Lumican negatively controls the pathogenicity of murine encephalitic TH17 cells

Abstract

TH17 cells play an essential role in the development of both human multiple sclerosis and animal experimental autoimmune encephalomyelitis (EAE). Nevertheless, it is not well understood how the pathogenicity of TH17 cells is controlled in the autoimmune neuro-inflammation. In vitro, we found Lumican (Lum), an extracellular matrix protein, is selectively expressed by TH17 cells among tested murine TH subsets. Lum-deficiency leads to earlier onset and enhanced severity of experimental autoimmune encephalomyelitis. This enhanced disease in Lum-deficient mice is associated with increased production of IL-17 and IL-21 and decreased TH17 cell apoptosis. Dysregulation in cytokine production appears to be specific to TH17 cells as TH1 and TH2 cell polarization and/or cytokine production were unaltered. Furthermore, adoptive transfer of MOG-specific TH17 cells derived from Lum-deficient mice led to earlier onset and increased severity of disease compared to controls highlighting a TH17 cell-intrinsic effect of Lum. Taken together, our results suggest that Lum negatively regulates encephalitic TH17 cells, implicating a potential therapeutic pathway in TH17 cell-mediated autoimmune and inflammatory diseases.

Introduction

Multiple sclerosis (MS) is a chronic autoimmune disease that attacks the central nervous system (CNS). The hallmarks of MS are demyelination of the fatty myelin sheaths around the axons of neurons and chronic inflammation caused by infiltration of cells from both the innate and adaptive immune system [1, 2]. Among the CNS infiltrates, autoreactive T cells play a critical role in the pathogenesis of MS as well as experimental autoimmune encephalomyelitis (EAE), an animal model of multiple sclerosis [1, 2]. Recent studies have implicated TH17 cells and their downstream pathways in the pathogenesis in EAE [3], as well as MS [4, 5].

CD4⁺ T cells are the central organizer of the adaptive immune system. In MS or EAE, the involved CD4⁺ T cells are mainly TH17 cells, TH1 cells and regulatory T cells [6]. TH17 cells can be differentiated by various combinations of the following cytokines TGFβ, IL-1, IL-6 and IL-23 [7]. Signature TH17 proinflammatory cytokines, IL-17, IL-17F, IL-21, and IL-22, mediate tissue inflammation by inducing many other inflammatory mediators or direct activation of other immune cells [8, 9]. IL-17 is highly expressed in MS. The frequency of TH17 cells is also significantly higher in the cerebrospinal fluid of relapsing remitting MS (RRMS) patients during relapse compared to RRMS patients in remission or to patients with other non-inflammatory neurological diseases [10]. In a recent clinic trial, hematopoietic stem cell transplantation treatment in patients with aggressive MS completely abrogated new clinical relapses and new focal inflammatory brain lesions throughout a 2-years immune monitoring, which was associated with sustained decrease and diminished TH17 responses but not TH1 or TH2 responses, suggesting a key role of TH17 cells in the disease progress and the maintenance of chronic autoimmune neuro-inflammation [5].

A major goal of the field is to determine how the immune infiltrates initiate and maintain the neuro-inflammation. Recently, based on EAE studies, TH17 cells have emerged as an initiating mediator of the disease. Nevertheless, there have been varying reports demonstrating differences in the initiating cytokines to generate pathogenic TH17 cells [11-13]. TH17 cells differentiated by TGFβ and IL-6 are non-pathogenic unless exposed to IL-23 [11]. Pathogenic TH17 cells are thought to initiate disease by producing IL-17, the signature cytokine produced by TH17 cells, which induces expression of matrix metalloproteases (MMPs) and the inflammatory cytokines IL-6 and TNF-α in endothelial cells. Together these mediators contribute to the disruption of blood-brain barrier (BBB) [8, 9, 14]. In humans, IL-17 and IL-22 produced by TH17 cells have been shown to disrupt BBB tight junctions [4]. In addition, TH17 cells highly express CCR6 and, in both mice and humans, TH17 cells transmigrate towards choroid plexus epithelial cells that constitutively express CCL20, the ligand of CCR6 [15-18]. Furthermore, TH17 cells also induce chemokines in the inflamed tissue that recruit more immune infiltrates and amplify inflammatory reactions [8, 9]. In summary, TH17 cells contribute to the initiation of autoimmune neuro-inflammation. Moreover, TH17 cells are involved in the maintenance of the chronic neuro-inflammation through induction of ectopic lymphoid follicles in CNS [19, 20]. Taken together, identifying immune mediators regulating TH17 cytokine production or TH17 maintenance has become a major focal point in MS and EAE studies.

