Leukocyte trafficking to sites of inflammation follows a defined temporal pattern, and evidence suggests that initial neutrophil transendothelial migration modifies endothelial cell phenotype. We tested the hypothesis that preconditioning of human umbilical vein endothelial cells (HUVEC) by neutrophils would also modify the subsequent transendothelial migration of T lymphocytes across cytokine-stimulated HUVEC in an in vitro flow assay. Using fluorescence microscopy, preconditioning of HUVEC by neutrophils was observed to significantly reduce the extent of subsequent stromal cell-derived factor-1 ␣ (SDF-1 ␣ [CXCL12])-mediated T lymphocyte transendothelial migration, without reducing accumulation. In contrast, recruitment of a second wave of neutrophils was unaltered. Conditioned medium harvested after transendothelial migration of neutrophils or supernatants from stimulated neutrophils mediated a similar blocking effect, which was negated using a specific neutrophil elastase inhibitor. Furthermore, T lymphocyte transendothelial migration was inhibited by treatment of HUVEC with purified neutrophil elastase, which selectively cleaved the amino terminus of HUVEC-bound SDF-1 ␣ , which is required for its chemotactic activity. The reduction in T lymphocyte transendothelial migration was not observed using a different chemokine, ELC (CCL19), and was not reversed by replenishment of SDF-1 ␣ , indicating endothelial retention of the inactivated chemokine. In summary, transmigrating neutrophils secrete localized elastase that is protected from plasma inhibitors, and thereby modulate trafficking of other leukocyte subsets by altering the endothelial-associated chemotactic activities.

Endothelial cell junctional adhesion molecule (JAM)-C has been proposed to regulate neutrophil migration. In the current study, we used function-blocking mAbs against human JAM-C to determine its role in human leukocyte adhesion and transendothelial cell migration under flow conditions. JAM-C surface expression in HUVEC was uniformly low, and treatment with inflammatory cytokines TNF-α, IL-1β, or LPS did not increase its surface expression as assessed by FACS analysis. By immunofluorescence microscopy, JAM-C staining showed sparse localization to cell-cell junctions on resting or cytokine-activated HUVEC. Surprisingly, staining of detergent-permeabilized HUVEC revealed a large intracellular pool of JAM-C that showed little colocalization with von Willebrand factor. Adhesion studies in an in vitro flow model showed that functional blocking JAM-C mAb alone had no inhibitory effect on polymorphonuclear leukocyte (PMN) adhesion or transmigration, whereas mAb to ICAM-1 significantly reduced transmigration. Interestingly, JAM-C-blocking mAbs synergized with a combination of PECAM-1, ICAM-1, and CD99-blocking mAbs to inhibit PMN transmigration. Overexpression of JAM-C by infection with a lentivirus JAM-C GFP fusion protein did not increase adhesion or extent of transmigration of PMN or evoke a role for JAM-C in transendothelial migration. These data suggest that JAM-C has a minimal role, if any, in PMN transmigration in this model and that ICAM-1 is the preferred endothelial-expressed ligand for PMN β2 integrins during transendothelial migration.

Invariant Natural Killer T (iNKT) cells are a T cell subset expressing an invariant T Cell Receptor (TCR) that recognizes glycolipid antigens rather than peptides. The cells have both innate-like rapid cytokine release, and adaptive-like thymic positive selection. iNKT cell activation has been implicated in the pathogenesis of allergic asthma and inflammatory diseases, while reduced iNKT cell activation promotes infectious disease, cancer and certain autoimmune diseases such as Type 1 diabetes (T1D). Therapeutic means to reduce or deplete iNKT cells could treat inflammatory diseases, while approaches to promote their activation may have potential in certain infectious diseases, cancer or autoimmunity. Thus, we developed invariant TCR-specific monoclonal antibodies to better understand the role of iNKT cells in disease. We report here the first monoclonal antibodies specific for the mouse invariant TCR that by modifying the Fc construct can specifically deplete or activate iNKT cells in vivo in otherwise fully immuno-competent animals. We have used both the depleting and activating version of the antibody in the NOD model of T1D. As demonstrated previously using genetically iNKT cell deficient NOD mice, and in studies of glycolipid antigen activated iNKT cells in standard NOD mice, we found that antibody mediated depletion or activation of iNKT cells respectively accelerated and retarded T1D onset. In BALB/c mice, ovalbumin (OVA) mediated airway hyper-reactivity (AHR) was abrogated with iNKT cell depletion prior to OVA sensitization, confirming studies in knockout mice. Depletion of iNKT cells after sensitization had no effect on AHR in the conducting airways but did reduce AHR in the lung periphery. This result raises caution in the interpretation of studies that use animals that are genetically iNKT cell deficient from birth. These activating and depleting antibodies provide a novel tool to assess the therapeutic potential of iNKT cell manipulation.

