SEARCH

SEARCH BY CITATION

Keywords:

  • Chemokines;
  • EAE/MS;
  • Inflammation;
  • T cells

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

The T-cell subsets, characterized by their cytokine production profiles and immune regulatory functions, depend on correct in vivo location to interact with accessory or target cells for effective immune responses. Differentiation of naive CD4+ T cells into effectors is accompanied by sequentially regulated expression of the chemokine receptors responsible for cell recruitment to specific tissues. We studied CCR6 function in EAE, a CD4+ T-cell-mediated CNS disease characterized by mononuclear infiltration and demyelination. CCR6−/− mice showed an altered time course of EAE development, with delayed onset, a higher clinical score, and more persistent symptoms than in controls. An imbalanced cytokine profile and reduced Foxp3+ cell frequency characterized CNS tissues from CCR6−/− compared with CCR6+/+ mice during the disease effector phase. Transfer of CCR6+/+ Treg to CCR6−/− mice the day before EAE induction reduced the clinical score associated with an increased in infiltrating Foxp3+ cells and recovery of the cytokine balance in CCR6−/− mouse CNS. Competitive assays between CCR6+/+ and CCR6−/− Treg adoptively transferred to CCR6−/− mice showed impaired ability of CCR6−/− Treg to infiltrate CNS tissues in EAE-affected mice. Our data indicate a CCR6 requirement by CD4+ Treg to downregulate the CNS inflammatory process and neurological signs associated with EAE.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

EAE is a CNS disease characterized by mononuclear cell infiltration and demyelination; it is used to study certain aspects of human MS. EAE can be induced via immunization with neural antigens such as myelin oligodendrocyte glycoprotein (MOG). The immunopathological event in EAE and MS initiates when autoreactive T cells in the systemic immune compartment are activated and cross the blood–brain barrier 1. Re-encounter of encephalitogenic T cells with their antigen leads to reactivation and expansion of autoreactive T cells, which in turn stimulate microglia/astrocyte activity, with increased release of proinflammatory cytokines and chemokines. The action of these mediators leads to demyelination and axon degeneration 2, 3. After antigen contact, naive CD4+ T cells differentiate into various effector-cell subsets characterized by the cytokines they produce and by their immune regulatory functions. Th1 and Th17 cytokine profile effector cells and antigen-specific Treg have a critical role in EAE pathogenesis 4–10. In MOG-induced EAE, both antigen-specific T-effector and Treg differentiate and proliferate in the periphery before migrating to the CNS 9. Differentiation is accompanied by sequential expression of selectins, integrins, and chemokine receptors responsible for T-cell-subset recruitment to and extravasation at inflammation sites. The correct in vivo location of these cell subsets, necessary for their interaction with accessory or target cells, is regulated by chemokines and their receptors 11, which are assumed to have a critical impact on MS and EAE pathogenesis 12.

Studies in EAE models suggest that the chemokine CCL20 and its receptor CCR6 are involved in MS and EAE. CCL20 expression is upregulated in both the CNS and the draining LN (DLN) during clinical disease; the major intracerebral source of CCL20 is infiltrating lymphocytes at disease onset and astrocytes during relapse 13, 14. Administration of neutralizing anti-CCL20 Ab at the time of EAE induction reduces disease severity, suggesting a role for this chemokine in the sensitization phase of EAE, although adoptive transfer experiments with reactive T cells indicate that CCL20 is not necessary for the effector phase of disease 14. After disease induction in mice, CCR6+ Treg are enriched in peripheral blood and accumulate in the CNS 15. This cell subset appears to represent a population of regulatory effector-memory T cells that control potentially destructive immune responses directly in inflamed tissues 15, 16. Autoantigen-specific natural Treg are expanded in the peripheral lymphoid compartment after MOG immunization and are targeted to the CNS during EAE, resulting in in situ accumulation 9. Naive T cells develop reciprocally into Treg or pathogenic Th17 cells, depending on the presence or absence of IL-6 in the local cytokine milieu 17–19. The inflammatory environment not only controls T-cell differentiation but also affects Treg suppressor function in the target tissue 9.

CCR6 is a common marker of certain tissue-homing Treg 15, 16 and Th17 cells 20; these two cell types show a clear functional dichotomy in inflammatory processes, and both are directly involved in autoimmune disease regulation 18, 19. EAE is thus an excellent model for analysis of the CD4/Th1/Th17/Treg pathway, and CCR6−/− mice are useful for determining the function of this receptor in in vivo EAE development.

We show delayed EAE onset in CCR6−/− mice, followed by more neurological damage than controls in the MOG-induced EAE model. CCR6−/− mice developed a standard peripheral response to the antigen; nonetheless, CCR6−/− Treg recruitment to inflamed CNS target tissue was impaired. This altered the balance between effector/Treg subsets during EAE in CCR6−/− mice. Our results indicate that CCR6 modulates the development of MOG-induced EAE in C57BL/6 mice and it has an important role in resolution of the inflammatory process in this model.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Effect of CCR6 deficiency on the course of MOG35–55-induced EAE

To examine the effect of CCR6 deficiency on the course of disease, we performed four independent sets of EAE induction experiments. Although both CCR6+/+ and CCR6−/− mice developed disease, the time course of EAE development was altered in CCR6−/− compared with CCR6+/+ mice. MOG35–55-induced EAE showed typical onset and course in controls, with most mice developing hind limb paralysis. In CCR6−/− mice, the effector phase of the disease was clearly altered. Disease incidence showed no significant differences (89% WT; 85% CCR6−/− mice), but disease onset was consistently delayed in CCR6−/− mice (Fig. 1A and B). At day 13–14 post immunization, 50% of WT mice manifested neurological symptoms (clinical score, CS>1), whereas CCR6−/− mice required 16 days. After disease onset, CCR6−/− mice developed more neurological damage than controls and achieved a higher mean CS (2.7 versus 2.2), showed increased mortality (15 versus 7%) and reduced resolution of the EAE attack (disease peak at day 20 versus day 15) (Fig. 1A). We observed almost complete disease remission (CS≤1) in approximately 50% of mice in both groups, at 30 days post EAE induction in CCR6+/+ and at 50 days in CCR6−/− mice (Fig. 1A). A summary of disease features for both groups is shown in Fig. 1C. These results indicate that absence of CCR6 modulates EAE development in C57BL/6 mice.

thumbnail image

Figure 1. CCR6−/− mice develop a severe course of EAE. CCR6+/+ and CCR6−/− mice on the C57BL/6 background were immunized with 100 μg MOG35–55 peptide. (A) CS for CCR6+/+ (open circles; n=28) and CCR6−/− mice (gray squares; n=28), 57 days post immunization. Scores for mice that died of EAE were included only until day of death. Results are expressed as mean EAE score±SEM for each group. Two-way ANOVA analysis renders a value of p<0.001 when disease development is compared for both groups. Significance level by Student's t-test comparing CS for both groups at the indicated days, *p<0.05, **p<0.01. (B) Delayed EAE onset in CCR6−/− mice. The data show the differences in days for disease onset between CCR6−/− and CCR6+/+ mice, with time 0 corresponding to day of EAE onset in CCR6+/+ mice. Each bar represents an independent experiment in which each experimental group contained n≥6 mice. *p<0.05; ranked Mann–Whitney U test. (C) Features of MOG35–55-induced EAE in CCR6+/+ and CCR6−/− mice; the mean value is shown for four independent experiments; p.i., post-induction.

