Antibody-Directed Enzyme Prodrug Therapy (ADEPT) is an experimental cancer treatment. First, an antibody-enzyme is targeted to the tumour. After clearance from healthy tissue, a prodrug is administered and activated by the enzyme. MFE-CP, an anti-carcinoembryonic antibody fragment fused to the enzyme carboxypeptidase G2, has been used for ADEPT in combination with a nitrogen mustard prodrug. Clinical trials are encouraging but highlight major challenges of sub-optimal MFE-CP pharmacokinetics and drug resistance. This thesis explores means to address these challenges. MFE-CP, manufactured in P. pastoris, clears rapidly from the circulation due to yeast mannosylation. This leads to excellent tumour:blood ratios but decreased opportunity for tumour uptake. MFE-CP was therefore mutated in an attempt to reduce glycosylation. Initially, N-linked glycosylated asparagine residues were mutated to glutamine. The enzyme remained active but cleared rapidly in vivo. O-linked residue mutations were then explored. Changes that would least impede enzyme function were predicted using bioinformatics and a series of mutated constructs generated. The T55V mutation generated a functional enzyme that also cleared rapidly in vivo. The DNA damage response was investigated as a mechanism of drug resistance. Using the comet assay, DNA interstrand cross-links were shown to form rapidly in carcinoma cells and xenografts in response to ADEPT, but these were unhooked over 48 hours. The γ-H2AX and RAD51 response indicated unhooking was due to DNA damage repair. Cell cycle studies showed that ADEPT treatment also led to G2/M arrest. G2/M arrest allowed DNA repair to occur and it was hypothesised that ADEPT could be made more effective by blocking arrest and driving entry into mitosis. To test this hypothesis, Chk1 inhibitors, UCN-01 and PF-477736, were investigated. Results demonstrated an enhanced anticancer effect with the ADEPT and PF-477736 combination, whereby increased cell death was observed at 48 hours post treatment. The approach has potential for clinical translation.

Tissue engineering is the regeneration of tissues through the use of cells with the aid of scaffolds. Such in vitro fabrication of tissues is beginning to become an option for the management of patients who have irreversible failure of an organ. There are many technical challenges to overcome before 'off-the-shelf' tissues that represent the translation of scientific discoveries into routine treatments (or millions of patients becomes a reality. To meet this demand, the products must be manufactured reproducibly and in large numbers employing current good tissue practice (cGTP). This thesis takes a whole bioprocess approach and includes: Bioreador design A new and novel methodology is described which potentially allows any tubular organ to be fabricated using an easily scalable liquid based technique suited to automated mass- production. It is novel in that it employs a single multitasking chamber which functions as both a "dynamic former" to create a precise tubular cell/polymer construct and then acts as the core of the bioreactor in which to grow the tissue. Optical coherence tomography (OCT) OCT is an advanced non-invasive high-resolution clinical imaging technique. This thesis describes both its initial application to the engineering of tissue as a means of imaging a construct within its individual bioreactor and also the use of Doppler enhanced OCT to qualitatively and quantitatively image the flow of culture medium through the construct. Polymer scaffold Typically, tissue engineers construct a scaffold and then seed it with cells. This has inherent difficulties including the achievement of homogeneous seeding densities. An alternative is described whereby cells and liquid polymer are mixed prior to scaffold formation. Stem cells Because of the novel approach to construct fabrication, rat smooth muscle cells were used for "range-finding" experiments. Later experiments deployed human adult mesenchymal stem cells (MSCs) which were shown to network within the tubular constructs.