Type 1 diabetes development in the NOD mouse model is widely reported to be dependent on high-level production by autoreactive CD4+ and CD8+ T cells of interferon-γ (IFN-γ), generally considered a proinflammatory cytokine. However, IFN-γ can also participate in tolerance-induction pathways, indicating it is not solely proinflammatory. This study addresses how IFN-γ can suppress activation of diabetogenic CD8+ T cells. CD8+ T cells transgenically expressing the diabetogenic AI4 T-cell receptor adoptively transferred disease to otherwise unmanipulated NOD.IFN-γnull, but not standard NOD, mice. AI4 T cells only underwent vigorous intrasplenic proliferation in NOD.IFN-γnull recipients. Disease-protective IFN-γ could be derived from any lymphocyte source and suppressed diabetogenic CD8+ T-cell responses both directly and through an intermediary nonlymphoid cell population. Suppression was not dependent on regulatory T cells, but was associated with increased inhibitory STAT1 to STAT4 expression levels in pathogenic AI4 T cells. Importantly, IFN-γ exposure during activation reduced the cytotoxicity of human-origin type 1 diabetes–relevant autoreactive CD8+ T cells. Collectively, these results indicate that rather than marking the most proinflammatory lymphocytes in diabetes development, IFN-γ production could represent an attempted limitation of pathogenic CD8+ T-cell activation. Thus, great care should be taken when designing possible diabetic intervention approaches modulating IFN-γ production.

Self-reactive T cells must escape thymic negative selection to mediate pathogenic autoimmunity. In the NOD mouse model of autoimmune diabetes, several β cell-cytotoxic CD8 T cell populations are known, with the most aggressive of these represented by AI4, a T cell clone with promiscuous antigen recognition characteristics. We have identified a long-elusive β cell-specific ligand for AI4 as an unusually short H-2Db-binding 7-mer peptide lacking a C-terminal anchor residue and derived from the insulin A chain (InsA14-20). Crystallography reveals that compensatory mechanisms permit peptides lacking a C-terminal anchor to bind sufficiently to the MHC to enable destructive T cell responses, yet allow cognate T cells to avoid negative selection. InsA14-20 shares two solvent-exposed residues with previously identified AI4 ligands, providing a structural explanation for AI4’s promiscuity. Detection of AI4-like T cells, using mimotopes of InsA14-20 with improved H-2Db binding characteristics, establishes the AI4-like T cell population as a consistent feature of the islet infiltrates of NOD mice. Our work establishes undersized peptides as previously unrecognized targets of autoreactive CD8 T cells and presents a strategy for their further exploration as antigens in autoimmune disease.