Download figure to PowerPoint

CCR6 deficiency does not affect encephalitogenic response or cytokine production in peripheral LN

To test whether T-cell priming or encephalitogenic T-cell responses were altered in CCR6-deficient mice, we evaluated proliferation and cytokine production by DLN cells from CCR6–/– and control mice after EAE induction. T cells from both groups showed equivalent proliferative responses to MOG-specific stimuli (Fig. 2A), suggesting that T-cell priming is unimpaired in CCR6−/− T cells. The percentage of IFN-γ-, IL-10- and IL-17-producing cells in DLN were unaffected by lack of CCR6. In both mouse groups, the frequency of DLN CD4+ cells producing IFN-γ or IL-17 showed a moderate increase during disease onset and the acute phase. IL-10+ cell frequency decreased from pre-onset to the acute phase, although values recovered at the peak of clinical symptoms (Fig. 2B). In MOG35–55 peptide-stimulated LN cell cultures, both the percentage of CD4+ T cells expressing IFN-γ or IL-10 (Fig. 2C) as well as cytokine accumulation (Fig. 2D) were equivalent in both mouse groups. We also analyzed the humoral response to the MOG35–55 peptide and the MOG protein in CCR6−/− and control mice, and found equivalent responses in both groups during the first month after EAE induction (data not shown). This indicates that the differences in EAE disease shown by CCR6−/− mice are not the consequence of an exacerbated humoral response.

thumbnail image

Figure 2. Peripheral cell immune responses to antigen are unaffected by CCR6 deficiency. DLN mononuclear cells from MOG/CFA-immunized mice in acute phase (days 1–3 post disease onset) were isolated from CCR6+/+ (n=12; open circles) and CCR6−/− mice (n=12; gray squares). (A) The T-cell-specific proliferative response was measured after stimulation with MOG35–55 peptide (10 μg/mL, 72 h) and pulsed with 1 μCi [3H]thymidine for the last 16 h. Curves show mean cpm±SD of triplicate values for one representative experiment of four performed. (B) DLN mononuclear cells isolated from CCR6+/+ (n=10) and CCR6−/− mice (n=10) were used for intracellular cytokine staining. Data represent the percentage of cells positive for the indicated cytokines within the DLN CD4+ T-cell population at distinct disease phases. Results show mean±SD for one representative experiment of two performed. (C and D) Cultures of mononuclear cells isolated from DLN and restimulated in vitro with MOG35–55 peptide (10 μg/mL, 72 h) were used for intracellular cytokine staining to determine percentages of (C) IFN-γ- and IL-10-producing cells and to quantify cytokine secretion into the culture medium (D). CCR6+/+ (n=6; open bars); CCR6−/− (n=6; gray bars). Results are indicated as mean±SD for one representative experiment of two performed.

Download figure to PowerPoint

Severe phenotype in CCR6-−/− mice is linked to increased inflammatory profile in EAE-target tissues

During EAE, genes encoding inflammatory molecules (cytokines, chemokines) are strongly induced in the CNS 4, 12, 21. We therefore used real-time quantitative PCR (RT-qPCR) analysis to measure the expression of several of these genes in brain and spinal cord of EAE-affected WT and CCR6−/− mice. We found a greater increase in proinflammatory molecule transcript levels in brain from EAE-affected CCR6−/− mice during the acute phase of disease. Figure 3 shows the fold increase in brain transcript levels in CCR6+/+ and CCR6−/− mice in the acute versus the pre-onset phase of disease. Mice from each group were pooled based on CS and number of days post EAE onset, rather than on days post disease induction. Each point corresponds to a pool of two-to-three mice corresponding to the same day post disease onset and showing a similar CS. Values for IL-1α and IL-6 were clearly higher in CCR6−/− than in WT mice (Fig. 3A). In contrast, the increase in transcript levels for suppressor cytokines in EAE pathology 22, 23 was similar (IL-10) or lower (IL-13) in the CCR6−/− pools (Fig. 3B). The increase in IL-12a, IFN-γ, IL-17A, and TNF-α cytokine transcripts showed by CCR6−/− was similar to or higher than in WT pools. The mice in each pool are homogeneous, but the distinct pools cover the early to late acute phase of disease; we therefore analyzed cytokine transcripts within each pool. Comparison of the mean increase in proinflammatory and downregulatory cytokine transcripts within each pool showed a clear bias toward an inflammatory response in CCR6−/− mice, with a more balanced response in WT mouse pools. The mean ΔΔCt (acute versus pre-onset phase) of proinflammatory (IL-1α, IL-12a, IFN-γ, IL-17, IL-6, TNF-α) and suppressive (IL-10, IL-13) cytokine transcripts is shown for each mouse pool (Fig. 3C). An independent experiment showed a similar pattern for spinal cord samples from both mouse pools (data not shown).

thumbnail image

Figure 3. Increased inflammatory profile in target tissues from EAE-affected CCR6−/− mice compared with WT controls. Data were obtained by RT-qPCR using RNA isolated from the brains of mice pooled according to EAE CS and days post disease onset (two to three mice/pool). Transcript levels were normalized to GAPDH. (A and B) Fold increase in cytokine transcript levels during the acute phase of disease compared with pre-onset phase in CCR6+/+ (open circles) and CCR6−/− pools (gray squares). Each point represents a pool of two to three mice corresponding to the same day post disease onset and showing a similar CS. (A) CCR6−/− pools showed a greater increase in proinflammatory cytokine transcripts, whereas (B) the suppressor cytokine increase is similar to or lower than those of CCR6+/+ groups; *Student's t-test: p<0.05. (C) The mean transcript level increase (ΔΔCt, acute versus pre-onset phase) is shown for proinflammatory (IL-1α+IL-12a+IFN-γ+IL-17+IL-6+TNF-α; dark gray) and suppressor (IL-10+IL-13; light gray) cytokine transcripts for each pool of CCR6+/+ and CCR6−/− mice. (D and E) Fold increase in transcript levels of (D) CCL20 and (E) CCR6 during clinical phases of EAE compared with the pre-onset phase; results are shown as mean±SD in CCR6+/+ (open bars) and CCR6−/− mice (gray bars). Data shown are representative of two independent experiments.

