Canonical Wnt Signaling in Dendritic Cells Regulates Th1/Th17 Responses and Suppresses Autoimmune Neuroinflammation

This article is featured in In This Issue, p.2957

Multiple sclerosis (MS) is a chronic autoimmune inflammatory condition that leads to multifocal demyelination in white matter of the human CNS. Using EAE, a mouse model for MS, studies have shown that dendritic cells (DCs) play a critical role in initiation and development of CNS pathology (1–3). Accordingly, innate immune receptors, including TLR-mediated signaling in DCs, play a critical role in the initiation of experimental autoimmune encephalomyelitis (EAE). These receptors on DCs sense various danger signals and induce the activation of several signaling networks with secretion of cytokines that drive the differentiation of naive CD4⁺ and CD8⁺ T cells to effector or regulatory T cells (Treg) (4). Activation of most TLRs on DCs induces secretion of IL-12p70 that promotes an IFN-γ⁺ CD4⁺ T cell (Th1) response, whereas dectin-1–mediated signals in DCs induce strong production of IL-6 and IL-23 that promote an IL-17A⁺ CD4⁺ T cell (Th17) response (4, 5). Other microbial stimuli that activate TLR2 or TLR9 induce immune regulatory molecules such as IL-10, retinoic acid (RA), TGF-β, and IDO that promote IL-4–producing CD4⁺ (Th2) or Treg responses (5). Similarly, DCs contribute to CNS pathology through differentiation and activation of naive CD4⁺ T cells to myelin-specific Th1 and Th17 cells (6, 7). In addition to CD4⁺ effector T cells, CD8⁺ T cells also contribute to CNS pathology during EAE and MS (8, 9). Conversely, emerging evidence suggests that DCs are also critical in resolving inflammation and limiting immune-mediated pathology in EAE by producing immune regulatory factors and initiating Treg activation (3, 5, 10). Thus, DCs play a key role in bridging innate and adaptive immunity. However, the receptors and signaling networks that program DCs to control inflammation and autoimmunity are not known. Thus, understanding these events may represent promising targets for therapeutic intervention of various autoimmune and chronic inflammatory conditions.

Low-density lipoprotein receptor-related protein (LRP)5 and LRP6 coreceptors are critical signaling mediators of the canonical wnt-signaling pathway (11). β-catenin, a transcriptional cofactor, is an important downstream mediator of LRP5 and LRP6 signaling (11). Alterations or defects in the LRP5- and LRP6-mediated signaling pathway are associated with several human inflammatory diseases (11, 12). Interestingly, increased wnt ligand expression is observed in several inflammatory diseases (13–17). Although DCs are present in low number in the CNS under homeostatic conditions, their numbers increase drastically during autoimmune inflammation, infection, or trauma (18). However, the functional and biological significance of the wnt signaling pathway in regulating ongoing inflammation and establishing immune homeostasis is poorly understood. In this context, our previous study has shown that the β-catenin pathway in intestinal DCs plays an important role in limiting inflammation and promoting gut homeostasis (19). We have also shown that the activation of β-catenin via the TLR2 pathway in DCs can suppress neuroinflammation (20). So, we hypothesize that wnt ligands in the tissue microenvironment activate the LRP5/6-β-catenin pathway in DCs and program them to limit inflammation and immune-mediated pathology.

In the current study, we report that, during the induction and effector phase of EAE, the canonical wnt-signaling pathway in DCs regulates the magnitude of inflammatory responses and limits collateral damage to the host. Accordingly, our data demonstrate that DC-specific deletion of both LRP5 and LRP6 coreceptors or the key downstream signal mediator, β-catenin, in mice results in severe EAE pathology. This was because of increased expression of proinflammatory cytokines by DCs lacking LRP5/6 or β-catenin with increased Th1 and Th17 cell polarization over Treg polarization by these DCs. In contrast, pharmacological activation of the β-catenin pathway significantly reduced EAE severity. Taken together, our data demonstrate that the canonical wnt/β-catenin pathway in DCs promotes Treg responses over pathogenic Th1 and Th17 cells in a LRP5/6 coreceptor-dependent manner. Thus, manipulating wnt/β-catenin signaling may represent a convenient therapeutic approach to improve the outcome of MS and other inflammatory diseases.

C57BL/6 male mice 6–12 wk of age were purchased from The Jackson Laboratory (Bar Harbor, ME). LRP5-floxed mice (21) and LRP6-floxed mice (21) were provided by B. Williams (Van Andel Research Institute) and were crossbred to generate LRP5/6-floxed mice. TCF/LEF-reporter mice (22), β-catenin–floxed mice (23), CD11c-cre (24), and 2D2 myelin oligodendrocyte glycoprotein (MOG)-specific TCR transgenic mice (25) were originally obtained from Jackson ImmunoResearch Laboratories and bred on-site. β-catenin–floxed mice were crossed with transgenic mice expressing cre recombinase enzyme under the control of CD11c promoter to generate mice lacking β-catenin in DCs (β-cat^ΔDC). Successful cre-mediated deletion was confirmed by PCR and protein expression analyses, as previously described (19). LRP5/6-floxed mice were crossed to CD11c-cre mice to generate mice in which LRP5/6 (LRP5/6^ΔDC) were deficient in DCs. Successful cre-mediated deletion was confirmed by PCR, as in our previous studies. β-Cat^flox(ex3) mice (26) were bred to CD11c-cre mice to generate mice that expresses active β-catenin specifically in DCs (Act-β-cat^DC). All of the mice were housed under specific pathogen-free conditions in the Laboratory Animal Services of Georgia Regents University. Animal care protocols were approved by the Institutional Animal Care and Use Committee of Georgia Regents University.

