From our early classes in biology, we have been taught that proteins are the ‘molecular machinery’ of the cell, and that they carry out the cellular functions within the cell. At a top-down level, the amount of proteins present within a cell is a balance between its synthesis and degradation, the dynamic regulation of which is referred to as proteostasis. Altered proteostasis can be the cause of many diseases such as cancer or neurological diseases. Therefore, being able to access tools that can study those processes or even actively perturb the balance to reduce or increase levels of a protein is of great scientific and therapeutic interest. In the first part of this study, I sought to develop novel ways to study the relative stability of Aurora Kinase A (AurKA). AurKA is a serine/threonine protein kinase that has well documented functions in the onset and progression of mitosis by promoting centrosome maturation and mitotic spindle assembly. The cognate pathway of AurKA degradation is through the ubiquitin proteasome system (UPS), a series of catalytic steps including the APC/C E3 ligase, resulting in the ubiquitination and subsequent degradation by the proteasome. Recently, more work has been done in exploring the non-mitotic roles of AurKA as its overexpression and mis-regulation has been observed in different cancers. These non-mitotic roles have often been thought to be kinase-independent and therefore cannot be targeted by conventional therapy with kinase inhibitors. Therefore, targeting the stability of the protein could be a better therapeutic strategy, and to this end, the second half of my work aimed to characterise a novel targeted protein degradation tool for AurKA.

In establishing a tool to monitor AurKA turnover, I utilised tandem fluorescent timer tags (tFTs) tagged to AurKA. This ratiometric method relies on the fluorescent intensities of two fluorophores with different maturation rates to provide a readout of the protein’s age and hence its relative stability. First, I created some MATLAB models that allowed the prediction of changes in tFT ratios in response to changes in synthesis and degradation parameters, in order to aid the interpretation of tFT behaviour. I next tested different combinations of tFTs for AurKA and found that the mRuby-mNeonGreen tag was optimal. After creating stable cell lines expressing the AurKA-mRuby-mNeonGreen tFT, I validated their functionality in reporting protein turnover in live imaging of single cells. The tFTs were successfully able to report on changes in the relative stability of AurKA through the cell cycle. I showed that tFTs can also respond to experimentally induced perturbations in AurKA stability. Knocking down its stabilising interacting partner, TPX2, or knocking down Cdh1, a critical component of AurKA’s cognate ubiquitin ligase, resulted in decreased and increased relative AurKA stability respectively, as reported by a decreased or increased tFT ratio (mRuby:mNeonGreen). I also tested the closely related Aurora Kinase B (AurKB) tFT and found that knocking down or overexpressing a de-ubiquitinating enzyme, USP35, recently shown to regulate AurKB turnover, resulted in a decrease and increase of tFT reported stability respectively. Further characterization experiments additionally demonstrated that tFTs show localised differences in AurKA-tFT ratio that may indicate differences in stability between the nucleus and cytoplasm. Importantly, I also demonstrated that analysis of single time-point images of tFTs was able to report changes in AurKA stability as well as time-lapse experimentation, making data acquisition and analysis simpler. The development of a tFT for measuring AurKA stability opens up the possibility of using it as a high-throughput tool to screen for unknown regulators and interactors of AurKA governing its stability, offering potentially new therapeutic targets.

In the latter half of the work for this PhD, I characterised a novel targeted protein degradation tool synthesised by AstraZeneca. The novel protein degrader belongs to a new family of compounds called proteolysis targeting chimeras (PROTACs) and consisted of MLN8237, a small molecule inhibitor of AurKA, linked to pomalidomide, a chemical ligand for the Cereblon (CRB) E3 ligase. The PROTAC works by ectopically recruiting AurKA to CRB, resulting in its ubiquitination and subsequent degradation by the UPS. I showed using live imaging of single cells, immunoblotting and immunofluorescence, that the PROTAC efficiently and specifically degrades AurKA. I also reported localisation-dependent effects on AurKA degradation, revealing that centrosomal AurKA is resistant to PROTAC-mediated depletion. In a similar vein, I showed that there was enhanced nuclear degradation of wild-type AurKA compared to cytoplasmic degradation. A more strongly nuclear localised version of AurKA (truncation mutant Δ67) was degraded better in response to PROTAC than the wild-type. I have found by immunofluorescence that CRB exhibits nuclear localisation. An interesting avenue for further characterisation of this PROTAC will be whether the preferential targeting of different subcellular pools of AurKA arises from the localisation of CRB or of the PROTAC, or of other parameters. I also investigated the phenotypic outcomes of PROTAC-mediated degradation compared to AurKA inhibition with MLN8237. Herein I reported that AurKA degradation by the PROTAC resulted in shorter mitotic spindles without significant mitotic defects. This was in contrast to severe mitotic defects seen with AurKA inhibition alone, even at very low doses. It’s believed that this shortening of the spindle phenotype arises from the selective degradation of AurKA on the spindle. Outside of mitosis, I found that PROTAC treatment was able to rescue mitochondrial fragmentation induced by stabilization of AurKA in interphase in Cdh1 knockout cells.

Having a tool in the form of a PROTAC that can rapidly and efficiently degrade AurKA not only opens up therapeutic avenues, but it is also a powerful experimental tool for further dissection of AurKA’s roles, both kinase-dependent and non-kinase roles. It is hoped that in the future, with further development and characterisation of tFT and PROTAC tools, the importance of AurKA and its regulated stability within and outside of mitosis can be further understood.

Finally, in bringing the two tools together and showing their functionality, I demonstrated that the AurKA tFT is able to report the increased turnover of AurKA in response to PROTAC. This raises the possibility of using ‘stability biosensors’ such as tFTs for drug discovery in the PROTAC field.