West Nile Virus spread and differential chemokine response in the central nervous system of mice: Role in pathogenic mechanisms of encephalitis.

Abstract

West Nile virus (WNV) is a zoonotic mosquito-borne flavivirus able to cause severe neurological disease in humans, horses and various avian species. The more severe pathological changes of neurotropic WNV infection are caused by virus neuroinvasion and/or the immunological response in the central nervous system (CNS). The extent in which inflammatory cell trafficking orchestrated by chemokines is involved in the pathogenesis of CNS lesions has not been entirely elucidated. To understand the sequence of pro-inflammatory chemokine induction during WNV encephalitis, a murine intranasal inoculation model was used. The relationship between lesional patterns in the mice CNS, the viral antigen distribution and the expression of pro-inflammatory chemokine (CCL2, CCL5 and CXCL10) were evaluated. Viral antigen was first observed in olfactory tract and pyriform cortex neurons, suggesting a retrograde neuronal infection from the olfactory nerve. A spatio-temporal association between WNV antigen and perivascular cuffs development was observed. Chemokine immunostaining was widely distributed in the brain from early stages. CCL2 immunolabelling was localised in neurons, astrocytes, microglia and endothelial cells as well as mononuclear leucocytes within perivascular cuffs. In contrast, CCL5 and CXCL10 immunostaining were mainly observed in astroglia and neurons, respectively. A strong correlation was demonstrated between the presence of perivascular cuffs and CCL2 and CCL5 expression in most brain areas, while CXCL10 was only associated with inflammatory lesions in few specific regions. Importantly, a strong correlation between WNV and CCL5 distribution was observed. However, no correlation was observed between CXCL10 and viral antigen. Neurons were confirmed as the main target cells of WNV, as well as one of the sources of CCL2, CCL5 and CXCL10. This study shows the sequence and comparative distribution pattern between histological lesions, WNV antigen and chemokine expression over the infection process. Furthermore, it identifies potential targets for immune intervention to suppress damaging chemokine responses.