CD4 (RM4-5), IL-17 (TC11-18H10), IFN-γ (XMG1.2), anti-CD3 (145.2C11), anti-CD28 (37.51), and brefeldin A were purchased from BD Biosciences. Foxp3-PE (FJK-16s), CD11c (N418), and CD11b (M1/70) were purchased from eBioscience. MOG_35–55 peptide (MEVGWYRSPFSRVVHLYRNGK) was purchased from Anaspec. Abs against mouse CD4 (GK1.5), CD8a (53-6.7), CD45 (30-F11), Foxp3 (clone FJK-16s), IL-10 (JES5-16E3), CD11c (clone N418), CD90.1 (HIS51), IL-17 (TC11-18H10), and IFN-γ (XMG1.2) were purchased from eBioscience. Nonphospho-active β-catenin, β-catenin, and TCF4 Abs were obtained from Cell Signaling Technology. β-galactosidase (β-gal) Ab was purchased from Abcam. Wnt agonist II/β-catenin agonist (SKL2001) and Wnt inhibitor (Porcn Inhibitior II, C59) were purchased from Calbiochem. Recombinant Wnt3a, Wnt2b, and Wnt5a were purchased from R&D Systems.

EAE induction experiments were performed as described in our previous study (27). EAE was induced by s.c. immunization in the hind flanks on day 0 using 100 μg MOG_35–55 emulsified in CFA containing 2.5 mg/ml heat-inactivated Mycobacterium tuberculosis (Difco). Mice also received 250 ng pertussis toxin (List Biological Laboratories) i.p. on days 0 and 2 postimmunization (pi). Disease severity was assessed on different days pi according to the following scale: 0, no disease; 1, flaccid tail; 2, hind limb weakness; 3, hind limb paralysis; 4, forelimb weakness; and 5, moribund. In some experiments, EAE-induced wild-type (WT) mice were treated with either Wnt inhibitor (5 mg/kg) on days 0, 3, and 5 pi or Wnt-agonist/β-catenin agonist (5 mg/kg) 2 d before EAE induction, followed by treatment on days 0, 3, and 5 pi. For therapeutic treatment, EAE-induced WT mice were treated with Wnt-agonist/β-catenin agonist (5 mg/kg) on day 10, followed by treatment on days 13 and 16 pi.

Mice were euthanized with CO₂ and perfused through the left ventricle with PBS. The brain and spinal cord were removed from each animal and dissected into small fragments, followed by digestion with collagenase type 4 (1 mg/ml) in complete DMEM plus 2% FBS for 30 min at 37°C. Leukocytes were isolated using 40% Percoll (Sigma-Aldrich) and then were stained for CD4⁺ and CD8⁺ T cells expressing Foxp3 and different intracellular cytokines.

CD11c⁺ DCs were purified from the CNS, draining lymph nodes (DLN), or spleen, as previously described (27). In brief, spleens or brain were cut into small fragments and then digested with collagenase type 4 (1 mg ml⁻¹) in complete DMEM plus 2% FBS for 30 min at 37°C. Cells were washed twice and enriched for CD11c⁺ DCs with the CD11c-specific microbeads from Miltenyi Biotec. The resulting purity of CD11c⁺ DCs was ∼95%.

Single-cell suspensions from CNS leukocytes, DLN, and spleen were resuspended in PBS containing 5% FBS. After incubation for 15 min at 4°C with the blocking Ab 2.4G2 (anti-FcγRIII/I), the cells were stained at 4°C for 30 min with the appropriately labeled Abs. Samples were then washed twice in PBS containing 5% FBS. The samples were either immediately analyzed at this point or fixed in PBS containing 2% paraformaldehyde and stored at 4°C. Intracellular staining for β-catenin, active β-catenin, and β-gal was performed using rabbit mAb or with appropriate isotype control in TBS containing 1% BSA, followed by incubation with Alexa Fluor 488–conjugated rabbit anti-goat Ig or goat anti-rabbit IgG (Molecular Probes, Eugene, OR). To measure cytokines, single-cell suspensions from the DLN or CNS were ex vivo stimulated with PMA/ionomycin and brefeldinA/monensin (eBioscience) for 6 h at 37°C. The cells were then stained for CD4 and CD8, followed by intracellular staining of IFN-γ, IL-17A, TNF-α, and IL-10. To measure Treg or β-catenin or β-gal expression, corresponding Abs were added after permeabilization and fixation of cells. Flow cytometric analyses were performed using a FACS LSRII system (BD Biosciences), and the data were analyzed using FlowJo software (Ashland, OR).

MOG_38–49–I-A^b tetramers (PE) were provided by the National Institutes of Health Tetramer Core Facility at Emory University. CNS-infiltrating leukocytes were resuspended in PBS containing 5% FBS, followed by tetramer staining for 1 h at 37°C, as described in a previous study (28). After 1 h, cells were stained with anti-CD4 (allophycocyanin) and anti-CD45 (Alexa 700) for 30 min on ice. Cells were washed three times, followed by acquisition on flow cytometer. A nonspecific peptide tetramer was used as negative control. The percentage of tetramer-PE–positive cells was determined in CD4-positive populations based on negative control staining.

