Abstract
Conventional αβ T cells require sphingosine 1-phosphate (S1P) receptor 1 (S1P1) for circulation through the lymph nodes (LN); however, it is unclear whether γδ T cells use similar mechanisms. In this study, we found that treatment with fingolimod (FTY720, 1 mg/kg, orally) markedly reduced not only conventional CD4 T cells but also circulating γδ T cells (Vγ4+ and Vγ4− subsets) in the blood of mice. In contrast, IL-17+Vγ4+, IL-17+Vγ4−, and IL-17−Vγ4− subsets were significantly accumulated in the LN after 6 h of FTY720 treatment. By skin application of a synthetic TLR7/8 agonist, Vγ4+ γδ T cells (IL-17+ and IL-17− subsets) were accumulated and expanded in the draining LN (DLN), whereas the IL-17+ subset predominantly migrated to the inflamed skin. FTY720 induced a marked sequestration of IL-17–producing Vγ4+ γδ T cells in the DLN and inhibited their infiltration into the inflamed skin. Similarly, FTY720 inhibited infiltration of Vγ4+ γδ T cells into the CNS by their sequestration into the DLN in experimental autoimmune encephalomyelitis. Vγ4+ γδ T cells expressed a significant level of S1P1 and showed a migratory response toward S1P. FTY720 treatment induced almost complete downregulation of S1P1 expression and S1P responsiveness in Vγ4+ γδ T cells. Our findings strongly suggest that IL-17–producing Vγ4+ γδ T cells require S1P1 for their egress from the LN under homeostatic and inflammatory conditions. Consequently, inhibition of S1P1-dependent egress of pathogenic IL-17–producing Vγ4+ γδ T cells from the DLN may partly contribute the clinical therapeutic effects of FTY720 in relapsing multiple sclerosis.
Introduction
Fingolimod (FTY720), the sphingosine 1-phosphate (S1P) receptor modulator, is shown to be highly effective in relapsing multiple sclerosis (MS) and has been widely used for the oral therapy of this disease (1, 2). It is well documented that phosphorylated FTY720 acts as a functional antagonist at S1P receptor 1 (S1P1) by internalization and degradation of the receptor and inhibits S1P1-dependent lymphocyte egress from the lymph nodes (LN) (3–6). Because it has been suggested that IL-17 rather than IFN-γ plays a pathogenic role in MS and experimental autoimmune encephalomyelitis (EAE), a CD4 T cell–dependent animal model for MS, many studies have focused on understanding the differentiation and effector functions of IL-17–producing Th cells (Th17 cells) in autoimmunity (7–9). The production of IL-17 by peripheral blood T cells stimulated with anti-CD3 and anti-CD28 mAbs was markedly reduced (>95% inhibition) in FTY720-administered MS patients, suggesting that FTY720 inhibits egress of Th17 cells from the LN (10, 11). Furthermore, FTY720 showed a marked therapeutic effect on mouse EAE by reducing infiltration of Th17 cells into the spinal cord (12–15). Based on these data, it has been thought that the therapeutic effects of FTY720 on MS and EAE are predominantly due to inhibition of infiltration of pathogenic Th17 cells into the CNS.
More recently, however, it has been noted that conventional Th17 cells are not the sole producers of IL-17 because larger amounts of IL-17 are produced by several types of γδ T cells rather than Th17 cells without the clonal expansion or additional TCR stimulation required for the adaptive response (16–18). Unlike conventional αβ T cells, γδ T cells preferentially reside in epithelial tissues, including the skin, intestine, lung, and reproductive tract, and they provide a first line of host defense against various pathogens (19, 20). In MS patients, the frequency of γδ T cells was significantly increased in the cerebrospinal fluid and blood, and the accumulation of Vδ2 T cells was found in chronic active lesions, suggesting a pathogenic role of IL-17–producing γδ T cells in MS (21, 22). Similarly, a significant infiltration of IL-17–producing γδ T cells, particularly the Vγ4+ subset, was found in the CNS of EAE mice (17, 23, 24). Although IL-17–producing murine γδ T cells mainly consist of Vγ4+ and Vγ6+ subsets (25), it is strongly suggested that IL-17–producing Vγ4+ γδ T cells play a pathogenic role in the development of EAE because activation of this subset by anti-Vγ4 mAb treatment exacerbated EAE (17, 26–28).
It has been reported that thymic γδ T cells express S1P1 mRNA and are accumulated in the thymus of S1P1-deficient mice, indicating that γδ T cells, similar to αβ T cells, require S1P1 for thymic egress (29, 30). Peripheral IL-17–producing Vγ4+ γδ T cells are shown to be migrated to the draining LN (DLN), considerably expand there, and recirculate back to the inflamed skin by application with a synthetic TLR7/8 agonist to mouse ear skin (31). However, it remains unclear whether γδ T cells, including the Vγ4+ subset, require S1P1 for their circulation. In this study, we investigated the role of the S1P/S1P1 axis in the circulation of γδ T cells, particularly IL-17–producing Vγ4+ γδ T cells between the blood, LN, and inflamed tissues. We found that FTY720 markedly reduces the numbers of circulating γδ T cells and IL-17–producing Vγ4+ γδ T cells by their sequestration into the LN. Our data clearly demonstrated that Vγ4+ γδ T cells express significant levels of S1P1 and show a migratory response toward S1P. FTY720 induced almost complete downregulation of S1P1 expression in Vγ4+ γδ T cells and reduced infiltration of Vγ4+ γδ T cells into the CNS of EAE mice by sequestration of these cells into the DLN. Thus, our findings strongly suggest that IL-17–producing Vγ4+ γδ T cells require S1P1 for egress from LN under homeostatic and inflammatory conditions.
