CaMK4-dependent activation of AKT/mTOR and CREM-α underlies autoimmunity-associated Th17 imbalance (original) (raw)
CaMK4 expression is induced preferentially during Th17 differentiation. CaMK4 expression and activity is increased in T cells from patients with SLE (19, 20) and MRL/lpr lupus-prone mice (refs. 19, 20, and Supplemental Figure 1; supplemental material available online with this article; doi:10.1172/JCI73411DS1). To gain a better understanding of the role that CaMK4 plays in T cell function, we isolated naive CD4 T cells from MRL/lpr mice and stimulated them in the absence (Th0) or presence of polarizing cytokines to generate either Th17 cells (TGF-β + IL-6) or Tregs (TGF-β + IL-2) (10, 23). As shown in Figure 1A, T cell stimulation caused an increase in the levels of CaMK4 at the protein level. This was particularly marked in T cells stimulated under Th17-polarizing conditions (Figure 1A, lane 3) and dampened by Treg-inducing cytokines (Figure 1A, lane 4). To determine whether this effect was specific for the Th17-polarizing program, we differentiated naive CD4 T cells into Th1, Th2, and Th17 cells and Tregs and quantified Camk4 expression by real-time PCR. As shown in Figure 1B, Camk4 induction was significantly stronger in Th17 cells than in the other CD4 functional subsets. In order to clarify how the Th17-polarizing cytokines enhance the expression of CaMK4, we measured CaMK4 expression upon IL-1β, TGF-β, and/or IL-6 stimulation. Importantly, Camk4 mRNA was induced modestly by IL-6, TGF-β, or IL-1β alone but was induced significantly more by the combination of IL-6 and TGF-β and that of IL-6, TGF-β, and IL-1β. Induced Camk4 was inhibited by STAT3 or SMAD3 inhibitors, indicating that both signals are necessary for Camk4 induction under Th17 conditions (Supplemental Figure 2). Although IL-1β is crucial for the induction of Th17-producing cells (24), IL-1β did not have an additional effect when combined with Th17-promoting cytokines (IL-6 and TGF-β) in increasing CaMK4 (Supplemental Figure 2). CaMK4 is a member of a family of serine/threonine kinases that includes CaMK1, CaMK2γ, CaMK2δ, and CaMK4 (25). To determine the specificity of Th17-induced expression of CaMK4, we stimulated naive CD4 T cells under neutral (Th0) or Th17-inducing conditions and analyzed the expression of members of the CaMK4 family by real-time PCR at different time points (Figure 1C). Expression of Camk1, Camk2g, and Camk2d did not differ when cells were stimulated under Th0- and Th17-polarizing conditions. In sharp contrast, Camk4 expression was significantly higher early during Th17 differentiation. These results indicate that CaMK4 is preferentially induced during Th17 polarization. This phenomenon is particularly relevant, since patients with SLE and MRL/lpr mice have an increased abundance of IL-17–producing CD4+ and CD4–CD8– T cells (5, 26) and IL-17 has been associated with organ damage in lupus (5, 23).
CaMK4 is induced during Th17 differentiation. (A) Western blot analysis of CaMK4 and phospho-STAT3 in unstimulated (UN) cells from MRL/lpr mice and cells stimulated under Th0, Th17, and Treg conditions. Cumulative data of densitometry is also shown (*P < 0.05; mean ± SEM; n = 3). (B) Real-time PCR analysis of Camk4 mRNA in naive CD4+ T cells from MRL/lpr mice stimulated 6 hours in Th0-, Th1-, Th2-, Th17-, or Treg-polarizing conditions. Results were normalized to Gapdh (*P < 0.05; mean ± SEM; n = 4). (C) Expression of Camk1, Camk2d, Camk2g, and Camk4 mRNA in naive CD4+ T cells at different time points during Th0 or Th17 differentiation (*P < 0.05; mean ± SEM; n = 4–5). Data are representative of 3 independent experiments.
