Pharma industry currently works on shortening the time required to manufacture drug candidates to support early preclinical testing, and reduces the failure rate by early screening of the molecules with the best potential to reach the market later on. This thesis focused on the development and integration of three aspects of manufacturing to contribute in achieving these objectives. The first one looked at the way to approach experimentation. The use of DOE was generalised to the whole development and the creation of a library of statistical tools to develop an algorithm for the analysis of the commonly used Central composite and Box-Behnken designs. This algorithm proved to be as efficient as commercially available statistical softwares but presents the advantage of automating the analysis of the DOE designs. The second aspect consisted in developing viable manufacturing steps such as an upstream PEI-mediated transfection process in CHO cells, capable of generating 100mg L-1 of product in less than 15 days, or a cation exchange purification platform using a Quality by Design approach. The third aspect focused on the integration of different processing steps to yield a whole integrated platform for the production, then purification of one gram of a reporter antibody. This platform proved to be a cost effective alternative to the development of a stable producing cell clone for the production of recombinant product. In addition, this research reveals that the adoption of a statistically driven approach to process development is as important as the implementation of innovative technologies to address the challenges ahead.

Background: The anaesthetic gas xenon provides long-lasting preservation of neuronal function when given several hours prior to neuronal injury. This phenomenon is described as preconditioning. However, little is known of the mechanisms by which xenon preconditioning mediates its protective effect. Interest has focused on the involvement of hypoxia-inducible factor-1α (HIF-1α) in preconditioning because: (i) Hypoxia activates HIF-1α and is an effective preconditioning stimulus; (ii) HIF-1α is a regulator of adaptive responses promoting cellular survival; (iii) Several genes that have hypoxia-responsive elements (HRE) for HIF-1 in the promoter region are capable of mediating preconditioning (e.g., erythropoietin). Therefore, this study speculates on the role of HIF-1α in xenon preconditioning. Method: Separate cohorts of male adult C57/BL6 mice were preconditioned by exposure to 75% xenon/25 % oxygen for 2 hours and thereafter immediately sacrificed for organ harvesting at 0-24 hours after xenon preconditioning. Semi-quantitative study of HIF-1α and EPO protein expression was performed by Western blot analysis, and RT-PCR was used as a measure of gene transcription. Results: Xenon preconditioning provokes a time-dependent increase in HIF-1α protein in brain as well as kidney. Xenon preconditioning also caused a time-dependent increase in EPO (a HIF-1 target gene) transcription and protein expression in a corresponding time course to xenon-induced HIF-1α. To elucidate the mechanisms of xenon-induced HIF-1α accumulation, the effect of xenon preconditioning on HIF-1α transcription, translation and degradation was studied. Xenon preconditioning does not induce a change in HIF-1α mRNA expression, nor does it significantly attenuate the expression of the PHD2 enzyme, required for HIF-1α degradation. To explore translation-dependent pathways, mice were treated with rapamycin before xenon preconditioning, to inhibit translation of HIF-1α through the mTOR pathway. Inhibition of this pathway prevented the xenon-induced increase of HIF-1α protein. To ascertain whether HIF-lα is required for xenon preconditioning, siRNA was used to knockdown HIF-lα expression in the kidney and xenon's renoprotective properties were shown to be abolished. Conclusions: Over the same time course as xenon's protection against subsequent injury in both brain and kidney, xenon preconditioning induces expression of HIF-1α. Increased HIF-1α expression is also associated with increased activity as evidenced by enhanced transcription and translation of the downstream effector, EPO. Xenon preconditioning does not regulate HIF-1α at the transcriptional level, nor does it inhibit HIF-lα degradation. However, these results suggest that xenon preconditioning upregulates expression of HIF-1α through translation-dependent mechanisms. Furthermore, xenon's action on HIF-1α is shown to be causally related to its organ protective effect. If these data can be extrapolated to the clinical setting, exposure to xenon would be beneficial prior to procedures in which organ perfusion is interrupted, preventing hypoxic-ischaemic as well as ischaemic-reperfusion injury.