SUMMARY Androgen deficiency in men increases body fat, but the mechanisms by which testosterone suppresses fat deposition have not been elucidated fully. Adipose tissue macrophages express the androgen receptor (AR) and regulate adipose tissue remodeling. Thus, testosterone signaling in macrophages could alter the paracrine function of these cells and thereby contribute to the metabolic effects of androgens in men. A metabolic phenotyping study was performed to determine whether the loss of AR signaling in hematopoietic cells results in greater fat accumulation in male mice. C57BL/6J male mice (ages 12–14 weeks) underwent bone marrow transplant from either wild-type (WT) or AR knockout (ARKO) donors (n = 11–13 per group). Mice were fed a high-fat diet (60% fat) for 16 weeks. At baseline, 8 and 16 weeks, glucose and insulin tolerance tests were performed, and body composition was analyzed with fat-water imaging by MRI. No differences in body weight were observed between mice transplanted with WT bone marrow [WT (WTbm)] or ARKO bone marrow [WT(ARKObm)] prior to initiation of the high-fat diet. After 8 weeks of high-fat feeding, WT(ARKObm) mice exhibited significantly more visceral and total fat mass than WT(WTbm) animals. Despite this, no differences between groups were observed in glucose tolerance, insulin sensitivity, or plasma concentrations of insulin, glucose, leptin, or cholesterol, although WT(ARKObm) mice had higher plasma levels of adiponectin. Resultant data indicate that AR signaling in hematopoietic cells influences body fat distribution in male mice, and the absence of hematopoietic AR plays a permissive role in visceral fat accumulation. These findings demonstrate a metabolic role for AR signaling in marrow-derived cells and suggest a novel mechanism by which androgen deficiency in men might promote increased adiposity. The relative contributions of AR signaling in macrophages and other marrow-derived cells require further investigation.

Vitamin D exerts important regulatory effects on the endocrine and immune systems. Autoimmune type 1 diabetes (T1D) development in the inbred NOD mouse strain can be accelerated by vitamin D insufficiency or suppressed by chronic treatment with high levels of 1α,25-dihydroxyvitamin D(3). Consequently, a report that T1D development was unaffected in NOD mice genetically lacking the vitamin D receptor (VDR) was unexpected. To further assess this result, the mutant stock was imported to The Jackson Laboratory, backcrossed once to NOD/ShiLtJ, and progeny rederived through embryo transfer. VDR-deficient NOD mice of both sexes showed significant acceleration of T1D. This acceleration was not associated with alterations in immune cells targeting pancreatic β-cells. Rather, the capacity of β-cells to produce and/or secrete insulin was severely impaired by the hypocalcaemia developing in VDR-deficient NOD mice fed a standard rodent chow diet. Feeding a high-lactose calcium rescue diet that circumvents a VDR requirement for calcium absorption from the intestine normalized serum calcium levels, restored β-cell insulin secretion, corrected glucose intolerance, and eliminated accelerated T1D in VDR-deficient NOD mice. These findings suggest that calcium and/or vitamin D supplementation may improve disease outcomes in some T1D-prone individuals that are calcium deficient.

In both NOD mice and humans, the development of type 1 diabetes (T1D) is dependent in part on autoreactive CD8 T cells recognizing pancreatic β cell peptides presented by often quite common MHC class I variants. Studies in NOD mice previously revealed that the common H2-K and/or H2-D class I molecules expressed by this strain aberrantly lose the ability to mediate the thymic deletion of pathogenic CD8 T cell responses through interactions with T1D susceptibility genes outside the MHC. A gene(s) mapping to proximal chromosome 7 was previously shown to be an important contributor to the failure of the common class I molecules expressed by NOD mice to mediate the normal thymic negative selection of diabetogenic CD8 T cells. Using an inducible model of thymic negative selection and mRNA transcript analyses, we initially identified an elevated expression variant as a likely NOD-proximal chromosome 7 region gene contributing to impaired thymic deletion of diabetogenic CD8 T cells. CRISPR/Cas9-mediated genetic attenuation of expression in NOD mice resulted in improved negative selection of autoreactive diabetogenic AI4 and NY8.3 CD8 T cells. These results indicated that allelic variants of contribute to the efficiency of intrathymic deletion of diabetogenic CD8 T cells. However, although enhancing thymic deletion of pathogenic CD8 T cells, ablating expression surprisingly accelerated T1D onset that was associated with numeric decreases in both regulatory T and B lymphocytes in NOD mice.