GPR65 inhibits experimental autoimmune encephalomyelitis through CD4+ T cell independent mechanisms that include effects on iNKT cells

Abstract
The G protein-coupled receptor 65 (GPR65) gene has been genetically associated with several autoimmune diseases, including multiple sclerosis (MS). GPR65 is predominantly expressed in lymphoid organs and is activated by extracellular protons. In this study, we tested whether GPR65 plays a functional role in demyelinating autoimmune disease. Using a murine model of MS, experimental autoimmune encephalomyelitis (EAE), we found that Gpr65- deficient mice develop exacerbated disease. CD4+ helper T cells are key drivers of EAE pathogenesis, however, Gpr65 deficiency in these cells did not contribute to the observed exacerbated disease. Instead, Gpr65 expression levels were found to be highest on invariant natural killer T (iNKT) cells. EAE severity in Gpr65-deficient mice was normalized in the absence of iNKT cells (CD1d-deficient mice), suggesting that GPR65 signals in iNKT cells are important for suppressing autoimmune disease. These findings provide functional support for the genetic association of GPR65 with MS and demonstrate GPR65 signals suppress autoimmune activity in EAE.

INTRODUCTION
Newly identified G protein-coupled receptors (GPCRs) are attractive therapeutic targets due to receptor-ligand specificity, their cell membrane localization and the discrete distribution of individual GPCRs in the body. As a testament to their suitability as drug targets, an estimated 36% of existing, marketed pharmaceuticals act at GPCRs.1 Genome wide association studies have identified the G protein-coupled receptor 65 (GPR65) locus to be associated with multiple sclerosis (MS)2,3, as well as type 1 diabetes,4 ankylosing spondylitis,5 inflammatory bowel diseases6,7 and rheumatoid arthritis.8 The GPR65 locus also contains the GALC gene and these two genes are in tight linkage disequilibrium.2,3 Thus, genetic association studies alone are unlikely to identify which gene is driving increased risk of MS. Expression of the inflammatory bowel disease-associated risk variant of GPR65 (rs3742704) or null deletion of GPR65 both resulted in autophagy defects, consistent with the loss of function of GPR65 in the disease-associated haplotype.9 Functional studies are required to determine the relevance and importance of GPR65 in MS.Gpr65-deficient mice display no overt phenotype prior to immunological challenge, nor any changes in antibody production in response to T-dependent or T-independent antigens.10 Gpr65-deficient mice develop exacerbated disease in models of arthritis and delayed-type hypersensitivity.11 These mice also exhibit increased immune cell recruitment, IL17A production and adverse outcomes in a model of myocardial infarction.12 Mutation of Gpr132, a Gpr65 family member, leads to the development of a lupus-like autoimmune condition in mice.13 These previous observations support the idea that GPR65 controls immunological processes relevant to MS, independent of unstimulated T-cell phenotype and antibody production.GPR65 encodes an acid (proton) sensing GPCR14,15 for which the biological outcomes of signaling are incompletely understood. Local tissue acidosis, due to glycolysis, is a hallmark of inflammation. Acidic pH has been reported in the CNS during experimental autoimmune encephalomyelitis (EAE) (pH 6.6 compared with 7.4 in healthy CNS)16 and lactate has been detected in MS patient brain biopsies.17 There is evidence that GPR65 signaling could control numerous aspects of immune function through the adenylate cyclase pathway. Increasing acidity results in GPR65-dependent Gas- mediated activation of adenylate cyclase and leads to the accumulation of cAMP in multiple primary cell types including eosinophils,18 macrophages,19 and thymocytes.20 In leukocytes, cAMP/PKA can influence many cellular functions including activation, motility and cytokine production.21Here, we test the hypothesis that GPR65 regulates EAE development. Our data point to a key regulatory role for GPR65 in demyelinating autoimmunity, at least partially through effects on iNKT cells.