β-cat^fl/fl, β-cat^ΔDC, and LRP5/6^ΔDC recipient mice were reconstituted with 2.5 × 10⁶ 2D2 TCR transgenic CD4⁺ T cells, followed by EAE immunization. Five days later, DLNs were removed and, after RBC lysis, in vitro recall responses were assayed by restimulating DLN cells (2 × 10⁶/ml) for 6 h with PMA/ionomycin in the presence of brefeldin A for intracellular cytokine detection.

CD11c⁺ splenic DCs (10⁶ cells/ml) were cultured with M. tuberculosis for 24 h. The supernatants were collected for cytokine analysis by ELISA, whereas cells were collected for gene expression analysis by RT-PCR. For bone marrow–derived DCs, single-cell suspension from mouse bone marrow of WT or LRP5/6^ΔDC mice was cultured for 7 d in the presence of GM-CSF (10 ng/ml; PeproTech) and IL-4 (1 ng/ml; PeproTech) in 10% FBS containing RPMI 1640. At end of 7-d stimulation, bone marrow–derived DCs were stimulated with M. tuberculosis for 24 h, followed by cytokine analysis in cell supernatants using ELISA.

In vitro stimulation was performed, as previously described (27). In brief, purified splenic CD11c⁺ DCs (10⁶ cells/ml) were stimulated with M. tuberculosis and washed with media three times. In some experiments, DCs were cultured with Wnt3a (R&D Systems; 0.5 μg/ml) and DKK1 (0.5 μg/ml). Activated DCs (2 × 10⁴) were cultured together with naive 2D2 CD4⁺CD62L⁺ T cells (10⁵) in 200 μl RPMI 1640 complete medium in 96-well round-bottom plates in the presence of MOG peptide (1 μg/ml). After 96 h, cells were restimulated with PMA/ionomycin for 6 h in the presence of brefeldin A for intracellular cytokine detection.

Total RNA was isolated from purified splenic, DLN, or CNS-infiltrating CD11c⁺ DCs using the Qiagen RNeasy Mini Kit, according to the manufacturer’s protocol (Qiagen). cDNA was generated using the superscript First-Strand Synthesis System for RT-PCR and random hexamer primers (Invitrogen), according to the manufacturer’s protocol. cDNA was used as a template for quantitative real-time PCR using SYBER Green Master Mix (Bio-Rad) and gene-specific primers, as described in our previous study (27), and gene expression was calculated relative to the housekeeping gene GAPDH. The following primers were used for analysis: wnt2b forward, 5′-CCGAGGGTATGACACAACTC-3′, and reverse, 5′-GTGGAGGGAAGAATGAGGTT-3′; wnt3 forward, 5′-TGTGAGGACACTTGAGCAGA-3′, and reverse, 5′-TTTGGATACAGCAGGTTGGT-3′; wnt5a forward, 5′-GCAGGACTTTCTCAAGGACA-3′, and reverse, 5′-CCCTGCCAAAGACAGAAGTA-3′; wnt5b forward, 5′-GGATGGATGGATGGATGATA-3′, and reverse, 5′-CTAATCCCCACCTGTCTCCT-3′; wnt7a forward, 5′-CAAAGTTTTCTACGGCAGGA-3′, and reverse, 5′-CTCCCAGTCCTAGCAAGTCA-3′; wnt8a forward, 5′-CCTGAGCATGCTTTTCAGTT-3′, and reverse, 5′-CCACCTGTTTTTCCATTTTG-3′; wnt8b forward, 5′-GGGTAAGAGGTAACCCCAGA-3′, and reverse, 5′-GCCAACCTGCCTACTACAGA-3′; wnt9a forward, 5′-GGCACAGGGTTACAAACAAC-3′, and reverse, 5′-GGACAGAGGCAACTGAGAAA-3′; wnt10a forward, 5′-ATGAGTGCCAGCATCAGTTC-3′, and reverse, 5′-GCCTTCAGTTTACCCAGAGC-3′; Ifn-g forward, 5′-CTCTTCTTGGATATCTGGAGG-3′, and reverse, 5′- CCTGATTGTCTTTCAAGACTTC-3′; IL17 forward, 5′-AATGCCCTGGTTTTGGTTGAA-3′, and reverse, 5′-CATTGATGCAGCCTGAGTGTCT-3′; IL22 forward, 5′-TCCGAGGAGTCAGTGCTAAA-3′, and reverse, 5′-AGAACGTCTTCCAGGGTGAA-3′; TNF-α forward, 5′-ATCATCTTCTCAAAATTCGAGTGA-3′, and reverse, 5′-TTGAGATCCATGCCGTTGG-3′; IL6 forward, 5′-AGACAAAGCCAGAGTCCTTCAGAGA-3′, and reverse, 5′-GCCACTCCTTCTGTGACTCCAGC-3′; IL1β forward, 5′-TGTAATGAAAGACGGCACACC-3′, and reverse, 5′-TCTTCTTTGGGTATTGCTTGG-3′; IL12p40 forward, 5′-CAATCAGGGCTTCGTAGGTA-3′, and reverse, 5′-GGCCCTGGTTTCTTATCAAT-3′; IL10 forward, 5′-CAGAGCCACATGCTCCTAGA-3′, and reverse, 5′-TGTCCAGCTGGTCCTTTGTT-3′; IL23 forward, 5′-AATAATGTGCCCCGTATCCA-3′, and reverse, 5-AGGCTCCCCTTTGAAGATGT-3′; Gmcsf forward, 5′-TTTACTTTTCCTGGGCATTG-3′, and reverse, 5′-TAGCTGGCTGTCATGTTCAA-3′; Il-27p28 forward, 5′-CTGAATCTCGATTGCCAGGAGTGA-3′, and reverse, 5′-AGCGAGGAAGCAGAGTCTCTCAGAG-3′.

