CD4+ Th1 cells produce IFN-γ, TNF-α, and IL-2. These Th1 cytokines play critical roles in both protective immunity and inflammatory responses. In this study we report that sphingosine kinase 1 (SPHK1), but not SPHK2, is highly expressed in DO11.10 Th1 cells. The expression of SPHK1 in Th1 cells requires TCR signaling and new protein synthesis. SPHK1 phosphorylates sphingosine to form sphingosine-1-phosphate. Sphingosine-1-phosphate plays important roles in inhibition of apoptosis, promotion of cell proliferation, cell migration, calcium mobilization, and activation of ERK1/2. When SPHK1 expression was knocked down by SPHK1 short interfering RNA, the production of IL-2, TNF-α, and IFN-γ by Th1 cells in response to TCR stimulation was enhanced. Consistently, overexpression of dominant-negative SPHK1 increased the production of IL-2, TNF-α, and IFN-γ in Th1 cells. Furthermore, overexpression of SPHK1 in Th1 and Th0 cells decreased the expression of IL-2, TNF-α, and IFN-γ. Several chemokines, including Th2 chemokines CCL17 and CCL22, were up-regulated by SPHK1 short interfering RNA and down-regulated by overexpression of SPHK1. We also showed that Th2 cells themselves express CCL17 and CCL22. Finally, we conclude that SPHK1 negatively regulates the inflammatory responses of Th1 cells by inhibiting the production of proinflammatory cytokines and chemokines.
Naive CD4+ T cells differentiate into two distinct subsets, Th1 and Th2, depending on their environment and the associated differentiation signals. Th1 cells are characterized by their ability to produce IFN-γ, but not IL-4, whereas Th2 cells produce IL-4, but not IFN-γ (1, 2). Th1 cells also produce more TNF-α and IL-2 than Th2 cells. Th1 cells play critical roles in cell-mediated immune response, whereas Th2 cells are involved in humoral immunity. Overactivation of CD4+ Th1 and Th2 cells may induce autoimmune and inflammatory diseases. For example, Th1 cells are involved in inflammatory bowel disease, arthritis, type I diabetes, and myocarditis. Crohn’s disease, one of the inflammatory bowel diseases, is driven by the production of IL-12 and IFN-γ (3, 4). Mice lacking IFN-γRs were highly susceptible to collagen-induced arthritis (5). TNF-α is the major cytokine in chronic destructive arthritis. Anti-TNF-α treatment shows great efficacy in rheumatoid arthritis patients (6). TNF-α is up-regulated in the mucosa of patients with inflammatory disease, and anti-TNF-α Abs have shown marked clinical benefit in the majority of patients with Crohn’s disease (7). Thus, inhibition of the above cytokines is critical for the treatment of autoimmune and inflammatory diseases.
There are two known sphingosine kinases (SPHK)2: SPHK1 and SPHK2. SPHK phosphorylates sphingosine to form sphingosine-1-phosphate (S1P). S1P is a bioactive lipid mediator that can function as an intracellular second messenger as well as a ligand of cell surface receptors (8). Intracellular S1P regulates several important pathways, such as inhibition of apoptosis, promotion of cell proliferation, calcium mobilization, and activation of ERK1/2 (8, 9). IgE Ag stimulation of human bone marrow-derived mast cells triggers SPHK1-mediated fast and transient Ca2+ release from intracellular stores, which depends primarily on the activation of both phospholipase D1 and SPHK1 (10). In L929 cells, TNF-α-induced cyclooxygenase-2 and PGE2 expression is significantly reduced by SPHK1 short-interfering RNA (siRNA) (11). Although SPHK2 is known to associate with the mouse IL-12Rβ1 cytoplasmic region and augment IL-12-induced STAT4-mediated transcriptional activation (12), the function of SPHK1 in CD4+ T cells is still poorly understood.
The immunomodulatory drug FTY720 is phosphorylated by SPHKs (13). The phosphorylated FTY720 is a partially selective S1P analog (14). It binds and activates four of five S1P receptors (15, 16). FTY720 inhibits lymphocyte emigration from lymphoid organs (16). Graeler et al. (17) showed that S1P enhanced chemotaxis of CD4+ T cells to CCL-21 and CCL-5 by up to 2.5-fold at 10 nM to 0.1 μM, whereas 0.3–3 μM S1P inhibits this chemotaxis by up to 70%. They also showed that S1P decreased CD4+ T cell generation of IFN-γ and IL-4, without affecting IL-2. Meanwhile, S1P inhibited chemotaxis to CCL-5 and CCL-21 (18). Thus, understanding the regulation of S1P production and downstream activity in CD4+ T cells is crucial.
