No description available.

Abnormalities in the coordinated activities of protein phosphatases (PPs) and protein kinases (PKs) contribute to the development of many diseases. Phosphatase of regenerating liver (PRL)-3 and Vaccinia H1-Related (VHR) are two members of the protein tyrosine phosphatase (PTP) family shown to be involved in cancer. PRL-3 is a member of the PRL phosphatases containing a unique post-translationally modifiable prenylation CAAX motif at the carboxy (C)- terminal end. There is an immense body of evidence to support a role for PRL-3 in the development of various types of cancer and in progression to metastasic disease. However, many questions are still pending, especially with respect to the identity of physiological substrates, and interacting partners in general, of PRL-3. VHR is a model for a group of atypical dual specificity protein phosphatases (DUSPs) with a role in cell cycle progression. Only a few of VHR’s physiological substrates have been reported to date and there are very few studies addressing its regulation and physiological role(s). Generally, in order to isolate and identify transient phosphatase-substrate interactions, substrate-trapping mutants of PTPs are employed. Mutants which can function as substrate traps have the ability to recognise and bind substrates, yet they lack functionality of the key catalytic residues and cannot efficiently process the hydrolysis of the substrate. However, it is acknowledged that the efficiency of such standard substrate-trapping mutants of PTPs is low. In this work, the expanded genetic code approach was applied to develop more efficient substrate trapping variants of the PTPs by incorporating the photocross-linkable amino acid para-benzoylphenylalanine (pBPA). The concept was optimised for PRL-3 and VHR, and for both proteins, pBPA-containing variantswere expressed at excellent yields and were highly purified. By utilizing the photo-cross-linkable F68pBPA variant of VHR, dimerisation of VHR was detected in an in vitro ultraviolet (UV) exposure-mediated crosslinking assay. VHR dimerisation was further demonstrated to be a potential novel regulatory mechanism for VHR, having a negative effect on the catalytic activity of the protein. A specific region in VHR known as the variable insert segment was pinpointed as a region in the protein, which is either at the dimer interface or heavily contributing to dimeric association. Furthermore, the intrinsic ability of VHR to self-associate was also demonstrated by complementary methods. For PRL-3 it was demonstrated that its D72pBPA variant could recognise and bind to lipids, with a stronger signal detected in the UV-exposed sample, and without altering the lipid binding profile with respect to the native protein. Lastly, the potential of exploiting photo-cross-linkable variants of the PTPs was also demonstrated by incubating PRL-3 variants, with or without selectively introduced pBPA, with mammalian cell lysates, followed by UV exposure. Western blot analysis detected new bands corresponding to covalently crosslinked PRL-3-protein complexes. Future work in our laboratory will follow up on these newly identified interactions.

Preservation of genetic integrity during mitotic chromosome segregation requires the attachment of sister chromatids to microtubules emanating from opposite poles of the mitotic spindle. Successful mitosis and cell cycle progression critically depend on the achievement of this attachment state, otherwise called amphitelic attachment or biorientation. Cells therefore operate an error‐correcting system at the kinetochore, which is the main microtubule attachment site on chromosomes. This system selectively removes incorrect attachments while leaving amphitelic attachments unaltered. Error correction of microtubule attachments heavily depends on integration of kinase‐phosphatase signalling. The data presented in this thesis suggest that one mechanism of fine‐tuning this phospho‐regulation at the kinetochore is the regulation of protein phosphatase 2A (PP2A) by the small kinetochore protein Biorientation defective 1 (Bod1).

