Figure 1. γδ T cells and γδ T-cell subsets in the blood and cerebrospinal fluid (CSF) of various populations of patients. A, Contour plots with parent gate on CD3+ T cells. γδ T-cell frequencies (in percentage) are shown in the peripheral blood and CSF of patients with clinically isolated syndrome/multiple sclerosis in relapse (CIS/MS in relapse) as compared with control groups (Kruskal-Wallis followed by a Dunn post hoc test). TCR indicates T-cell receptor. B, Contour plots with parent gate on γδ T cells. CD161high CCR6+ γδ T-cell frequencies (in percentage) in blood and CSF samples of patients with CIS/MS in relapse and control patient groups are plotted (Kruskal-Wallis followed by a Dunn post hoc test). Significant differences between groups are marked by a capped line.
Figure 2. Cerebrospinal fluid (CSF) accumulation and functional phenotype of CD161high CCR6+ γδ T cells. A, Histograms with parent gate on γδ T cells. Differential expression of CCR6 (upper panel) and CD161 (lower panel) on γδ T cells in peripheral blood and CSF. B, Scatterplots showing differences in CD161high CCR6+ γδ T-cell frequencies between blood and CSF in patients with clinically isolated syndrome/multiple sclerosis in relapse (CIS/MS in relapse) and control patient groups (Wilcoxon signed rank test). (C) Contour and dot plots with parent gate on γδ T cells (contour plot) and subpopulations of γδ T cells according to their CD161 and CCR6 status (lower panel). Note that interleukin 17 (IL-17)–producing γδ T cells are exclusively found in the CD161high CCR6+ subpopulation. All numbers are given as a percentage of parent population (parts A and C), and in dot plots (part C), numbers are given as mean (SD). Three independent experiments.
Schirmer L, Rothhammer V, Hemmer B, Korn T. Enriched CD161high CCR6+ γδ T Cells in the Cerebrospinal Fluid of Patients With Multiple Sclerosis. JAMA Neurol. 2013;70(3):345-351. doi:10.1001/2013.jamaneurol.409
Author Affiliations: Department of Neurology, Klinikum rechts der Isar, Technische Universität München, Munich, Germany.
Objective To investigate the expression of CD161 (KLRB1) and CCR6 on human γδ T cells in blood and cerebrospinal fluid (CSF) of patients with a clinically isolated syndrome (CIS) and multiple sclerosis (MS) in relapse.
Design Flow cytometry analysis of CD161 and CCR6 expression and intracellular cytokine staining for interleukin 17 and interferon-γ on human γδ T cells in blood and CSF samples.
Setting Department of Neurology, Klinikum rechts der Isar, Technische Universität München, a tertiary referral center.
Patients Twenty-six patients with CIS/MS in active relapse, 10 patients with other autoimmune disorders, 12 patients with neuroinfectious diseases, and 15 patients with noninflammatory neurological diseases.
Main Outcome Measures Frequencies of CD161high and CCR6+ γδ T cells in blood and CSF samples of patients with CIS/MS in relapse and control patients.
Results γδ T cells were increased in both blood and CSF of patients with CIS/MS in relapse as compared with controls with noninflammatory disease. The fraction of CD161high CCR6+ γδ T cells was significantly higher in the CSF of patients with CIS/MS in relapse than of those with systemic autoimmune disorders or controls with noninflammatory disease. The CD161high CCR6+ double-positive γδ T-cell population was further enriched in the CSF in relation to blood in patients with CIS/MS in relapse but not in patients with infectious disease or the other control groups. The CD161high CCR6+ γδ T-cell population was characterized by its capacity to produce interleukin 17.
Conclusion Interleukin 17–producing CD161high CCR6+ γδ T cells might contribute to the compartmentalized inflammatory process in the central nervous system of patients with MS.
