published_or_final_version / Chemistry / Doctoral / Doctor of Philosophy

Ubiquitin carboxy-terminal hydrolase L1 (UCH-L1), a protein of 223 amino acids, is a member of deubiquitinating enzymes and it is one of the most abundant proteins in the brain. Although the in vivo functions of UCH-L1 are still unclear, its abundance and specificity for neurons indicate that it may serve an important role in neuronal cell function or dysfunction. Indeed, an isoleucine 93 to methionine amino acid mutation (I93M) in UCH-L1 was identified to be linked to an autosomal dominant form of Parkinson’s disease, while the serine 18 to tyrosine amino acid mutation (S18Y) in UCH-L1 is linked to a decreased susceptibility to Parkinson’s disease. To investigate the effects of these mutations on the structure of human UCH-L1, the mutant proteins have been successfully over-expressed, biophysically characterized and compared with the wild-type UCH-L1 using circular dichroism and NMR spectroscopy. While the data from circular dichroism and NMR chemical shift perturbation analysis suggested that the S18Y point mutation only slightly perturbs the global structure, the effect of the I93M point mutation was found to be more profound. In particular, the structural perturbations caused by I93M substitution are not only observed near the site of mutation, but are also found at more distant sites. These structural perturbations may be significant for the function of UCH-L1 and explain the reduced hydrolase activity (~55 % of wild-type) observed in UCH-L1-I93M, as the geometry of the catalytic triad (C90, H161 and D176) is likely to be distorted by this substitution. To provide further insights into the effect of serine 18 to tyrosine (S18Y) mutation on the structure and function of UCH-L1, the three-dimensional solution structure of UCH-L1-S18Y was determined by NMR spectroscopy. The solution structure of UCH-L1-S18Y reveals a monomer with a typical fold of papain-like cysteine proteases and consists of a six-membered antiparallel β-sheet surrounded by eight α-helices. Although the global structure is very similar to the crystal structure of wild-type UCH-L1, both the altered hydrogen bond network and the surface charge distributions have demonstrated that the S18Y substitution could lead to profound structural changes. In particular, the analysis of the difference in the dimeric interfaces of the wild-type and the S18Y mutant showed that the serine to tyrosine mutation can significantly affect the distribution of the surface-exposed residues involved in the dimeric interface. It is thought that such observed differences might weaken the stability of the UCH-L1 dimer and hence may explain the reduced dimerization-dependent ligase activity of UCH-L1-S18Y in comparison to the wild-type UCH-L1. / published_or_final_version / Chemistry / Doctoral / Doctor of Philosophy

DNA aptamers have been studied since their inception in 1990, but have only targeted membrane and serum proteins in therapeutics. Their potential as inhibitors of protein function is hampered by their inability to efficiently enter cells in order to function. Surmounting this hurdle is worthwhile since the inhibition of protein-protein interactions is not achievable by small molecule pharmaceuticals alone. Herein we target an intracellular ubiquitin ligase WWP1, which is known to complex with Schnurri3 and polyubiquitinate Runx2, thus targeting it for proteosomal destruction. Since Runx2 is the key transcriptional regulator of osteoblast differentiation, WWP1 inhibition may encourage osteoblast differentiation, and by extension force bone deposition in osteoporosis sufferers. By targeting WWP1 we attempt to intervene in intracellular protein interactions with an aptamer for the first time. To begin this effort we cloned, expressed, and purified three functionally important truncations of WWP1. The final protein pools were highly concentrated above 2 mg/mL, approximately 95% pure, and were found to be acceptably soluble after assessment in various buffers. DNA aptamers were then selected against these WWP1 truncations using the established SELEX method while monitoring the progression of the enrichment with PCR. After 12 selection rounds of increasing stringency, pools were sequenced and assessed for homogeneity and secondary structure. Several groups of enriched and identical DNA sequences were obtained with no obvious pattern in secondary structure seen between them. While focusing on sequences specific for the active site containing C-lobe, we then evaluated the aptamers for their ability to bind key functional regions of WWP1 and inhibit its function as an enzyme. For the most potent aptamer from the C-lobe pool, an Electrophoretic Mobility Shift Assay (EMSA) estimated a Ki of around 2 μM. Furthermore, a HECT ubiquitin ligase activity assay was developed to evaluate inhibition, and an IC50 of around 100 μM was found for the most inhibitory of three C-lobe aptamers. This aptamer was then transfected into SaOS-2 osteoblastic cells so that localization could be assessed with fluorescence microscopy. Surprisingly, both the C-lobe specific aptamer and a control sequence were found to enter the cells with or without the employment of transfection reagent. Moreover, approximately 60% migrated to the nucleus and remained there over a period of days, which implies diffusion through the Nuclear Pore Complex. Taken together, this work introduces an alternative approach to disease therapy by targeting intracellular proteins with aptamers, and may have significant implications for expanding the therapeutic applications of nucleic acid aptamers in the future. / published_or_final_version / Biochemistry / Doctoral / Doctor of Philosophy