The Twin Arginine Translocase (Tat) system is a membrane-bound transport system present in plants and bacteria that has the remarkable ability to export fully folded proteins across a lipid bilayer, powered by the protonmotive force. In Escherichia coli the Tat system is composed of only three essential membrane proteins, the homologous TatA and TatB proteins, and the highly hydrophobic TatC core subunit. The transport mechanism involves a TatBC recognition complex which binds substrates through their signal peptides, with TatA being recruited to form an oligomeric structure that facilitates protein transport. Current models suggest that TatA oligomers locally disrupt membrane lipids allowing the substrate to move from the cytoplasm to the periplasm. Here, biochemical methods were used to further understand interactions between these components. Affinity purification experiments in detergent solution, using a His-tagged TatC variant, resulted in co-purification of TatB along with a small amount of associated TatA. The amount of co-purifying TatA was notably increased when TatB was absent, suggesting that the proteins may compete for the same binding site on TatC. Disulphide crosslinking experiments performed in vivo identified a binding site for TatA at the sixth transmembrane helix of TatC occupied in the resting state, indicating that TatA interacts constitutively with the TatBC recognition complex. Further crosslinking experiments found that the protonmotive force was required for TatA to occupy this constitutive site, explaining the poor yields of TatA co-purifying with TatC in detergent solution. In response to increased substrate flux through the Tat pathway, disulphide crosslinking demonstrated that TatA no longer occupied its constitutive site, instead binding at the fifth transmembrane helix of TatC, a site previously shown to be occupied by TatB. This supports the proposition that TatA and TatB occupy this site differentially, with TatB potentially acting as a “gatekeeper” to modulate TatA polymerisation. Interaction sites were identified between the transmembrane regions of TatA and TatB when substrate was overexpressed. Using these crosslinking data, a model was produced which presented interfaces between TatA, TatB and TatC in the resting state and how these change during interaction with a Tat substrate.

Cullin RING E3 ubiquitin ligases (CRLs) function in the ubiquitin proteasome system by catalysing the transfer of ubiquitin from E2 conjugating enzymes to specific substrate proteins. CRLs are large dynamic multi-subunit complexes that control the fate of many proteins in cells and, therefore, constitute attractive drug targets for the development of small-molecule tools and potential drug leads, such as inhibitors and chemical inducers of protein degradation. This work presents the first crystal structure of the pentameric human CRL2VHL complex, composed of Cul2, Rbx1, Elongin B (EloB), Elongin C (EloC) and pVHL. The structure presents a closed state of full-length Cul2 and a new conformation of Rbx1, thought to be in a trajectory from inactive to active state. The thermodynamic signature of the interaction between Cul2 and pVHL-EloBC (VBC) was determined as well as mutations that contribute toward a selectivity switch for Cul2 versus Cul5 recognition. In addition, this work focused on an extensive approach to probe the VBC surface with peptides. A first methodology involved the structure-based design aimed at targeting the Cul2 VBC interaction. Three peptides have been shown to bind at the Cul2 binding site on EloC, however, with very weak affinities that could not be optimised, suggesting its poor ligandability. The second methodology included an unbiased screening using phage display libraries of bicyclic peptides. The screening campaign yielded peptide hits but their physicochemical properties, especially poor solubility, constituted an obstacle to the progress toward a biophysical characterisation of their binding to VBC and the development into high-affinity probes. The findings of this work provide structural and biophysical contributions into the whole CRL2VHL complex assembly and functioning and provide insights and tools that could aid future targeting of this CRL.

