DEF6 is an atypical guanine nucleotide exchange factor (GEF) for Rho GTPases Rac1 and Cdc42, which is highly expressed in mature T cells. DEF6 is recruited to the immunological synapse (IS) following phosphorylation by LCK in response to T cell receptor (TCR) stimulation during inflammation. In resting Jurkat T cells, DEF6 has been known to associate with polysomes and under cellular stress conditions to form cytoplasmic granules colocalising with mRNA decapping enzyme subunit 1 (DCP1), a marker of P-bodies. Hence DEF6 has been linked to mRNA regulation, which includes mRNA translation initiation and repression by P-body formation. To elucidate the structural requirements for P-body localisation of DEF6, N-terminal truncation mutants lacking either the C-terminal coiled coil domain or the pleckstrin homology (PH) domain in conjunction with the coiled coil domain were established and coexpressed with DCP1 in COS7. Both GFP-tagged mutant DEF6 proteins spontaneously colocalised with DCP1 indicating that the N-terminal end of DEF6 is sufficient for P- body association and that the coiled coil domain facilitates a confirmation that masks the N-terminal end of DEF6. Exchange of serine and/or threonine residues in the C- terminal end with either phosphomimic or phosphosilent amino acids resulted in formation of GFP-DEF6 aggregates or granules, respectively. While GFP-DEF6 granules partly overlapped with DCP1, GFP-DEF6 aggregates appeared to trap DCP1. In resting Jurkat cells endogenous DEF6 associated with nascent mRNA translation and colocalised with eukaryotic translation initiation factor 4E (eIF4E) as well as poly A binding protein (PABP). eIF4E and the eIF4E-binding protein 4E-T but not PABP were shown to interact with DEF6 in vitro. Moreover, siRNA-mediated knockdown and ectopic overexpression of DEF6 in Jurkat cells established a positive correlation of DEF6 expression and eIF4E, PABP and 4E-T protein expression. RT-PCR revealed that the alteration in expression of these proteins was not due to transcriptional regulation and inhibition of protein degradation pathways could not rescue downregulation of eIF4E. Colocalisation of DEF6 with eIF4E and PABP but not with 4E-T was also observed in the IS upon TCR-mediated signalling. Together these results strongly suggest that DEF6 is involved in active mRNA translation in resting and activated Jurkat T cells controlling the expression of components of the translational initiation complex as well as P-bodies.

Essential or primary hypertension is a complex polygenic disease with genetic heritability averaging approximately 30% and with strong influence of environmental factors and gene-environment interaction. Heterogeneity in the general population and the polygenic complexities of the disease has meant that identification and functional validation of candidate genes has proved extremely difficult in humans. Several strategies have been developed to dissect genetic determinants of hypertension, one of which is the use of rodent models (1;2). Animal models of heritable hypertension offer more favourable investigative opportunities because of reduced genetic heterogeneity, the capacity for controlled breeding and environmental conditions, and the ability to produce genetic crosses and analyse large numbers of progeny. The stroke-prone spontaneously hypertensive rat (SHRSP) is a commonly used model of human essential hypertension. Previous studies conducted in our laboratory utilizing a combination of congenic strain construction and genome-wide microarray expression profiling in the SHRSP have allowed us to identify the positional candidate gene, glutathione S-transferase μ-type 1 (Gstm1), which is involved in the defence against oxidative stress and is significantly down-regulated in the SHRSP (3;4). Genomic DNA sequencing of Gstm1 in SHRSP and WKY identified 13 single nucleotide polymorphisms (SNPs), an insertion and a deletion (5). Luciferase reporter gene assays implicated five SNPs to be responsible for significant reduction in luciferase activity measurements (6). In consideration of these previous studies, it is hypothesized that Gstm1 deficiency in the SHRSP plays a causative role in the development of oxidative stress and hypertension. To establish definitive proof that reduced Gstm1 expression affects blood pressure regulation and oxidative stress, two independent transgenic lines (referred to as Trans1 and Trans2) of SHRSP were created with the aim of rescuing Gstm1 deficiency by incorporation of a normal Gstm1 gene into the SHRSP genome. Generation of these transgenic SHRSP rats involved microinjection with a 2.7 kb linear construct encoding wild type (WKY) Gstm1 under the control of the universal EF-1α promoter. They were generated using the same expression platform and microinjection fragment purification protocol employed in the successful production of the CD-36 transgenic, rat as previously described (7). The transgenic protocol was carried out in collaboration with Dr 17 Michal Pravenec (Prague), who is an expert in transgenic rat production, using male and female SHRSP rats from the University of Glasgow