Materials and Methods
Animals
Male BALB/c and C57BL/6 mice were purchased from Charles River Japan and were used at 6–10 wk of age. All animal experiments were performed under experimental protocols approved by the Ethics Review Committee for Animal Experimentation of Research Division, Mitsubishi Tanabe Pharma.
Compound treatment
Cell preparation
BALB/c mice were treated with 5 mg 5% imiquimod (IMQ) cream (Beselna cream; Mochida Pharmaceuticals) on the right ear daily for 6 d. Inguinal and auricular DLN were digested with 1.25 mg/ml collagenase type IV (Roche Applied Science) and 200 μg/ml DNase I (Roche Applied Science) for 30 min at 37°C. Whole ear skins were minced and digested with collagenase, DNase I, and 500 μg/ml dispase (Life Technologies) for 1 h. Spleen cells were prepared by gently disrupting in RPMI 1640 medium (Sigma-Aldrich) using glass slides. Removal of RBC in spleen cells and peripheral blood was performed using Tris-NH4Cl buffer and an automated TQ-Prep system (Beckman Coulter), respectively. Spinal cords of four mice were minced and digested with collagenase and DNase I for 1 h. The resulting cell pellets were resuspended in 30% Percoll (GE Healthcare) and layered on 70% Percoll. After centrifugation at 500 × g for 25 min at 25°C, cells at the interface were collected. For migration assays, γδ T cells in the auricular LN of IMQ-treated mice were prepared using a TCRγ/δ+ T cell isolation kit (Miltenyi Biotec). Single-cell suspensions were disaggregated by passage through a 70-μm nylon cell strainer.
Flow cytometry
Single-cell suspensions were stained with the following mAbs (all from BD Biosciences or eBioscience); anti-CD3 (145-2C11), anti-CD4 (GK1.5), anti-CD8 (53-6.7), anti-TCRγδ (GL3), anti-Vγ4 (UC3-10A6), and anti–IL-17A (TC11-18H10). For S1P1 staining, LN cells were stained with rat anti-S1P1 mAb (713412, R&D Systems), biotinylated donkey anti-rat IgG (Jackson ImmunoResearch Laboratories), and PE-conjugated streptavidin (eBioscience) (34). For detection of intracellular IL-17, cells were cultured in the presence of 50 ng/ml PMA (Sigma-Aldrich), 1 μg/ml ionomycin (Sigma-Aldrich), and 2 μM monensin (eBioscience) for 4 h. After blocking by rat anti-mouse CD16/32 mAb (2.4G2; FcR Block, BD Biosciences), the cells were labeled with surface Ags, treated with Cytofix buffer and Perm/Wash buffer (both from BD Biosciences), and then stained with anti–IL-17 mAb. Samples were acquired on a flow cytometer (LSR II; BD Biosciences). For cell counting, the number of each cell subset was determined using comparison with a known number of beads as an internal standard of Flow-Count fluorosphere (Beckman Coulter) by flow cytometry with a Cytomics FC500 (Beckman Coulter). The numbers of IL-17+ and IL-17− γδ T cells were calculated by using the absolute number of γδ T cells and the frequencies (%) of IL-17+ and IL-17− γδ T cells in total γδ T cells, respectively.
Immunohistochemistry
Samples of auricular LN from mice were embedded in Tissue-Tek OCT compound (Sakura FineTek) and fixed with acetone. LN sections were stained with hamster anti-mouse TCRγδ polyclonal Ab (GL3; BD Biosciences), incubated with biotin-goat anti-hamster IgG (Vector Laboratories) and peroxidase-conjugated streptavidin (Dako), and colorized with 3,3′-diaminobenzidine in the presence of peroxidase. Images of the stained sections were inspected under an Olympus BX51 microscope equipped with cellSens software (Olympus).
In vivo BrdU labeling
Mice were given 0.8 mg/ml BrdU (Sigma-Aldrich) and 2% glucose in their drinking water during IMQ applications. Intracellular BrdU staining was performed using a BrdU flow kit (BD Biosciences).
