The coiled-coil is a very common protein structural motif, consisting of two or more alpha helices intertwined in a supercoil. In biological systems it is found in proteins which fulfil a number of roles, for example in structural proteins such as keratin, as well as a large number of transcription factors and other DNA binding proteins, where it functions as a dimerisation domain. This versatile motif has also been adapted for a number of applications including the production of responsive hydrogels; the construction of self-assembling fibres for tissue engineering applications; as a cross-linking agent in drug delivery applications, as well as in the creation of biosensor surfaces. In this study two different types of protein containing coiled-coil domains are examined. It is the aim of this work to increase the understanding of how behaviour of this motif at the molecular level relates to those at the macroscopic. The first set of proteins of interest in this thesis form pH-responsive hydrogel systems utilising a variant of the coiled-coil motif, the leucine zipper, as their cross-linking domain. These consist of two or more leucine zippers separated by random coil spacer sections. When the leucine zipper sequences dimerise, they cross-link the proteins forming a gel matrix. As charges within the leucine zippers alter, as the pH of their environment is altered, their relative stability changes causing them to associate or dissociate. This is reflected in a change in the physical characteristics of gels from visco-elastic solids to viscous liquids which is reversible. The bulk properties of these materials is well characterised, but relatively little information is known at the molecular level. Starting at the single molecule level and scaling up to the mesoscale, the behaviour of hydrogel-forming proteins has been probed using a range of techniques, particularly with respect to the interactions of the coiled-coil forming domains. Studies described in this thesis have concentrated on the change of physical characteristics of these proteins and their assemblies as pH was altered. We have thus been able to connect behaviour of these proteins at the molecular scale to bulk properties of solutions and materials formed from them. In the final experimental chapter a coiled-coil structure of more complex architecture is examined. This is formed of a ternary complex of three proteins which assemble into a two-stranded leucine zipper domain, designed as a cross-linking motif for nano-particle assemble. Force was used to probe the kinetic stability and mechanical strength of this assembly. With the increasing use of the coiled-coil motif as a cross-linking domain in the use of biomaterials it is increasingly important to be able to connect the behaviour of such materials to the behaviour of the coiled-coil motif at the molecular scale, to gain further insight into the mechanistics behind their behaviour.

Previously, the sitABC operon of S. epidermidis has been shown to be regulated in response to the extracellular concentration of both iron and manganese, though the identity of the true substrate for this putative ABC-type import system remains to be determined. Through the course of these investigations, the metal binding characteristics of the substrate binding component, SitC, were investigated further. Recombinant protein was over-expressed and purified as an oligomeric form. The metal binding properties of purified SitC were studied using circular dichroism (CD) spectroscopy, which demonstrated a primary specificity for Mn2+ ions with additional lower affinities for Fe2+ and Zn2+ ions. Hill analysis showed metal ion binding proceeded with dissociation constants within the micromolar range and with the strongest degree of cooperativity with Mn2+. As only a single metal ion binding site is expected to reside within SitC this cooperativity was proposed to occur intermolecularly. Inductively coupled plasma atomic emission spectroscopy (ICP-AES) revealed dissociation between metal ions and SitC with Mn2+ almost completely dissociating, and which may be due to conformational changes required to relay the substrate(s) through the transport system. A molar ratio of bound metal ion to protein approaching 1:1, suggests SitC possesses only a single metal ion binding site. These data suggest SitABC may function primarily in manganese uptake. MntC from S. aureus is regarded as an orthologue of SitC, and which has previously been shown by mutational analysis to function solely in manganese accumulation. An amino-terminal truncate of MntC (His(6)-MntC-N) was over-expressed also in an oligomeric form and the metal binding properties compared to SitC using CD analysis. Using this method, His(6)-MntC-N demonstrated cooperative primary specificity for Mn2+, with secondary affinity for Zn2+ and Cu2+ but no evidence of Fe2+ binding was apparent. These data support the previous hypothesis that SitC and MntC are orthologous in their primary metal ion specificity, though attempts to provide confirmatory functional evidence through mutational analysis of sitABC were unsuccessful.

