Y EC has previously been hampered by the requirement for MHC-matched EC and T cells. Some studies using MHC matched donors support the model that cultured human EC are able to present antigen and activated CD4+ T cells [9?1]. Moreover, mouse T cell clones or T cells from TCR-transgenic mice can be stimulated to proliferate in a peptide-antigen-specific manner by co-culture with MHCmatched ECs and the relevant protein antigen [14,34]. Additionally, as presented in this study with our HBEC line, co-cultures ofMHC-mismatched EC and T cells result in the activation of CD4+ and CD8+ T cells demonstrating that EC are able to present 11089-65-9 biological activity alloantigens [15,16]. In this study we have used a widely accepted assay of allogenic T cell stimulation without well characterised antigens in order to prepare for future experiments that will involve defined malarial antigens. In this assay, the separation of HBEC and T cells resulted in reduced T cell proliferation, indicating the role of cell-cell contact in this phenomenon. The costimulatory molecules CD40 and ICOSL are likely to be mediating this effect. ICOSL, a B-7 co-stimulatory UKI-1 family member was upregulated on HBECs following cytokine stimulation. Moreover, ICOSL has been shown previously to be a major costimulator in Human umbilical vein EC-mediated T cell activation, particularly in the re-activation of effector/memory T cells [12,26]. Another co-stimulatory molecule, CD40, was constitutively expressed on HBEC and upregulated after IFNc stimulation (Fig. 1). CD40 regulates the adhesion of CD4+ T cells to brain endothelium via the interaction with its ligand, CD40L on T cells, suggesting a potential mechanism by which activated CD40L expressing T cells could enhance adhesion and migration of inflammatory cells across the BBB to sites of inflammation in the human central nervous system [23]. This increase in HBEC MHC II expression has relevance for CM pathogenesis as MHC II expression 
on isolated mouse brain EC has been associated with the genetic susceptibility to CM [35]. Moreover, more recently the HLA ligand, HLA-C1 along with its cognate natural killer (NK) cell immunoglobulin-like receptor were shown to be significantly associated with the development of CMBrain Endothelium and T Cell Proliferationin humans [36]. EC, at least from lymph nodes, can be modulators of immune responses as they express multiple peripheral tissue antigens, independent of the autoimmune regulator, AIRE [37], and can even induce anergy [38]. This, together with our observation of malarial antigen transfer to brain EC surfaces [3], opens more possibilities for endothelial-mediated immunopathological mechanisms in CM. The findings described here are not only a major interest for understanding CM pathogenesis but also other neuroinfections involving disruption of endothelial cell barriers such as neurocysticercosis and toxoplasmosis [39,40]. In summary, we 
have shown that human brain endothelium cells express molecules important for T cell stimulation and activation including CD40, MHC II and ICOSL. They readily can take up fluorescently labeled antigens via clathrin-coated pits and macropinocytosis. Moreover, these cells are able to bind to and promote the proliferation of allogeneic T cells in vitro. Data presented here supports the hypothesis that HBEC are able to act as APC. This is particularly pertinent in neuroinfections such as CM where the diameter of microvessels is smaller than the size of lymphocytes; the lymphocyt.Y EC has previously been hampered by the requirement for MHC-matched EC and T cells. Some studies using MHC matched donors support the model that cultured human EC are able to present antigen and activated CD4+ T cells [9?1]. Moreover, mouse T cell clones or T cells from TCR-transgenic mice can be stimulated to proliferate in a peptide-antigen-specific manner by co-culture with MHCmatched ECs and the relevant protein antigen [14,34]. Additionally, as presented in this study with our HBEC line, co-cultures ofMHC-mismatched EC and T cells result in the activation of CD4+ and CD8+ T cells demonstrating that EC are able to present alloantigens [15,16]. In this study we have used a widely accepted assay of allogenic T cell stimulation without well characterised antigens in order to prepare for future experiments that will involve defined malarial antigens. In this assay, the separation of HBEC and T cells resulted in reduced T cell proliferation, indicating the role of cell-cell contact in this phenomenon. The costimulatory molecules CD40 and ICOSL are likely to be mediating this effect. ICOSL, a B-7 co-stimulatory family member was upregulated on HBECs following cytokine stimulation. Moreover, ICOSL has been shown previously to be a major costimulator in Human umbilical vein EC-mediated T cell activation, particularly in the re-activation of effector/memory T cells [12,26]. Another co-stimulatory molecule, CD40, was constitutively expressed on HBEC and upregulated after IFNc stimulation (Fig. 1). CD40 regulates the adhesion of CD4+ T cells to brain endothelium via the interaction with its ligand, CD40L on T cells, suggesting a potential mechanism by which activated CD40L expressing T cells could enhance adhesion and migration of inflammatory cells across the BBB to sites of inflammation in the human central nervous system [23]. This increase in HBEC MHC II expression has relevance for CM pathogenesis as MHC II expression on isolated mouse brain EC has been associated with the genetic susceptibility to CM [35]. Moreover, more recently the HLA ligand, HLA-C1 along with its cognate natural killer (NK) cell immunoglobulin-like receptor were shown to be significantly associated with the development of CMBrain Endothelium and T Cell Proliferationin humans [36]. EC, at least from lymph nodes, can be modulators of immune responses as they express multiple peripheral tissue antigens, independent of the autoimmune regulator, AIRE [37], and can even induce anergy [38]. This, together with our observation of malarial antigen transfer to brain EC surfaces [3], opens more possibilities for endothelial-mediated immunopathological mechanisms in CM. The findings described here are not only a major interest for understanding CM pathogenesis but also other neuroinfections involving disruption of endothelial cell barriers such as neurocysticercosis and toxoplasmosis [39,40]. In summary, we have shown that human brain endothelium cells express molecules important for T cell stimulation and activation including CD40, MHC II and ICOSL. They readily can take up fluorescently labeled antigens via clathrin-coated pits and macropinocytosis. Moreover, these cells are able to bind to and promote the proliferation of allogeneic T cells in vitro. Data presented here supports the hypothesis that HBEC are able to act as APC. This is particularly pertinent in neuroinfections such as CM where the diameter of microvessels is smaller than the size of lymphocytes; the lymphocyt.