Real time-PCR
Total RNA was isolated from the DLN of mice using TRIzol reagents (Invitrogen). cDNA was synthesized with a TaqMan reverse transcription kit (Life Technologies) using random hexamers and 0.5 μg total RNA. Real-time PCR for the detection of Il2 (Mm00434256_m1), Il4 (Mm00445259_m1), Il7 (Mm01295803_m1), Il9 (Mm00434305_m1), Il15 (Mm00434210_m1), and Il21 (Mm00517640_m1) mRNA was performed with predesigned TaqMan gene expression assays (Life Technologies). For normalization, internal control GAPDH mRNA was detected using TaqMan rodent GAPDH control reagents (Life Technologies). Samples were assayed on an ABI Prism 7900 sequence detector (Life Technologies) and the data were analyzed using sequence detector software (Life Technologies).
Migration assay
Migration assays were performed according to the method described previously (5). Lymphocytes (106 cells) or γδ T cells (105 cells) prepared from the DLN of mice applied with IMQ for 6 d were added to the upper wells of 5-μm pore, polycarbonate 24-well tissue culture inserts (Costar). S1P (10 nM; Sigma-Aldrich) was added to the lower chamber. The migration assays were conducted in RPMI 1640 medium with 0.5% fatty acid–free BSA (Sigma-Aldrich) for 4 h at 37°C in 5% CO2. The numbers of CD4 T cells and Vγ4+ γδ T cells in the starting and the migrated population were determined by flow cytometry described above, and the migration index was calculated from these values.
EAE induction
C57BL/6 mice were immunized s.c. with 200 μg myelin oligodendrocyte glycoprotein (MOG)35–55 peptide (Peptide Institute) emulsified in CFA (containing 4 mg/ml killed Mycobacterium tuberculosis, H37Ra; Chondrex) on day 0, followed by i.v. injection with 200 ng pertussis toxin (List Biological Laboratories) on day 0 and day 2 (12, 13). The severity of EAE was daily monitored and graded on a scale of 0–5 using the following criteria: 0, no paralysis; 0.5, stiff tail; 1, limp tail; 1.5, limp tail with inability to right; 2, paralysis of one limb; 2.5, paralysis of one limb and weakness of one other limb; 3, complete paralysis of both hindlimbs; 4, moribund state; 5, death (12, 13).
Statistical analysis
The results were expressed as the means ± SEM. Statistical differences were calculated by a Student t test and considered significant when p < 0.05.
Results
FTY720 sequesters circulating Vγ4+ γδ T cells into the LN under homeostatic conditions
We investigated the influence of FTY720 on the number of Vγ4+ γδ T cells in the blood of normal BALB/c mice because several γδ T cell subsets are known to be circulating in the blood under homeostatic conditions (35, 36). As shown in Fig. 1A, FTY720 (1 mg/kg, a single oral administration) induced a marked reduction of the numbers of CD4 T cells, γδ T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells with similar time course changes (at 3–24 h after the administration). In contrast, the numbers of these T cells (CD4 T cells, γδ T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells) were significantly increased in the peripheral LN at 6 h after FTY720 administration (Fig. 1B). The intracellular cytokine staining revealed that both IL-17+ and IL-17− γδ T cell subsets were accumulated in the LN of mice given FTY720 (Fig. 1C, 1D). Particularly, more than half (55.1%) of Vγ4+ γδ T cells expressed significant levels of intracellular IL-17 and FTY720 increased the frequency of IL-17–expressing Vγ4+ γδ T cells in the LN (Fig. 1C). FTY720 significantly showed an ∼1.5-fold increase in the number of γδ T cells (IL-17+Vγ4+, IL-17+Vγ4−, and IL-17−Vγ4− subsets) but not in the number of the IL-17−Vγ4+ γδ T cell subset. (Fig. 1D). These results suggest that FTY720 sequesters circulating γδ T cells, including IL-17–expressing Vγ4+ γδ T cells into the LN under homeostatic conditions.
FTY720 modulates circulation of Vγ4+ γδ T cells through the LN under homeostatic conditions. After a single oral administration of FTY720 (1 mg/kg) or vehicle, the numbers of total lymphocytes, CD4 T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells in the blood (A) and inguinal LN (B) were determined by flow cytometry. (C) Intracellular staining of IL-17 in the LN lymphocytes at 6 h after administration of FTY720. Representative dot plots of IL-17–expressing Vγ4+ γδ T cells are shown gated on γδ TCR+ cells at 4 h after stimulation with PMA plus ionomycin. (D) The numbers of IL-17+ and IL-17− γδ T cell subsets in the LN. All results are expressed as the means ± SEM (n = 3). Statistical differences were calculated by a Student t test. *p < 0.05, **p < 0.01.
FTY720 modulates circulation of Vγ4+ γδ T cells through the LN under homeostatic conditions. After a single oral administration of FTY720 (1 mg/kg) or vehicle, the numbers of total lymphocytes, CD4 T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells in the blood (A) and inguinal LN (B) were determined by flow cytometry. (C) Intracellular staining of IL-17 in the LN lymphocytes at 6 h after administration of FTY720. Representative dot plots of IL-17–expressing Vγ4+ γδ T cells are shown gated on γδ TCR+ cells at 4 h after stimulation with PMA plus ionomycin. (D) The numbers of IL-17+ and IL-17− γδ T cell subsets in the LN. All results are expressed as the means ± SEM (n = 3). Statistical differences were calculated by a Student t test. *p < 0.05, **p < 0.01.
