Ankylosing spondylitis (AS) is a chronic inflammatory arthritis characterised by severe inflammation of the axial skeleton followed by bone formation that can lead to ankylosis.The mechanisms mediating the transition from inflammation to bone formation are poorly understood due to the limited availability of relevant bone samples. Therefore, we used the proteoglycan-induced spondylitis (PGISp) mouse model to delineate the morphological changes during axial disease development. This mouse model develops spondylitis followed by ankylosis, mimicking the clinical progression seen in patients with AS.The first part of this project aimed to characterise the disease progression across a 43-week time-course in the PGISp mouse model. The principal assessment for analysing spinal disease was a semi-quantitative histological score that assessed joint inflammation, joint destruction (including intervertebral disc (IVD), cartilage and bone) and excessive tissue formation (peri-joint mesenchymal tissue expansion or fibrocartilage formation).Early inflammation initiated at the periphery of the IVD and was followed by varying levels of disc, cartilage and bone damage. In the advanced stages of disease, excessive tissue formation and ectopic chondrocyte expansion, ectopic bone and osteophyte formation were the key features. Abnormal tissue formation was always associated with IVD destruction, which was the result of inflammation, indicating that inflammation-derived IVD destruction is a prerequisite for induction of excessive tissue formation and the subsequent osteoproliferation.Current therapies for AS include non-steroidal anti-inflammatory drugs (NSAIDs), tumour necrosis factor (TNF) inhibitors and physiotherapy. However, these therapies aim to alleviate symptoms but cannot prevent syndesmophyte formation; hence new, more effective, therapeutic strategies are required. Genome-wide association studies have shown that AS is strongly associated with prostaglandin E2 (PGE2) receptor subtype 4 (EP4), which plays regulatory roles in both inflammation and bone formation. The involvement of EP4 in AS has not been examined. The second objective of this thesis was to examine the hypothesis that blocking EP4 can suppress disease progression by inhibiting both inflammation and bone formation. PGISp mice were prophylactically administered with ONO-AE1-329 (an EP4 agonist) or ONO-AE2-227 (an EP4 antagonist) ii for 16 weeks. Indomethacin, an NSAID, was included as a positive control of PGE2 signalling inhibitor. None of the treatments significantly altered peripheral arthritis or axial disease progression. Use of EP4 agonist or antagonist did not change bone mineral density and bone mineral content as measured by dual-energy X-ray absorptiometry. The