Novel function of Extracellular matrix protein 1 in suppressing Th17 cell development in experimental autoimmune encephalomyelitis

Abstract

Multiple sclerosis (MS) is a chronic inflammatory disease of the CNS characterized by demyelination and axonal damage. Experimental autoimmune encephalomyelitis (EAE) is a well-established animal model for human MS. While Th17 cells are important for the disease induction, Th2 cells are inhibitory in this process. Here, we report the effect of a Th2 cell product, extracellular matrix protein 1 (ECM1), on the differentiation of Th17 cells and the development of experimental autoimmune encephalomyelitis (EAE). Our results demonstrated that ECM1 administration from day 1 to day 7 following the EAE induction could ameliorate the Th17 cell responses and EAE development in vivo. Further mechanism study revealed that ECM1 could interact with αv integrin on DC cells and block the αv integrin-mediated activation of latent TGF-β, resulting in an inhibition of Th17 differentiation at early stage of EAE induction. Furthermore, overexpression of ECM1 in vivo significantly inhibited Th17 cell response and EAE induction in ECM1 transgenic mouse. Overall, our work has identified a novel function of ECM1 in inhibiting Th17 differentiation in the EAE model, suggesting that ECM1 may have a potential to be used in clinical applications for understanding the pathogenesis of MS and its diagnosis.

Introduction

Multiple sclerosis (MS), also known as disseminated sclerosis or encephalomyelitis, is a chronic inflammatory demyelinating autoimmune disease of the central nervous system (CNS) (1–3) in which the insulating sheaths of neurons in the brain and spinal cord are damaged. Experimental autoimmune encephalomyelitis (EAE) is a widely used animal model for human MS. EAE is a Th1/Th17 cell-mediated autoimmune demyelination of the CNS. A novel strategy to control Th1/Th17 cells development in EAE could be beneficial for treatment of MS (4).

Previous studies have demonstrated that naive CD4⁺ T cells differentiate into at least four major lineages: Th1, Th2, Th17 and Treg cells (Tregs). Cytokine networks and cross-regulation among Th cell subsets as well as transcription factor regulation in each Th subset are critical for determining CD4⁺ T-cell fates and effector cytokine production (5). As demonstrated in previous studies on the differentiation of naive CD4⁺ T cells, Th cell differentiation involves a positive feedback loop mediated by the effector cytokines that they produce (6). The process of Th differentiation is also actively controlled by cross-inhibition of cytokines from other Th cell lineages. Reciprocal suppression between the IFN-γ and interleukin (IL)-4 signaling pathways in Th1 and Th2 cells (7, 8), suppression of both Th1 and Th2 differentiation by TGF-β (9) and inhibition of Th17 cell differentiation by both IL-4 and IFN-γ (10, 11) have been documented.

IL-17-producing T helper (Th17) cells have been shown to play a critical role in several mouse autoimmune disease models and are thought to be similarly involved in human disease (12, 13). Th17 cell generation requires exposure of naive T cells to the cytokine TGF-β in combination with pro-inflammatory cytokines (14), although in some contexts Th17 cell differentiation does appear to occur in the absence of TGF-β (15). Despite this potential TGF-β-independent Th17 cell development, the important function of TGF-β in Th17 cell differentiation has been observed in several in vivo mouse models (16, 17). Additionally, TGF-β promotes peripherally induced Treg (pTreg) differentiation (18–20). TGF-β is synthesized as an inactive latent precursor that requires cleavage and/or dissociation from the latency-associated peptide (LAP) to engage the TGF-β receptor complex (21–24). Integrins, a family of heterodimeric cell surface receptors consisting of α and β subunit, are constitutively expressed on dendritic cells, leukocytes and many other cell types. To date, 24 total integrin subunits (18 α and 6 β) have been identified (25). Among the integrins, five share the αν subunit (αvβ1, αvβ3, αvβ5, αvβ6, and αvβ8) and are capable of binding the RGD tripeptide sequence on the LAP of TGF-β (26). The αv integrins are important physiological regulators of TGF-β activation and hence determinants of Th17 cell differentiation (27–29). The deletion of αv integrins or the disruption of the αv-binding site in TGF-β causes the failure of Th17 differentiation and EAE induction (29, 30).

ECM1, an 85-kDa glycoprotein, is involved in skin physiology (31–33), angiogenesis (34), tumor progression and malignancies (35). But very few reports about the role of ECM1 on regulation of the immune response exist. We have previously demonstrated that ECM1 is specifically secreted by Th2 cells and promotes their egress from draining lymph nodes in an animal model of asthma (36).

IL-4 drives Th2 cell differentiation, but it suppresses other Th cell lineages (37, 38). Since ECM1 is a secreted protein by Th2 cells and natural modulator(s) from host cells in regulating Th17 cells remain to be identified, we hypothesize that ECM1 could be such a candidate modulator that has a suppressive effect on Th17 differentiation and EAE development. Thus, we investigated the regulatory role of ECM1 on Th1 and Th17 cell development in the EAE mouse model. The results demonstrate that ECM1 significantly ameliorated EAE development. Mechanistic studies revealed that ECM1 interacts with αv integrin and blocks the αv integrin mediated activation of latent TGF-β, which is critical for activating latent TGF-β, and consequently reduces Th17 cell differentiation. We further confirmed its effect in ECM1 transgenic mice. Taken together, our study identifies a novel function of ECM1 in inhibiting pathogenic Th17 cell development during EAE induction.

Materials and Methods

Animals

C57BL/6 mice were purchased from Model Animal Research Center of Nanjing University (Nanjing, China). ECM1-Tg mice were obtained from Shanghai Research Center for Model Organization (Shanghai, China). Mice expressing a transgenic TCR specific for MOG_35–55 (2D2 mice) were developed in the laboratory of V.K. Kuchroo, Harvard University. Mice were maintained under specific pathogen-free conditions at the Animal Care Facility of the Chinese Academy of Sciences (Shanghai, China). Animal care and use were in compliance with guidelines of the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences.

To generate T cell-specific ECM1 transgenic mice, cDNA encoding mouse ECM1 was cloned into the plasmid Va-hCD2 (39). ECM1 Transgene-positive mice were determined by PCR and Western Blot. The sequences are as follows: sense primer ACCACATGGCTGAGTTCG on the hCD2 promoter, anti-sense primer AAGGCTGCTCTGGATACG on the ECM1 gene. Mice were initially created on the C57BL/6 background.

