Antigen dependently activated cluster of differentiation 8-positive T cells cause perforin-mediated neurotoxicity in experimental stroke.

Neuroinflammation plays a key role in secondary brain damage after stroke. Although deleterious effects of proinflammatory cytokines are well characterized, direct cytotoxic effects of invading immune cells on the ischemic brain and the importance of their antigen-dependent activation are essentially unknown. Here we examined the effects of adaptive and innate immune cells-cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells-that share the direct perforin-mediated cytotoxic pathway on outcome after cerebral ischemia in mice. Although CTLs and NK cells both invaded the ischemic brain, only brain-infiltrating CTLs but not NK cells were more activated than their splenic counterparts. Depletion of CTLs decreased infarct volumes and behavioral deficit in two ischemia models, whereas NK cell depletion had no effect. Correspondingly, adoptive CTL transfer from wild-type into Rag1 knock-out mice increased infarct size. Adoptive CTL transfer from perforin knock-out or interferon-γ knock-out mice into Rag1 knock-out mice revealed that CTL neurotoxicity was mediated by perforin. Accordingly, CTLs isolated from wild-type or interferon-γ knock-out but not from perforin knock-out mice induced neuronal cell death in vitro. CTLs derived from ovalbumin-specific T-cell receptor transgenic mice were not activated and infiltrated less into the ischemic brain compared with wild-type CTLs. Their transfer did not increase the infarct size of Rag1 knock-out mice, indicating antigen-dependent activation as an essential component of CTL neurotoxicity. Our findings underscore the importance of antigen-dependent, direct cytotoxic immune responses in stroke and suggest modulation of CTLs and their effector pathways as a potential new strategy for stroke therapy.

Boosting regulatory T cells limits neuroinflammation in permanent cortical stroke.

Inflammatory mechanisms contribute substantially to secondary tissue injury after brain ischemia. Regulatory T cells (Tregs) are key endogenous modulators of postischemic neuroinflammation. We investigated the potential of histone deacetylase inhibition (HDACi) to enhance Treg potency for experimental stroke in mice. HDACi using trichostatin A increased the number of Tregs and boosted their immunosuppressive capacity and interleukin (IL)-10 expression. In vivo treatment reduced infarct volumes and behavioral deficits after cortical brain ischemia, attenuated cerebral proinflammatory cytokine expression, and increased numbers of brain-invading Tregs. A similar effect was obtained using tubastatin, a specific inhibitor of HDAC6 and a key HDAC in Foxp3 regulation. The neuroprotective effect of HDACi depended on the presence of Foxp3(+) Tregs, and in vivo and in vitro studies showed that the anti-inflammatory cytokine IL-10 was their main mediator. In summary, modulation of Treg function by HDACi is a novel and potent target to intervene at the center of neuroinflammation. Furthermore, this novel concept of modulating endogenous immune mechanisms might be translated to a broad spectrum of diseases, including primary neuroinflammatory and neurodegenerative disorders.

Differential effects of sympathetic nervous system and hypothalamic-pituitary-adrenal axis on systemic immune cells after severe experimental stroke.

Infectious complications are the leading cause of death in the post-acute phase of stroke. Post-stroke immunodeficiency is believed to result from neurohormonal dysregulation of the sympathetic nervous system (SNS) and hypothalamic-pituitary-adrenal (HPA) axis. However, the differential effects of these neuroendocrine systems on the peripheral immune cells are only partially understood. Here, we determined the impact of the hormones of the SNS and HPA on distinct immune cell populations and characterized their interactions after stroke. At various time points after cortical or extensive hemispheric cerebral ischemia, plasma cortisone, corticosterone, metanephrine and adrenocorticotropic hormone (ACTH) levels were measured in mice. Leukocyte subpopulations were flow cytometrically analyzed in spleen and blood. To investigate their differential sensitivity to stress hormones, splenocytes were incubated in vitro with prednisolone, epinephrine and their respective receptor blockers. Glucocorticoid receptor (GCR) and beta2-adrenergic receptor (ß2-AR) on leukocyte subpopulations were quantified by flow cytometry. In vivo effects of GCR and selective ß2-AR blockade, respectively, were defined on serum hormone concentrations, lymphopenia and interferon-γ production after severe ischemia. We found elevated cortisone, corticosterone and metanephrine levels and associated lymphocytopenia only after extensive brain infarction. Prednisolone resulted in a 5 times higher cell death rate of splenocytes than epinephrine in vitro. Prednisolone and epinephrine-induced leukocyte cell death was prevented by GCR and ß2-AR blockade, respectively. In vivo, only GCR blockade prevented post ischemic lymphopenia whereas ß2-AR preserved interferon-γ secretion by lymphocytes. GCR blockade increased metanephrine levels in vivo and prednisolone, in turn, decreased ß2-AR expression on lymphocytes. In conclusion, mediators of the SNS and the HPA axis differentially affect the systemic immune system after stroke. Moreover, our findings suggest a negative-feedback of corticosteroids on the sympathetic axis which may control the post-stroke stress-reaction. This complex interplay between the HPA and the SNS after stroke has to be considered when targeting the neurohormonal systems in the post acute phase of severe stroke.