colony. Oxidative stress is an important pathogenic factor in the development of cardiovascular disease. Glutathione S-transferases protect against oxidative stress-induced injury through the detoxification of reactive oxygen species. It is hypothesised that Gstm1 deficiency in the SHRSP plays a causative role in the development of oxidative stress and hypertension. Thus the aims of this study were to establish definitive proof that reduced Gstm1 expression in the SHRSP plays a causative role in the development of hypertension and oxidative stress through utilizing a combinational approach of in vivo and ex vivo studies alongside molecular analysis to fully characterize the Gstm1 transgenic SHRSP rat. Additionally, information and insights gained from this investigation from the Gstm1 transgenic SHRSP will be applied to a translation aspect for the investigation of GSTM family in humans. Functional validation through hemodynamic and cardiac analysis included measurement of systolic, diastolic and mean arterial blood pressures, pulse pressure and heart rate using the Dataquest IV telemetry system (Data Sciences International) and transthoracic echocardiography was used to assess cardiac geometry and contractility. Telemetry data show that there is a significant reduction in systolic blood pressure, diastolic blood pressures, and pulse pressure in both of the transgenic lines when compared to the SHRSP suggesting that incorporation of a WKY type Gstm1 gene into the SHRSP genome does indeed reduce the hypertensive phenotype. Moreover, the observed reduction in systolic blood pressure is remarkably similar in magnitude to that demonstrated in the Chromosome 2 congenic strain, SP.WKYGla2c*, in which Gstm1 was identified as a candidate gene for hypertension. In order to investigate the potential role of Gstm1 deficiency in the salt-sensitivity phenotype in SHRSP rats, parental strain rats and Trans1 animals underwent 1% salt loading starting at 18 weeks of age. This resulted in Trans1 displaying a trend towards salt-sensitivity (i.e. exaggerated night-time daytime blood pressure variation) similar to that of the SHRSP, however, the Trans1 line still maintained a significant decrease in systolic and diastolic blood pressure compared to the SHRSP during salt loading. 18 In parallel with the significantly lower SBP, DBP and PP we also observe significantly improved cardiac function and reduced cardiac hypertrophy in the two independently generated transgenic lines. While there was no significant changes in both fractional shortening (FS) and ejection fraction (EF), between the four strains, relative wall thickness was significantly reduced in WKY, Trans1, and Trans2 rats when compared to the SHRSP with Trans1 and Trans2 rats showing an intermediate phenotype between the parental strains. Analysis of genetic and molecular changes resulting from the random insertion of Gstm1 into the SHRSP genome included assessment of transgene (WKY form) and total Gstm1 gene expression, protein quantification, immunohistochemistry (IHC), transgene insertion and copy number. Both transgenic lines demonstrated an increase in total and transgene specific expression of Gstm1 in kidneys at 5 weeks of age as well as increased transgene expression in several other cardiovascular tissues. Protein expression was also similarly increased in the kidney at 5 weeks of age and showed a similar expression pattern to that of the WKY. Additionally, we saw increased total Gstm1 expression in a range of cardiovascular tissues at 21 weeks of age without changes of other Gstm family members (Gstm2 and Gstm3). Although it was not possible to identify the exact location of the transgene insertion site in both transgenic lines, data presented indicate that they are not identically inserted. Furthermore, sequencing data shows that each transgenic line contains multiple copies of the transgene across a number of generations. To assess renal function in the Gstm1 transgenic lines, rats from each line that were implanted with telemetry probes were assessed by 24-hr metabolic cage measurements which allowed for analysis of indirect glomerular filtration rate along with proteinuria and urinary electrolyte measurements. Histological analysis was used to assess renal morphology by examining haematoxylin and eosin (H&E) stained sections. Fibrosis was examined by staining with picrosirius red. At 21 weeks, we saw evidence of reduced renal pathology as indicated by the absence of renal vessel hyperplasia and reduced proteinuria in the WKY, Trans1, and Trans2 rats. H&E staining showed a more similar morphology to the WKY in the transgenic lines with no signs of accelerated hypertension. These improvements in renal pathology were also apparent in salt-loaded Trans1 rats. 19 Oxidative stress and myography measurements were also carried out in order to ascertain the impact of increased Gstm1 expression on the SHRSP genetic background. The data presented in this study clearly shows a reduction in renal oxidative stress in both transgenic lines. Furthermore, these improvements in oxidative stress were also apparent in salt-loaded Trans1 rats.