Vascular injury and endothelial dysfunction are recognised features of atherosclerosis, transplant vasculopathy, SLE and rheumatoid. Heme oxygenase-1 (HO-1) and complement inhibitory proteins (CIP) decay-accelerating factor (DAF) limit vascular injury. There is considerable interest in therapeutic manipulation of cytoprotective genes and statins induce HO-1 and DAF expression, while rapamycin (R) enhances HO-1 and protects against post-transplant vasculopathy. A therapeutic regimen combining immunosuppression with vasculoprotection has the potential to prevent accelerated atherosclerosis in systemic inflammatory diseases. Based on previous data generated in the lab, my project explored the combination of atorvastatin and rapamycin, using DAF and HO-1 as target genes. A combination of AT 0.5μM and R 1μM led to a synergistic increase in EC DAF expression 24-72h post-treatment (>200%, p<0.05). In contrast, no change in CD59 was seen. In vivo, AT+R enhanced DAF expression in the murine aorta when compared to treatment with AT or R alone. The functional relevance of this synergy was revealed by enhanced protection against complement-mediated lysis, when compared to EC treated with R or AT (35% v 75% and 55% respectively, p<0.05). The role of DAF was confirmed by loss of protection in the presence of a DAF inhibitory mAb. Mechanistically, AT+R increased HO-1 expression and activity, while HO-1 inhibition with ZnPP attenuated synergy. As a consequence of HO-1-activation, AT+R increased intracellular ferritin, while Fe2+ chelation with DFO suggested depletion of intracellular Fe2+ is important for synergy. AT+R enhanced PKCα phosphorylation, while a DN-PKCα adenovirus and a PKCα inhibitory peptide abrogated DAF induction (AT + R 250% v DN-adv 95%, p<0.05). AT and R-induced p38 MAPK phosphorylation and inhibition attenuated DAF upregulation (AT+R 270% v 120%, p<0.05). Transcriptionally, DAF induction by AT+R required activation and binding of CREB to the DAF promoter. Synergistic upregulation of DAF expression was reproduced by simvastatin and rapamycin, while combinations of AT and cyclosporine or mycophenolate were ineffective. Treatment with rapamycin and atorvastatin synergistically enhances DAF expression and protection against complement-mediated injury. Rapamycin increases HO-1 activity and intracellular ferritin, and this combined with HMG-CoA reductase inhibition and activation of PKCα, p38 MAPK and CREB, results in optimal DAF induction. Thus, rapamycin and statin combination therapy may represent a means by which vascular endothelium can be therapeutically conditioned against complement-mediated injury and is worthy of further study in vascular diseases.

Growing pressures in the pharmaceutical industry are driving the need to optimise processes used for the manufacture of drugs at commercial-scale, in order to improve cost of goods, product throughput and production times. Evaluating the impacts of process optimisation upon these metrics presents a challenge due to complexities and trade-offs that are often encountered when developing a typical bioprocess. Such factors have resulted in a range of novel simulation- and experimental- based techniques being developed which enable rapid, accurate and cost effective assessment of manufacturing options for commercial-scale production. This thesis proposes a combination of modelling and experimental methods for evaluating the business- and process-related impacts of implementing changes to pre-existing commercial-scale pharmaceutical manufacturing processes. The approaches are illustrated through an industrial case study, focusing upon a process operated by Protherics U.K. Limited for the manufacture of the FDA-approved rattlesnake anti-venom CroFab (Crotalidae Polyvalent Immune Fab (Ovine)). The novel methods developed and illustrated in this thesis include: Investigating the effects of process changes upon calculated yields and processing times within the production framework for a pre-existing FDA-approved bio-manufacturing process Evaluating the impacts of both developing and implementing process changes, combining output metrics into a single value to simplify the assessment Developing a multi-layered simulation methodology for the rapid and efficient evaluation of bio- manufacturing process options Applying advanced sensitivity analysis techniques to identify the most critical factors that influence product yield and throughput Evaluating a novel synthetic Protein A matrix for the recovery and purification of polyclonal antibodies from hyperimmunised ovine serum Developing decision-support software to aid the design of chromatography steps for antibody purification at industrial scale Demonstrating the utility of such models by application to data and constraints derived from a full-scale industrial facility.