Download figure to PowerPoint

Our results confirmed the overexpression of CCL20 described in CNS during EAE 13, with highest transcript levels during onset and early acute phase. CCR6+/+ and CCR6−/− mice showed similar increases in transcript levels and expression pattern of CCL20 in CNS (Fig. 3D); CCR6 was also upregulated in WT mice, and reached the highest values at the peak stage of disease (Fig. 3E).

Reduced frequency of CD4+Foxp3+ T cells in CNS from EAE-affected CCR6−/−mice

Since EAE-mediated neurological impairment is usually related to leukocyte infiltrates in the acute phase, we used FACS and immunohistochemistry to analyze spinal cord-infiltrating cells in EAE-affected mice. We found no significant differences between spinal cord-infiltrating cell populations in the two mouse groups. Glia activation (measured by CD45, MHC I, CD11b and F4/80 expression levels; data not shown) and percentages of myeloid (CD45highCD11b+) and lymphoid cells (CD45highCD11b) within CNS infiltrates were also similar in WT and CCR6−/− mice during the effector phase of disease (Fig. 4A). The relative percentages of macrophages (CD45highF4/80+) and CD8+ cells infiltrating the spinal cord were also comparable, and there were no significant differences in the distribution and extent of F4/80 immunoreactivity on axial sections of lumbar spinal cord from CCR6+/+ and CCR6−/− mice (Fig. 4B).

thumbnail image

Figure 4. CNS pathology in EAE-affected CCR6+/+ and CCR6−/− mice. Mononuclear cells were isolated from spinal cords of affected mice at different times after EAE induction and examined for leukocyte surface marker expression. (A) The percentages of myeloid (CD45highCD11b+) and lymphoid cells (CD45highCD11b), macrophages (CD45highF4/80+) and CD8+ T cells infiltrating the spinal cord are shown, indicating mean±SD in CCR6+/+ (n=5; open bars) and CCR6−/− mice (n=5; gray bars). (B) Representative photomicrographs of spinal cord sections from CCR6+/+ and CCR6−/− mice, stained with F4/80, anti-CD4, or -Foxp3 mAb. Arrows indicate Foxp3+ cells. (C and D) Reduced infiltration of Treg in spinal cord from EAE-affected CCR6−/− mice compared with WT counterparts. Percentages of (C) CD4+ and (D) Foxp3+ T cells in CD45+-cell populations in spinal cord during the acute disease phase are shown for CCR6+/+ (n=10; open circles) and CCR6−/− mice (n=9; gray squares). *Student's t-test: p<0.05. (E) Ratio of effector (Foxp3) versus regulatory (Foxp3+) cell subsets in CD4+CD25+ T cells infiltrating spinal cord from CCR6+/+ (n=5; open circles) and CCR6−/− mice (n=5; gray squares). Individual mice and the mean (horizontal bar) of all data points are shown. *Student's t-test: p<0.05. (F) Fold increase in the indicated leukocyte surface marker levels in brain-infiltrating cells from CCR6+/+ (n=10; open bars) and CCR6−/− mice (n=9; gray bars), (G) increase in total Foxp3 transcript levels in CCR6+/+ (open circles) and CCR6−/− mice (gray squares), and (H) increase in Foxp3 relative to CD4 brain transcript levels, during the EAE effector phase compared with pre-onset phase. *Student's t-test: p<0.05. (I) Similar percentages of Foxp3+ cells are present in the DLN CD4+ T-cell population from EAE-affected CCR6+/+ (n=10; open bars) and CCR6−/− mice (n=9; gray bars). (J) CCR6 deficiency does not affect the in vitro suppressor function of Treg. Suppression of WT responder cell proliferation (3×105 cells) in response to MOG35–55 peptide by Treg (1.5×105) from CCR6+/+ or CCR6−/− mice. Data show mean cpm±SD of triplicate values. Results show mean±SD for one representative experiment of three performed.

Download figure to PowerPoint

The similar leukocyte infiltration in the EAE effector phase and the higher inflammatory response in CCR6−/− mice suggested a defect in the regulatory population. We compared the time course of CD4+ and CD4+Foxp3+ T-cell infiltration into CCR6−/− and WT mouse target tissues. The percentage of CD4+ cells increased in spinal cord during onset and acute phases, and returned to baseline values in the remission phase in both groups. The relative abundance of CD4+ cells within the spinal cord CD45+-cell population was nonetheless higher in WT (24%) than in CCR6−/− mice (16%) at the peak of disease (Fig. 4C). In addition, the percentage of Foxp3+ cells increased in WT mice from pre-onset to the effector phase, with no significant change in CCR6−/− mice (Fig. 4D). In WT mice, spinal cord-infiltrating Foxp3+ cells represented 30–40% of CD4+ cells at pre-onset, dropping to approximately 7% during onset-acute phase, increasing again at the disease peak and remaining at approximately 30% in remission phase. The percentage of effector CD4+CD25+Foxp3 cells was highest during onset-acute phase, decreasing at peak and remission phases. As a result of the distribution of these T-cell subsets, the ratio between effector/Treg (CD4+CD25+Foxp3versus CD4+CD25+Foxp3+) varied from 1 to 2 in WT mice at the acute-peak phase of disease. In contrast, the balance was biased toward the effector T subset in CCR6−/− mice, most of which showed a ratio > 2 at the same disease phase (Fig. 4E). The relative abundance of Foxp3+ T cells within the CD4+ T-cell subset infiltrating spinal cord was evaluated on axial sections of lumbar spinal cord from CCR6+/+ and CCR6−/− mice. Figure 4B illustrates the reduction in the percentage of Foxp3+ cells in the CD4+ T-cell population infiltrating spinal cord in CCR6–/–versus CCR6+/+ controls.

We also studied brain tissue, the other main EAE target. We used RT-qPCR to measure the relative abundance of lymphocyte lineage marker mRNA in both groups of EAE-affected mice. No significant differences were detected in transcript level increase for lymphocyte markers CD3, CD4, CD8, CD19, or CD25 in brain from CCR6+/+ and CCR6−/− mice during the acute disease phase (Fig. 4F), although the increase in Foxp3 transcripts was greater in WT than in CCR6−/− mice (Fig. 4G). WT mice showed clear enrichment in Foxp3 transcript levels relative to the CD4 transcript from the acute to peak and early remission phases; this was not observed in the CCR6−/− group (Fig. 4H). These results support the evidence for differences between CCR6+/+ and CCR6−/− mice in the relative frequency of Treg in EAE target tissues.