Lumican (Lum), a small leucine-rich repeat proteoglycan, is known to be expressed in various tissues, including skin, artery, lung, vertebral discs, kidney, bone, aorta, and articular cartilage [21]. Lum is an important non-collagenous constituent of extracellular matrix [21]. Lum plays an important role in collagen fibrillogenesis accounting for the formation of transparent cornea [22], and also serve as a matrikine that regulates epithelial cell proliferation, migration, and adhesion during healing of epithelium debridement [23-28]. More recently, increased levels of Lum have been found in fibrosis of various organ systems, e.g., heart, liver and lung [29]. Besides, Lum plays an important role in innate immunity. Lum is required for neutrophil migration [30, 31], which is likely through regulation of the formation of chemokine gradient [32]. In addition, Lum also promotes bacterial phagocytosis by macrophages [33]. Nevertheless, it remains unknown whether Lum is involved in the regulation of function of helper T cells, such as TH17 cells in adaptive immunity.

Clearly, TH17 cells are important mediators in autoimmune neuro-inflammation; however, it is unclear if cell intrinsic or extrinsic factors control the pathogenicity of TH17 cells. In this report, we demonstrate Lum is selectively expressed by TH17 cells among subtypes of TH cells and, regulates TH17 cell survival, and modulates EAE disease progression.

Results

Lum alters cytokine expression but not the differentiation of TH17 cells in vitro

Lum, one of the extracellular matrix proteins has emerged as modulators of the inflammatory responses. We asked whether Lum plays a role in CD4⁺ TH cell differentiation and function. We polarized naïve CD4⁺ T cells isolated from Lum^−/− or Lum^+/− mice using plate-bound anti-CD3 and anti-CD28. Under the TH1 and TH2 conditions, Lum did not affect TH1 (Fig. 1A-B) and TH2 (Fig. 1C-D) cell differentiation and cytokine production. There were no visibly detectable IL-17⁺ cells in either Lum^−/− or Lum^+/− cells under these conditions (data not shown). Additionally, our results demonstrated that Lum only marginally affected TH17 cell development in the presence (Fig. 1E) or absence (Fig. 1G) of TGF-β (not reach statistical significance, data not shown). Despite no major differences in TH17 differentiation, we found that when skewed in the presence of TGF-β, Lum-deficiency significantly increased TH17 cell cytokines IL-17, IL-21 and GM-CSF outputs as measured by ELISA after re-stimulation with plate-bound anti-CD3 and anti-CD28 (Fig. 1F); similarly, when skewed in the absence of TGF-β, Lum-deficiency increased IL-17 and IL-21 (but not GM-CSF) expression (Fig. 1H). Furthermore, Lum-deficiency did not alter the development of CD4⁺ and CD8⁺ T cells and Foxp3⁺ regulator T (Treg) cells in thymus (Fig. 1I) and the suppressive function of Treg cells (Fig. 1J). Taken together, Lum selectively regulates cytokine production of TH17 but not TH1, TH2 and Treg cells.

Figure 1. Lum-deficiency increases IL-17 expression in vitro.