The aim of this project was to explore and refine the conformational rationale for the activity of capsazepine (CPZ) as a blocker of the ion channel TRPV1 (transient receptor potential vanilloid type 1), by the synthesis and biological evaluation of further conformational constrained capsaicin analogues. The resolution of the stereoisomers of Af-(4-chlorophenethylthiocarbamoyl)-6,7-dihydroxy-1-methyl-1,2,3,4-tetrahydroisoquino-line 29, Af-(4-chlorophenethylthio-carbamoyl)-6,7-dihydroxy-3-methyl-1,2,3,4-tetra-hydroisoquinoline 30 and N-(4-chlorophenethylthiocarbamoyl)-6,7-dihydroxy-1,3-dimethyl-1,2,3,4-tetrahydroiso-quinoline 31 by stereoselective synthetic methodology is described, and some of the more unusual and interesting mechanisms are discussed. The novel asyrnmetric chemistry described includes the separation of the enantiomers of 2-(3,4-dimethoxyphenyl)-l-methylethylamine 38 by crystallisation with the enantiomers of mandelic acid, the use of sodium triacetoxyborohydride in the stereoselective reduction of 6,7-dimethoxy-l,3-dimethyl-3,4-dihydroiso-quinoline 171, to give the cw-diastereomers of 6,7-dimethoxy-1,3-dimethyl-1,2,3,4-tetrahydroisoquinoline 44, and the novel stereoselective route to the trans-diastereomers of 6,7-dimethoxy-l,3-dimethyl-l,2,3,4-tetrahydroisoquinoline 44 from the enantiomers of 2-(3,4-dimethoxyphenyl)-l-methylethylamine 38 by the Michael addition of A-benzyl-2-(3,4-dimethoxyphenyl)-l-methylethylamine 155 to ethynyl-4-tolylsulfone 150, followed by TFA-mediated cyclisation, single electron reductive desulfonylation and palladium-catalysed hydrogenolysis. The results of investigations into the conformational behaviour of the resolved stereoisomers of 29, 30 and 31 by techniques of NMR spectroscopy and molecular modelling, the evaluation of their biological activity at the rat and human orthologues of the ion channel TRPV1, and the attempted correlation of the two sets of data, with respect to the published conformational rationale for the activity of CPZ, are also described.

Intracellular recordings were made from pyramidal neurones in the CA1 region of the rat hippocampus to investigate aspects of cholinergic synaptic transmission in the mammalian brain. In the presence of ionotropic glutamate and GABA receptor antagonists, single shock stimulation of the septohippocampal cholinergic input in <I>stratum oriens</I> evoked a slow excitatory postsynaptic potential (EPSP) which was associated with an increase in input resistance. Stimulation at intensities sub-threshold for evoking a slow EPSP caused a reduction of spike frequency adaptation. Pharmacological analysis of these responses revealed that they were mediated by muscarinic acetylcholine receptors (mAChRs). Having established a protocol for evoking reproducible mAChR-mediated EPSPs (EPSP<SUB>M</SUB>s) and mAChR-mediated inhibition of spike frequency adaptation, experiments were conducted to investigate the modulation of these responses by other neurotransmitter receptors. In this respect, activation of adenosine receptors by the broad spectrum adenosine receptor agonist 2-chloroadenosine (CADO) reversibly inhibited mAChR-mediated synaptic responses in a concentration dependent manner. The pharmacological profile of this effect established that it was mediated by adenosine A<SUB>1</SUB> receptors. In conclusion, it is possible to evoke isolated muscarinic acetylcholine receptor mediated synaptic responses with a single stimulus in the CA1 region of the rat hippocampus. Adenosine A<SUB>1</SUB> receptors are present on cholinergic terminals in this region and act to inhibit these responses by an unknown mechanism. GABA<SUB>B</SUB> receptor activation also inhibits cholinergic synaptic transmission although the mechanism and site of action of this is unclear. Finally, endogenous adenosine and GABA, acting at adenosine A<SUB>1</SUB> and GABA<SUB>B</SUB> receptors respectively, inhibit cholinergic synaptic transmission <I>in vitro</I>. This modulation provides a potentially important mechanism for the control of neuronal excitability in the hippocampus <I>in vivo</I>.

