Biomolecular and Biomedical Science Theses

Cilia are evolutionarily conserved microtubule based organelles extending from the surface of most cells serving important sensory and signalling functions. Defects in cilia cause a variety of disorders with overlapping phenotypes, termed ciliopathies. The ciliary base, acting as a ciliary gate, plays a key role in regulating ciliary protein composition and cilia-related signal transduction pathways, forming membrane and cytosolic diffusion barriers that prevent exchange between ciliary and non-ciliary compartments. The transition zone at the proximal ~1 µm of the axoneme is part of the ciliary base and ciliopathy proteins localising specifically to the transition zone form two functional modules (MKS and NPHP). Together, these modules are required for ciliogenesis, and are implicated in membrane diffusion barrier function in Caenorhabditis elegans cilia. Active transport across the transition zone diffusion barrier is thought to be facilitated by intraflagellar transport. The protein composition of the ciliary base is not fully known and molecular mechanisms underlying the diffusion barriers are poorly understood.The work presented in this thesis is focussed on the ciliary base and the transition zone in C. elegans ciliated sensory neurons. Specifically, Chapter III focusses on characterisation of a candidate ciliary component, K04F10.2. Exhaustive phenotypic analysis indicates that K04F10.2 serves ciliary functions. Consistent with a ciliary role, subtle IFT defects were observed, as was a functional interaction with the ciliopathy protein, Joubert Syndrome-associated ARL-13. The interactome of human KIAA0556 (K04F10.2 ortholog) was identified and subsequent analysis in worms revealed novel ciliary localisations and transport properties for RAB-28 and F47G4.5/KATNBL1.Chapter IV describes the development of a Fluorescence Recovery After Photobleaching (FRAP)-based assay to validate the existence of a transition zone membrane diffusion barrier. The subsequent use of this assay to investigate the molecular mechanisms that establish and maintain this barrier reveals that various MKS- and NPHP-module components are differentially required for diffusion barrier integrity. Additionally, IFT components are required for active transport to overcome the transition zone membrane diffusion barrier. This FRAP assay is the first such described in a multicellular system and allows for in vivo investigation of exchange kinetics across the transition zone membrane diffusion barrier in real time.

AbstractIntroduction: The differentiation of stem cells constitutes a process poorly characterized at the epigenetic level. The contribution to the understanding of this process is extremely valuable in the advance of molecular-based therapies. After introducing the applicability of a new MS- based approach we here presented its potential in the investigation of changes in histone post-translational modifications during cell differentiation. The investigation of the protein epigenetic complexes is the other piece that completes the information about this process. The changes during differentiation of one of these complexes, the Polycomb group, were also attempted in this work.Methods: The sensitivity offered by mass spectrometry is extremely helpful in the identification of histone epigenetic marks. To increase the detection of low abundant species we presented here the combination of the accurate inclusion mass screening approach with the chemical derivatization of histones. After proving the robustness of this combined approach in the two different cell lines, NT2 and HEK293, we used to identify histone modifications during the cell differentiation of NT2 cells. The sensitivity of mass spectrometry was also helpful in the detection of new interactor of Polycomb members during the differentiation; the cell lines used were HMEC and HEK293.Results: The used of the combination of the two technologies presented here increased in some cases by two-fold the detection of histone marks. The application of this combined approach to the NT2 differentiation allowed the classification of 41 histone marks as constant, early or late expressed during this process; 9 of them were novel. The identification of a novel Polycomb interactor, the demethylase NO66, was also achieved.Discussion: The combined approach proved to identify an almost similar number of histone marks that more sophisticated methodologies. It also associated 9 new histone marks with not reported function to date with the differentiation of NT2 cells. Future work employing semisynthetic nucleosomes is proposed to investigate the role of these marks. The discovery of the interaction of NO66 with Polycomb led to further investigation that concluded in the publication of biological model of its role within the cell.