Cell line and cell culture

The human LAC cell line, A549, was purchased from ATCC and cultured in RPMI-1640 medium, supplemented with 10% FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin at 37°C in a humidified atmosphere containing 5% CO₂. The HEK293T cells were cultured in DMEM medium supplemented with 10% FBS, 100 U/ml penicillin, and 100 µg/ml streptomycin at 37°C in a humidified atmosphere containing 5% CO₂.

EAE induction and assessment

For C57BL/6 mice, 6- to 8-week-old mice were immunized subcutaneously (s.c.) with 200 µg of MOG_35–55 emulsified in CFA (supplemented with 5 mg/mL of heat-inactived Mycobacterium tuberculosis) and injected intraperitoneally (i.p.) with 300 ng pertussis toxin (Sigma) on day 0 and day 2. ECM1 or control protein (Human IgG) was intravenously (i.v.) injected into mice on days 1, 3, 5 and 7 post-immunization. Assessment of classical EAE was performed as follows: 0, no disease; 1, decreased tail tone; 2, hind limb weakness or partial paralysis; 3, complete hind limb paralysis; 4, front and hind limb paralysis; and 5, moribund state. For histopathological studies, spinal cords were immersion fixed in 4% paraformaldehyde and paraffin embedded. Sections were stained using H&E and Luxol fast blue, and stained sections were evaluated for immune cell infiltration and demyelination. During histological evaluation, we assessed the inflammation (inflammatory index) as described previously (40). The principle of the inflammatory index as follows: 0, no inflammation in the CNS; 1, a few infiltrated inflammatory cells in the perivascular areas and meninges; 2, mild cellular infiltration in the parenchyma; 3, moderate cellular infiltration in the parenchyma; 4, severe cellular infiltration in parenchyma. CNS-infiltrating mononuclear cells were prepared by Percoll gradient separation.

Immunohistochemistry

Differentially treated EAE mice were anesthetized with chloral hydrate (3 µg/g via intraperitoneal injection) and the spinal cords were removed and fixed in 4% paraformaldehyde and paraffin embedded. Sections were stained to detect the myelin, oligodendrocytes, and inflammatory cells. The following primary antibodies were utilized in immunolabeling: Rabbit anti-mouse MBP antibody (ab40390, Abcam), Rabbit anti-mouse Oligodendrocyte Specific Protein antibody (ab53041, Abcam), and rabbit anti-mouse CD3 antibody (ab5690, Abcam). The secondary antibody is Cy3-AffiniPure Goat Anti-Rabbit IgG (JAC-111165045, ELITechGroup). Sections were examined under an Olympus BX51 microscope (Olympus).

Preparation of mass spectrometry samples

Splenic cells obtain from C57BL/6 mice with the methods previously described (36), and the cells were lysed with Cell lysis buffer (Life technology), sonicated briefly on ice (5 × 10 s at full power), and cleaned extracts by centrifugation at 2,800×g (GH3.8 rotor; Beckman Coulter GS-6) for 10 min at 4 °C. Subsequently, the cell lysates were incubated with the ECM1 antibody and protein A/G Plus-agarose immunoprecipitation beads (Santa Cruz Biotechnology) at 4°C for 3 h or overnight. After three washes, proteins were separated by one-dimensional SDS-polyacrylamide gel electrophoresis on a Bio-Rad Mini-Protean II system using 1-mm-thick 10% polyacrylamide gels. Aliquots of protein stock solutions prepared in 1% formic acid were diluted in sample buffer to a final concentration of ~1.0 or 3.0 µM, and 5 µL of this solution was loaded onto a gel. After electrophoresis, proteins were visualized by Coomassie Brilliant Blue R250 staining (Serva Electrophoresis GmbH, Heidelberg, Germany). Protein bands were excised, cut into 1 mm³ cubes, put into 0.65-mL PCR microtubes (Roth, Karlsruhe, Germany), and in-gel digested using modified trypsin as described previously (41). And the proteins binding to ECM1 antibody and beads were harvested and analyzed by mass spectrometry.

Expression of recombinant ECM1 protein

The Bac-to-Bac Baculovirus Expression System (Invitrogen) was used for recombinant ECM1 production. ECM1 cDNA fused to the human-Fc sequence at the C-terminus was cloned into the pFastBac vector (Invitrogen). In our following experiments, we used the human IgG protein as the control group. The resulting plasmid was then used to generate recombinant baculoviruses that were in turn used to infect High-Five insect cells, which were grown at 27°C in suspension culture in SF-900II medium (Invitrogen). High-Five cell cultures were infected at a density of ~2 × 10⁶ cells/ml and used for experiments after 72 h of infection. After 72 h of infection at 27°C, medium containing secreted ECM1 was centrifuged (500×g for 10 min) and frozen at −80°C. Conditioned medium supplemented with protease inhibitors was then centrifuged (12,000 rpm for 60 min) to remove cellular debris and applied to an anti-human Fc Affinity Gel column (Orgma) equilibrated with 20 mM sodium phosphate buffer, pH 7.0. The bound proteins were eluted with 0.1 M glycine solution, pH 2.8. All purification steps were performed at 4°C.

Cell purification and T cell differentiation in vitro

CD4⁺ T cells were purified by a CD4⁺ Naive T cell Negative Isolation Kit (Stem Cell). CD11c⁺ DCs were purified by a CD11c Positive Isolation Kit (Miltenyi Biotec) (>95% purity). The method for generating bone-marrow dendritic cells (BMDCs) has been described previously (42). For in vitro differentiation, CD4⁺ Naive T cells were maintained in RPMI 1640 supplemented with 10% fetal calf serum (GIBCO) and stimulated with 5 µg/ml plate-bound anti-CD3 and 1 µg/ml soluble anti-CD28 under the appropriate conditions to obtain either Th0 (50 U/ml IL-2, 10 µg/ml anti-IL-12 and 10 µg/ml anti-IL-4), Th1 (10 ng/ml IL-12, 10 µg/ml anti-IL-4 and 50 U/ml IL-2) or Th17 (3 ng/ml TGF-β1, 20 ng/ml IL-6, 10 ng/ml IL-1β, 10 ng/ml IL-23, 10 µg/ml anti-IFN-γ, and 10 µg/ml anti-IL-4) cells in the presence or absence of ECM1 protein or human IgG control.

