CCR6 (CC Chemokine Receptor 6) Is Essential for the Migration of Detrimental Natural Interleukin-17–Producing γδ T Cells in Stroke
Background and Purpose—Immune-mediated tissue damage after stroke evolves within the first days, and lymphocytes contribute to the secondary injury. Our goal was to identify T-cell subpopulations, which trigger the immune response.
Methods—In a model of experimental stroke, we analyzed the immune phenotype of interleukin-17 (IL-17)–producing γδ T cells and explored the therapeutic potential of neutralizing anti-IL-17 antibodies in combination with mild therapeutic hypothermia.
Results—We show that brain-infiltrating IL-17–positive γδ T cells expressed the Vγ6 segment of the γδ T cells receptor and were largely positive for the chemokine receptor CCR6 (CC chemokine receptor 6), which is a characteristic for natural IL-17–producing γδ T cells. These innate lymphocytes are established as major initial IL-17 producers in acute infections. Genetic deficiency in Ccr6 was associated with diminished infiltration of natural IL-17–producing γδ T cells and a significantly improved neurological outcome. In the ischemic brain, IL-17 together with tumor necrosis factor-α triggered the expression of CXC chemokines and neutrophil infiltration. Therapeutic targeting of synergistic IL-17 and tumor necrosis factor-α pathways by IL-17 neutralization and therapeutic hypothermia resulted in additional protective effects in comparison to an anti-IL-17 antibody treatment or therapeutic hypothermia alone.
Conclusions—Brain-infiltrating IL-17–producing γδ T cells belong to the subset of natural IL-17–producing γδ T cells. In stroke, these previously unrecognized innate lymphocytes trigger a highly conserved immune reaction, which is known from host responses toward pathogens. We demonstrate that therapeutic approaches targeting synergistic IL-17 and tumor necrosis factor-α pathways in parallel offer additional neuroprotection in stroke.
Stroke is worldwide ranked as the second most common cause of death and the third most common cause of disability-adjusted life years.1 Numerous experimental and clinical studies demonstrate that the activation of the systemic immune compartment and brain resident immune cells is a major element of the pathophysiology of stroke. Recently, clinical studies targeting T cells highlighted the therapeutic potential of immune modulatory treatment strategies.2
T-cell–related detrimental mechanisms in cerebral ischemia encompass perforin-mediated neurotoxicity and the production of the proinflammatory cytokines interleukin-17 (IL-17) and IL-21.3–5 Whereas IL-21 and perforin are thought to be directly neurotoxic, IL-17 effects mainly depend on the induction of proinflammatory chemokines, which are leading to a rapid neutrophil infiltration into ischemic hemispheres. Major sources of IL-17 in ischemic brains are γδ T cells, which are producing IL-17 within 24 hours after the ischemic attack.5 The temporal dynamics of the IL-17 production in γδ T cells indicate that antigen-specific priming is not essential for their activation.5,6 In models of acute infection, similarly rapid IL-17 responses are mounted by γδ T cells before the antigen-specific activation of T cells and B cells.7
IL-17–producing γδ T cells can be divided into several subsets. So-called natural IL-17–producing γδ T cells (nTγδ17 cells) are producing IL-17 within 12 hours.8 The rapid antigen-independent activation of nTγδ17 cells can occur through the engagement of pathogen pattern recognition receptors and the cytokine receptors for IL-1β and IL-23, whereas the antigen receptor is not required. nTγδ17 cells express CCR6 (CC chemokine receptor 6)7 and can further be classified based on the expression of the variable (v) segment of the T-cell receptor. nTγδ17 cells express mostly Vγ4, or Vγ6, according to the classification proposed by Heilig and Tonegawa.9
Although several studies suggest detrimental functions of IL-17–producing γδ T cells in stroke, their exact phenotype is unclear.5,10,11 In this study, we classified γδ T cells in detail, studied the role of the chemokine receptor CCR6 for their migration, and investigated their effects on stroke outcome. To further explore the therapeutic potential of an IL-17 neutralization, we combined the treatment with anti-IL-17 antibodies with mild therapeutic hypothermia (TH). TH is already established as a treatment for several central nervous system diseases and has a major effect on proinflammatory cytokines, that is, tumor necrosis factor-α (TNF-α) and IL-1β.12 In connection with IL-17, TH effects on TNF-α are potentially crucial because both cytokines have synergistic effects on the expression of neutrophil-attracting CXC chemokines.