In this study we show that SPHK1 is highly expressed in Th1 cells. Overexpression of SPHK1 in these cells leads to decreased expression of IL-2, TNF-α, IFN-γ, CCL17, and CCL22, whereas knocking down of SPHK1 by SPHK1 siRNA shows increased induction of IL-2, TNF-α, IFN-γ, CCL17, and CCL22 by TCR stimulation. Thus, SPHK1 plays an inhibitory function in the proinflammatory responses of Th1 cells.
Materials and Methods
Mice, reagents, cytokines, and Abs
DO11.10 αβ-TCR transgenic mice have been described previously (19); they were maintained in our pathogen-free facility. Female BALB/c mice were bought from The Jackson Laboratory. Murine IL-4 and IL-12 were purchased from Calbiochem. Murine IL-2 and monoclonal anti-IL-4 (11B11) were obtained from eBioscience. Polyclonal anti-mouse IL-12 was purchased from Cell Sciences. Monoclonal anti-mouse CD3ε (500A2) and anti-mouse CD28 (37.51) were purchased from BD Biosciences. OVA, cyclosporin A, PMA, ionomycin, and Histopaque-1119 were all obtained from Sigma-Aldrich. Cycloheximide, genistein, AG9 (tyrphostin A1), and AG126 (tyrphostin AG 126) were purchased from Calbiochem.
Differentiation of DO11.10 T cells to Th1 and Th2 phenotypes and infection by retrovirus were conducted as previously described (20, 21). Briefly, splenic T cells were activated by OVA (500 μg/ml) in 10% FBS/IMDM in the presence of IL-2 (2 ng/ml). IL-12 (5 ng/ml) and anti-IL-4 (11B11; 2 μg/ml) were added to the culture for Th1 development, whereas IL-4 (20 ng/ml) and anti-IL12 (1 μg/ml) were used for Th2 development. For the generation of Th0 cells, no additional cytokines or Abs were added to the culture. The T cells were passaged on a weekly basis by activation with OVA and irradiated BALB/c splenocytes (2000 rad). For anti-CD3 stimulation, plates were coated overnight at 4°C with 0.63–10 μg/ml anti-CD3 Ab (500A2) diluted in PBS. T cells were applied at 1 × 106 cells/ml as indicated in the figure legends. For GeneChip experiments, T cells were briefly spun down in anti-CD3-coated plates and stimulated with anti-CD3 for 2 h.
Human T cells were purified from peripheral blood donated by healthy donors. RBC were separated by Ficoll-Paque centrifugation. The remaining cells were incubated according to label directions with StemSep Human Naive CD4+ T Cell Enrichment Cocktail (Stem Cell Technologies), which purifies T cells that are CD4+, CD45RA+, CD45RO−, and CD8− by negative selection. The T cells were cultured in 10% FBS/IMDM. These naive T cells were then stimulated with plate-bound anti-CD3 and anti-CD28 in the presence of 2.5 ng/ml human (h) IL-2 (BD Biosciences), 4 ng/ml hIL-12 (Calbiochem), and 200 ng/ml anti-hIL-4 (eBioscience) at a concentration of 2 × 106 cells/ml for 7 days. The cells were restimulated for another week before being transfected with siRNA.
Th1 and Th2 cells were stimulated with anti-CD3 for 18 h before being harvested. Cells then were lysed by SDS loading buffer (187.5 mM Tris-HCl (pH 6.8), 6% (w/v) SDS, 30% glycerol, 0.03% (w/v) bromphenol blue, and 42 mM DTT; New England Biolabs). One hundred microliters of the SDS loading buffer was used to lyse 1 × 107 cells. The lysates were then boiled at 100°C for 5 min, centrifuged at 13,000 rpm, resolved by SDS-PAGE, and transferred to nitrocellulose membrane. The membrane was probed with rabbit polyclonal anti-SPHK1 (Exalpha Biologicals) and developed using an anti-rabbit HRP-conjugated Ab and SuperSignal West Pico Substrate (Pierce). The membrane was then striped and reprobed with mouse anti-p38 MAPK (5F11) mAb (Cell Signaling) and developed using an anti-mouse-HRP-conjugated Ab.
