Search

Patients with epidermal growth factor receptor (EGFR) sensitizing mutations in non small cell lung cancer (NSCLC) receive benefit from Tyrosine Kinase inhibitors. Accurate selection of patients before treatment is highly dependent on precise molecular diagnosis of EGFR mutations. Presently in the clinic, the diagnostic samples routinely used tumour biopsy and/or cell free DNA (cfDNA), are not sufficiently effective for precise diagnosis. Circulatory tumour cells (CTC) in blood have been explored successfully as alternative and complementary diagnostic markers to the current clinical tools. However, utility in the clinics has been hampered by the relatively low concentration of CTC in blood, and the lack of robust technologies that are adaptable for routine use. The present study describes the design and optimization of an immunomagnetic based microfluidic device (Lung card version II) that isolates CTC expressing the epithelial cell adhesion molecule (EpCAM) from blood with high capture efficiency and purity. The device is a 2-part system comprising a disposable chip that is simple in design and a reusable microfluidic unit that contains a mobile magnetic arm. The simple design and work-flow process of the device ensures cost efficiency for scalability and, ultimately, use in the clinic. The device was initially validated for its capability to isolate EpCAM positive cells. Results from spiking carboxylfluorescein succinimidyl ester stained EpCAM positive cells in media/blood showed a capture efficiency of ≥ 65% and a purity ≥ 97% from a 13ml sample in 50 minutes. The isolated CTC from NSCLC patients (n=38) were analysed for mRNA markers specific to malignant cells and were characterized for EGFR mutations following PCR and next generation sequencing. The mutational status of CTC was compared to that obtained from matched, tumour biopsy, samples. Significantly more mutations (P=0.0173) were detected in CTC enriched samples than the matched biopsy. Interestingly, mutations were detected in only 4 biopsy samples and the mutations detected in the biopsy were only concordant with results from CTC enriched samples for 1 patient. Exon 19 deletion was the most frequent mutation detected (86.7%) with rare mutations such as: L792P, C797S, H509R also been detected in CTC, and the present study reports the detection of K708R mutation in NSCLC for the first time. The clinical outcomes of patients who were positive for EGFR mutation from CTC, but had been placed on therapies based on mutation results from tissue biopsy were evaluated in this study. The results showed that no significant progression free survival (PFS) benefit was attained when comparing treatment response between patients whose CTC possessed an EGFR mutation and patients whose CTC possessed no EGFR mutation (10 months vs26 months p value-0.3420 HR- 0.76 95% CI- 0.2498-2.319). In summary the results from this study showed that the microfluidic device captured CTC with efficiency equal to other immuno-affinity based devices but had better purity rates and throughput and also that the device can be utilized for CTC processing for downstream analysis. Results from this current study further demonstrated the clinical potential of CTC+NGS matrix for the detection of EGFR mutations and the prospective impact it would have for precision oncology in NSCLC are discussed.

Despite major advances in our understanding of cancer progression mechanisms, cancer remains a leading cause of death in the UK and globally. Solid tumours are complex, and heterogeneous, existing in a dynamic microenvironment subjected to a variety of biomechanical cues. Conventional cancer research methods have long relied on two-dimensional (2D) static cultures which neglect the dynamic, three-dimensional (3D) nature of the tumour microenvironment (TME), especially the role of interstitial fluid (ISF) and the impact of interstitial fluid flow (IFF). In order to address this, we have developed and optimised a spheroid-on-chip microfluidic platform which allows 3D cancer spheroids to be integrated into extracellular matrices (ECM)-like hydrogels and exposed to continuous perfusion, mimicking IFF in the TME. Our platform enables clear imaging of spheroids, analysis of effluent media, and harvesting of spheroids for further analyses. Where many studies have focused on perfusion as a tool mainly for high-throughput methods, we aim to identify how perfusion, mimicking IFF in the TME, alters the biology of cancer spheroids. Spheroids exposed IFF-like or static conditions were harvested, and gene and protein expression analysis was performed. These data indicated that gene expression was altered by flow when compared to static conditions, but clear functional patterns were not identifiable. Subsequently, a large-scale transcriptomic analysis by RNA-sequencing was used to conduct an unbiased analysis, revealing robust gene expression changes in genome preservation pathways and leading to the identification of a potential biomarker of IFF in cancer. Our spheroid-on-chip flow platform has revealed that exposure to IFF-like conditions in a 3D environment significantly altered the transcriptome of spheroids after just 24 hours, leading to changes in the expression of key factors involved in various aspects of tumour progression. These results will direct our future work on further elucidating the role of ISF flow in cancer biology and spread.