CaMK4 is necessary for in vitro Th17 differentiation. To determine whether CaMK4 plays a role in Th17 cell differentiation, we isolated naive CD4 T cells from WT or Camk4–/– OT-II mice and stimulated them under Th17-polarizing conditions. As shown in Figure 2A, absence of Camk4 caused a significant decrease in the percentage of IL-17–producing T cells (P = 0.0046). To expand these results, we transfected cells from _Camk4_-sufficient (Camk4+/+) or -deficient (Camk4–/–) OT-II mice with either an empty vector or a _Camk4_-encoding plasmid. This allowed us to evaluate WT cells with normal and augmented abundance of CaMK4 side by side _Camk4_-deficient cells before and after CaMK4 reconstitution (Supplemental Figure 3). Cells were then stimulated in Th17-polarizing conditions, and the frequency of IL-17–producing cells was quantified by flow cytometry (Figure 2B). As expected, absence of Camk4 resulted in fewer IL-17+ T cells (P = 0.0321). However, reconstitution of Camk4 restored the IL-17 production defect, and, importantly, Camk4 overexpression led to increased numbers of IL-17–producing cells (P = 0.0378; Figure 2B). KN-93 is an inhibitor of CaMK4 (27) that ameliorates disease in lupus-prone mice (21). We stimulated naive CD4 T cells in Th1-, Th2-, and Th17-polarizing conditions in the presence of 2 concentrations of KN-93 (4 and 10 μM). As shown in Figure 2C, KN-93 inhibited Th17 differentiation and IL-17 production in a dose-dependent manner (PBS vs. 4 μM, P = 0.0142; PBS vs. 10 μM, P < 0.0001). Consistent with these observations, mRNA levels of Th17 transcription factors and Th17 cell–associated cytokines were also decreased in the presence of KN-93 (Supplemental Figure 4). Conversely, its effects on the differentiation of Th1 and Th2 cells were negligible (Figure 2, C and D). Taken together, these data indicate that CaMK4 is a necessary element in Th17 differentiation and IL-17 production that can be modulated by a pharmacologic inhibitor of CaMK4.
CaMK4 controls the development of Th17 cells. (A) Naive T cells differentiated for 72 hours in Th1, Th2, or Th17 conditions from spleens of B6 or B6.Camk4–/– mice were gated on TCRβ+CD4+ and stained for intracellular expression of IFN-γ, IL-4, and IL-17A. A profile representative of 4 mice per group is shown (**P < 0.01; mean ± SEM). (B) OT-II (WT or Camk4–/–) cells were transfected with either empty vector or pCMV-CaMK4. 4 hours after transfection cells were stimulated under Th17 conditions. After 48 hours, IL-17–producing TCRβ+CD4+ T cells were measured by intracellular cytokine staining. A profile representative of 4 mice per group is shown (*P < 0.05, **P < 0.01; mean ± SEM). (C) Naive T cells differentiated for 72 hours in Th1, Th2, or Th17 conditions in the presence of different concentrations of KN-93 from spleens of MRL/lpr mice (16 weeks old) were gated on TCRβ+CD4+ and stained for intracellular expression of IFN-γ, IL-4, and IL-17A. A profile representative of 3 independent experiments with 3 to 5 mice per group is shown (*P < 0.05, **P < 0.01; mean ± SEM). (D) ELISA of IFN-γ, IL-4, and IL-17 in supernatants of naive T cells from MRL/lpr mice differentiated for 72 hours in Th1, Th2, or Th17 conditions in the presence of different concentrations of KN-93 (μM). (*P < 0.05; mean ± SEM; n = 3–5). Data are representative of 3 independent experiments with 3 to 5 mice.
Camk4 deficiency ameliorates EAE. To evaluate the relevance of CaMK4 in an IL-17–dependent inflammatory condition, we induced EAE (28) in _Camk4_-sufficient (WT) and -deficient mice by immunizing with myelin oligodendrocyte glycoprotein (MOG35–55) (29). WT mice developed signs of EAE on day 11 and reached the peak of the disease on day 15 after immunization. In contrast, in Camk4–/– mice, disease onset was delayed and was significantly less severe when quantified as a clinical score (P = 0.0318; Figure 3A) or percentage of weight loss (P = 0.0464; Figure 3B).
