Previous reports have characterized the immune infiltrate present in the brain during TE but there has been little insight into the behavior of these populations. The ability to visualize total endogenous T cell populations in vivo
in the brain during infection provided the impetus for the reductionist approach described above, allowing us to focus on the anti-parasite CD8+
T cell response. Recent studies have proposed a requirement for an antigen specific entry mechanism for CD8+
T cells to the brain (Galea et al., 2007
). The inability to detect significant numbers of peripherally activated OTI CD8+
T cells in the brains of mice infected with parental parasites is in agreement with this concept. However, our findings do not distinguish between a requirement for cognate antigen in the brain for retention versus antigen specific recruitment.
Based on previous models (Hunter and Remington, 1994
) there was the expectation that once T. gondii
- specific CD8+
T cells gained entry to the brain they would display directed migration towards areas of parasite replication where they would accumulate, stop and mediate effector functions. Indeed, endogenous and transferred T cells formed foci associated with infected cells, and imaging of OTIGFP
cells revealed clusters of stationary rounded cells. These latter behaviors are consistent with T cells interacting with either infected targets or cells involved in cross presentation of class I restricted antigens. Indeed, during TE there are multiple candidates that could be involved in these processes including dendritic cells, macrophages, microglia or astrocytes (Figure S9
). What was unexpected was the large number of T cells that were highly motile, whether imaging the endogenous populations or the transferred OTIGFP
cells. This highly motile population was most apparent when there were maximum effector cells in the brain and antigen load was reduced. These findings are consistent with a model in which antigen availability is a major determinant of T cell behavior at the population level. Alternatively, the observation of increasing PD-1 expression on the endogenous CD8+
T cells, and its delayed kinetics on the transferred OTIGFP
population, raises the possibility that the mechanisms that limit T cell activity during TE may underlie the reduced migration and velocity of chronically activated T cells.
The use of intra-vital imaging highlighted the close association of T cells with fibroblastic reticular cells and ECM structures in the LN and led to an appreciation that T cell motility in lymphoid organs is highly directed (Bajenoff et al., 2006
). There is also evidence that even in uninflamed tissues such as the liver or skin, which contain circulating populations of immune cells, similar scaffolds already exist (Friedl and Weigelin, 2008
; Yang et al., 2007
). This report, describes the presence of an analogous reticular system at inflammatory sites within the brain, that was not present in the normal brain tissues examined, but was associated with areas of parasite replication and local inflammation. At present the source of this network is unclear, but astrocytes have been implicated in the generation of collagen-like networks in the brain during glial scar formation (Heck et al., 2007
), consistent with the localization of activated astrocytes in these lesions. Relevant to this observation are reports of an ECM structure composed of proteoglycans and glycoproteins that guides axonal migration during neuronal development (Fitch and Silver, 2007
; Heck et al., 2007
; Tessier-Lavigne and Goodman, 1996
). It may be that the structures important during this developmental process may also be used to regulate lymphocyte migration in this immunologically privileged site.
In the lymph node, the presence of CCR7 ligands coating the ECM structure is a requirement for efficient T cell motility. The observation that in mice with TE there is increased expression of CCL21 with a fibrous appearance that is closely associated with CD8+ T cells supports the comparison with secondary lymphoid organs. Moreover, although the role of CCR7 and its ligands during T. gondii infection has not been fully elucidated, our studies have identified CD8+ T cells expressing CCR7 during TE and which are functionally responsive to CCL21 (unpublished observations). Taken together, these observations suggest that reticular fibers in the brain, analogous to the role of ECM in the lymph node, may provide a scaffold that supports T cell migration during inflammation. While further studies will be required to define the composition of these reticular networks, targeting these structures may provide therapeutic approaches to manage T cell mediated inflammatory conditions that affect the brain.