Inhibition of lymphocyte trafficking shields the brain against deleterious neuroinflammation after stroke.

T lymphocytes are increasingly recognized as key modulators of detrimental inflammatory cascades in acute ischaemic stroke, but the potential of T cell-targeted therapy in brain ischaemia is largely unexplored. Here, we characterize the effect of inhibiting leukocyte very late antigen-4 and endothelial vascular cell adhesion molecule-1-mediated brain invasion-currently the most effective strategy in primary neuroinflammatory brain disease in murine ischaemic stroke models. Very late antigen-4 blockade by monoclonal antibodies improved outcome in models of moderate stroke lesions by inhibiting cerebral leukocyte invasion and neurotoxic cytokine production without increasing the susceptibility to bacterial infections. Gene silencing of the endothelial very late antigen-4 counterpart vascular cell adhesion molecule-1 by in vivo small interfering RNA injection resulted in an equally potent reduction of infarct volume and post-ischaemic neuroinflammation. Furthermore, very late antigen-4-inhibition effectively reduced the post-ischaemic vascular cell adhesion molecule-1 upregulation, suggesting an additional cross-signalling between invading leukocytes and the cerebral endothelium. Dissecting the specific impact of leukocyte subpopulations showed that invading T cells, via their humoral secretion (interferon-γ) and immediate cytotoxic mechanisms (perforin), were the principal pathways for delayed post-ischaemic tissue injury. Thus, targeting T lymphocyte-migration represents a promising therapeutic approach for ischaemic stroke.

Spectratype analysis of clonal T cell expansion in murine experimental stroke.

T lymphocytes are major contributors to post-ischemic neuroinflammation following ischemic stroke. However, the mode of postischemic T cell activation remains a controversial issue. Therefore, this study aimed to directly investigate antigen-dependent clonal T cell expansion after experimental brain ischemia using spectratype/immunoscope analysis. We detected clonal T cell expansion in ischemic brains and to a lesser extent in secondary lymphatic organs and characterized its time course. This is the first study presenting direct evidence of clonal T cell expansion after stroke - however, with delayed kinetics that make antigen-dependent mechanisms unlikely in acute post-ischemic neuroinflammation.

Comparison of humoral neuroinflammation and adhesion molecule expression in two models of experimental intracerebral hemorrhage.

BACKGROUND: Inflammatory cascades contribute to secondary injury after intracerebral hemorrhage (ICH) via humoral factors and cell-mediated cytotoxicity. Several experimental models were previously developed to analyze post-hemorrhagic neuroinflammation. However, neuroinflammatory markers have not been compared face-to-face between these models so far, and therefore, pathophysiological conclusions drawn from only one individual model may not be valid. METHODS: We compared neuroinflammatory pathways in the two most common murine models: striatal injection of autologous blood or collagenase. Expression of pro- and anti-inflammatory cytokines (IL-1, TNF-α, IFN-γ, IL-6, TGF-ß and IL-10) as well adhesion molecule expression (VCAM-1, ICAM-1) was analyzed by RT-PCR at several time points after ICH induction. Outcome and physiological parameters were compared between the models. RESULTS: Both models induced a profound and dynamic increase in the expression of pro-inflammatory cytokines and adhesion molecules. However, blood injection resulted in significantly more pronounced alteration of these markers than collagenase injection. This difference was associated with worse outcome after blood injection compared to the collagenase model despite equal ICH volumes. CONCLUSIONS: This is the first study performing a face-to-face comparison of neuroinflammatory pathways in the two most widely used murine ICH models, revealing substantial differences between the models. This discrepancies need to be taken into account in designing future studies employing experimental ICH models, especially when analyzing neuroinflammatory pathways and therapies.