This research project attempts to identify the genetic determinants of SHRSP/Gla phenotypes by using mRNA and micro(mi)RNA expression profiling data, in combination with the genome sequence of the SHRSP/Gla and WKY/Gla, to facilitate human translational studies for hypertension and vice versa.

Despite the diverse structures and functions sampled by the proteome, all proteins comprise 20 canonical amino acids that sample only a small percentage of available chemistry. This limitation is lifted somewhat through the use of post-translational modifications, however the limit imposed by the restricted number of amino acids inherently limits the variety of protein function and control that can be accessed. One powerful route to diversify the chemistry sampled by proteins is through genetically encoded unnatural amino acid (uAA) incorporation. The uAA p-azido-L-phenylalanine (AzPhe) can introduce two novel methods of control, photochemical covalent rearrangement and Click chemistry. AzPhe incorporation combined with these two methods of novel post-translational control were used to modulate the function of two distinct proteins; TEM β-lactamase and sfGFP. This thesis introduces the use of uAAs and the technical modifications required to enable uAA incorporation in vivo. It describes the in silico approach taken to evaluate potential mutations based on the likelihood of them imparting novel changes to protein function. Nine positions in TEM β-lactamase were chosen for uAA incorporation and the effect on activity was then determined using kinetic analyses. AzPhe incorporation alone resulted in a variety of effects on enzyme activity, ranging from small increases to complete loss of activity. Subsequent post-translational modification using UV light resulted in only slight changes in activity. Modification via Click chemistry using dibenzyl cyclo-octyne (DBCO) derivatives resulted in either inhibition or increased catalytic activity, depending on the position of AzPhe incorporation and the type of adduct used. Click chemistry was then used to modify TEM β-lactamase with other chemical modifications that enable the immobilization of proteins onto two different surfaces. The π-π stacking interaction between a DBCO-pyrene moiety and graphene was exploited to attach TEM β-lactamase to graphene in a defined and controlled manner, placing the active site in close proximity to the electron cloud of the sp2-bonded material. TEM β-lactamase was then modified using two DNA oligonucleotides that define assembly of a DNA origami “tile”. DNA origami can be used to immobilize multiple proteins at several defined positions, enabling the re-creation of enzyme pathways or signalling cascades in vitro. Finally, AzPhe was incorporated into sfGFP and the effects of its incorporation and subsequent modification on fluorescence were explored. The incorporation of AzPhe resulted in a blue shifted λmax, a change that was reversed upon UV irradiation. X-ray crystallography suggested that a hydrogen-bonding network involving the chromophore and surrounding residues was disrupted upon AzPhe incorporation, but then reformed upon modification of the uAA. Click chemistry had a variable effect on fluorescence depending on the modification used. Modification of AzPhe with a large fluorescent dye had no effect on the sfGFP fluorescence spectrum, but enabled FRET between the two chromophores. Modification with a DBCO-amine had the same effect as UV irradiation. Overall, this thesis has shown that the use of genetically encoded uAA incorporation coupled with novel post-translational modifications is a powerful approach for modifying protein function, and facilitating defined interfacing with new and useful materials.