Background and Aim of the study: Lung cancer is the third most common cancer in the UK. A significant number of these patients are diagnosed with inoperable advanced-stage non-small cell lung cancer. Until recently, the standard of care for these patients was radiotherapy with or without chemotherapy but the overall survival remained poor, with a 5-year survival of only 13%. Recent studies showed significant improvement in overallsurvival in patients who are suitable for, and have received, immunotherapy, in addition to chemotherapy and radiotherapy. Radiotherapy aims to deliver tumoricidal doses to the target volume whilst minimising doses to the surrounding organs at risk (OAR). However, achieving this goal could be challenging especially when treating advanced-stage tumours, as it could increase OAR doses and increase toxicities to an unacceptable level. Furthermore, several factors affect the achieved dose distribution including, patient's geometry, treatment technique, planner's experience, beam geometry, optimisation parameters, and interventions used during treatments. It is therefore important to develop methods to personalise treatment plan optimisation for advanced-stage non-small cell lung cancer (NSCLC) patients to achieve minimum OAR doses without compromising target doses using patient-specific parameters. This study aims to develop knowledge-based planning models (KBP) using patient-specific factors todetermine personalised treatment planning optimisation for advanced-stage NSCLC patients treated with volumetric modulated arc therapy (VMAT) to reduce OAR doses whilst delivering intended doses to the target volume. Methods: Four KBP models were developed using patient-specific dose and volume parameters to predict minimum achievable OAR doses, identify optimal arc parameters, trigger adaptive radiotherapy and estimate doses to adapted gross tumour volume using patients' geometry. The KBP models were verified using independent patient data sets. Change in treatment plan optimisation could increase modulation and affect plandeliverability therefore, several modulation indexes were calculated and plans were measured on the clinical linear accelerator to assess the effect of change in optimisation on treatment plan delivery. Results: The KBP models developed showed that relatively simple models can predict OAR doses and arc parameters and help identify patients for adaptive radiotherapy. The models can accurately estimate personalised and progressive dose escalation. The KBP resulted in a significant reduction in plan variability in all three studied dosimetric parameters, volume of lungs receiving 5Gy (V5), 20Gy (V20) and mean lung dose (MLD)by 4.9% (p=0.007, 10.8% to 5.9%), 1.3% (p=0.038, 4.0% to 2.7%) and 0.9Gy (p=0.012, 2.5Gy to 1.6Gy), respectively. The individualised arc geometry resulted in a significant reduction in lungs (V5 = - 15.1%, MLD = - 1.0Gy) and heart (MHD = - 1.4Gy) doses without compromising target coverage. The models, which were developed to predict changes in PTV coverage (ΔV95^PTV) using a specific biomarker (Programmed death-ligand 1 (PD-L1 expression)) and the difference in 'planning' and 'fraction' planning target volume (PTV) centre of the mass (characterised by mean square difference, MSD), could predict change in PTV coverage within ℗ł 1.0% for 77% of the total fractions. Furthermore, the models developed for predicting personalised and progressive dose escalation predicted doses within 0.4% and 0.7% respectively. Additionally, the plan complexity and deliverability measurements show that plan complexity could increase but may not affect treatment delivery significantly. Conclusion: The studies performed show that altering the 'standard' treatment planning optimisation approach could significantly reduce OAR doses and improve target coverage. This will help reduce toxicities and improve local control and overall survival and outcome for inoperable advanced-stage NSCLC patients.

Glioblastoma (GBM) is a deadly disease with poor prognosis. The gold standard treatment is surgery followed by chemoradiotherapy almost always including Temozolomide. Despite extensive research for new therapeutic approaches relapse after treatment still occurs in approximately 80% of patients. The median survival, after first diagnosis, remains at 15 months, with a 5-year survival rate still less than 5%. Translationally-relevant models are urgently needed to allow reliable identification and validation of new potential therapeutic targets. The current study aims to establish a microfluidic flow-based system for identifying targets via gene knockdown in a GBM spheroid model. Two genes, PRMT2 and RAB21, were chosen as suitable targets for doing knockdown experiments, following a review of the literature and demonstration that both were expressed in the U87 GBM cell line. Gene knockdown was initially undertaken using lipofectamine based siRNA transfection of U87 cultured in monolayer, 3D static and 3D flowing systems. Gene expression was assessed by qPCR whereas western blotting and MTS assays were used to measure protein expression and cell proliferation, respectively. Accel™ siRNA, a chemically modified form of siRNA, were used to try and improve the gene repression levels. Finally, a comparison of the whole transcriptome profile between spheroids from three, independent, GBM cell lines was conducted both pre- and post-maintenance on the microfluidic device. The microfluidic chip was demonstrated to successfully maintain GBM spheroids in a viable state for up to 7 days, during which the RNA yield was significantly higher compared to the 3D static conditions for three independent genes (PRMT2, RAB21 and FUS). Respectively, PRMT2 and RAB21 siRNA silencing levels showed a gradual decrease from monolayer (75.66±5.13% vs 86.33±9.01%), to 3D static (43.00±24.75% vs 59.00±4.58%) and 3D flow system (26.33±2.51% vs 19.67±8.02%) using 5nM of siRNA. Gene repression of PRMT2 and RAB21 had no effect on the protein expression or on the proliferation rate. This study also showed that Accell siRNA was effective in inducing gene silencing in U87 spheroids maintained in static conditions (with 1%FBS, 49±0.09% of FUS knockdown level and 29.71±0.03% of protein depletion using 0.5µM of Accell siFUS). Next Generation Sequencing analysis of the transcriptome of three GBM cell lines showed that the correlation in gene expression pattern was higher between two patient derived cells (PDCs) than between them and U87. Furthermore, PDCs presented a heterogeneity in activated and deactivated pathways similar to developing tumour tissues, compared to U87 which was characterised by a predominance of pro-proliferative pathways. In addition, gene ontology and Kyoto Encyclopedia of Genes and Genomes enrichment analysis demonstrated that GBM spheroids when cultured in the microfluidic system compared to the static conditions, exhibited a more in vivo-like profile by activating pathways considered as hallmarks of cancer. These included pathways and cellular functions involved in: proliferation, inflammation, metabolism dysregulation, senescence, necrosis, and invasion. In conclusion, this study showed that siRNA mediated gene knockdown was successful in the 3D flowing microfluidic platform, however no effects were observed on protein expression or cell proliferation; this may be due to the half-life of mRNA and will require additional study. Furthermore, PDCs spheroids cultured in this flow system compared to static state, exhibited properties more comparable to that of in vivo tumour tissues. Further investigation using a range of other genes or comparing spheroids with primary human tissue would be likely to give additional evidence about the effectiveness of the current flow system for mimicking the in vivo microenvironment.

