Molecular and functional characterisation of unanchored polyubiquitin

Proteins are the most abundant molecules in all living organisms and the life process relies on their controlled interactions with other proteins. One way in which interactions between proteins can be switched on is by a process known a post-translational modification. In this case, a first protein is modified by attaching another molecule on to its surface, and this appended modification is then recognised by a second protein which binds to it; thus, two proteins can be made to interact. Cells use this process as a way of signalling that a particular protein is worn out or damaged and needs to be destroyed. The protein that needs to be disposed of becomes modified by another molecule called ubiquitin, and a specialised protein called the proteasome, which itself is a protease (a protein that is able to digest other proteins) recognises this ubiquitin-modified target and removes it. Modification with ubiquitin does not however always signal disposal of the target; sometimes the modification simply serves to bring two proteins in to close proximity so they can function together. In each case, in fact usually several copies of the ubiquitin modifier become attached on to the target protein in a structure known as a 'polyubiquitin chain'. A very important recent scientific paper has demonstrated that 'unanchored' forms of polyubiquitin chains exist in cells - these are equivalent to the modifications found on other proteins, but are not actually attached to targets - and they themselves have very specific functions. However, to date no one has ever purified these unanchored polyubiquitin chains and analysed precisely what they are composed of. We have developed a completely new method that allows for the first time unanchored polyubiquitin chains to be purified. We intend to purify these molecules from different biological samples, and work out what they contain. We will also investigate what happens to unanchored polyubiquitin chains when cells switch on their protein disposal systems, and when they respond to signals or messages from outside the cell. By investigating what other proteins control the formation or removal of unanchored polyubiquitin chains, and what jobs different unanchored polyubiquitin chains actually do, we will have a much clearer picture of how ubiquitin controls some of the absolutely essential functions of the cell. It is really important to know this information, because although 'normal' processes in the cell are controlled by ubiquitin, many human diseases are actually caused by defects in these processes. For example, in neurodegenerative diseases such as Alzheimer's and Parkinson's, the protein disposal system which ubiquitin normally controls does not function properly. Likewise, in some bone diseases the cellular systems which use ubiquitin to respond to signals from outside the cell are defective. We can only start to develop really effective treatments for these conditions with a proper understanding of the processes that the ubiquitin molecule normally controls.

The influences of ubiquitin (Ub) brought about by the post-translational modification of other proteins have been extensively studied. However, very recent evidence also suggests a physiological role for unanchored (non-substrate linked) polyUb chains in the direct activation of certain protein kinases, and earlier studies indicated unanchored chains may also act as regulators of Ub-mediated proteolysis. To date, the existence of unanchored polyUb in vivo has only been indirectly inferred; using a newly established purification method based on the intrinsic specificity of a Ub-binding domain for the free C-terminus of Ub we can for the first time selectively purify unanchored polyUb (but not substrate-anchored Ub) from cell and tissue sources. Using peptide mass fingerprinting (PMF) and MS/MS we will define the different isopeptide linkages within individual unanchored chains and using the AQUA MS strategy quantitate levels of these different linkages within total unanchored polyUb pools. In addition, we will determine how the constitution of unanchored polyUb pools (individual chains and the overall pool) change upon activation of signalling or proteolysis, and in response to expression changes of known (USP5, CYLD, E225k) and novel (HDAC6) regulators. Further, using 'top-down' protein MS we will map the connectivity (i.e. isopeptide linkages present and order in which they are assembled) of shorter unanchored polyUb chains, information which is lost in 'bottom-up' MS analyses such as PMF and MS/MS. Finally, we will purify individual unanchored polyUb chains and for those of defined linkage(s), assess their biological activity using in vitro kinase and proteasome function assays. These studies will provide the first insight into the molecular composition (isopeptide linkages) of pools of unanchored polyUb, detail of how they respond to physiological stimuli and how they are regulated, and define the 'activity code' of different unanchored polyUb structures.

The major impact of this work is likely to be within the academic research community as it will change the way the field thinks about the regulation of Ub-mediated processes (which themselves are fundamental to the life process). By understanding the role played by unanchored polyUb within different physiological processes, and in particular the functional significance of different polyUb chains of defined stuctures, a re-evaluation of current thinking may be necessary. For example, our own work almost a decade ago [PNAS (2000) 97, 9902-6] indicated that a mutant form of the Ub protein termed Ubb+1 exerts its deleterious effects in Alzheimer's disease brains by 'capping' unanchored polyUb chains leading to proteasome inhibition. New knowledge that unanchored polyUb also directly activates protein kinases (combined with existing knowledge that kinase activity is also aberrant in Alzheimer's disease) means that further investigations in to whether Ubb+1 capped polyUb chains affect kinase activity are now required. However, as indicated in the Impact Statement and Details of Beneficiaries, the work will also be more broadly of interest to those (academic, clinical academic, and pharma environments) in other research fields e.g. cell signalling and inflammatory diseases, or technological areas e.g. the protein MS community. Significantly, our work has the potential to revolutionise the way in which Ub PTMs are analysed, in a similar way that the application of MS to the study of protein glycosylation had a major impact on the analyses of these PTMs.