Fragment complementation between EF1 and EF2, calcium binding protein subdomains of the protein calbindin D9k, forms the basis behind a novel affinity protein purification system enabling rapid, highly specific capture and elution of pure EF1-tagged protein in physiological conditions. Construction of an affinity resin for this purification system relies on efficient production of the EF2 peptide as the affinity ligand to be displayed on a nanoscale scaffold.In this thesis a protocol was developed for production of the EF2 peptide via a recombinant fusion protein construct incorporating the EF2 peptide with an added terminal GGC. Isolation of pure EF2-GGC peptide from this fusion protein followed. This protocol involved expression of this recombinant protein through the bacterial cell host, Escherichia coli (E.coli), and purification via native cell lysis and immobilized metal affinity chromatography (IMAC), both optimized for yield and purity. Subsequent processing steps of this recombinant fusion protein consisted of TEV protease cleavage and additional affinity purification steps to obtain the peptide in a pure formulation for further steps in affinity resin manufacture.Initial protein expression investigations demonstrated that the EF2-GGC recombinant fusion protein was soluble and expressed well in both autoinduction and isopropyl β-D-1-thiogalactopyranoside (IPTG) induction systems. Ni-NTA IMAC purification of this recombinant fusion protein was successful with a yield of approximately ~50 mg/L. Proteolytic cleavage of the EF2-GGC recombinant fusion protein successfully releasing the EF2-GGC peptide of interest was optimised. A second round of Ni-NTA IMAC, ion exchange chromatography (IEX) and a final polishing step of size exclusion chromatography (SEC) retrieved the peptide in a highly pure form, suitable for subsequent resin coupling steps.While this protocol was optimised to provide 6 - 12 mg/L of recombinant EF2-GGC peptide, SEC analysis confirmed that the peptide was prone to significant aggregation. Further investigations into the peptide buffer formulation failed to completely reverse aggregation. Therefore, while the expression and purification protocols developed during this project have been effective in isolating a pure EF2-GGC peptide, further work to identify an effective formulation for the purified peptide is required.

The single use plastic polyethylene terephthalate (PET) has greatly contributed to the convenience of our modern lifestyle. However, its widespread use in single use plastic packaging and synthetic fibres, and recalcitrant properties are a major contributor to plastic pollution. Meanwhile, the microbially produced polyesters, polyhydroxyalkanoate (PHA) have received increased attention as a possible solution to the accumulation of plastic waste due to their biodegradable properties. The upcycling of PET waste into a material such as PHA creates an opportunity to improve both resource efficiency and contribute to a circular economy. This is possible as intermediates in the chemical recycling of PET, disodium terephthalate (Na2TA) and ethylene glycol (EG) can be harnessed as cheap microbial substrates for the cultivation of Pseudomonas umsongensis GO16 and polyhydroxyalkanoate (PHA) production. Previous studies have successfully employed Na2TA as a microbial substrate for growth and PHA accumulation. However, issues surrounding its feasibility as a microbial feedstock, such as solubility have not been tackled. This thesis presents the model-based optimisation of a liquid feeding regime with the purpose of improving the automation and operational ease of Na2TA conversion into P. umsongensis GO16 biomass. The model was parameterised and validated using data from dynamic liquid-phase and solid-phase feeding experiments. The validated model identified sodium ion accumulation as a key determinant of bioprocess performance and was used to design a liquid-phase feeding strategy that maximises P. umsongensis GO16 biomass synthesis. The obtained biomass concentrations of 10.5 g/L (liquid-phase feeding) and 15.3 g/L (solid-phase feeding) are the highest ever reported using Na2TA as a sole carbon source for microbial growth. However, even though a solid pulse feeding regime of Na2TA was successful in maximising biomass, it was found to limit PHA accumulation, achieving a 5-fold lower PHA content compared to previously achieved yields using Na2TA as the sole source of carbon and energy. Secondly, this thesis demonstrates a completely biotechnological process for upcycling of PET into PHA. PET was enzymatically hydrolysed to provide Na2TA and EG. Enzymatically hydrolysed PET was then supplied as a sole source of carbon and energy to P. umsongensis GO16. Using nitrogen limiting conditions to stimulate PHA accumulation a biomass of 1.6 g/L was achieved in 24 h with 7% of that biomass representing medium chain length PHA, which was equal to that achieved when a synthetic mixture of both monomers mimicking enzymatically hydrolysed PET was employed. Furthermore, adaptive laboratory evolution carried out on P. umsongensis GO16 was successful in improving the biomass accumulation and total PHA content of the strain when cultivated on enzymatically hydrolysed PET. Recently the genome sequence of P. umsongensis GO16 revealed that it is equipped with the genes required for both medium chain length (mcl) and short chain length (scl) PHA production. The deletion of PHA depolymerases have been known to improve PHA productivity, although its impact is not consistent across different bacterial strains and substrates employed. To date no study has investigated the impact of a PHA depolymerase gene knockout on a scl and mclPHA producer. Accordingly, the final major finding of this study was that the deletion of both scl and mclPHA depolymerases have varying effects on growth and PHA accumulation depending on the depolymerase knocked out and the substrate used. Deleting both scl and/or mclPHA depolymerases significantly impairs growth on sodium octanoate, while the deletion of the sclPHA depolymerase negatively impacted PHA accumulation in P. umsongensis GO16 while also affecting the monomer composition of the polymer with an over 4-fold increase in C4 monomer fraction and over 1.7-fold decrease in PHA content.