Failure modalities of bone anchoring and skin penetrating devices, such as Intraosseous Transcutaneous Amputation Prosthesis (ITAP), stem from the lack of dermal cell and tissue attachment at the skin implant interface. As a result the implant shaft offers a direct route for harmful pathogens into the body. The aim of my thesis is to develop techniques that will enhance dermal cell and tissue attachment, in order to create a tight biological seal of implant biomaterials currently used for ITAP. I compared the effects of fibronectin against its cell binding region- a 3-amino- acid region know as Arginine-Glycine-Aspartic Acid (RGD)- on human dermal fibroblast (HDF) attachment to ITAP substrates. Silanization of RGD- polypeptides provided a durable biological functionalisation technique that significantly increased HDF attachment compared with controls. No significant difference was observed between fibronectin and RGD-polypeptide functionalised substrates in vitro. I then investigated the effects of porosity (as a function of pore and strut diameter) on soft fibrosis tissue in-growth and vascularisation. Electron Beam Manufacturing (EBM) provided an effective method of creating precise 3- dimensional porous structures. Porosity as a function of pore and strut diameter led to optimal soft fibrosis tissue in-growth with 700μm pore and 300μm strut diameters exhibiting significantly increased vascularisation over other implants groups after 4 weeks in vivo. The development of these functionalisation techniques, which utilised porous structures and silanization of RGD sequences, were combined and applied to the ITAP device. An EBM porous flange, of 700μm pore and 300μm strut diameters, were silanised with RGD-polypeptide and compared with the current clinical standard ITAP device. Implants were tested in a trans-tibial, transcutaneous ovine model for 5-months. Functionalisation techniques employed in this study did not eradicate the failure modalities of ITAP devices, however they did not detrimentally affect the formation of a stable transcutaneous interface compared with current clinical standards. Significant positive effects were observed, with the biological functionalisation, using silanised RGD-polypeptides, significantly increasing dermal tissue infiltration and porous structures, manufactured by EBM, significantly increasing vascularisation. Functionalisation with silanised RGD-polypeptides to porous structures may provide an opportunity to enhance the skin implant interface and tight biological seal for bone anchoring and skin penetrating devices, such as ITAP, clinically.

Production of polymeric magnetic micro and nano-particles is a rapidly emerging area in pharmaceutical and biomedical science. In this thesis, the capability of the electrohydrodynamic atomization (EHDA) process for preparing biodegradable polymeric magnetic particles with different sizes was explored. The EHDA processing method offers several advantages over conventional coprecipitation and emulsification techniques for the preparation of magnetic particles. Most significant are the process efficiency and preservation of the iron oxide nanoparticles and/or therapeutic agents functionality, as complex multistep processing involving harsh solvents, additives and elevated temperatures or pressure are avoided. The first part of the thesis describes a detailed investigation of how the size, morphology and shape of the particles generated can be controlled through the operating parameters; specifically the flow rate and applied voltage. The particle diameter was greatly influenced by flow rate and applied voltage. The mean size of the particles changed from1.6µ m to 17.8µm as the flow rate increased from 100µl/min-1 to 400µl/min-1. The research also focuses on the effects of these parameters on the jetting modes of the E H DA process, in particular the con-jet mode operation. Magnetic nanospheres were also produced using single needle processing with mean size of 56nm with a corresponding polydispersivity index of 16%. Nanospheres exhibited a high saturation magnetization at room temperature (67emu/g). Chlorotoxin, a scorpion venom was chosen as the therapeutic agent model because it is non-toxic, non-immunogenic along with other favourable characteristics such as small size high stability and most importantly only binds to tumour cells and not healthy cells. Scorpion venom loaded magnetic microspheres were produced using single needle processing, with a particle size of 2µm. This work demonstrates a powerful method of generating micro and nano magnetic polymeric particles, with control over the size of particles prepared.

No description available.