FTY720 reduces infiltration of IL-17–producing Vγ4+ γδ T cells into the inflamed skin of mice treated with TLR7/8 agonist
It has been reported that IL-17–producing Vγ4+ γδ T cells are increased in the inflamed skin and DLN by multiple skin application of a synthetic TLR7/8 agonist, IMQ (31, 37, 38). Based on these findings, we examined the time course change of the numbers of total lymphocytes, CD4 T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells in the blood, inflamed skin, and auricular DLN by the consecutive application of IMQ for 6 d. In the blood, the numbers of total lymphocytes, CD4 T cells, and Vγ4+ γδ T cells were increased on day 6 during IMQ applications whereas those were significantly reduced by FTY720 treatment (Fig. 2A). Although the number of Vγ4− γδ T cells in the blood was not increased by IMQ treatment, these T cells were also reduced by FTY720 administration (Fig. 2A). A noticeable infiltration of CD4 T cells and Vγ4+ γδ T cells into the ear skin was induced by IMQ applications, and the infiltration of these T cells was significantly reduced by FTY720 (Fig. 2B). IMQ applications led to a significant and time-dependent increase in the numbers of total lymphocytes, CD4 T cells, and Vγ4+ γδ T cells in the DLN, and FTY720 significantly inhibited the increase of total lymphocytes and CD4 T cells (Fig. 2C). In contrast, interestingly, a dramatic increase in the number of Vγ4+ γδ T cells was observed in the DLN of IMQ-applied mice treated with FTY720 (Fig. 2C). Alternatively, there was no clear alteration of Vγ4− γδ T cell counts in the ear skin or DLN in mice treated with IMQ and/or FTY720 (Fig. 2B, 2C). Intracellular cytokine staining revealed that FTY720 treatment showed 80% reduction in the number of IL-17–expressing Vγ4+ γδ T cells in the inflamed skin whereas FTY720 induced a 20-fold increase in the number of the Vγ4+ subset in the DLN (Fig. 2D, 2E). Based on these results, FTY720 appeared to affect the trafficking of IL-17–expressing Vγ4+ γδ T cells rather than IL-17−Vγ4+ γδ T cells between the inflamed skin and DLN.
FTY720 reduces infiltration of IL-17–producing Vγ4+ γδ T cells into the inflamed ear skin of mice treated with IMQ. BALB/c mice were treated with 5 mg 5% IMQ cream on the right ear daily for 6 d. FTY720 (1 mg/kg, orally) or vehicle was administered daily from day 0 to day 6. The numbers of total lymphocytes, CD4 T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells in the blood (A), ear skin (B), and auricular DLN (C) were determined by flow cytometry on days 0, 3, and 6. (D) Intracellular staining of IL-17 in the DLN lymphocytes and infiltrated lymphocytes into the ear skin on day 6. Representative dot plots of IL-17–expressing Vγ4+ γδ T cells are shown gated on γδ TCR+ cells at 4 h after stimulation with PMA plus ionomycin. (E) Numbers of IL-17+ and IL-17− γδ T cell subsets in the ear skin (upper panels) and DLN (lower panels). All results are expressed as the means ± SEM (n = 4). Statistical differences were calculated by a Student t test. *p < 0.05, **p < 0.01.
FTY720 reduces infiltration of IL-17–producing Vγ4+ γδ T cells into the inflamed ear skin of mice treated with IMQ. BALB/c mice were treated with 5 mg 5% IMQ cream on the right ear daily for 6 d. FTY720 (1 mg/kg, orally) or vehicle was administered daily from day 0 to day 6. The numbers of total lymphocytes, CD4 T cells, Vγ4+ γδ T cells, and Vγ4− γδ T cells in the blood (A), ear skin (B), and auricular DLN (C) were determined by flow cytometry on days 0, 3, and 6. (D) Intracellular staining of IL-17 in the DLN lymphocytes and infiltrated lymphocytes into the ear skin on day 6. Representative dot plots of IL-17–expressing Vγ4+ γδ T cells are shown gated on γδ TCR+ cells at 4 h after stimulation with PMA plus ionomycin. (E) Numbers of IL-17+ and IL-17− γδ T cell subsets in the ear skin (upper panels) and DLN (lower panels). All results are expressed as the means ± SEM (n = 4). Statistical differences were calculated by a Student t test. *p < 0.05, **p < 0.01.
FTY720 sequesters Vγ4+ γδ T cells into the DLN under inflammatory conditions
Immunohistochemical analyses demonstrated that there were few γδ T cells in the auricular LN of normal mice, whereas a significant increase in the number of γδ T cells was found to be localized in the T cell area of the auricular DLN by IMQ application (Fig. 3). Consistent with the flow cytometry data shown in Fig. 2C, a marked accumulation of γδ T cells was seen in the T cell area surrounding the lymphatic sinus of the DLN by FTY720 treatment (Fig. 3). These results strongly suggest that FTY720 induces sequestration of IL-17–producing Vγ4+ γδ T cells into the DLN and thereby inhibiting infiltration of the γδ T cells into the inflamed tissues under inflammatory conditions.
