Abstract
Cellular functions are largely carried out by noncovalent protein complexes that may exist within the cell as stable modules or as assemblies of dynamically changing composition, whose formation and decomposition are triggered in response to extracellular stimuli. The protein constituents of complexes often exhibit post-translational modifications such as phosphorylation that can impact their ability to interact with other proteins and thus to form multicomponent complexes. A complete characterization of a particular protein complex thus requires determining both, the identity of interacting proteins and their covalent modifications, in terms of attachment sites and stoichiometry. We have previously developed a protocol which identifies genuine constituents of partially purified protein complexes and concurrently determines their phosphorylation sites and levels in a single LC-MS/MS analysis performed on a MALDI-TOF/TOF instrument (Pflieger, D.; Junger, M. A.; Muller, M.; Rinner, O.; Lee, H.; Gehrig, P. M.; Gstaiger, M.; Aebersold, R. Mol. Cell. Proteomics 2008 , 7 , 326 - 346). The method combines fourplex iTRAQ labeling (isobaric tags for relative and absolute quantification) and phosphatase treatment of peptide samples derived from the tryptic digestion of isolated complexes. To test the performances of this method with nanoESI and different peptide fragmentation modes, possibly better suited for the identification of phosphorylated sequences than MALDI-TOF/TOF-MS, we have implemented it on the nanoESI-LTQ-Orbitrap instrument. The model protein beta-casein was used to optimize the conditions with respect to sensitivity and quantitative accuracy: a combination of CID fragmentation in the linear ion trap and Higher energy Collision Dissociation (HCD) appeared optimal to obtain reliable and robust identification and quantification data. The optimized conditions were then applied to identify and estimate the respective levels of phosphorylation sites on the purified, autoactivated tyrosine kinase domain of Fibroblast Growth Factor Receptor 3 (FGFR3-KD) and to analyze complexes formed around the insulin receptor substrate homologue CHICO immunopurified from Drosophila melanogaster cells that were either stimulated with insulin or left untreated. These new analyses allowed us to improve the assignment of the phosphorylation sites of some peptides previously detected by MALDI-TOF/TOF analysis and to identify additional phosphorylated sequences in CHICO and in the insulin receptor.