RESULTS
The Gpr65gfp/gfp mouse line has GFP spliced into the second exon of the Gpr65 gene, generating a complete deletion of Gpr65 with GFP as a reporter of GPR65 expression.10 To test the functional relevance of GPR65 in demyelinating autoimmune disease, we induced “active” EAE in groups of Gpr65gfp/gfp and Gpr65+/+ mice by immunizing with MOG35-55 peptide in incomplete Freund’s adjuvant supplemented with heat-killed Mycobacterium tuberculosis (M. Tb). In our colony, inclusion of pertussis toxin led to highly lethal form of disease (d.n.s); we therefore omitted pertussis toxin in our disease induction strategy. Omission of pertussis toxin also avoided disrupting Gai association with GPCRs,22 although GPR65 is not thought to associate with Gai.15 This strategy produced EAE with mild severity, allowing for the detection of any changes in disease in the absence of GPR65.Mice were monitored for 30 days after immunization. Average disease scores were higher in the Gpr65gfp/gfp group compared to Gpr65+/+ controls and area under the curve (AUC) analysis indicated that Gpr65gfp/gfp mice developed significantly exacerbated disease (Figure 1a), with individual Gpr65gfp/gfp animals reaching significantly higher peak clinical scores than Gpr65+/+ mice (Figure 1b). The median onset of disease was 17 days for Gpr65gfp/gfp mice and 22 days for Gpr65+/+ mice. A trend for increased incidence of mice developing symptoms of disease was observed in Gpr65gfp/gfp mice, although this did not reach statistical significance (Figure 1c). Thus, in contrast to a recent report, in which adoptively transferred Gpr65gfp/gfp CD4+ T cells were unable to confer disease,23 we found that Gpr65gfp/gfp mice were indeed susceptible to typical active EAE induction and developed more severe disease than their wildtype counterparts.

Loss of Gpr65 expression in MOG35-55-specific CD4+ T cells does not significantly exacerbate EAE development in an adoptive transfer model
In the EAE model, demyelinating disease is induced by activated MOG35-55 peptide-specific CD4+ T cells.24 To analyze the effects of Gpr65 deficiency within CD4+ T cells in the context of EAE, we generated mixed bone marrow (BM) chimeras comprising 50% Gpr65+/+ or Gpr65gfp/gfp (Thy1.2 expressing) BM mixed with 50% WT C57BL6 (Thy1.1 expressing) BM. In the steady-state, Gpr65gfp/gfp CD4+ T cells repopulate irradiated hosts and differentiate normally (Figure 2a).To test CD4+ T cell function upon challenge, mixed BM chimeras were immunized with MOG35-55 in CFA (as for EAE induction) and the draining inguinal LN and central nervous system (CNS) infiltrating leukocytes collected at D15 post immunization, corresponding to the time of disease onset. Tissues were subjected to MOG35-55 re-stimulation ex vivo to allow quantitation of intracellular cytokines in antigen-specific CD4+ T cells. Intracellular staining was performed to test IL-17, IFNc and Foxp3 expression in CD4+ T cells.