FTY720 induces accumulation of γδ T cells around the lymphatic sinus of the auricular DLN in IMQ-treated mice. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 6 d. The auricular LN sections from normal mice (top) and mice treated with IMQ (middle) or mice treated with IMQ and FTY720 (bottom) were stained with anti-mouse γδ-TCR mAb. Images are representative of four mice from two independent experiments in each group. Scale bars, 400 μm (left) and 50 μm (right).
FTY720 induces accumulation of γδ T cells around the lymphatic sinus of the auricular DLN in IMQ-treated mice. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 6 d. The auricular LN sections from normal mice (top) and mice treated with IMQ (middle) or mice treated with IMQ and FTY720 (bottom) were stained with anti-mouse γδ-TCR mAb. Images are representative of four mice from two independent experiments in each group. Scale bars, 400 μm (left) and 50 μm (right).
Next, we investigated whether a marked increase of Vγ4+ γδ T cells by FTTY720 selectively occurred in the DLN after IMQ applications. The proportion and total cell number of Vγ4+ γδ T cells were markedly increased in the auricular DLN but not in the distal (inguinal) LN by FTY720 administration (Fig. 4A–D). In the spleen, although there was no change in the proportion of Vγ4+ γδ T cells by treatment with IMQ or FTY720, the number of Vγ4+ γδ T cells was significantly increased by IMQ application as compared with normal control and were reduced markedly by FTY720 treatment (Fig. 4E, 4F). From these results, it is highly probable that a striking increase of Vγ4+ γδ T cells by FTY720 is selectively induced in the DLN and is not due to alteration of systemic lymphocyte distribution.
FTY720 induces a striking increase of Vγ4+ γδ T cells in the DLN but not in the distal LN or spleen of IMQ-treated mice. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 6 d to normal mice or mice treated with IMQ. Frequencies and absolute numbers of the Vγ4+ γδ T cells in the auricular DLN (A and B), distal (inguinal) LN (C and D), and spleen (E and F) were determined by flow cytometry. The results are expressed as the mean ± SEM (n = 4). Statistical differences were calculated by a Student t test. **p < 0.01, normal versus IMQ-treated mice; ##p < 0.01, IMQ-treated versus IMQ/FTY720-treated mice.
FTY720 induces a striking increase of Vγ4+ γδ T cells in the DLN but not in the distal LN or spleen of IMQ-treated mice. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 6 d to normal mice or mice treated with IMQ. Frequencies and absolute numbers of the Vγ4+ γδ T cells in the auricular DLN (A and B), distal (inguinal) LN (C and D), and spleen (E and F) were determined by flow cytometry. The results are expressed as the mean ± SEM (n = 4). Statistical differences were calculated by a Student t test. **p < 0.01, normal versus IMQ-treated mice; ##p < 0.01, IMQ-treated versus IMQ/FTY720-treated mice.
As shown in Fig. 5A, FTY720 treatment induced a marked increase in the number of Vγ4+ γδ T cells but not Vγ4− γδ T cells in the DLN on day 6 after IMQ applications. The proportion of BrdU-incorporated Vγ4+ γδ T cells in total LN lymphocytes was not altered on days 5 and 6 after IMQ application (Fig. 5B). In contrast, FTY720 treatment induced a significant increase in the number of BrdU-incorporated Vγ4+ γδ T cells in the DLN on day 6 (Fig. 5B). As Vγ4+ γδ T cells were comparably increased in the DLN even by IMQ application for 3 d, we examined the mRNA expressions of γc cytokines on days 3 and 6 (Fig. 5C). The mRNA expressions of IL-4 and IL-21 were significantly elevated in the DLN on day 6 after IMQ application and there was no clear change by FTY720 treatment (Fig. 5D), implying that IL-4 and/or IL-21 may induce proliferation of γδ T cells. From these results, it is probable that pathogenic IL-17–producing Vγ4+ γδ T cells, after their proliferation, can emigrate from the DLN and subsequently migrate to the inflamed tissues. Because FTY720 markedly reduced the infiltration of Vγ4+ γδ T cells into the inflamed skin (Fig. 2B), it is highly probable that FTY720 sequesters Vγ4+ γδ T cells into the DLN without affecting their proliferation.
FTY720 induces a marked increase in the number of BrdU-incorporated Vγ4+ γδ T cells in the DLN of mice treated with IMQ. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 6 d to mice treated with IMQ. (A) The numbers of γδ T cells, Vγ4+ γδ T cells, and Vγ4- γδ T cells in the auricular DLN were determined by flow cytometry on days 5 and 6. (B) IMQ-treated mice were fed with BrdU-containing water (0.8 mg/ml) from day 3 and the frequencies of BrdU-incorporated Vγ4+ γδ T cells among total lymphocytes were determined on days 5 and 6. (C) Mice were applied with IMQ once (day 0), three times (days 0–2), or six times (days 0–5). FTY720 or vehicle was administered daily for 6 d to mice treated with IMQ. On day 6, the numbers of Vγ4+ γδ T cells and CD4 T cells in the DLN were determined by flow cytometry. (D) The mRNA expressions of IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21 in the DLN were evaluated by real-time PCR normalized to GAPDH. All results are expressed as the means ± SEM (n = 4). Statistical differences were calculated by a Student t test. **p < 0.01.