Portal, systemic, cardiopulmonary and renal haemodynamics of 96 patients with alcohol related cirrhosis were measured. Their inter-relationship and predictive value for variceal haemorrhage and survival during a mean follow-up of 19 months was investigated. Severity of liver disease was related to the hepatic venous pressure gradient (HVPG), azgos blood flow (AzBF) and systemic hypotension. During follow-up, HVPG predicted survival and variceal bleeding. Propranolol and isosorbide-5-mononitrate are widely used in the prophylaxis of variceal haemorrhage, but recent reports have suggested they may compromise renal function. Renal blood flow (RBF), HVPG, AzBF and systemic haemodynamics were measured in 26 cirrhotic patients before and after each drug or combination therapy. Despite significant changes in the other parameters, no fall in RBF was detected. The novel vasodilating beta-blocker carvedilol offers potential in the treatment of portal hypertension, but little data currently exist. Portal, cardiopulmonary and systemic haemodynamics and renal function were assessed in 17 cirrhotic patients after acute and chronic (1 month) therapy. Although carvedilol had a continued portal hypotensive effect after 1 month with no detrimental effect on liver blood flow or renal function, a significant minority of patients were unable to tolerate chronic therapy. Adenosine-antagonism offers a new therapeutic approach to cirrhotic ascites and renal dysfunction. The effects on renal and systemic haemodynamics and renal function were assessed following administration of FK352 (a novel adenosine-1 antagonist) to 12 cirrhotic patients with ascites. An improvement in natriuresis, diuresis and RBF was detected.

In this thesis the effects of chronic antidepressant treatments and MDMA treatment on the density and affinity of a variety of regional rat brain 5-HT and non-5-HT transporters and receptors is investigated using radioligand binding and immunological techniques. The influence of potentially neuroprotective drugs on the effects of MDMA is examined. Since SERT mRNA has been detected in the rat adrenal gland, the effects of antidepressant and MDMA treatment on adrenal SERT has also been determined. A novel [³H]5-CT membrane binding assay, in the presence of specific 5-HT receptor subtype masking drugs, revealed a pharmacology consistent with 5-HT₇ receptor binding. The pharmacological profile of [³H]GR125, 743 membrane binding was consistent with binding to 5-HT_1B/1D receptors, known to regulate 5-HT neurotransmission. Decreases in 5-HT₇, 5-HT_1B and 5-HT_1Dreceptor density were observed in rat frontal neocortex after chronic SSR1, but not tianeptine (atypical antidepressant), treatment. No adaptive changes were observed for noradrenaline transporters, 5-HT_1A or dopamine₁receptors. Adrenal gland SERT displayed an identical pharmacology to brain and platelet SERT and was also unaffected by such treatments. A library of site directed antibodies raised against SERT were characterised. A monoclonal antibody which specifically recognised denatured SERT and displayed the appropriate immunohistochemical labelling, also immunoprecipitated SERT and specifically recognised the native form from rat cortex and platelets. SERT protein was specifically detected in the rat adrenal medulla by this antibody. MDMA exposure caused a similar depletion in the number of SERT binding sites in both the brain and adrenal gland. GK506 protected brain neurotoxicity, but did not protect adrenal SERT depletion. The MDMA SERT reduction was still apparent after 13 weeks in the brain, but not in the adrenal glands. In the brain, MDMA neurotoxicity was also rescued by a free radical scavenger.