Figure 1. Gpr65-deficient mice develop exacerbated EAE. Gpr65+/+ (n = 24) or Gpr65gfp/gfp (n = 28) mice were immunized in the flank with 50 lg MOG35-55 in incomplete Freund’s adjuvant containing 400 lg M. Tb. Clinical symptoms of paralysis were monitored over 30 days. (a) Average clinical score in each group, bars indicate s.e.m.(b) Maximum clinical scores reached, each symbol represents an individual mouse. (c) Incidence of EAE symptoms over time. Data are pooled from two independent cohorts. For clinical scores, P < 0.05 was determined by area under the curve for each mouse and groups compared using an unpaired t-test. For maximum scores, P < 0.05 was determined using an unpaired t-test. For incidence, P < 0.05 was determined using a Gehan-Breslow-Wilcoxon test. EAE, experimental autoimmune encephalomyelitis of these factors in either WT (CD45.2+Thy1.2+) or Gpr65gfp/gfp (CD45.2+Thy1.2+) CD4+ T cells was compared to the expression observed in the C57BL6 (CD45.2+Thy1.1+) CD4+ T cells present in the same host mouse. The ratio in each mouse was calculated, as this overcomes the variation in disease severity between host mice. Raw values are provided in Supplementary figure 1. The data indicate that there are no T cell intrinsic differences in IL-17, IFNc or Foxp3 expression in Gpr65gfp/gfp CD4+ T cells. Likewise, when CD4+ T cells from nonchimeric Gpr65gfp/gfp and Gpr65+/+ mice were MOG35-55 restimulated ex vivo, no significant differences in cytokine production were observed (d.n.s).MOG35-55 specific CD4+ T cells from 2D2.Tg mice were used to specifically test the function of Gpr65gfp/gfp CD4+ cells in vivo in the EAE setting. In this transfer model of EAE, 2D2 MOG35-55 specific CD4+ T cells were transferred into immunologically replete wildtype recipients. 2D2.Tg Gpr65+/+ or 2D2.Tg Gpr65gfp/gfp donor cells induced comparable disease in recipients (Figure 2c). Proliferation of CTV-labeled, adoptively transferred 2D2.Tg T cells was similar 3 days after immunization with MOG35-55 in CFA (Figure 2d). Thus, CD4+ T cell differentiation and function, and the anti- MOG response appears to be intact in Gpr65gfp/gfp mice.

Since GPR65 activates MEK/ERK25 and enhances the survival of eosinophils and lymphoma cells,18,25 we also tested whether survival was effected in Gpr65gfp/gfp CD4+ T cells. Indeed, a survival defect was observed in Gpr65gfp/gfp CD4+ T cells in culture either with or without anti-CD3/ anti-CD28 stimulation (Figure 2e and f).GPR65 is highly expressed by iNKT cells and Gpr65 deficiency does not alter the course of EAE in the absence of iNKT cells Since changes in the conventional CD4+ T cell anti-MOG response could not explain exacerbated EAE in Gpr65gfp/gfp mice, we next investigated Gpr65 expression in other immune cells. The GFP reporter in Gpr65gfp knock-in mice has been used previously to show expression of Gpr65 in the stages of developing thymocytes as well as broadly in mature populations of CD4+ and CD8+ T cells, B cells, granulocytes, macrophages, NK (CD3-/ DX5+) cells and DC.10 We extended this to investigate the regulation of GPR65 during T-cell activation and in additional cell types. The intensity of the GFP reporter in heterozygous mice (Gpr65+/gfp) was used to characterize expression (Figure 3a). Since we observed that the CD4+CD44+CD62Llo (effector CD4+ cell) gate contained up to 40% iNKT cells in immunologically na€ıve mice in our colony, we were careful to use aGalCer-CD1d tetramer to differentiate iNKT cells and conventional T (T-conv) cells. Gpr65 was widely expressed, with reporter GFP detected in all immune cells tested (Figure 3a and b). Gpr65 expression was high in DN2 thymocytes, NK cells, DC, cd T cells and was most highly expressed by iNKT cells, which had approximately twice the GFP level of na€ıve CD4+ T cells (Figure 3a and b). By contrast, effector CD4+ T cells had slightly reduced GFP signal compared to na€ıve CD4+ cells (Figure 3a and b). Gpr65 mRNA quantitation in iNKT cells showed an even more.

