Interleukin-1 Mediates Ischemic Brain Injury via Induction of IL-17A in γδ T Cells and CXCL1 in Astrocytes

Abstract

As a prototypical proinflammatory cytokine, interleukin-1 (IL-1) exacerbates the early post-stroke inflammation, whereas its neutralization is protective. To further investigate the underlying cell-type-specific IL-1 effects, we subjected IL-1 (α/β) knockout (Il1^-/-) and wildtype (WT) littermate mice to permanent middle cerebral artery occlusion (pMCAO) and assessed immune cell infiltration and cytokine production in the ischemic hemisphere by flow cytometry 24 h and 72 h after stroke. Il1^-/- mice showed smaller infarcts and reduced neutrophil infiltration into the ischemic brain. We identified γδ T cells and astrocytes as target cells of IL-1 signaling-mediated neutrophil recruitment. First, IL-1-induced IL-17A production in γδ T cells in vivo, and IL-17A enhanced the expression of the main neutrophil attracting chemokine CXCL1 by astrocytes in the presence of tumor necrosis factor (TNF) in vitro. Second, IL-1 itself was a potent activator of astrocytic CXCL1 production in vitro. By employing a novel FACS sorting strategy for the acute isolation of astrocytes from ischemic brains, we confirmed that IL-1 is pivotal for Cxcl1 upregulation in astrocytes in vivo. Our results underscore the pleiotropic effects of IL-1 on immune and non-immune cells within the CNS to mount and amplify the post-stroke inflammatory response.

Keywords: Astrocytes; CXCL1; Inflammation; Interleukin-1; Interleukin-17A; Ischemic stroke; γδ T cells.

Fig. 1

Infarct volumes and motorsensory deficits are reduced in Il1^−/− mice post-stroke. A Representative toluidine blue-stained sections for infarct volume analysis of WT littermates and Il1^−/− mice 24 h after pMCAO. Scale bar indicates 1 mm. B Infarct volumes in WT littermates and Il1^−/− mice 24 h after pMCAO, n = 16/group, unpaired Student’s t-test. (C) Total grip strength in WT littermates (n = 16) and Il1^−/− mice (n = 11) 24 h after pMCAO relative to their total grip strength at baseline, which was normalized to 100%, paired Student’s t-test. Data are presented as mean ± SD, **p ≤ 0.01

Fig. 2

Neutrophil infiltration into the ischemic hemisphere is reduced in Il1^−/− mice. A Gating strategy: Microglia were gated as CD45^intermediateCD11b⁺ cells. Neutrophils were gated as CD45^highSSC^highCD11b⁺Ly6G⁺, dendritic cells as CD45^highSSC^highLy6G^lowMHCII⁺CD11c⁺ and macrophages as CD45^highSSC^highLy6G^lowCD11b⁺F4/80⁺. Lymphocyte populations were gated as CD45^highSSC^low. B cells were excluded via B220, and NK cells were identified as CD45^highSSC^lowCD3⁻NK1.1⁺. CD3⁺ T cells were divided into CD45^highSSC^lowCD3⁺γδTCR⁺ γδ T cells, CD45^highSSC^lowCD3⁺γδTCR⁻CD4⁺ CD4⁺ T cells and CD45^highSSC^lowCD3⁺γδTCR⁻CD8⁺ CD8⁺ T cells. Respective parent populations are indicated above the plots. B Neutrophil cell counts (#) in the ischemic hemisphere of WT littermates and Il1^−/− mice 24 h and 72 h after pMCAO and representative flow cytometry plots of CD45^highSSC^highCD11b⁺Ly6G⁺ neutrophils. C Frequency of CD45^highSSC^highCD11b⁺Ly6G⁺ neutrophils of all CD45⁺ cells in cervical lymph nodes 24 h and 72 h after pMCAO. D Absolute numbers of microglia and macrophages in the ischemic hemisphere of WT littermates and Il1^−/− mice 24 h and 72 h after pMCAO. E Absolute numbers of γδ, CD4⁺ and CD8⁺ T cells in the ischemic hemisphere of WT littermates and Il1^−/− mice 24 h and 72 h after pMCAO. All data are presented as mean ± SD, n = 4–7/group, Mann Whitney test, *p ≤ 0.05, **p ≤ 0.01