Autoreactive helper T subtype 17 (TH17) cells are important inducers of immunopathology in various animal models of organ-specific autoimmunity.1 In addition, TH17 cells appear to be enriched in the peripheral blood and cerebrospinal fluid (CSF) of patients with multiple sclerosis (MS).2 However, TH17-associated effector cytokines including interleukin 17A (IL-17A), IL-17F, IL-21, and IL-22 are also produced by γδ T cells.3 Based on a series of recent studies in murine model systems, it is believed that γδ T cells adopt distinct functional phenotypes according to the signature cytokines that they produce. Both the interferon-γ (IFN-γ)–producing (γδT1 cells) and IL-17–producing subset of γδ T cells (γδT17 cells) seem to be committed to their respective phenotype already during thymic development.4 Elevated fractions of T cells with a γδ T-cell receptor have been observed in blood samples of patients with MS and γδ T cells were found to occur in early MS lesions.5- 7 In human peripheral blood, γδ T cells compose only a small population of less than 5% of the entire CD3+ T-cell population. As opposed to CD4+ TH and CD8+ cytotoxic T cells with conventional αβ T-cell receptors, γδ T cells are not HLA antigen restricted and are capable of responding rapidly to pattern recognition receptor signals as a first line of defense. Hence, γδ T cells are assumed to link innate with adaptive immune responses.8 It is possible that γδ T cells also play a role in lymphoid stress surveillance responses, implying their potential significance in tumor immunology and autoimmunity.9 However, it is unclear whether specific subsets of γδ T cells, ie, γδT1 vs γδT17 cells, are preferentially involved either in distinct autoimmune disorders or anatomical niches that are affected by these autoimmune disorders. Indeed, increased fractions of IL-17–producing γδ T cells have been recovered from the peripheral blood of patients with ankylosing spondylitis10 and from psoriasis lesions.11
By using the surface markers CD161 and CCR6 that have been associated with production of IL-17 in other immune cell subsets,12,13 we herein characterized γδ T-cell subpopulations in fresh blood and CSF samples of patients with a clinically isolated syndrome (CIS) or with MS in relapse in comparison with inflammatory and noninflammatory control groups. Overall, we found increased fractions of γδ T cells in both peripheral blood and CSF samples of patients with CIS/MS in relapse. Only in patients with CIS/MS in relapse, but not in controls, the CD161high CCR6+ γδ T-cell subset was enriched in the CSF as compared with peripheral blood. By intracellular cytokine staining, we determined that the CD161high CCR6+ γδ T-cell subpopulation was characterized by the capacity to produce IL-17 whereas CD161low CCR6− γδ T cells produced large amounts of IFN-γ but no IL-17.
We examined fresh peripheral blood and CSF samples from 26 patients with CIS/MS in relapse. The patients were in active relapse and not yet receiving steroid treatment. The control groups included 10 patients with other autoimmune disorders, 12 patients with neuroinfectious diseases, and 15 patients with noninflammatory neurological diseases (Table). The patients were between 19 and 82 years of age (mean [SD] 44  years for all patients and 33  years in the CIS/MS in relapse group) (Table). In patients with CIS/MS in relapse, the mean (SD) Expanded Disability Status Scale score at the time of blood and CSF sampling was 2.6 (1.3). This study was approved by the local ethics committee and informed consent was obtained from all patients.
Erythrocytes in fresh EDTA blood were lysed and leukocytes were washed with 2% fetal calf serum/phosphate-buffered saline. Fresh CSF samples were directly washed and processed. Ex vivo staining was carried out with the following antibodies: pan-γδ-TCR-FITC (clone Immu510; Beckman Coulter), CD161-PE (clone 191B8; Miltenyi Biotec Inc), CCR6-PE-Cy7 (clone 53103; R&D Systems), and CD3-APC-Cy7 (clone SK7; BD Biosciences) for 30 minutes at 4°C. For functional analysis, fresh healthy donor peripheral blood mononuclear cells were isolated using a density Ficoll-Paque (Biocoll; Biochrom) gradient centrifugation; 5 × 105 cells/well were then stimulated in T-cell medium14 for 4 hours in the presence of phorbol 12–myristate 13–acetate (50 ng/mL; Sigma), ionomycin (1 μg/mL; Sigma), and monensin (GolgiStop, 1 μg/mL; BD Biosciences). On stimulation, surface staining was carried out with antibodies to pan-γδ-TCR-FITC, CD161-PE, CCR6-PeCy7, and CD3-ECD (clone UCHT1; Beckman Coulter) followed by fixation and permeabilization using cytofix/cytoperm and perm/wash buffer (BD Biosciences). Intracellular cytokine staining was carried out overnight at 4°C with antibodies to IFN-γ–PacificBlue (clone B27; Invitrogen) and IL-17A–APC (clone eBio64CAP17; eBiosciences). IgG isotype controls were run in parallel. A live/dead staining was applied (Invitrogen) to exclude dead cells in all cases.
All cells were analyzed using an FACS cytometer (CYAN; Beckman Coulter) and FlowJo Software (Treestar). For statistical evaluation, the nonparametric Kruskal-Wallis test with a Dunn multiple-comparison post hoc test was applied to compare different γδ T-cell populations between patients with CIS/MS in relapse and control patients. A Wilcoxon signed rank test was applied for comparison of γδ T-cell fractions between peripheral blood and CSF samples. All tests were classified as significant if P was <.05. GraphPad PRISM (Graph Pad Software Inc) software was used for statistical calculations and graph generation.
