Human cytochromes P450 (CYP) constitute one of the most important and studied classes of phase I drug metabolizing enzymes. A small group of 6-7 isoforms of the 57 identified until now accounts for 90-95% of the metabolism of clinically used drugs and can contain mutations (single nucleotide polymorphisms) that, when located in the coding regions, can lead to absent, deficient or enhanced enzyme activity. Thanks to the development of pharmacogenetics and later on pharmacogenomics, the presence of these single nucleotide polymorphisms (SNP) has been associated to inter-individual and inter-ethnic variability in the response to several important therapeutic agents. Due to this direct correlation between polymorphism and efficacy of drug treatment, the pharmaceutical industry is particularly interested in developing new analytical tools for the determination of the cytochrome P450 in vitro metabolism to correlate to the in vivo situation. The use of amperometric sensing systems represents an interesting alternative to the traditional methods. In the present study an electrochemical characterisation of the human cytochrome P450 2C9 and its two main allelic variants the CYP2C9*2 and CYP2C9*3 have been performed. Different methods of immobilisation and electrodes have been used to investigate the conditions that are more suitable to maintain and guarantee the active state and biocatalytic response of an immobilised cytochrome P450. The CYP2C9 and its two main allelic variants CYP2C9*2 and CYP2C9*3 have been inserted into engineered constructs where the human cytochrome P450 gene is linked to artificial redox chains to regulate the electron flow from the electrode surface to the haem. All the constructs have been successfully expressed and purified in heterologous E.Coli cells, with the CYP2C9FLD chimeras showing higher yields. Preliminary spectroelectrochemical studies on semi-conductive and optically transparent tin-dioxide allowed the combination of absorption spectroscopy and electrochemistry (cyclic voltammetry) to ascertain the native state of these P450 enzymes once immobilised. The results showed how the conversion from the active P450 form to the inactive P420 one can be achieved by modifying the surface with polycations with slightly different chemical properties. A proper characterisation of the two species was performed for the first time and, under certain conditions, the inactive P420 species appeared to dominate the cyclic voltammogram (CV). Similar findings have been observed when the CYP2C9 wild type have been electrochemically characterised on DDAB and PDDA modified glassy carbon electrodes. Electrocatalysis in presence of S-warfarin and FT-IR spectra of the CYP2C9/DDAB/GC electrodes revealed higher catalytic activity and maintenance of the native secondary structure of the enzyme. Immobilisation of the CYP2C9FLD, CYP2C9*2FLD and CYP2C9*3FLD on DDAB modified glassy carbon electrodes showed well defined redox couples on the oxygen-free cyclic voltammograms (CVs) and mid point potentials of all enzymes were calculated. Electrocatalysis in presence of substrate and quantification of the product formed showed lower catalytic activities for the CYP2C9*3FLD and CYP2C9*2FLD compared to the wild type CYP2C9FLD as it was expected from literature data. When the CYP2C9FLD, CYP2C9*2FLD and CYP2C9*3FLD were immobilised on alkanethiol modified gold electrodes, the system was tested as both amperometric sensor and catalyser. Parameters such as the apparent Michaelis– Menten constant Km and the Vmax were determined and the metabolic profile of S-warfarin was confirmed to be in line with literature findings. The fundamental knowledge acquired was then transferred to a commercial prototype sensor for testing the metabolic profile of known drugs. Encouraging results in line with the previous findings were achieved.

The metal chelating agent hexasodium N, N, N', N'-ethylenediaminetetramethylenephosphonate (EDITEMP. Na6) disturbed rat bone metabolism both in vivo and in vitro. In immature male rats (starting weight 65-85 g), for example, biochemical and histological changes indicative of reduced bone mineralization and resorption were observed immediately following the treatment with EDITEMP. Na6 (70 umol/kg/day, ip for 7 days). Withdrawal of the compound led to the recovery of bone resorption, as indicated by bone Acp activities, and a persistent increase in bone and serum Alp activities. Post-treatment mineralization of hypertropic cartilage and osteoid tissue laid down during and immediately following the treatment with EDITEMP. Na6 was observed and may have accounted for the increase in Alp activities. In vitro EDITEMP. Na6 reduced (mature) osteoblastic cell function. This was reflected by reductions in Alp activity, PTH-induced cAMP response, and in vitro mineralization (although the actual number of nodules formed was not reduced), and may well contribute to the in vivo inhibition of both bone mineralization and resorption. The presence of hydroxyapatite in vitro slightly modulated this effect of EDITEMP. Na6 on bone cell function. EDITEMP. Na6 did not appear to directly affect osteoclastic cell function in vitro. EDITEMPA is a carcinogen, and the significance of the disturbance of bone metabolism and possible mechanisms of carcinogenicity are discussed in the light of its lack of genotoxic potential.