Retroviral constructs, transfection, and infection
We obtained a mouse full-length SPHK1 cDNA clone from an established sequence tag IMAGE clone from OpenBiosystems. The SPHK1 cDNA insert was generated by PCR using the IMAGE clone as a template. The sequences of the primers for PCR are as follows: murine (m) SPHK1-1, 5′-TAAGAAGCTGAACGCAGGAG; and mSPHK1-3, 5′-AAAGGCACAGAGTTATGGTTC. The PCR products were then ligated into the blunted BglII/XhoI site of internal ribosome entry site GFP retrovirus (GFP-RV) as described previously (20, 21). The final SPHK1-RV was confirmed by sequencing. Changing hSPHK1 glycine 26 or glycine 80 to aspartic acid 26 or aspartic acid 80 by point mutation results in an catalytically inactive kinase (22). We designed the following oligonucleotides to generate SPHK1 G26D- and G80D-RV: mSPHK1 G26D-F, 5′-GCTGCTGAACCCCCAGGGTGACAAGGGCAAGGCTCTGCAGC; mSPHK1G26D-R, 5′-GCTGCAGAGCCTTGCCCTTGTCACCCTGGGGGTTCAGCAG; mSPHK1G80D-F, 5′-CGCCCTGGCAGTCATGTCCGATGATGGTCTGATGCATGAGG; and mSPHK1G80D-R, 5′-CCTCATGCATCAGACCATCATCGGACATGACTGCCAGGGCG. The SPHK1-RV was used as a template to generate SPHK1 G26D-RV and SPHK1 G80D-RV using the QuikChange Site-Directed Mutagenesis kit (Stratagene). Phoenix-Eco packaging cell line (licensed from Dr. G. Nolan, Stanford University, CA) was transfected to generate the retroviral supernatants according to Dr. Nolan’s protocol, as described previously (21). T cells were activated as described above and infected 48 h later. The viral supernatant and 6 μg/ml polybrene (Sigma-Aldrich) were added to the T cells in six-well plates and centrifuged at 1800 rpm for 30 min at room temperature. Cells were then cultured for 24 h before a 1:3 split. Infected T cells were purified by cell sorting on day 7 after primary activation by the expression of murine CD4 and GFP. Cells were then restimulated with OVA and irradiated BALB/c splenocytes (2000 rad; 5 × 106 cells/ml) for another week. Finally, cells were counted, washed, suspended with 1 × 106 cells/ml, and then restimulated with plate-bound anti-CD3 for 2–18 h.
Transfection of siRNA
Human SPHK1-specific 21-nucleotide siRNA oligo (Dharmacon Research; GAGCUGCAAGGCCUUGCCCdTdT and GGGCAAGGCCUUGCAGCUCdTdT) targets 70 nt downstream of the start codon as described previously (23, 24, 25). These previous studies also showed that the above SPHK1 siRNA specifically inhibits SPHK1 mRNA expression. Bioinformatics study shows that the siRNA sequence is unique to SPHK1 mRNA. The nonspecific control GL2 siRNA (5′-CGUACGCGGAAUACUUCGATT and TTGCAUGCGCCUUAUGAAGCU-5′) were described previously (26). T cells were harvested by centrifugation and resuspended at 3 × 107 cells/ml in Human T Cell kit buffer (VPA-1002; Amaxa), and 100 μl was dispensed to electroporation cuvettes containing 6 μl of siRNA and 1 μl of FITC-control siRNA (Invitrogen Life Technologies). The final concentration of siRNA was 100 nM. Cells were electroporated using the specific program T23 by the Nucleofector device (Amaxa) and were incubated for 72 h in IMDM with hIL-2. Cells were then counted and placed into wells at a concentration of 1 × 106 cells/ml. Finally, the T cells were stimulated with anti-CD3 for 18 h before supernatants and cell pellets were harvested for additional analysis.
Cytometric bead array (CBA)
The cytokine titers in the supernatants were analyzed by CBA (BD Biosciences). Mouse or hTh1/Th2 cytokine CBA kits were used (BD Biosciences). Briefly, 25 μl of each sample or standard dilution was mixed with 25 μl of mixed capture beads and 25 μl of the mouse Th1/Th2 PE detection reagent. After the samples were incubated at room temperature for 2 h in the dark, they were washed once and resuspended in 200 μl of wash buffer before acquisition on a FACScan. Data were analyzed using the CBA software. The concentration of each cytokine in the supernatants was calculated with the corresponding standard curve.