All animal experiments were approved by the local animal care committee (Behörde für Lebensmittelsicherheit und Veterinärwesen Hamburg). We conducted the experiments according to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 83-123, revised 1996) and performed all procedures in accordance with the ARRIVE guidelines (Animal Research: Reporting of In Vivo Experiments; http://www.nc3rs.org/ ARRIVE) between January 2015 and April 2017. Ccr6−/− homozygous mice (C57BL/6 background; strain B6.129P2-Ccr6−/−) were described before.13 Transgenic mice were back-crossed at least 10 generations to the C57BL/6 background. Age-matched male wild-type littermates served as controls. We randomized all mice and conducted transient middle cerebral artery occlusion for 45 minutes as previously described5 using the intraluminal filament method (6-0 nylon) in a blinded fashion. The detailed experimental description can be found in the online-only Data Supplement. Experimental groups and animal numbers are shown in Table I in the online-only Data Supplement.
Animals were treated with 500 µg of mouse monoclonal anti–murine IL-17A antibody5 (Clone MM17F3; 16.6 mg/kg of bodyweight) or with 500 µg of isotype control antibody (IgG1). Mice were injected intravenously once at 90 minutes after onset of ischemia.
Analysis of Infarct Size by 2,3,5-Triphenyl-2-Hydroxy-Tetrazolium Chloride Staining
We analyzed infarct size by harvesting brains and cutting them into 1 mm slices (Braintree Scientific, 1 mm) followed by vital staining using 2% (wt/vol) 2,3,5-triphenyl-2-hydroxy-tetrazolium chloride in phosphate buffer. We determined infarct volumes in a blinded fashion using NIH ImageJ software.
Stroke Assessment by Magnetic Resonance Imaging
Magnetic resonance imaging was performed in normal mice on a dedicated 7-T magnetic resonance small animal imaging system (ClinScan, Bruker). The image protocol comprised T2-weighted imaging, diffusion-weighted imaging, and 3-dimensional time-of-flight angiography.
Temperature Modulation Methodology
Core body temperature was maintained at 37°C throughout surgery for normothermia and 33 to 34°C for mild hypothermia as described before by using a feedback-controlled heating device (DC temperature controller, FHC Apparatus). The detailed experimental description can be found in the online-only Data Supplement.
Antibodies and Flow Cytometry
RNA Isolation and Quantitative Real-Time Polymerase Chain Reaction
We obtained real-time PCR primers from Applied Biosystems. The detailed experimental description can be found in the online-only Data Supplement.
We stained brains after standard immunohistochemistry procedures with antibodies against lymphocyte antigen 6 complex, locus G (1:200; 1A8; Biolegend) glial fibrillary acidic protein (1:400; 4A11; BD Biosciences), and ionized calcium binding adaptor molecule 1 clone (1:100; A1F1; Proteintech). The detailed experimental description can be found in the online-only Data Supplement.
Data are reported as mean±SD. Statistical analyses were performed using the appropriate test indicated in the figure legends. Briefly, Student t test was used to compare infarct volumes and quantitative polymerase chain reaction data; Mann–Whitney U test for the comparison of clinical scores and 1-way ANOVA for multiple comparisons with Bonferroni post hoc test, after validating the normal distribution of these data sets (Kolmogorov–Smirnov test). P values <0.05 were considered statistically significant.
Vγ6+/CCR6+ γδ T Cells Are the Major Producers of IL-17 in the Ischemic Brain
Others and we have shown that γδ T cells are detrimental in stroke, but the exact phenotype of these innate-like T lymphocytes is unknown.5
When we measured the infiltration of γδ T cells on day 3 after experimental stroke, we found ≈1664±387.6 γδ T cells per ischemic hemisphere (Figure 1A). Analysis of the cellular infiltrate and intracellular cytokines levels at early time points revealed that γδ T cells were already present at 12 hours, and that 41±3% of the infiltrating γδ T cells were positive for IL-17. The IL-17 levels persisted on similarly elevated levels until day 3 (45±12%). In comparison to IL-17, interferon-γ levels were significantly lower with 10±12% (12 hours) and 15±9% (3 days), respectively. We did not observe γδ T cells double positive for IL-17 and interferon-γ. Seventy-nine ±15% of infiltrating γδ T cells were positive for TNF-α on day 3.