Despite intense research, the prognosis of patients with glioblastoma (GBM) remains extremely poor, with a median survival of 1.5 years after diagnosis. The need for new therapeutic options is therefore a vital step towards securing a better outlook for patients. Protein Arginine Methyltransferases (PRMTs) are novel targets in current oncology clinical trials due to their ability to augment a vast variety of cellular processes including growth factor signalling as well as Deoxyribonucleic acid (DNA) damage repair and proliferation. There are three known types of PRMT with varying target specificities and activities. All PRMT's are able to transfer a single methyl group onto the guanidino nitrogen on the target arginine, producing the monomethyl-arginine (MMA) mark. Type I PRMTs (PRMT1, PRMT2, PRMT3, PRMT4, PRMT6 and PRMT8) then transfer a second methyl group on the same nitrogen to produce the asymmetrical dimethyl-arginine (ADMA) mark. Type II PRMTs (PRMT5 & PRMT9) transfer a second methyl group to the opposite nitrogen on the arginine functional group, making symmetrical dimethyl-arginine (SDMA). The type III PRMT, PRMT7, is only able to carry out the first methylation step. The generally accepted view is that PRMT expression is increased in tumour cells, including GBM. Therefore, the overarching hypothesis here is that GBM cells are dependent on PRMT expression, and the inhibition of these enzymes will result in cell death or reduced proliferation. This research aims to determine the efficacy of PRMT inhibitors as a therapeutic avenue in GBM treatment by measuring the anti-proliferative effects of PRMT inhibiting drugs on GBM cells. A further aim of this research is to characterise an ex vivo microfluidic model that would allow for patient specific testing of pharmacological drugs in a simulated tumour microenvironment. Using the U-87MG glioma cell line as a model system, differing levels of efficacy of PRMT inhibitors were observed. PRMT1 inhibition by Furamidine resulted in a reduction in cellular viability, however, inhibition of PRMT1 and other type I PRMTs by MS023 had minimal effect, as measured by MTS assays in Two-dimensional (2D) and Three-dimensional (3D) culture U-87MG models. Western blotting revealed an increase in the symmetrical dimethylation of a ~70 kDa protein following inhibition of Type I PRMTs with MS023, suggesting cross-talk between Type I and Type II PRMTs. To further investigate the use of PRMT inhibitors and possible cross-talk mechanisms, an ex vivo microfluidic model was utilised, where patient biopsies were perfused with growth medium with and without MS023. Patient samples could be maintained in a viable state on the microfluidic device for up to 192 hr (8 days), as determined by the minimal release of the membrane contained enzyme lactate dehydrogenase into the effluent (LDH assay). However, a reduction in proliferation was observed in samples incubated on chip compared with the matched fresh biopsies. Treatment of patient samples with MS023 did not result in an increase in cell death as determined by LDH assay or a reduction in proliferation as determined by the quantification of ki-67 positivity by immunohistochemistry (IHC). The cross-talk activity seen in the earlier cell line models was further demonstrated in the ex vivo microfluidic model. In patient biopsies six out of nine and four out of five patient samples showing increases in SDMA and MMA following MS023 treatment, respectively. To identify specific proteins that had undergone changes in methylation status, heavy methyl SILAC mass spectrometry was used. With this methodology, the RNA binding protein FUS was found to be methylated in the presence of MS023 only. To further understand the effects of the ex vivo incubation of patient samples and explore the presence of intra-tumour heterogeneity, samples were treated with Temozolomide (TMZ), the current chemotherapy drug used is clinical practice, and cell death and proliferative activity determined by LDH assay and ki-67 positivity through IHC. Treatment with 10 μM TMZ did not result in an increase in cell death or a reduction in proliferation. However, a reduction in ki-67 was again seen in the samples incubated on chip compared with fresh biopsies. To explore intra-tumour heterogeneity, multiple slices from the same patient biopsy were ran in parallel. The slices appeared to have variation in ki-67 positivity, suggesting that the rate of proliferation is not uniform within the tumour mass and could perhaps mask any response to drug treatment during analysis. In order to elicit a tumour response in GBM patient biopsies, MS023 was given in combination with TMZ, or GSK591, a PRMT5 inhibitor. Despite this aggressive approach, there were no significant changes in cell death or proliferative activity. It is hypothesised that the reduction in proliferative activity seen following incubation on the microfluidic device could reduce sensitivity to anti-tumour drugs and mute any changes that would otherwise be observed. To conclude, this study has explored the feasibility of using a GBM-on-chip model to investigate emerging drugs, that is, PRMT inhibitors, towards the treatment of GBM. Further improvements to the microfluidic model, or broader analysis of the effect of TMZ and PRMT inhibitors on other cellular processes, such as apoptosis and senescence, are needed. This study has presented exciting cross- talk activity amongst PRMT enzymes not previously demonstrated in GBM tissue which warrants further investigation.