Nogalamycin is a member of the anthracycline family of antitumour antibiotics. These are potent cytotoxic agents and are routinely used in cancer chemotherapy. Though nogalamycin is clinically insignificant, it does exhibit three distinct types of non-covalent binding to DNA. Since most other anthracyclines bind to DNA by only one or two of these mechanisms, nogalamycin is an excellent model with which to probe the interaction of this class of anti-tumour agents with DNA. Here, we investigate the binding orientation and stoichiometry of nogalamycin in adjacent TpG(CpA) (and CpG(CpG)) intercalation sites using a combination of NMR techniques and NOE-restrained molecular dynamics simulations. These methods are also employed to investigate the structure of GNA hairpin loops, which are considered to have important biological functions, and assess how their structure and stability are influenced by the introduction of nogalamycin at an adjacent site. The effect of nogalamycin on extrahelical thymine bases incorporated onto either face of the intercalation sites is also investigated in this context. Binding of quadruplex-specific antibodies to telomeric DNA in Stylonychia lemnae macronuclei has recently been detected using immunofluorescence, providing direct evidence for the formation of quadruplex DNA structures in vivo. Guanine-rich quadruplex structures have been extensively studied by NMR and x-ray crystallographic methods. Previous structural studies have failed to unambiguously resolve the conformation preferred by less-stable A-tetrads incorporated into DNA quadruplexes. Additionally, little effort has been made to address the exact number of ions bound to these adenine-containing structures. This forms the basis of our study into quadruplex DNA. Finally, we endeavour to investigate the extent of hydration of both duplex and quadruplex structures using rMD methods, and to comapre hydration patterns in the liquid- and solid-state.

This study considers the molecular weight, conformation, stability and nanoparticle formation of the polysaccharide chitosan in relation to its pharmaceutical applications in the drug delivery. Chitosan has bioadhesive and absorption enhancing properties that augment its use in the delivery of therapeutic proteins and peptides across mucosal membranes. The molecular weights of a range of chitosans, with degree of acetylation (DA) ranging form 7 to 30%, were studied using the hydrodynamic techniques of SEC-MALLS (size exclusion chromatography coupled with multi-angle laser light scattering) and analytical ultracentrifugation. It was found that there was reasonable agreement between the two techniques with molecular weights obtained ranging between 42000 and 200000 kDa. These results indicated that the use of SEC-MALLS as a measure of degradation during storage studies would be valid and this technique would be more convenient due to the much shorter time required for each sample. Conformational studies indicated that in 0.2M acetate buffer, pH 4.3, the chitosan samples studied appeared to have an extended conformation which is in agreement with studies performed by other researchers (e.g. Anthonsen, M. W., K. M. Vårum and O. Smidsrød (1993) Carbohydrate Polymers 11: 193-201 and Dyer, A. M., M. Hinchcliffe, P. Watts, J. Castile, I. Jabbal-Gill, R. Nankervis, A. Smith and L. Illum (2002) Pharmaceutical Research 19(7): 998-1008), non-withstanding complications in interpretation through possible solvent draining effects (Berth, G., Colfen, H., Dautzenberg H (2002) Prog. Coil. Int. Sci. 119, 50-57). Degradation studies were performed using viscosity measurements and SECMALLS to measure molecular weight. The effect of solution parameters such as pH and ionic strength were examined as well as structural parameters, e.g. molecular weight and DA. Degradation was found to increase as the pH of the solution decreased indicating that acid hydrolysis was occurring. The range of ionic strength studied (0.1 - 0.3M) did not have any significant effect on degradation rate. For the samples studied, molecular weight appeared to have little effect on degradation, however, the more deacetylated the polysaccharide, the slower the rate of degradation due to increased charged residues along the chain. Finally chitosan nanoparticles were then prepared using ionotropic gelation with tripolyphosphate pentasodium. This technique produced nanoparticles in the size range 320 - 380 run with an insulin loading capacity of 30-50%. Insulin-loaded nanoparticles were prepared and resuspended in acetate buffer at various pHs. The molecular weight and insulin loading was tested at 2-week intervals and results indicated that chitosan degradation was reduced by 30% compared to a control chitosan solution, and the total amount of insulin incorporated remained constant over the storage period.