RNA extraction, TaqMan quantitative RT-PCR, and GeneChip analysis
Total RNA was extracted using RNeasy Mini kit (Qiagen). The possible remaining DNA was digested with the RNase-free DNase I (Qiagen) during the extraction of total RNA. cDNA synthesis, TaqMan quantitative RT-PCR, and GeneChip analysis were described previously (27, 28). TaqMan real-time RT-PCRs were performed using ABI 7700 to measure the mRNA expression level. TaqMan signals were calculated as mRNA copy numbers per sample input generated using a standard curve and normalized to endogenous GAPDH (28). The TaqMan signals provide relative quantitation of gene expression within each experiment. TaqMan primers and probes are as follows: mSPHK1: mSPHK1-539 forward, CCATCCAGAAACCCCTGTGTA; mSPHK1-615 reverse, CCCAGCATAGTGGTTCACAGAA; and mSPHK1-563T-FAM minor groove binder (MGB) probe, TCCCTGGAGGCTCCGGCAATG; hSPHK: hSPHK1 forward, CTTGCAGCTCTTCCGGAGTC; SPHK1 reverse, GCTCAGTGAGCATCAGCGTG; and SPHK1-FAM MGB probe, CCCTTTTGGCTGAGGCTGAAATCTCC; mGAPDH: mGAPDH-864 forward, GCTACACTGAGGACCAGGTTGTCT; mGAPDH-980 reverse, ACCAGGAAATGAGCTTGACAAAGT; and mGAPDH-898T-FAM MGB probe, CAACAGCAACTCCCACTCTTCCACCTTC; hGAPDH: hGAPDH forward, CCAGGTGGTCTCCTCTGACTTC; hGAPDH reverse, GTGGTCGTTGAGGGCAATG; and hGAPDH-FAM MGB probe, AGCGACACCCACTCCTCCACCTTTG. Human and mouse CCL17 and CCL22 and hSPHK1 TaqMan primers and probes were purchased from Applied Biosystems.
For TaqMan and cytokine assays, statistical analysis was performed using Student’s t test by Microsoft Excel. A value of p < 0.05 was considered significant.
For Affymetrix GeneChip studies, levels of gene expression were quantitated from the hybridization intensities of 11 pairs of perfectly matched and mismatched control probes using the Affymetrix GeneChip Operating Software (GCOS) software. The software computes how the expression level of each transcript has changed between the baseline and experimental samples (change call). A change call describes whether a transcript in an experimental array has changed compared with a baseline array. Wilcoxon’s signed rank is used to generate a change p value using hybridization signals from all 22 probes for two comparison experiments. A change call is assigned based on analysis parameters. Change p values between 0.00 and 0.0025 are given an increase call. Change p values between 0.0025 and 0.003 are given a marginal increase call. Change p values between 0.997 and 0.998 are given a marginal decrease call. Change p values between 0.998 and 1.00 are given a decrease call (Affymetrix User Manual).
SPHK1 is highly induced in Th1 cells by TCR stimulation
To identify genes that are specifically induced in Th1 or Th2 cells, we performed DNA microarray studies using Affymetrix U74Av2 chips. DO11.10 splenocytes were differentiated under Th1 or Th2 conditions for 7–14 days, and the cells were restimulated with anti-CD3, IL-12, and/or IL-18 for 2 h. RNA was extracted from these cells and subjected to GeneChip analysis. We first analyzed Th1/Th2 cytokine mRNA levels from the GeneChip data. As expected, the Th1 cells expressed high levels of IFN-γ mRNA and very low levels of IL-4 mRNA, whereas Th2 cells expressed high levels of IL-4 and very low levels of IFN-γ mRNA (data not shown). Many other genes were also differentially induced in Th1 and Th2 cells. Among them, we identified that SPHK1 was specifically induced in Th1 cells by anti-CD3, but not IL-12/IL-18, stimulation (Fig. 1,A). Anti-CD3 stimulation had much less effect on SPHK1 mRNA expression in Th2 cells (Fig. 1,A). In contrast, SPHK2 showed similar expression levels between Th1 and Th2 cells (Fig. 1,B). To confirm the GeneChip results, we designed TaqMan probes and primers which have no overlap with the SPHK1 oligos printed on the microarray chip. Real-time TaqMan PCR analysis showed that the SPHK1 was highly induced in Th1 cells by anti-CD3 stimulation (Fig. 1,C). There was an ∼10-fold induction after Th1 cells were stimulated with anti-CD3. This finding is consistent with the GeneChip result. Meanwhile, day 2 Th1 and Th2 cells did not express high levels of SPHK1 (Fig. 1,C), suggesting that SPHK1 is highly expressed in differentiated Th1 cells. We also detected the protein level of SPHK1 in Th1 and Th2 cells. Fig. 1 D shows that Th1 cells expressed higher levels of SPHK1 protein than Th2 cells after cells were stimulated with anti-CD3 for 18 h. We used p38 MAPK levels as a loading control. Because SPHK1 phosphorylates sphingosine to form S1P, we determined the level of S1P in Th1 and Th2 cells by mass spectrometric analysis. We found that Th1 cells produce ∼7-fold more intracellular S1P than Th2 cells (data not shown). Finally, Th0 cells expressed moderate levels of SPHK1 mRNA (data not shown).