Thrombotic complications are common in cancer patients and are amongst the leading causes of morbidity and mortality. Recently, it was reported that treatment of cancer patients with the low molecular weight heparin (LMWH) family of anticoagulants can result in a significant reduction in mortality compared with those receiving other conventional anticoagulants. Therefore, the mechanisms by which different anticoagulants interact with cancer cells were investigated. First, the influence of two preparations of LMWH (tinzaparin and dalteparin) and two direct fXa inhibitors (apixaban and rivaroxaban) on cancer invasion, angiogenesis and tumour growth was explored. Tinzaparin and to a lesser extent dalteparin, were shown to inhibit cancer cell invasion and angiogenesis, using matrix invasion and chorioallantoic membrane (CAM) assays, respectively. In addition, apixaban, but not rivaroxaban inhibited the growth of CAM-implanted xenografts which indicated a mechanism independent of the anti-fXa property of apixaban. The anti-tumour property of apixaban was further explored by comparing the influence of apixaban and rivaroxaban on cancer cell proliferation and microvesicle (MV) release. Apixaban was shown to inhibit the release of MV in response to exogenous fXa and fVIIa, as well as endogenous fVIIa expressed by the cancer cells. In contrast rivaroxaban was effective in suppressing MV release in response to fXa only. The mechanism of inhibition of endogenous fVIIa by apixaban was also shown to include the suppression of complex formation with TF and therefore, preventing the activation of protease activated receptor (PAR) 2 by the TF/fVIIa complex. In addition to signalling through PAR2, the interaction of tissue factor (TF) with β1 integrin is known to be required for induction of cell proliferation and contact-independent survival of cancer cells. Therefore, in the next section of the study an attempt was made to identify the function of the potential interacting domains between TF and β1 integrin. The cDNA corresponding to the upper or lower fibronectin-like domains, and also that encoding the complete extracellular domain of TF were separately cloned into an expression plasmid. These were expressed in human dermal blood endothelial cells (HDBEC) and MDA-MB-231 cell line and the ability of the resulting peptides to bind to β1 integrin was assessed in situ, using the proximity ligation assay (PLA) and also, by co-immunoprecipitation. The upper fibronectin-like domain (UED) was shown to bind to the EGF4 domain of β1 integrin and the lower fibronectin-like domain (LED) with the βTD domain. To examine the proliferative properties of these peptides, the expression of cyclin D1 and ERK1/2 phosphorylation were measured and changes in cell numbers determined. The interaction of either the UED or the LED peptide with β1 integrin in HDBEC, which do not express endogenous TF, was shown to induce ERK1/2 phosphorylation. However, displacement of endogenous TF by the UED peptide in MDA-MB-231 cells resulted in a reduction in ERK1/2 phosphorylation, cyclin D1 expression and cell numbers. In contrast, expression of LED peptide in MDA-MB-231 cells did not influence these parameters for proliferative signalling, indicating that the TF-signalling was maintained by the LED peptide. Collectively, these findings suggested that although both the upper and lower extracellular domains of TF can induce cell signalling, the lower domain may be essential for the promotion of proliferation. The last section of the study was based on previous work showing that prevention of the phosphorylation of TF can block the release of the protein in MV, resulting in a build-up of TF within cells. This study explored the role of TF depalmitoylation which is also known to precede and be essential for TF phosphorylation. It was hypothesised that such alterations may regulate the translocation of TF to lipid rafts and its release within MV. Mutant forms of TF with substitutions to prevent, or to mimic palmitoylation at cysteine 245, or with altered lengths of the transmembrane domain were prepared and expressed in cells. The influence of these proteins on cell proliferation and apoptosis were then examined. Expression of mutant forms of TF which could not be palmitoylated or containing a longer transmembrane domain resulted in decreased levels of apoptosis. However, neither the mutant mimicking palmitoylation, or containing a shortened transmembrane domain were not released within MV and resulted in increased cell apoptosis. In conclusion, this study has demonstrated the anti-cancer properties and some of the mechanisms involved, in a set of anticoagulants. In particular, apixaban was shown to suppress cancer cell growth by preventing the activation of PAR2 by the TF/fVIIa complex. In addition, this study has elucidated the role of the various extracellular and intracellular domains within TF, in promoting the signals that regulate cell proliferation and apoptosis.