Spinal cords from EAE-induced mice were removed after perfusion and fixed using 10% formalin in PBS. Fixed spinal cords were embedded in paraffin, followed by sectioning and staining with H&E to study leukocyte infiltration and pathology as well as luxol fast blue staining to demonstrate CNS demyelination.

Statistical analyses were conducted using GraphPad Prism (Software for Science). Mean clinical scores were analyzed using the Mann–Whitney nonparametric t test. The statistical significance of differences in the means ± SEM of cytokines released by cells of various groups was calculated with the Student's t test (one tailed).

To investigate the role of wnt/β-catenin signaling in regulating autoimmune neuroinflammation, we induced EAE by immunizing mice with MOG_35–55 peptide emulsified in CFA, as previously described (20, 27). First, we characterized wnt gene expression in draining lymph nodes (DLN) on day 2 (induction phase) and in the CNS on day 11 (effector phase) of EAE. We noted a significant increase in the expression of wnt3a and wnt5a in DLNs during the induction phase of EAE (Supplemental Fig. 1A). Similarly, there was a significant increase in the expression of wnt2b and wnt3a in the CNS during the effector phase of EAE (Supplemental Fig. 1B). Next, we determined whether DCs express coreceptors LRP5 and LRP6 that mediate canonical wnt signaling. Enriched splenic CD11c⁺ DCs from WT B6 mice showed both LRP5 and LRP6 expression under steady state conditions (Supplemental Fig. 1C).

To address the role of LRP5- and LRP6-mediated signaling in DC function during autoimmune neuroinflammation, we generated DC-specific deletion of LRP5 or LRP6 by crossing LRP5- or LRP6-floxed mice (21) to CD11c-cre mice (24). We immunized mice specifically lacking LRP5 or LRP6 in DCs (LRP5^ΔDC or LRP6^ΔDC) and WT mice (littermate-floxed mice; WT^FL/FL) to induce EAE, and disease progression was monitored at different days pi. DC-specific deletion of either LRP5 or LRP6 individually had no effect on disease progression and severity compared with WT control mice (data not shown). Because it is possible that LRP5 and LRP6 may play a compensatory role in DCs in response to inflammation, we then crossed LRP5^ΔDC and LRP6^ΔDC together to generate DC-specific deletion of both LRP5 and LRP6 (LRP5/6^ΔDC). Next, we induced EAE in WT and LRP5/6^ΔDC mice, and disease severity was compared at different days pi. In contrast to WT control mice, LRP5/6^ΔDC mice were more susceptible and developed a more severe form of EAE (Fig. 1A). The histopathological analysis of CNS showed a marked increase in leukocyte infiltration and intense demyelination in LRP5/6^ΔDC mice compared with WT mice (Fig. 1B).

CNS-infiltrating CD4 and CD8 effector T cells contribute to the pathogenesis of EAE (6, 7, 9). Thus, we next determined whether increased EAE disease severity observed in LRP5/6^ΔDC mice was due to effector T cell subsets. As shown in Fig. 1C–E, there was a significant increase in the total number of CD4⁺ and CD8⁺ T cells as well as the frequency of MOG_33–55 tetramer-specific CD4⁺ T cells in the CNS of LRP5/6^ΔDC mice when compared with control mice. Furthermore, intracellular cytokine analysis showed significant increase in the frequencies of IFN-γ⁺, IL-17⁺, and TNF-α⁺ CD4⁺ T cells (Fig. 1F, 1G). Moreover, there were significantly higher frequencies of IFN-γ⁺ and TNF-α⁺ CD8⁺ T cells in the CNS of LRP5/6^ΔDC mice than in WT mice (Fig. 1H). Consistent with these observations, treatment of mice with wnt inhibitor resulted in a more severe disease compared with control mice (Supplemental Fig. 2A). Likewise, treatment of mice with wnt inhibitor resulted in an increase in the frequencies of CD4 and CD8 effector T cells in the CNS compared with control mice (Supplemental Fig. 2B, 2C). In contrast, we observed a decrease in the frequency of IL-10⁺ CD4⁺ regulatory (Tr1) cells in the CNS of LRP5/6^ΔDC mice or upon wnt inhibitor-treated mice than in the CNS of WT control mice, whereas the frequency of Foxp3⁺ CD4⁺ T cells was similar in both groups (Fig. 1I, Supplemental Fig. 2D). Consistent with the above observations, CNS-infiltrating leukocytes from LRP5/6^ΔDC mice showed increased mRNA expression for IFN-γ, IL-17A, IL-22, TNF-α, and GM-CSF as compared with WT control mice (Fig. 1J). Collectively, these data suggest that DC-specific expression of at least either LRP5 or LRP6 is critical for limiting autoimmune CNS pathology. Our results also show that the absence of both of these coreceptors results in increased EAE disease severity through promotion of effector T cell responses, indicating a possible regulatory role for LRP5/6-mediated signaling in DCs during ongoing neuroinflammation.

