Information Processing by Post-translational Modification

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Most proteins in an organism are reversibly modified by the covalent attachment of chemical groups to specific amino-acid residues. Such "post-translational modification" (PTM) allows protein function to be modulated on a physiological time-scale. PTMs are involved in regulating most cellular processes and they are frequently disordered in major diseases such as cancer and Alzheimer's. Our understanding of their role has been hampered by their extraordinary complexity. It is not uncommon for a protein to have many types of modification (phosphorylation, methylation, ADP-ribosylation, ubiquitination, etc.) on multiple sites and for modifications on different sites to interact combinatorially to influence protein function. This can create an explosion of combinatorial modification states. For instance, the tumor suppressor and "guardian of the genome" p53, on which this proposal will focus, has more than 100 sites of modification, creating the potential for more than 1030 combinatorial modification states on this single protein. Of course, very few of these states will be present in any cellular condition but different conditions may elicit different patterns of modification. This had led many researchers to suggest that combinatorial PTMs provide some form of "code" for cellular information processing, such as the "p53 code", the "tubulin code" and the "histone code". However, while correlations between patterns of modification and downstream responses have been shown, causality has not been demonstrated and the concept of code remains, at best, a metaphor. In previous work, we have laid a foundation for addressing this problem. We have introduced the concepts of "mod-form", for a pattern of modification across the whole protein, and "mod-form distribution", for the proportions of each mod-form in a sample. We have developed mass- spectrometry (MS) techniques in which conventional "bottom-up" peptide-based MS is combined with newer "top-down", or whole-protein, MS to quantitatively constrain the mod-form distribution of typical cellular proteins with small numbers (<5) of modified sites. We have also developed a new mathematical framework which shows that the combinatorial explosion in mod-forms disappears mathematically at steady-state, offering a way to overcome the complexity barrier. Here, we build on this foundation by bringing together an outstanding group of collaborators, with expertise that spans cell biology, mass spectrometry, mathematics and computation, to unravel whether and how endogenous p53 in human cells uses PTMs to encode information. We have already obtained the first top-down mass-spectrum of intact p53. By focusing on such a challenging exemplar, we expect to learn a great deal about p53 itself while developing concepts and methods that can be widely applied to other cellular proteins in which PTMs play a key role.