This study examines the potential for increasing natural L-DOPA drug biosynthesis in Mucuna pruriens by silencing or “knocking down” expression of putative DOPA/tyrosine decarboxylase (Mp-ty/ddc) in situ. Mp-ty/ddc codes for DOPA/tyrosine decarboxylase (Mp-TY/DDC) which converts L-DOPA to dopamine in plants. The hypothesis of the work was that silencing the Mp-ty/ddc gene would result in accumulation of L-DOPA in the plant tissues. This work involved isolation and characterisation of 1.73 kb putative full-length ORF of Mp-ty/ddc. The gene showed 74% homology with TY/DDC protein alignments of other plants in the same taxa, although no enzyme activity was detected when the gene product was heterologously expressed. In addition, a protocol was developed for Agrobacterium mediated transformation of M. pruriens so as to be able to manipulate expression of the DOPA genes in situ. The cotyledonary nodal and hypocotyl tip explants regenerated shoots on M.S media supplemented with 50 μM BA, 0.5 μM NAA and 50 mg l-1 kanamycin selection also the nptII transgene was detected by PCR. The Agrobacteria strains GV3101 harbouring a pGREEN vector and carrying an Mp-ty/ddc antisense were used for the plant transformation experiments. Further work showed that the Mp-ty/ddc gene copy number was 1, the gene expression was highest in roots and stems, followed by seeds and was very low in leaves. On the other hand, L-DOPA-content in seeds was 17-fold higher relative to leaves and 15 fold relative to stems and roots.

SWI/SNF chromatin remodelling complexes are one of the well-characterized cellular machineries capable of regulation of gene expression. Numerous lines of evidence indicate that SWI/SNF complexes are involved in a wide range of cellular processes and the maintenance of homeostasis whereas aberrant expression of those proteins contributes towards cancer development. Colorectal cancer remains one of the most clinically significant cancers due to its high incidence in developed countries and previous studies have demonstrated that SWI/SNF complexes are aberrantly regulated in a significant proportion of patients with this disease. However, whilst the sequence of molecular events leading to CRC has been well-established, the role of SWI/SNF chromatin remodelling complex ATPase subunits Brm and its paralogue Brg1 in the colorectal tumorigenesis remains elusive. The chromatin remodelling catalytic subunit Brm has been found to interact with the Notch pathway effectors ICD-22 and CBF-1 and also to be necessary for expression of the Wnt target gene CD44 and for Rb-mediated cell cycle arrest. In this PhD thesis, the potential of Brm to modulate Wnt-driven intestinal tumorigenesis was addressed. Initially, a murine model carrying constitutively deleted Brm was used to assess the consequences of this loss on homeostasis of the small intestinal and colonic epithelia. The effects of Brm deficiency were also examined in the context of Wnt-activated epithelium via conditional loss of Apc. Additionally, the effect of concomitant loss of Brm and Brg1 was addressed in the contexts of both normal homeostasis and aberrant Wnt signalling. The results presented here demonstrate that Brm plays an important role in the small intestine by regulating the distribution of proliferating cells and cell fate decisions mediated through Notch pathway effectors. Furthermore, Brm deficiency was found to modulate intestinal phenotype of Wnt activation through the attenuation of the Wnt transcriptional programme and the suppressed expression of the intestinal stem cell marker Olfm4. Thus while Brg1 has been widely characterized as a bone fide tumour suppressor, the function of Brm continues to remain elusive especially in the light of contrasting effects co-mediated by Brm on proliferation, differentiation and gene expression. Taken together, these results elucidate the tissue-specific role of Brm, the catalytic subunit of SWI/SNF chromatin remodelling complex, on both normal intestinal homeostasis and acute activation of Wnt pathway while the extent of these Brm-dependent effects depend upon the gradient of Wnt signalling throughout the epithelium of small and large intestine.

