Murine models of osteoarthritis (OA) provide a potentially powerful tool to elucidate mechanisms responsible for OA pain. However, few studies have examined pain behaviours in relevant OA models in mice. Several models including spontaneous, surgical and chemically induced OA were explored before selecting partial medial meniscectomy of C57BI/6 mice for in-depth study. The development of degenerative joint disease was confirmed using histology which illustrated progressive cartilage damage in a time dependent manner. Pain was assessed by monitoring weight bearing, mechanical hyperalgesia, cold allodynia, mechanical allodynia and vocalisation in response to knee compression for 12 weeks post-surgery. No significant weight bearing deficits were observed during the course of the study. Significant mechanical allodynia was present in the ipsilateral hind limb from 9 weeks following surgery. Hind limb mechanical hyperalgesia and cold allodynia, and increased vocalisation in response to knee compression developed in the ipsilateral limb in two phases. An early phase of hypersensitivities lasted for up to 3 weeks and was reversed by treatment with an NSAID, diclofenac. Pain then resolved for several weeks followed by a second phase of NSAID-insensitive pain after 6 weeks post-surgery. During this phase, all pain behaviours could be reversed by morphine. In contrast, other analgesic drugs (paracetamol, gabapentin and tramadol) had selective effects on only one or two modalities. Pain fluctuated during the second phase with transient periods of reduced pain. At these times underlying hypersensitivities could be unmasked by administration of naloxone indicating that reduced pain was due to endogenous opioids. The role of specific pain receptors in OA pain were investigated in this model of OA using genetically modified mice, novel analgesics and calcium imaging. The role of specific pain receptors in OA pain were investigated in this model of OA using genetically modified mice, novel analgesics and calcium imaging. Partial medial meniscectomy has proven to be a useful tool for the investigation of the pathophysiology of OA pain and also for the discovery of potential new therapeutics.

The existing medical model for managing rheumatoid arthritis (RA) focuses on minimising joint inflammation using suppressive treatments. However, patients have broader concerns spanning other symptoms like fatigue and pain and the way their health care is delivered. This thesis used qualitative and quantitative research methods to address three inter-related aspects of clinical care. Firstly, identifying critical challenges for providing patient-centred care. Secondly, defining outcomes important to patients like fatigue and pain and concomitant fibromyalgia. Finally, examining temporal changes in the RA management and evaluating aspects of clinical decision making. Firstly the thesis shows current care is not optimal. Key limitations include: being insufficiently patient-centred, failing to integrate management across the primary/secondary divide, over-emphasising drug treatment and overlooking "whole-person" care. Secondly, care overlooks several crucially important areas to RA patients. Many patients had high levels of fatigue, associated with pain, disability and psychological factors. Their fatigue spanned several domains. Current fatigue questionnaires are heavily weighted towards psychological aspects, and a more balanced assessment is needed. Pain, a dominant RA symptom, is often not directly addressed. The research showed central sensitisation causes persistent pain in many RA patients; it may require different management approaches. The research also characterised patients with the fibromyalgic rheumatoid clinical phenotypes; their higher disease activity scores may not fully reflect disease activity. Despite changes in treatment over the years their disease activity scores have not improved significantly, unlike RA patients in general. Different treatment strategies are needed to improve their outcomes. Finally although patients with high disease activity usually have their treatment changed when reviewed in rheumatology clinics, patients with moderate disease activity often have insufficient treatment changes; patients' age has a significant influence on treatment decisions. Strategies are needed to better target moderate disease activity and overcome the limiting effect of age on treatment decisions.

Systemic lupus erythematosus (SLE) is an autoimmune disease, with a strong genetic component. It is characterised by hyperactive T and B cells, chronic inflammation and the production of antinuclear autoantibodies. SLE affects mostly women of child baring age, with a 9:1 ratio, women to men and has been reported to be more prevalent in people of non-European ancestry. In the era of genome-wide association studies (GWAS), elucidating the genetic factors present in SLE has been very successful, with over 28 confirmed disease susceptibility loci mapped and a number of candidate genes identified. During this thesis I fine mapped IL18 as it had previously been reported to be associated with SLE, SNP rs360719. After fine mapping and subphenotype analysis in UK and African American cohorts, I was unable to replicate the published association. Although genetic data did not confirm IL18 to be associated with SLE, I demonstrated increased IL-18 serum levels in SLE renal patients compared to SLE patients. I further analysed IL10, another previously associated SLE candidate loci in our current SLE GWAS cohort (4000 cases and 9000 controls) of European ancestry. I again was unable to replicate the previous association, however using other SLE GWAS data showed SNP rs3024505 to be associated in Northern European samples. Further analysing our SLE GWAS, I located IKZF3 as a candidate loci. I identified an associated block of 56 SNPS and located the association to a single SNP rs2941509, p=1.46xlQ-8. Furthermore, I demonstrated an allelic imbalance in this SNP, with the protective G allele being expressed 1.5 times greater than the risk A allele, in controls. These data here demonstrated in this thesis indicates the importance of fine mapping candidate loci and verification of previously associated loci. This thesis contributes to the current knowledge of SLE by demonstrating discrepancies in published association data and showing the importance of larger studies.