γδ T cells have been implicated in lymphoid stress surveillance responses and, thus, might play a role in various human autoimmune disorders. Therefore, we determined the fractions of γδ T cells in the blood and CSF of patients with active MS as compared with control subjects. By ex vivo surface staining, we found significantly higher fractions of γδ T cells in both blood and CSF samples of patients with CIS/MS in relapse as compared with controls with noninflammatory diseases (mean [SD], blood: 7.4% [5.2%] of CD3+ T cells vs 3.6% [2.8%]; CSF: 2.4% [1.6%] vs 1.2% [0.8%]) (Figure 1A). The fraction of γδ T cells in the peripheral blood is independent of sex but decreases with age.15,16 Thus, we wished to focus on functionally defined γδ T-cell subsets and compare them between distinct anatomical compartments within each group. To determine the potential functional phenotype of γδ T cells, we used surface markers that were recently associated with production of IL-17 in other immune cell populations. We focused on CCR6 and CD161, which seem to be robust markers for IL-17 production in human TH cells. Herein, we detected significantly higher frequencies of CD161high CCR6+ γδ T cells in the CSF of patients with CIS/MS in relapse (mean [SD], 16.5% [11%]) as compared with CSF samples from patients with other autoimmune disorders (mean [SD], 5.9% [5.4%]) and patients with noninflammatory disease (mean [SD], 6.5% [6%]) (Figure 1B). The fraction of CD161high CCR6+ γδ T cells was also elevated in the CSF of patients with neuroinfectious disease (mean [SD], 9.5% [7.3%]). However, when we compared paired samples of peripheral blood and CSF within the same group, we found CD161high CCR6+ γδ T cells to accumulate in CSF only in patients with CIS/MS in relapse (P = .02) (Figure 2A and B), but not in other control groups including patients with viral or bacterial meningitis (Figure 2B). Notably, the CD161high γδ T-cell population (ie, not the double-positive population) even decreased from blood to CSF in the group with noninflammatory disease (P = .001). These data suggest that compartmentalized inflammatory conditions in the central nervous system (CNS) might be associated with an increase in the CSF fraction of CD161high CCR6+ γδ T cells and that this specific γδ T-cell subset is particularly enriched in the CSF in relation to peripheral blood in patients with active CNS-specific autoimmune diseases such as CIS/MS in relapse. In our cohort, we did not find a correlation between the Expanded Disability Status Scale score of the patients with CIS/MS in relapse with the frequency of either the entire γδ T-cell population or the fraction of the CD161high CCR6+ γδ T cells in the peripheral blood or CSF (data not shown).
Intracellular cytokine staining revealed that IL-17–producing γδ T cells were predominantly found in the CD161high CCR6+ double-positive subpopulation (mean [SD], 4.1% [2.5%] IL-17–only producers; 5.9% [8.1%] IL-17 and IFN-γ double producers; and 58.4% [7.7%] IFN-γ–only producers) (Figure 2C). The CD161low CCR6− compartment contained only 0.1% of IL-17 producers but was characterized by the production of large amounts of IFN-γ (mean [SD], 85.4% [7.7%]). CD161− CCR6− γδ T cells only produced low amounts of IFN-γ (mean [SD], 47.9% [17.9%]) and no IL-17 (mean [SD], 0.11% [0.14%]). Taken together, our results suggest that CD161high CCR6+ γδ T cells that have the capacity to produce IL-17 might preferentially accumulate in the CSF in autoimmune inflammatory processes that target the CNS.
In the present study, we found that CD161high CCR6+ γδ T cells are increased in the CSF of patients with infectious CNS diseases and with CIS/MS in relapse. However, only patients with CIS/MS in relapse showed a preferential enrichment of the CD161high CCR6+ γδ T-cell subset in the CSF as compared with the systemic compartment. Similar to TH cells, the combined expression of CD161 and CCR6 defined a subset of IL-17–producing γδ T cells in adult human individuals. It is unclear whether CSF-derived γδ T cells upregulate CD161 and CCR6 on local inflammatory stimuli in the CNS or whether peripherally activated CD161high CCR6+ γδ T cells are recruited to the CSF. However, the fact that in viral meningitis, which is usually subsequent to or accompanied by a systemic infection, no clear enrichment of CD161high CCR6+ γδ T cells was detected in CSF vs blood supports the hypothesis that local inflammatory stimuli in the CSF compartment might maintain the CD161high CCR6+ γδ T-cell population more efficiently in the CSF than in the blood of patients with MS.
