A20 regulates lymphocyte adhesion in murine neuroinflammation by restricting endothelial ICOSL expression in the CNS
Lisa Johann, Sasha Soldati, Kristin Müller, Josephine Lampe, Federico Marini, Matthias Klein, Eva Schramm, Nathalie Ries, Carsten Schelmbauer, Ilaria Palagi, Khalad Karram, Julian C. Assmann, Mahtab A. Khan, Jan Wenzel, Mirko H.H. Schmidt, Jakob Körbelin, Dirk Schlüter, Geert van Loo, Tobias Bopp, Britta Engelhardt, Markus Schwaninger, Ari Waisman
Lisa Johann, Sasha Soldati, Kristin Müller, Josephine Lampe, Federico Marini, Matthias Klein, Eva Schramm, Nathalie Ries, Carsten Schelmbauer, Ilaria Palagi, Khalad Karram, Julian C. Assmann, Mahtab A. Khan, Jan Wenzel, Mirko H.H. Schmidt, Jakob Körbelin, Dirk Schlüter, Geert van Loo, Tobias Bopp, Britta Engelhardt, Markus Schwaninger, Ari Waisman
View: Text | PDF
Research Article Autoimmunity Neuroscience

A20 regulates lymphocyte adhesion in murine neuroinflammation by restricting endothelial ICOSL expression in the CNS

Abstract

A20 is a ubiquitin-modifying protein that negatively regulates NF-κB signaling. Mutations in A20/TNFAIP3 are associated with a variety of autoimmune diseases, including multiple sclerosis (MS). We found that deletion of A20 in central nervous system (CNS) endothelial cells (ECs) enhances experimental autoimmune encephalomyelitis (EAE), a mouse model of MS. A20ΔCNS-EC mice showed increased numbers of CNS-infiltrating immune cells during neuroinflammation and in the steady state. While the integrity of the blood-brain barrier (BBB) was not impaired, we observed a strong activation of CNS-ECs in these mice, with dramatically increased levels of the adhesion molecules ICAM-1 and VCAM-1. We discovered ICOSL to be expressed by A20-deficient CNS-ECs, which we found to function as adhesion molecules. Silencing of ICOSL in CNS microvascular ECs partly reversed the phenotype of A20ΔCNS-EC mice without reaching statistical significance and delayed the onset of EAE symptoms in WT mice. In addition, blocking of ICOSL on primary mouse brain microvascular ECs impaired the adhesion of T cells in vitro. Taken together, we propose that CNS EC-ICOSL contributes to the firm adhesion of T cells to the BBB, promoting their entry into the CNS and eventually driving neuroinflammation.

Authors

Lisa Johann, Sasha Soldati, Kristin Müller, Josephine Lampe, Federico Marini, Matthias Klein, Eva Schramm, Nathalie Ries, Carsten Schelmbauer, Ilaria Palagi, Khalad Karram, Julian C. Assmann, Mahtab A. Khan, Jan Wenzel, Mirko H.H. Schmidt, Jakob Körbelin, Dirk Schlüter, Geert van Loo, Tobias Bopp, Britta Engelhardt, Markus Schwaninger, Ari Waisman

×

Figure 3

A20-deficiency in CNS-ECs does not impair BBB integrity.

Options: View larger image (or click on image) Download as PowerPoint
A20-deficiency in CNS-ECs does not impair BBB integrity. (A–C) Brain sec...
(A–C) Brain sections from A20ΔCNS-EC and A20fl/fl mice were stained for collagen IV (red) and CD31 (green) to determine vessel length and string vessel length. String vessels were identified as capillaries that have lost CD31-positive endothelial cells and only consist of the basement membrane protein collagen IV. Representative microscopic images are shown in (A). Scale bar: 50 μm. (B) Vessel length (in mm per mm2) and (C) string vessel length (%) normalized to A20fl/fl littermate controls were quantified (n = 5–6 mice per group). (D) Brain weight of A20ΔCNS-EC and A20fl/fl mice (n = 8 mice per group) shown as representative of 2 individual cohorts. (E and F) Brain sections from A20ΔCNS-EC and A20fl/fl mice were stained for occludin (red). Representative microscopic images are shown in (E). Scale bar: 50 μm. (F) Occludin intensity was quantified and is presented relative to A20fl/fl littermate control mice (n = 5–6 mice per group) as representative of 2 individual cohorts. (G and H) A20ΔCNS-EC and A20fl/fl mice were injected with 2 mM 3-5 kDa FITC-dextran in PBS i.p. After 15 minutes, SC and brain were isolated and homogenized in PBS. Fluorescence was measured in the supernatant and raw fluorescence units (RFU) were normalized to tissue weight. PBS-injected mice were used as sham controls; mice at the peak of an active EAE (scores between 1.5 and 3.5) were used as positive controls. Normalized RFUs are shown for SC (G) and brain (H) relative to A20fl/fl controls (SC: n = 5–7; brain: n = 3–5 mice per group). Statistical significance was determined by 2-tailed unpaired Student’s t test (B–D and F) or ordinary 1-way ANOVA with Tukey’s multiple comparisons test (G and H). **P<0.01, ****P<0.0001.