Figure 2. CD4+ T cells do not contribute to the exacerbated EAE observed in Gpr65-deficient mice. (a-b) 50:50 mixed BM chimeras were generated using 50% Gpr65gfp/gfp BM (CD45.2+Thy1.2+) mixed with 50% Thy1.1 congenic WT BM (CD45.2+Thy1.1+). Control group mice were generated with 50% Gpr65+/+ BM (CD45.2+Thy1.2+) mixed with 50% Thy1.1 congenic WT BM (CD45.2+Thy1.2+). Recipient mice were irradiated CD45.1 mice. (a) Reconstitution was assessed at 8 weeks post-transfer and the percentage of CD4+ na€ıve, effector and regulatory T cells in the Gpr65+/+ (closed circles) or Gpr65gfp/gfp (open squares) gate is shown relative to the CD45.2+Thy1.1+ WT donor cells in the same mouse. (b) Mixed BM chimeras were immunized with MOG35-55 as in Figure 1 and CNS infiltrating leukocytes were harvested at day 15 post immunization. Cells were re-stimulated with 100 lg mL-1 MOG35-55 ex vivo for 24 h. The percentage of cytokine positive cells in the Gpr65+/+ (closed circles) or Gpr65gfp/gfp (open squares) gate is shown relative to the CD45.2+Thy1.1+ WT donor cells in the same mouse. Data are representative of three independent experiments. No statistically significant differences were observed using a t-test. (C and D) 2D2.Tg Gpr65+/+ or 2D2.Tg Gpr65gfp/gfp CD4+ T cells were transferred into WT.CD45.1+ recipients, simultaneously immunized with MOG35-55 as in Figure 1. (c) Average clinical score for each group (left panel) over 26 days, bars indicate s.e.m. Maximum scores reached by individual animals (right panel; each symbol represents an individual mouse, bars indicate the mean). 2D2.Tg Gpr65+/+, n = 8; 2D2.Tg Gpr65gfp/gfp, n = 10. Data is representative of two independent experiments. (d) 2D2.Tg Gpr65+/+ and 2D2.Tg Gpr65gfp/gfp CD4+ na€ıve T cells were sorted, labeled with CTV and equal numbers of each were mixed. WT.CD45.1+ recipient mice received a total of 5 9 105 cells. Recipients were simultaneously immunized as in (c). Mice were sacrificed on day 3 post-transfer and iLN harvested for flow cytometric analysis. Representative CTV plots and the cell number recovered from each mouse for unimmunized (—MOG35-55) and immunized (+MOG35-55) mice. Each symbol represents an individual mouse, bars indicate the mean. No
statistically significant differences were observed. The data presented are representative of two independent experiments. (e & f) Sorted na€ıve
CD4+ T cells from Gpr65+/+ or Gpr65 gfp/gfp mice were left untreated (unt) or stimulated with anti-CD3 alone (aCD3) or along with anti-CD28 (aCD3 + aCD28) for 48 h. (e) Live 7AAD-negative cells as a percentage of the total lymphocyte gate and (f) calculation of total live cell number per well. Each point represents an experimental replicate. Representative of four independent experiments. Bars indicate mean s.d. Unpaired t-test ****P < 0.0001. BM, bone marrow; EAE, experimental autoimmune encephalomyelitis striking increase, with at least 40-fold higher Gpr65 mRNA observed in sorted iNKT cells compared with any of the T-conv cell subsets tested (Figure 3c). Investigation of GFP expression in iNKT cell subsets, divided based on CD4 and NK1.1 expression, revealed uniformly high expression (Supplementary figure 2). These data collectively demonstrate that Gpr65 is expressed widely on leukocytes, including all types of peripheral T cells tested, and the expression was particularly high on innate and innate-like lymphocytes.
It was noted that effector CD4+ and CD8+ T cells had reduced GFP expression relative to na€ıve T cells (Figure 3a and b).