FTY720 induces a marked increase in the number of BrdU-incorporated Vγ4+ γδ T cells in the DLN of mice treated with IMQ. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 6 d to mice treated with IMQ. (A) The numbers of γδ T cells, Vγ4+ γδ T cells, and Vγ4- γδ T cells in the auricular DLN were determined by flow cytometry on days 5 and 6. (B) IMQ-treated mice were fed with BrdU-containing water (0.8 mg/ml) from day 3 and the frequencies of BrdU-incorporated Vγ4+ γδ T cells among total lymphocytes were determined on days 5 and 6. (C) Mice were applied with IMQ once (day 0), three times (days 0–2), or six times (days 0–5). FTY720 or vehicle was administered daily for 6 d to mice treated with IMQ. On day 6, the numbers of Vγ4+ γδ T cells and CD4 T cells in the DLN were determined by flow cytometry. (D) The mRNA expressions of IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21 in the DLN were evaluated by real-time PCR normalized to GAPDH. All results are expressed as the means ± SEM (n = 4). Statistical differences were calculated by a Student t test. **p < 0.01.
The S1P/S1P1 axis regulates emigration of Vγ4+ γδ T cell from the DLN
It has been widely accepted that T cells, including effector T cells, require S1P1 for efficient egress from the LN (3, 4, 39, 40). However, it remains unknown whether S1P1 regulates circulation and trafficking of γδ T cells through the DLN under inflammatory conditions. Significant levels of S1P1 were expressed in Vγ4+ γδ T cells as well as CD4 T cells in the DLN of mice with IMQ application, and the S1P1 expressions in both T cell subsets were almost completely downregulated by in vivo treatment with FTY720 (Fig. 6A, 6C). As expected, similar to conventional CD4 T cells, Vγ4+ γδ T cells prepared from mice treated with IMQ showed a significant migratory response toward S1P (10 nM), and their S1P responsiveness was almost completely inhibited by in vivo treatment with FTY720 (Fig. 6B, 6D). Similarly, by in vivo administration of an S1P lyase inhibitor, THI (100 mg/kg, orally), the number of Vγ4+ γδ T cells was significantly reduced in the blood, increased in the DLN, and decreased in the inflamed skin of mice treated with IMQ (Fig. 6E). As noted, the S1P1 expression in Vγ4+ γδ T cells was almost completely downregulated by THI treatment (Fig. 6F). Taken together, these results indicate that the S1P/S1P1 axis regulates trafficking of Vγ4+ γδ T cells from the DLN to inflamed tissues.
The S1P/S1P1 axis regulates emigration of Vγ4+ γδ T cells from the DLN. FTY720 [(A–D) 1 mg/kg, orally], THI [(E and F) 100 mg/kg, orally], or vehicle was administered daily for 6 d to mice treated with IMQ. (A, C, and F) The mean fluorescence intensity of S1P1 on CD4 T cells (A) and Vγ4+ γδ T cells (C and F) from the auricular DLN was determined by flow cytometry. Histograms are representative of four mice from two independent experiments in each group. Isotype control (shaded), mice treated with IMQ and FTY720 [(A and C) blue], IMQ and THI [(F) green], or IMQ and vehicle (red). (B and D) Sorted γδ T cells or total lymphocytes from the DLN of mice treated with IMQ (and FTY720 or vehicle) for 6 d were performed to migration assays toward 10 nM S1P for 4 h. The percentage of migration of CD4 T cells (B) and Vγ4+ γδ T cells (D) was determined by flow cytometry. (E) The numbers of Vγ4+ γδ T cells in the blood, auricular DLN, and ear skin of mice treated with IMQ (and THI or vehicle) were determined by flow cytometry on day 6. All results are expressed as the mean ± SEM. (A, C, E, and F) n = 4. (B and D) n = 3. Statistical differences were calculated by a Student t test. **p < 0.01.
The S1P/S1P1 axis regulates emigration of Vγ4+ γδ T cells from the DLN. FTY720 [(A–D) 1 mg/kg, orally], THI [(E and F) 100 mg/kg, orally], or vehicle was administered daily for 6 d to mice treated with IMQ. (A, C, and F) The mean fluorescence intensity of S1P1 on CD4 T cells (A) and Vγ4+ γδ T cells (C and F) from the auricular DLN was determined by flow cytometry. Histograms are representative of four mice from two independent experiments in each group. Isotype control (shaded), mice treated with IMQ and FTY720 [(A and C) blue], IMQ and THI [(F) green], or IMQ and vehicle (red). (B and D) Sorted γδ T cells or total lymphocytes from the DLN of mice treated with IMQ (and FTY720 or vehicle) for 6 d were performed to migration assays toward 10 nM S1P for 4 h. The percentage of migration of CD4 T cells (B) and Vγ4+ γδ T cells (D) was determined by flow cytometry. (E) The numbers of Vγ4+ γδ T cells in the blood, auricular DLN, and ear skin of mice treated with IMQ (and THI or vehicle) were determined by flow cytometry on day 6. All results are expressed as the mean ± SEM. (A, C, E, and F) n = 4. (B and D) n = 3. Statistical differences were calculated by a Student t test. **p < 0.01.