To differentiate infiltrating γδ T-cell subsets in our stroke model, we analyzed the expression of the variable (v) segments Vγ1, Vγ4, and Vγ6 of the T-cell receptor, according to the classification proposed by Heilig and Tonegawa.9 Staining for Vγ6 was performed as described before, after pretreatment with anti-Cδ.14 We found that 18±2% were positive for Vγ1, 19±2% for Vγ4, 40±5% for Vγ6, and 24±9% were stained by none of the antibodies against Vγ1/4/6 (Figure 1B). In the next step, we measured IL-17 levels in the different subsets and found that the vast majority of IL-17+ γδ T cells expressed Vγ6 (95±5%; Figure 1B). In contrast, we did not observe Vγ1+/IL-17+ γδ T cells. Only few IL-17+ γδ T cells were positive for Vγ4 (6±2%) or negative for antibodies against Vγ1/4/6 (1±1%; Figure 1B). It is known from other studies that subsets of IL-17–producing γδ T cells express the chemokine receptor CCR6.15 We observed that 59±7% of the infiltrating γδ T cells were positive for CCR6 (Figure 1C). When we further gated on Vγ6+ γδ T cells, we determined that CCR6 was expressed on 80±6% of Vγ6+ γδ T cells and on 47±8% of IL-17+/Vγ6+ γδ T cells, whereas Vγ1+ and Vγ4+ γδ T cells were negative for CCR6 (Figure 1D).
Taken together, we show that the vast majority of the infiltrating IL-17+ γδ T cells display a T-cell receptor that comprises Vγ6+. It is this Vγ6+ subset of γδ T cells, which is positive for the chemokine receptor CCR6. Thus, Vγ6+/CCR6+ γδ T cells are the major producers of IL-17 in the ischemic brain.
Ccr6−/− Mice Are Protected From Experimental Stroke
To investigate whether CCR6 has effects on the recovery of animals, we analyzed the neurological outcome in Ccr6−/− mice and littermates in our middle cerebral artery occlusion model. On day 3, Ccr6−/− mice showed significantly reduced infarct sizes and milder disability compared with littermate controls (Figure 2A and 2B). Mortality, physiological parameters, vessel occlusion during transient middle cerebral artery occlusion, and vasculature were not altered between genotypes (Figure IA and Table II in the online-only Data Supplement; Figure 2C).
CCR6 Is Required for the Infiltration of IL-17+ γδ T Cells in Experimental Stroke
To determine whether the expression of CCR6 is required for the migration of γδ T cells, we analyzed the cellular infiltrate in Ccr6−/− mice and littermate controls by flow cytometry. On day 3, we detected a significant reduction in absolute numbers of γδ T cells in Ccr6−/− mice in comparison to littermate controls (Figure 3A), whereas CD4 and CD8 T cells (Figure IB in the online-only Data Supplement) were unchanged. When we measured intracellular cytokine levels in infiltrating γδ T cells, we determined reduced IL-17 levels in Ccr6−/− mice, whereas interferon-γ and TNF-α levels were not affected (Figure 3A). In contrast, cytokine levels in infiltrating αβ T cells were not altered (Figure IC and ID in the online-only Data Supplement). To analyze whether Ccr6−/− mice exhibit an altered repertoire of γδ T cells in the peripheral immune system, we measured γδ T cells subsets and cytokine levels in γδ T cells in cervical and mesenteric lymph nodes without detecting significant changes between genotypes (Figure IIA and IIB in the online-only Data Supplement). The analysis of Ccl20 RNA levels in whole brain RNA of ischemic hemisphere revealed a robust upregulation of the CCR6 ligand at 12 and 24 hours (Figure 3B). We next analyzed the expression of Vγ1, Vγ4, and Vγ6 on infiltrating γδ T cells in Ccr6−/− mice and observed a significant reduction in absolute numbers selectively for the Vγ6+ γδ T cell subset (Figure 3C). This was paralleled by a dramatic decrease of IL-17 levels in the Vγ6+ subset of Ccr6−/− mice (Figure 3D).
Taken together, we show that the expression of the chemokine receptor CCR6 is required for the infiltration of IL-17+ Vγ6+ γδ T cells in stroke.