This thesis reports on the synthesis, development and validation of a novel, non-toxic dual modal bio-imaging probe that targets a conformational molecular switch and relocalisation of cell surface Endo180 into endosomes that occurs during the early stages of aggressive prostate cancer. InP/ZnS quantum dots were synthesised using a successive heating methodology following the introduction of a cooling phase between the addition of 'shelling' precursors. Photoluminescent quantum yields following the synthesis of InP/ZnS (2) quantum dots are reported at 85.30% in hexane and 57% in water (following phase transfer). An OI/MR dual-modal imaging probe was successfully synthesised following the surface modification of InP/ZnS (2) - TGA with Ln(III) DOTA chelate derivative (Ln = Gd, Eu and Tb). Up to 190 Gd(III) chelates were added to the surface of InP/ZnS (2) - TGA resulting in a combined r1 value of 1609.94 mM-1s-1 (InP/ZnS (2) - Gd.5a) when InP/ZnS (2) is synthesised with InI3. High values of photoluminescent quantum yield were achieved of up to 19.9% (InP/ZnS (2) - Gd.5.b) with no shift in emission properties observed following surface functionalisation when compared to InP/ZnS (2) - TGA. A targeted dual modal OI/MR imaging probe for the early detection of aggressive prostate cancer has been developed, via a two-step surface modification of InP/ZnS (2) - TGA with Endo180 conformational specific antibodies (A5.158, 39.10) and Gd(III) chelate (Gd.5a). Work presented in chapter 2 has formed part of the publication, Clarke, M. T.; Viscomi, N.; Chamberlain, T. W.; Hondow, N.; Adawi, A. M.; Sturge, J.; Erwin, S. C.; Bouillard, J.-S. G.; Tamang, S.; Stasiuk, G. J. Near Unity Quantum Yield in InP Based Quantum Dots through Thermal Diffusion. Commun. Chem. 2019, 2.

Inflammatory Bowel Diseases (IBD) including Crohn's Disease and Ulcerative Colitis are chronic conditions characterised by inflammation of the wall of the gastrointestinal tract. IBD has also been shown to have systemic impacts including on the central nervous system. Traditional models including the animal systems provide only limited information due to a lack of clinical relevance. Microfluidic technology offers a solution, allowing for the creation of human models which better consider the biophysical properties seen within an organ. This study aimed to develop and optimise a dual flow microfluidic device for the study of the gut and blood-brain-barrier (BBB) systems. Devices were designed in-house and consisted of two channels separated by a semi-permeable membrane. A series of iterations of the device were examined for gut-chip studies, with the device refined and optimised to allow a culture of colonic epithelial cells to be maintained for 7 days. Permeability studies and visualisation of ZO-1 expression showed the maintenance of barrier properties during this time. Following optimisation of the gut-chip, the inflammatory effects of bacterial products on epithelial cells were examined. Treatment with bacterial products induced an inflammatory response in the model, however this was lowered in comparison with a static model. Adaption of the device to culture endothelial cells and astrocyte cells in a BBB model was also carried out. Viability tests showed the device could maintain a variety of cell lines for at least 96 h on chip. The gut and BBB-chip were then connected in series, creating a dual model. This platform could maintain a co-culture of epithelial cells within the gut-chip and endothelial cells within a BBB-chip for at least 48 h, showing the potential of the dual flow device to allow for more systemic studies. Preliminary studies were undertaken using a modification of the gut-chip for the maintenance of full thickness gut tissue biopsies for up to 72 h on chip, however morphology of the tissue was not well-preserved. In summary, this study examined the development and optimisation of a dual flow microfluidic device for the study of barrier systems. The final iterations of the device were both robust and reliable and are suitable for investigating a wide variety of physiological and pathological barriers and potentially provide an alternative to existing animal and cellular models.

The signalling arising from tissue factor (TF) has been associated with cellular homeostasis as well as haemostasis and, is particularly pertinent in carcinogenesis and cancer progression. The formation of the TF-FVIIa complex is known to activate protease activated receptor 2 (PAR2) and initiates multiple signalling pathways which can determine the fate of the cell. Among these mechanisms PI3K-Akt is known as a major promoter of cell survival and proliferation. The regulation of Akt is mediated by PTEN, a tumour suppressor that negatively regulates the PI3K-Akt pathway resulting in the loss of downstream proliferative and pro-survival signalling. In this study, seven tumour cell lines from different tissues were activated with PAR 2-activating peptide (SLIGKV), or incubated with a range of concentrations of rec-TF (65- 1300 pg/ml). The outcome on the phosphorylation state and the lipid-phosphatase activity of PTEN, and also Akt activity was assessed. Incubation of cells with rec-TF (65 pg/ml) resulted in de-phosphorylation of PTEN in all the cell lines tested which was concurrent with increased PTEN activity and reduced Akt activity. Moreover, activation of PAR2 reduced PTEN phosphorylation and Akt activity in the majority, but not all of the cell lines tested. However, prolonged exposure of cells to rec-TF over 5 days, resulted in decreased PTEN antigen levels together with enhanced Akt activity and increased cell proliferation in the cell lines tested. To explore the mechanism of regulation of PTEN by PAR2, the association of membrane- associated guanylate kinase-with inverted configuration (MAGI)1-3 proteins with PTEN was assessed by the proximity ligation assay (PLA) and co-IP. The interaction of PTEN with all three MAGI proteins was reduced following PAR2 activation and explains the alterations in PTEN activity and stability. The structure of MAGI proteins contains a guanylate kinase-like (GK) domain, two WW domains and six PDZ domains. This structure allows MAGI to restrain proteins at the cell junction, regulating the activity of the proteins and stabilising cell-cell contacts. In the second part of the study, analysis of a motif within the cytoplasmic domain of TF (ENSPL) indicated the possible interaction with MAGI proteins. PLA and co-IP studies showed that TF may be restrained at the tight junctions through interaction with MAGI-1. Interestingly the activation of PAR2 transiently dissociated TF from MAGI-1. By expressing the individual PDZ domains as tagged-hybrid proteins, the first PDZ domain was identified to be capable of binding TF and competing with MAGI-1. The expression of this hybrid protein also resulted in augmented TF activity, similar to that following PAR2 activation. Analysis of the interactions of the N-terminal region of MAGI-1 with or without PDZ1 further confirmed the role of PDZ1 in the constrainment of TF and regulation of its procoagulant activity. In conclusion, this study has demonstrated that PAR2 activation promotes increased PTEN activity in the short term, allowing for the elimination of severely injured cells. However, prolonged exposure TF reduces the cellular PTEN antigen levels allowing aberrant cell survival and proliferation, through increased Akt activity. This study also demonstrated the interaction of TF with the PDZ1 domain of MAGI-1, suggesting a mechanism for restraining cell-surface TF from contact with the surrounding fluid, and may explain the regulation of both coagulation and signalling mechanisms.