Osteoarthritis is a common, multifactorial disease, characterised by the progressive loss of articular cartilage from synovial joints. It has a major genetic component that is polygenic in nature. The aim of my thesis was to explore functional effects of several osteoarthritis susceptibility loci. I have carried out functional studies on five genes in total, covering three osteoarthritis susceptibility loci, including GDF5, MICAL3 and CHST11. I used protein, gene and allelic expression analyses. Firstly, I assessed whether human chondrocytes would respond in a predictable manner to GDF5. Reduced expression of GDF5 (chromosome 20q11.22) correlates with an increased risk for osteoarthritis and I assessed what effect providing exogenous GDF5 to chondrocytes had on expression of target genes. I observed that chondrocytes responded to exogenous GDF5 in a highly discordant manner. Secondly, I carried out gene and allelic expression analyses on MICAL3 (chromosome 22q11.21), to investigate if the association to osteoarthritis at this locus is mediated by an influence on gene expression. I observed allelic expression imbalance (AEI) at this locus, however, this did not correlate with genotype at the associated polymorphism in the human joint tissues investigated. I then focused on the susceptibility locus at CHST11 (chromosome 12q23.3). AEI was common at this gene, although a correlation with genotype at the associated polymorphism was not observed in the human joint tissues investigated. I carried out sequence analysis of CHST11 in osteoarthritic patients to identify rare amino acid coding variants, and found no evidence for any accounting for disease susceptibility. I explored the role of CHST11 in mesenchymal stem cells during chondrogenesis and observed that CHST11 knockdown leads to gene expression changes in chondrocyte marker genes and a reduction in cartilage extracellular matrix synthesis. My studies highlight the complexity in performing functional studies to identify and characterise the causal polymorphisms influencing osteoarthritis susceptibility.

Background: Despite improvements in manufacture, sterilization and storage of tibial inserts, polyethylene wear continues to restrict the survivorship of total knee replacements. The problem is particularly evident in the young population undergoing knee arthroplasty. Aims and Methods: The objectives of the thesis were to ascertain: • If polyethylene wear may be reduced by utilizing scratch resistant femoral coats. • The timing of significant femoral scratching in the laboratory setting • If retrieved femoral components exhibited scratches and whether the roughening of the components is significant Results: • Both zirconium oxide and titanium niobium nitride femoral coats provided scratch resistance. However, no polyethylene wear benefit was conferred by the titanium niobium coat. • Significant scratching was noted as early as 100 cycles. • Retrieval specimens exhibited femoral scratching which were of the same orientation of those noted from the simulator studies. Not all condyles had significant roughening. Conclusions: Improving the scratch resistance of femoral components has the potential to improve the longevity of total knee replacements. However, alternate bearing surfaces should be used with caution, as the standard modern day knee replacement allows reproducibility and good mid to long term results which are lacking for alternate bearings. Improvements in polyethylene wear may be achieved by engineering a femoral coat that reduces abrasive, adhesive wear and delamination or by optimizing the ultra high molecular weight polyethylene components.

