Neuroinvasive Listeria monocytogenes infection triggers accumulation of brain CD8(+) tissue-resident memory T cells in a miR-155-dependent fashion

BACKGROUND: Brain inflammation is a key cause of cognitive decline after central nervous system (CNS) infections. A thorough understanding of immune responses to CNS infection is essential for developing anti-inflammatory interventions that improve outcomes. Tissue-resident memory T cells (T(RM)) are non-recirculating memory T cells that provide surveillance of previously infected tissues. However, in addition to protecting the brain against reinfection, brain T(RM) can contribute to post-infectious neuroinflammation. We hypothesized that accumulation of CD8(+) T(RM) in the brain could be reduced by inhibiting microRNA (miR)-155, a microRNA that influences development of cytotoxic CD8(+) T lymphocytes during infection. METHODS: C57BL/6J mice were infected by intraperitoneal injection with a lethal inoculum of Listeria monocytogenes (Lm) then treated with antibiotics. Flow cytometry was used to quantify specific populations of brain leukocytes 28–29 days (d) post-infection (p.i.). To test the degree to which miR-155 altered leukocyte influxes into the brain, infected mice were injected with a miR-155 inhibitor or locked nucleic acid (LNA) scramble control 2d, 4d, 6d, and 8d p.i. along with antibiotic treatment. Bacterial loads in spleen and liver and body weights were measured up to 7d p.i. Brain leukocytes were analyzed 14d and 28d p.i. Confirmatory studies were performed in mutated mice lacking miR-155 (miR-155(−/−)) RESULTS: Lm infection significantly increased the numbers of brain CD3(+)CD8(+) lymphocytes at 28d p.i. These cells were extravascular, and displayed markers characteristic of T(RM), with the predominant phenotype of CD44(+)CD62L(-)CD69(+)CX3CR1(−). Further analysis showed that > 75% of brain T(RM) also expressed CD49a, PD-1, Ly6C, CD103, and CD127. Mice injected with miR-155 inhibitor lost less weight through 7d p.i. than did control mice, whereas bacterial loads in brain, liver, and spleen were not different from controls. By 28d p.i., the numbers of brain CD8(+) T(RM) cells were significantly decreased in mice treated with the inhibitor compared with controls. Similarly, miR-155(−/−) mice showed significantly reduced numbers of brain CD8(+) T(RM) cells by 28d p.i. CONCLUSIONS: Brain CD8(+) T(RM) populations are established during neuroinvasive Lm infection. Accumulation of brain CD8(+) T(RM) cells is reduced by blocking miR-155 and in miR-155(−/−) mice, indicating that this molecule has a critical role in development of these specialized cells. Administering anti-miR-155 during infection could provide a novel avenue for reducing post-infectious neuroinflammation.