Cellular Responses of the Retina to West Nile Virus Infection

Abstract

Age-related macular degeneration (AMD) is the leading cause of blindness in the developing world in people aged over 60 years, manifested as a loss of central vision in one or both eyes, with significant morbidity including loss of mobility and depression. This condition involves the degeneration of the macula, and although the exact aetiology of this disease is unknown, various epidemiological studies have shown it to be multifactorial. Current research points towards the involvement of a dysregulated immune system in the pathogenesis and progression of the disease: as the body ages, the immune system increasingly adopts a more inflammatory basal state. However, not all of the aged population develops AMD and it is highly likely that an additional stimulus or stimuli is/are needed to exploit this dysregulated immune environment to initiate this disease. Given the range of pathogens that can infect the retina, we hypothesize that this breaking point could manifest as a chronic inflammation as a result of a low-level infection. West Nile Virus (WNV) is a flavivirus that has come into international prominence ever since its spread into previously WNV-free regions following the 1999 New York outbreak. As several case reports have shown that WNV is capable of infecting the retina, and given its immunopathogenic properties, we believe the virus is a useful tool to model key immune pathways and responses that may be involved in the development and progression of AMD. Of significant interest are the processes involved in the breakdown of the outer blood-retinal barrier (BRB), which is an important step in the progression of AMD from an early stage to a more severe one. Additionally, deciphering and understanding the profile and populations of leukocytes that are recruited during an immunopathic infection in an organ regarded as being immunoprivileged is of great appeal. With this in mind, we set out to investigate the effects of WNV infection on the retinal pigment epithelium (RPE), which comprises the outer BRB. Previously, our laboratory established the WNV BRB model by quantitating various parameters, such as level of infectivity, viral output by WNV-infected RPE and effects of WNV infection on RPE proliferation/migration. The effect of WNV on the extracellular matrix (ECM) production by RPE was also investigated and increases in collagen I, IV and fibronectin were noted. Global ECM production induced a lowered rate of proliferation of RPE seeded on WNV-infected RPE ECM as opposed to mock-infected ECM. A full genome microarray was also undertaken on WNV-infected RPE to analyse differentially regulated gene mRNA production, and increases in several immune genes, as well as genes involved in the stress-response pathway and the TGFβ pathway were found. This current investigation expanded upon these results, and found that WNV infection produces a predominantly CCL5 chemokine response rather than a CCL2 response. Additionally, a lack of TNF production was noted, despite a high initial upregulation of the TNF gene in WNV-infected cells. WNV attenuation was found to be predominantly IFNβ-1-driven, while induction of indoleamine 2,3 dioxygenase activity was induced in part by IFNλ-1 and -2. The effects of WNV infection on RPE barrier integrity was investigated, and an initial increase in infected cells of barrier integrity was observed. Several investigations resulted in a conclusion of a soluble-mediator as the likely mechanism behind this initial increase, and while none of the chemokines tested appeared to contribute to this change, the results suggest that it may be TLR3/RIG-I independent. Finally, establishment of a murine WNV intravitreal model was also undertaken, and several key parameters were determined, including confirmation of WNV-infection of the murine retina, effect of WNV titre on mortality, and histological analysis of the effects of WNV infection on the murine retina. Quantification of the leukocyte profile recruited into the WNV-infected murine retina and choroid revealed significant increases in inflammatory Ly6Chi monocytes, as well as significant differences between immune mice and naïve mice intravitreally infected with WNV, and differences between 2 month old and 5 month old mice. Collectively, these results highlight the importance of the interferon response in both direct and indirect anti-WNV activities and immunomodulation, the changes in outer BRB integrity and possible contributors to its degradation, and the establishment of the murine intravitreal WNV model along with identification of several key leukocytes that are recruited at the peak of infection. These results will help guide further research and highlight possible immune pathways that may contribute to dysregulated inflammatory processes that may occur during the pathogenesis of AMD.