Huntington’s disease (HD) is a neurodegenerative disease caused by a mutation in the huntingtin gene (HTT). The extended CAG repeat ultimately leads to loss of medium spiny neurons (MSNs) in the striatum of the HD brain. Cell replacement therapy using primary human fetal tissue as a source of “genuine” MSNs has shown ‘proof of principle’ as a strategy to treat this genetically inherited disease1. However, renewable cell sources need to be identified to overcome the ethical and logistical issues that are associated with using human fetuses. Here we attempted to generate iPS cells by introducing reprogramming factors using the piggyBac Transposon2 transduction system in human fetal fibroblasts and fetal neural stem cells. We wish to test the hypothesis that these cells are more easily reprogrammable and/or are more readily directed towards an MSN phenotype. The established iPS cell lines were similar to human embryonic stem (ES) cells in terms of their morphology, surface antigen, and proliferation. These iPS cells lines have been successfully manipulated to differentiate into MSNs in culture according to their expression of standard molecular markers of premature and mature MSNs - Ctip2 and Darrp32. Differentiation following transplantation into the quinolinic acid (QA) lesion model showed that grafts of these striatal progenitors derived from human fetal iPS cells could differentiate into neural progenitors according to expression of human nuclei marker (HuNu) and nestin.

There has been a continuous production of high quality reports focussing on fruitless as the genetic switch for male sexual behaviour in Drosophila melanogaster, and on fruitless’s contributions to creating a male-specific neural circuit within the CNS. However it has become increasingly clear that fruitless is not sufficient in itself to specify the full complement of male-specific behavioural repertoires. One obvious genetic candidate that contributes to the male neural circuit is doublesex. doublesex has long been known to be pivotal to the specification of the sexually dimorphic adult soma but it’s function in specifying sex-specific neural substrates has, up till now, been largely unexplored. While fruitless has so far shown to be found only in insects, doublesex is a more ancient gene and, as member of the Dmrt family of genes, is both structurally and functionally conserved throughout the animal kingdom. Thus the study of doublesex offers great potential for understanding the neuronal, developmental and physiological logic underlying innate and species-specific behaviours, in not one but both sexes, in organisms throughout the animal kingdom. Using the novel dsxGAL4 transgenic tool, generated by ends-in homologous recombination at the doublesex locus, I have been able to perform a systematic temporal and spatial survey of doublesex expression both within, and outwith, the nervous system. Excitingly, as doublesex is endogenously expressed in both males and females, this has uncovered profound dimorphic differences in male and female neural substrates. In the male this circuit is shared with fruitless (whose expression is restricted to adult males) and has allowed myself, and my colleagues in the Goodwin lab, through functional behavioural analyses, to gain greater understanding into how male-specific behavioural outputs may be generated. Further though, functional analyses impinging on the novel doublesex female circuitry has allowed us to gain new insight into the (largely unstudied) role that females play in the courtship ritual. The dsx GAL4 transgenic tool, and the insights gained in this study, are also of import in relation to dissecting out mechanisms involved in the post-mating physiological and behavioural changes the female undergoes after successful copulation with a male. As well as this, as doublesex is known to play a pivotal role in establishing the dimorphic morphology of the fly, this tool has begun to allow us an understanding of how the assembly of these dimorphic neural circuits is coordinated with the development, and maintenance, of a sex-specific anatomy and physiology to produce the complete male or female ‘state’; Integrating both mind (fly brain) and body (fly soma).