Co-culture of CD4⁺ T cells with DCs

Splenic DCs were sorted to more than 95% purity using the CD11c Positive Isolation Kit (Miltenyi Biotec). Naive 2D2 T cells were isolated from the lymph nodes and spleens of 2D2 TCR transgenic mice using the CD4⁺CD62L⁺ T-cell isolation kit (Miltenyi Biotec). Co-culture of CD4⁺ T cells with splenic DCs has been described previously (29, 30). Briefly, purified naive 2D2 CD4⁺ T cells were cultured with splenic DCs at a ratio of 5:1 in the presence of the MOG_35–55 peptide (10 µg/ml), IL-6 (50 ng/ml; Peprotech), IL-1β (10 ng/ml; Peprotech), FICZ (6-formylindolo [3,2-b] carbazole; 300 nM; Biomol) and TGF-β (1 ng/ml; R&D Systems) or latent-TGF-β (50 ng/ml). For RGD blockade experiments, cRGD peptide was added at 4 µg/ml. For the co-culture assay, co-cultured cells were maintained in X-VIVO™ 15 Chemically Defined, Serum-free Hematopoietic Cell Medium (Lonza). The 2D2 T cells were analyzed by flow cytometry after 3 days of culture. Cytokines in the culture supernatants were measured using an IL-17A ELISA kit (R&D, USA). The activation of TGF-β was measured with the Plasminogen Activator Inhibitor-1 Promoter Luciferase Assay (details described below).

Luciferase assays

Luciferase reporter genes containing the human plasminogen activator inhibitor-1 (PAI-1) promoter (−740 to +44) were prepared by PCR of genomic DNA. Subsequently, the PCR product was subcloned into the pGL3-Basic luciferase vector (Promega). The luciferase assays were performed as described previously (43). Briefly, HEK293T cells were seeded into 24-well culture plates and transfected with the PAI-1 reporter plasmid (Luciferase, 1 µg/well) and pRL-TK-Luciferase (1 µg/well) using Lipofectamine reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. At 24-h post-transfection, the cells were lysed in passive lysis buffer (Promega). The luciferase activity of the lysates was analyzed using a dual luciferase reporter assay system (Promega).

Intracellular cytokine staining

For analyses of intracellular IFN-γ, IL-17, IL-4, and Foxp3, T cells were stimulated with PMA (50 ng/ml) and ionomycin (1 µM) for 6 h with the addition of Brefeldin A (10 µg/ml; Sigma-Aldrich) for the last 4 h of stimulation. Cells were harvested, washed, fixed, permeabilized (CALTAG FIX AND PERM), and stained with the indicated fluorescein-labeled antibodies according to the manufacturer’s instructions. The follow antibodies were used: FITC-labeled anti-CD4 antibody (L3T4, eBioscience), APC-labeled anti-IFN-γ antibody (XMG1.2, eBioscience), Percp-labeled anti-IL-17 antibody (eBio17B7, eBioscience), PE-labeled anti-IL-4 antibody (444389, BD Pharmingen), APC-labeled IL-4 antibody (11B11, eBioscience), and PE-labeled anti-Foxp3 antibody (NRRF30, eBioscience). Appropriate fluorescein-conjugated, isotype-matched, irrelevant mAbs were used as negative controls. Cells were analyzed on a FACSCalibur cytometer (BD Biosciences).

Proliferation analysis and ELISA

Purified CD4⁺ T cells were stimulated with plate-bound anti-CD3, and the indicated proteins were added 3 days prior to 12-h pulse with [³H]-thymidine. For recall experiments, splenocytes isolated from EAE mice were stimulated with MOG_35–55 (50 µg prior to 12-h pulse with [³H]-thymidine). Cells were collected using a cell harvester, and [³H]-thymidine was quantified by scintillation counting. The proliferation of CD4⁺ T cells isolated from draining lymph nodes and CNS of the differentially treated EAE mice was measured with CSFE staining and Cell Titer-Glo Luminescent Cell Viability Assay (Promega). Briefly, the purified CD4⁺ T cells from draining lymph nodes and CNS were stained (or not) and stimulated with with plate-bound anti-CD3. Then cells were collected and measured with Flow Cytometry (or BioTek SynergyNEO). To measure cytokines production in the supernatant, IL-17A and IFN-γ ELISA Duoset kits were purchased from R&D Systems and used according to the manufacturer's protocol.

RT-PCR analysis

Total RNA was extracted from cells using TRIzol reagent (Invitrogen; Carlsbad, CA) according to the manufacturer’s protocol. RT-PCR kits (TaKaRa; Kyoto, Japan) were used for RT-PCR experiments. Four micrograms of total RNA template were used to make cDNA using the Prime Script RT Master Mix kit (TaKaRa; Kyoto, Japan). Synthesized cDNA was used in qPCR experiments. Quantitative real-time PCR (qRT-PCR) was performed using a 40-cycle two-step PCR with sequence-specific primer pairs using the ABI PRISM 7900HT Sequence Detection System (PE Applied Biosystems). Data were analyzed using SDS software. Primers were designed using Primer Express 3.0 software. The mRNA expression was evaluated as a ratio based on qRT-PCR results for cells HPRT mRNA. Sequences of primers used were: HPRT forward (fwd), 5’- TGC TCG AGA TGT CAT GAA GGA G-3’ and HPRT reverse (rev), 5’- CAG AGG GCC ACA ATG TG-3’; IL-1β fwd, 5’- GAC CTG GGC TGT CCT GAT GA-3’ and IL-1β rev, 5’- GTG CTG CTG CGA GAT TTG AA-3’; IL-23p19 fwd, 5’- TCC GTT CCA AGA TCC TTC GA-3’ and IL-23p19 rev, 5’-GGG CAG CTA TGG CCA AAA A-3’; IL-6 fwd, 5’-TTC CAT CCA GTT GCC TTC TTG-3’ and IL-6 rev, 5’-GAA GGC CGT GGT TGT CAC C-3’.