The AMP-activated protein kinase (AMPK) is a sensor of cellular energy that is activated by increases in the intracellular AMP:ATP or ADP:ATP ratios. Once active, AMPK activates catabolic pathways and inhibits anabolic pathways to restore cellular energy homeostasis. AMPK activity can be increased by phosphorylation of Thr172, a conserved residue in the activation loop of the kinase domain, by either LKB1 or CaMKKß. Increases in Thr172 phosphorylation can be mediated by either protecting this site against dephosphorylation by protein phosphatases, or by promoting its phosphorylation by upstream kinases, both of which were proposed to be mediated by binding of AMP or ADP to the ?-subunit of AMPK. AMPK is also allosterically activated by binding of AMP, but not ADP. Recently, ADP has been proposed as the major regulator of AMPK and the role of AMP has been questioned. The reasons for this are: (i) while both AMP and ADP can increase phosphorylation of Thr172, ADP is present at higher concentrations in the cell and it was proposed that AMP would be unable to compete with ADP for binding at the ?-subunit; (ii) allosteric activation by AMP was reported to increase AMPK activity by less than 2-fold, whereas changes in the phosphorylation of Thr172 can increase AMPK activity by 100-fold. Using cell-free systems and intact cells, the regulation of Thr172 phosphorylation and AMPK activity by AMP, ADP and ATP was re-investigated. AMP promoted Thr172 phosphorylation by LKB1 but not by CaMKKß, while ADP had no effect on Thr172 phosphorylation by either upstream kinase. Additionally, while both AMP and ADP could protect Thr172 against dephosphorylation, AMP was more potent. The allosteric activation of AMPK by AMP was demonstrated to be a significant component of the overall activation mechanism. Using phosphorylation of ACC (a downstream target of AMPK) as a marker, allosteric activation of AMPK was observed under conditions where there were no changes in Thr172 phosphorylation. The changes in Thr172 phosphorylation and AMPK activity observed in intact cells in response to AMPK activators were also examined, and demonstrated to be far lower than the maximal effects observed in cell-free systems. Taken together, these results demonstrate that AMP is a true physiological regulator of AMPK activity. The regulation of LKB1 has also been the subject of much interest. Whilst it appears that LKB1 is constitutively active, several groups have reported that LKB1 activity is regulated by post-translational modifications, particularly phosphorylation. The role of phosphorylation of LKB1 on Ser31 by AMPK was investigated. While AMPK could phosphorylate LKB1 in a cell-free system, an effect lost in an LKB1 [S31A] mutant, this phosphorylation was not observed in intact cells. Additionally, mutation of Ser31 did not appear to alter LKB1 activity. These results suggest that phosphorylation of Ser31 of LKB1 by AMPK, whilst possible, may not be a physiologically relevant event. The mTOR pathway is a key regulator of cell growth and protein synthesis in response to a number of stimuli, including growth factors, nutrients and energy status. The mTOR pathway is frequently deregulated in a number of cancers and cancer syndromes, and inhibition of mTOR is a promising therapeutic approach. The nature of the genetic lesion in tumours may determine the response of cells to mTOR inhibition. The LKB1-AMPK pathway, as well as being a negative regulator of the mTOR pathway, is also frequently inactivated in human cancers. The role that the LKB1-AMPK pathway plays in determining the cellular response to AZ4, a dual mTORC1 and mTORC2 inhibitor, was investigated. AZ4 induced apoptosis in a number of cell lines, confirming its potential as an anti-cancer therapeutic. However, the presence of an active LKB1-AMPK pathway protected cells against AZ4-induced apoptosis, suggesting that the status of this pathway may determine how cells respond to mTOR inhibition.

The chromosomal passenger complex (CPC) is a conserved, essential protein complex in eukaryotes, consisting of Ipl1/Aurora B kinase and three regulatory subunits: Sli15/INCENP, Bir1/Survivin and Nbl1/Borealin. The CPC has numerous cell cycle roles, the best characterised of which is the regulation of kinetochore-microtubule attachments during metaphase to promote chromosome bi-orientation on the mitotic spindle. To efficiently perform these functions, the CPC must be properly regulated and localised. The original defining characteristic of 'chromosomal passenger' proteins was their dynamic localisation: associated with chromatin and centromeres during early mitosis followed by translocation to the spindle midzone during anaphase. However, exactly how the CPC is properly localised and regulated is not fully understood. Recent studies demonstrated that phosphorylation of the CPC regulatory subunits by cyclin-dependent kinase (CDK) regulates localisation of the complex. For example, in budding yeast and human cells the dephosphorylation of CDKphosphorylated sites in Sli15/INCENP is required for the efficient translocation of the CPC from centromeres in metaphase to the spindle in anaphase. Many Ipl1/Aurora B homologues also contain CDK phosphorylation motifs but whether modification of these sites is required for proper CPC localisation and function is not known - here this was investigated in budding yeast. Preventing phosphorylation of CDK consensus sites in the N-terminus of Ipl1 was found to result in its premature localisation to the spindle in metaphase; this was dependent on the yeast microtubule binding EB1 homologue Bim1. However, preventing or mimicking phosphorylation at these sites had no effect on cell viability or chromosome bi-orientation. During the course of this study, several groups demonstrated that the conserved Haspin kinase in fission yeast, Xenopus and human cells contributes to CPC localisation at the centromere during metaphase by phosphorylating histone H3 on Threonine 3 (H3- T3). In light of these findings, two budding yeast Haspin kinase homologues (Alk1 and Alk2) were investigated. Although roles for Alk1 and Alk2 in histone H3 phosphorylation and CPC recruitment could not be conclusively determined, unusual phenotypes were observed in cells lacking both Alk1 and Alk2. In these cells, the spindle was frequently confined to the bud and incorrectly aligned. If these cells were arrested in metaphase, the bud became significantly larger than the mother cell. When anaphase was eventually triggered, the spindle elongated and then broke down entirely within the bud. These abnormalities were independent of histone H3-T3 phosphorylation, since mutation of H3-T3 to non-phosphorylatable alanine did not result in these phenotypes. Thus Alk1 and Alk2 share novel and redundant roles in spindle positioning, cell polarity and bud growth that are particularly important during a metaphase arrest.