Mycobacterium abscessus subspecies abscessus is a highly drug resistant mycobacteria and the most common respiratory pathogen among the rapidly growing non-tuberculous mycobacteria. We report here the first multi-omics approach to characterize the primary transcriptome, coding potential and potential regulatory regions of the Mycobacterium abscessus genome utilizing RNA-seq, dRNA-seq, ribosome profiling and LC-MS proteomics. In addition, we attempt to address the genome’s contribution to the molecular systems that underlie Mycobacterium abscessus’ adaptation and persistence in the human host through an examination of Mycobacterium abscessus' transcriptional responses to a number of clinically relevant conditions. These include hypoxia, exposure to sub-inhibitory concentrations of antibiotics and growth in an artificial sputum designed to mimic the conditions within the cystic fibrosis lung. To computationally infer the gene regulatory network for Mycobacterium abscessus we propose a novel statistical computational modelling approach: BayesIan gene regulatory Networks inferreD via gene Expression and compaRative genomics (BINDER). In tandem with derived experimental expression data, the property of genomic conservation is exploited to probabilistically infer a gene regulatory network in Mycobacterium abscessus. In particular, inference on regulatory interactions is conducted by combining ‘primary data’ from RNA-seq experiments derived from Mycobacterium abscessus and ‘auxiliary’ ChIP-seq data from the related Mycobacterium tuberculosis. The inferred relationships provide insight to regulon groupings in Mycobacterium abscessus. We construct an inter-conditional snapshot of the transcriptional landscape in Mycobacterium abscessus across a range of stress-inducing conditions comprising exposure to antimicrobial compounds as well as nutrient starvation and iron depletion. The research herein provides valuable elucidation on the transcriptional means through which Mycobacterium abscessus persists in hostile environments and mediates virulence in the human host.

Thrombosis remains a major cause of mortality and morbidity worldwide. Common risk factors include pro-inflammatory conditions, endothelial dysfunction, and aberrant platelet activation, all augmenting the patient’s risk of developing a thrombus. Intriguingly, these risk factors also manifest in increased levels of circulating extracellular vesicles (EVs). EVs are a heterogenous group of membrane vesicles released from all cells. Due to the expression of tissue factor and phospholipids on their surface, EVs can accelerate thrombus formation. To circumvent adverse events in thrombotic disease, pharmacological thromboprophylaxis is frequently indicated. Rivaroxaban, a direct FXa inhibitor, mediates anti-inflammatory and cardiovascular-protective effects besides its well-established anticoagulant properties, however, these remain poorly characterized. Given the prevalent role of EVs in thrombotic diseases, we hypothesized that Rivaroxaban’s anti-inflammatory properties are reflected upon differential molecular profiles of circulating EVs. Here, we used single vesicle analysis and comparative proteomics to, for the first time, characterise small (<200nm) and large (200-1000nm) plasma EVs from patients with non-valvular atrial fibrillation (AF), venous thromboembolism (VTE) and stable cardiovascular disease (CVD) treated with Rivaroxaban compared with cohort-specific controls. We identified profound changes in circulating EV profiles and proteomic signatures that may contribute to Rivaroxaban’s pleiotropic effects. Circulating EV profiles were fundamentally altered across the cohorts analysed. While AF patients displayed profoundly reduced levels of total small EVs, we identified a decrease in the proportion of large EVs between 400-700 nm in VTE patients. Patients with stable CVD, on the other hand, indicated increased small EV mode size with a concomitant increase in the proportion of large EVs between 600-800 nm. Comparative proteomic characterisation of enriched EV fractions revealed differential expression of proteins involved in the regulation of complement activation, vascular integrity, and inflammation, collectively mirroring the reported anti-inflammatory and cardioprotective characteristics associated with Rivaroxaban therapy. In summary, we have determined that circulating EV profiles are powerful surrogate markers of the Rivaroxaban-mediated pleiotropic effects. Regardless of the underlying thrombotic disease, Rivaroxaban appears to confer vascular protective effects in both venous and arterial thrombotic conditions. While several mechanisms, including inhibition of protease activated receptor or NF-¿B signalling, have been proposed to contribute to these vascular protective characteristics mediated by Rivaroxaban, we here provide original evidence that the modulation of circulating EV signatures may pose a distinct mechanism through which Rivaroxaban therapy may elicit such pleiotropic effects, an exciting finding that may be leveraged when planning future studies exploring personalized management strategies, such as identification of those patients most likely to benefit from this therapy, particularly in jurisdictions where access to therapy may be restricted. Alleviating the underlying pro-inflammatory state of patients with thrombosis poses a promising future therapeutic target to avert major adverse thrombotic events. Further investigations into Rivaroxaban-mediated changes in circulating EV signatures may provide a future avenue to open Rivaroxaban’s use for extended patient cohorts, such as rheumatoid arthritis.

