Depression is a major disease burden worldwide and, despite its prevalence and socioeconomic costs, around 30% of patients do not respond to currently available treatments. Inflammation is increasingly associated with, not only depressive illness but also resistance to existing therapies. This highlights the need for investigation of the mechanisms of neuro-inflammation, particularly in the context of peripheral inflammatory stimuli. Specifically, the chemokine molecular family is increasingly associated with human depressive illness, and neuro-inflammation and behavioural change in rodent models, making this an attractive molecular family for study. This thesis describes research aimed at investigating the association of these molecules with human depression and analysis of their role in an animal model of peripherally stimulated neuro-inflammation, the Aldara model of psoriasis-like inflammation.Systematic review and meta-analysis of the human biomarker literature using a random effects, inverse variance model revealed that a number of chemokines (CCL2, CCL3, CCL4, CCL11, CXCL4, CXCL7, CXCL8) are significantly associated with depressive illness in a human population. However this work revealed that there are a number of limitations of the human literature primarily associated with the methodological challenges of studies in human populations and confounding factors.Alongside this work, the Aldara model, which utilises the toll-like receptor 7 (TLR7) ligand imiquimod (IMQ), was investigated as a tool for studying neuroinflammation. Initial time-course investigation revealed that significant chemokine and cytokine transcriptional alterations occur within four hours at the local site of cutaneous treatment, the peripheral tissues and the brain. In addition, protein quantification in the brain confirmed that many of these transcriptional responses are translated to protein. Interestingly, it was shown that the brain response was temporally distinct from that of the peripheral tissues, and that in general brain responses were induced slightly more slowly and persisted for a longer period of time than those in the periphery. Investigation of Iba1+ (microglia/monocytes), GFAP+ (astrocytes) and CD3+ (T-cells) cells within the brain revealed significant changes in the microglial and T-cell populations, which were consistent with microgliosis and T-cell recruitment to the brain parenchyma. Changes in astrocyte populations were more equivocal although there was evidence of astrogliosis.Mechanistic investigations into responses to the Aldara model in inflammatory chemokine receptor (iCCR) KO mice did not reveal significant alterations in chemokine and cytokine transcription or in microglial responses to cutaneous Aldara treatment in the absence of the iCCRs (CCR1, CCR2, CCR3 and CCR5), but there did appear to be evidence of reduced CD3+ T-cell recruitment. In contrast, investigations in type I interferon receptor (IFNAR) KO mice identified a clear role for type I IFN signalling through IFNAR in the induction of chemokine and cytokine gene expression in the brain, and associated changes in Iba1+ microglial and CD3+ T-cell populations in response to cutaneous Aldara treatment. Mass spectrometric analysis of IMQ, the main active ingredient of Aldara, revealed that within four hours it enters both the circulation and the brain. The finding of IMQ within the brain parenchyma suggests that, while it is not an appropriate tool for studying peripheral-central immune crosstalk, it is a useful non-invasive model of TLR7 mediated neuroinflammation.These data provide compelling evidence of a role for chemokines in human depression and in neuro-inflammation, although the precise actions of this family of molecules remain unclear. In addition, building on previous work, the Aldara model appears to be a suitable tool for the study of neuro-inflammation, particularly interferon-driven immune responses, but is less appropriate for studying peripherally driven CNS immune reactions. Further work into the specific role of chemokines and associated cellular populations will hopefully provide additional insight into how CNS immune reactions are co-ordinated.