Because SPHK1 is highly induced in Th1 cells by anti-CD3 stimulation, we sought to confirm the requirement for a TCR-mediated signal for the expression of SPHK1. Different inhibitors of TCR signaling were first cultured with T cells for 30 min before cells were added to an anti-CD3-coated plate. Cells were then briefly spun down and cultured for 2 h. As shown in Fig. 2, genistein and cyclosporin A almost completely inhibited SPHK1 expression, whereas the inactive inhibitor tyrphostin A1 and the p42MAPK inhibitor tyrphostin AG 126 had no effect. Genistein is a potent inhibitor of tyrosine phosphorylation, whereas cyclosporin A inhibits calcineurin and the downstream NF-AT pathway (29). It is well known that TCR signaling induces both tyrosine phosphorylation and the activation of calcineurin and NF-AT. Taken together, these data strongly suggest that SPHK1 mRNA expression is induced by TCR signaling. We also asked whether the expression of SPHK1 is the result of direct TCR signaling or requires de novo protein synthesis. When cells were treated with cycloheximide to block new protein synthesis after TCR stimulation, the induction of SPHK1 mRNA expression was suppressed (Fig. 2). This finding suggests that the SPHK1 promoter is not activated by direct signaling of the TCR, but requires de novo protein synthesis.
SPHK1 inhibits the expression of IL-2, TNF-α, and IFN-γ
Because SPHK1 is highly expressed in Th1 cells, we asked whether knocking down or overexpression of SPHK1 in Th1 cells would affect Th1 cytokine expression. RNA interference by specific siRNA oligos is now a common tool used to specifically knock down a gene of interest in mammalian cells (26, 30). However, there is no effective way to efficiently transfect siRNA oligos into mouse primary T cells. Thus, we used human primary CD4+ T cells, because siRNA oligos can be efficiently transfected into these cells using the electroporation technology developed by Amaxa (31). We purified human naive CD4+CD45RA+CD45RO− T cells and then stimulated the cells under Th1 conditions. After 2 wk of differentiation, Th1 cells were transiently transfected with SPHK1 siRNA or control siRNA (GL2). The transfection efficiency was ∼95% as detected by FITC-siRNA (data not shown). Cells were cultured for an additional 3 days after siRNA transfection and then restimulated with anti-CD3 for 18 h. The cells were harvested for RT-PCR analysis, and the supernatants were used to determine the cytokine titers by BD CBA. SPHK1 siRNA knocked down ∼70% of the SPHK1 mRNA level as determined by TaqMan real-time RT-PCR (Fig. 3,A). We then measured cytokine titers of Th1 cytokines in the supernatants treated with SPHK1 siRNA or GL2 siRNA. Surprisingly, SPHK1 siRNA-treated cells produced higher titers of IL-2, TNF-α, and IFN-γ than the control cells (Fig. 3,B). SPHK1 siRNA did not inhibit SPHK2 mRNA expression (Fig. 3,A). A subsequent SPHK2 siRNA study showed that this siRNA specifically inhibited SPHK2 mRNA expression, but not cytokine expression (Fig. 3, A and B). This result indicates that SPHK1 is a negative regulator of cytokine production in these cells.
