Neuroimmune interactions related to development of affective behavioural disturbances in neuropathic pain states

Abstract

Nerve damage leads to the development of disabling neuropathic pain in susceptible individuals, where patients present with pain as well as co-morbid behavioural changes, such as anhedonia, decreased motivation and depression. The pathophysiology of neuropathic pain remains unknown, however accumulating evidence suggests that neuroimmune interactions play a key role in its pathogenesis and development of co-morbid behavioural disturbances. Complex regional pain syndrome (CRPS) is a debilitating neuropathic disorder where trauma to a limb results in chronic pain. Mass cytometry (CyTOF) was used to systematically analyse circulating immune cells with a panel of 38 phenotypic and activation markers in the blood of CRPS patients and healthy controls. CyTOF revealed an expansion and increased activation of signalling pathways in several distinct populations of central memory CD8+ and CD4+ T lymphocytes. Regarding emotional state, CD8+ T lymphocytes were correlated with clinical scores for stress and CD4+ Th1 lymphocytes correlated with clinical scores for anxiety. There was also a reduction in circulating Dendritic cells (DC), indicative of DC tissue trafficking and potential involvement in lymphocyte activation. These data highlight a pathogenic role for T lymphocyte mediated chronic inflammation in CRPS and co-morbid behavioural disabilities. To further explore to role of neuroimmune interactions in the development of neuropathic pain and co-morbid behavioural changes, a rodent nerve injury model was utilized to evaluate whether individual differences in radial maze behaviour and neuroimmune interactions in the hippocampus (HP) and medial prefrontal cortex (mPFC) occurred in rats after sciatic nerve chronic constriction injury (CCI). CCI reduced mechanical withdrawal thresholds in all rats, whilst pellet-seeking behaviours were altered in some but not all rats. One group, termed ‘No effect’, had no behavioural changes compared to sham rats. Another group, termed ‘Acute effect’, had a temporary alteration to their exploration pattern, displaying more risk-assessment behaviour in the early phase post-injury. In a third group, termed ‘Lasting effect’, exploratory behaviours were remarkably different for the entire post-injury period, showing a withdrawal from pellet-seeking. Immunohistochemical analysis throughout the dorso-ventral axis of the HP revealed that the withdrawal from pellet-seeking observed in Lasting effect rats was concomitant with distinct glial-cytokine-neuronal adaptations within the contralateral ventral HP, including; increased expression of IL-1b and MCP-1; astrocyte atrophy and decreased area in the dentate gyrus (DG); reactive microglia and increased FosB/DFosB expression in the cornu ammonis (CA) subfield. These data highlight that glial-cytokine-neuronal adaptations in the ventral HP may mediate individual differences in radial maze behaviour following CCI. A follow up experiment explored whether pre-injury learning on the maze altered the effects of nerve injury on exploratory behaviour and spatial memory function. Whilst CCI again produced three distinct patterns of behaviour on the radial maze, Acute effect rats had improved working spatial memory outcomes after CCI. This indicates that the increased risk-assessment behaviours employed by Acute effect rats after injury may be considered advantageous when pellet-seeking, as it reduces unnecessary exploration during reward-seeking. The behavioural disruptions observed in Lasting effect rats were accompanied by neuroimmune activation within the contralateral ventral HP and mPFC. Multiplex immunoassay analysis revealed an increase in IL-1b, IL-6 and MCP-1 within the contralateral mPFC and ventral HP. Detailed immunohistochemical analysis of the mPFC and HP revealed an increased expression of IL-6, increased phospho-p38 MAPK expression in neurons and microglia, and a shift to a reactive microglial morphology in the caudal prelimbic and infralimbic cortex, ventral CA1 and DG. There was also a reduction in astrocyte cell size and BDNF expression in the contralateral ventral DG. These data provide further evidence that neuroinflammation in the mPFC and ventral HP may influence individual differences in radial maze behaviour following CCI. Collectively, these data provide evidence that individual differences in circulating immune cell activation and neuroimmune signature in the interconnected ventral HP-mPFC circuitry may play a significant role in the divergent behavioural trajectories in the neuropathic pain state, contributing to co-morbid behavioural changes in susceptible individuals.