Mutations in the ruv gene of Escherichia coli K-12 result in an increased sensitivity to agents that damage DNA. Studies presented in this thesis demonstrate that the ruv gene product is required for conjugational recombination in certain genetic backgrounds. From this it was inferred that the role of the ruv gene product was in the recombination repair of daughter strand gaps and double strand breaks in damaged DNA. In addition, the ruv gene product is shown to be required for the efficient recovery of F' transconjugants in certain genetic backgrounds, suggesting that recombination between transferred F' and the recipient chromosome may be an obligatory event in these strains. Expression of ruv is regulated as part of the SOS response to DNA damage, by the lexA and recA gene products. The ruv gene product appears not to have any major role in its own regulation, however the basal level of expression of other SOS genes is increased in strains carrying ruv mutations. The ruv gene has been cloned on a 10.4kb HindIII fragment into the low copy number vector pHSG4l5, to give plasmid pPVA101, which has been demonstrated to complement the UV sensitivity of strains carrying any of the 10 different ruv mutations tested. Analysis of the proteins synthesised by pPVA101, its deletion derivatives, and derivatives with Tnl1000 insertions inactivating the ruv gene, allowed the identification of the ruv coding region, and suggested that the ruv gene encoded a 4lkd protein. In addition, regions of the cloned DNA coding for two further proteins of approximately 24kd and 33kd were identified. The sites of insertion of Mud(Ap)Rlac and Tn10 elements in the ruv gene were mapped, which allowed the direction of transcription to be determined, and suggested that the 4lkd protein may be cotranscribed with the 24kd protein from a promoter upstream of the smaller protein. This was substantiated by the demonstration that two of the ruv mutations studied were chromosomal inversions, one of which had its end point within the coding region for the 24kd protein, and by the isolation of an SOS inducible promoter derived from the region upstream of the 24kd protein. The nucleotide sequence of the ruv region revealed two open reading frames, designated ruvA and ruvB, with coding potential for proteins of 22087 daltons and 37177 daltons respectively, corresponding to the proteins with molecular weights estimated as 24kd and 41kd from SOS-polyacrylamide gels. A possible promoter, and two sequences with homology to the LexA binding site consensus sequence were identified upstream of the coding region of the 22kd protein. An amino acid sequence within the proposed RuvB protein was identified with homology to ATP binding sites of other proteins 950involved in DNA metabolism.

Density functional theory (DFT) and combined quantum mechanics/molecular mechanics (QM/MM) calculations have been used to model inter- and intra-molecular non-covalent interactions of transition metal complexes and where applicable their interactions with DNA. Two DFT functionals, BHandH and B97-D, which have shown to be efficient in modelling systems containing non-covalent interactions, have been tested against high level ab initio calculations on test transition metal complexes, designed to represent the intermolecular interactions present in the benzene dimer and methane benzene systems. The DFT functionals above show good agreement with the benchmark calculations and have been used to study ruthenium arene 'piano stool' type complexes, of the general form [h6(arene)Ru(en)Cl]+, which have shown potential as anticancer agents. The intramolecular interactions of these ruthenium complexes through coordination to guanine and adenine through the N7 nitrogen, has been explored using a selection of pure DFT, hybrid DFT, and post Hartree-Fock approaches against benchmark correlated wavefunction methods, where the best methods were found to be BHandH, B97-D2, and MP2(0.25). The B97-D2 functional was used to model these ruthenium complexes, with a selection of extended aromatic ligands with potential to act as intercalators, interacting with base pair steps. Calculated binding energies show a sensitivity to the nature of the arenes, where the more flexible ligands form more non-covalent interactions with DNA, as demonstrated by QTAIM analysis. Conformations and binding energies of a relatively new platinum anticancer drug, kiteplatin, with small single strand fragments of DNA, have been studied using B97-D and semi-empirical methods and compared to established drugs cisplatin and oxaliplatin. Isotropic shielding values and J coupling constants have also been calculated for these systems to relate these values to conformational data. Extended dual strand kiteplatin-DNA adducts have been studied using the QM/MM method ONIOM, combining BHandH with AMBER, to calculate binding energies and optimised structures. These results show that as the DNA adduct increases in size the values of the kiteplatin energies start to converge and comparison of base pair parameters show that around the site of coordination all fragments show comparable geometrical distortions.