beta subunit structures derived from the secondary and tertiary structure predictions were each used as the starting structure in docking simulations with ClusPro’s as well as in GRAMM algorithms for obtaining the bound structures of the alpha and beta1 subunits. ClusPro has the option of selecting either DOT, or ZDOCK to perform rigid-body docking and both are based on fast Fourier transform (FFT) correlation techniques., DOT was selected for the present work, since it allows for the use of an electrostatic potential in the scoring function and in the surface complementarities between the two structures. DOT runs on a 128A × 128A × 128A grid, with a grid spacing of 1Å. It performs 13,000 rotations, initially obtaining over 2.7 × 1010 structures and finally retaining only 20,000 structures with the best surface complementary scores. These docked structures are then filtered using distance-dependent electrostatics and an empirical potential energy. The 2000 conformations retained after filtering are clustered based on the pairwise RMSD (root mean square deviation) and the best conformational structure is selected. Finally, the representative conformation selected is refined using CHARMM minimization. The GRAMM program was used because it runs in helix mode, ie, the search can be limited to helix pairs in antiparallel and parallel orientations, discarding configurations with large displacements along the helix axes and crossing angles larger than 10°.The best structural model for the complex of alpha and beta subunits obtained from the docking simulations was subjected to MD simulation to refine the protein interface. However, no explicit constraint functions were used to maintain the initial docking contacts during the simulation. The structures were first energy minimized using 1000 steps of steepest descent and 2000 steps of conjugate gradient minimization implemented in GROMACS. A distance dependent dielectric function was used with the dielectric constant set to 1 and the non-bonded cutoff was set to 8Å. Energy minimization with classical force field can be used to remove unrealistically close steric clashes and large deviations from ideal geometry resulting from the conformational changes of amino acid side chains after docking. The binding free energy was calculated by MM-PBSA implemented in Amber 7.0. […]

Pipeline specifications

Software tools tmap, TMHMM, TMpred, TopPred, HMMTOP, BLASTN, HHPred, PROCHECK
Databases UniProt ExPASy
Applications Protein structure analysis, Membrane protein analysis
Chemicals Calcium, Hydrogen, Potassium