Rhodococcus equi is an intracellular pathogen of macrophages, which often causes life-threatening foal pneumonia. The intracellular replication of R. equi is dependent on a virulence plasmid containing a pathogenicity island (PAI). The vapA, virR, virS genes located in the PAI are essential and sufficient for intracellular growth of R. equi. VapA is the key virulence factor, which functions as a pH-neutralizing factor during phagocytosis. The LysR-type transcriptional regulator VirR and the two-component system orphan response regulator VirS are required for transcription of vapA and mediate crosstalk with the R. equi chromosome for host-specific niche adaptation. VirR and VirS are products of the virR operon. Transcription of the virR operon is driven by the PvirR promoter located upstream of virR and by an inducible promoter Porf5 located within the virR gene which is regulated by both temperature and pH in a synergistic manner. Temperature is the primary environmental signal activating Porf5. During growth at high temperatures, Porf5 is more actively responding to pH signals. Although VirS was considered as a promising candidate involved in pH regulation, deletion of virS does not affect pH regulated activity of Porf5. Surprisingly, a frameshift mutation in virR abolished pH regulation of Porf5, which could be restored by complementation with a trans-acting VirR. This demonstrated that the effect of the virR frameshift mutation on Porf5 activity was not due to disruption of Porf5 but by the inactivation of the virR gene. VirR is therefore required for pH regulation of Porf5 and hence expression of virS. Transcriptome analysis of different R. equi strains grown at high and low pH revealed that R. equi has at least two pH regulons: a general pH regulon and a VirR-mediated virulence-associated pH regulon. The former contains 423 genes involved in transportation, nitrogen metabolism and amino acid metabolism. The latter only contains two chromosomal genes (REQ01110 and REQ08750), both of which are homologues of icgA, a gene located immediately downstream of virR and encoding major facilitator superfamily transporters. Considering the high similarity between icgA, REQ01110, and REQ08750, and the fact that they are all regulated by pH in a VirR-dependent manner, they potentially play a role in VirR-mediated pH regulation. Mutation of a conserved aspartate residue (D57A) abolished VirS-dependent transcription of vapA, suggesting phosphorylation is required for activation of VirS. The phosphorylation of VirS is highly likely mediated by a sensor kinase. Sensor kinases including VirX, VirY, MprB, REQ45230, and KdpD were chosen as candidates involved in the activation of VirS. However, multiple deletions of these sensor kinase genes did not affect VirS-dependent gene transcription. The result suggests three possible scenarios of activation of VirS: 1. The sensor kinase responsible for the activation of VirS is not chosen in this study, which requires further investigation. 2. VirS is phosphorylated in a sensor kinase-independent manner, potentially, by a low molecular weight phosphodonor. 3. VirS is a stand-alone response regulator whose activity is independent of phosphorylation. KdpD/KdpE is the most upregulated two-component system in the presence of VirR and VirS during the growth under VapA inducing growth conditions, and therefore may play a role in pathogenicity. The KdpD/KdpE in R. equi regulates the transcription of the high-affinity potassium transporting system KdpFABC. Deletion of kdpD leads to increased transcription of the kdp genes, suggesting KdpD is a transcriptional repressor of the Kdp system. However, the deletion of kdpD did not affect the intracellular replication of R. equi in the observed period. The result suggests that kdpD is not required for intracellular growth, which may be due to the increased expression of the kdpFABC genes in the absence of KdpD.