We therefore examined the expression of GFP in iNKT cells at day 10 after the induction of EAE. In line with the observations in T-conv cells, GFP MFI in iNKT cells was reduced during EAE (Figure 3d). However the GFP levels in iNKT cells remained higher than those observed in CD44+ CD4+ T-conv cells (Figure 3d).Given the elevated expression of Gpr65 in iNKT cells, the impact of Gpr65 deficiency on the steady state populations within the thymus and in peripheral lymphoid organs was investigated. The percentages and total cell numbers of iNKT were comparable between Gpr65+/+ and Gpr65gfp/gfp mice within thymus, spleen and iLN (Figure 3e).TCR-mediated activation of iNKT cells via aGalCer administration during EAE suppressed disease severity in groups of both Gpr65+/+ and Gpr65gfp/gfp mice (d.n.s). However, the interpretation of this data is complicated by the fact that all iNKT cell subsets, including those producing IFNc, IL-4 and IL-17, are highly stimulated by aGalCer, skewing the cytokine profile independently of GPR65. Also, the TCR mediated nature of the activation differs from conventional EAE where iNKT cells are activated indirectly.26 We therefore used an iNKT deficient model.
Development of iNKT cells is dependent upon interaction with cells expressing the MHC-like molecule CD1d. To assess whether Gpr65gfp/gfp mice displayed exacerbated EAE in the absence of iNKT cells, CD1d—/— Gpr65gfp/gfp mice were generated. CD1d—/— Gpr65gfp/gfp and CD1d—/— Gpr65+/+ mice exhibited similar disease severity (Figure 3f). The observation of disease exacerbation in WT Gpr65gfp/gfp mice, but not CD1d—/— Gpr65gfp/gfp mice suggests that loss of Gpr65 in iNKT cells could contribute to the pathogenesis of EAE.

DISCUSSION
The increased severity of EAE in the absence of GPR65 provides functional evidence to support the genetic association of GPR65 in autoimmune demyelinating disease.2,3 Our data indicate that GPR65 plays a protective role during EAE, since exacerbated EAE was observed in Gpr65-deficient mice compared with wildtype controls. Our data contrast with a published study, in which wildtype or Gpr65 deficient CD4+ T cells (from the same strain used in our investigation10) were isolated and transferred into Rag1—/— recipients, and EAE induced by immunization two weeks later.23 That study found that disease was delayed in mice that received Gpr65 deficient CD4+ T cells, compared with those that received wildtype cells.23 The differing effect on disease in that model could be a consequence of low numbers of Gpr65 deficient CD4+ T cells in Rag1—/— recipient mice after the parking period,
as we have noted a survival defect of na€ıve CD4+ T cells from Gpr65-deficient mice (Figure 2e and f). GPR65 has been shown to enhance cell survival in eosinophils18 and lymphoma cell lines through MEK/ERK activation.25 Therefore it is possible that the “parked” CD4+ T cells, which would have been undergoing homeostatic proliferation in the Rag1—/— host in the 2 weeks prior to immunization,27 may have proliferated or survived less efficiently than wildtype counterparts. In addition, recipient mice in that study lacked B cells, CD8+ T cells and iNKT cells. We have demonstrated that Gpr65- deficient mice develop exacerbated disease in a system with minimal manipulation and in immunologically replete animals. Furthermore, our analysis of mixed BM chimeric mice provides strong evidence for normal function of Gpr65gfp/gfp CD4+ T cells in EAE. In line with our findings, GPR65 was also required to attenuate inflammation in models of arthritis11 and myocardial infarction.12

Our investigation uncovered strikingly high expression of GPR65 in iNKT cells, along with normalized EAE levels in Gpr65-deficient mice in the absence of iNKT cells. Gpr65 mRNA levels were ~40 fold higher in iNKT cells compared with na€ıve T cells. We were not able to confirm changes at the protein level due to a lack of specific anti- GPR65 antibodies. The percentage of iNKT cells in various organs was not altered in the absence of Gpr65,
however this does not preclude changes in iNKT cell function. Innate-like T cells such as iNKT cells, mucosa- associated invariant T (MAIT) cells and cdT cells exist in a pre-activated state and can have potent effects in EAE through the early production of cytokines.28-30 Although iNKT cells are less numerous in human tissue, iNKT and MAIT cells have been found within MS lesions, their numbers negatively correlate with disease and functional data suggest an immunosuppressive role for these cells.29- 33 These data are consistent with a homeostatic role for GRP65 in acidic conditions, preventing excessive immune response through regulation of iNKT cells. The Gpr65gfp reporter data presented here also shows high levels of expression in other innate-like lymphoid cells such cdT cells and NK cells. MAIT cells are infrequent in mice