Nek2 is a cell cycle-regulated protein kinase localised to the centrosome. Nek2 is involved in regulating mitotic events and centrosomal activity through phosphorylation of various substrates including C-Nap1, Rootletin and PP1. Nek2 is overexpressed in cancer cell lines and primary tumours. Overexpression of active Nek2 is characterised by the premature splitting of centrosomes and altered microtubule nucleation. Activation of Nek2 is dependent upon dimerisation and autophosphorylation. Using mass spectrometry, 13 sites of autophosphorylation have been identified in the Nek2 kinase. Individual phosphorylation sites were mutated and the effect that this had on Nek2 activity in vitro and in vivo was characterised. Phosphomimiking mutation of T170 or S171 created hyperactive Nek2A kinases. However, mutation of the C-terminal autophosphorylation sites had no effect upon Nek2A activity. The proposal that the C-terminal domain of Nek2 acts as an autoinhibitory domain was investigated although the removal of C-terminal residues did not create a hyperactive Nek2 kinase. An investigation into the role of a novel αT-helix found at the N-terminal end of the T-loop of an inactive Nek2 crystal structure, found that mutation of αT-helix residues resulted in loss of Nek2A kinase activity. Phospho specific antiserum directed against the Nek2 T175 autophosphorylation site was purified and characterised. This reagent should prove to be a valuable tool for detection of active Nek2 kinase in tumours as well as cells in culture. Furthermore, the intramolecular interactions of C-Nap1 and Nek2 were investigated to determine how centrosome cohesion is regulated. Protein binding assays revealed that C-Nap1 can form hetero- and homo-dimers and that the ability of C-Nap1 to heterodimerise is compromised as a consequence of phosphorylation by Nek2. Finally, the consequences of Cdk1 inhibition upon Nek2A activity and centrosome cohesion were investigated. Inhibition of Cdk1 in cells induced a substantial G_2/M arrest accompanied by premature centriole disengagement resulting in multipolar spindle formation observed in cells released from Cdk1 inhibition.

The spliceosomal E complex is the first spliceosomal complex that is committed to pre-mRNA splicing and therefore the components of this complex are early targets for regulation. With the majority of splicing regulation occurring at these early stages a greater understanding of these early complexes will bring a greater understanding of how splicing is regulated. Even at this early stage the functional groups at either end of the intron are in close proximity but the proteins involved in bringing the splice sites together are still poorly characterised. This thesis studied three distinct threads: the first thread investigated the interactions of one of the proposed bridges between the two splice sites, PRPF40A. Interaction studies in this thesis showed that the WW domains of PRPF40A were able to precipitate the U2 snRNP. The FF domains of PRPF40A were also able to interact with another component of the E complex, the SMN protein. As yet it is unclear whether the interaction between PRPF40A and SMN is a direct interaction. What was striking in this study was that no clear interaction with U1 snRNP could be seen with any of the fragments of the PRPF40A protein. This leads me to believe that, unlike yeast PRP40, PRPF40A is a U2 snRNP related protein, not a U1 snRNP protein. The second thread was to develop a new method to isolate U snRNPs using Flag-tagged U snRNP proteins; in this study the U1 snRNP was purified using Flag-tagged U1A and a U1-U2 di-snRNP was isolated using Flag-tagged U2 snRNP. This method of epitope-tagging core U snRNP proteins is useful as an alternative to antibodies for isolation of specific U snRNPs from nuclear extracts, as it gives flexibility in the target protein. The final thread was the development of a method to isolate and analyse the composition of ATP independent complexes formed on Exon and intron defined MINX constructs. The RNA composition was successfully analysed in an isolated MINX intron defined complex and an isolated MINX exon defined complex. Both exon- and intron-defined MINX constructs precipitated U1 and U2 snRNA. This study has focused on the interactions of PRPF40A; it has confirmed the interaction of PRPF40A with U2 snRNP (Makarov et al., 2012) and shown that this interaction is through the WW domains of the protein. The study has also begun to elucidate the interactions of the FF domains of the PRPF40A protein. Finally the study has developed two methods for isolating and studying the formation of the spliceosomal E complex.