FTY720 reduces infiltration of Vγ4+ γδ T cells into the CNS in mouse EAE
Because Vγ4+ γδ T cells are known to produce larger amounts of IL-17 compared with the conventional Th17 cells and are found in the CNS of mouse EAE (17, 24), it is strongly suggested that IL-17–producing Vγ4+ γδ T cells play a pathogenic role in EAE. Although FTY720 showed marked therapeutic effects on EAE by inhibiting infiltration of Th17 cells into the CNS (14, 39), it remains unclear whether FTY720 affects trafficking of Vγ4+ γδ T cells to the CNS of EAE mice. As shown in Fig. 7A, prophylactic administration of FTY720 (1 mg/kg, orally) showed almost complete preventative effects on EAE induced by immunization with MOG35–55 peptide to C57BL/6 mice. Along with the development of EAE, significant numbers of infiltrated Vγ4+ γδ T cells as well as CD4 T cells were found in the spinal cord of EAE mice, and the infiltration of these T cells into the CNS was of an extremely low level in EAE mice treated with FTY720 (Fig. 7B, 7C, 7E). In the inguinal DLN, significant numbers of Vγ4+ γδ T cells were found in EAE mice on days 10–21, although Vγ4+ γδ T cells were very rare in mice before immunization (Fig. 7D, 7E). Interestingly, a more marked increase in the number of Vγ4+ γδ T cells was seen in the DLN on day 21 in EAE mice treated with FTY720 (Fig. 7D, 7E), suggesting sequestration of these γδ T cells into the DLN. Consequently, our findings imply that the preventative effects of FTY720 on EAE development may be partly due to inhibiting egress of pathogenic IL-17–producing Vγ4+ γδ T cells from the DLN via downregulation of S1P1.
FTY720 reduces infiltration of Vγ4+ γδ T cells into the spinal cord of EAE mice. C57BL/6 mice were immunized with MOG35–55 peptide and CFA. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 21 d. (A) Clinical scores of EAE symptoms are expressed as the means ± SEM (n = 3–4). Statistical differences were calculated by a Mann–Whitney U test. *p < 0.05, **p < 0.01. (B–E) The numbers of CD4 T cells (B) and Vγ4+ γδ T cells in the spinal cord and inguinal DLN (C and D) were determined by flow cytometry on days 10, 14, and 21. The results are expressed as the means ± SEM (n = 3–4). Statistical differences were calculated by a Student t test. *p < 0.05, **p < 0.01. (E) Representative dot plots of Vγ4+ γδ T cells and CD4 T cells in the spinal cord and DLN on day 21 are shown.
FTY720 reduces infiltration of Vγ4+ γδ T cells into the spinal cord of EAE mice. C57BL/6 mice were immunized with MOG35–55 peptide and CFA. FTY720 (1 mg/kg, orally) or vehicle was administered daily for 21 d. (A) Clinical scores of EAE symptoms are expressed as the means ± SEM (n = 3–4). Statistical differences were calculated by a Mann–Whitney U test. *p < 0.05, **p < 0.01. (B–E) The numbers of CD4 T cells (B) and Vγ4+ γδ T cells in the spinal cord and inguinal DLN (C and D) were determined by flow cytometry on days 10, 14, and 21. The results are expressed as the means ± SEM (n = 3–4). Statistical differences were calculated by a Student t test. *p < 0.05, **p < 0.01. (E) Representative dot plots of Vγ4+ γδ T cells and CD4 T cells in the spinal cord and DLN on day 21 are shown.
Discussion
It has been well known that FTY720 markedly decreases the number of circulating lymphocytes including conventional αβ T cells in the blood by inhibiting S1P1-dependent lymphocyte egress from the LN (3, 41). Several years ago, it was initially demonstrated that FTY720 is highly effective in the cerebral infarction in mice, and one possible mechanism was proposed that this therapeutic effect is caused by inhibiting infiltration of γδ T cells into the mouse brain (42). However, there was no description or discussion of where the γδ T cells come from or what molecular mechanisms are underlying in the infiltration of γδ T cells into the inflamed lesions. Although it has been reported that thymic γδ T cells express S1P1 mRNA and S1P1-deficient γδ T cells accumulate in the thymus, suggesting S1P1-dependent thymic egress of γδ T cells (29, 30), there were few reports on the effect of FTY720 or other S1P receptor modulators on homeostatic circulation or trafficking of γδ T cells in the periphery.