(A-J) Differentiation of naïve CD4⁺ T cell (A-G), thymic development of CD4⁺, CD8⁺ and CD4⁺Foxp3⁺ Treg cell compartments (I), and suppressive function of Treg cells (J) of Lum^−/− and Lum^+/− mice. (A, B, D) Flow cytometry of intracellular cytokine stain in cells skewed under TH1 (A), TH2 (C) or TH17 in the presence (E) or absence of TGF-β (G) conditions. (B, D, F, H) ELISA of cytokine expression in TH1 (B), TH2 (D), or TH17 (F, with TGF-β; H, without TGF-β) cells. (B, D, F, H) Data are shown as mean + SD (n = 3 in each group). Student's t-test, *, p ≤ 0.05; **, p ≤ 0.005. (I) Profile of CD4⁺ and CD8⁺ cells in thymus and Foxp3⁺ Treg cells in CD4⁺ thymocytes. (J) Suppression assay of Treg cells. (A-J) Data shown are representative of 3 independent experiments.

Lum negatively regulates experimental autoimmune encephalomyelitis

In Fig. 1, we observed Lum-deficiency increased cytokine output in TH17 but not TH1 and TH2 cells. Thus, it is plausible to speculate that Lum may play a role in Th17 cell-mediated immunity specifically an impact on the pathogenicity of TH17 cells in autoimmune encephalomyelitis. To investigate the role of Lum in TH17 pathogenicity, we examined the susceptibility to T cell-mediated EAE in Lum^−/−and Lum^+/− mice. Interestingly, Lum^−/− mice had earlier onset of the disease and exhibited more severe EAE symptoms (as determined by disease score) and incidence (Fig. 2A). Furthermore, we determined the number of inflammatory cells accumulated in the CNS of the EAE mice. When compared with Lum^+/− littermate controls, Lum^−/− mice had significant increase in the number of inflammatory cells in the CNS (Fig. 2B-C). Specifically, we observed more CD11b⁺ and CD4⁺ cells in Lum^−/− EAE mice (Fig. 2C).

Figure 2. Lum-deficiency enhances EAE disease in mice.

(A-D) Lum^−/− and Lum^+/− mice were examined for their susceptibility to T cell-mediated EAE. (A) (Left) Clinical scores and (right) disease incidence of EAE disease in Lum^−/− and Lum^+/− mice. (B) Flow cytometry of CNS infiltrates. IL-17 and IFN-γ expressing CD4⁺ cells were assessed after intracellular stain. (C) Cellular profile of the CNS infiltrates. (D) Quantification of IL-17, IFN-γ, and IL-17 and IFN-γ expressing CD4⁺ T cells in the CNS infiltrates. (A, C, and D) Data are shown as means + SD and (B, C and D) are representative of two independent experiments (n = 4–5 per group). (A) Data shown are pooled from two independent experiments (n = 9 mice/group). Student's t-test, *, p ≤ 0.05, **, p ≤ 0.005.

To further determine how Lum-deficiency contributes to the severity of EAE disease, we assessed cytokine production of the CNS infiltrating CD4⁺ T cells by intracellular stain. Lum-deficient mice displayed increased frequencies of single-positive IL-17⁺, IFN-γ⁺ and double-positive IL-17⁺ IFN-γ⁺ CD4⁺ T cells (Fig. 2D). These results suggest that Lum^−/− mice are more susceptible to autoimmune neuro-inflammation compared to their heterozygote littermates, implicating that Lum may have a suppressive role on the pathogenesis mediated likely via modulating the function of TH17 (and possibly TH1) cells.