The epigenetic modifications of DNA shape cell fate in development, differentiation, and disease. The existing gold standard sequencing technologies for epigenetic DNA modifications are based on sodium bisulfite, which is a harsh chemical treatment resulting in DNA degradation. A novel bisulfite-free and base-resolution sequencing method, TET Assisted Pyridine-borane Sequencing (TAPS) was developed to detect the most abundant DNA modifications. In comparison to bisulfite sequencing (BS), TAPS relies on mild reactions for the detection of modified bases. From a bioinformatics perspective, sodium bisulfite substantially reduces information content, complicating data processing and the detection of genetic variation. In fact, most existing modification calling tools for bisulfite-treated data do not distinguish between modifications and genetic variants, which results in false positives. A computational tool, asTair, was created to process DNA modification sequencing data. It was designed primarily for handling TAPS sequencing output, but also contains functions that are useful for bisulfite sequencing data analyses. It was shown that TAPS has more even coverage than BS while having a comparable conversion rate over CpGs, and is applicable to low input samples. A Deep Neural Network (DNN) model that detects single-nucleotide variants in TAPS- and BS-converted sequencing data was created to enable sensitive modification and variant calling. The algorithm showed precision and recall above 0.9 for classifying variants, modifications and reference positions. The model outperformed available variant callers for whole-genome sequencing and BS data. Applying such a model on real datasets could improve the accuracy of identifying real DNA modifications masked by genetic variation and errors, as around a sixth of all SNPs could be misclassified as modifications.

Microparticles (MP) are procoagulant due to tissue factor and phospholipid exposure on the surface. MP are tumour-derived and can be a beneficial biomarker of cancer to recognize individuals who are susceptible to venous thrombosis. The aim of the presented work was to develop and validate an in vitro microfluidic system consisting of two distinct microfluidic biochips to enable the investigation of the relationship between tumour MP and endothelial cells in vitro. Firstly, a range of tumour cell lines were assessed for procoagulant activity (PCA) of the cells and also MP released into the media. Pancreatic AsPC-1, human glioma U87 ovarian ES-2 and SKOV-3, were found to have the highest PCA in both cell suspension and cell- free media, while pancreatic MIAPaCa-2 and ovarian A2780 had a lower PCA. Cell lines were then investigated as to whether or not they could form stable spheroids in 3D cell culture U87, AsPC-1 and ES-2 produced the most compact spheroids and had the fastest PCA. In contrast, PANC-1, MIAPaCa-2 and A2780 formed loose shaped spheroids and slower PCA. However, SKOV-3 showed small compact spheroid and slower PCA. Following the application of media flow, ES-2 and U87 were selected and transferred into the developed dual microfluidic biochips model. Labelled MP were quantified via flow cytometry and this showed MP concentration reduced over time suggesting attachment of tumour MP to HUVECs. This reduction in MP was further reflected with a loss of PCA associated with the media. The effect of Doxorubicin on tumour spheroids resulted in an increased PCA of an endothelial cell layer under flow condition. In conclusion, in this study a microfluidic two-chip dynamic model mimics the interstitial fluid flow showed that tumour MPs released from tumour spheroids attach to endothelial cells and potentially could be a mechanism of clot formation in cancer patients.

Acute wound healing involves a tightly regulated cascade of cellular signalling and functional events. Deterioration at any stage of this sophisticated system can lead to healing impairment and chronic, non-healing wounds. Chronic wounds, which are prevalent in the elderly and diabetic, are a global socioeconomic burden and remain a major area of clinical unmet need. Improved understanding of the cellular and molecular aetiology of chronic wounds is essential to develop new therapies. The aim of this work was to explore the role of cellular senescence and the metallome in governing normal and pathological wound repair. Novel data presented in this thesis show increased senescence in both aged and diabetic wounds, while biologically-important metals, such as calcium, were reduced. Transcriptional profiling of wounds strongly linked the transcriptome, metallome and senescence. A direct role for senescence in pathological healing was mechanistically demonstrated in vitro, ex vivo and in vivo. Crucially, pharmacological inhibition of the explicated senescence receptor, Cxcr2, accelerated diabetic wound healing in vivo. Collectively, these data reveal a hitherto unappreciated role for Cxcr2 in mediating cellular senescence during pathological skin repair. Global profiling of the wound metallome highlighted significant changes in wound iron levels during late-stage healing. In vitro studies uncovered a new role for iron in mediating extracellular matrix deposition during wound remodelling, while reduced iron levels in diabetic wounds correlated with impaired ECM deposition. In summary, temporospatial metallome profiling identified multiple defects in metal-linked cellular processes in the pathological wound environment. Taken together, the research platform delivered in this work will provide an unprecedented opportunity to further interrogate transcriptional and functional relationships between cellular senescence and the metallome in wound repair. Indeed, this research may underpin the development of novel, efficacious metal-targeted therapies for chronic healing wounds in the future.