Osteoarthritis (OA) often affects the medial tibiofemoral joint (TFJt) of the knee. Foot posture could explain inconclusive study findings evaluating laterally wedged orthoses (LWOs) as an intervention. The overall aim of this PhD is to establish clinical lower limb alignment profiles in OA and non-OA participants. The thesis will then evaluate the mediating effect, if any, of foot posture on dynamic in-shoe loading patterns in OA gait and non-OA gait. The thesis further investigated the influence of LWOs on patient centred outcomes scores and in-shoe patterns in OA patients in a randomised controlled trial (RCT). The most prevalent foot posture in the non-OA group and the OA group was mildly pronated using the Foot Posture Index (FPI-6) (p>0.01). There were significant differences in the static and dynamic TFJt angles (p<0.01). In-shoe plantar pressure measurements also demonstrated a greater loading area and peak magnitude of loading in an OA pronated foot compared to a non-OA pronated foot (p<0.05). A 12 week RCT compared the effects of LWOs and a neutral orthotic on patient-centred outcome scores and in-plantar pressure measurements. While there were no significant differences between the groups for pain, symptoms, activities of daily living, recreational ability and quality of life (QoL) (p>0.05), there were significant differences within the groups over time (p<0.01). In the OA control group, there was an increase in peak pressure (p=0.03) in the lateral rearfoot compared with the treatment group. This suggests the LWOs may have had an offloading effect. Foot posture did not have an effect on patient centred outcome scores parameters (p>0.05), however, there was a footwear effect on in-shoe plantar pressure measurements (p<0.01). Clinical measures of foot posture did have a mediating effect on in-shoe plantar pressure outcomes in OA and healthy gait but not on patient outcome scores. The role of footwear alongside LWOs was a significant insight as it has rarely been considered. Their combined effect could help to reduce pain, manage symptoms and improve QoL in OA patients.

A large number of bone substitute materials are available; for which some authors claim osteoconductivity and some osteoinductivity. In order to rank these materials an in vivo analysis was carried out. These materials were chosen based on their availability and claimed mode of action. Silicon substituted Hydroxyapatite (SiHA), Hydroxyapatite (HA), Resorbable Calcium Phosphate Silicon, Skelite [siliconstabilized tricalcium phosphate-based bone substitute], Pro Osteon 500R [coralline HA], BiIonic [Yttrium stabilized SiHA] and two non-calcium phosphate, Dimeneralised Bone Matrix (DBM) based biomaterials: Accell Connexus DBM putty and Grafton crunch DBM were implanted in sheep femoral condyle defects for 6 weeks. Implanted calcium phosphate (CaP) based biomaterials demonstrated superior bone formation in comparison with the DBM samples. Silicon within CaPs increased the rate of bone formation in vivo. Silicon substituted HA showed increased proliferation rate (P<0.05) of human marrow stromal cells compared to pure HA in vitro. Expression of osteoblastic marker genes RUNX2, Osterix and Osteopontin within the hMSCs indicated the differentiation of preosteoblasts into osteoblasts, and osteogenic development on both HA and SiHA. Expression of osteocalcin and bone sialoprotein genes on HA and SiHA samples indicated the activation of mineralisation process. Differentiation of hMSCs into osteoblasts in vitro suggested a role in promotion of osteoinduction by both HA and SiHA. Implantation of porous SiHA and HA in paraspinous muscle of sheep, exhibited new bone formation through osteoinduction. SiHA indicated significantly higher new bone formation (P<0.01) compared to HA. SiHA and HA biomaterials with higher strut porosity (30%) indicated greater bone formation (P<0.05). In conclusion, CaP based biomaterials demonstrate superior bone formation in comparison with DBM biomaterials. Silicon substitution within HA enhances the cellular activity of hMSCs. Osteoinduction was greatest on SiHA with higher strut porosity. This result is believed to be due to a combination of the effect of interconnected porosity and chemical composition of the bone substitute.

The congenital muscular dystrophies are a clinically and genetically heterogeneous group of disorders characterised by a congenital onset of weakness and hypotonia, typically associated with dystrophic-appearing muscle biopsy findings. The spectrum of clinical phenotypes associated with the congenital muscular dystrophy subgroup resulting from a deficiency of collagen VI in the extracellular matrix of muscle are collectively termed ‘collagen VI-related myopathies’ and include the early onset Ullrich congenital muscular dystrophy and the milder and later onset Bethlem myopathy as well as a phenotype of intermediate severity called ‘intermediate collagen VI-related myopathy.’ A major goal of this research has been to study the natural history of respiratory insufficiency in the collagen VI-related myopathies by analyzing longitudinal forced vital capacity data in a large, international cohort. A total of 486 forced vital capacity measurements obtained in 145 genetically and/or biochemically confirmed collagen VI-related myopathy patients from 10 neuromuscular centres [United States (2), United Kingdom (2), Australia (2), Italy (2), France (1) and Belgium (1)] were analysed and the resulting clarification of the phenotypic stratification of collagen VI-related myopathies reported. Another focus of this research has been the evaluation and the refining of the challenging diagnostic pathway for collagen VI-related myopathy patients, including evaluations of the diagnostic role of muscle histopathology, skin fibroblast immunocytochemistry and flow cytometry studies as well as muscle ultrasound and muscle magnetic resonance imaging studies. Finally, this research has studied the role of next generation genetic sequencing technologies including whole-genome and exome sequencing in the assessment of patients evaluated for collagen VI-related myopathies and related conditions.