DCs control the fate of naive T cell differentiation through the secretion of various inflammatory or anti-inflammatory cytokines (3, 4). The above results prompted us to assess whether LRP5/6-mediated signals regulate DC function through the expression of inflammatory and anti-inflammatory cytokines. So, we assessed the cytokine production by DCs upon adjuvant (M. tuberculosis) treatment. As shown in Fig. 2A, M. tuberculosis stimulation of LRP5/6-deficient DCs led to significantly increased production of IL-6, TNF-α, IL-1β, IL-12p40, and IL-12p70 compared with WT DCs. In contrast, there was significantly reduced production of IL-10 by LRP5/6-deficient DCs compared with control DCs (Fig. 2A). These results prompted us to determine whether LRP5/6 signaling regulates the cytokine expression by DCs during the EAE induction phase. We observed significantly increased mRNA levels for IL-6, TNF-α, IL-1β, and IL-23 in DLN DCs isolated from EAE-induced LRP5/6^ΔDC mice compared with WT control mice (Fig. 2B). In contrast, there was significant reduction in mRNA levels for immune regulatory genes such as IL-10, TGF-β, and IL-27 (p28) by LRP5/6^ΔDC DCs compared with control DCs (Fig. 2B). In addition to effector T cell differentiation during the EAE induction phase, DCs infiltrating the CNS during the effector phase reactivate primed T cells and function as effector cells to cause CNS lesions. So, we examined the expression of inflammatory and anti-inflammatory cytokines in CNS-infiltrating DCs. As shown in Fig. 2C, there were significant increases in mRNA levels of Il-6 and Tnfa, but decreased levels of Il-10 in LRP5/6^ΔDC CNS DCs compared with control CNS DCs. Thus, our data demonstrate that LRP5/6-mediated signaling in DCs is critical for limiting inflammatory cytokine expression and inducing anti-inflammatory cytokines during the induction and effector phase of EAE.

As DCs dictate the fate of naive CD4⁺ T cells through differential production of pro- and anti-inflammatory cytokines (4), we further considered the functional relevance of DC-specific LRP5/6-mediated signaling in naive CD4⁺ T cell differentiation. We adoptively transferred naive CD4⁺CD25⁻ T cells from 2D2 mice (25) into WT and LRP5/6^ΔDC mice and then challenged these mice with MOG_35–55 plus CFA. Intracellular cytokine analysis on day 5 postchallenge showed a significant increase in naive 2D2 T cell differentiation toward Th1 and Th17 cells in LRP5/6^ΔDC mice compared with WT mice (Fig. 3A, 3B). Next, we characterized whether increased inflammatory Th1 and Th17 cell responses and reduced Treg responses observed in LRP5/6^ΔDC mice are due to altered DC maturation and activation. However, characterization of DLN DCs from LRP5/6^ΔDC mice on day 2 pi showed no significant difference in the expression of MHC class II, CD40, CD80, and CD86 as compared with control WT mice upon CFA immunization (Fig. 3C). Thus, these data indicate that activation of LRP5/6-mediated signaling in DCs regulates the Th1 and Th17 cell differentiation through differential expression of pro- and anti-inflammatory cytokine production in the DLN as well as in the CNS.

Next, we determined whether EAE immunization affects β-catenin expression and activation in DCs during the initiation phase (in DLN) and effector phase (in CNS) of the EAE. To test this, we quantified β-catenin expression in DLN DCs on day 2 and in CNS DCs on day 16 post-EAE induction. As shown in Fig. 3D, DCs isolated from the DLN and CNS showed similar levels of β-catenin expression compared with control DCs. However, we observed a marked increase in active β-catenin levels in both DLN and CNS DCs (Fig. 3D). Upon activation, β-catenin translocates into the nucleus, where it binds to T cell factor (TCF) family members to form a complex and regulate target gene expression. Therefore, we next determined whether β-catenin activation in DCs during the effector phase of EAE activates TCF expression using β-catenin/TCF reporter mice (22). We observed markedly higher levels of β-gal expression in DCs isolated from the DLN and CNS of EAE-induced mice (Fig. 3D). These data suggest that the β-catenin/TCF pathway is activated in DCs in the DLN and CNS during the initiation and effector phases of EAE.

Next, we studied whether activation of β-catenin in DCs affects the differentiation of naive CD4⁺ T cells. To test this, we adoptively transferred naive CD4⁺ T cells from 2D2 mice into the WT mice and mice lacking β-catenin expression in DCs (β-cat^ΔDC mice), followed by challenge with MOG_35–55 in CFA. As shown in Fig. 3E and 3F, donor T cells isolated from β-cat^ΔDC mice showed significantly increased frequencies of Th1 and Th17 cells compared with control WT mice. Collectively, our results indicate that LRP5/6-mediated activation of β-catenin in DCs regulates the differentiation of naive CD4⁺ T cells into Th1 and Th17 cells through regulation of DC-specific cytokine production. Furthermore, our data indicate a regulatory role for the canonical wnt pathway during ongoing neuroinflammation, in which activation of this pathway in DCs limits the uncontrolled differentiation of naive CD4⁺ T cells to Th1 and Th17 cells.