To study this role, we asked whether overexpression of SPHK1 could inhibit Th1 cytokine expression. To do this, we generated a mouse SPHK1 retroviral vector (SPHK1-RV) that also expressed GFP. The vectors were transiently transfected into a Phoenix E packaging cell line to produce retroviral supernatants, which were then used to infect mouse DO11.10 Th0 or Th1 cells. On day 7, sorting of GFP- and CD4-positive cells resulted in >95% GFP-positive cells (data not shown). The sorted cells were also stimulated with OVA and irradiated BALB/C spleen cells for another week before restimulation with different concentrations of anti-CD3 for 18 h. SPHK1 mRNA increased ∼10-fold in SPHK1-RV-infected T cells (data not shown). Western blot analysis confirmed that SPHK1 protein expression was enhanced in SPHK1-RV-infected Th0 cells (Fig. 4,A). Cytokine titers of the resulting supernatants were analyzed using the CBA Th1/Th2 cytokine kit. Fig. 4 shows that SPHK1-RV-infected Th0 cells produced less IL-2, TNF-α, IFN-γ, and IL-5 than the GFP control. However, no significant reduction of IL-4 was observed (Fig. 4,B). Table I shows a similar pattern of expression when TaqMan analysis was used to examine changes in the mRNA levels of these four cytokines. Additionally, an increase in IL-10 was observed in Th0 cells infected with SPHK1-RV compared with the control (Table I). As expected, a decrease in cytokine expression of IL-2, TNF-α, and IFN-γ was also seen by CBA analysis of Th1 cells infected with SPHK1-RV (Fig. 5 A). Although a slight inhibition was observed (∼20–30% inhibition), it should be noted that the result was reproducible in four separate experiments. Because endogenous SPHK1 is already highly expressed in activated T cells, additional overexpression may not show a dramatic effect on the cytokine production.
|.||IL-2 .||TNF-α .||IFN-γ .||IL-4 .||IL-5 .||IL-10 .|
|.||IL-2 .||TNF-α .||IFN-γ .||IL-4 .||IL-5 .||IL-10 .|
The percentage of mRNA level compared with cells infected with SPHK1-RV.
Next, we asked whether the inhibitory effect was due to the enzymatic activity of SPHK1. Mutation of human SPHK1 glycine 26 or glycine 80 to aspartic acid 26 or aspartic acid 80 results in a catalytically inactive sphingosine kinase (22). Therefore, we generated SPHK1 G26D and SPHK1 G80D retroviral vectors. Th1 cells infected with SPHK1 G26D-RV and SPHK1 G80D-RV produced more IL-2, TNF-α, and IFN-γ than cells infected with GFP-RV (Fig. 5,B). Furthermore, SPHK1 G26D-RV and SPHK1 G80D-RV produced much more IL-2, TNF-α, and IFN-γ than cells infected with SPHK1-RV (Fig. 5 B). The repeated observations that IL-2, TNF-α, and IFN-γ were down-regulated by SPHK1 overexpression and up-regulated by SPHK1 G26D-RV, and SPHK1 G80D-RV are consistent with the siRNA data, indicating that SPHK1 negatively controls the expression of these Th1 cytokines.
CCL17 and CCL22 are highly expressed in Th2 cells and are inhibited by SPHK1
To obtain a broader view of SPHK1-controlled gene expression in T cells, we performed Affymetrix GeneChip microarray analysis of T cells overexpressing SPHK1. Th0 cells infected with SPHK1 GFP-RV or control GFP-RV were stimulated with anti-CD3 for 2 h, and the total RNA was extracted for the GeneChip analysis. As expected, we found that IL-2, TNF-α, and IFN-γ were inhibited in Th0 cells infected with SPHK1-RV (data not shown). We also discovered that the mRNA levels for several chemokines were decreased in SPHK1 RV-infected Th0 cells. Specifically, the mRNA level of chemokines CCL17 and CCL22 in SPHK1 RV-infected cells was significantly lower than that in control GFP-RV-infected cells (data not shown).
CCL17 (also known as thymus- and activation-regulated chemokine (TARC)) and CCL22 (also known as macrophage-derived chemokine), mainly produced by dendritic cells and epithelial cells (32, 33, 34), are known as Th2 chemokines that attract Th2 cells. However, there is no report showing that Th1 and Th2 cells themselves express both chemokines. From our GeneChip data, we found that both CCL17 and CCL22 were expressed in Th1 and Th2 cells (Fig. 6,A). Th2 cells expressed higher mRNA levels of CCL17 and CCL22 than Th1 cells. Anti-CD3 stimulation slightly enhanced their expression. A similar observation was made for Th1 and Th2 cells from the C57BL/6 background (data not shown). We also performed TaqMan real-time PCR analysis to detect the expression of CCL17 and CCL22 in Th1 and Th2 cells. Fig. 6 B shows that Th2 cells expressed higher levels of CCL17 and CCL22 than Th1 cells after cells were differentiated for 48 h or 1 wk. Higher levels of CCL17 and CCL22 mRNA were observed after cells were differentiated for 1 wk. Anti-CD3 stimulation also slightly increased their expression. The TaqMan real-time PCR results were consistent with the GeneChip data.