The mouse cytochrome P450 4a family of genes is poorly characterised. This thesis describes a detailed study of these genes. Using primers derived from exon 1 of the mouse cyp3a10 cDNA sequence, a 1.4kb genomic fragment was cloned which contained the promoter region of the gene. Additionally, a lambda genomic library was screened with probes derived from the partial cDNA clones of the mouse cyp4a10 and cyp4a12. Originally, two classes of phage lambda clones were isolated. One clone, lambda 16, was partially sequenced and was found to contain the 3' end of the cyp4a12 gene. The second clone, lambda 4, was subcloned and found to be distinct from previously known murine cyp4a genes. This novel gene has been designated cyp4a14. To obtain the full cyp4a14 gene, the genomic library was rescreened. Lambda clone 12 was subsequently isolated from the library, which was found to contain exons 1 to 5 of Cyp4a14. In order to study the promoter region of cyp4a14, a pcr-based approach was undertaken to clone 4.4 kb upstream of exon 1 of the gene and a 1.2kb fragment has been sequenced. Cyp4a14 RNA was also found to be highly induced by a peroxisome proliferator, methylclofenapate. Primer extension analyses were performed to determine the start of transcription of Cyp4a14, which was mapped to a single T nucleotide 26bp upstream of the putative start site of protein translation. The promoter region of Cyp4a14 is highly similar to the corresponding regions of the rat CYP4A2 genes. however, similarity is dramatically reduced about 350bp upstream of the start of transcription, which coincides with the presence of two 358bp repeats in teh CYP4A2 gene. The promoter also contains a highly conserved 19bp element which is also found in the promoter regions of the CYP4A1 and CYP4A2 genes. Cyp4a14 is a novel member of the murine Cyp4a subfamily. It contains 12 exons and is highly similar to the rat CYP4A2/3 genes. However, there is high conservation in exon 3 between Cyp4a14 and cyp4a3, which makes it marginally more related to this gene. In addition, cyp4a14 shows very high similarity in exons 4,8, 11 and 12 with other known CYP4A genes. Correspondingly, the peptide sequences encoded by these exons are highly conserved. Exon 8 encodes a 16 amino acid motif, LRAEVDTFMGEGHDTT, which is highly conserved in the CYP4 family. Exons 22 and 12 encode the well known RNCIG motif, containing the haem-binding domain, which is the proposed signature for a P450 protein. Exon 4 in the CYP4A genes encodes for a highly conserved, but previously unreported motif, HRRMLLTPGFHYDIL. Primary sequence alignments indicate that these motifs map very close to or within predicted substrate recognition sites and thus could have an important role in the enzyme activities of the CYP4A enzyme. The identification of Cyp4a14 also enables the analysis of the evolution of the murine Cyp4a subfamily. The mouse has 4 genes in the Cyp4a subfamily; however, the closely related rat has four CYP4A members. As the rat CYP4A2, which is very similar to CYP4AA3, was not to have a homologue in the mouse, it must have arisen after the mouse lineage had diverged from the rat in evolutionary history. As all three mouse Cyp4a genes are significantly more similar to their equivalents in the rat than to other murine Cyp4a genes, the gene duplication events giving rise to these three genes must have occurred before the mouse had diverged from the rat. Phylogenetic analyses of the CYP4A, CYP4B and CYp4F subfamilies demonstrate that CYP4A genes are more similar to the CYP4B family than to the CYP4F subfamily. This suggests that the gene duplication event giving rise to the CYP4A and CYP4B genes must be a more recent event than that giving rise to the CYP4F subfamily. In addition, the CYP4A subfamily of genes is more divergent than the CYP4B subfamily across species.