Fig. 3

IL-17A production by γδ T cells is reduced in Il1^−/− mice following stroke. A Proportion of IL-17A producing γδ T cells of all γδ T cells and absolute numbers (#) of IL-17A positive γδ T cells in the ischemic hemisphere of WT littermates and Il1^−/− mice 24 h and 72 h after pMCAO (n = 5–7/group) and representative flow cytometry plots of IL-17A⁺ γδ T cells (CD45^highSSC^lowCD3⁺γδTCR⁺). B Proportion of IL-17A producing CD4⁺ and CD8⁺ T cells of all CD4⁺ and CD8⁺ T cells, respectively, in the ischemic hemisphere of WT littermates and Il1^−/− mice 24 h and 72 h after pMCAO (n = 5–7/group). C Immunohistochemistry of brain sections from Tcrd-H2BeGFP mice 24 h after tMCAO for visualization of GFP (green) and CD3 (AlexaFluor-555, red) double-positive γδ T cells (arrows) and GFAP (AlexaFluor-647, pink) positive astrocytes in the ischemic penumbra. 10 × magnification, scale bar indicates 50 µm. Data are presented as mean ± SD, Mann Whitney test, **p ≤ 0.01

Fig. 4

Cxcl1 levels are reduced in the ischemic hemisphere of Il1^−/− mice, and astrocytes produce CXCL1 upon stimulation with IL-1ß in vitro. A Expression of Cxcl1, Cxcl2, and Ccl2 in whole-brain mRNA from the ischemic hemisphere of n = 5 Il1^−/− mice normalized to levels in n = 3 WT littermate mice (normalized to 1) 24 h after pMCAO, data are presented as mean ± SEM, unpaired Student’s t-test. B CXCL1 levels in the supernatant of cultivated astrocytes 24 h after stimulation with IL-1ß, IL-17A, TNF, LPS and medium and combinations of two or all three cytokines, respectively, n = 4 independent experiments. Data are presented as mean ± SD, repeated measures one-way ANOVA. C Fold change of Cxcl1 expression in astrocytes stimulated with IL-1ß, IL-17A, TNF, LPS and combinations of two or all three cytokines, respectively, over unstimulated astrocytes, n = 5 independent experiments. Data are presented as mean ± SEM, repeated measures one-way ANOVA. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001

Fig. 5

Cxcl1 expression is reduced in astrocytes from Il1^−/− mice in vivo. A Gating strategy for FACS sorting of astrocytes. B Fold change over input material of astrocyte markers Aqp4 and Slc1a3, oligodendrocyte marker Mog, neuronal marker Tubb3, endothelial cell markers Cldn5 and Tie1, and microglia marker P2ry12 in n = 8 astrocyte samples sorted from the contralateral hemisphere of  n= 8 Il1^−/− mice and n = 10 WT littermates at both time points after pMCAO. C Cxcl1 expression 24 h and 72 h after pMCAO. Cxcl1 expression at 24 h in n = 3 astrocyte samples sorted from the ischemic hemisphere of n = 6 Il1^−/− mice relative to the expression in n  = 3 samples from n = 6 WT littermates, which was normalized to 1. Cxcl1 expression at 72 h in n  = 4 astrocyte samples sorted from the ischemic hemisphere of n = 7 Il1^−/− mice relative to the expression in n = 4 samples from n = 9 WT littermates, which was normalized to 1. D Expression of reactivity markers lipocalin 2 (Lcn2), tissue inhibitor of metalloproteinase 1 (Timp1), CCAAT/enhancer-binding protein beta (Cebpb) and pentraxin 3 (Ptx3) 24 h and 72 h after pMCAO in n = 3 astrocyte samples sorted from the ischemic hemisphere of n = 6 Il1^−/− mice relative to the expression in n = 3 samples from n = 6 WT littermates and n = 4 astrocyte samples sorted from the ischemic hemisphere of n = 7 Il1^−/− mice relative to the expression in n = 4 samples from n = 9 WT littermates, respectively. E Immunofluorescent co-staining for GFAP (Cy3, red), CXCL1 (AlexaFluor-488, green), and DAPI (blue) in WT littermate mice 12 h after pMCAO, 40 × magnification, scale bar indicates 20 µm. F Immunofluorescent staining for GFAP (Cy3, red), Iba1 (AlexaFluor-488, green), and DAPI (blue) in the penumbra of WT littermates and Il1^−/− mice 24 h after pMCAO, 40 × magnification, scale bar indicates 50 µm. All astrocyte gene expression data are presented as mean ± SEM, unpaired Student’s t-test, *p ≤ 0.05