The application of composite tissue techniques is constrained by the susceptibility of skin to rejection. The aim of this thesis is to improve our understanding of skin rejection and find ways to avoid it, in order to enable expansion of the application of composite tissue transplantation techniques. The first part of the thesis explores the consequences and mechanism of skin rejection in rat models. These studies indicate that in the event of allograft failure, there is minimal damage to the vascular pedicle of a composite tissue allotransplant, even after full rejection, making retransplantation possible. Furthermore, there is only mild damage to the recipient tissues, indicating that the second transplant would not be limited in form or function by recipient tissue bed damage. Finally, the studies indicate that there are significant differences between the mechanism of rejection of skin in composite tissue transplants and conventional skin grafts. This means that much of the historical data relating to skin graft rejection is not necessarily relevant to composite tissue allotransplantation. The second part of the thesis uses swine models to explore ways to overcome skin rejection while avoiding the toxicity of chronic systemic immunosuppression, through tolerance induction, and site specific therapy. Previous experience in organ and composite tissue allotransplantation models are analysed to develop the hypothesis that high-level chimeras are tolerant to vascularised skin allotransplants. In utero and adult chimerism induction models are then used in an attempt to attain moderate-level chimeras. A vascularised skin allotransplant model is developed. Finally, the hypothesis is confirmed with the transplantation of a vascularised skin allotransplant on to moderate-level chimeras with the achievement of tolerance. In addition, site-specific therapy is used in an attempt to avoid the side-effects of chronic high-dose systemic immunosuppression. This led to prolongation of skin survival, but eventual skin rejection.

Ischaemia reperfusion (IR) injury is a major contributor to morbidity and mortality following liver resection and transplantation. Hindlimb remote ischaemic preconditioning (RIPC) reduces liver IR injury but the mechanisms underlying this protection are unknown. This thesis evaluated the effects of RIPC on the early phase of liver IR injury in a new mouse model. This model was used to assess the role of the nitric oxide (NO) pathway in mediating the protective effects of RIPC on liver IR injury. In the initial experiments wild type C57BL/6 mice were used to establish a new model of hindlimb RIPC that protects against liver IR injury. The new experimental model hindlimb protocol consisted of 6 cycles of 4 minutes of femoral vessel clamping followed by 4 minutes of reperfusion. The IR protocol consisted of a well-described 40 minutes of lobar liver (70%) ischaemia followed by 2 hours reperfusion. This hindlimb RIPC model resulted in significant reductions in liver IR injury, as evaluated by plasma liver transaminases levels, histopathological scores, and ultrastructural assessment of cellular damage in the liver. In addition hindlimb RIPC preserved the hepatic microcirculatory blood flow (MBF) in livers subjected to IR. In subsequent experiments administration of the selective NO scavenger C-PTIO to wild type mice abrogated the protective effects of limb RIPC on liver IR injury. In addition the RIPC-induced preservation of hepatic MBF was attenuated by C-PTIO administration. In contrast to the protective effects of limb RIPC on liver IR injury in wild type mice; mice lacking the constitutively expressed NO synthase (eNOS-/-) enzyme were not protected against liver IR injury by hindlimb RIPC, and MBF measurements in these mice showed no benefit of RIPC on the microcirculation. In wild type mice hepatic and limb eNOS protein expression was similar among preconditioned and non-preconditioned animals. In comparison expression of the inducible NOS (iNOS) isoform was only seen in preconditioned animals. In order to elicit the pathway through which RIPC-derived NO protects against liver IR injury, plasma nitrite and nitrate (NOx) levels were quantified in wild type animals and shown to be significantly elevated in preconditioned compared to non-preconditioned animals. However, intravenous administration of exogenous nitrite to eNOS-/- mice undergoing liver IR injury failed to mimic the protective effects of RIPC-derived endogenous NOx in wild types. Administration of the soluble guanylyl cyclase (sGC) inhibitor ODQ showed a trend indicating reversal of the protective effects of RIPC on liver IR injury but this did not reach statistical significance. ODQ significantly annulled the protective effects of limb RIPC on hepatic MBF during liver reperfusion. Measurement of hepatic cyclic GMP levels in wild type animals showed a significant increase in animals subjected to limb RIPC only compared to sham. However there was no difference in the cGMP levels in the RIPC + IR compared to the IR group. In conclusion this thesis has described a new mouse model of limb RIPC that protects against liver IR injury. It was also shown that NO and eNOS are essential in mediating the protective effects of limb RIPC on liver IR injury. Endogenous NOx metabolites of NO play a crucial role in RIPC induced protection. The hepatic sGCcGMP pathway is at least partially involved in RIPC-induced liver protection.