Next, we sought to determine whether increased wnt ligand expression in the DLN (day 2) and CNS (day 11) activates β-catenin in DCs and its effect on naive CD4⁺ T cell differentiation. Previous studies have shown that wnt3a specifically activates β-catenin in DCs (29–31). Consistent with these studies (29–31), the wnt3a treatment of DCs from TCF reporter mice showed an increase in the reporter gene expression compared with untreated DCs (Fig. 4A). In contrast, wnt2b and wnt5a failed to activate β-catenin in DCs (data not shown). Next, we examined whether wnt3a exerts a regulatory effect on DCs to drive T cell differentiation. Accordingly, M. tuberculosis–treated DCs cocultured with naive CD4⁺ T cells led to an increase in Th1 and Th17 differentiation compared with control DCs (Fig. 4B). Interestingly, the addition of DKK1, a negative regulator of wnt ligands, during M. tuberculosis treatment of DCs significantly increased the frequencies of Th1 and Th17 cells (Fig. 4B). Conversely, wnt3a treatment during M. tuberculosis pulsing of DCs significantly reduced Th1 and Th17 cell differentiation (Fig. 4B). Furthermore, addition of wnt3a along with DKK1 inhibited the regulatory effect of wnt3a, as observed by increased frequencies of Th1 and Th17 cells to levels equal to that of M. tuberculosis–treated DCs (Fig. 4B). Consistent with these observations, DCs conditioned with wnt3a produced significantly lower levels of inflammatory cytokines and more IL-10 and TGF-β compared with the control (Fig. 4C).

To further explore the role of LRP5/6-mediated activation of the wnt/β-catenin signaling pathway during neuroinflammation, we analyzed whether activation of β-catenin in DCs is critical for limiting EAE pathology. To test this, we monitored EAE disease severity over time in mice that specifically lack β-catenin in DCs (19). Similar to the deletion of LRP5/6 in DCs, deletion of β-catenin in DCs significantly increased EAE severity (Fig. 5A). Accordingly, histopathological analysis of the CNS of β-cat^ΔDC mice showed increased inflammation, as observed by increased leukocyte infiltration and severe demyelination compared with WT control mice (Fig. 5B). Moreover, the CNS of β-cat^ΔDC mice had higher numbers of CD4⁺ and CD8⁺ T cells as well as increased frequency of MOG_35–55-specific CD4⁺ T cells compared with control mice (Fig. 5C–E). Consistent with these data, intracellular cytokine analysis showed higher frequencies of Th1, Th17, TNF-α⁺ CD4⁺ T cells, IFN-γ⁺ CD8⁺, and TNF-α⁺ CD8⁺ T cells with lower percentage of Tr1 cells and no change in Tregs in the CNS of β-cat^ΔDC mice compared with control mice (Fig. 5F–H).

To further confirm the regulatory role of the wnt/β-catenin pathway in DCs during autoimmune neuroinflammation, we immunized mice expressing constitutively active β-catenin in DCs (Act-β-cat^DC) and monitored EAE progression. In contrast to WT control mice, Act-β-cat^DC mice showed significantly diminished EAE disease severity (Fig. 6A). Consistent with this observation, we also noted a significant reduction in MOG_35–55-specific CD4⁺ T cell, Th1, and Th17 cell responses, whereas an increase in Tr1 cells was observed in Act-β-cat^DC mice compared with control mice (Fig. 6B–E). In conclusion, our data indicate that EAE induction leads to the expression of wnt ligands in the CNS that activates the β-catenin pathway in DCs and potentially limits the expansion of Th1 and Th17 cells. Furthermore, our data suggest a possible feedback regulatory role for the wnt/β-catenin pathway in controlling excessive neuroinflammation through induction of wnt ligands, which activate the β-catenin pathway in DCs and modulate T cell differentiation.

Collectively, the above data indicated that β-catenin is downstream of the LRP5/6 signaling pathway in DCs, and its activation is critical for limiting EAE pathology. So, we explored the possible therapeutic implication of pharmacologically activating the canonical wnt/β-catenin pathway during ongoing neuroinflammation. First, we examined the effect of activating the β-catenin pathway in DCs on the expression of inflammatory and anti-inflammatory cytokines in response to M. tuberculosis. As shown in Fig. 7A, M. tuberculosis treatment of DCs led to increased production of IL-1β, IL-6, TNF-α, and IL-12p40 compared with untreated DCs. Interestingly, the addition of β-catenin agonist to M. tuberculosis–treated DCs significantly reduced IL-1β, IL-6, TNF-α, and IL-12p40 production by DCs compared with M. tuberculosis–treated DCs alone (Fig. 7A). In contrast, we observed a significant increase in IL-10 levels in M. tuberculosis plus β-catenin–activated DCs compared with M. tuberculosis–treated DCs (Fig. 7A). These in vitro data further confirmed that the β-catenin pathway in DCs regulates the expression of anti-inflammatory cytokines and proinflammatory cytokines.