RaLP is a recently identified signalling adaptor protein found to be up-regulated in invasive melanoma. To characterise RaLP and investigate its cellular role, a polyclonal antibody was raised against RaLP and purified successfully; it was considered to be a good tool for RaLP characterisation. Moreover, stable cell lines that express either GFPRaLP or FLAG-RaLP were generated. These cells showed a slightly higher increase in the percentage of G1 cells when compared to the parental counterparts using FACS analysis, and increased levels of p21/cip, a marker of G1 arrest. The underlying mechanism for the increased level of p21/cip remains unclear, but this may contribute to enhanced melanoma migration or survival induced by RaLP. The subcellular localisation of RaLP, as revealed by fluorescence microscopy and subcellular fractionation, was not restricted to the cytoplasm and cell membrane as has been reported previously, but a proportion of the protein was found to be present in the nucleus and mitochondria. Herein we show for the first time that RaLP shuttles between the nucleus and cytoplasm, and that the RaLP-CH2 domain has a critical role in RaLP export from the nucleus. Analysis of deletion and point mutants revealed that the sequence [superscript 83]LCTLIPRM[superscript 90] represents a functional nuclear export signal within the CH2 domain. Moreover, we have demonstrated that RaLP accumulates in the nucleus upon oxidative stress. Preliminary experiments showed that RaLP associates with the DNA and might play a role in gene transcription. Moreover, using mass spectrometry we were able to identify a phosphorylation site at Thr156/159 within the CH2 domain that is induced upon oxidative stress, while Ser132 was phosphorylated in unstressed conditions as well. These phosphorylation events may play a role in nuclear localisation or function of RaLP. Further support of a role for RaLP in the nucleus and mitochondria was provided by analysis of protein complexes containing GFP-RaLP by mass spectrometry. Proteins identified included a nuclear pore complex protein, importin-β transcriptional regulators, and mitochondrial proteins, in addition to the receptor tyrosine kinases Ret and TrkC. In a previous study, the invasion suppressor Nischarin was identified as an interacting partner with the CH2 domain of RaLP using a yeast two-hybrid library screen. Here we confirm that RaLP co-immunoprecipitates with Nischarin when overexpressed in cells or when present at endogenous levels. Mapping the RaLP-CH2 domain for the sequence required for Nischarin binding revealed that the first 93 amino acids are required for RaLP/Nischarin interaction, while interaction is greatly reduced upon deletion of the first 24 amino acids. Interestingly, Nischarin was also identified by mass spectrometry as part of a complex precipitated by GFP-RaLP from N1E-115 cells. Confocal microscopy provided additional evidence of co-localisation within melanoma cells. While we were unable to observe any effect of the RaLP/Nischarin association on the LIMK/cofilin pathway, very preliminary evidence suggested that RaLP might relieve suppression of Erk activation induced by Nischarin. Taken together, our results point to diverse roles for RaLP in cellular signalling and migration.

No description available.

The MonoCarboxylate Transporter (MCT) family of transmembrane proteins contain 14 members of which 6 have been functionally characterized. Of these characterised MCTs only MCTs 1-4 have been shown to transport lactate. These MCTs also facilitate the movement of pyruvate and ketone bodies across the plasma membrane (PM) in cotransport with a proton. For trafficking to and function at the PM MCTl, MCT3 and MCT4 require association with the monotopic ancillary glycoprotein basigin whereas MCT2 prefers association with embigin. This thesis has investigated the sensitivity of MCTl, MCT2 and MCT4 to the highly potent and selective MCTI inhibitor, ARC155858, discovered by AstraZeneca. Chimeras of MCTI and MCT4 were constructed and expressed in Xenopus laevis oocytes for transport studies to determine their inhibitor sensitivity. These identified a region between transmembrane domains (TMs) 7 and 10 of MCTI with which AR-C155858 binds from the cytoplasmic side. ARC155858 was shown to inhibit MCT2 but sensitivity was found to be dependent on the ancillary protein with which it is associated. Co-expression with embigin decreased the sensitivity of MCT2, but not MCTl, to AR-CI55858. The MCT C-terminus was shown to playa role in the interaction between MCT and ancillary protein which is secondary to interactions between the TM of the ancillary protein and TMs3 and 6 of the MCT. Additional studies were performed to characterise the substrate specificity of the orphan transporter, MCT6. Initial work suggested that products of pyruvate decarboxylation or polymerisation will provide lead compounds in the continuing search for the physiological substrate of MCT6, with formate another potential substrate. During this work it was also discovered that MCTI can catalyse the transport of specific dicarboxylates at low pH.