Lum impacts cytokine expression ex vivo by autoreactive TH17 but not TH1 cells

To assess MOG-specific TH17 and TH1 cell frequency in the periphery, splenocytes from the EAE mice were stimulated overnight with MOG peptide as previously described [34]. Surprisingly, Lum-deficient splenocytes expressed significantly more IL-17 but not IFN-γ after ex vivo recall compared to littermate controls, although Lum did not alter the frequencies of TH17 nor TH1 cells after ex vivo recall with MOG peptide (Fig. 3A-B), in agreement with our in vitro observations (Fig. 1). Taken together, our data demonstrates that Lum attenuates EAE disease by inhibiting expression of IL-17, a TH17 cell cytokine; but not IFN-γ, a TH1 cell cytokine. Our data thus support the hypothesis that Lum controls the pathogenicity of TH17 cells in EAE.

Figure 3. Lum-deficiency promotes effector cytokine expression in peripheral autoreactive TH17 cells.

(A-B) Frequencies and cytokine production of autoreactive TH1 and TH17 cells in splenocytes from Lum^−/− and Lum^+/− EAE mice. (A) Intracellular stain of IL-17 and IFN-γ (recalled by MOG peptide) from splenocytes collected from Lum^−/− and Lum^+/− EAE mice on a CD4⁺ gate. (B) ELISA of IL-17 and IFN-γ produced from splenocytes from the Lum^−/− and Lum^+/− EAE mice after ex vivo re-stimulation with MOG peptide. (A and B) Data are shown as mean + SD (n = 4-5/group) and are representative of 2 independent experiments. Plots in (A) are pooled from 2 independent experiments. Student's t-test; *, p ≤ 0.05, **, p ≤ 0.005

Lumican is highly expressed in TH17 cells

The results above indicate Lum preferentially effects TH17 cytokine production both in vitro and in vivo. The observations in vitro using an antigen presentation cell-free system suggest that Lum modulates TH17 cell cytokine production in a TH17 cell-intrinsic manner. We thus hypothesized this specific effect on TH17 cells is due to differences in Lum expression. To determine the expression of Lum in various subtypes of CD4⁺ helper T cells, effector helper T cells were differentiated from CD4⁺CD25⁻CD44^loCD62L^hi naïve cells using specific polarization conditions (as described in materials and methods). The relative expression of Lum mRNA was examined in naïve CD4⁺ T cells, TH1, TH2, TH17, and inducible (i) and natural (n) Treg cells by RT-quantitative (q) PCR. Lum mRNA was highly expressed by TH17 cells, weakly expressed by TH1 cells with minimal to no detectable levels in all other CD4⁺ subsets (Fig. 4A). Furthermore, both TH1 and TH17 cells expressed increased amounts of Lum mRNA overtime, and TH17 cells expressed consistently higher amounts Lum mRNA compared to TH1 cells in a time course experiment (Fig. 4B). Additionally, Lum protein expression was assessed by western blot. Lum has been detected in almost all tissue types as well as serum [29], therefore, serum-free media was used for effector T cell differentiation. Aligning with the qPCR data, TH17 cells displayed increased protein levels of Lum compared to other CD4⁺ T cell subsets (Fig. 4C). Given this increase of Lum at both mRNA and protein level, we asked which TH17 cell skewing cytokine was involved in regulating Lum expression. When naïve CD4⁺ T cells were differentiated in the presence of IL-6 only a slight enhancement of Lum mRNA expression was observed compared to only anti-CD3/CD28 stimulation (Fig. 4D). Addition of IL-6 with TGFβ or TGFβ and IL-23 further enhanced Lum mRNA expression (Fig. 4D). These data suggest Lum expression is induced by the collective stimulation of TH17-inducing cytokines.

Figure 4. Lum is preferentially expressed in TH17 cells.

(A) RT-qPCR of Lum mRNA expression in indicated cell populations. Actb served as an internal control. (B) Time course of Lum mRNA expression in TH1 and TH17 cells as determined by RT-qPCR. (C) Immunoblot analysis of Lum protein expression in indicated T cell populations. β-Actin was used as a loading control. Blots are representative of 2 independent experiments. (D). RT-qPCR of Lum mRNA expression in CD4+ T cells differentiated in the presence of indicated cytokines. (A, B and D) Results were normalized to Actb. Data are shown as mean+ SD (n= 3/group and are representative of 3 (A) or 2 (B-D) independent experiments. Student's t-test; *, p ≤ 0.05; **, p ≤ 0.005.

