Numerous neurological disorders are characterized by central nervous system (CNS) vascular permeability. However, the underlying contribution of inflammatory-derived factors leading to pathology associated with blood-brain barrier (BBB) disruption remains poorly understood. In order to address this, we developed an inducible model of BBB disruption using a variation of the Theiler's murine encephalomyelitis virus (TMEV) model of multiple sclerosis. This peptide induced fatal syndrome (PIFS) model is initiated by virus-specific CD8 T cells and results in severe CNS vascular permeability and death in the C57BL/6 mouse strain. While perforin is required for BBB disruption, the cellular source of perforin has remained unidentified. In addition to CD8 T cells, various innate immune cells also express perforin and therefore could also contribute to BBB disruption. To investigate this, we isolated the CD8 T cell as the sole perforin-expressing cell type in the PIFS model through adoptive transfer techniques. We determined that C57BL/6 perforin−/− mice reconstituted with perforin competent CD8 T cells and induced to undergo PIFS exhibited: 1) heightened CNS vascular permeability, 2) increased astrocyte activation as measured by GFAP expression, and 3) loss of linear organization of BBB tight junction proteins claudin-5 and occludin in areas of CNS vascular permeability when compared to mock-treated controls. These results are consistent with the characteristics associated with PIFS in perforin competent mice. Therefore, CD8 T cells are sufficient as a sole perforin-expressing cell type to cause BBB disruption in the PIFS model.

BackgroundCD8 T cell-mediated blood-brain barrier (BBB) disruption is dependent on the effector molecule perforin. Human perforin has extensive single nucleotide variants (SNVs), the significance of which is not fully understood. These SNVs can result in reduced, but not ablated, perforin activity or expression. However, complete loss of perforin expression or activity results in the lethal disease familial hemophagocytic lymphohistiocytosis type 2 (FHL 2). In this study, we address the hypothesis that a single perforin allele can alter the severity of BBB disruption in vivo using a well-established model of CNS vascular permeability in C57Bl/6 mice. The results of this study provide insight into the significance of perforin SNVs in the human population.MethodsWe isolated the effect a single perforin allele has on CNS vascular permeability through the use of perforin-heterozygous (perforin+/−) C57BL/6 mice in the peptide-induced fatal syndrome (PIFS) model of immune-mediated BBB disruption. Seven days following Theiler’s murine encephalomyelitis virus (TMEV) CNS infection, neuroinflammation and TMEV viral control were assessed through flow cytometric analysis and quantitative real-time PCR of the viral genome, respectively. Following immune-mediated BBB disruption, gadolinium-enhanced T1-weighted MRI, with 3D volumetric analysis, and confocal microscopy were used to define CNS vascular permeability. Finally, the open field behavior test was used to assess locomotor activity of mice following immune-mediated BBB disruption.ResultsPerforin-null mice had negligible CNS vascular permeability. Perforin-WT mice have extensive CNS vascular permeability. Interestingly, perforin-heterozygous mice had an intermediate level of CNS vascular permeability as measured by both gadolinium-enhanced T1-weighted MRI and fibrinogen leakage in the brain parenchyma. Differences in BBB disruption were not a result of increased CNS immune infiltrate. Additionally, TMEV was controlled in a perforin dose-dependent manner. Furthermore, a single perforin allele is sufficient to induce locomotor deficit during immune-mediated BBB disruption.ConclusionsPerforin modulates BBB disruption in a dose-dependent manner. This study demonstrates a potentially advantageous role for decreased perforin expression in reducing BBB disruption. This study also provides insight into the effect SNVs in a single perforin allele could have on functional deficit in neurological disease.

Many viruses utilize cell adhesion molecules of the immunoglobulin superfamily as receptors. In particular, viruses of different classes exploit nectins. The large DNA viruses, herpes simplex and pseudorabies viruses, use ubiquitous nectins 1 and 2. The negative-strand RNA virus measles virus (MeV) uses tissue-specific nectin-4, and the positive-strand RNA virus poliovirus uses nectin-like 5 (necl-5), also known as poliovirus receptor. These viruses contact the BC, C=C؆, and FG loops on the upper tip of their receptor's most membrane-distal domain. This location corresponds to the newly defined canonical adhesive interface of nectins, but how viruses utilize this interface has remained unclear. Here we show that the same key residues in the BC and FG loops of nectin-4 govern binding to the MeV attachment protein hemagglutinin (H) and cell entry, nectin-4 homodimerization, and heterodimerization with nectin-1. On the other hand, residues in the C=C؆ loop necessary for homo-and heterotypic interactions are dispensable for MeV-induced fusion and cell entry. Remarkably, the C=C؆ loop governs dissociation of the nectin-4 and H ectodomains. We provide formal proof that H can interfere with the formation of stable nectin-1/nectin-4 heterodimers. Finally, while developing an alternative model to study MeV spread, we observed that polarized primary pig airway epithelial sheets cannot be infected. We show that a single amino acid variant in the BC loop of pig nectin-4 fully accounts for restricted MeV entry. Thus, the three loops forming the adhesive interface of nectin-4 have different roles in supporting MeV H association and dissociation and MeV-induced fusion. IMPORTANCEDifferent viruses utilize nectins as receptors. Nectins are immunoglobulin superfamily glycoproteins that mediate cell-cell adhesion in vertebrate tissues. They interact through an adhesive interface located at the top of their membrane-distal domain. How viruses utilize the three loops forming this interface has remained unclear. We demonstrate that while nectin-nectin interactions require residues in all three loops, the association of nectin-4 with the measles virus hemagglutinin requires only the BC and FG loops. However, we discovered that residues in the C=C؆ loop modulate the dissociation of nectin-4 from the viral hemagglutinin. Analogous mechanisms may support cell entry of other viruses that utilize nectins or other cell adhesion molecules of the immunoglobulin superfamily as receptors.