Figure 3. iNKT cells express high levels of Gpr65 and contribute to the exacerbated EAE observed in Gpr65-deficient mice (a) Mean fluorescence = intensity (MFI) of the GPR65 GFP reporter was determined by flow cytometry, gating various immune cell populations of Gpr65gfp/+ mice (filled bars) and compared to autofluorescence in the same populations in Gpr65+/+ mice (empty bars). Populations were defined by the following markers: Thymocytes DN1 (CD4—CD8—CD44+CD25—), DN2 (CD4—CD8—CD44+CD25+), DN3 (CD4—CD8—CD44—CD25+), DN4 (CD4—CD8—CD44—CD25—) and DP (CD4+CD8+CD44—CD25—), cd T cells (CD3+cdTCR+), iNKT cells (CD3+CD1dtet+), NK cells (CD3—NK1.1+), Treg (CD4+CD25+), CD4+ and CD8+ na€ıve (CD1dtet—CD62L+CD44—), effector (CD1dtet—CD62LloCD44+) and central memory (CD1dtet—CD62L+CD44+) T cells, B cells (B220+), neutrophils (CD11b+Ly6G+), DC (CD11c+) and macrophages (CD11b+). Bars indicate the mean of 2-3 individual mice and S.D. (b) Representative histogram overlays for GFP levels in Gpr65gfp/+ iNKT cells (red line), CD4+ na€ıve (green line) and effector T cells (blue line). Background fluorescence in Gpr65+/+ iNKT cells is given (grey filled histogram). (c) Relative Gpr65 mRNA levels in FACS purified wildtype CD4+ na€ıve (Tn), CD4+ central memory (Tcm), CD4+ effector memory (Tem), CD8+ na€ıve (Tn) and iNKT cells from C57BL/6 mice. Data are given as the fold change relative to Rpl19 x1000. Histograms indicate the mean of three replicates and error bars indicate s.d. Data is representative of at least two independent experiments for each data point. (d) The relative GFP MFI (KO:WT) was calculated for iNKT cells and CD4+ Tconv cells from na€ıve mice (black bars), or from mice 10 days after EAE induction (grey bars). (e) The percentage of iNKT (CD3+CD1d-tet+) cells present in the spleen, inguinal LN and thymus was quantified in groups of Gpr65+/+ and Gpr65gfp/gfp mice by flow cytometry. No statistically significant difference was observed, data are representative of three independent experiments. (f) Wildtype and CD1d—/— mice were immunized with 50 lg MOG35-55/incomplete Freund’s adjuvant /M.Tb. Clinical scores were monitored for 24 days. Average clinical score is given for each group of mice. The data are pooled from four independent experiments, CD1d—/—Gpr65+/+: n = 53 and CD1d—/—Gpr65gfp/gfp: n = 55. No significant difference was observed by comparing the area under the curve values between the two groups by t-test.EAE, experimental autoimmune encephalomyelitis were therefore not tested for expression here. Investigating GPR65 expression on MAIT cells and other CD1- restricted cell subsets in humans will be informative as to the effects of GPR65 in human disease.