Lum regulates EAE disease in a TH17 cell-intrinsic manner

Although in the T cell compartment, TH17 cells highly express Lum, many other tissues/cells also express this molecule. Therefore, we tested the TH17 cell-intrinsic role of Lum in EAE disease by using an adoptive transfer model. We immunized Lum^−/−and Lum^+/− mice with MOG emulsified in CFA and then expanded MOG-specific TH17 cells from splenocytes and draining lymph node cells from the immunized mice with MOG in the presence of IL-23 and IL-6 as previously described [35]. The expanded TH17 cells were then enriched and i.v. injected into congenic Rag1^−/− mice as previously described [35]. Similar with our observation in Lum^−/− mice, the recipients receiving Lum^−/− cells had earlier onset of the disease and exhibited much severer clinical symptoms than those receiving Lum^+/− cells (Fig. 5A). When compared with Lum^+/− controls, mice receiving Lum^−/− cells had significant increase in the number of inflammatory cells including CD4⁺ cells (CD11b⁺ cells were also increased but did not reach statistical significance) in the CNS (Fig. 5B). In the CNS infiltrating CD4⁺ T cells, mice receiving Lum^−/− cells contained increased frequencies of single-positive IL-17⁺ and IFN-γ⁺ cells (Fig. 5C-D). These results suggest that Lum regulates the pathogenicity of TH17 cells in a TH17 cell-intrinsic way.

Figure 5. Lum regulates EAE disease in a TH17 cell-specific manner.

(A) Clinical scores of EAE disease in Lum^−/− and Lum^+/− mice. (B) Cellular profile of the CNS infiltrates. (C) Intracellular stain of IL-17⁺ and IFN-γ⁺ cells in the CNS infiltrates on a CD4⁺ gate. (D) Quantification of IL-17, IFN-γ, and IL-17 and IFN-γ expressing CD4⁺ T cells in the CNS infiltrates. (A, B and D) Data are shown as mean+ SD (n = 4/group) and are representative of 2 independent experiments. Student's t-test,*, p ≤ 0.05.

Lum promotes the apoptosis of TH17 cells

The elevated expression of IL-17 by Lum-deficient cells may result from the following possibilities: First, Lum-deficiency may promote TH17 cell cycling, and lead to more TH17 cells that express higher amounts of TH17 cell cytokine. Second, Lum-deficiency may inhibit apoptosis of TH17 cells and, therefore, result in survival of more TH17 cells that express more cytokines. Although we expected that Lum-deficiency would enhance TH17 cell proliferation, we did not observe differences between Lum^+/− and Lum^−/− TH17 cells (Fig. 6A). These results rule out the possibility that Lum-deficiency affects TH17 cell cytokine expression via enhanced proliferation.

Figure 6. Lum-deficiency promotes survival of TH17 but not TH1 cells.

(A) CFSE dilution from Lum^+/− and Lum^−/− naïve CD4⁺ T cells polarized under TH17 conditions for 3 days. (B) Lum^+/− and Lum^−/− naïve CD4⁺ T cells (0.5 × 10⁶ each) were polarized under TH17 cell conditions. The numbers of resulting cells are shown as mean +SD (n=3/group).Student's t-test, *, p ≤ 0.05. (C) Flow cytometric expression of Fas on Lum^+/− and Lum^−/− TH17 cells. Filled, isotype control; solid, Lum^+/−; dotted, Lum^−/−. (D) Annexin V stain of Lum^+/− and Lum^−/− TH17 or TH1 cells after restimulation with anti-CD3 (E) Annexin V stain of Lum^+/− and Lum^−/− TH17 cells following re-stimulation with anti-CD3 in the presence or absence of anti-FasL. (F) Anti-Lum antibody co-immunoprecipitated FasL and Lum. (G) Anti-FasL antibody co-immunoprecipitated Lum and FasL. Iso, Isotype antibody. β-actin was used as a loading control (F, G). (A-G) Data shown represent 2 independent experiments.