Uveal melanoma (UM) is a rare, ocular cancer that arises from melanocytes within the uveal tract. This cancer imposes the threat of visual impairment, ocular pain, enucleation, and in half of all cases, death from metastatic disease. Approximately 50% of UM patients will develop metastases, which occur most frequently in the liver. Despite advances in control of the primary tumour, there is currently no approved therapy that can prevent or halt the growth of UM metastases. The prognosis for metastatic UM patients is extremely poor; the median overall survival is approximately 13.4 months, with as few as 8% of patients surviving beyond 2 years. Recent reports suggest that Ireland has one of the highest incidence rates in the world, with an average of 45 new cases diagnosed annually. The development of therapies that can prolong the life of UM patients is an area of urgent unmet need. This research evaluates the clinical relevance and therapeutic potential of cysteinyl leukotriene receptors (CysLT1 and CysLT2) in UM. The cysteinyl leukotrienes (CysLTs) are a group of inflammatory, lipid mediators that signal through G-protein couple receptors, CysLT1 and CysLT2. The role of CysLT receptor expression, and signalling, in several cancers has recently emerged. This, coupled with the identification of an oncogenic mutation in CYSLTR2 in a subset of UM patients, led us to hypothesise that these receptors could be targeted therapeutically in the disease. We have shown that high expression of CYSLTR1 and CYSLTR2 are significantly associated with reduced disease-free survival and reduced overall survival in UM patients. The expression of both receptors was further investigated by IHC. In two, independent patient cohorts we validated that high expression of CysLT1 is significantly associated with reduced overall survival. These findings solidified the importance of CysLT receptor expression in UM and its link to patient outcomes. As CysLT receptors are druggable targets, we hypothesised that pharmacological antagonists may attenuate cancer phenotypes including viability, proliferation, angiogenesis, inflammation, and metabolism, of UM cells in culture. We found that CysLT1 antagonists, but not CysLT2 antagonist, HAMI 3379, produced significant anti-cancer effects in primary and metastatic UM cell lines through the inhibition of cell survival and cell proliferation. Novel CysLT1 antagonists, quininib and 1,4-dihydroxy quininib, produce cell line-dependent effects on the cancer secretome of UM cell lines. In terms of metabolism, CysLT1 antagonists significantly reduce oxidative phosphorylation in primary and metastatic UM cells. We sought to validate our findings in in vivo and ex vivo preclinical models of UM. In zebrafish cell line-derived xenograft models, CysLT1 antagonists significantly inhibit the growth of primary and metastatic cell lines in vivo and have a greater effect in zebrafish ocular orthoxenograft models. In a cell line-derived orthotopic rodent xenograft model of metastatic UM, treatment with CysLT1 antagonist, 1,4-dihydroxy quininib, did not significantly decrease tumour weight versus vehicle control but did decrease tumour weight and expression of Ki-67, a marker of proliferation, versus standard-of-care, dacarbazine. 1,4-dihydroxy quininib significantly decreases ATP5B, a marker of oxidative phosphorylation, versus vehicle, mimicking our in vitro data. Treatment of UM ex vivo explants derived from primary UM with 1,4-dihydroxy quininib significantly alters the secretion of inflammatory mediators in the tumour microenvironment. The secretion of IL-13, IL-2 and TNF-a was significantly increased following treatment of primary UM tumours for 72 hours. These preclinical data strengthen the importance of CysLT signalling in UM. Our findings suggest that high expression of CysLT1 in UM could act as a biomarker and that antagonism of CysLT1 may be of therapeutic interest in the treatment of UM.