Finally, we asked whether CCL17 and CCL22 mRNA were indeed inhibited by SPHK1. As expected, TaqMan analysis showed that CCL17 and CCL22 mRNA were decreased in SPHK1-RV-infected T cells (Fig. 7,A). In addition, increased expression of CCL17 and CCL22 mRNA was observed in human Th1 cells when SPHK1 was knocked down by SPHK1 siRNA (Fig. 7 B). These results strongly support the idea that SPHK1 is a negative regulator of the expression of these chemokines.
In this report we showed that SPHK1, but not SPHK2, is selectively expressed in activated Th1 cells. The expression of SPHK1 in Th1 cells requires TCR signaling and new protein synthesis. When human SPHK1 expression was knocked down by SPHK1-specific siRNA, the production of IL-2, TNF-α, and IFN-γ from human Th1 cells in response to TCR stimulation was enhanced. Consistently, overexpression of enzyme-dead, dominant-negative SPHK1 mutants increased the production of IL-2, TNF-α, and IFN-γ in Th1 cells. Furthermore, overexpression of wide-type SPHK1 in Th1 and Th0 cells decreased the expression of IL-2, TNF-α, and IFN-γ. Moreover, overexpression of SPHK1 also inhibited IL-5 and increased IL-10 mRNA expression. To our knowledge, this is the first report showing that SPHK1 is highly induced in Th1 cells and that it is a negative regulator of cytokine production in these cells. In addition, we report that the Th2 chemokines CCL17 and CCL22 are highly expressed in Th2 cells. Indeed, Zhang et al. (35) first reported that CCL22 is differentially expressed in Th2 cells. The expression of CCL17 and CCL22 was also inhibited by SPHK1. We observed that SPHK1 inhibits cytokine and chemokine production in both human and mouse T cells, suggesting that it is a conserved mechanism. However, it is unclear how SPHK1 regulates the cytokine and chemokine expression.
The molecular mechanism of the inhibitory role of SPHK1 in cytokine production is not clear. SPHK1 activity may alter the half-life of transiently expressed messages or regulate cytokine mRNA expression at the transcription level. Yoshimoto et al. (12) showed that SPHK2 associates with the mouse IL-12Rβ1 and augments IL-12-induced STAT4-mediated transcriptional activation. In T cells overexpressing SPHK1, we observed no change in the mRNA level for the IL-12R β1- and β2-chains, IL-18R1, and IL-18R accessory protein (GeneChip results; data not shown). The levels of the transcription factors GATA3 and T-bet were not affected (data not shown). IL-12/IL-18-induced IFN-γ, growth arrest and DNA damage-inducible (GADD) 45β, and GADD45γ were also not significantly impaired (data not shown). Thus, SPHK1 does not inhibit the process of Th1 differentiation.
SPHK1 phosphorylates sphingosine to form S1P. The function of S1P is through its binding and activation of S1P receptors. Rosen et al. (14) reported that S1P receptor agonists S1P and FTY720 induced emptying of lymphoid sinuses by retention of lymphocytes on the abluminal side of the sinus-lining endothelium and inhibition of egress into lymph. They also showed that FTY720 induces immunosuppression through inhibition of both the recirculation of naive T cells and the release of Ag-activated T cells from the draining lymph node (36). Matloubian et al. (16) showed that S1PR1 is essential for lymphocyte recirculation and that it regulates egress from both thymus and peripheral lymphoid organs. Mature T cells are unable to exit the thymus in mice whose hemopoietic cells lack a single S1PR (16). They also discovered that S1PR1 promotes B cell localization in splenic marginal zone (37). Furthermore, S1P (0.3–3 μM) inhibits chemotaxis of CD4+ T cells to CCL-21 and CCL-5 (17, 18). These discoveries show the close relationship between S1P receptors and the migration of lymphocytes. We found that chemokine CCL17 and CCL22 are highly expressed in Th2 cells, whereas Th1 cells express moderate level of CCL17 and CCL22 (Fig. 6). CCR4, which is expressed in Th2 and T regulatory cells, is the receptor of CCL17 and CCL22 (38, 39, 40). Both CCL17 and CCL22 could chemoattract Th2 and T regulatory cells via CCR4 (38, 41, 42). The low levels of CCL17 and CCL22 in Th1 cells may be due to the inhibitory role of SPHK1. Thus, SPHK1 may play an important role in preventing Th2 and T regulatory cells from migrating into the inflammatory sites of Th1-related autoimmune and inflammatory diseases as well as Th1-related protective immune reaction.