Because we observed Lum-deficiency did not alter TH17 cell differentiation or proliferation but increased IL-17 expression in TH17 cells, we expected to observe Lum-deficiency protects TH17 cells from apoptosis. Indeed, after differentiation, Lum^−/− TH17 cell culture had increased numbers of cells than that of Lum^+/− TH17 culture (Fig. 6B). Previous studies have shown Lum aids in Fas-Fas ligand (FasL) mediated apoptosis in MEFs and various cancer cell lines [36, 37]. Although Lum is implicated in induction of Fas through binding to FasL [36], we observed that Lum did not alter Fas expression by TH17 cells (Fig. 6C), which was similar to an earlier report [37]. Nevertheless, after re-stimulation with anti-CD3, fewer Annexin V positive Lum^−/− TH17 cells (skewed with TGF-β) were detected compared to Lum^+/− cells (Fig. 6D), suggesting a pro-apoptotic role for Lum. We also observed similar results when the TH17 cells were skewed in the absence of TGF-β (data now shown). Interestingly, Lum-deficiency did not affect TH1 cell apoptosis (Fig. 6D). Because Lum did not affect Fas expression by TH17 cells, we asked whether the pro-apoptotic effect of Lum requires FasL. Therefore, we re-stimulated Lum^−/− and Lum^+/− TH17 cells with anti-CD3 in the presence of anti-FasL. In the presence of anti-FasL, the difference in Annexin V stain diminished between Lum^−/− and Lum^+/− TH17 cells (Fig. 6E), suggesting that Lum regulates TH17 cell apoptosis in a FasL-dependent manner. To verify Lum/FasL interaction on TH17 cells, naïve CD4 T cells were differentiated into TH17 cells in serum free media. Cell lysates were immunoprecipitated with anti-Lum, anti-FasL or isotype control antibodies and subjected to Western blot with anti-FasL and anti-Lum antibodies, respectively. FasL was co-precipitated with anti-Lum antibodies (Fig. 6F), in agreement with a previous report [38]. Reciprocally, Lum co-precipitated with anti-FasL antibodies (Fig. 6G). These data suggest Lum binds FasL and regulates the survival of TH17 cells via modulation of the Fas-FasL mediated apoptotic pathway. Taken together, Lum, selectively expressed by TH17 cells, promotes TH17 cell apoptosis in vitro and inhibits TH17 cell cytokine expression, therefore contributing to the repression of TH17 cell mediated CNS pathology.

Discussion

Collectively, our study has demonstrated that Lum, an extracellular matrix protein, is selectively expressed by TH17 cells and Lum-deficiency leads to earlier onset and increased severity of EAE, which is associated with increased accumulation of immune infiltrates. IL-23 has been shown to play a critical role in the migration of TH17 cells from draining lymph nodes into the peripheral and the CNS [39]. However, we observed that IL-23 did not inhibit Lum expression (Fig. 4D), suggest that IL-23 regulates TH17 cell migration independent of Lum. Integrins, α4β1 (VLA-4) and αLβ2 (LFA-1), play a pivotal role in antigen presentation and migration through the brain blood barrier (BBB) of CD4⁺ T cells [40-42]. Lum has been found to bind β1 or β2 integrins [26, 31, 43], suggesting a role of Lum in the function of pathogenic TH17 cells in EAE. However, murine Lum is unable to bind β1 integrin due to lacking a RGD motif [30]. Whether Lum regulates TH17 cell migration via binding β2 integrin is not clear at this time. Our study demonstrates that Lum inhibits IL-17 production which is likely through promoting TH17 cell apoptosis through its interaction with FasL, and in turn, impacts the pathogenicity of TH17 cells.