Cancer is a major cause of morbidity and mortality globally. There were 23.6 million cancer cases and 10 million cancer deaths globally in 2019 and cases are estimated to reach 28.5 million in 2040. Around 30-50 % of cancer cases can be prevented by avoiding risk factors and implementation of prevention strategies. Exposure to chemical carcinogens can increase a person’s risk of developing cancer in their lifetime. Chemical carcinogens are substances which directly induce malignant tumour formation, increase tumour incidence, or decrease the time taken for a tumour to form at an increased rate compared to background. Carcinogens Aristolochic acid (ARAI) and Ochratoxin A (OTA) are major contaminants of herbal medicines and foodstuffs globally. The kidneys are one of the principal organs affected by carcinogens due to their function in maintaining tissue homeostasis. Renal cell carcinoma arises from the epithelium of the nephrons of the kidney which are disproportionately affected by chemical carcinogens due to their role in filtering plasma for removal of waste products and xenobiotics. In most eukaryotic organisms, oxygen is essential for normal cellular functions. During cellular respiration and metabolism, by-products such as reactive nitrogen species and reactive oxygen species (ROS) are produced. Antioxidants remove free radicals and help restore cellular balance. Oxidative stress occurs when excess free radicals are produced in cells which overwhelms the normal antioxidant capacity. This can be caused by both intrinsic (e.g., metabolism) and extrinsic (e.g., xenobiotics) factors. Excess ROS can damage DNA, proteins and lipids and is associated with many pathophysiological processes including carcinogenesis. The Nuclear Factor Erythroid 2-related factor (Nrf2) pathway is an intrinsic, inducible stress mitigation response which regulates antioxidant response genes to reduce oxidative stress. Bardoxolone methyl (CDDO-Me), a synthetic triterpenoid induces Nrf2 through inhibition of the endogenous repressor Keap-1. Several human carcinogens elicit tumourigenic effects by modulating associated oxidative stress pathways including the Nrf2 pathway. For this research the normal (RPTEC/TERT1) cell line was exposed to carcinogens ARAI and OTA, pre-treated with CDDO-Me. Several effects associated with cellular transformation were characterised including loss of junctional protein expression and barrier function, lipid peroxidation and effects on the primary cilium. Pre-treatment with CDDO-Me affected these endpoints and elicited a spectrum of effects dependent on cell type and the carcinogen in question. To further characterise and contextualise the effects observed, whole cell proteome analysis and systems biology modelling was performed. In summary, the current research suggests that the beneficial effects of Nrf2 induction are highly dependent on the nature of the carcinogenic insult applied. In some cases, Nrf2 enhancement abrogated some of the deleterious effects whilst in others, it appeared to potentiate the cellular damage caused.

Retinal degeneration is the leading cause of blindness in the industrialised world1 and is characterised by progressive loss of the light sensing cells, photoreceptors, in the retina. Retinal degeneration occurs in both inherited retinal degenerations (IRD) and age-related macular degeneration (AMD). Limited therapies are available for both conditions and there is a pressing need to uncover novel therapies to rescue/preserve vision. With the advent of computational technology and new screening techniques, novel compounds restoring vision can be uncovered by combining computational and phenotypic drug discovery methodologies. Here, I present two complementary workflows to identify compounds rescuing vision. Firstly, ligand-based virtual screening was used to uncover 3D analogues of 7,8-DHF using Cresset Ltd, Blaze’s software. The second workflow utilised orthogonal pooling to screen 720 compounds from the Chembridge DIVERSet™ compound library for hit compounds. Screening for both workflows was conducted using the optokinetic response assay in the atp6voe1-/- zebrafish model of inherited blindness. Three hit compounds were discovered to rescue vision. Compounds UCD-OPGG-A2 and UCD-OPGG-B15 were identified as 3D analogues of 7,8-DHF restoring vision. Compound UCD-OPGG-3E was uncovered during the randomised library screen. RT-qPCR of 3E and A2 did not confirm alterations in inflammatory or oxidative stress related genes in atp6voe1-/- after drug treatment. Light microscopy analysis suggests 3E may reduce cell death within the ciliary marginal zone of atp6voe1-/- larvae. Tolerability studies performed in collaboration with Experimentica Ltd, indicate that intravitreal injections of A2 and 3E are tolerated in mice. In conclusion, I helped develop two complementary workflows to efficiently detect compounds rescuing vision. Although additional experimentation is needed, compounds A2 and 3E are promising starting points for the discovery of novel compounds to restore vision.

Dairy side streams, namely, whey permeate (WP) and delactosed permeate (DLP), are a key challenge for dairy processing. Excess WP and DLP volumes can be a sustainable bottleneck for the expansion of dairy industries and finding a sustainable disposal route with environmental and economic incentive for these dairy side streams can be challenging. This thesis explores a second-generation (2G) biotechnological process for the bioconversion of dairy side streams into bio-based L-lactic acid. WP and DLP are supersaturated in fermentable lactose that can be converted biochemically into biodegradable bio-based monomers such as lactic acid (LA). The bioconversion of whey side streams into LA and subsequent polymerization into biodegradable polymers such as polylactic acid creates new value chains in the dairy industry but also the opportunity to reduce the environmental and economic burden associated with WP and DLP processing. However, WP and to a greater extent, DLP, have a high mineral content which can affect microbial fermentation performance and increase the cost and complexity of the downstream purification process, thus, requires a pre-treatment demineralisation step. At pilot-scale, the pre-treatment of dairy side streams into a suitable fermentation substrate was investigated using two different pre-treatment methods, NaOH and Ca(OH)2 pre-treatment, applying different dairy side stream feedstock types. Furthermore, microbial L-LA fermentation performance was then investigated at pilot-scale (100 L and 3000 L) using both NaOH and Ca(OH)2 pre-treated dairy side streams as the fermentation substrate. Analytical data was compiled and assessed after completing all stages of the 2G LA biotechnological process, namely, pre-treatment demineralisation, microbial L-LA fermentation, and the downstream purification of the lactate salt into high-quality L-LA. The generated data was then used to develop a techno-economic assessment to validate if the valorisation of dairy side streams into LA was economically feasible at industrial scale production. To sum up, this thesis underpins a feasible bioprocessing valorisation route for WP and DLP to produce L-LA. A 2G LA valorisation platform solves the issue of finding a sustainable solution for excess dairy side stream volumes, creates a new value chain, and supports the transition of dairy industries into a circular dairy bioeconomy.