No description available.

No description available.

No description available.

The ubiquitin-proteasome system regulates protein degradation. Although proteasomes localize in the nucleus of proliferating Saccharomyces cerevisiae, they are sequestered into cytoplasmic proteasome storage granules (PSG) in quiescence. Although important for cell cycle regulation and mediating external stressors, the content and structure of these membraneless PSGs remain unknown. Yeast deletion genetic screens identified several ubiquitin-related genes involved in proteasome sequestration into PSGs. This study aims to determine whether changes in free ubiquitin levels or ubiquitin post-translational modifications affect proteasome dynamics in quiescence. Unlike the wild-type, PSGs were not seen in catalytically inactive Ubp6 mutant strains in quiescence and proteasomes failed to be imported into the nucleus upon the resumption of cell growth. Although no significant differences in proteasome configurations were observed, Western blot analysis of these mutants suggests the presence of post-translationally modified monoubiquitinated proteasomes in quiescence. Ubiquitin modification may target proteasomes towards lysozomal degradation rather than into PSGs for storage.

The cellular levels of a number of proteins have been found to be regulated by the ubiquitin-proteasome pathway. In this pathway, proteins are covalently tagged (&ldquo / ubiquitinated&rdquo / ) by ubiquitin, which acts as a signal for degradation by the proteasome. A number of key cellular processes, including cell-cycle progression, transcription and DNA repair, are regulated in this way. In recent years a number of cellular proteins resembling ubiquitin in structure or function, the so-called ubiquitin-like proteins, have been identified. Ubiquitin-like proteins can be divided into two classes-the so-called &ldquo / ubiquitin-like modifiers&rdquo / , which consist of a single domain that structurally resembles ubiquitin, and &ldquo / ubiquitin-domain&rdquo / proteins, which are multi-domain proteins, which include domains that resemble ubiquitin.<br /> <br /> This thesis describes the recombinant expression, purification and full backbone assignment of the human DWNN domain, a novel ubiquitin-like domain. The DWNN domain occurs at the N-terminus of RBBP6, a protein that has been shown to interact with p53 and Rb as well as to be involved in mRNA processing and apoptosis. A bacterial expression system was used to overexpress the DWNN domain as a GST fusion protein. The domain was labelled with 15N and 13C to perform triple-resonance heteronuclear NMR experiments, from which full backbone assignments were obtained.<br /> <br /> Although full structure determination of the DWNN domain falls outside the scope of this thesis, the backbone assignments formed the basis for the subsequent structure determination, which confirmed that the DWNN domain is indeed a novel ubiquitin-like domain. The RBBP6 protein may therefore represent a novel E3 ubiquitin ligase that plays a role in regulating the cellular levels of p53 and Rb.