Camk4–/– mice are resistant to EAE. EAE was induced in WT and Camk4–/– mice by immunization with MOG35–55 emulsified in CFA. (A) The clinical score of EAE and (B) body weight in these mice were monitored (*P < 0.05; mean ± SEM; cumulative results of 3 independent experiments with 3 to 5 mice per group). (C) Flow cytometry of intracellular IL-17 and IFN-γ at day 8 in CD4+ T cells obtained from mononuclear cells isolated from the draining inguinal lymph nodes (dLNs) and spleens of WT and Camk4–/– mice immunized with MOG35–55 emulsified in CFA to induce EAE and then activated in vitro with PMA and ionomycin (**P < 0.01; mean ± SEM; n = 3–4). Data are representative of 2 independent experiments with 3 to 4 mice per group.
To determine whether CaMK4 inhibition ameliorated EAE by decreasing the differentiation of Th17 cells, we immunized WT and Camk4–/– mice with MOG35–55 and quantified the frequency of IL-17– and IFN-γ–producing cells 8 days later. As predicted by our in vitro results, CaMK4 deficiency reduced significantly the number of IL-17+ CD4 T cells in the spleens and draining inguinal lymph nodes of immunized mice, without affecting the production of IFN-γ (Figure 3C). Independent histological analysis of spinal cords demonstrated significantly decreased inflammation and demyelination in Camk4–/– mice (Figure 4).
Camk4–/– mice display less inflammation and demyelination in EAE. Spinal cord sections from WT and Camk4–/– mice obtained at 14 days after immunization. Sections were stained with (A and B) H&E to assess inflammation and (D and E) luxol fast blue to assess myelin content. Arrows indicate inflammatory cellular infiltrates. Scale bars: 50 μm (top rows); 200 μm (bottom rows). Quantitative cumulative data (n = 4 mice per group) are shown in C (*P < 0.05).
CaMK4 inhibition limits the production of IL-17 in MRL/lpr mice. CaMK4 expression and activity is increased in T cells from patients with SLE and lupus-prone mice (19, 20), and IL-17 has been suggested to play a role in target organ damage in lupus, including glomerulonephritis (30, 31). Therefore, we treated MRL/lpr mice with KN-93 for 10 weeks and examined IL-17A expression in spleens and lymph nodes during the peak of the disease (~16 weeks of age). KN-93 treatment led to a significant decrease in IL-17–producing CD4+ and CD4–CD8– (double-negative [DN]) T cells in spleens and in lymph nodes from mice treated with KN-93 (Figure 5, A and B). In line with these observations, mRNA levels of Rorc, the master regulator of Th17 cells (32), were also reduced by pharmacologic inhibition of CaMK4 (Figure 5C). The inhibitory effects of KN-93 were specific for IL-17, since it did not modify the frequency of IFN-γ–producing cells or the expression of the Th1- and Th2-associated transcription factors Tbx21 and Gata3, respectively (Figure 5, B and C). KN-93 treatment decreased significantly the mortality of MRL/lpr mice (Figure 5D) as well as serum titers of anti–double-stranded DNA antibodies at 12 and 16 weeks of age and proteinuria at 16 weeks of age (Supplemental Figure 5), confirming the relevance of IL-17 inhibition in this lupus model.
CaMK4 inhibition limits the production of IL-17 cytokines in MRL/lpr mice. (A) Cells from spleens and lymph nodes of PBS-treated and KN-93–treated MRL/lpr mice (16 weeks old) were gated on TCRβ+CD4+ and stained for intracellular expression of IL-17A and IFN-γ. Data are representative of 4 independent experiments with 4 to 5 mice per group. (B) Percentage of IL-17A–producing and IFN-γ–producing T subsets in spleens (n = 4–6) and LNs (n = 4–6). DN T cells are CD3+CD4–CD8– (*P < 0.05; mean ± SEM). (C) Quantitative real-time PCR analysis of the expression of Tbx21, Gata3, and Rorc mRNA in memory CD4+ T cells (*P < 0.05; mean ± SEM; n = 4). (D) Survival of MRL/lpr mice treated with PBS or KN-93 is depicted. Mice were observed until 20 weeks of age (*P < 0.05; n = 8–10 mice per group). Cumulative results of 3 independent experiments with 3 to 5 mice per group.