Naive CCR6−/− mice show no defects in the generation or peripheral distribution (lymphoid or non-lymphoid) of Treg 16. We found similar percentages of Foxp3+ cells in DLN from MOG35–55-immunized CCR6−/− or WT mice during disease development (Fig. 4I). We also tested whether the suppressive ability of Treg was affected by CCR6 deficiency. CCR6+/+ and CCR6−/− DLN-derived Treg equally suppressed the proliferative response of MOG-reactive cells to peptide stimulus in vitro (Fig. 4J). These data rule out a defect in Treg generation associated with CCR6 deficiency.

Our results suggest a defect in the percentage and relative frequency of Treg in target tissues of EAE-affected CCR6−/− mice, not associated with defective peripheral Treg generation in response to immunization.

Adoptive transfer of WT Treg to CCR6−/− mice re-established WT disease kinetics

To evaluate whether CCR6−/− mice are able to downregulate the EAE-associated inflammatory response and to clear inflammatory cytokines from the CNS, we adoptively transferred WT Treg to CCR6−/− mice the day before EAE induction. Our rationale was to analyze whether CCR6 expression was necessary for Treg suppressor function in target tissues. Transfer of WT Treg to CCR6−/− mice in these conditions re-established the maximum score and disease remission rate values to those for CCR6+/+ mice, although the delay in disease onset was unaffected. The mean CS for CCR6−/− mice dropped from 2.75 to 2.1 after adoptive transfer of WT Treg; at 25 days, 50% of the mice achieved a CS of 1, compared with 36 days in the non-transferred CCR6−/− group (Fig. 5A). Transfer of WT Treg to CCR6+/+ mice did not affect the kinetic parameters of EAE, altering neither the mean CS nor time to remission (data not shown); this suggests that the changes in transferred CCR6−/− mice were associated to CCR6 and not to differences in Treg numbers in recipients. The percentage of Foxp3+ cells in spinal cord increased threefold in the adoptively transferred CCR6−/− group (1.26 versus 0.45% in non-transferred CCR6−/− mice), whereas no significant differences were found in the transferred CCR6+/+ group. The percentage of Foxp3+ cells in WT Treg-transferred CCR6+/+ (1.4%) and CCR6−/− (1.26%) mouse spinal cord were equivalent to CCR6+/+ controls (1.6%; Fig. 5B). The relative frequency of Foxp3+ cells within the CD4+ T-cell population also recovered in the WT Treg-transferred CCR6−/− group (13%) compared with non-transferred CCR6−/− mice (6.6%; Fig. 5C). As a result, the effector/regulatory CD4+ T-cell subset ratio in spinal cord returned to WT values (1.6 in CCR6+/+ and in WT Treg-transferred CCR6−/− mice versus 3.01 in CCR6−/− mice; Fig. 5D). We detected no alterations in the transferred CCR6+/+ group, in which the percentages of Foxp3+ cells in spinal cord as well as the effector/regulatory CD4+ T-cell subset ratio remained at control values.

thumbnail image

Figure 5. Adoptive transfer of CCR6+/+ Treg to CCR6−/− mice before EAE induction abolished the severe disease phenotype. WT and CCR6−/− mice were adoptively transferred with CD4+CCR6+ Treg (3×105) 24 h before MOG/CFA immunization. (A) EAE disease development in CCR6−/− recipients (n=5; black triangles) of CCR6+/+ Treg non-transferred CCR6+/+ (n=9; open circles) and CCR6−/− mice (n=9; gray squares) are shown. Significance level using the Mann–Whitney test, *p<0.05 comparing non-transferred CCR6−/− and WT Treg-transferred CCR6−/− mouse groups. Error bars represent +SEM. (B–D) CCR6+/+ Treg adoptively transferred into CCR6−/− mice re-established (B) the percentages of Foxp3+ cells, (C) the frequency of Foxp3+ cells within the CD4+ T-cell population, and (D) the effector/regulatory CD4+ T-cell ratio in spinal cord of EAE-affected CCR6−/− mice to WT values (non-transferred CCR6+/+). Results indicate mean±SD in non-transferred CCR6+/+ controls (n=6; open bars), non-transferred CCR6−/− (n=5; light gray bars), WT Treg-transferred CCR6+/+ (n=7; dark gray bars), and WT Treg-transferred CCR6−/− mouse groups (n=7; black bars). Student's t-test: *p<0.05 comparing non-transferred CCR6−/− with WT Treg-transferred CCR6−/− mouse groups. (E) CCR6+/+ Treg downregulate the inflammatory profile associated with EAE-affected CCR6−/− mice. Results indicate the mean fold increase of indicated cytokine transcript levels in spinal cord from WT Treg-transferred (n=3; CS=3) versus non-transferred CCR6−/− mouse groups (n=3; CS=3). (F) CCR6+/+ Treg transferred to CCR6−/− mice re-established Foxp3 and IL-10 transcript levels in recipient CCR6−/− mouse brain to values seen in non-transferred CCR6+/+ mice. Mouse groups are labeled as in (B—D). Student's t-test: *p<0.05 for non-transferred CCR6−/−versus WT Treg-transferred CCR6−/− mice. The experiments were performed twice with similar results.

Download figure to PowerPoint

The adoptively transferred WT Treg reduced the inflammatory profile of CCR6−/− mouse EAE target tissues. In the acute disease phase, IL-17, TNF-α, IFN-γ, and IL-6 transcript levels decreased and those of IL-10 increased in spinal cord of adoptively transferred CCR6−/− mice compared with the non-transferred CCR6−/− group (for both groups, we compared mice with CS 3 at day 3 post disease onset); Fig. 5E shows the fold increase or decrease for each transcript compared with values for non-transferred CCR6−/− mice. We used RT-qPCR to measure Foxp3 and IL-10 transcript levels in brain samples from non-transferred and WT Treg-transferred CCR6+/+ and CCR6−/− mice. Values were relative to CD4 transcript levels in the samples. WT mouse brain was enriched in Foxp3 transcripts (relative to the CD4 transcript) compared with the CCR6−/− group (Fig. 4H). WT Treg transfer to CCR6−/− mice the day before EAE induction yielded a 2.2-fold increase in Foxp3 transcript levels in brain tissue, which reached WT values (Fig. 5F). IL-10 transcript levels in WT Treg-transferred CCR6−/− mouse brain also returned to WT values (Fig. 5F).

These results showed that transfer of WT Treg to CCR6−/− mice before EAE induction downregulated the associated inflammatory profile in target tissues and ameliorated neurological symptoms.