Next, we assessed whether activation of this pathway could ameliorate EAE pathology. To study this, EAE-induced WT mice were treated with β-catenin agonist (5 mg/kg) i.p. before and after disease induction. The treatment of WT mice with β-catenin agonist significantly delayed EAE onset and markedly reduced EAE severity compared with control group mice (Fig. 7B). Consistent with disease severity score, histopathological analysis of the CNS showed reduced inflammation as marked by reduced infiltration of leukocytes and diminished demyelination in the β-catenin agonist-treated group compared with control vehicle-treated mice (Fig. 7C). Further characterization of CNS-infiltrating leukocytes showed a significant reduction in total number of CD4⁺, CD8⁺ T cells, MOG_35–55-specific CD4⁺ T cells, Th1, Th17, and TNF-α⁺ CD4⁺ T cells (Fig. 7D–G, Supplemental Fig. 3A). Similarly, there was a significant reduction in IFN-γ⁺ and TNF-α⁺ CD8⁺ T cells in the CNS of β-catenin agonist-treated mice compared with control mice (Fig. 7H, Supplemental Fig. 3B). In contrast, we observed a significant increase in frequency of Tr1 cells with no change in Foxp3⁺ Treg cells in β-catenin agonist-treated animals compared with control mice (Fig. 7I, Supplemental Fig. 3C). To confirm further the role of β-catenin activation during clinical stages of EAE, β-catenin agonist treatment was initiated from day 10 post-EAE induction when EAE clinical signs became evident, followed by treatment on days 13 and 16. This therapeutic activation of β-catenin also reduced the clinical severity of EAE and CNS pathology (Supplemental Fig. 3D, 3E), further confirming the role of β-catenin pathway activation during late stages of EAE pathology.

Because β-catenin is downstream of LRP5/6 signaling, we next investigated whether activation of β-catenin in LRP5/6^ΔDC mice can rescue EAE disease severity. To test this, we treated LRP5/6^ΔDC mice with β-catenin agonist following EAE induction. Accordingly, β-catenin agonist-treated LRP5/6^ΔDC mice showed significantly reduced EAE disease severity compared with untreated LRP5/6^ΔDC mice (Supplemental Fig. 4A). Also, we noted significantly diminished frequencies of MOG_35–55-specific CD4⁺ T cells as well as Th1, Th17, and TNF-α⁺ CD4⁺ T cells in β-catenin agonist-treated LRP5/6^ΔDC mice compared with untreated mice (Supplemental Fig. 4B–D). Conversely, there was an increase in the frequency of Tr1 cells in β-catenin agonist-treated LRP5/6^ΔDC mice compared with untreated mice (Supplemental Fig. 4E). Thus, these results indicate that activation of the β-catenin pathway independent of LRP5/6-mediated signaling is equally effective in suppressing autoimmune pathology. Taken together, our overall data demonstrate that LRP5/6-mediated activation of the canonical wnt/β-catenin pathway in DCs during EAE limits CNS pathology via suppression of proinflammatory responses and promotion of anti-inflammatory responses.

In the current study, we show that increased wnt ligand expression during the induction and effector phase of EAE regulates neuropathology through activation of LRP5/6-mediated canonical β-catenin signaling in DCs. Accordingly, DC-specific deletion of LRP5/6 or β-catenin led to an increased expression of IL-6, TNF-α, IL-1β, IL-12p40, and IL-12p70 with diminished production of IL-10 and TGF-β. Consequently, the absence of LRP5/6-β-catenin–mediated signaling in DCs suppressed regulatory cell responses, yet promoted Th1 and Th17 cell differentiation. Accordingly, mice lacking LRP5/6 or β-catenin in DCs exhibited severe EAE pathology characterized by intense demyelination in CNS with increased effector T cell infiltration in the brain. Finally, DC-specific expression of active β-catenin or pharmacological induction of β-catenin signaling during ongoing neuroinflammation delayed EAE onset with diminished neuropathology. These data indicate an important role for the canonical wnt/β-catenin pathway in controlling CNS inflammation during EAE and represent an important target to control chronic inflammatory conditions such as MS.

In models of gut tolerance and tumor-induced immune tolerance, our recent studies have shown that activation of β-catenin in DCs induces Treg responses and suppresses intestinal inflammation and antitumor immunity (19, 31). However, whether this pathway in DCs regulates peripheral immune homeostasis and neuroinflammation is not known. The data presented in this study clearly demonstrate that wnt-ligand–mediated activation of the β-catenin pathway in DC is critical for limiting inflammation and the magnitude of disease severity. Our data also suggest that either deletion of LRP5/6 or blocking LRP5/6 signaling in DCs programmed them to a potent inflammatory state inducing Th1/Th17 responses that exacerbated EAE. This was due to high levels of proinflammatory cytokines secreted by DCs that promoted Th1/Th17 cell differentiation and expansion. In addition, T cells isolated from the CNS of LRP5/6^ΔDC mice showed increased expression of GM-CSF, a key effector cytokine involved in CNS pathology during EAE (32–35). These data suggest that wnt-mediated canonical signaling in DCs is critical in regulating both differentiation and effector function of T cells.