TH17 cells mediate inflammatory response through secretion of proinflammatory cytokines, IL-17, IL-17F, IL-21, IL-22 and GM-CSF [8, 9, 44, 45]. We observed that Lum-deficiency led to increased expression of IL-17, IL-21 (maybe also GM-CSF) by TH17 cells. IL-21 is an autocrine cytokine that sustains TH17 cell differentiation and cytokine production [46, 47]; therefore increased IL-21 expression in Lum-deficient TH17 cells may feedforward enhance TH17 cell function. IL-17, the signature cytokine of TH17 cells, induces the expression of matrix metalloproteases and the inflammatory cytokines, TNF-α and IL-6, in endothelial cells that together disrupt the BBB [8, 9, 14]. The BBB disruption is an early and central event in MS pathogenesis [44], which promotes early attack by autoreactive T cells. IL-17 also induces expression of chemokines in the inflamed tissue that promotes recruitment of infiltration of immune cells, including both TH17 and TH1 cells and macrophages [9]; accumulation of immune infiltrates causes tissue damage. Furthermore, increased TH17 cells in the CNS of Lum-deficient mice may also enhance macrophage function through expression of GM-CSF [45]. Consistently, TH17 cell-intrinsic defect in Lum is sufficient to promote EAE disease in an adoptive transfer experiment. Taken together, our model supports the notion that autocrine Lum attenuates the pathogenicity of TH17 cells.

Given the importance of TH17 cells in inflammatory disease, targeting TH17 cells emerges a potential therapeutic approach for such diseases. Current disease-modifying therapies for MS mainly use immunosuppressant's that broadly repress immune responses and so occasionally display severe adverse effects, including viral reactivation, susceptibility to viral infection and malignancies [48-50]. Therapies that specifically target TH17 cells but leaving intact other arms of TH cells in the immune system may reduce the inflammatory environment that drives disease with less side effects of broad immunosuppressive therapy. Lum is an endogenous inhibitor of TH17 cells that is selective expressed by TH17 cells; therefore enhancement of the Lum pathway may benefit Th17 cell-mediated inflammatory diseases, such as MS.

Materials and Methods

Mice

Lum-deficient mice (Lum^−/−) on the C57BL/6 background were described previously [27]. Because of male Lum^−/− has limited fertility due to age-related skin lesion [27], we had to breed female Lum^−/− with male Lum^+/−, which produced a half of Lum^+/− mice. To reduce the use of animals, we used littermate Lum^+/− mice as controls in most of experiments (in fact, in a pilot EAE experiment, Lum^+/− mice developed similar severity of disease as Lum^+/+ mice; data not shown). All mice were housed in the specific pathogen-free animal facility at the University of New Mexico Health Sciences Center. All experiments were performed with mice 6-8 weeks old at the initial of experiments with protocols approved by the Institutional Animal Care and Use Committee of the University of New Mexico.

In vitro T cell differentiation and flow cytometric analysis

Naïve CD4⁺ CD25⁻ CD62L^hi CD44^lo cells were sorted and activated with plate-bound anti-CD3 and anti-CD28 in serum-free media under the following conditions for polarization of TH cells: IL-12, IL-2 and anti-IL-4 for TH1 cells; IL-4 and anti-IFN-γ for TH2 cells; IL-6, IL-23 and anti-IFN-γ and anti-IL-4 with or without TGF-β as indicated for TH17 cells; TGF-β and IL-2 for inducible regulatory T cells. For intracellular cytokine stain, the differentiated cells were stimulated with phorbol 12-myristate 13-acetate (PMA) and ionomycin in the presence of Golgi-stop for 4 hr. For proliferation analysis, the naïve cells were stained with carboxyfluorescein succinimidyl ester (CFSE) (Invitrogen) prior to differentiation. Following 3-day differentiation, CFSE dilution was assessed by flow cytometry. For Treg cell suppression assay, CFSE labelled naïve cells were co-cultured with or without CD4⁺CD25⁺ Treg cells isolated from splenocytes of indicated mice for 3 days in the presence of plate-bound anti-CD3 and paraformaldehyde-fixed splenic antigen presenting cells.