Polyhydroxyalkanoates (PHAs) are biodegradable polymers which can be used a starting material for plastic, offering an alternative to petrochemical based plastic. PHA is produced by many bacteria as a form of carbon storage and is stored in a granule composed of a polymer core, surrounded by a layer of proteins known as granule-associated proteins (GAPs). A better understanding of PHA production is critical for their use as a bioplastic. In this thesis, modern proteomic approaches were used to investigate PHA production in Pseudomonas putida KT2440. A combination of three approaches were used to study PHA production: expression proteomics, subcellular fractionation proteomics and protein interaction studies. While many studies have focused on specific enzymes involved in the production of PHA, the changes that occur on a whole proteome level are less understood. Therefore, we firstly implemented a liquid chromatography mass spectrometry (LCMS) approach to gain an insight into proteome status at various timepoints in both PHA accumulating and PHA non-accumulating conditions, using glucose as a carbon source. Using this approach, 52% of the theoretical P.putida proteome was successfully analysed. Statistical analysis revealed that proteins mapping to key pathways such as nitrogen metabolism and ABC transporters showed coordinated expression changes. Proteins involved in the PHA production pathway were also upregulated in response to PHA accumulating conditions, including PhaI and PhaF, which play an important role in PHA granule formation and segregation. Secondly, subcellular fractionation proteomics was employed to isolate pure PHA granules from P.putida and the resulting preparation analysed by LCMS to determine the protein composition. Over 1,000 proteins were identified, requiring data analysis and validation steps to prioritise the most likely GAPs. All 6 proteins expressed from the PHA operon were identified in the granule proteome. Other proteins were evaluated based on factors including a) enrichment on the granule b) up-regulation under conditions known to induce granule formation c) reports from the scientific literature d) biophysical factors predicted using bioinformatics. As well as confirming previous descriptions of the PHA granule proteome, my work expanded the set of potential granule proteins, including 87 proteins that were enriched on the PHA granule. Thirdly, protein interaction studies were carried, focused on proteins involved in PHA synthesis. Few protein interaction studies have been carried out on PHA-related proteins, and little is known about how these proteins interact with each other. I implemented a fusion protein strategy, fusing enhanced yellow fluorescent protein (eYFP) to the C-terminal of five proteins that were known to be involved in PHA production: Crc global regulator, PhaI, PhaD PhaZ and PhaC1. Fusion proteins were subjected to affinity purification mass spectrometry (APMS) to identify potential interactors. I constructed a database of the results, supplemented with information describing the bait and prey proteins. Only one previously identified interaction between PHA pathway components was observed, between PhaI and PhaF. 79 potential interactions were observed with significance and evaluated using additional criteria. 3 proteins identified were selected for further validation studies: pyruvate dehydrogenase subunit E1 (Q88QZ5), putative lipoprotein (Q88F99) and peptidoglycan-associated lipoprotein (P0A138). Reverse pulldown and microscopy studies were used to validate protein interactions and confirm subcellular localisation. We confirmed the interaction of P0A138 and PhaF and the localisation of P0A138 to the PHA granule in vivo. We also confirmed the interaction between Q88QZ5 and PhaF and that Q88QZ5 localises in vivo to the PHA granule. This finding reveals a potential link between PHA granule production during stress conditions and central metabolism.