RNA-mediated interference

The αv integrin siRNA and negative control were synthesized from GenePharma (GenePharm, Shanghai, China). The siRNA sequences as follows: sense: 5’-GAC CCG UUG UCA CUG UAA ATT-3’, anti-sense: 5’-UUU ACA GUG ACA ACG GGU CTT-3’. Bone-Marrow derived Dendritic Cells (BMDCs) were transfected with siRNA delivered by Lipofectamine 2000 (Life Technologies). At 24 h after transfection, the cells were used for DC and T cell co-culture experiments.

Immunoprecipitation and immunoblot analysis

Immunoprecipitation and immunoblot analysis were performed as described previously (36). Briefly, HEK293T cells were transfected with various combinations of plasmids. Twenty-four hours after transfection, the cell lysates were prepared in lysis buffer and incubated with the indicated antibodies and protein A/G Plus-agarose immunoprecipitation reagent (Santa Cruz Biotechnology) at 4°C for 3 h or overnight. After three washes, the immunoprecipitates were boiled in SDS sample buffer for 10 min and analyzed by western blot. For immunoprecipitation analysis, protease inhibitor cocktail was obtained from Roche, and anti-hemagglutinin antibody (HA.11; 16B12; CO-MMS-101R) was obtained from Covance. For western blot analysis, the rabbit anti-ECM1 polyclonal antibody used in our experiments was raised against bacterially produced recombinant mouse ECM1 protein fragments encompassing the N- and C-terminal peptides. The anti-actin antibody was purchased from Sigma. The mouse anti-CD51 (αv Integrin) antibody was purchased from BD Transduction Laboratories.

Adoptive transfer experiments

C57BL/6 mice were immunized with subcutaneous injection of 200 µl of phosphate buffered saline (PBS) containing 300 µg of MOG_35–55 emulsified in complete Freund’s Adjuvant (CFA) containing 6.5 mg/ml Mycobacterium tuberculosis H37 RA (Difco Laboratories, Detroit, MI, USA). At day 0 and 2 after immunization, the mice received 300 ng of pertussis toxin intraperitoneally (Sigma, Chemical). Then, at day 1, 3, 5, and 7 after EAE immunization, the mice were treated with IgG or ECM1 protein via tail vein injection. Draining LN cells were collected 10 days after EAE immunization, and single-cell suspensions were prepared and re-stimulated in 24-well plates at 6 × 10⁶ cells/well in a total volume of 2 ml of RPIM1640 medium in vitro with MOG_35–55 (50 µg/ml) and IL-23 (20 ng/ml) for 4 days to generate encephalitogenic T cells. The encephalitogenic T cells were then injected i.v. into sublethally irradiated (450 rads) C57BL/6 mice (2×10⁷ cells per mouse) within 24 h. And each mouse received 200 ng of pertussis toxin via i.p. injection on the same day and 2 days after adoptive transfer. The mice were monitored daily for clinical signs of disease and assigned disease scores of 0-5 based on the severity of EAE.

Statistical analysis

Quantitative measures were summarized with descriptive statistics (mean ± SEM). Statistical significance was determined using unpaired Student’s t test or ANOVA. Pearson’s correlation coefficients were performed to assess the relationship between the ECM1 expression level in MS patients and IL-17, TGF-β, etc. P-values were two-tailed and a significant level of 0.05 was used. Statistical analysis was conducted with Prism version 5 (GraphPad Software).

Results

ECM1 treatment ameliorates EAE and suppresses Th17 cell differentiation in vivo

Because ECM1 is secreted by Th2 cells, we asked whether recombinant ECM1 protein could regulate Th1/Th17 cell development in an EAE animal model. We produced mouse recombinant ECM1 protein using the Baculovirus Expression System (Invitrogen, Life Technology) and treated mice intravenous (i.v.) injection with recombinant ECM1 protein on day 1, 3, 5, and 7 after the initiation of EAE induction. ECM1 treatment significantly reduced the clinical severity (Fig. 1A) and the incidence of disease (Fig. 1B) relative to human IgG control treatment. Moreover, the mean maximum EAE score (Fig. 1C) and the cumulative EAE scores (Fig. 1D) were also lower in ECM1-treated mice compared to the human IgG-treated control group. These results demonstrate that ECM1 treatment attenuates the severity of EAE.

Figure 1. ECM1 protein treatment during early EAE induction ameliorates EAE.

C57BL/6 mice were immunized with MOG_35–55, and mice were treated i.v. with recombinant ECM1 protein or human IgG protein on day 1, 3, 5, and 7 after EAE induction. (A): EAE clinical scores of ECM1 protein-treated mice (n=10) and human IgG control-treated mice (n=10). (B): Incidence of EAE clinical signs in MOG_35–55 induced EAE mice treated with control human IgG and recombinant ECM1 protein. (C and D): The mean ± SEM of the maximum (C) and cumulative (D) EAE scores. (E and F): Luxol fast blue (E) and H&E (F) staining of paraffin sections of spinal cords isolated from representative of human IgG control-treated and recombinant ECM1 protein-treated mice on day 21 after immunization. IL-17 (G) and IFN-γ (H) production by splenocytes were determined by ELISA. The splenocytes were isolated from differentially treated mice and restimulated with MOG_35–55 (50 µg/ml) for 72 h in the presence of IL-23. (I): Splenocytes were pulsed with [³H]-thymidine in the last 12 h for proliferation analysis. Data are representative of three independent experiments. Mean ± SEM is shown. *, P<0.05; **, P<0.01; ***, P<0.001; NS, not significant.

Histopathology results revealed a clear reduction in demyelination (Fig. 1E) and inflammation (Fig. 1F) in the affected spinal cord of ECM1 protein-treated EAE mice compared with the human IgG-treated control EAE mice (Supplementary Fig. 1A). Meanwhile, the accumulation of inflammatory cells in the spinal cord (Fig. 1F) of EAE mice was decreased by ECM1 treatment (Supplementary Fig. 1A). Since naïve CD4⁺ T cells are activated by antigen and dendritic cells and differentiate into distinct effector Th cell subsets, depending on their cytokine microenvironment during the first week of activation after EAE induction, we reasoned that ECM1 protein inhibits EAE is possibly by affecting differentiation of Th1/Th17 cells in vivo.