CCR6 expression enables Treg recruitment to EAE target tissues

The adoptive transfer data suggested a defect in CCR6−/− Treg recruitment to or retention in EAE-damaged tissues. We used in vivo competitive assays between WT and CCR6−/− Treg to examine this possibility. CD4+ Treg from CCR6+/+GFP-expressing mice (WTTREGgfp) were adoptively transferred, alone or mixed (1:1 ratio) with unlabeled CCR6+/+ or CCR6−/− Treg, into CCR6−/− mice the day before EAE induction (total 3×105 cells in all conditions). We quantified GFP-labeled and unlabeled Treg in DLN and in spinal cord infiltrate in the acute EAE phase. In mice that received WTTREGgfp cells alone, 50–60% of Treg infiltrating the spinal cord were GFP+; this percentage was reduced to ∼30% in WTTREGgfp/CCR6+/+ Treg-transferred CCR6−/− mice (Fig. 6A). When we transferred WTTREGgfp/CCR6−/− Treg, however, there was no dilution effect in the percentage of WTTREGgfp-expressing cells relative to total Treg in the spinal cord; ∼70% of Treg were GFP+ (Fig. 6A). The relative frequency of total Treg (Foxp3+) within the CD4+ T-cell population infiltrating the spinal cord was similar in all conditions tested, suggesting a specific defect of CCR6−/− Treg in EAE target tissue infiltration (Fig. 6B). The frequency of GFP-expressing cells within the Treg population (Fig. 6C) and the percentage of Treg in DLN (Fig. 6D) were also similar in both mouse groups. These data support our hypothesis that CCR6 expression is required for Treg infiltration to or retention in the inflamed CNS, but not for homing to LN.

thumbnail image

Figure 6. Impaired CCR6−/− Treg recruitment to EAE target tissues compared with WT Treg. CCR6−/− mice received 3×105 CD4+CCR6+ Treg GFP+ cells (WTTREGgfp (n=3; open bars), 1.5×105 WTTREGgfp plus 1.5×105 non-GFP-expressing CD4+ CCR6+/+ Treg (n=3; light gray bars), or 1.5×105 WTTREGgfp plus 1.5×105 non-GFP-expressing CD4+CCR6−/− Treg (n=3; dark gray bars) 24 h before EAE induction with MOG/CFA. After 15 days, (A and B) spinal cord and (C and D) DLN CD4+ T cells were analyzed for each mouse group. The frequency of GFP-expressing cells within the Treg population infiltrating (A) spinal cord and (C) DLN, and the percentage of Treg in (B) spinal cord and (D) DLN are shown (mean±SD) for each group of transferred CCR6−/− mice. Student's t-test: *p<0.05. Data are shown from one representative experiment of two performed.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Here, we show altered time course development of disease-associated symptoms in MOG35–55-induced EAE in CCR6−/− mice compared with WT controls. Although disease onset was consistently delayed in the CCR6−/− group, the course of disease was more severe in these mice following the appearance of clinical symptoms. This indicates that CCR6 is not an absolute requirement for disease development; nonetheless, given the delayed onset in CCR6−/− mice, our data suggest CCR6 influence on an effector-cell population at the onset of disease. CCR6 expression defines functional antigen-specific Th17 cells that induce EAE 20. Deficiency in IL-17 or in IL-23, which increases or stabilizes IL-17 production, confers partial or complete resistance to MOG-induced EAE in C57BL/6 mice 5, 24. Very recently, Yamazaki et al. 25 described a reduction in clinical symptoms in an EAE model driven by MOG-reactive CD4+CCR6−/− T cells in vitro polarized to the Th17 phenotype. Overall, these results suggest a role for CCR6 in the Th17 population in EAE, probably facilitating homing of these cells to inflamed tissues. Nonetheless, CCR6 deficiency confers only transient resistance to EAE disease in mice. Studies of two different CCR6−/− mouse strains (those from Dong's group 25 and our own) showed that these mice develop disease with CS similar to 25 or higher than those of WT controls. O'Connor et al. 26 nonetheless established that in the absence of IFN-γ+ cells, Th17 cells do not induce EAE. Using Th1 and Th17 cells and avoiding cross-contamination between the polarized cell populations, these authors show that only Th1 cells have access to the non-inflamed CNS. Once Th1 cells have provoked the experimental lesions, Th17 cells appear in the CNS.

CCR6-mediated Th17-cell migration to the CNS could be important for driving CNS inflammation at early time points. Since the disease score in CCR6−/− mice reached or exceeded that of WT mice despite delayed onset, however, other chemokine receptors expressed by effector Th17 cells might assist the migration of these cells to the CNS. Alternatively, other effector T-cell populations replace their function. In EAE-affected WT mice, only 50% of Th17 cells express CCR6 25; in addition, IL--17A transcript can be detected in the CNS of EAE-affected CCR6−/− mice.

We cannot discard a role for CCR6 early in immune response generation. Reduced disease severity following administration of neutralizing anti-CCL20 mAb at the time of EAE induction led Kohler et al. to suggest a role for CCL20 also in the sensitization phase 14. Whether or not the CCR6/CCL20 pair facilitates antigen presentation during preclinical disease, our results show that CCR6 deficiency does not alter the development or function of MOG-primed cells in lymphoid organs. MOG peptide-reactive cell generation, percentages of cytokine (IFN-γ, IL-10, IL-17)-expressing CD4+ T cells, and cytokine production in response to T-cell stimulation were similar in DLN from CCR6−/− and CCR6+/+ mice at all times after disease onset. During the clinical phase of disease, CCR6−/− mice showed no defects in Ab generation to MOG peptide or to myelin protein. Furthermore, final CS for EAE-associated symptoms were higher in CCR6−/− mice, indicating that once clinical disease appears, CCR6 is not essential during the effector phase. Our hypothesis concurs with reports that CCL20 neutralization does not prevent passive transfer of EAE using encephalitogenic T cells 14, indicating that the CCR6/CCL20 pair is not necessary for the disease effector phase.

We show that Th17 cells development in vivo is unaffected by CCR6 deficiency, either in naive mice 27 or in response to antigen; percentages of IL-17-expressing CD4+ T cells in DLN were similar in MOG-immunized CCR6−/− and WT mice. Semiquantitative RT-qPCR analysis of brain and spinal cord from EAE-affected mice confirmed similar or higher IL-17 transcript levels in CCR6−/− target tissues. The cytokine profile in CCR6−/− target tissues showed a preferential increase in proinflammatory (IL-1α, IL-6) versus downregulatory (IL-10, IL-13) transcript levels compared with WT counterparts, in accordance with the greater neurological impairment in CCR6−/− mice. As a result, these mice show an imbalanced cytokine profile, which could account for their recovery delay and longer-lasting neurological damage.

Other authors have shown that distinct effector-cell populations can mediate similar clinical effects in EAE 28. Adoptive transfer of IL-12p70- or IL-23-polarized myelin-reactive T cells into naive hosts results in paralysis that is clinically indistinguishable between the two groups; the disease induced by each of these cell lines nonetheless differs in CNS chemokine expression patterns as well as in the composition and location of infiltrating leukocyte populations within the spinal cord 28. In our system, we detected no differences in the characteristics or distribution of the cellular infiltrate in spinal cord tissue from EAE-affected CCR6−/− and CCR6+/+ mice, except for the CD4+ T-cell population.