EAE induction and isolation of CNS-infiltrating cells

EAE was induced using a protocol described previously with minor modifications [34, 51]. In brief, the mice were immunized with 0.5-1.5 mg/ml of myelin oligodendrocyte glycoprotein (MOG) peptide (aa. 35-55, MEVGWYRSPFSROVHLYRNGK) emulsified in complete Freund's adjuvant (CFA) (day 0) or incomplete Freund's adjuvant (IFA) (day 7) subcutaneously at flanks, 0.05 ml per site, total 2 sites. The mice peritoneally received 2 doses of pertussis toxin (0.5 μg in 100 μl PBS per mouse) on day 1 and day 8. The mice were evaluated daily. Clinical symptoms were scored: 0, No clinical signs; 1, Loss of tail tone; 2, Wobbly gait; 3, Hind limb paralysis; 4, Hind and fore limb paralysis; 5, Death. The animal was sacrificed when it reached a score 4. The experiment was terminated if 2 out of 5-6 animals in a group were sacrificed or died. To isolate the CNS infiltrates, both brain and spinal cord were collected from the EAE mice following intracardiac perfusion immediately after euthanasia. Mononuclear cells were prepared into single cell suspensions by percoll gradient and subjected to FACS analysis for cell profile.

Adoptive transfer of TH17 cells induced EAE

Sex and age matched Lum^−/− and Lum^+/− donor mice were immunized with MOG-CFA, and 7 d later, the splenocytes and draining lymph node cells were collected and cultured with 100 μg/ml MOG in the presence of 20 ng/ml IL-23 and 20 ng/ml IL-6 for 5-7 days. Expanded TH17 cells were then enriched by selection of CD4⁺ cells with magnetic beads and i.v. injected into congenic Rag1^−/− mice (2.5 × 10⁶ cells/mouse) followed one dose of pertussis toxin as previously described [35].

ELISA

Supernatants were harvested from in vitro differentiation after recall with plate-bound anti-CD3 and anti-CD28 for 20 h or from ex vivo cultures of splenocytes from the EAE mice with MOG re-stimulation for 3 d. Expression of indicated cytokines was assessed by ELISA.

Western blot and co-immunoprecipitation analysis

Whole-cell lysates were subjected to immunoblot analysis using a standard protocol. The antibodies were as follows: anti-Lum (AF2745; RnD systems) and anti-β-Actin (BA3R; MA5-15739; Thermo Fisher Scientific). For co-immunoprecipitations of Lum to FasL or FasL to Lum, we incubated protein A/G magnetic beads with 2 μg of anti-Lum or anti-FasL (MFL3, eBioscience) antibody or isotype control for 4 hours at 4°C. The bead-antibody complexes were washed with cold PBS and then incubated with whole-cell lysates for overnight at 4°C. Following washes with lysis buffer and cold PBS, the bead-immune complexes were then resuspended in Laemmli's sample buffer and boiled for 5 minutes. Samples were then prepared for immuoblot analysis.

RT-PCR and Quantitative real-time RT-PCR

Gene expression level was determined by Quantitative real-time RT-PCR as described previously [34, 51]. Data were normalized to an Actb reference gene. The primers were as follows: Lum forward, 5′-tggctgatagtggggtacct, and reverse, 5′-aggattgccatccaagcgca; Actb forward, 5′-gacggccaggtcatcactattg, and reverse, 5′-aggaaggctggaaaagagcc.

Statistical analysis

The statistical significance of differences between groups was calculated with the unpaired Student's t test. P values of 0.05 or less were considered significant.