Total hip arthroplasty is a well established treatment modality for the diseased hip. The number implanted rises annually on a global scale which is mirrored by increasing indications. After aseptic loosening and infection, periprosthetic fracture remains one of the commonest complications of this otherwise successful surgery. Management is geared towards restoring function through fixation of the fracture. The general aim of this thesis is to validate the classification of periprosthetic fractures of the femur around total hip arthroplasty, provide evidence towards the outcomes of methods of fixation of these fractures, and present supplementary biomechanical data regarding fixation and implant stress. It is hypothesised that the Vancouver classification will be a reliable and reproducible system to use, that strut grafts, cables and long-stemmed implants will improve function and outcome when used to manage these injuries, and that biomechanical models will provide evidence on why the use of the implants is successful. Study I The purpose of this study was to ensure that the Vancouver Classification of periprosthetic fractures which is most widely used classification system of periprosthetic fractures is repeatable. It was hypothesised that the system would be reliable amongst for all grades of clinician. The inter-rater agreement ranged from 0.61-0.74 and the intra-rater agreement ranged from 0.59-0.67. Validity analysis was scored at 77% (κ = 0.67). The Vancouver Classification was shown to be reliable and reproducible. Study II The purpose of this study was to evaluate the clinical and radiographic outcomes of 40 periprosthetic femoral fractures around stable hip implants treated with cortical onlay strut allografts without revision of the stem. It was hypothesised that this treatment would improve function and result in bony union. At a mean follow-up of 28 months, 98% of patients had radiological evidence of union with all but one of the surviving patients returned to their preoperative functional level within one year. Study III The purpose of this biomechanical cadaveric study was to determine the effect of allograft cortical strut length, configuration, cable number, cable tension and the use of wire or cable on the fixation of periprosthetic femoral fractures. It was hypothesised that an increasing number of struts and the use of cable would improve fracture stability. Fracture stability was found to increase with the use of two rather than one strut, and by using cables rather than wires. Study IV The purpose of this study was to evaluate the clinical and radiological outcomes of using cementless femoral stems in conjunction with cortical struts, cable plating systems, bone allograft and demineralised bone matrix in 26 patients with Vancouver B2 or B3 fractures. It was hypothesised that this treatment would improve function and lead to radiological union. It was found that all fractures were healed clinically and radiologically, and all patients were reported to be satisfied with the outcome. Study V The purpose of this biomechanical study was to determine the strain exerted by an uncemented femoral implant upon a synthetic, composite femur modelling various clinical scenarios. It was hypothesised that strain would be reduced when using a grip, strut or cables. It was found that these devices did reduce the strain exerted upon the femur and may be useful in preventing femoral stem fractures. Study VI The purpose of this study was to evaluate the clinical and radiographic outcomes of treating periprosthetic femoral fractures around unstable hip implants treated with revision to an uncoated locked Kent Hip prosthesis. It was hypothesised that this method of treatment would improve clinical and radiological outcome in the 36 patients included in the study group. Harris Hip Scores improved and fracture union was seen in all but one patient; there were three patients in whom the implant was subsequently revised. Study VII The purpose of this study was to clinically evaluate interlocking long stem femoral prostheses as either temporary functional spacers or as definitive implants in cases of infected periprosthetic femoral fractures. It was hypothesised that these devices would improve the clinical and radiological outcomes of these patients. The Cannulok uncoated stem was used in twelve cases and the Kent Hip Prosthesis in five cases. Patients were asked post-operatively they were satisfied with the outcome achieved. All patients were satisfied and in eleven cases, revision to a definitive stem was undertaken after successful control of the infection and fracture union. Conclusions The management of periprosthetic fractures is a complex issue. There are numerous ways to manage this injury and treatment must be tailored to the patient and to the specific injury sustained. The results of this work demonstrate that classifying periprosthetic fractures using the Vancouver system is valid. Furthermore cortical struts are an effective adjunct with proven biomechanical advantages in non-infected cases around stable implants, whilst long cementless stems lead to excellent outcomes in the presence of a loose implant irrespective of infection.