The inhibition of Th1 cytokines by SPHK1 could be associated with S1P production. Indeed, Graeler et al. (17) and Dorsam et al. (18) reported that S1P decreased CD4+ T cell production of IFN-γ and IL-4. Idzko et al. (43) showed that S1P reduces IL-12 and TNF-α production and augments IL-10 release in maturating dendritic cells. Furthermore, S1P in maturating dendritic cells inhibits their capacity to induce Th1 immune response, resulting in a Th2 response. Although their finding was in dendritic cells, it is consistent with our observations in T cells. We showed SPHK1 reduced the expression of Th1 cytokine mRNA and increased IL-10 mRNA expression (Table I). The increased IL-10 expression was consistent with the report by Idzko et al. (43). However, Jin et al. (44) reported that S1P enhanced IL-2 and IFN-γ production by T cells stimulated with anti-CD3 and anti-CD28, although T cell proliferation was inhibited by S1P. This apparent contradictory report may be due to the different cell types used. Jin et al. (44) used whole human T cells, a mixture of naive and memory CD4+ T cells and CD8+ T cells, for their experiments. Graeler et al. (17, 18) used mouse CD4+ T cells for their S1P experiments. In our experiments we used human naive CD4+ T cells and mouse DO11.10 TCR-transgenic CD4+ T cells. In addition, we differentiated the CD4+ T cells into Th1 cells. An additional explanation could be the difference between intracellular S1P and extracellular S1P. Other intracellular functions of SPHK1 may also be considered. It has been recently shown that SPHK1 is an intracellular effector of phosphatidic acid (45). In addition to sphingosine and S1P, SPHK1 expressed in Th1 cells could alter the intracellular balance of other sphingolipid metabolites, including ceramide and sphingomyelin, which also play important roles in T cell signaling. Furthermore, SPHK1 could have another substrate in addition to sphingosine. Indeed, Gijsbers et al. (46) showed that 1-O-hexadecyl-2-desoxy-2-amino-sn-glycerol is a substrate for human SPHK. During our submission of this manuscript, Allende et al. (47) reported that lymphocyte distribution is unaffected in lymphoid organs of SPHK1-deficient mice, although S1P signaling regulated lymphocyte trafficking. They also showed that the S1P level in most tissues from SPHK1-deficient mice is not markedly decreased. Thus, SPHK1 is not essential for the generation of S1P, which meditates T lymphocyte trafficking through S1P1R. In SPHK1−/− mice, SPHK2 or another enzyme(s) is able to phosphorylate sphingosine to produce S1P. The SPHK1−/− study indicates that the suppressive function of SPHK1 might occur because of its alteration of the intracellular function of S1P and other sphingosine metabolites. It also indicates the possibility that the suppression is independent of S1P.
Our findings suggest that SPHK1 is a negative regulator of cytokine production in CD4+ Th1 cells. The present work implies that SPHK1 plays a role in the balance of cytokine and chemokine expression levels in Th1 cells. It is well known that overexpression of Th1 cytokines induces inflammation. The production of SPHK1 after T cell activation may serve as a negative feedback to limit the overexpression of Th1 cytokines. It also may play an important role in preventing the chemoattraction of Th2 and T regulatory cells by inhibiting CCL17 and CCL22. Our current work indicates that inhibition of SPHK1 might induce inflammatory and autoimmune diseases by overproduction of cytokines and chemokines. It remains unclear how SPHK1 inhibits the expression of these proinflammatory cytokines and chemokines. Additional studies will be required to understand this inhibitory mechanism.
We thank Drs. Jeanne Magram, Sheenah Mische, and Mark Labadia for their discussion of this manuscript.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Abbreviations used in this paper: SPHK, sphingosine kinase; CBA, cytometric bead array; m, murine; GFP-RV, GFP retrovirus; siRNA, short-interfering RNA; S1P, sphingosine-1-phosphate; MGB, minor groove binder; GADD, growth arrest and DNA damage-inducible.