Critical size bone defects after fracture or malignant tumour resection are still a challenge to repair in orthopaedics. Stem cell therapy combined with cytokines and bone grafts has the potential to improve outcomes. However, this application has its drawbacks preventing it from popular clinical use. The large number of stem cells required for transplantation is often a limiting factor. The goal of this thesis was to make the stem cells more effective and efficient in bone repair thus potentially reducing the required number of cells. In this study I engineered stem cells by the introduction of a gene to over express stromal cellderived factor-1 (SDF-1), a pivotal chemokine that has been proved to regulate cell migration, with the hypothesis that these cells would effectively increase the migration of native cells to the site of the repair, thus, enhancing bone repair. In vitro treatment of recombinant SDF-1 to human mesenchymal stem cells induced significantly greater osteogenic differentiation compared with control cells (p=0.024) and increased the migration of non-infected cells in a trans-well migration test (p=0.04). In a rat femoral bone defect model, using a low number of bone marrow cells, resulted in no difference in bone formation compared control defects without cells. Interestingly, the same number of bone marrow cells overexpressing SDF-1 showed significantly (86% increase, p=0.02) more new bone formation within the gap and less bone mineral loss at the area adjacent to the defect site during the early bone healing stage. A greater number of donor cells transfected with SDF-1 remained in the repair site compared with the control non-transfected site. An additional second cell injection of cells at 3 weeks was applied to the fracture but did not result in increased new bone formation but did reduce bone mineral loss at this time point. This thesis demonstrates that by applying stem cells transfected with SDF-1, bone fracture healing was improved using a low cell number, which is a non-optimal condition for normal stem cell transplantation. This suggests that SDF-1 transfected cells recruited more host’s stem cells into the fracture gap or resulted in greater osteogenic differentiation, preventing bone loss and increasing bone formation. These findings need further investigations to reveal the mechanism of SDF-1 in bone healing by studying the effects of down stream signaling pathway of SDF-1, the cell type of the recruited cells and the angiogenesis in the defect site.

Autoimmune arthritis in childhood, know as juvenile idiopathic arthritis (JIA), has a heterogeneity, ranging from monoarthritis to recalcitrant polyarthritis, making it a model disease in which to study immuno-regulation. Regulatory T cells, key players in peripheral immune homeostasis are enriched in the joints of JIA patients, particularly those with mild arthritis. To test the factors that lead to this enrichment, Treg trafficking was examined in the context of JIA. Synovial Treg showed enhanced chemotaxis to the inflammatory chemokine CCL5, widely detectable within the joint, when compared to Treg from peripheral blood. The trafficking of a related, but highly inflammatory T cell subset, Th17 cells, was also investigated. Th17 cells play a dominant role in murine models of arthritis, yet their contribution to human disease is unknown. The data presented here, showed that IL-17+ CD4 T cells were enriched in the joints of JIA patients, by a CCR6 dependent mechanism and importantly, their frequency correlated with the severity of disease course. The majority of synovial IL-17+ CD4 T cells expressed a cytokine and chemokine receptor phenotype intermediate between Th17 and Th1. Here it was shown that these cells (Th17/1) expressed high levels of both Th17 and Th1 specific transcription factors, RORC2 and T-bet. Modelling the generation of Th17/1 in vitro, Th17 cells ‘converted’ to Th17/1 under conditions which mimicked the disease site, namely low TGFβ and high IL-12 levels. Using CD161, a human Th17 marker, it was shown that synovial Th17/1 cells, and unexpectedly, a large proportion of Th1 cells expressed CD161. This study provided evidence to support a Th17 origin for Th1 cells expressing CD161. In vitro, Th17 cells which converted to a Th1 phenotype maintained CD161 expression, whilst in the joint CD161+ Th1 cells shared features with Th17 cells, with shared T cell receptor clonality and expression of RORC2, although they were IL-17 negative. We propose that Th17 cells may 'convert' to Th17/1 and Th1 cells in human arthritis. Therefore therapies targeting the induction of Th17 cells could also attenuate the Th17/1 and Th1 effector populations within the inflamed joint.