Bioresorbable polymers are used extensively in a wide range of medical applications, including orthopaedic fixation devices and bone scaffolds. They have excellent mechanical properties and the fact they do not· require removal roves advantageous over other materials. However, a relatively slow degradation rate has been observed and furthermore, undesirable acidic products are produced late in its degradation process. Incorporating bioactive fillers such as, calcium carbonate (CaC03) into bioresorbable polymers has the potential benefit of: (1) neutralising these acidic products and (2) releasing calcium (Ca) into the local environment to enhance bone formation (osteogenesis). The long term use of such composites has proven problematic due to the unpredictability of their degradation mechanisms. Therefore, a need exists to optimise the incorporation of bioactive fillers, with the ability to evaluate the long term resorption of the polymer and release mechanism of the bioactive fillers. The aim of this research was to evaluate the effect of incorporating CaC03 as a bioactive filler, into a bioresorbable polymer via twin screw extrusion, looking in particular at the Ca release mechanism, what influence particle size and shape has on this and the overall degra~ation profile. The accelerated methodology presented seems appropriate for rapid evaluation of slow-degrading polymer-composites, although caution is advised in terms of precise interpretation of results in relation to in vivo behaviour. It was found that minimal mass loss occurred at 37.5°C, which correlated with h low level of Ca release within the same period. Accelerating the study, using increased temperature, saw the extent of mass loss increase as a function of time. The increase in mass loss also correlated with the amount of filler material, with the smallest particle size relating to the largest mass loss. This work clearly demonstrates that CaC03 characteristics (particle size, shape and morphology) influence mechanical properties, degradation behaviour and calcium release from the polymer. Results also offered important insight into the mechanism by which CaC03 is released from PLLA, clearly showing there is minimal diffusion-controlled release (typical for many drug-release systems), with release instead being strongly dependent on the stage of polymer degradation. This offers exciting prospects for controlled bioactive release (including calcium phosphates) via manipulation of the polymer degradation profile. It is believed, once the correct levels of Ca required for optimising bone remodelling have been established, devices with improved release efficacy can be developed to fully aid osteogenesis.

Ceramide accumulates in Cystic Fibrosis (CF) airways and ceramide reduction decreased inflammation, susceptibility to Pseudomonas aeruginosa (Pa) infection, mortality and mobility. The research undertaken here aimed to evaluate the anti-inflammatory effect of sphingolipid-modifying drugs in LPS-stimulated bronchial epithelial cells. Moreover, the use of Raman Spectroscopy to identify-the distribution of biomembrane lipids was evaluated. Firstly, Raman markers of selected lipids were determined and their specificity and changes of lipid-derived Raman peaks due to bio-compound interaction were confirmed in solution. Secondly, membrane lipids within cells were investigated using Surface-enhanced Raman Spectroscopy (SERS) employing silver nanoparticles (NPs) and self-assembled thiols monolayers I (SAMs). NPs and S03--terminated SAMs promote a weak SERS, compared to choline-, methyl- or OHterminated SAMs. However, cellular lipid signals were not detected, SERS combined with confocal Raman microscopy gave enriched Raman spectra, but the signal was unstable and was not suitable for analyses , Thirdly, the anti-inflammatory effect of amitriptyline and miglustat on Pa LPS-induced inflammation in CF and non-CF cell lines was investigated. LPS from a clinical Pa isolate failed to induce inflammation, compared to commercially available LPS and bacterial lysate. Neither of the drugs suppressed LPS-induced inflammation, in contrast to dexamethasone, and cell lines responded to treatments similarly. Finally, using Raman spectroscopy combined with principal component analysis, the distribution of cellular components (DNA, RNA, proteins), and lipids (phosphatidylcholine, sphingomyelin, phosphatidylinositol, cholesterol) in LPS-stimulated cells pretreated with amitriptyline or sphingomyelinase was semiquantitatively imaged: CF cells expressed more sphingomyelin than non-CF cells. Sphingomyelin was reduced by sphingomyelinase, but increased by amitriptyline and LPS, suggesting that sphingomyelin was not associated with inflammation. Overall Raman spectroscopy techniques have potential for the label-free study of the cellular lipid composition; however, further improvements in the sensitivity are needed. The observation that miglustat and amitriptyline target LPS-independent immune responses needs further investigations.