Tendon injuries of the hand are common with nearly one-third of a million digital flexor tendon injuries per year in the United States. Injuries in zone II of the flexor tendon are notoriously difficult to repair and the main complications are either tendon rupture or adhesion formation. Adhesions remain a problem despite many attempts at prevention using various chemicals and physical barrier techniques. The overall aim of this thesis was to further understand the biology of tendon adhesion formation and to develop novel treatments targeting this process. Uninjured flexor tendons were obtained from New Zealand White rabbits. Tenocytes derived from different parts of the flexor tendon-sheath complex were grown using standard tissue culture techniques. Each of the three different cell types (endotenon, epitenon and tendon sheath) was subjected to various assays (proliferation/toxicology, cell adhesion, and mRNA expression,) using TGF-β1 and our proposed treatments; epigallocatechin-3-gallate (EGCG), Resveratrol and Pumactant. A further study then compared the three treatments in vivo. New Zealand White rabbits (n=8 per group; 32 in total) were anaesthetised and the flexor digitorum profundus (FDP) of digits 2 and 4 of the forepaw was subjected to a partial tenotomy. The three treatments (compared with control groups) were infiltrated into the flexor sheath of immobilised tendons and the wound was then sutured closed. After two weeks the tendons were harvested and randomised to either mechanical or histological assessment of adhesion formation. The major findings from the in vitro study were as follows: TGF-β1 showed a statistically significant increase in collagen type I gene expression in epitenon cells at 24 and 48 hours and an increase in collagen type III in sheath cells between 6 and 24 hours. There was a statistically significant down-regulation of collagen type III in endotenon and epitenon cells at various time points. Resveratrol showed a statistically significant increase in collagen type I gene expression in epitenon cells with a corresponding down-regulation of fibronectin and PAI-1 in both epitenon and sheath cells. Resveratrol also up-regulated collagen type III at late time points in tendon sheath cells. Pumactant also showed some therapeutic advantages at the gene expression level with a statistically significant increase in collagen type III in endotenon cells at late time points, corresponding with an overall down-regulation of PAI expression in the same cell type and sheath cells. The results from the in vivo study were that all three treatments showed a statistically significant reduction of tendon adhesion formation when compared to operated controls in both mechanical and histological assessments (p<0.05). However, Pumactant was the only treatment to demonstrate a statistically significant reduction in adhesion formation when compared to the H20-group using both methods (p<0.05). All three treatments displayed potential therapeutic advantages. However, Pumactant showed the most promising in vivo results and it would be worthwhile investigating this further with an in vivo model of tendon healing and if successful a pilot clinical trial. Hopefully this could act as a suitable adjunct to tendon repair in the future and improve the lives of patients with disabling tendon injuries.