The thesis describes investigations concerning the effects of lithium on ATPases and ion transport. Effects of potassium and lithium on sodium transport from blood to cerebrospinal fluid and brain were studied using a ventriculo-cisternal perfusion technique. The fast component of sodium entry into cerebrospinal fluid was correlated with the cisternal potassium concentration. This together with results showing the potassium-dependance of choroid plexus Na/K ATPase activity were interpreted as showing choroid plexus sodium transport to be sensitive to cerebrospinal fluid potassium concentration and to be due to Na/IC ATPase activity in the apical membranes of the plexuses. Effects of lithium on sodium transport were interpreted as a stimulation of the sodium pump when presented to the potassium-sensitive side and an inhibition when presented to the sodium-sensitive side. The effects of chronic lithium administration on ATPase activities were investigated in subcellular fractions prepared from rat cerebral cortex. Lithium was found to have no effect on synaptic plasma membrane Na/K ATPase activity but Mg ATPase specific activity was increased both in the membrane fraction and in mitochondria, after 21 and 7 days administration respectively. The changes in enzyme specific activity were not due to changes in fraction composition. Lithium administration for 21 days caused an increase in the concentration of homovanillic acid and 3,4-dihydroxyphenylacetic acid in the striatum. This was interpreted as showing an increased release and turnover of dopamine. ATPase activities were also studied in erythrocyte membrane preparations. Depressive patients were shown to have a reduced erythrocyte membrane Na/ATPase specific acitivity compared to controls. It is suggested that such a reduced enzyme activity could explain many of the previously observed derangements of sodium metabolism in depression. Lithium-treated patients showed increased specific activities of both Mg ATPase and Na/K ATPase in erythrocyte membranes; it is suggested that the increase in Na/K ATPase activity was due to recovery of the patients but that the increase in Mg ATPase activity was due to lithium itself. Lithium administration had no effect on erythrocyte membrane ATPase activities in the rat. It is concluded that chronic lithium administration affects the specific activities of certain Mg ATPases but not the Na/K ATPase. It is speculated that the changes in membrane ATPase activities reflect changes in actomyosin-like proteins which could lead to increases in the release of neurotransmitter in the brain.

Acutely ovariectomised adult female Wistar rats were used to study the ability of oestrogen and progesterone to alter central 5-HT neurotransmission, with particular interest in changes in 5-HT_2A receptor and serotonin transporter (SERT) binding site density and the firing characteristics of 5-HT neurones in the dorsal raphe nucleus (DRN). Changes in 5-HT_2Abinding site density were evaluated using quantitative autoradiography on serial coronal sections using [³H]MDL 100,907 and [³H]ketanserin with non-specific binding being assessed using RP 62203. Binding site density was measured in the dorsal and median raphe nuclei, which contain serotonergic cell bodies, and regions of the forebrain which receive projections from these raphe nuclei e.g. the cortex and hypothalamus. The experiments examined the effects of oestrogen and progesterone on binding site density while also comparing the binding patterns of the two 5-HT_2A receptor ligands. Preliminary experiments were carried out using coronal sections containing cingulate cortex to obtain the optimal conditions for [³H]MDL 100,907 binding experiments. Similar binding patterns were observed throughout the brain using [³H]MDL 100,907 or [³H]ketanserin although higher non-specific binding was observed with ketanserin. Specific binding levels differed between the two ligands with higher levels being observed in the cingulate cortex, nucleus accumbens and olfactory tubercle and lower levels in the amygdala and dorsal raphe using [³H]MDL 100,907 compared to [³H]ketanserin. Acute oestrogen treatment produced no significant differences in binding site density in any of the regions examined when [³H[MDL 100,907 was used; however, an increase in binding was observed with [³H]ketanserin in the cingulate and frontal cortex. Progesterone treatment alone or in combination with oestrogen produced no significant differences in binding site density in any of the regions examined with either [³H]MDL 100,907 or [³H]ketanserin. These results suggest that progesterone has no effect alone but can attenuate the effect of oestrogen on ketanserin binding in specific brain regions.