To test whether administration of recombinant ECM1 protein on day 1, 3, 5, and 7 after EAE induction affects the Th1/Th17 lineage commitment, we immunized C57BL/6 mice with MOG_35–55 peptide and treated mice i.v. with ECM1 protein on days 1, 3, 5, and 7 post-EAE immunization. MOG_35–55-primed T cells were isolated from the draining LN cells on day 8 post-EAE induction. The MOG_35–55-primed T cells were re-stimulated ex vivo with MOG_35–55 peptide and IL-23 for 72 h and then IL-17 was measured by ELISA. Interestingly, ex vivo re-stimulated lymphocytes from mice treated with ECM1 early during the induction period produced less IL-17 than did lymphocytes from human IgG-treated control EAE mice (P<0.001) (Fig. 1G). However, IFN-γ production (Fig. 1H) and cell proliferation (Fig. 1I) were unchanged following MOG_35–55 peptide re-stimulation.

To examine whether the ECM1 effect occurs at the transcriptional level, CD4⁺ T cells were purified by MACS from spleens on day 8 following EAE induction. After re-stimulation with MOG_35–55 and IL-23 in vitro for 72 h, real-time PCR analyses of various master transcription factors and classic cytokines associated with each Th subset were performed. The data revealed that ECM1 treatment lead to decreased transcription of Rorc (P<0.001) (Fig. 2A), which was consistent with the dramatically decreased expression of Il17a (P<0.001) (Fig. 2B). Interestingly, ECM1 treatment resulted in only a slight increase in the transcription of Foxp3 and Il10. However, we did not observe a significant change in the expression of Tbx21, Gata3, Ifng and Il4 between the ECM1-treated and untreated groups. These observations suggest that ECM1 specifically inhibits the generation of Th17 cells in vivo.

Figure 2. ECM1 protein attenuates EAE by inhibiting Th17 generation.

C57BL/6 mice were immunized with MOG_35–55 and treated with recombinant ECM1 protein or human IgG on days 1, 3, 5, and 7. Draining lymph node cells were isolated on day 8 and re-stimulated by MOG_35–55 (50 µg/ml) and IL-23(10 ng/ml) in vitro for 72 h to generate encephalitogenic T cells. The encephalitogenic T cells were then injected i.v. into sublethally irradiated (450 rads) C57BL/6 mice (2×10⁷ cells per mouse) within 24 h. (A and B): The re-stimulated CD4⁺ T cells were isolated with MACS column from differentially treated EAE mice, then the transcription of T-bet, GATA-3, ROR-γt and Foxp3 (A) and the cytokines (B) IL-4, IFN-γ, IL-17 and IL-10 were evaluated using real-time PCR analysis. (C): EAE clinical scores of the different groups of mice (n=10 per group) were evaluated. (D): Incidence of EAE clinical signs. (E and F): The mean± SEM of the maximum (E) and cumulative (F) EAE scores. (G and H): Luxol fast blue (G) and H&E (H) staining of paraffin sections of spinal cords isolated from the indicated groups of mice. (I): Histological scores (inflammatory index) of different groups of mice for inflammatory lesions. Data are representative of three independent experiments. Mean ± SEM is shown. *, P<0.05; **, P<0.01; ***, P<0.001; NS, not significant.

Encephalitogenic T cells, obtained from ECM1-treated donors and re-stimulated with MOG_35–55 peptide and IL-23 for 72 h, were injected intravenously into sublethally irradiated (450 rads) recipient C57BL/6 wild-type mice. Adoptive transfer of ECM1-treated donor T cells into recipient mice had significantly reduced EAE disease scores (Fig. 2C) and the incidence of EAE relative to the IgG-treated control group (Fig. 2D). ECM1 treatment also decreased the mean maximum EAE score (Fig. 2E) and the cumulative EAE scores (Fig. 2F) relative to the IgG control treatment. The histopathology was consistent with these results and revealed significantly decreased inflammation and demyelination in the spinal cord of ECM1-treated EAE mice relative to control EAE mice (Fig. 2G, H, I and Supplementary Fig. 1B). Taken together, these results indicate that recombinant ECM1 protein is capable of suppressing EAE induction by affecting Th17 cell development in vivo.

ECM1 suppresses Th17 response in EAE mice

Based on the above results, we concluded that ECM1 could attenuate MOG induced EAE via inhibiting Th17 cells differentiation in vivo. To further evaluate the effect of ECM1 on the Th17 responses in EAE mice, the different Th cell populations present in periphery and in CNS were analyzed. The apoptotic and proliferated cells were also assessed by flow cytometry. As shown in Fig. 3A, we observed that ECM1 treatment inhibited the proportion of IL-17⁺ Th17 cells in peripheral lymph nodes (Fig. 3A) and in the CNS (Supplementary Fig. 2). On the other hand, the proportion of IFN-γ⁺ Th1 cells and IL-4⁺ Th2 cells were unchanged (Fig. 3A and supplementary Fig. 2). Meanwhile, the proportion of Foxp3⁺ Treg cells in the CNS has a slightly increased compared with those from IgG treated mice (Supplementary Fig. 2). The results of ELISA assay showed that ECM1 treatment decreased the production of IL-17A and slightly increase IL-10 production in peripheral lymph nodes (Fig. 3B). These results were consistent with the data in Fig. 1 and Fig. 2. Additionally, we also measured cell proliferation using CFSE staining and cell apoptosis using Annexin V/PI staining of the CD4⁺ T cells from peripheral draining lymph node and the CNS. The data showed that the cell proliferation and apoptosis had no significant changes between ECM1 and IgG treated EAE mice either in peripheral draining lymph node or CNS (Data not show). Taken together, these results indicate that ECM1 attenuates MOG induced EAE through suppressing Th17 responses in vivo.

Figure 3. FACS analysis of the differentiated Th cell populations in the peripheral lymph nodes.

C57BL/6 mice were immunized with MOG_35–55 (300 µg) in CFA, and mice were treated i.v. with recombinant ECM1 or IgG protein on day 1, 3, 5, and 7 after EAE induction. Cells from the draining lymph nodes were isolated from differentially treated EAE mice and re-stimulated with MOG_35–55 (50 µg/ml) for 72h in the presence of IL-23. (A) CD4⁺IFN-γ⁺Th1, CD4⁺IL-17⁺Th17, CD4⁺IL-4⁺Th2, and CD4⁺Foxp3⁺Treg cells were analysis by FACS. (B) The production of IFN-γ, IL-17, IL-4, and IL-10 were determined by ELISA. Data are representative of three independent experiments. Mean ± SEM is shown. *, P<0.05; **, P<0.01.