CaMK4 promotes transcription of Il17 through CREM-α. Methylation of CpG-DNA is associated with decreased transcription of neighboring genes (33). To investigate the mechanisms that regulate IL-17A expression during Th17 differentiation, we examined the CpG-DNA methylation of regulatory conserved noncoding sequences (CNSs) of the Il17 gene in CD4+ T cells from MRL/lpr and MRL/lpr.Camk4–/– mice (Supplemental Figure 6). After 24 hours of activation with anti-CD3 and anti-CD28 antibodies, CD4+ T cells from MRL/lpr mice exhibited low degrees of CpG-DNA methylation in all investigated regions of the Il17 locus (Figure 6A). In contrast, cells from MRL/lpr mice deficient in Camk4 had significantly higher levels of methylation (CNS1, P = 0.0382; CNS2, P = 0.0353). These results suggested that CaMK4 regulates IL-17 production by controlling its transcription. Since the phosphorylation and DNA-binding activity of the transcription factor CREM-α is regulated by CaMK4 (19, 20) and CREM-α has been shown to modulate IL17 transcription in T cells from patients with SLE (34, 35), we hypothesized that CaMK4 might control IL-17 expression through CREM-α. To evaluate this possibility, we analyzed the binding of CREM-α to consensus cAMP response element (CRE) sites within the Il17 promoters of CD4+ T cells from MRL/lpr and MRL/lpr.Camk4–/– mice by ChIP. As shown in Figure 6B, we detected reduced recruitment of CREM-α to CREs in the Il17 promoters of _Camk4_-deficient MRL/lpr mice. These results indicate that, through promoting the DNA-binding activity of CREM-α, CaMK4 facilitates Il17 transcription.
CaMK4 mediates CpG-DNA methylation of the Il17a gene through CREM-α. (A) CD4+ T cells from 14-week-old MRL/lpr and MRL/lpr.Camk4–/– mice were sorted for the assessment of CpG-DNA methylation of the Il17a promoter region using methylated CpG-DNA immunoprecipitation (*P < 0.05; mean ± SEM). Methylation index (MI), as assessed relative to methylated (100%) and unmethylated (0%) control DNA, is shown. Data are representative of 2 independent experiments with 4 mice per group. (B) CREM-α recruitment to the Il17a CRE-binding site in CD4+ T cells from MRL/MPJ, MRL/lpr, or MRL/lpr.Camk4–/– mice was detected by ChIP assay (*P < 0.05; mean ± SEM). Data are representative of 2 independent experiments with 4 mice per group. (C and D) Alignment of the CRE consensus sequence with the CRE site (–111/–104) and the ROR element binding site (–140/–135) of the proximal human IL-17A promoter is shown. Jurkat T cells were treated with (C) KN-93 (10 μM) or transfected with (D) either control siRNA or _CAMK4_-specific siRNA for luciferase assays using the IL-17A reporter plasmids. IL17Ap (-195mut)-luc indicates a reporter plasmid containing a site-directed mutation at the CRE site (–111/–104) and the ROR element binding site (–140/–135) (*P < 0.05; mean ± SEM; n =3–4). Cumulative results of 3 independent experiments.
To investigate whether CaMK4 regulates IL-17 expression at the transcriptional level, we cloned the IL17 promoter into a luciferase reporter system (36). This promoter region includes CRE and ROR element binding sites (35). As shown in Figure 6C, the cloned promoter region possessed transcriptional activity that was completely abrogated in cells treated with KN-93 or _Camk4_-specific siRNA, indicating that CaMK4 promotes the transcriptional activity of IL-17. As expected, the activity of the promoter was partially abrogated when the CRE site (–111/–104) was mutated and was completely lost when both the CRE and the ROR element binding sites (–140/–135) were mutated. CaMK4 inhibition by KN-93 or siRNA was still able to decrease the transcriptional activity of the promoter in the absence of the CRE site, suggesting that CaMK4 promotes Il17 transcription both through the CRE and the ROR element binding sites.