Biomedical devices are commonly used in all areas of healthcare, These devices, which range from contact lenses through to endotracheal tubes, are most often fashioned from materials which allow the device to carry out its function thoroughly, but in doing so render the device susceptible to a number of complications. Two of the most major complications are that of device infection and poor frictional behaviour at the interface of the device and human tissue. This thesis details the development and characterisation of various polymeric systems which allow the resolution of these problems. With regard to infection, well established photodynamic techniques are further developed to provide a system which can bring about effective prevention of infection for prolonged durations of time, leading to a wide range of advantages, potentiating the function of the device. Biomaterial frictional behaviour is improved in a number of ways, including the development of next generation device coatings which are more easily wetted, offer improved biocompatibility, and also offer an improved tenacity of effect. Moreover, further work in this thesis has led to the development of successful photochemical attachment pathways for the addition of such coatings to the surface of commonly used biomaterial substrates.

In the surgical field, the trend tends increasingly towards performing minimally invasive procedures as they offer improved aesthetics, a reduction in blood loss, tissue trauma and post operative complications. However, the limited accessibility of the surgical site due to the use of small incisions does generate a few drawbacks. One of the main concerns is that surgeons cannot manually palpate the tissue they are operating on as they would in traditional open surgery. Manual palpation of the surgical site is performed to feel any tissue abnormalities such as tumours that may be present. This inability to feel the organs through minimally invasive surgery (MIS) is known as a lack of haptic feedback. Over the last decade, there has been an extensive amount of research in tactile and force sensing for MIS in an effort to overcome this limitation. The air cushion sensor for MIS is one of those. This novel sensing concept employs a spherical component that is surrounded by a cushion of air, to roll over the surface under inspection. The proposed sensing concept employs optical sensing when it is rolled to determine the areas of increased stiffness in a tissue. Using this concept, three sensors were designed and built. Their ability to detect tissue abnormalities in ex vivo tissues was assessed and so was their capability at estimating the force distribution over that area. The results show that all of the sensors were capable of detecting the tissue abnormalities and that one sensor was able to generate a force distribution map of the area under investigation.

In biomedical sciences, the increasing amount of available high through­ put data brings many challenges. The collection of such data usually results in large number of predictor genes and few samples, possibly also with high noise levels. Such problems are associated to the so called "curse of dimensionality", i.e. the small n large p problem. Therefore, the development and application of computational protocols in bioinformatics is necessary in order to tackle these problems, translate knowledge discovery from genome-scale studies and infer new knowledge combining the different types of post-genomics data. Data mining methods, including machine learning approaches, aim to identify patterns in high-throughput data and extract information about the underlying biological interactions. Research questions that are discussed in this thesis are disease strati­fication, biomarker discovery, network inference and data integration. The methodological contributions of this thesis focus on the problem encountered, nowadays, by clinicians where patients appearing to have the same disease may not respond to the same treatment. First, using supervised and unsupervised learning techniques, a machine learning strategy based on ensembles of decision trees was used to define sub­ phenotypes based on gene expression patterns and generate potential biomarkers for disease progression. Second, we developed a network inference algorithm (NetCFS) that uses feature selection to select a number of genes (n) that are highly correlated with the phenotype of interest so as to generate n different regression problems. Third, a "top-down" approach was implemented where gene sets corresponding to biochemical pathways are used to develop a disease classification framework. A multi-stage procedure was developed to uncover func­tional modules that are closely associated to the phenotype of interest and relevant to disease pathology. Phenotype-Responsive Genes (PRGs) are identified based on non-overlapping constraints of the classification procedure and association rules are used to estimate the activity level of each pathway. Applications discussed in this thesis include skin inflammation where an integrative approach combining clinically relevant in vivo mod­els with molecular network analysis was developed to infer disease biomarkers and to translate the rapidly growing body of data into knowledge usable at the bedside. Other disease cases studied involve cancer analyses to illustrate contributions in systems medicine. Overall, this thesis presents methodological contributions on predictive models based on machine learning techniques and mathematical programming together with relevant insights in disease mechanisms and potential treatment options.