Multiple sclerosis (MS) is a progressive neurodegenerative disorder which occurs when autoreactive T-lymphocytes infiltrate the central nervous system (CNS) and damage the oligodendrocytes responsible for maintaining the myelin sheath. Initially, the damage is repaired through the activation of a repair process called remyelination. However, as the disease progresses, remyelination begins to fail leaving the denuded axons vulnerable to damage and subsequent degeneration. The accumulating loss of chronically demyelinated axons causes a steady decline in neuronal activity, resulting in progressive disability. Current therapies target the immune component of MS but do not address the remyelination deficits underlying the disease progression. Thus, there is a need for therapies which actively encourage remyelination to improve treatment outcomes. Our group has found that the nootropic nefiracetam accelerates remyelination in vitro and in vivo. Moreover, in the experimental autoimmune encephalomyelitis (EAE) model, the gold standard model of MS, nefiracetam reduces white matter lesions in the spinal cord and, when used together with an immunosuppressant, restores normal motor function. In vitro studies with oligodendrocyte precursor cells (OPC) found that nefiracetam modulates key phases of remyelination, including OPC migration. Microarray analysis of corpus callosal tissue taken from the in vivo cuprizone model of demyelination highlighted several genes and pathways which may be central to nefiracetam’s effect on remyelination. In particular, the data implicated the regulation of glutamate signalling. Indeed, calcium-imaging studies with OPCs found that nefiracetam modulates glutamate receptor signalling in these cells. The aim of this thesis was to further elucidate the mechanism underlying nefiracetam-mediated acceleration of remyelination. We provide further evidence for the modulation of glutamate pathways by nefiracetam in OPCs. Notably, we found that the regulation of glutamate signalling is linked to its ability to accelerate OPC migration. Furthermore, RNA sequencing analysis of the cortex of cuprizone mice revealed additional pathways regulated by nefiracetam which may be relevant to its mechanism of action, including those related to immune cell migration, inflammation and extracellular matrix organisation. We found that nefiracetam directly modulates monocyte, macrophage and microglial signalling. To date, our in vitro studies exploring the mechanism of action of nefiracetam have predominantly focused on OPCs. As such, another aim was to explore the effect of nefiracetam on cells other than OPCs. We found that nefiracetam reduces the secretion of the pro-inflammatory cytokine TNF-a from macrophages and microglia. Nefiracetam also downregulates the expression of CD14, a toll-like receptor co-receptor that mediates pro-inflammatory signalling, in monocytes and macrophages. Furthermore, nefiracetam alters the expression of several matrix metalloproteinases known to be dysregulated in MS in macrophages. Overall, these results suggest that nefiracetam may also alter immune cell signalling within lesions to modulate remyelination. Finally, we identified two serum-based proteins which could be developed as a companion diagnostic for monitoring nefiracetam’s therapeutic effect. A barrier to the development of remyelination therapies is the lack of biomarkers which are specific and sensitive for myelin repair. The identification of biomarkers which reflect changes in myelin would assist in assessing the therapeutic efficacy of nefiracetam in subsequent clinical trials. Analysis of the serum of EAE mice detected changes in ceruloplasmin and Ig kappa chain V-II region 17S29.1 in response to nefiracetam. Interestingly, these proteins were dysregulated by the model before being normalised by nefiracetam; thus, they warrant further investigation as biomarkers of remyelination and/or nefiracetam’s therapeutic effect.

Inherited retinal diseases constitute a rare and heterogeneous group of eye disorders affecting more than four million people worldwide, characterised by progressive dysfunction of the retina and degeneration of retinal cells causing subsequent vision loss. They collectively represent the leading cause of blindness in the working-age population with varying inheritance patterns, clinical phenotypes and ages of onset, and treatment options for patients are limited. The causative factor is one or more pathogenic variants in more than 280 retinal genes with diverse functions in visual perception. In the retina at the back of the eye, the photoreceptor cells (rods and cones) contain outer segments; specialized and modified cilia which are imperative to the visual process of phototransduction. Their importance in vision necessitates their regular renewal to maintain their structural health and function. As a result of disruptions to genes involved in the structure, function or renewal of outer segments, there is progressive dysfunction of photoreceptors causing visual impairment. This thesis aims to phenotypically characterise a novel zebrafish model of inherited blindness, known as the raifteirí (raf) model which was first identified in an N-ethyl-N-nitrosourea mutagenesis screen. Mutant zebrafish harbour a single point mutation in the emc1 gene, the largest subunit of the ubiquitously expressed endoplasmic reticulum membrane protein complex which functions in protein folding and synthesis. This includes crucial retinal proteins which are synthesised and trafficked through photoreceptors for localisation in outer segments to function in vision-associated processes. Previous genetic studies have implicated human EMC1 variants in numerous blindness phenotypes, but there is a need for additional studies employing animal research models to uncover the disease-associated cellular and molecular mechanisms.