ECM1 inhibits Th17 differentiation by blocking TGF-β activation

To evaluate whether ECM1 inhibits Th17 cell development in a T cell-intrinsic fashion, we induced Th17 differentiation in vitro in the presence of TGF-β and IL-6 with recombinant ECM1 protein or human IgG control protein. To our surprise, ECM1 did not affect Th17 or Th1 differentiation (Fig. 4A) or the production of IL-17 or IFN-γ (Fig. 4B) in CD4⁺ T cells cultured in the absence of DCs. These data indicate that the effect of ECM1 on Th17 differentiation is not T cell-intrinsic, i.e., that ECM1 does not directly target differentiating Th17 cells. Next, we asked whether ECM1 performs its inhibitory function on Th17 cells by targeting dendritic cells.

Figure 4. ECM1 does not affect Th1 and Th17 differentiation in vitro but suppresses the activation of TGF-β in DCs.

Purified CD4⁺ T cells were cultured under Th1 or Th17 differentiation conditions for 4 days. (A): Intracellular staining of IL-17 and IFN-γ was assessed by flow cytometry. (B): IL-17 and IFN-γ produced by T cells were measured by ELISA. (C): The expression of IL-6, IL-23p19, IL-1β, and TGF-β in bone marrow-derived DCs cultured in the presence of recombinant ECM1 or human IgG proteins. (D): The activation of TGF-β in the supernatant of BMDCs cultured in the presence of recombinant ECM1 or human IgG proteins. Data are representative of three independent experiments. Mean ± SEM is shown. *, P<0.05; **, P<0.01; ***, P<0.01; NS, not significant.

Th17 differentiation is critically dependent on TGF-β, in combination with IL-6, IL-23, and IL-1β. DCs are believed to be one of the major producers of these cytokines, especially TGF-β, and are essential to the differentiation of Th17 in vivo (44). Thus, we tested whether DC production of these cytokines, including TGF-β, IL-6, IL-23, and IL-1β, is modulated by ECM1. TGF-β is synthesized as an inactive latent precursor (LTGF-β) that forms a complex with latency-associated peptide (LAP) and the αv integrin on the surface of DCs (24). Subsequently, the LAP must be cleaved and/or dissociated from the complex to produce mature, active TGF-β (aTGF-β), which is essential in promoting Th17 cell differentiation (24, 29, 30, 44). Therefore, we constructed a Plasminogen Activator Inhibitor-1 Promoter Luciferase plasmid (43) to detect the activity of TGF-β using a luciferase assay in the presence or absence of ECM1 in DC culture.

To address this possibility, we generated Bone-Marrow-Derived DCs and cultured these cells in the presence of ECM1 protein. We measured TGF-β, IL-6, IL-23p19, and IL-1β mRNA expression using real-time PCR. ECM1 did not affect TGF-β, IL-6, IL-23p19 and IL-1β expression in DCs (Fig. 4C). However, when ECM1 protein and L-TGF-β were added to BMDC culture, active TGF-β in the culture supernatants was significantly reduced (Fig. 4D). Based on these results, we conclude that ECM1 has the potential to inhibit TGF-β activation on DCs.

ECM1 binds to αv integrin and inhibits TGF-β activation mediated by αv integrin in DCs, consequently suppresses Th17 cell differentiation

To elucidate the molecular mechanisms responsible for ECM1-mediated inhibition of TGF-β activation in DCs, we first performed mass spectrometry (MS) to identify the candidate proteins that interact with ECM1. Among these proteins, an interesting candidate is αv integrin as shown in Fig. 5A. Thus, we hypothesized that ECM1 targets the αv integrin on the surface of DCs and inhibits the activation of TGF-β mediated by αv integrin in DCs. To test this hypothesis, we investigated whether ECM1 interacts with αv integrin using co-immunoprecipitation. C-myc-tagged ECM1 and HA-tagged αv integrin, transfected into HEK293T cells, strongly interacted with each other as shown in Fig. 5B and 5C.

Figure 5. ECM1 inhibits Th17 differentiation by blocking αv integrin-mediated TGF-β activation.

(A): Mass spectrometry of the ECM1 interaction with αv integrin. (B and C): ECM1 interacts with αv integrin. HEK293T cells were transfected with the indicated plasmids. After 24 h, the cells were lysed, and proteins were immunoprecipitated using HA antibody and protein A/G agarose beads. The precipitated complexes were resolved by SDS-PAGE and immunoblotted with ECM1 antibody. (D–F): Th17 differentiation was induced in vitro during co-culture of DCs and naive 2D2 CD4⁺ T cells in the presence of IL-6, IL-23, IL-1β, and either IgG, ECM1 or cRGD, then added either active or Latent TGF-β to the culture. 72 h after culture, (D): The intracellular staining of IL-17 of the co-cultured T cells was investigated by flow cytometry. (E): The quantitative analysis of CD4⁺IL-17⁺ T cells in DC-T cell co-culture system. (F): The activity of active TGF-β in the supernatant of the co-culture system. Vec, vector; WCL, whole cell lysate. Data are representative of three independent experiments. Mean ± SEM is shown. *P<0.05.

We then examined whether ECM1 affects latent TGF-β (L-TGF-β) maturation and subsequently inhibits Th17 cell differentiation. Th17 cell differentiation was induced in vitro during co-culture of splenic DCs and naive 2D2 CD4⁺ T cells in the presence of IL-6, IL-23, IL-1β, and either IgG, ECM1 or cRGD, then added either active or L-TGF-β to the cells. Since the peptide cRGD specifically inhibits L-TGF-β activation and inhibits TGF-β activation by blocking the interaction of αv integrins and L-TGF-β (45), we used cRGD as a positive control. As shown in Fig. 5D and 5E, the percentage of Th17 cells was significantly reduced from 13.1% to 9.12% by ECM1 in the presence of L-TGF-β but did not change in the presence of active TGF-β by FACS analysis. As expected, ECM1-mediated inhibition had a similar effect as that of cRGD, which reduced the percentage of Th17 cells from 13.1 to 8.06% (Fig. 5D, E). We also detected the level of active TGF-β in the co-cultured supernatants using PAI-1 luciferase assay, these results (Fig. 5F) were consistent with the data of Th17 cell differentiation (Fig. 5D) and IL-17 secretion (data not show). These results suggest that ECM1 is as effective as cRGD in suppressing TGF-β maturation and thus Th17 cell differentiation.