In this study, we demonstrate that similar to conventional CD4+ αβ T cells, circulating Vγ4+ γδ T cells are markedly reduced in the blood and sequestrated into the LN by administration of FTY720 under homeostatic conditions. Furthermore, FTY720 strikingly reduced infiltration of IL-17–producing Vγ4+ γδ T cells into the inflamed tissues by their sequestration into the DLN in mice treated with TLR7/8 agonist. Vγ4+ γδ T cells express a significant level of S1P1 expression on the cell surface and show a migratory response toward a physiological concentration (10 nM) of S1P. Of note, the expression of S1P1 and the S1P responsiveness in Vγ4+ γδ T cells were almost completely lost in mice treated with FTY720 or THI, an inhibitor of S1P lyase. These results strongly suggest that the S1P/S1P1 axis plays an important role in the egress of IL-17–producing Vγ4+ γδ T cells from the LN.
Recent analyses using two-photon intravital microscopy revealed that conventional αβ T cells exit LN through cortical sinuses in an S1P1-dependent manner (43, 44). After S1P1-dependent emigration into cortical sinuses, T cells become rounded, flow unidirectionally into medullary sinuses, and are subsequently flushed into the subcapsular space and efferent lymph (44). However, it is still unclear whether Vγ4+ γδ T cells can exit from the LN with a similar mechanism to conventional αβ T cells. Our immunohistochemical analyses clearly showed that significant numbers of γδ T cells were found to be located in the T cell area of the DLN by IMQ applications. More marked accumulation of γδ T cells were seen in the T cell area surrounding the lymphatic sinus of the DLN by FTY720 treatment. These results strongly support that FTY720 induces sequestration of IL-17–producing Vγ4+ γδ T cells into the DLN and thereby inhibiting infiltration of these γδ T cells into the sites of inflammation. Consequently, it is highly probable that Vγ4+ γδ T cells, similar to conventional αβ T cells, exit from LN by an S1P1-dependent mechanism.
BrdU-incorporated Vγ4+ γδ T cells were increased in the DLN on days 5 and 6 in IMQ-treated mice and significant infiltration of Vγ4+ γδ T cells was found in inflamed skin on day 6. It has been reported that IL-2, IL-4, and IL-21 can induce proliferation of human γδ T cells (45). Although the mRNA expressions of IL-4 and IL-21 were elevated in the DLN on day 6 and these γc cytokines may induce proliferation of Vγ4+ γδ T cells in mice, it would be necessary to perform in vitro or in vivo studies using these cytokines or their neutralizing mAbs. Our results imply that after TLR7/8 stimulation, activated Vγ4+ γδ T cells are migrated to the DLN, are expanded there, and then recirculate back to the inflamed skin. Interestingly, we stably observed that FTY720 treatment induced a marked increase of BrdU-incorporated Vγ4+ γδ T cells in the DLN and strikingly reduced infiltration of Vγ4+ γδ T cells into the inflamed skin on day 6. When S1P1 is downregulated by treatment with FTY720, it is highly likely that Vγ4+ γδ T cells are unable to exit from the DLN and therefore they are expanded and accumulated there. Consequently, it is presumed that the dramatic increase of Vγ4+ γδ T cells by FTY720 is due to sequestration of these cells in the DLN.
It has been reported that IL-17–producing Vγ4+ γδ T cells are detectable in the CNS prior to the onset of EAE and thus contribute to the pathogenesis of EAE (17). Our results clearly demonstrate that FTY720 inhibits infiltration of Vγ4+ γδ T cells into the spinal cord and shows a marked preventing effect on MOG35–55-induced EAE. We and others have reported previously that FTY720 shows marked prophylactic and therapeutic effects and reduces infiltration of Th17 and Th1 cells into the CNS by inhibiting S1P1-dependent egress of these Th cells in mouse EAE (12–15). Alternatively, recent studies demonstrated that Th17 cells may not be a sole producer of IL-17 in mouse EAE, because Vγ4+ γδ T cells can produce larger amounts of IL-17 compared with Th17 cells (17). It has been impressive that the infiltration of Vγ4+ γδ T cells in the CNS correlated with disease severity in EAE in SJL/J mice; a significant number of Vγ4+ γδ T cells were found in the spinal cord during onset, peak, and relapse of EAE and these γδ T cells were temporarily lost during remission (24). Studies in MS patients have described that accumulation of Vδ2 T cells is found in the chronic active lesions (21). Vγ9Vδ2 T cells produced large amounts of IL-17 in the presence of IL-1β, IL-23, IL-6, and TGF-β (46). Moreover, it has been reported that the frequency of γδ T cells is significantly increased in the PBL from MS patients, and FTY720 administration resulted in a marked reduction of IL-17 production by PBL stimulated with anti-CD3 and anti-CD28 mAbs in vitro (11, 22, 47). Based on these findings, it is highly probable that inhibition of S1P1-dependent egress of pathogenic IL-17–producing γδ T cells partly contributes to the therapeutic effects of FTY720 on mouse EAE and relapsing MS.
Acknowledgements
We thank Dr. Kunitomo Adachi for synthesizing THI, and Shoji Wada and Yasuko Ogawa for excellent technical assistance.
Footnotes
References
Disclosures
The authors have no financial conflicts of interest.