The lipid kinase VPS34 is an essential mediator of multiple aspects of intracellular trafficking via the formation of phosphoinositide-3-phosphate (PI(3)P) on membranes, this is critical to mediate efficient trafficking of cargo by endocytosis. In addition, VPS34 kinase activity is essential for inducing autophagy in combination with the protein kinase ULK1. Autophagy acts as a catabolic pathway that is up regulated in response to stress, but is negatively regulated by the master growth protein kinase mTOR. This study sought to identify novel control mechanisms that may mediate the regulation of VPS34 between the pathways of endocytosis and autophagy. During nutrient rich conditions a binding protein of VPS34, UVRAG, was identified to be phosphorylated. Further analysis has identified that phosphorylation is mediated by mTOR and that this occurs at two sites, S550 and S571. Multiple lines of evidence suggest that phosphorylation does not alter the stoichiometry or localisation of the complex nor does it mediate recruitment of additional factors. Phosphorylation of UVRAG acts to increase lipid kinase activity in vitro and cellular PI(3)P levels by ~ 2 fold, mutation of S550 and S571 to alanine residues abrogate this increase in activity. Examination of autophagy, receptor mediated endocytosis and recycling have demonstrated no effect of UVRAG phosphorylation upon their rate of trafficking. Mutation of UVRAG phosphorylation sites however leads to a significant lysosome abnormality that is demonstrated by a dispersed phenotype. Preliminary analysis suggests that this may occur due to abnormalities in the process of autophagic lysosome reformation, a process that is dependent upon the activity of mTOR. This suggests a previously uncharacterised role of PI(3)P in lysosomal regulation and adds to current understanding of regulation between VPS34 and mTOR. Additionally data presented here examines a novel PI3KC3 inhibitor that demonstrates profound selectivity over other PI3K isoforms and lipid kinases. This will be important to further examine the functional role of VPS34 and UVRAG.

Proteins are the fundamental building blocks of all living cells and they carry out nearly all cell functions. Proteins predominantly act as part of multiprotein complexes to carry out these functions. The association of protein isoforms and post-translationally modified forms in protein complexes can influence their subcellular location, activity and substrate specificity. It is therefore crucial to characterise protein complexes at the level of the protein isoforms and post-translationally modified forms they contain to fully decipher the network of signalling and regulatory pathways within cells. The aim of my work has been to develop a technique to study protein complexes through the use of size exclusion chromatography (SEC) combined with tandem mass spectrometry. Combining these approaches has enabled an in-depth analysis of protein complexes in U2OS cells, including those involving post-translationally modified proteins and protein isoforms. The data presented in this thesis provide a proof of concept, together with forming a useful resource, which can be used alongside pull-down analyses to differentiate the interaction partners involved in different protein complexes. This combined approach minimises the need for multiple IP analyses and facilitates a more targeted approach in dissecting the components of individual protein complexes. I developed this technique further by utilising in vivo crosslinking prior to denaturing SEC. This approach enabled more efficient recovery and detection of proteins previously underrepresented using native SEC analysis, including many membrane complexes, thereby providing a more complete picture of endogenous protein complexes. I have applied the native SEC/MS approach in a study of the interactions between the MRFAP1-MORF4L1 proteins. I demonstrated that this complex is distinct from the larger complex involving interactions between the MRGBP-MORF4L1 proteins. In addition, I also demonstrate that the native SEC/MS technique can be extended to assess the effect of drugs on protein-protein interactions. Overall, the methods I present in this thesis enable the rapid, proteome-wide analysis of endogenous protein complexes, which will advance the future study of protein complexes in biology.

No description available.