Deficiency in CCR6 Affects Neutrophil Infiltration
Others and we have previously reported that γδ T cells are detrimental in stroke via an IL-17–mediated induction of the neutrophil-attracting chemokine CXCL1.5 To assess whether diminished IL-17 levels in Ccr6−/− mice are associated with decreased neutrophil levels, we analyzed cellular infiltrates by flow cytometry and immunohistochemistry and observed a selective significant reduction in neutrophil numbers on day 3 in Ccr6−/− mice (Figure 4A). Cell numbers and intracellular TNF-α levels of microglia, monocytes/macrophages, and dendritic cells were similar between genotypes (Figure 4B and 4C). When we analyzed levels of the Cxcl1 transcript in whole brain RNA, we found reduced levels in Ccr6−/− mice (Figure 4D). These findings link the CCR6+/IL-17+ γδ T cells with the IL-17–dependent induction of CXC chemokines and the subsequent infiltration of neutrophils.
Combined Disruption of IL-17 and TNF-α Pathways Results in an Additional Decrease of Infiltrating Neutrophil
Both mild hypothermia and neutralization of IL-17 are neuroprotective in experimental stroke. Protective effects of mild hypothermia include a robust downregulation of TNF-α.16 Others and we have previously shown that TNF-α and IL-17 synergistically induce CXC chemokines. The synergistic function of both cytokines let us speculate that a combined treatment—hypothermia and neutralization of IL-17—might exert additive beneficial effects on stroke outcome. When we investigated effects of mild hypothermia, we found significantly diminished TNF-α levels in microglia and monocytes/macrophages under TH (Figure 5A). We next observed that the deficiency in CCR6 in combination with TH led to a significant reduction of infiltrating γδ T cells and IL-17 levels when compared with hypothermic wild-type animals, indicating that γδ T cells were still functional under TH (Figure 5B). Finally, we found that TH in Ccr6−/− mice significantly further diminished Cxcl1 levels and numbers of infiltrating neutrophil in comparison to hypothermic littermate controls (Figure 5C and 5D).
Taken together, we show that the combined reduction of TNF-α and IL-17 levels results in a significant decrease of neutrophil levels in ischemic brains in comparison to approaches, which are targeting either TNF-α or IL-17 pathways alone.
Neutralization of IL-17 Provides Additional Protective Effects Under Conditions of Mild Hypothermia
We next analyzed the neurological outcome after middle cerebral artery occlusion in hypothermic mice with and without disruption of the CCR6/IL-17-axis. Three days after the ischemic attack, hypothermic Ccr6−/− mice showed a significant reduction in infarct size and improved neurological outcome in comparison to hypothermic littermate controls (Figure 6A). In addition, and much more interesting for therapy, administration of a single dose of anti-IL-17A (500 µg) antibody 90 minutes after stroke induction led to a significant reduction in infarct sizes and improved clinical outcomes in hypothermic wild-type animals compared with the injection of an isotype control antibody (Figure 6B). To demonstrate that the detrimental effects of CCR6+ γδ T cells were IL-17 dependent, we finally analyzed the effect of the anti-IL-17A neutralization in Ccr6−/− mice and found that the anti-IL-17A antibody did not exert additional protective effects.
Our study in a model of experimental stroke clarifies the mechanism how detrimental IL-17+ γδ T cells are recruited to the site of ischemic tissue damage. We show that the IL-17+ γδ T cells can be classified as so-called nTγδ17 cells8 and that the migration of these harmful nTγδ17 cells into ischemic hemispheres is CCR6 dependent.
Although a proinflammatory role of IL-17–positive γδ T cells in stroke was suggested by several studies, recruitment mechanisms and the phenotype of these innate lymphocytes were unclear.5,10 The observation that brain-infiltrating IL-17+ γδ T cells display the phenotype of CCR6+/Vγ6+ nTγδ17 cells now explains why these innate-like lymphocytes are capable to exert their detrimental function already in the first days. The physiological role of nTγδ17 cells is to immediately produce IL-17 in response to pathogens,8 which can be triggered through IL-1, IL-23, and toll-like receptor agonists without explicit involvement of the T-cell receptor.17,18 nTγδ17 cells are the major innate producers of IL-17, which is the central to induce host defenses.19 IL-17 immediately induces chemokines, which are orchestrating the neutrophils influx to the site of injury. In mice, nTγδ17 can be characterized by the expression of CCR6 and either Vγ1, Vγ4, or Vγ6.8 We now show that equally to bacterial infections, sterile inflammation after stroke triggers similarly conserved pathways: (1) production of IL-17 by Vγ6+/CCR6+ nTγδ17 cells; (2) induction of CXC chemokines; and (3) infiltration of neutrophils. Notably, consequences for affected tissues differ. Whereas the neutrophil influx is protective in infections, the infiltration into the highly vulnerable ischemic brain leads to additional collateral damage. Accordingly, blocking of neutrophils responses reduces pathology in ischemia reperfusion injury.5,10,20
To demonstrate the pivotal role of the IL-17 axis and the CCR6 expression on nTγδ17, we further analyzed effects of an IL-17 neutralization and genetic depletion of CCR6. We found that both the neutralization of IL-17 and the inhibition of the nTγδ17 cell migration into ischemic brains in Ccr6−/− mice had beneficial effects on the neurological outcome. These results are underlining the essential function of CCR6 for the migration of nTγδ17 in stroke and are in correspondence with previous reports, demonstrating neuroprotective effects for the intracerebroventricular neutralization of CCL20,21 which is the main ligand for CCR6.22 In principle, reduction of γδ T cells in ischemic brain of Ccr6−/− mice can also be attributed to smaller infarct volumes. However, the finding of a selective reduction of Vγ6+ γδ T cells and IL-17+/Vγ6+ γδ T cells in Ccr6−/− mice argues for specific effects of the CCR6 deficiency on the IL-17+ subset of γδ T cells.