CCR6 is found not only in effector T cells but also in antigen-experienced effector/Treg subsets 15, 16. The prominence of Treg-produced IL-10 as a suppressor cytokine in inflammatory conditions, the greater increase in proinflammatory cytokine transcript levels in target tissues of EAE-affected CCR6−/− mice compared with WT, as well as the more persistent clinical symptoms in CCR6−/− mice suggested impairment of Treg function in vivo. We reported normal Treg distribution in the periphery of untreated CCR6−/− mice 16. Our results here show similar generation of Foxp3+ cells in DLN from MOG35–55-immunized WT and CCR6−/− mice. In addition, in vitro suppression assays showed that Treg from both mouse groups have similar inhibitory potential for MOG-reactive T-cell proliferation. Although the overall results indicated no CCR6-deficiency-associated anomalies either in the generation or the in vitro suppressive ability of Treg, CCR6−/− mice showed a 75% reduction in Foxp3+ cells infiltrating spinal cord at the peak phase of disease (Fig. 4). As a result, the T effector:Treg ratio was significantly higher in CCR6−/− than in WT mice. In CD4+ T cells infiltrating the brain, the Foxp3 transcript level was lower in CCR6−/− mice than in WT counterparts.

Treg have a role in modulating EAE 8, 10, 29. Here we found that CCR6 is required for appropriate downregulation of the inflammatory process in target tissues, suggesting that this receptor is necessary for Treg recruitment to these tissues. To test this hypothesis, we used two complementary adoptive transfer approaches. In the first, we analyzed whether CCR6−/− mice generated a normal downregulatory response after adoptive transfer of WT Treg prior to EAE induction. In the second, we compared the ability of WT and CCR6−/− Treg to reach EAE target tissues in direct in vivo competitive assays. WT Treg adoptively transferred to CCR6−/− mice re-established CS and remission kinetics to WT values, although disease onset continued to be delayed. The inflammatory profile of EAE target tissues in CCR6−/− mice was downregulated by adoptively transferred WT Treg, and the percentage of Foxp3+ cells infiltrating EAE target tissues recovered. Kohm et al. reported that adoptive transfer of relatively large numbers (2×106) of CD4+CD25+ Treg from naive mice protected recipients from MOG-induced EAE 8. We found no protective effect in either WT or CCR6−/− recipients; differences in the number of Treg transferred, which was sevenfold higher in the experiments of Kohm et al., may account for this discrepancy.

TNF-α and IL-6 produced by encephalitogenic effector T cells in the inflamed CNS are reported to suppress Treg function in this tissue. Treg that have migrated to the CNS therefore cannot regulate encephalitogenic effector T cells in mice that develop EAE 9. Not only Treg numbers but also the inflammatory cytokine milieu might determine whether Treg control autoimmune inflammation. Our data also indicate that transferred WT Treg are subject to this inhibitory cytokine milieu in the CNS, explaining why the effector phase of disease takes place even in Treg-transferred CCR6+/+ and CCR6−/− mice. It thus appears that the effector/regulatory cell ratio within the CNS and the resulting cytokine levels are the main factors driving disease remission.

To further test CCR6 function in the Treg subset, we analyzed the ability of CCR6−/− Treg to compete in vivo with WT counterparts in reaching EAE target tissues. Approximately 70% of Treg from WT donors infiltrated CCR6−/− target tissues, even when a 1:1 mixture of CCR6+/+/CCR6−/− Treg was adoptively transferred to CCR6−/− hosts. There were no differences in Treg percentages in spinal cord or LN, nor did we find alterations in the relative frequency of Treg from either donor in LN. Autoantigen-specific natural Treg are expanded in the peripheral lymphoid compartment after MOG immunization and are targeted to the CNS during EAE, resulting in in situ accumulation 9. Our results indicate that this Treg recruitment to EAE-affected CNS tissues is CCR6-dependent.

Defective Treg recruitment to damaged CNS was also described by Yamazaki et al. 25, who analyzed EAE outcome in WT mice receiving in vitro-differentiated CCR6+/+ or CCR6−/− Th17 cells and immunized with MOG. They argue that as Treg still express CCR6 in host mice, the absence of Treg in the CNS of CCR6−/− Th17-cell recipients indicates that Treg do not infiltrate the damaged tissue due to lack of Th17-cell-produced CCL20 in the CNS. Although Th17 cells might have a role as CCL20 producers and so attract Treg migration to the damaged CNS, our results in the MOG-induced model show no defects in CNS CCL20 levels in affected CCR6−/− mice compared with WT mice at the same disease stage. The kinetics of CCL20 transcript increase is the same, with higher levels at onset and early acute phases that drop in the late acute stage and during remission. Furthermore, WT Treg transfer to EAE-affected CCR6−/− mice downregulates the inflammatory response, with recovery of the clinical parameters shown by WT mice in terms of disease score and remission; delayed onset nonetheless persists. Our results thus demonstrate that WT Treg migrate to the CNS of CCR6−/− EAE-affected mice.

In this study, we demonstrate a requirement for CCR6 in CD4+ Treg to downregulate the CNS inflammatory process and neurological signs associated with EAE pathogenesis.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

Mice

All mice were bred in-house. C57BL/6 mice were originally purchased from the Jackson Laboratory. CCR6−/− mice were generated in our laboratory 30 and backcrossed for nine generations onto the C57BL/6 background. The C57BL/6-Tg (ACTB-EGFP)1Osb strain was kindly provided by Dr. Masaru Okabe (Osaka University, Japan). Experiments were performed with gender-matched CCR6+/+ (WT, controls) and CCR6−/− mice derived from heterozygote breeding, conducted according to national and European Union guidelines for experimentation in animals, and were approved by the CNB Ethics Committee.

EAE induction and assessment

EAE was induced as described previously 31. Eight- to 10-wk-old male mice received s.c. injections of 100 μg MOG35–55 peptide (MEVGWYRSPFSRVVHLYRNGK) emulsified in CFA (Difco) containing 100 μg Mycobacterium tuberculosis. Mice received i.v. injections of 200 ng pertussis toxin (Sigma) on days 0 and 2 post-immunization. Mice were weighed and graded daily after disease onset in a blind manner for neurological signs, as follows: 0, no disease; 0.5, decreased tail tone; 1, toneless tail; 2, hindlimb weakness, with difficulty in recovering upright position; 3, monolateral hindlimb paralysis; 4, bilateral hindlimb paralysis; 5, death, moribund state or inability to feed, leading to euthanasia.

The mean CS was calculated as the arithmetic mean of individual scores. The average day of EAE onset was calculated by adding the first day that each mouse showed CS 1 and dividing by the number of mice in the group. Disease remission was determined as the day at which the mean CS reached a value of 1 in surviving mice.