Inflammatory bowel disease is a chronic inflammatory condition of bowel. CT and barium fluoroscopy are main stay of radiological investigation but impart high radiation dose. MRI is a safe and less invasive technique to assess bowel. The thesis examines the use of magnetic resonance Imaging in enteric inflammatory bowel disease. A discussion on inflammatory bowel disease and overview of MRI techniques, diagnostic features and review of literature is described. A national survey about the uptake of MRI for the investigation of IBD showed that 38% of radiology departments offered enteric MRI and barium studies remaining the main imaging investigation performed. A proforma administered to clinicians in OPD showed significant increase in their diagnostic confidence for small bowel disease after MRE, which had positive impact on therapeutic strategy of 61% of patients. The results of a prospective studies investigating MR colonography as a biomarker of disease activity are then presented. Quantitative measurements of contrast enhancement in normal colon have shown intersegmental differences. Three proposed qualitative MRI scores of disease activity correlated with endoscopic disease activity, but correlation with histopathological scores was less apparent. The use of unprepared colonic MRI in assessment of acute colitis is then investigated. A qualitative total colonic inflammation score (TCIS) proposed and validated against clinical standards including stool frequency and CRP. It also has prognostic ability for length of hospital stay. Region of interest derived quantitative measurements from the colon wall including T2 signal and contrast enhancement are then compared to a validated clinical score of colitis activity. Quantitative markers seemed less robust then qualitative scores, although quantified contrast enhancement is correlated with disease severity. Patient experiences of MR Colonography and colonoscopy are investigated by using face-to-face qualitative interviews, together with a quantitative questionnaire. Patient preference is highly complex but patients expressed overall preference for MRC.

Bladder cancer is the second most common urological cancer after prostate cancer and is one of the leading causes of cancer mortality in most western countries. For organ-confined, muscle-invasive disease, the standard of care, in terms of definitive cure, remains radical surgery (cystectomy) with lymphadenectomy. However, survival rates remain poor following supposedly curative treatment. Radical radiotherapy and more recently, multimodality treatment incorporating chemo-radiotherapy, are alternatives which allow bladder preservation in those choosing not to undergo or are unsuitable for radical surgery. However, survival rates following radiotherapy are generally lower relative to radical cystectomy and multimodality treatments can only be offered to select cases in few institutions. Biomarkers which can accurately predict tumour response to radiotherapy or chemotherapy can aid the selection of patients who are likely to respond well to treatment options incorporating radiotherapy and/or chemotherapy, as alternatives to radical cystectomy, in the management of bladder cancer. Such a strategy would allow personalised cancer care with patients likely to benefit from treatments that they are likely to respond well to and concomitantly avoid complications arising from other treatments less likely to benefit them. This thesis investigated the novel tumour suppressor gene, AIMP3 which is involved in the DNA damage response (DDR) pathway following exposure to genotoxic insults such as irradiation and chemotherapy. The expression and cellular localisation of AIMP3 protein was characterised in a panel of bladder cancer cell lines. Expression of AIMP3 was altered by gene knockdown with siRNA transfection and survival outcomes assessed following irradiation and chemotherapy. The predictive value of AIMP3 expression in determining survival outcome of patients with muscle-invasive bladder cancer who had undergone radical radiotherapy, with or without carbogen supplementation, in the BCON trial, was assessed. Prognostic significance was evaluated by interrogating a control cohort of patients who had undergone radical cystectomy and had not had exposure to radiotherapy or either neoadjuvant or adjuvant chemotherapy. Reportedly important DDR proteins, including Mre11, p53 and ERCC1, were also interrogated in the BCON, Radical Cystectomy, Neodjuvant and LaMB trial TMA datasets. Clonogenic survival outcomes following AIMP3 knockdown were also investigated in cisplatin-sensitive (RT112) and cisplatin-resistant (RT112CP) cell lines following cisplatin exposure. Survival outcome, stratified for AIMP3 as well as ERCC1, Mre11 and p53 status, were interrogated in the Neoadjuvant set, which incorporated a cohort of patients who had undergone cisplatin-based neoadjuvant chemotherapy prior to radical treatment. This was validated in a second cohort of patients who had undergone cisplatin-based chemotherapy as part of the LaMB trial.