To confirm the inhibitory effect of ECM1 on Th17 differentiation through blocking αv integrin-mediated TGF-β activation, siRNA was used to silence αv integrin on Bone-Marrow-Derived-DC cells (BMDCs) under the same experimental conditions. In the absence of αv integrin, ECM1 was unable to inhibit Th17 differentiation (Supplementary Fig. 3). Taken together, these data indicate that ECM1 binds to αv integrin, blocking the maturation of latent TGF-β mediated by αv integrin, which is critical for the Th17 cell differentiation.

ECM1 directly binds αv integrins to competitively inhibit the interaction between αv integrin and RGD sequence

To clarify whether the interaction between ECM1 and αv integrin affected the binding of αv integrin to RGD sequence in latent-TGF-β, the competition assays were performed using cRGD. HA-tagged αv integrin (HA-αv Integrin) and c-myc tagged ECM1(c-myc-ECM1) overexpressed cell lysates were incubated with cRGD in different concentrations. Pull-down of αv integrin using protein A/G Plus-agarose beads resulted in coprecipitation of ECM1 in a dose-dependent fashion (Fig. 6A). The results showed that ECM1 could compete with the binding of RGD sequence to αv integrin. Therefore, ECM1 could bind αv integrins as a competitive inhibitor of L-TGF-β.

Figure 6. ECM1 inhibits the interaction of αv integrin and RGD sequence in latent-TGF-β.

(A): HEK293T cells were overexpressed with HA-tagged αv integrin (HA-αv Integrin), and c-myc-tagged ECM1 (c-myc-ECM1) separately, and then the cells were lysed. The cell lysates were mixed with cRGD reagent at indicated concentrations. After 6 h of incubation, protein-A/G Plus-agarose beads were added to the mixture and incubated further. After another 6 h, the beads were washed thoroughly, and the proteins bound to the beads were eluted. The elution was subjected to SDS/PAGE and Western blot analysis with indicated antibodies. (B): The phosphorylation of Smad2 in differentially treated A549 cells. C57BL/6 mice were immunized with MOG_35–55 (300 µg) in CFA, and mice were treated i.v. with recombinant ECM1 or cRGD or human IgG protein on day 1, 3, 5, and 7 after EAE induction. (C): EAE clinical scores from mice treated with recombinant ECM1 protein (n=10), human IgG (n=10) and cRGD (n=10). (D): Luxol fast blue (up panel) and H&E (down panel) staining of paraffin sections of spinal cords isolated from differentially treated groups of EAE mice on day 21 after immunization. Data are representative of three independent experiments. Mean ± SEM is shown. *, P<0.05; **, P<0.01, ***, P<0.001, #, P<0.05; ##, P<0.01; ###, P<0.001; *, Represent ECM1 group vs human IgG group; #, Represent cRGD group vs human IgG group.

Subsequently, we measured the phosphorylation of Smad2, a downstream member of the TGF-β signaling pathway in A549 cells. To examine the level of active TGF-β in the supernatants of DCs and T cells co-culture and the activation of TGF-β pathway, we added the supernatants to the culture of A549 cells and then the phosphorylation of Smad2 was determined using Western Blot assay. We observed that phosphorylated Smad2 levels were reduced in A549 cells in presence of recombinant ECM1 protein compared with cells treated with human IgG protein (Fig. 6B). These results indicate that the ECM1 protein inhibits the activation of TGF-β signaling cascade by targeting αv integrin-mediated TGF-β maturation. Then to confirm whether the inhibitory effect of ECM1 (or cRGD) on the activation of TGF-β signaling cascade could suppress the development of EAE, we treated mice intravenous (i.v.) injection with recombinant ECM1 protein or cRGD or Human IgG on day 1, 3, 5, and 7 after the initiation of EAE induction. As shown in Fig. 6C, we observed that ECM1 or cRGD treatment significantly attenuated the severity of disease relative to human IgG control treatment (Fig. 6C). Furthermore, this observation was also supported by the histopathology results (Fig. 6D).

ECM1 transgenic mice are protected from EAE induction and exhibit impaired IL-17 production

The above data demonstrated that the ECM1 protein suppresses Th17 cell development by inhibiting αv integrin-mediated activation of TGF-β, and the administration of recombinant ECM1 protein to MOG-induced EAE mice protects mice from EAE induction. To further test whether T cell-produced endogenous ECM1 also attenuates MOG-induced EAE, we generated ECM1-transgenic mice that constitutively express ECM1 in T cells under the control of a human CD2 promoter. ECM1 transgene mRNA expression was confirmed by RT-PCR (Fig. 7A), and ECM1 protein levels were measured by western blot in CD4⁺ T cells (Fig. 7B). Lymphocytes from EAE-induced ECM1^Tg mice that were re-stimulated ex vivo with MOG^35–55 peptide produced less IL-17 than those from WT littermates (Fig. 7C). However, no apparent reduction in IFN-γ production was noted (Fig. 7D). When WT and ECM1^Tg mice were immunized with MOG_35–55, the ECM1^Tg mice expressed elevated ECM1 levels and were resistant to EAE (Fig. 7E). Histopathology revealed significantly reduced inflammatory cell infiltration and neuronal demyelination in ECM1^Tg mice compared with WT (Fig. 7F and supplementary Fig. 4). These results demonstrate that T cell-produced ECM1 is able to protect mice from (Th17 dependent) EAE pathology.

Figure 7. ECM1^Tg mice are resistant to EAE.