CaMK4 promotes AKT/mTOR signaling. The activation of mTORC1 enhances Th17 differentiation (37) and disruption of mTORC1 caused by deletion of Rheb or Raptor impairs Th17 differentiation (38, 39). CaMKs, including CamK4, have been reported to modulate the AKT signaling pathway (40, 41). Therefore, CaMK4 might promote IL-17 production by facilitating AKT/mTOR signaling. To test this hypothesis, we first established the physical association between AKT and CaMK4 by performing coimmunoprecipitation experiments (Figure 7A). Next, we analyzed the effect of KN-93 on the AKT/mTOR pathway activity induced by T cell stimulation. Western blots of cell lysates revealed that KN-93 significantly inhibited AKT phosphorylation in a dose-dependent manner (Figure 7B). KN-93 also inhibited the phosphorylation of p70S6, a substrate of mTOR (Figure 7C). These results were confirmed using T cells from Camk4–/– mice cultured in vitro under Th0- or Th17-polarizing conditions in the presence of OVA peptide (5 μM). As shown in Figure 7, D and E, phosphorylation of AKT and p70S6 was clearly decreased in the absence of Camk4. To further establish these findings, we incubated Jurkat T cells with KN-93 and quantified AKT and S6K phosphorylation by flow cytometry. KN-93 decreased AKT and S6K phosphorylation in Jurkat T cells (Supplemental Figure 7A). In concordance, overexpression of CaMK4 in Jurkat T cells induced increased AKT and S6K phosphorylation (Supplemental Figure 7B). Taken together, these results demonstrate that CaMK4 facilitates AKT/mTOR signaling. To determine whether CaMK4 promotes IL-17 production by facilitating AKT/mTOR signaling, we treated _Camk4_-overexpressing T cells with the mTORC1 inhibitor rapamycin (100 nM). mTORC1 blockade abrogated IL-17 production induced by CaMK4 overexpression (Supplemental Figure 8).
Inhibition of CaMK4 decreased Th17 differentiation through the blocking of the AKT/mTOR signaling pathway. (A) Immunoprecipitation and protein immunoblot analysis of AKT expression is shown. Jurkat T cells were stimulated with anti-CD3 and anti-CD28 antibodies for 20 minutes. Cell lysates were then prepared and immunoprecipitated with anti-CaMK4 or with a control rabbit antibody. The immunoprecipitates were then analyzed by immunoblotting using AKT antibody. The data are representative of 2 independent experiments. Western blotting analysis of (B) phospho-AKT and (C) phospho-S6K in CD4+ T cells from MRL/lpr mice in unstimulated and Th17 conditions for the indicated times. Western blotting analysis of (D) phospho-AKT and (E) phospho-S6K in OT-II cells (WT or Camk4–/–) in unstimulated, Th0 (with or without TGF-β), and Th17 conditions. The graphs in B–E show cumulative data of densitometry (*P < 0.05, **P < 0.01; mean ± SEM; n = 3–4). Data are representative of 3 independent experiments with 3 to 4 mice per group.
Silencing of CaMK4 suppresses Th17 cells in human T cells. To determine the relevance of our findings in human T cells, we analyzed the effect of CaMK4 inhibition in T cells from healthy donors or patients with SLE. We first asked whether KN-93 can inhibit Th17 differentiation in controls. As expected, KN-93–treated T cells displayed a substantial reduction of IL-17–producing cells in a dose-dependent manner (Figure 8A). To determine the effects of CaMK4 inhibition on IL-17 expression in T cells from patients with SLE, we stimulated cells transfected with _CAMK4_-specific or control siRNA with anti-CD3, anti-CD28, IL-6, and TGF-β. As shown in Figure 8B, CaMK4 inhibition decreased significantly IL17A and IL17F mRNA levels in cells from healthy donors and patients with SLE. Taken together, our results indicate that CaMK4 positively regulates IL-17 production in T cells from healthy donors and patients with SLE.
Silencing of CaMK4 decreases Th17 cells in patients with SLE. (A) T cells from healthy donors were stimulated in Th17 conditions in the absence or presence of KN-93 for 72 hours, and IFN-γ–producing CD4+ T cells or IL-17A–producing CD4+ T cells were determined by intracellular cytokine staining. A representative experiment from 1 of 4 donors is shown. (B) T cells from normal controls (n = 4) or patients with SLE (n = 6) were transfected with either control siRNA or _CAMK4_-specific siRNA. 24 hours after transfection, cells were differentiated for 72 hours under Th17 conditions and analyzed by quantitative real-time PCR of IL17A or IL17F mRNA (*P < 0.05; mean ± SEM).