T-Lymphocytes Traffic into the Brain across the Blood-CSF Barrier: Evidence Using a Reconstituted Choroid Plexus Epithelium

Abstract

An emerging concept of normal brain immune surveillance proposes that recently and moderately activated central memory T lymphocytes enter the central nervous system (CNS) directly into the cerebrospinal fluid (CSF) via the choroid plexus. Within the CSF space, T cells inspect the CNS environment for cognate antigens. This gate of entry into the CNS could also prevail at the initial stage of neuroinflammatory processes. To actually demonstrate T cell migration across the choroidal epithelium forming the blood-CSF barrier, an in vitro model of the rat blood-CSF barrier was established in an "inverse" configuration that enables cell transmigration studies in the basolateral to apical, i.e. blood/stroma to CSF direction. Structural barrier features were evaluated by immunocytochemical analysis of tight junction proteins, functional barrier properties were assessed by measuring the monolayer permeability to sucrose and the active efflux transport of organic anions. The migratory behaviour of activated T cells across the choroidal epithelium was analysed in the presence and absence of chemokines. The migration pathway was examined by confocal microscopy. The inverse rat BCSFB model reproduces the continuous distribution of tight junction proteins at cell margins, the restricted paracellular permeability, and polarized active transport mechanisms, which all contribute to the barrier phenotype in vivo. Using this model, we present experimental evidence of T cell migration across the choroidal epithelium. Cell migration appears to occur via a paracellular route without disrupting the restrictive barrier properties of the epithelial interface. Apical chemokine addition strongly stimulates T cell migration across the choroidal epithelium. The present data provide evidence for the controlled migration of T cells across the blood-CSF barrier into brain. They further indicate that this recruitment route is sensitive to CSF-borne chemokines, extending the relevance of this migration pathway to neuroinflammatory and neuroinfectious disorders which are typified by elevated chemokine levels in CSF.

Fig 1. The cellular model of the blood-CSF barrier.

(A) Schematic representation of a choroidal villus and of the experimental set up illustrating the two-chamber culture device. Left: the choroidal epithelium which forms the actual tight barrier controlling access into the CSF [18] delimits a stroma, in which the fenestrated vessels lacking a typical blood-brain barrier phenotype, express P-selectin [4]. Right: The epithelial cell monolayer grown on the lower side of the filter separates the upper chamber corresponding to the blood/stromal or basolateral space, from the bottom chamber representing the CSF or apical compartment (adapted from [18,41]. (B and C) Immunofluorescent staining of occludin (B) and claudin 1/3 (C) showing a typical intercellular distribution of the tight junction proteins in the confluent inverted monolayers of choroidal epithelial cells. (D) Immunofluorescent staining of Na⁺K⁺ ATPase and ABCC1, showing the expected respective apical and basolateral membrane localization in the choroidal epithelial cells. The left image is a close up of a single cell to better appreciate the polarity of distribution of the 2 proteins. Nuclei appear in blue. Arrows show the lateral cellular membranes best seen in the z direction by confocal analysis, arrowheads show the basal labeling of ABCC1.

Fig 2. Schematic representation of the Transwell migration assay.

Fluorescently labeled lymphocytes (10⁶ cells/ml) are added into the upper chamber, in the presence or absence of chemokines in the lower chamber. Counting of cells retrieved from the lower chamber enables to determine the percentage of lymphocytes that fully migated through the choroidal epithelial cells. Staining protocols of the cell covered filters following the migration study allow defining the localization of lymphocytes within the filter/epithelium system, and approaching the route of transmigration.

Fig 3. Distribution of claudins and activated T cell localization following transmigration assay across the choroidal epithelium.

T cell adhesion to and migration across the epithelium do not cause any alteration in the continuous peripheral distribution of epithelial tight junction proteins claudin 1/3 and 2. Stacks of optical Z-sections obtained by confocal microscopy across the whole filter + epithelial layer allowed visualizing all T cells associated with the system. Their relative position with respect to the tight junctions revealed by claudin1/3 (A, B, D, G) or claudin 2 (C, E, F) immunocytochemistry was determined in xz or yz views and used to discriminate transmigrated immune cells adhering to the apical membrane from those still located basolaterally or remaining in the upper chamber (example in B). Activated T cells contacting either the basolateral or the apical membrane domain of the epithelium are often located at tricellular corners (A, C, E, F) hinting at a paracellular route of T cell migration across the monolayer. Most T cells present a non-homogeneous fluorescence intensity or a typical comma-like shape with a uropod and a cellular protrusion that could be a leading edge (stars in A, D), which reflect their motility. In G, the T cell “a” extends a long cytoplasmic projection seemingly scanning the intercellular junction towards the tricellular corner.

Fig 4. Activated T cell migration across the choroidal epithelium.

(A) Transepithelial chemotaxis of PHA-activated T cells in response to CCL2, CCL5 and CXCL10. **: statistically different from migration without chemokines p<0.01, two-tail Student's t test for unequal variance. The migration indices, measured after a 7-hour period, represent means ± SD, n = 3 or 4. (B) T cell chemotaxis does not alter the paracellular gate function of the blood-CSF barrier. Epithelial permeability coefficients for the paracellular marker sucrose are calculated and expressed in cm.min^-1 [21]. ns: not statistically different from control group (without T cells) p<0.05, one-way ANOVA followed by ‘a posteriori’ Dunnett’s test. Data represent means ± SD, n = 3.

Fig 5. Confocal analysis of T cell transmigration route.

Successive optical Z-sections through the filter + epithelial cell system are represented in an apical to basal order. (A) The most apical confocal plane is at the level of the tight junction protein claudin 2. The T cell, barely visible in this section, emerges on the following plane at a tricellular corner (white arrowhead). The leading edge is then visualized in section 3, followed by the cell body in more basal focal planes. A second, presumably transmigrating T cell similarly appears at another tricellular corner (Section 4, white arrow). (B) The T cell visible in the first confocal plane (white arrowhead) has migrated through the epithelial monolayer. It is located apically relative to the sections displaying the tight junction protein claudin 2 (visible from section 3), and associated with a tricellular corner (Section 5), through which it may have migrated. A second T cell (Section 7, white arrow) is visible in the second half of the plane series. It is mostly present in sections that are basal to the tight junction network.