Translating the ubiquitin code in mitotic cells

Regulated gene expression ensures that cells make the correct selection of genetically encoded protein components required for their function. However, cells also require a method to get rid of proteins once they are no longer required, or if they are faulty. Cellular 'digestion' of unwanted proteins is called proteolysis, and occurs very rapidly inside subcellular machines called proteasomes. Proteins are directed into proteasomes by specific tagging with multiple copies of a small ubiquitous protein known as 'ubiquitin', which forms chains that can be recognized by receptors on the lid of the proteasome. Ubiquitin tags also mediate other functions unrelated to proteolysis, and our growing knowledge of how different types of ubiquitin chain direct different outcomes has given rise to the concept of a 'ubiquitin code'.
A large fraction of the human genome encodes components of the Ubiquitin-Proteasome System (UPS), because almost all cellular processes require ubiquitin-mediated control. For example in regulation of cell division (mitosis) precise destruction of several key components, at exactly the right time and place, drives the whole process. The complexity of the UPS means that ubiquitin chains come in many varieties, and some are better than others at directing targeted proteins to the proteasome. In other words, the proteolysis of different proteins happens at different rates, depending on what type of ubiquitin chain they carry.
We still don't know very much about this part of the ubiquitin code, and how it is translated into proteolysis.
This proposal focuses on proteolysis of two key regulators of mitosis, called the Aurora kinases (A and B). Both are targeted by the same UPS pathway, but they are destroyed at very different rates at the end of mitosis. The resulting difference in timing of their eventual disappearance is critical to the correct sequence of events at mitotic exit. Our preliminary studies indicate that the different rates of proteolysis arise from differences in the ubiquitin code assembled on Aurora A versus Aurora B. We propose a detailed study of the ubiquitin code applied to Aurora kinases, to understand the part of the ubiquitin code that directs rapid destruction at the proteasome. We will use our new knowledge to design reagents that can be used to artificially manipulate cellular levels of Aurora kinases, or of other cellular targets to which they bind, by harnessing the UPS. A new generation of targeted therapies currently in development (called Protein Targeting Chimeras, or PROTACS) will in the future deliver the ability to target cellular proteins that are faulty, or expressed in the wrong time and place, as is often the case in disease. Understanding the ubiquitin code will assist the future design of these tools.

Progress in deciphering the ubiquitin code has relied on in vitro approaches using artificial ubiquitin chains. How are ubiquitin chains assembled and processed in vivo? In recent years we have developed tools to purify ubiquitinated proteins from cells and interrogate the composition of ubiquitin chains they carry, using linkage-specific antibodies and deubiquitinases. We discovered that two substrates, AURKA and AURKB, strongly conserved and both targeted by the Anaphase-Promoting Complex (APC/C) ubiquitin ligase, carry different configurations of ubiquitin chains. This difference translates into markedly different rates of proteolysis of the two substrates. AURKA and AURKB therefore present an important model for studying parameters of ubiquitin chain assembly and processing.
We propose to identify substrate-specific and signaling parameters mediating differential ubiquitin chain assembly on AURKA and AURKB, and the switch in ubiquitin chain specificity required for AURKA destruction. The study will focus on divergent N-terminal IDRs that are proposed to contain a number of functional Short Linear Interacting Motifs (SLiMs). We will use cell-based degradation assays and ubiquitination assays to correlate measurements of substrate proteolysis with substrate-specific ubiquitin linkage assembly, and in vitro proteasome binding and proteolysis assays using endogenously ubiquitinated substrates purified from mitotic cells. We will examine the role of SLiMs, and of cellular signaling pathways proposed to regulate them. Integration of bioinformatic and biophysical approaches with functional assays will advance the SLiM model of the proteome, describing molecular detail that underpins the notion of AURKA as a signalling hub.
Finally, we propose to design constrained peptide tools based on AURKA SLiMs and to test their ability to modulate AURK stability and function. These can be used to design chimeric peptides (PROTACS) to target other proteins in the cell.

Our research will have beneficiaries within academia and within the commercial sector (biotechnology, pharmaceutical) sectors and will benefit the general public on various timescales.
Academia will benefit through our acquisition and dissemination of new knowledge, through training of young scientists and future scientists in a world-class environment and through the new tools and methodology that we will generate, that will be shared with fellow academics upon request.
The commercial sector will benefit through new knowledge of cellular mechanisms relevant to disease and to the design of new therapeutic tools. New methodologies that we develop and promote in this research (purification of ubiquitin conjugates, characterization of degrons, measurements of protein stability, peptide design) are highly exploitable for drug discovery in ubiquitin-mediated pathways.
The general public will benefit from the potential healthcare benefits of the theoretical knowledge we generate and the biological tools we develop that may have therapeutic application in the future.