Disease phases were assigned according to CS and days after disease onset as follows: onset, score 0.5–1, with symptoms arising in the last 24 h; acute phase, score 1.5–3 or 2–3 days post onset; peak, score 3.5–5 or 4–5 days after onset.

Isolation of spinal cord, spleen, and LN cells

Spinal cords were extruded by flushing the vertebral canal with PBS and rinsed in PBS. Single-cell suspensions were obtained by forcing tissues through 70 μm nylon cell strainers (BD Falcon). CNS mononuclear cells were resuspended in 40% isotonic Percoll, overlaid on 70% Percoll (BD Healthcare), and centrifuged (1200g, 4°C, 20 min). The top myelin layer was removed before harvesting mononuclear cells at the interface. Cells were washed and used immediately for ex vivo surface staining or in vitro proliferation studies. For direct comparison, DLN (axillary and inguinal LN) and spleen were removed and single-cell suspensions obtained by forcing tissue through 70 μm nylon cell strainers.

Treg used in adoptive transfer experiments were isolated from spleen. After erythrocyte lysis in 0.83% w/v NH4Cl, cells were stained with anti-mouse CD4-APC and CD25-PE mAb (BD-Pharmingen) and selected as CD4+CD25+ (Treg) and CD4+CD25 populations (>98%) in an Epics Altra sorter (Beckman Coulter). Foxp3 expression was detected in ≥90% of these CD4+CD25+ Treg.

Suppression and T-cell proliferation assays and cytokine production

For T-cell proliferation and quantification of cytokine production, DLN cells were harvested from MOG35–55-immunized mice and incubated in 96-well plates with MOG35–55 peptide (10 μg/mL). Suppression assays were performed in 96-well plates with 3×105 CD4+ T cells and titrated amounts of Treg. All tests were performed in triplicate. At 16 h before termination of culture, cells were pulsed with 1 μCi/well (3H)thymidine (Amersham Biosciences) to evaluate proliferation. For cytokine production by MOG-induced (4×105 cells/200 μL/well) DLN cells, supernatants were collected at 72 h and tested using the multiplex Biomarker Immunoassay (Linco).

Intracellular staining

Percentages of cytokine-producing cells were determined by intracellular staining with the appropriate mAb after cell incubation (4 h) with PMA (50 ng/mL)+ionomycin (1 μg/mL) and brefeldin A (2 μg/mL). The percentage of Treg was monitored by Foxp3 transcription factor expression. All intracellular staining kits were from BD Bioscience.

Immunohistochemistry

Mice from each group with a CS near average group outcome score were used for immunohistochemical analysis. Spinal cord samples were embedded in Jung tissue freezing medium (Leica) and snap-frozen. Sections (10 μm) were air-dried, acetone-fixed (−20°C, 10 min), and stored at −80°C. Endogenous peroxidase was inactivated in buffered 0.3% H2O2 (20 min). Endogenous biotin was quenched with an avidin/biotin blocking kit (Vector Laboratories). We used the Tyramide Signal Amplification kit (PerkinElmer Life Science); biotinylated Foxp3, CD4, and F4/80 mAb were from Pharmingen. Diaminobenzidine/hydrogen peroxide (Sigma) was used for chromogenic visualization.

RT-qPCR analysis

PolyA+ RNA was extracted using TRI-reagent (Sigma) and the oligotex mRNA kit (Qiagen). Reverse transcription was performed with the SuperScript first strand synthesis system for RT-PCR (Invitrogen), using random hexamers as primers. GAPDH was used as an endogenous control. RT-qPCR was performed with TaqMan low-density arrays (TLDA Mouse Immune panel, Applied Biosystems), in a 7900HT Sequence Detection System (Applied Biosystems). Relative expression levels of genes of interest were calculated by the 2ΔΔCt method.

Statistical analyses

Student's t-test was used to compare levels of T-cell proliferation, cytokine production and expression, and CNS-cell infiltration. Two-way ANOVA, Mann–Whitney test and Student's t-test were used to compare disease parameters between CCR6+/+ and CCR6−/− mice. A p value <0.05 was considered significant.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References

We thank Drs. S. Mañes, R. Lacalle and L. Planelles for critical reading and helpful discussion of the manuscript, Dr. M. Okabe for the gift of C57BL/6-Tg (ACTB-EGFP)1Osb mice, M.C. Moreno-Ortiz, A. Moreno, and S. Escudero for help with flow cytometry, L. Gómez, A. Morales, and S. Rodríguez for animal handling, and C. Mark for editorial assistance. This work was supported in part by a EU grant (INNOCHEM 518187). The Department of Immunology and Oncology was founded and supported by the Spanish National Research Council (CSIC) and by Pfizer.