It is a known scheme in immunology that proinflammatory cascades converge and synergistically amplify downstream signaling pathways.23 The combined inhibition of several pathways in parallel might, therefore, offer new therapeutic options. A prototypical pair of cytokines, which are synergistically enhancing the gene expression of proinflammatory CXC chemokines, are IL-17 and TNF-α.5,24 Similar to IL-17, expression levels of TNF-α are rapidly upregulated in ischemic brains, and anti-TNF therapy proved to be neuroprotective in experimental stroke.25 Besides antibody-mediated neutralization, mild TH diminishes TNF-α levels in ischemic brains.12 Because TH is already established as a treatment for several central nervous system diseases, that is, cerebral hypoxia, we performed a proof of principle study, combining IL-17 neutralization and TH. Supporting the concept that the inhibition of synergistic immunologic pathways can provide additional neuroprotection, we found that the neutralization of IL-17 in combination with TH led to an additional decrease of CXC chemokine and neutrophil levels and improved neurological outcome in comparison to anti-IL-17 antibody treatment or TH alone.
Taken together, our study shows that the initiation of the sterile immune response in stroke follows highly conserved patterns. Here, we describe for the first time IL-17–producing Vγ6+/CCR6+ γδ T cells as the main source of IL-17, which seem to play a similar role in ischemic stroke as previously described for infectious diseases.
We thank Ellen Orthey and Anika Ruhl for excellent technical assistance.
Sources of Funding
This work was supported by grants from the European Research Area Network/nEUROSurv (Dr Magnus) and the Hermann und Lilly Schilling-Foundation for medical research (Dr Magnus).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.117.016753/-/DC1.
- Received January 18, 2017.
- Revision received April 20, 2017.
- Accepted May 10, 2017.
- © 2017 American Heart Association, Inc.
- Feigin VL,
- Forouzanfar MH,
- Krishnamurthi R,
- Mensah GA,
- Connor M,
- Bennett DA,
- et al
- Zhu Z,
- Fu Y,
- Tian D,
- Sun N,
- Han W,
- Chang G,
- et al
- Mracsko E,
- Liesz A,
- Stojanovic A,
- Lou WP,
- Osswald M,
- Zhou W,
- et al
- Clarkson BD,
- Ling C,
- Shi Y,
- Harris MG,
- Rayasam A,
- Sun D,
- et al
- Gelderblom M,
- Weymar A,
- Bernreuther C,
- Velden J,
- Arunachalam P,
- Steinbach K,
- et al
- Gelderblom M,
- Leypoldt F,
- Steinbach K,
- Behrens D,
- Choe CU,
- Siler DA,
- et al
- Turner JE,
- Paust HJ,
- Steinmetz OM,
- Peters A,
- Riedel JH,
- Erhardt A,
- et al
- Roark CL,
- Aydintug MK,
- Lewis J,
- Yin X,
- Lahn M,
- Hahn YS,
- et al
- Hamada S,
- Umemura M,
- Shiono T,
- Tanaka K,
- Yahagi A,
- Begum MD,
- et al
- Lee AY,
- Phan TK,
- Hulett MD,
- Körner H
- Hartupee J,
- Liu C,
- Novotny M,
- Li X,
- Hamilton T