Emerging evidence suggests that DCs are critical in resolving inflammation and limiting immune-mediated pathology in several autoimmune disease settings. However, the immune modulatory factors in the tissue microenvironment that are critical in programming DCs to containing inflammation are not known. In this context, our data suggest that the central role of increased wnt ligand expression in the brain is as a feedback mechanism to regulate CNS pathology through activation of the β-catenin pathway in DCs. Accordingly, we observed increased wnt3a and wnt5a expression in the DLN during disease induction and in the CNS during EAE effector phase. Conditioning of DCs with wnt3a or wnt5a programs DCs to a regulatory state and limits the expression of proinflammatory cytokines in response to different TLR ligands (29, 30). Consistent with these studies (29, 30), we observed that wnt3a conditioning of DCs limits the expression of proinflammatory cytokines and increases the expression of anti-inflammatory factors in response to M. tuberculosis. Interestingly, dysregulated wnt signaling has been linked to neurodegenerative and inflammatory disorders (36). Sustained activation of the wnt pathway restrains inflammation, incites neuroprotection, and promotes neurogeneration (36). In addition to DCs, macrophages and microglial cells play an important role in EAE pathogenesis (37, 38). Wnt5a, which signals independently of the β-catenin pathway, also reprograms DCs to limit the expression of inflammatory cytokines (29, 30). It is possible that both the canonical and noncanonical pathway might act in concert to regulate the level of inflammation. In addition to canonical wnt signaling, TLR2- and E-cadherin–mediated signaling can also activate β-catenin in DCs suppressing chronic inflammation and EAE disease severity (20, 39). Although our data indicated the role for wnt3a in modulating DC activity during EAE, it is quite possible that increased wnt ligand expression could also regulate EAE pathology by modulating the functions of macrophage and microglial cells. Furthermore, it will be interesting to investigate the cellular source or factors inducing wnt ligand expression during ongoing neuroinflammation.

DC-derived signals and the types of cytokines produced by DCs dictate the type of Th cell differentiation and adaptive immune responses. In general, activation of the β-catenin pathway in DCs promotes the expression of immune regulatory molecules such as IL-10, TGF-β, and retinoic acid (19, 20, 29–31, 39). There is accumulating evidence suggesting that these factors regulate the immune response by acting directly on DCs and limiting the expression of proinflammatory cytokines under different autoimmune settings (7, 40). For example, increased IL-10 production downstream of β-catenin activation in DCs could further induce suppressor of cytokine signaling (SOCS)3 expression in DCs and regulate proinflammatory cytokines. Accordingly, our previous study has shown that TLR2/IL-10/RA–mediated SOCS3 induction in DCs regulates IL-6, IL-12, and IL-23 through suppression of p38 (27). Moreover, p38 expression in DCs is critical for Th17 differentiation and plays an important role in EAE pathology (41). Thus, activation of β-catenin in DCs could regulate EAE pathology through IL-10/SOCS3–mediated suppression of p38 in DCs. Consistent with these studies, we show that activation of the β-catenin pathway using wnt3a ligand or β-catenin agonist results in increased expression of IL-10 and TGF-β. In contrast, blockade of wnt signaling or deletion of β-catenin in DCs results in increased expression of proinflammatory cytokines and enhanced Th1/Th17 responses. Similarly, recent studies have shown the immunoregulatory role for DC-specific IL-27 signaling in limiting EAE pathology. TLR-mediated activation of DCs induces IL-27 production, which limits Th1 and Th17 cell differentiation through the induction of immunoregulatory molecule CD39 (42–46). Accordingly, our data showed reduced production of IL-27p28 by DCs, increased Th1 and Th17 cell differentiation, and severe EAE pathology in the absence of LRP5/6-β-catenin signaling in DCs (Figs. 1–3, 5). These data indicate that LRP5/6-β-catenin–mediated signaling in DCs could limit EAE pathology through induction of IL-27. However, further studies are warranted to elucidate mechanistic details of this pathway.

What evolutionary benefits might accrue to the host from activating the canonical wnt-β-catenin pathway during ongoing inflammation? Accumulating evidence suggests that high levels of wnt expression occur in several inflammatory diseases such as arthritis (13), arthrosclerosis (14), psoriasis (15), inflammatory bowel disease (16), and neurodegenerative (17) and neuroinflammatory diseases (36). Although moderate inflammation is essential for normal immune responses, uncontrolled chronic and excessive inflammation leads to allergic and autoimmune diseases. Generally, wnt signaling represents a molecular switch in APCs to dampen excessive inflammation, thereby conferring host protection from immune-mediated pathology. Although the activation of β-catenin during chronic inflammatory conditions is an excellent strategy to limit excessive inflammation (19, 20), tumor cells (31, 47) and persistent pathogens possibly exploit the same molecular mechanisms to suppress effective immune responses. Thus, wnt-mediated activation of LRP5/6-β-catenin signaling in DCs limits Th1, Th17, and CD8⁺ T cell responses and promotes Treg responses. Finally, our findings suggest that fine-tuning of the wnt/β-catenin pathway in DCs might be an attractive approach to limit neuroinflammation in MS patients.

We thank Dr. Brat O. Williams (Van Andel Research Institute) for providing LRP5- and LRP6-floxed mice, Jeanene Pihkala and William King for technical help with FACS sorting and analysis, Janice Randall for expert technical assistance with mice used in this study, the Georgia Regents University Histopathology core facility for CNS histopathology studies, as well as our colleagues in the Georgia Regents University Cancer Immunology, Inflammation, and Tolerance program for constructive comments on various aspects of this study.

The authors have no financial conflicts of interest.