EAE was induced in C57BL/6 WT and ECM1^Tg mice with MOG_35–55 in CFA. CD4⁺ T cells from wild type and ECM1 transgenic mice were activated in vitro for 2 days and harvested for the determination of ECM1 expression by qPCR (A) and western blotting (B). (C) IL-17 and (D) IFN-γ production by draining LN cells. C57BL/6 WT and ECM1^Tg mice were immunized with MOG_35–55. After 8 days, draining LN cells were separated and re-stimulated with MOG_35–55 (50 µg/ml) for 3 days in the presence of IL-23. (E): Clinical scores of WT mice (n=10) and ECM1^Tg mice (n=10) were evaluated. (F): Luxol fast blue staining of paraffin sections of spinal cords isolated from representative WT and ECM1^Tg mice at 21 day after immunization. Data are representative of three independent experiments. Mean ± SEM is shown. *, P<0.05; **, P<0.01; ***, P<0.001.

Discussion

Multiple sclerosis (MS) is a CD4⁺ T cell-mediated autoimmune disease that affects the central nervous system (2). Th17 cells are widely believed to be critically involved in the pathogenesis of MS and experimental autoimmune encephalomyelitis (EAE), an animal model of MS (4). As we know, the cross-inhibition between different Th subsets exhibits valuable potential for clinical application to autoimmune diseases that are mediated by a specific Th subset. The present study demonstrates that ECM1, a protein that is specifically secreted by Th2 cells, is critically involved in the pathogenesis of EAE by regulating Th17 cell differentiation and lineage commitment. The effect of ECM1 is not T cell-intrinsic, but rather is mediated by inhibition of latent TGF-β activation, through competitive inhibition of the interaction between latent TGF-β and αv integrins on the surface of the DCs. These results suggest that the Th2-specific secreted protein ECM1 may play a regulatory role in immune homeostasis in EAE.

Previous studies have been reported that αv integrin play an important role in the regulation of TGF-β activation. The αv integrin could interact with latent TGF-β and then promote the release of active TGF-β into the microenvironment and regulate Th17 cell differentiation. In our present study, we found ECM1 protein suppressed the differentiation of Th17 in DCs-T cell co-culture system, and then we observed that ECM1 protein inhibited the production of active TGF-β. These results suggested that ECM1 might affect the process of TGF-β activation. For initial mechanism study, mass spectrometry (MS) was used to identify the proteins interacting with ECM1 from Th2 cells lysates. The results showed that αv integrin is one of proteins, which interact with ECM1. This was further confirmed in co-immunoprecipitation assay. These observations indicate that ECM1 inhibits the differentiation of Th17 and the development of EAE through blocking αv integrin-mediated TGF-β activation.

There are a number of studies (14, 44, 46) have demonstrated that TGF-β is important for mouse and human Th17 cell differentiation. While, considerable controversy regarding the requirement of TGF-β in directing Th17 cell development remains. Das (47) reported that TGF-β does not directly promote Th17 differentiation but instead acts indirectly through inhibiting expression of the transcription factors signal transducer and activator of transcription (STAT) 4 and GATA-3, thus blocking Th1 and Th2 differentiation and promoting Th17 generation. Ghoreschi (15) showed that Th17 differentiation could occur in the absence of TGF-β in vivo and in vitro. Gutcher’s study (14) found that T cell-produced TGF-β1 could act on T cells to promote Th17 cell differentiation and the development of EAE. Based on our experiments, it is likely that ECM1 blocks the maturation of TGF-β, which could be secreted by DCs, and thus suppresses the generation of Th17 cells.

Additionally, it is well documented that Th17 and Treg cells share a common requirement for TGF-β signaling during early differentiation. Therefore, the beneficial effect of ECM1 on disease through dampening of the Th17 response might be mitigated by a parallel dampening effect on the Treg response. Sugimoto et al (48) reported that ECM1 gene is predominantly transcribed in fresh and activated natural Treg as well as in Foxp3-transduced cells, indicating that ECM1 may play potential role in Tregs. In our study, we only detected the expression of ECM1 at mRNA level, but the production of ECM1 protein in Treg cells was undetectable (data no shown). Moreover, we observed that ECM1 treatment significantly reduced the expression of ROR-γt and IL-17, but the mRNA expression of Foxp3 and IL-10 in purified encephalitogenic CD4⁺ T cells from ECM1-treated EAE mice was slightly changed as compared to those from human IgG control-treated EAE mice (Fig. 2A and 2B), suggesting that ECM1 does not compromise Treg cell development. The reason is not clear, but the possibility is that there is a different sensitivity between Th17 and Treg cells in their responding to ECM1 treatment. It has been reported that Tregs, which produce TGF-β themselves, may be less dependent on DC-mediated TGF-β activation than Th17 cells due to their different sensitivity to TGF-β (49). Further studies would be needed to address the effect of ECM1 on Treg function in detail.

The immunosuppressive drugs now approved for the clinical treatment of MS work mainly by increasing the frequency of Treg cells or by changing the Th1-Th2 bias (50). In present study, we demonstrated that ECM1 significantly ameliorates EAE development by affecting the generation of pathogenic Th17 cells. This is equivalent to an effect on the afferent (priming stage) of disease, in line with the ability of ECM1 treatment to prevent, but not to reverse, EAE. Although in ongoing autoimmunity the effector T cells have already been generated, ECM1 may continue to affect pathogenesis even after disease has already developed. In relapsing-remitting diseases such as MS, new effector T cells may be differentiated and recruited into the effector pool in order to maintain chronicity. Therefore, an afferent-acting treatment may still be useful and it is not inconceivable that ECM1 augmentation could be explored as a treatment strategy in MS.

The questions that remain to be answered are: What is the importance of ECM1 function in MS patients? What is the mechanism by which ECM1 might regulate Th17 cell differentiation in MS patients? Does ECM1 also play a regulatory role in other autoimmune diseases? Future studies will address these questions.

In summary, we demonstrate that ECM1, a Th2-specific protein, has a robust therapeutic effect on EAE. The mechanism involves inhibition of the conversion of latent to active TGF-β on the surface of (antigen-presenting) DC by blocking its interaction with αv integrin, resulting in inhibition of Th17 differentiation. Our study suggests that ECM1 augmentation could be explored as a potential therapeutic approach in Th17 cell-mediated diseases.

Abbreviations

aTGF-β

active TGF-β

CNS

central nervous system

EAE

experimental autoimmune encephalomyelitis

ECM1

extracellular matrix protein 1

LAP

latency-associated peptide

L-TGF-β

latent TGF-β

MOG

myelin oligodendrocyte glycoprotein

MS

multiple sclerosis