C57BL/6JAusb and B6.SJL-PtprcaPepcb/BoyJAusb (CD45.1 congenic) mice were obtained from Australian Bioresources (ABR; Mossvale, Australia) and B6.SJL-Ptprc (CD45.1 congenic), B6.129S7-RAG1 (Rag1—/—) mice were obtained from the Animal Resource Centre (Perth, Australia). Gpr65gfp10 mice were purchased from JAX® (stock number 008577) and backcrossed to C57BL/6J mice (n > 11). To avoid genetic drift in experimental cohorts, heterozygotes were paired, then Gpr65gfp/gfp or Gpr65+/+ homozygous F1 were selected for breeding, cohorts Gpr65gfp/gfp or Gpr65+/+ F2 progeny were used in experiments. 2D2 TCR transgenic (2D2.Tg) mice24 were supplied by A/Professor David Brown, The Westmead Institute, Australia. The two strains were intercrossed to generate the 2D2.Gpr65gfp line. CD1d—/— mice for breeding and B6.PL- Thy1a/CyJ (Thy1.1) congenic mice were originally purchased from Jackson labs (stock numbers 003814 and 000406). All mice were backcrossed at least 10 times to C57BL/6J. Animals were housed in specific-pathogen free (SPF) conditions (ABR), Moss Vale, or within the BTF at the Garvan Institute. All animal procedures were approved by the Institutional Animal Care and Use Committee at The Garvan Institute/St. Vincent’s Animal Experimentation Ethics Committee.EAE was induced in 8–13 week old mice by immunizing mice subcutaneously with 100 lL of an emulsification containing 50 lg/mouse of myelin oligodendrocyte glycoprotein peptide amino acids 35-55 (MOG35-55) (MEVGWYRSPFSRVVHLYR
NGK; ProSpec-Tany Technogene Ltd.) in incomplete Freund’s adjuvant supplemented with 400 lg mouse—1 heat killed Mycobacterium tuberculosis (M.Tb; H37RA, Difco). Mice were monitored daily for symptoms and were scored as follows: 0, no disease; 1, loss of tail tone; 2, tail paralysis; 3, hind limb weakness; 4, hind limb paralysis; and 5, hind limb paralysis and forelimb weakness. Mice were euthanized if they reached a score >4, according to Garvan Institute Ethics guidelines. All scoring was performed by researchers blinded to the genotype of the mice.

Flow cytometric staining was performed using panels of anti- mouse antibodies that included those against the following mouse proteins: CD4, CD3, CD45.2, Thy1.1, Thy1.2, CD24, NK1.1, CD44, CD62L, IFNc, IL-10, IL-17, B220, Ly6G, CD11b(BD Biosciences) CD4, CD25, Foxp3, cdTCR, CD11c (eBiosciences). PE and BV421 labelled CD1d-aGalCer tetramer was purchased from Professor Dale Godfrey at the University of Melbourne. 7AAD (BD Biosciences) was used to exclude nonviable cells.Briefly, cells were stained with desired surface markers for 30 min on ice. For intracellular staining, cells were stained first with surface markers followed by fixation and permeabilization (eBioscience). Flow cytometry data were acquired on the BD-CANTOII cytometer and analyzed using the FlowJo v9.7.7software (Tree Star).Cells were harvested under sterile conditions from spleen and lymph nodes (LN) by mechanical disruption and CD4+ T cells enriched by magnetic depletion. Cells were incubated for 20 min with biotin-anti B220+, biotin anti-CD11b+, biotin anti-CD11c+ and biotin anti-CD8a+ antibodies (eBioSciences), rinsed and then incubated with anti-biotin microbeads (MACS; Miltenyi). Labeled cells were depleted using Miltenyi LS columns and magnets.The antibody cocktail used for FACS of CD4+ T cells and iNKT cells included PE-conjugated CD1d-aGalCer tetramer, PerCP Cy5.5 anti-CD62L, PE Cy7 anti-CD3, APC Cy7 anti- CD44, eFluor450 anti-CD4. Cells were sorted using a BD FACS Aria II at the Garvan Institute Flow Cytometry facility. Purity of all samples exceeded 95% for all populations.For EAE experiments, 2.5 9 105 2D2.Tg CD4+ T cells were transferred into CD45.1 recipient mice by i.v. injection and EAE induced by immunization as described, on the same day.For analysis of proliferation, cells were labeled with Cell TraceTM Violet (CTV; Invitrogen Molecular Probes), Myelin Oligodendrocyte Glycoprotein 35-55 counted and 2D2.Tg Gpr65+/+ and 2D2.Tg Gpr65gfp/gfp cells mixed to achieve equal numbers of 2D2.Tg CD4+ T cells. Each WT.CD45.1 recipient mouse received a total of 5 9 105 2D2.Tg CD4+ T cells. Recipients were simultaneously immunized to induce suboptimal EAE, as described above (+MOG35-55) and one mouse was left as an unimmunized control (-MOG35-55).