|
Understanding of many cellular processes requires structural knowledge about large protein assemblies. It is not enough to know the
interacting protein components to understand the function of a large protein complex. The structure determination of protein complexes
is a challenging task and usually requires combination of several experimental techniques [1]. Recently there have been several attempts
to produce computationally either atomic models or more coarse-grained architectural models of large protein complexes. Aloy et al. started with a set of yeast protein complexes identified by tandem affinity purification and selected the most promising ones for electron microscopy [2]. They modelled the interactions between the component proteins using their similarity (sequence or structural) to interacting proteins of known structure and the obtained electron microscopy maps. Alber and coworkers have recently published structural models of the nuclear pore complex, which were produced by combining together diverse biochemical
and biophysical data of different resolution levels [3].
The goal of this work is to combine experimental protein - protein interaction data of various accuracy and computational interaction modeling techniques in order to produce reliable structures of large protein complexes. The produced models, on the other hand, can be used to study interaction interfaces and the amino acids involved in binding. These predictions can be validated with experimental methods and the obtained information used to further improve the modeling in an iterative process. The main idea is to build large protein complexes from structures of individual subunits with the help of various constraints. Molecular docking was used to obtain binary protein - protein structures. Then the pairwise structures were combined using experimental and computational constraints for filtering/decreasing the combination possibilities. Many structural alternatives of the binary structures were considered which leads to a high computational complexity. High-performance computing was used to handle the computational costs. Molecular dynamics simulations were used to calculate binding energies at various stages of the assembly process. The developed approach was applied to 3-dimensional modeling of Set1 histone methyltransferase complex from Saccharomyces cerevisae. Set1 complex methylates lysine 4 of histone 3 and, hence, is involved in the regulation of gene expression. Experimental high-quality data set produced by protein tagging and tandem affinity purification (TAP) in combination with mass spectrometry as well as bacterial two-hybrid method formed constraints for the modeling process [4]. An ensemble of structural models were obtained for this 8-subunit protein complex. 1. Sali A, Glaeser R, Earnest T, Baumeister W. From words to literature in structural proteomics. Nature 2003, 422:216-225 2. Aloy P, Böttcher B, Ceulemans H, Leutwein C, Mellwig C, Fischer S, Gavin AC, Bork P, Superti-Furga G, Serrano L, Russell RB. Structure-based assembly of protein complexes in yeast. Science 2004, 26:2026-2039 3. Alber F, Dokudovskaya S, Veenhoff LM, Zhang W, Kipper J, Devos D, Suprapto A, Karni-Schmidt O, Williams R, Chait BT, Rout MP, Sali A. Determining the architectures of macromolecular assemblies. Nature 2007, 450:683-694 4. Dehé PM, Dichtl B, Schaft D, Roguev A, Pamblanco M, Lebrun R, Rodríguez-Gil A, Mkandawire M, Landsberg K, Shevchenko A, Shevchenko A, Rosaleny LE, Tordera V, Chávez S, Stewart AF, Géli V. Protein interactions within the Set1 complex and their roles in the regulation of histone 3 lysine 4 methylation. J Biol Chem. 2006, 281:35404-35412 |
|