In the current understanding, γδ T cells that populate the secondary lymphoid tissue (noncanonical γδ T cells) are committed to functionally distinct subsets already in the thymus. In a series of studies in murine systems, transcriptional networks and surface markers of IFN-γ–producing γδT1 cells vs IL-17–producing γδT17 cells have been identified.17,18 It is likely that similar developmental pathways are operational for γδ T cells in humans as well. In mice, there is strong evidence that γδT17 cells are self-renewing in the peripheral immune compartment and are only produced by the embryonic but not the adult thymus.19 This is unclear in humans but the most convincing data that γδT17 cells might be involved in human disease conditions were collected in a study on bacterial meningitis in young children.20 Our data in patients with active MS suggest that CD161high CCR6+ γδ T cells that have the potential to produce IL-17 are still detectable in young adult individuals. Similarly, γδT17 cells were reported to be present either systemically or within lesions in patients with Mycobacterium tuberculosis infection,21 ankylosing spondylitis, and psoriasis.10,11 Some of these reports used the expression of the IL-23 receptor as a marker for γδT17 cells. It is true that the expression of the IL-23 receptor faithfully reports the capacity of γδ T cells to produce IL-17, which has been elegantly shown in murine IL-23 receptor reporter models.14,22 However, in our hands, none of the commercially available antibodies to the human IL-23 receptor was either sensitive or specific enough to faithfully stain those γδ T cells that were to express the molecule based on either real-time messenger RNA analysis or intracellular cytokine staining for IL-17. Therefore, we relied on alternative markers, ie, CD161 and CCR6, that have been proven to be associated with the production of IL-17.12,23 Herein, we show that CCR6 expression together with CD161 on γδ T cells discriminates between IL-17–producing (CD161high CCR6+) and IL-17–negative (CD161low CCR6−) γδ T cells.
Since we found an increase in CD161high–expressing γδ T cells in the CSF under inflammatory conditions (be it infectious or autoimmune), we hypothesize that the expression of the C-type lectin-like receptor CD161 might indicate target tissue tropism of γδ T cells into the CSF compartment under inflammatory conditions. Although binding to its ligand LLT1 might promote the entry of CD161-positive immune cells into specific tissue niches, it is unclear whether CD161 has a direct role in the trafficking of immune cell populations to sites of inflammation.24,25 Nevertheless, CD161high CD8+ T cells were recently found to markedly contribute to intralesional immune cell infiltrates in MS lesions24 and CD161+ TH cells appear to accumulate in inflammatory gut lesions.26 Moreover, the CD161 gene encoding the CD161 (KLRB1) receptor was identified as a possible susceptibility candidate gene in MS by the International Multiple Sclerosis Genetics Consortium.27
The expression of the chemokine receptor CCR6 has been linked to the capacity to produce IL-17 in human TH17 cells.12,13 Binding of CCR6 to its ligand CCL20, which is produced by epithelial cells including the plexus epithelium in the CNS, was proposed to guide CCR6+ T cells into the inflamed CNS.28 We and others have found that CCR6 strongly segregates with the capacity to produce IL-17 in noncanonical murine γδ T cells,14,18 and in young human patients with meningitis, IL-17–producing γδ T cells were strongly enriched in the CD161+ CCR6+ T-cell receptor γ9δ2 compartment.20 Herein, we found that CD161high CCR6+ γδ T cells are enriched in the CSF of patients with CIS/MS in relapse, suggesting that this subset of γδ T cells is not only associated with infectious diseases but might also play a role in the immunopathology of CNS-specific autoimmune disorders as well.
In conclusion, in adult individuals, CD161high CCR6+ γδ T cells were specifically enriched in the CSF of patients with CIS/MS in relapse in comparison with systemic autoimmune disorders and noninflammatory neurological diseases. The CD161high CCR6+ subpopulation of γδ T cells is characterized by their IL-17-producing capacity. Hence, we assume that CD161high CCR6+ γδ T cells might be important in the immunopathology of MS and represent a therapeutic target in compartmentalized autoimmune reactions in the CNS.
Correspondence: Thomas Korn, MD, Department of Neurology, Klinikum rechts der Isar, Technische Universität München, Ismaninger Str 22, 81675 Munich, Germany (email@example.com).
Accepted for Publication: August 17, 2012.
Published Online: December 17, 2012. doi:10.1001/2013.jamaneurol.409
Author Contributions:Study concept and design: Schirmer, Rothhammer, Hemmer, and Korn. Acquisition of data: Schirmer, Rothhammer, and Korn. Analysis and interpretation of data: Schirmer, Rothhammer, and Korn. Drafting of the manuscript: Schirmer and Korn. Critical revision of the manuscript for important intellectual content: Schirmer, Rothhammer, Hemmer, and Korn. Statistical analysis: Schirmer, Rothhammer, and Korn. Obtained funding: Schirmer, Hemmer, and Korn. Administrative, technical, and material support: Rothhammer, Hemmer, and Korn. Study supervision: Korn.
Conflict of Interest Disclosures: None reported.
Funding/Support: Dr Schirmer was supported by a Du Pré Grant of the Multiple Sclerosis International Federation and intramural funding of the medical faculty of the Technische Universität München. Dr Korn is the recipient of a Heisenberg grant from the DFG and is supported by other grants of the DFG. Dr Hemmer was supported by a grant from the German Ministry for Education and Research (“German Competence Network Multiple Sclerosis,” Control-MS).
Additional Contributions: We thank Verena Grummel and Tobias Keiner for skillful technical assistance.