The role of deoxyribonucleic acids in the cell has made DNA an attractive target for drug molecules. The anthracycline antitumour antibiotics are potent cytotoxic agents that have found widespread use in cancer chemotherapy. Nogalamycin binds DNA through intercalation, preferentially to 5'-TpG and 5'-CpG sites, by threading through the DNA helix and interacting with both the major and minor grooves simultaneously. In this thesis, the interaction of nogalamycin with the 5'-TpG site has been investigated using synthetic oligonucleotide duplexes and a combination of high-resolution NMR techniques and NOE-restrained molecular dynamics simulations. The solution structure of the 1: 1 complex with d(ATGCAT)2 is described with NOE data unambiguously identifying the position and orientation of the bound drug molecule, allowing conclusions to be drawn regarding the specificity for the TpG site. Binding at one TpG site sterically blocks the interaction at the symmetrically equivalent CpA site. The structural studies are extended to investigate by NMR the role of solvation in drug- DNA recognition and binding. Based on the sign and magnitude of solute-solvent NOEs, it is shown that only a small subset of water molecules visible in the crystal and MD structures are found to be bound in the solution complex, and that a number of these are involved in mediating drug-DNA interactions. The role of the dynamic network of water molecules in stabilising the complex in solution is discussed. Finally, the binding of nogalamycin at a TpG site carrying a DNA strand break has been investigated using a novel designed single-stranded intermolecular duplex consisting of two hairpins stabilised by GAA loops [d(ACGAAGTGCGAAGC)]. Although stacking of the two hairpins is weak, nogalamycin is shown to bind and stabilise a 1: 1 complex by binding at the intercalation site. The complex is discussed in terms of the mechanism by which nogalamycin is able to bind to premelted duplex DNA.

Haem proteins are functionally and structurally extremely diverse biomolecules and play a vital role in aerobic life. They perform a vast range of functions like transport of oxygen and electron transfer, gene regulation, redox-sensing and drug metabolism. This many-sidedness of haem-proteins is due to the extremely versatile chemical properties of the iron in the haem prosthetic group. Iron, or iron in the form of haem, plays a key role in many biological processes and it is an essential nutrient for the majority of living organisms. Despite being one of the most abundant chemical elements, iron is scarcely available under physiological conditions, because of its insolubility and toxicity. Pathogenic bacteria rely on their host as a source of haem and/or iron and a strong link between iron / haem acquisition, virulence factors and pathogenicity exists. Therefore they have evolved a set of specialised haem receptors and carriers to circumvent their iron dependency, often involving the 'stealing' of haem as a source of iron from host's haem-proteins, which in the host is the most abundant and relatively available source of iron. These proteins are both vital and unique to bacteria and so have been considered as possible drug targets. At the beginning of this thesis work, the fascinating cell and molecular biology mechanisms of these novel haem binding proteins were still largely unexplored. In this thesis the cloning, expression, purification of four novel bacterial haem transport proteins for biochemical and biophysical characterisation and structural studies is described: HemS and HemT from Yersinia enterocolitica and Shp and HtsA from Streptococcus pyogenes. HtsA in complex with haem was crystallised and a preliminary X-ray diffraction analysis was carried out. Furthermore HemS was crystallised in its apo- and haem bound form and both structures were determined. Comparison of the apo- and haem-bound crystal structures provide penetrating insights into its mechanism of haem binding and release.

Many structurally unrelated chemicals can induce members of the cytochrome P450 superfamily with phenobarbital (PB) being a typical example. PB induces CYP2B1/2, which are most highly expressed in the liver. Their mechanism of activation has not yet been elucidated, with advances hampered by the absence of a suitable cell culture system to mimic the in vivo PB-mediated induction. During this thesis a primary rat hepatocyte culture system has been developed which is highly responsive to PB at both RNA and protein levels. A sensitive and specific RNAse protection assay (RPA) has been used to demonstrate that CYP2B2 mRNA is highly inducible in vitro by PB. This response occurs in a time and dose-dependent manner. The use of RPA and Western blotting has demonstrated that this primary rat hepatocyte culture system supports the induction of CYP2B2 mRNA and protein levels by PB. Sequencing -1.4kb of the 5' flanking region of the CYP2B2 gene identified genomic regulatory elements and highlighted the location of the phenobarbital response element (PBRE). The PBRE was sub-cloned into various reporter constructs and transfection technology was used to determine its PB-mediated induction. A comparative study of the constructs generated in this thesis to that of a construct provided by Anderson's group (Trottier et al., 1995) was undertaken and no differences were found in their PB-responsiveness. The Anderson construct containing the PBRE was shown here to confer a 3.3-fold PB-mediated induction of the CYP2B2 gene by CAT reporter assays. This induction was shown to be both dose and time dependent. The induction is lower than that obtained by other workers due primarily to assay conditions which were not yet optimal. However, the effects of androstane on the constitutively active receptor (CAR) may also play a role in the small inductive response of the phenobarbital response element to PB.