Berezovsky et al.'s closed loop folding hypothesis suggests that the locks of closed loops (i.e. the ends of the interacting loop) of length ~25-35 residues interact through hydrophobic interactions and that this is a vital event in protein folding. Here we investigate a possible link between the lock residues and drug resistance. The hypothesis is that drug resistance can be limited if the drug binds to the high connectivity regions such as the lock regions. To follow this perspective, the work has moved on to develop an additional method for determining closed loops by using the X-ray crystallographic B-factors as a measure of flexibility. From previous kinetics studies, by using the correlation of folding rate with contact metrics evaluated over the locks of closed loops and their neighbours, we observed that the conservation, particularly maximum proportion is possibly the best method for determining closed loops; we also found that the closed loop hypothesis requires modification to include the neighbours of the lock residues. Berezovsky et al.'s hypothesis has been evaluated using literature results from time-resolved dynamic non-radiative excitation energy transfer measurements for bovine pancreatic ribonuclease A protein (RNase A), as studied by Haas' group. An analysis of the experimental data and molecular dynamics (MD) data in the light of Berezovsky et al.'s hypothesis shows that the MD results are consistent with the experiment results. To investigate the lock residues to look for any additional properties that may help in drug design, we investigated the lock residues of HIV reverse transcriptase and protease by analysing the positions of lock residues vis a vis positions of drug resistant mutations in both enzymes, focussing on the closed loops in accordance to the Berezovsky et al.'s hypothesis. Our results indicate that if these lock residues were to be targeted by drugs, they may be less likely to generate resistance mutations.

No description available.

The eight-cysteine motif is a protein repeat sequence that has been identified in extracellular matrix proteins of the fibrillin superfamily which includes isoforms fibrillin-l and fibrillin-2 and the closely related latent transforming growth factor-B binding proteins (LTBPs). These multidomain glycoproteins contain a number of cysteine-rich sequence repeats which resemble precursor epidermal growth factor domains (EGF-like domains) interspersed with eight-cysteine motifs and so-called "hybrid" domains, which exhibit some features of both these repeats. The signature of the eight-cysteine motif is a conserved pattern of eight"cysteine residues, three of which are contiguous. The structure of fibrillin eight-cysteine motifs has not been determined but mutations within fibrillin-l eight-cysteine motifs cause the heritable connective tissue disorder Marfan syndrome, confirming their critical role in fibrillin function and microfibril integrity. This study investigated the structure of fibrillin eight-cysteine motifs. Computer predictions of eight-cysteine motif secondary structure suggest limited amounts of defined secondary structure, most of which is beta-sheet (B-sheet), and a high proportion of loop regions. Several of the cysteine residues were predicted to be involved in disulphide bond formation. Eight-cysteine motif sequences did not fit to any known protein fold structures. Attempts to express a recombinant DNA molecule encoding a polyhistidine tagged human fibrillin-2 hybrid motif in mammalian and yeast cells, using plasmid expression vectors pCR™3 and pCS69 respectively, failed to produce detectable levels of recombinant protein, although a functional transcript and full-length polypeptide were produced using cell-free transcription and translation systems. However, a recombinant protein consisting of human fibrillin-l sixth eight-cysteine motif, flanked on either side by single EGF-like domains, was subsequently expressed using vector pSPEK in COS-I mammalian cells. This recombinant protein was N-glycosylated and exhibited anomalous migration during SDS-P AGE analysis, appearing larger than its predicted molecular weight of 18 kDa. Mass spectrometry analysis of the purified recombinant protein revealed products of masses 18 863 and 20 783 Da.. thought to correspond to non-glycosylated and glycosylated forms of. the protein respectively. Circular dichroism analysis of the recombinant protein revealed a high amount of ~-sheet structure (58% +/- 0.59), a small but significant alpha-helix (a-helix) content (1.0 +/- 0.56) and high amounts of residual structure (41 +/- 1.0), indicating significant amounts of ~-turn or loop regions. This study has provided insights into the secondary structure of eight-cysteine motifs. Computer predictions of secondary structure support data obtained by structural analysis of a recombinant protein containing a fibrillin eight-cysteine motif It is proposed that the eight-cysteine motif assumes a novel protein fold which contains a limited amount of defined secondary structure, most of which is ~-sheet with a small but significant amount of a-helix, the predominant feature of this motif being a high proportion of loop regions, which are held in place by disulphide bonding between pairs of the eight cysteine residues.