Conflict of interest: The authors declare no financial or commercial conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements
  8. References
  • 1
    Magnus, T. and Rao, M. S., Neural stem cells in inflammatory CNS diseases: mechanisms and therapy. J. Cell. Mol. Med. 2005. 9: 303319.
  • 2
    Engelhardt, B. and Ransohoff, R. M., The ins and outs of T-lymphocyte trafficking to the CNS: anatomical sites and molecular mechanisms. Trends Immunol. 2005. 26: 485495.
  • 3
    Rebenko-Moll, N. M., Liu, L., Cardona, A. and Ransohoff, R. M., Chemokines, mononuclear cells and the nervous system: heaven (or hell) is in the details. Curr. Opin. Immunol. 2006. 8: 683689.
  • 4
    Lock, C., Hermans, G., Pedotti, R., Brendolan, A., Schadt, E., Garren, H., Langer-Gould, A. et al., Gene-microarray analysis of multiple sclerosis lesions yields new targets validated in autoimmune encephalomyelitis. Nat. Med. 2002. 8: 500508.
  • 5
    Komiyama, Y., Nakae, S., Matsuki, T., Nambu, A., Ishigame, H., Kakuta, S., Sudo, K. and Iwakura, Y., IL-17 plays an important role in the development of experimental autoimmune encephalomyelitis. J. Immunol. 2006. 177: 566573.
  • 6
    Weaver, C. T., Hatton, R. D., Mangan, P. R. and Harrington, L. E., IL-17 family cytokines and the expanding diversity of effector T cell lineages. Annu. Rev. Immunol. 2007. 25: 821852.
  • 7
    Bettelli, E., Korn, T., Oukka, M. and Kuchroo, V. K., Induction and effector functions of TH17 cells. Nature 2008. 453: 10511057.
  • 8
    Kohm, A. P., Carpentier, P. A., Anger, H. A. and Miller, S. D., Cutting edge: CD4+CD25+ regulatory T cells suppress antigen-specific autoreactive immune responses and central nervous system inflammation during active experimental autoimmune encephalomyelitis. J. Immunol. 2002. 169: 47124716.
  • 9
    Korn, T., Reddy, J., Gao, W., Betelli, E., Awasthi, A., Petersen, T. R., Backstrom, B. T. et al., Myelin-specific regulatory T cells accumulate in the CNS but fail to control autoimmune inflammation. Nat. Med. 2007. 13: 423431.
  • 10
    Matsumoto, Y., Sakuma, H., Kohyama, K. and Park, I. K., Paralysis of CD4+CD25+ regulatory T cell response in chronic autoimmune encephalomyelitis. J. Neuroimmunol. 2007. 187: 4454.
  • 11
    Rot, A. and von Andrian, U. H., Chemokines in innate and adaptive host defense: basic chemokinese grammar for immune cells. Annu. Rev. Immunol. 2004. 22: 891928.
  • 12
    Ubogu, E. E., Cossoy, M. B. and Ransohoff, R. M., The expression and function of chemokines involved in CNS inflammation. Trends Pharmacol. Sci. 2006. 27: 4855.
  • 13
    Ambrosini, E., Columba-Cabezas, S., Serafini, B., Muscell, A. and Aloisi, F., Astrocytes are the major intracerebral source of macrophage inflammatory protein-3alpha/CCL20 in relapsing experimental autoimmune encephalomyelitis and in vitro. Glia 2003. 41: 290300.
  • 14
    Kohler, R. E., Caon, A. C., Willenborg, D. O., Clark-Lewis, I. and McColl, S. R., A role for macrophage inflammatory protein-3 alpha/CC chemokine ligand 20 in immune priming during T cell-mediated inflammation of the central nervous system. J. Immunol. 2003. 170: 62986306.
  • 15
    Kleinewietfeld, M., Puentes, F., Borsellino, G., Battistini, L., Rötzschke, O. and Falk, K., CCR6 expression defines regulatory effector/memory-like cells within the CD25+CD4+ T-cell subset. Blood 2005. 105: 28772886.
  • 16
    Varona, R., Cadenas, V., Lozano, M., Moreno-Ortiz, M. C., Kremer, L., Martínez-A, C. and Márquez, G., CCR6 regulates the function of alloreactive and regulatory CD4+ T cells during acute graft-versus-host disease. Leuk. Lymphoma 2006. 47:14691476.
  • 17
    Veldhoen, M., Hocking, R. J., Atkins, C. J., Locksley, R. M. and Stockinger, B., TGFbeta in the context of an inflammatory cytokine milieu supports de novo differentiation of IL-17 producing T cells. Immunity 2006. 24: 179189.
  • 18
    Bettelli, E., Carrier, Y., Gao, W., Korn, T., Strom, T. B., Oukka, M., Weiner, H. L. and Kuchroo, V. K., Reciprocal developmental pathways for the generation of pathogenic effector Th17 and regulatory T cells. Nature 2006. 441: 235238.
  • 19
    Yang, X. O., Nurieva, R., Martinez, G. J., Kang, H. S., Chung, Y., Pappu, B. P., Shah, B. et al., Molecular antagonism and plasticity of regulatory and inflammatory T cell programs. Immunity 2008. 29: 113.
  • 20
    Potzl, J., Botteron, C., Tausch, E., Pedre, X., Mueller, A. M., Mannel, D. N. and Lechner, A., Tracing functional antigen-specific CCR6+ Th17 cells after vaccination. PLoS ONE 2008. 3: e2951.
  • 21
    Sorensen, T. L., Tani, M., Jensen, J., Pierce, V., Lucchinetti, C., Folcik, V. A., Qin, S. et al., Expression of specific chemokines and chemokine receptors in the central nervous system of multiple sclerosis patients. J. Clin. Invest. 1999. 103: 807815.
  • 22
    Zhang, X., Koldzic, D. N., Izikson, L., Reddy, J., Nazareno, R. F., Sakaguchi, S., Kuchroo, V. K. and Weiner, H. L., IL-10 is involved in the suppression of experimental autoimmune encephalomyelitis by CD25+CD4+ regulatory T cells. Int. Immunol. 2004. 16: 249256.
  • 23
    Kleinschek, M. A., Owyang, A. M., Joyce-Shaikh, B., Langrish, C. L., Chen, Y., Gorman, D. M., Blumenschein, W. M. et al., IL-25 regulates Th17 function in autoimmune inflammation. J. Exp. Med. 2007. 204: 161170.
  • 24
    Cua, D. J., Sherlock, J., Chen, Y., Murphy, C. A., Joyce, B., Seymour, B., Lucian, L. et al., Interleukin-23 rather than interleukin-12 is the critical cytokine for autoimmune inflammation of the brain. Nature 2003. 421: 744748.
  • 25
    Yamazaki, T., Yang, O. X., Chung, Y., Fukunaga, A., Nurieva, R., Pappu, B., Martin-Orozco, N. et al., CCR6 regulates the migration of inflammatory and regulatory T cells. J. Immunol. 2008. 181: 83918401.
  • 26
    O'Connor, R. A., Prendergast, C. T., Sabatos, C.A., Lau, C. W. Z., Leech, M. D., Wraith, D. C. and Anderton, S. M., Cutting edge: Th1 cells facilitate the entry of Th17 cells to the central nervous system during experimental autoinmune encephalomyelitis. J. Immunol. 2008. 181: 37503754.
  • 27
    Lochner, M., Peduto, L., Cherrier, M., Sawa, S., Langa, F., Varona, R., Riethmacher, D. et al., In vivo equilibrium of proinflammatory IL-17+ and regulatory IL-10+ Foxp3+ RORgamma t+ T cells. J. Exp. Med. 2008. 205: 13811393.
  • 28
    Kroenke, M. A., Carlson, T. J., Andjelkovic, A. V. and Segal, B. M., IL12- and IL23-modulated T cells induce distinct types of EAE based on histology, CNS chemokine profile, and response to cytokine inhibition. J. Exp. Med. 2008. 205: 15351541.
  • 29
    Hirata, S., Matsuyoshi, H., Fukuma, D., Kurisaki, A., Uemura, Y., Nishimura, Y. and Senju, S., Involvement of regulatory T cells in the experimental autoimmune encephalomyelitis-preventive effect of dendritic cells expressing myelin oligodendrocyte glycoprotein plus TRAIL. J. Immunol. 2007. 178: 918925.
  • 30
    Varona, R., Villares, R., Carramolino, L., Goya, I., Zaballos, A., Gutierrez, J., Torres, M. et al., CCR6-deficient mice have impaired leucocyte homeostasis and altered contact hypersensitivity and delayed-type hypersensitivity responses. J. Clin. Invest. 2001. 107: 3745.
  • 31
    Huang, D., Tani, M., Wang, J., Han, Y., He, T. T., Weaver, J., Charo, I. F. et al., Pertussis toxin-induced reversible encephalopathy dependent on monocyte chemoattractant protein-1 overexpression in mice. J. Neurosci. 2002. 22: 1063310642.