Nerve autografting is the “gold standard” technique to repair nerve defects with a gap larger than 30 mm. The current commercially available FDA and CE approved nerve conduits offer considerable benefits to the patients suffering from completely transected nerve. They fail, however, to support neural regeneration in gaps over 30 mm. The aim of this research was to design, develop and evaluate new nerve conduits made from a biodegradable nanocomposite material known as polyhedral oligomeric silsesquioxanes incorporated poly (caprolactone) urea/ urethane (POSS-PCL). This material has been previously shown to have favourable cellular interactions. The biomechanical properties of POSS-PCL nanocomposite with varying POSS concentration were evaluated. Increasing POSS concentration resulted in less viscous polymer solution while increased the surface hydrophobicity, surface roughness and in vitro protein absorption. This increase, however, caused a considerable reduction in Schwann cells proliferation and rounded morphology. To enhance cellular interactions of the POSS-PCL surface, it was functionalized with synthetic RGD peptide, which resulted in an increase of SCs average process length while reducing the hydrophobicity of POSS-PCL surface. Furthermore, human adipose derived stem cells were successfully differentiated into Schwann-like cells as determined by S-100 expression and NGF production. The porogen concentration used for making porous conduits was also investigated using solvent evaporation technique combined with porogen leaching. It was demonstrated that 2% POSS-PCL with 30% porogen had the favourable viscoelastic properties for nerve conduit manufacturing. Two types of nerve conduit with varying wall thicknesses were fabricated and examined for their physiochemical properties. Double layered conduits were considered more suitable for the short-term pilot in vivo study as they had higher compressive resistance and suture retention ability than the single layered ones. Following in vivo implantation of conduits, Visual observations showed good interaction of conduits with surrounding tissue and no obvious inflammation at the repair site after 6 weeks. Histology revealed that the porous conduit (17.53±6.44 μm pore size) improved myelin sheath formation compared to non-porous POSS-PCL nerve conduit. Further investigations of POSS-PCL conduits are required to determine if this implant can overcome the limitation of commercially available nerve conduits.

Introduction: The circulating renin-angiotensin system (RAS) was originally described as a key endocrine regulator of intravascular homeostasis; however, the existence of a local (tissue) RAS has become increasingly reported in a variety of tissues including liver. RAS components have now also been detected in rat heart, brain and smooth muscle cell mitochondria as well as within intramitochondrial dense bodies of rat adrenal tissue. Further, reduced RAS levels have been associated with improved endurance performance and fatigue resistance in human skeletal muscle, suggesting that low RAS activity is associated with metabolic efficiency, potentially via RAS action upon, or within, mitochondria. However, such investigation has often relied heavily upon qualitative techniques (e.g. Western blotting, immunofluorescence and electron microscopy), which contain inherent limitations in that they completely rely upon the limited specificity of antibodies to demonstrate the existence of intra-mitochondrial RAS components. Methods: The presence of RAS components within the mitochondria of rat hepatic tissue and liver cell-lines was investigated via sub-fractionation of rat liver tissue and cell-lines, followed by Western blotting, as well as via immunofluorescence and confocal microscopy, and electron microscopy. The mitochondrial effects of stimulating or antagonizing hepatic RAS were assessed via functional fluorescence microscopy (for assessment of NADH, calcium and mitochondrial membrane potential) and measurement of oxygen consumption within live cells of a liver cell-line. Results: Western blotting, immunofluorescence and electron microscopy suggested the presence of RAS components within mitochondria; however, there was a lack of results consistency between techniques and the staining patterns were largely non-specific. Western blotting further demonstrated the presence of a prominent 55 kDa band, when immunostaining a mitochondrial fraction with (angiotensin-converting enzyme) ACE Cterminal antibody (usual size 180 kDa). This was further explored via isolation of the 55 kDa molecule and mass spectrometry to yield results consistent with non-specific staining only. Addition of RAS agonists or antagonists to live liver cell-lines demonstrated no consistent results, except at supra-physiological levels, where RAS antagonists improved oxygen consumption. Conclusions: Such data suggest that the previous descriptions of RAS components within mitochondria are likely to be secondary to methodological flaws, particularly the reliance upon single antibodies, which have subsequently been shown to have poor specificity. Thus, the effect of ang II on liver mitochondria is unlikely to be direct and any such action is likely to occur via one of several intracellular pathways, regulation of gene expression or mitochondrial biogenesis.