Proteolysis is a component of protein turnover, controlled by multiple proteolytic systems. Alterations in system components within skeletal muscle has been associated with hypertrophy, remodelling, atrophy, apoptosis and metabolic dysregulation. Key components may have novel regulatory roles, e. g. calpain-3 and cathepsin-L. Experiments described within this thesis investigated the hypothesis that the gene expression of specific proteolytic system components within skeletal muscle may be co-ordinately regulated and altered during nutritional and pharmacological states known to modify protein turnover and induce muscle growth. Gene expression for multiple components of the calpain system was analysed in calf LD (Longissmus dorsi) by Quantitative Real-Time PCR in a plane of nutrition trial. There were three groups: low (LOW), high (HIGH) plane of nutrition and LOW to HIGH (REFED). Half of each group were slaughtered 48 hrs after refeeding, whilst the remainder were slaughtered 13 days later. Total RNA yield/g LD increased (P < 0.05) across all groups between slaughter dates. Calpain-3 expression increased in LOW and REFED and calpastatin in all groups between slaughter dates, with a trend towards significance (P = 0.073, P=0.085, respectively). In the 1St slaughter, calpain-3 expression had a trend to be lower in the LOW group and values for REFED were similar to HIGH value level. cDNA probes for unique and novel proteolytic system components were generated by RT-PCR and used to investigate the effects of acute and chronic Q-adrenergic stimulation, on the gene expression of those specific components in pig LD, by northern blotting. The ß2-adrenergic agonist clenbuterol (5 ppm) decreased glycogen levels (mg/g LD) (P < 0.001), increased cathepsin-L expression (P < 0.001) and increased E2G 1 values numerically within 24 hrs of treatment. Cathepsin-L was unchanged by adrenaline administration. Calpain-3 was unchanged with either clenbuterol or adrenaline treatment. The significance and implications of the data are discussed.

Antisera raised against recombinant human CYP4Z1 (4Z1), mouse Aryl hydrocarbon receptor (AhR), and C.elegans Latrophilin (LpH) proteins were titred over the course of several bleeds. The sensitivity of the antibodies shows an increase over the course of several immunizations, with the minimum amount detected being 1 ng of 4Z1, 0.3 ng of AhR, and 0.3 ng of LpH. Terminal bleeds were taken for the AhR and LpH antisera. The AhR antisera detects proteins in rat and mouse liver cytosol consistent with previous reports of the AhR. LpH is predicted to be localized in a membrane compartment on the basis of its primary structure, so sub-cellular fractions of C.elegans were isolated and tested, revealing a protein of ~113 kDa in a crude membrane preparation. This protein was not solubilized in < 1% Emulgen 913, but was soluble in 2% dodecylmaltoside. A fresh 15k supernatant treated with 2% dodecylmaltoside showed a strong band of LpH protein at ~113 kDa. However, 66 kDa band was detected when the samples were stored overnight at 4 degrees centigrade due to presence of a G protein coupled receptor proteolysis site (GPS) between position M536 and C547 of LpH, which is a characteristic feature in the LpH protein family. In order to study the expression of the CYP 4X1 in mouse tissues, RNase protection assays were performed. Different tissues were assayed at the same time and the same riboprobe was used to hybridise all the samples. The CYP 4\x1 probe was shown to be full length (-ve control). However, the +ve control (RNase positive) shows an absence of signal when hybridised to the yeast tRNA demonstrating the specificity of the signal in the samples. Several RNA samples were hybridised with mouse cyp4X1 gene probe, such as aorta, brain and heart and liver. The mouse cyp4X1 gene appears to have 12 exon from the genomic sequence and encodes a protein which high identity with the human and rat cyp4X1 gene. The full-length of the probe was 424 b.p and the protected fragments were 177 b.p. The murine cyp4X1 was not expressed in control liver, but is expressed in brain at high level. Cyp4X1 gene was also investigated in aorta tissue and found to be expressed at low levels. Known inducers of hepatic cytochrome P450 were used (Ciprofibrate, TCDD, PB, and dexamethasone), but had no induction effect in the samples. Western blotting of brain confirmed that the cyp4X1 protein is expressed in brain, and quantification showed that this is a major cytochrome P450 in brain.