One in two people born after 1960 will develop cancer, with 90% of all cancer deaths arising from metastasis. Conventional 2D in vitro metastasis models do not fully replicate tumour complexity. In vivo models can address tumour complexity, but do not fully represent human tumour biology. Multicellular spheroids are widely used 3D models of cancer. Spheroids contain internal zonal differentiation of oxygen, metabolite, and nutrient gradients, associated with regions of proliferative, quiescent, and necrotic cells. Current in vitro static spheroid methodologies do not recapitulate factors of cellular spread including continuous flow or shear stress. Therefore, better 3D in vitro models to investigate metastasis represent an area of unmet need. This research aims to develop spheroid-on-chip models, providing a novel strategy to investigate cancer spread. The microfluidic devices used in this study feature weirs for spheroid inclusion, and a borosilicate base coverslip for optical clarity. The devices feature an access port allowing direct access to the microwell. Spheroids derived from established cancer cell lines, MCF7 and U-87 MG, were formed off-chip and incorporated into the device by pipetting, before being perfused with complete media at 3 μL min-1 for 72 h. Cell viability was assessed in effluent, using the CytoTox Glo assay, demonstrating spheroid viability is robustly maintained on-chip. In situ analysis of cell viability, through FDA/PI live/dead staining indicated an increased proportion of viable cells and decreased dead cells on-chip compared to off-chip. VEGF (Vascular Endothelial Growth Factor) ELISA showed that VEGF secretion, as evaluated by its presence in conditioned media, was comparable across all testing conditions. The chip model has been further developed to allow spheroids to be embedded in ECM (extracellular matrix)-like matrices. VEGF ELISA and IL-6 ELISA showed that both VEGF and IL-6 secretion, as evaluated by its presence in conditioned media, was comparable across all testing conditions. ELISA also showed that IL-6 and VEGF was increased in the on-chip models within hydrogel conditions. The device has also allowed the direct imaging of spheroids on chip over 72 h. The analysis of U-87 MG spheroids on chip showed that invasion through Matrigel was comparable to the off-chip static models. Whilst migration on chip, through collagen analysis, was increased over the off-chip counterparts. The work shown offers a novel insight into cancer metastasis; on a more replicative model than the current conventional in vitro techniques.

Dendritic cell (DC) activation underpins the patient immune response in multiple cancers, including ovarian cancer. While DCs are detected within the margins of ovarian tumours, DC populations are maintained in an immature state, displaying low levels of co-stimulatory receptor expression and cytokine production, with impaired antigen presentation capacity. Engagement of the DC surface receptor CD40 (CD40R) is required for efficient immune activation. This thesis investigated the potential of a Monovalent Targeted Peptide targeting CD40R (MTP40) to bind the CD40R of a murine DC cell line (tsDC), and whether the interaction could be enhanced by multivalent presentation. This work aimed to demonstrate the suitability of two potential scaffolds for multivalent presentation. The first system investigated the activity of MTP40 presented on streptavidin as a tetramer (Tet40), which improved the binding kinetics of MTP40 (KD 1.32μM) when presented as Tet40 (KD 0.072μM). Binding capacity was confirmed as MTP40 conjugated to streptavidin Dynabeads successfully isolated the CD40R by pull down assay. The second system explored MTP40 attached to the surface of gold nanoparticles through thiol-Au bonds, to produce targeted polyvalent gold nanoparticles (TPN40) presenting up to 2000 peptide units per molecule, which demonstrated binding to CD40R by coomassie stain. Changes in outputs associated with tsDC activation were observed in response to CD40R targeted treatments. Surface expression of CD40R was upregulated in response to Tet40 or TPN40, while TPN40 induced increased expression of activating receptors CD86 and CD54. tsDC upregulated expression of IL-2 and IL-12, with a corresponding down-regulation of IL-10, in response to Tet40 or TPN40. TPN40 treatments also stimulated uptake of the model antigen FITC-Ova by tsDC. After confirming the tsDC cell line induced functional changes in response to CD40R targeted activation, second goal was to determine whether the "activated" tsDCs were able to induce priming of effector T cells in an antigen specific manner. T cell lines significantly upregulated production of IL-2 and IL-12 in response to TPN40, while the cytotoxic T cell line specifically upregulated IFN-. T cell induced cytotoxicity and proliferation were not significantly affected by the addition of ID8 tumour cell lysates, suggesting further optimizations are needed to produce an antigen-specific response. This study provided evidence that multivalent engagement of CD40R is required to induce effective tsDC maturation, highlighting the importance of ligand presentation in immunity. The work also provides a multistep model of immune activation as stimulated tsDCs were capable of inducing changes in T cell lines. This demonstrates proof-of-principle that multivalent activation at the DC level is sufficient to prime an effector T cell response. These conclusions have wider significance in the field of immunology, where they suggest multivalent interactions can produce optimised immune responses. Gold nanoparticles provided a stable, well-tolerated platform for multivalent drug delivery. Future work should investigate the bimodal application of gold nanoparticles for targeted drug delivery with tumour imaging.

