An investigation of neurochemical, structural and functional characteristics in the murine model of collagen induced arthritis

Depression and rheumatoid arthritis have an estimated co-morbidity of 13-20%. However, the mechanisms whereby peripheral inflammation might alter brain function are unknown. We hypothesised that pro-inflammatory cytokines released in the periphery will result in neurochemical, structural and functional changes related to depression. Therefore, the aim of this thesis was to investigate altered central nervous system function in a rodent model of rheumatoid arthritis, the murine model of collagen induced arthritis (CIA). The CIA model is an established model used to investigate novel anti-inflammatory agents and resembles rheumatoid arthritis as the model is chronic and involves an autoimmune response to type II collagen. To our knowledge brain function in the CIA model has not previously been examined. Therefore, we began by identifying key neurochemical, cellular and functional changes associated with depression which had the greatest likelihood of being influenced by pro-inflammatory cytokines. Serotonin and dopamine transporter densities. The serotonergic system is implicated in the pathology of depression and there is evidence that pro-inflammatory cytokines may influence the serotonin transporter (SERT) in vitro. In vitro autoradiography binding of [125I]-β-carbomethoxy-3-β-(4 iodophenyl)tropane ([125I]-β-CIT) in the presence of mazindol and [3H]-citalopram was used to determine SERT binding in mice with CIA and controls. Out of 15 regions of interest investigated a significant change in SERT binding was identified by [125I]-β-CIT binding, in the nucleus accumbens (58%), thalamus (62%), and dentate gyrus (-60%) in CIA mice compared to controls. However, no significant difference in SERT density was detected in any region by [3H]-citalopram binding. Dopamine transporter (DAT) binding sites were also examined using [125I]-βCIT in the presence of displacer fluoxetine and [3H]-WIN 35,428. Out of 14 regions investigated a significant difference in DAT binding was only observed in the caudate putamen (95%) in the CIA group in comparison to the control group. However, no significant difference in DAT binding was detected in any region by [3H]-WIN 35,428 binding. A limitation of this study was the small group sizes and the degree of clinical symptoms in the CIA group. The data suggest that SERT and DAT transporter densities are not altered by CIA. [14C]-2-Deoxyglucose autoradiographic study of local cerebral glucose utilisation. To investigate brain function the [14C]-2-deoxyglucose ([14C]-2-DG) autoradiographic technique and a challenge to the serotonergic system were employed to identify any abnormalities in regional cerebral glucose utilisation. Overall there was no significant difference in the index of cerebral glucose utilization (iLCMRglu) in mice with chronic clinical symptoms of CIA. To investigate altered serotonergic function in the CIA model fenfluramine, a drug which stimulates serotonin release and blocks serotonin re-uptake was employed. Fenfluramine challenge in the CIA group resulted in only 3 out of the 35 regions of interest examined being significantly different from fenfluramine challenged controls. The orbital cortex (-41%) and the molecular level of the hippocampus (-26%) demonstrated a significant difference in iLCMRglu Overall the data suggest minimal influence of CIA on brain function. Cell proliferation and cell survival in the hippocampus. Hippocampal atrophy is implicated in the pathology of depression and there is evidence to suggest that pro-inflammatory cytokines reduce cell proliferation in vitro. To investigate hippocampal cell proliferation mice were administered 5’ –bromo-2’-deoxyuridine (BrdU), a marker of proliferating cells, prior to and after developing chronic clinical symptoms of CIA. There was no significant difference in cell proliferation prior to the development of clinical symptoms. There was a statistically significant increase in cell proliferation after chronic clinical symptoms in the CIA model in one out of two separate experiments. The data has been interpreted cautiously due to the fact the significant increase in cell proliferation was not reproduced. Cell survival was also investigated during the onset of clinical symptoms and the data demonstrated no significant effect of CIA on cell survival. Conclusion. The data indicate minimal influence of peripheral inflammation on the central nervous system, at least in the murine CIA model. Two possible explanations are that the CIA murine model is not a suitable model to detect changes in brain function associated with rheumatoid arthritis or that uninvestigated neurochemical systems play a role. This thesis highlights our limited understanding of the CIA model and whether or not it represents the features associated with rheumatoid arthritis other then peripheral inflammation. Further characterisation of the brain and development of the CIA model is required to establish if it is a suitable model to investigate the association between depression and rheumatoid arthritis. This is important as understanding the cause of depression and how the cause influences the brain will allow for the development of more specific treatments