Liver organ shortage is an increasing problem worldwide and many die each year waiting for a new liver. In case of acute liver failure a bioartificial liver (BAL) device could “buy” time until a donor liver is available or until the liver has spontaneously undergone self-repair. For the clinical application of a BAL large quantities of cells should be available immediately necessitating cryo-banking. However, cryopreservation of large volumes results in increased ice formation and increased cell death. Ice formation can be prevented by vitrification, but the high cryoprotectant agent (CPA) concentrations needed are normally toxic to mammalian cells. Short exposure time minimizes toxicity but can only be achieved in small samples where fast cooling rates can be reached. To reduce CPA toxicity a vitrification machine (Liquidus Tracker) designed by Planer plc was used, which provides the lowest toxic effect that can be established for a given CPA concentration by decreasing the sample temperature to just above the melting point of that particular mix. The CPA concentration is then gradually increased as temperature is decreased along the liquidus curve. The first aspect of this thesis was to standardise a rapid and reliable method to describe post-stress viability. A digital imaging system was used to evaluate membrane integrity and enzyme activity by quantifying the fluorescence signal of fluorescein and propidium iodide. To understand the Liquidus Tracking (LT) process but also to pre-test conditions for automatic LT, different methods to carry out manual LT were established, evaluated and improved. To further increase cell viability a low-toxicity CPA solution was developed with the requirement of low viscosity so that it may be used within the Liquidus Tracker. Finally improvements were applied to automatic Liquidus Tracking. The development of a new stirring system substantially increased post-warming viability. In conclusion, an optimised large scale slow cooling vitrification protocol was developed for alginate encapsulated liver cells which may be used in a BAL.

A third of patients with critical limb ischaemia (CLI) eventually require amputation. Inconsistencies between successful revascularisation and functional outcomes exist, and underlying musculopathy in CLI patients has been identified. Erythropoietin (EPO) has tissue-protective effects in response to ischaemic injury, but its clinical use is often precluded by thromboembolic side effects. Non-haematopoietic EPO-derivatives have been designed to retain only tissueprotective functions of EPO. We hypothesised that ARA-290 (EPO-derivative) may have tissue-protective potential that could represent a novel therapeutic adjunct in patients with CLI. The effect of EPO and ARA-290 in mediating cytoprotection in an in vitro simulated ischaemia model of skeletal muscle was assessed firstly in the immortalised murine C2C12 myoblast cell line and subsequently in skeletal myoblasts isolated from CLI and control donors. In human and murine cells, simulated ischaemia alone demonstrated a detrimental effect on cell function and survival. Addition of EPO or ARA-290 demonstrated significant improvements in function and survival and utilised JAK2/STAT3, PI3k/Akt and NF!B signalling molecules. Isolation of human skeletal myoblasts from CLI patients has not previously been described. Comparison of CLI and control myoblasts elucidated significant differences in their function and survival under both normoxic and simulated ischaemic conditions. CLI myoblasts and myotubes exhibited increased proliferative capacity but reduced migratory and contractile function and importantly a reduced susceptibility to a second ischaemic-insult compared with control myoblasts and myotubes. Evaluation of several variations in the hindlimb ischaemia model allowed the creation of a model which closely recapitulated the muscular pathology observed in human CLI patients. ARA-290 demonstrated improved functional, histological and perfusion outcomes compared to EPO or vehicle-control treated animals. These studies demonstrate the potential of ARA-290 to protect tissues and cells from ischaemic-injury and encourages the development of novel pharmacological therapies for use in patients with “no option” CLI or severe functional deficit.