Tumour infiltrating lymphocytes (TIL) in head neck squamous cell carcinoma (HNSCC) are enriched in Treg, a finding confirmed in the current study, which is thought to be a contributing factor to immunosuppression and tumour evasion. Soluble factors released by malignant HNSCC cells may contribute to the enrichment of Treg by inducing differentiation from naïve T cells and their migration to the tumour. Using multi-colour flow cytometry, increases in CD4+CD25hi, CD4+CD25+CD39+, CD4+CD25+CD26− and CD4+CD25+FoxP3+ Treg phenotypes were observed in TIL from HNSCC compared to PBMC. Following culture of CD4+CD25− T cells with conditioned medium (CM) collected from dispersed tumour samples, there was an increase in CD39 expression but not FoxP3 compared to control cultures that were cultured in complete growth medium. Furthermore these cells were unable to suppress the proliferation of CD4+CD25− T cells in CFSE assays. Soluble factors in CM from UMSCC cell lines and tumour-derived fibroblasts were unable to induce the expression of any Treg markers following culture with CD4+CD25− T cells. Culture with CM also had no effect on T cell apoptosis, with no significant increase in PI-AnexinV+ of CaspACE-binding in T cells cultured with CM compared to controls. The expression of the chemokine receptors, CXCR3, CCR4, CCR5 and CCR6 was analysed on T cell populations from HNSCC PBMC and TIL and healthy control PBMC using five-colour flow cytometry. Increased proportions of CXCR3+ and CCR5+ T cells were observed in HNSCC TIL compared to HNSCC PBMC but were no different between the patient and healthy control PBMC. CCR4 and CCR6 were expressed on a higher proportion of Treg from HNSCC PBMC compared to healthy controls but no difference was observed on CTL. In TIL the percentage expression of CCR4 and CCR6 were no different in that of HNSCC patient PBMC. Despite these observed differences in receptor expression, soluble factors in tumour dispersed CM was unable to induce T cell chemotaxis. Overall, although limited effects were observed from soluble factors in tumour CM, the different expression of chemokine receptors suggests there may be a role for soluble factors in Treg recruitment. However whether this is responsible for Treg accumulation or general to all T cells is unclear.

Trypanosoma brucei is a medically important protozoan parasite that causes African sleeping sickness in humans. Although disease treatment is possible, it is hindered by the limited availability of effective drugs and the rapid emergence of drug resistance. It is, therefore, important to identify novel drug targets for the development of new, more effective drugs. Mitochondrial carrier family (MCF) proteins transport a wide range of key metabolites across the mitochondrial inner membrane. They are important for the maintenance of key metabolic pathways in all eukaryotic cells. In particular, the MCF proteins that have been implicated in mitochondrial adenosine triphosphate (ATP) import appear to be essential for cell function and survival due to their roles in defence against oxidative stress and in the provision of the ATP required for mitochondrial electron transport and associated ATP production. Based on their important physiological roles, mitochondrial nucleotide transporters are potential novel drug targets. The main aim of this thesis was to identify and functionally characterise the putative mitochondrial nucleotide transporters in T. brucei. Sequence analysis revealed that the T. brucei genome contains 5 MCF proteins: TbMCP1, TbMCP15, TbMCP16, TbMCP20 and TbMCP23. The proteins were predicted to have roles in mitochondrial nucleotide transport based on their homology to functionally characterised MCF proteins that transport nucleotides. Of these TbMCPs, TbMCP1 has the highest sequence similarity to functionally characterised mitochondrial flavin carriers and peroxisomal ATP/adenosine monophosphate carriers in other eukaryotes, such as Flx1 in Saccharomyces cerevisiae. Sequence analysis further predicted that TbMCP15 and TbMCP16 were closely related to mitochondrial adenosine diphosphate (ADP/ATP) transporters, such as AAC2 in S. cerevisiae, whereas TbMCP20 appeared to be closely related to mitochondrial S-adenosylmethionine transporters, such as SAM5 in S. cerevisiae, and TbMCP23 appeared to be closely related to mitochondrial pyrimidine transporters, such as S. cerevisiae RIM2.

Protein interaction networks (PINs) are mathematical representations of interactions between proteins within organisms. Studying their properties can give insights into biological functions and the importance of proteins, and it can therefore aid in drug- discovery. Graphs are the most common mathematical object used to represent PINs. In this thesis, we investigate generalised mathematical representations of PINs. In particular, we examine multilayer networks (MLNs) and node-weighted networks. These mathematical objects allow the construction of temporal PINs and tissue-specific PINs by combining gene-expression data with PINs. We introduce promiscuity as an information-theoretical measure of a node's distribution of neighbours across different layers in MLNs. We examine promiscuity in synthetic networks and tissue-specific PINs and find that the vast majority of proteins are not cell-type specific. Integrating temporal gene-expression data with PINs allows us to create temporal PINs in the form of MLNs. We investigate an eigenvector-based temporal centrality in a temporal PIN of yeast during the cell cycle. We thereby examine the change of proteins' importance over time, which reflects their activity during the cell cycle. We then discuss the detection of community structure in node-weighted networks. For synthetic networks, we show that considering node weights can alter detected community structure. We combine a human PIN with gene-expression data to construct tissue-specific PINs and investigate their community structure. Comparing the detected communities with gene-ontology information, we find some tissue-specific functions of these PINs. Overall, the case studies in this thesis suggest that MLN and node-weighted networks are suitable for the integration of protein-interaction data with other biological data sets.