We aim to advance our understanding of the structure and function of chromatin by describing in vivo proteinprotein interactions in large scale. After having spent years developing tools and testing their capabilities on static and isolated multi-protein complexes, this year we have added the time dimension (dynamics) and moved into whole cell analysis. Our analysis of a static, 15-subunit, 670 kDa complex of Pol II with the initiation factor TFIIF has established cross-linking/mass spectrometry (3D proteomics) as an integrated structure analysis tool for large multi-protein complexes (Chen et al., EMBO J. 2010). We now add to this tool box the ability to quantify structural changes in solution. While establishing this technique, using the complement C3 to C3b conversion as a model system, we realised our tools are so precise that it was possible not only to follow the known structural changes accompanying this structurally well-defined process, but also to elucidate the previously unknown structure of the spontaneously arising complement C3(H2O), in the presence of C3 and C3b, in solution. C3(H2O) adopts a conformation that displays the same surfaces that C3b uses to bind factor B in C3bB, and suggests that the C3(H2O)Bb complex recapitulates the structure of C3bBb, a finding that structurally unifies the classical and the alternative pathways.

Stepping up vastly in complexity to cross-linking Mycoplasma pneumoniae, we found our tools now ready to provide mechanistic insights into biological processes in the context of whole cell lysates. The provision of a novel type of mass spectrometer through the Wellcome Trust, and further method improvements that this allowed us, increased the amount of information from such a proteomic-level cross-linking analysis. Consequently, we revealed protein pairs from ribosome and RNA polymerase that participate in coupling transcription and translation in Mycoplasma pneumoniae. The identified ribosome cross-links form a protein network that perfectly fits to that expected from the homologues in the E. coli crystal structure and thus serves as an internal quality standard of our analysis. After having recently defined the protein composition of mitotic chromosomes in collaboration with the Earnshaw lab (Ohta et al., Cell 2010), we believe we now have the tools in hand to decipher changes in protein conformation and interactions in whole mitotic chromosomes.