The anaphase promoting complex (APC/C) is an ubiquitin ligase that is an essential regulator of multiple steps in the cell cycle. The complex consists of at least 12 subunits with a catalytic core formed by a scaffold protein, APC2, and a RING-H2 protein, APC11. The Parapoxvirus, Orf virus (OV), encodes a RING-H2 protein, B5L, with clear sequence similarities to APC11. The disruption of APC/C function leads to pre-mature entry into S phase and a delayed M phase exit and, potentially, apoptosis. This investigation explored the functional significance of the similarity between B5L and APC11 and specifically sought to determine if B5L manipulates cell cycle regulation by targeting APC/C function. Co-immunoprecipitation experiments from lysates of cells expressing a range of constructs revealed an interaction between B5L and APC2 in the same manner as seen with APC11. Furthermore, B5L was found to associate with endogenous APC/C. However, although APC11 promoted the formation of polyubiquitin chains in substrate-independent in vitro assays, B5L was inactive in this assay. Bioinformatics comparisons of APC11 and other known RING ubiquitin ligases with B5L and its poxviral homologues revealed some subtle differences. In particular a domain of APC11 (amino acids 61-74), that is essential for its ubiquitin ligase activity is not conserved in B5L or its homologues. When this APC11 domain was incorporated in place of the corresponding region of B5L (amino acids 59-67), the mutated B5L acquired ubiquitin ligase activity. On the other hand, APC11 protein in which the domain was replaced with that of B5L lost ubiquitin ligase activity. Stable cell lines expressing B5L showed an increased number of cells in G2/M phase (30�4%) compared with cell lines expressing APC11 (11�2%, n=3, p<0.05, ANOVA, Tukey�s), consistent with impaired APC/C function. APC/C substrates such as cyclin A, cyclin B and the thymidine kinase were stablized in B5L-expressing cells compared with control cells. Furthermore, transient hyper-expression of B5L induced apoptosis in 25�2% (n=3, p<0.05) of the cell population compared with only 6�1% apoptotic cells when APC11 was hyper-expressed. Analysis of the DNA content of OV-infected cells revealed enhanced DNA synthesis compared with cells infected with a B5L knockout OV. These observations indicate that B5L is a non-functional mimic of APC11. It associates with APC/C, but lacks ubiquitin ligase activity, and hence disrupts APC/C function. These abilities may enable OV to induce a cellular environment that enhances viral replication.

The anaphase promoting complex (APC/C) is an ubiquitin ligase that is an essential regulator of multiple steps in the cell cycle. The complex consists of at least 12 subunits with a catalytic core formed by a scaffold protein, APC2, and a RING-H2 protein, APC11. The Parapoxvirus, Orf virus (OV), encodes a RING-H2 protein, B5L, with clear sequence similarities to APC11. The disruption of APC/C function leads to pre-mature entry into S phase and a delayed M phase exit and, potentially, apoptosis. This investigation explored the functional significance of the similarity between B5L and APC11 and specifically sought to determine if B5L manipulates cell cycle regulation by targeting APC/C function. Co-immunoprecipitation experiments from lysates of cells expressing a range of constructs revealed an interaction between B5L and APC2 in the same manner as seen with APC11. Furthermore, B5L was found to associate with endogenous APC/C. However, although APC11 promoted the formation of polyubiquitin chains in substrate-independent in vitro assays, B5L was inactive in this assay. Bioinformatics comparisons of APC11 and other known RING ubiquitin ligases with B5L and its poxviral homologues revealed some subtle differences. In particular a domain of APC11 (amino acids 61-74), that is essential for its ubiquitin ligase activity is not conserved in B5L or its homologues. When this APC11 domain was incorporated in place of the corresponding region of B5L (amino acids 59-67), the mutated B5L acquired ubiquitin ligase activity. On the other hand, APC11 protein in which the domain was replaced with that of B5L lost ubiquitin ligase activity. Stable cell lines expressing B5L showed an increased number of cells in G2/M phase (30�4%) compared with cell lines expressing APC11 (11�2%, n=3, p<0.05, ANOVA, Tukey�s), consistent with impaired APC/C function. APC/C substrates such as cyclin A, cyclin B and the thymidine kinase were stablized in B5L-expressing cells compared with control cells. Furthermore, transient hyper-expression of B5L induced apoptosis in 25�2% (n=3, p<0.05) of the cell population compared with only 6�1% apoptotic cells when APC11 was hyper-expressed. Analysis of the DNA content of OV-infected cells revealed enhanced DNA synthesis compared with cells infected with a B5L knockout OV. These observations indicate that B5L is a non-functional mimic of APC11. It associates with APC/C, but lacks ubiquitin ligase activity, and hence disrupts APC/C function. These abilities may enable OV to induce a cellular environment that enhances viral replication.