c-Met is expressed by highly autoreactive encephalitogenic CD8+ cells

Abstract

Background: CD8⁺ T lymphocytes are critical mediators of neuroinflammatory diseases. Understanding the mechanisms that govern the function of this T cell population is crucial to better understanding central nervous system autoimmune disease pathology. We recently identified a novel population of highly cytotoxic c-Met-expressing CD8⁺ T lymphocytes and found that hepatocyte growth factor (HGF) limits effective murine cytotoxic T cell responses in cancer models. Here, we examined the role of c-Met-expressing CD8⁺ T cells by using a MOG_35-55 T cell-mediated EAE model.

Methods: Mice were subcutaneously immunized with myelin oligodendrocyte glycoprotein peptide (MOG)_35-55 in complete Freund's adjuvant (CFA). Peripheral and CNS inflammation was evaluated at peak disease and chronic phase, and c-Met expression by CD8 was evaluated by flow cytometry and immunofluorescence. Molecular, cellular, and killing function analysis were performed by real-time PCR, ELISA, flow cytometry, and killing assay.

Results: In the present study, we observed that a fraction of murine effector CD8⁺ T cells expressed c-Met receptor (c-Met⁺CD8⁺) in an experimental autoimmune encephalitis (EAE) model. Phenotypic and functional analysis of c-Met⁺CD8⁺ T cells revealed that they recognize the encephalitogenic epitope myelin oligodendrocyte glycoprotein_37-50. We demonstrated that this T cell population produces higher levels of interferon-γ and granzyme B ex vivo and that HGF directly restrains the cytolytic function of c-Met⁺CD8⁺ T cells in cell-mediated cytotoxicity reactions CONCLUSIONS: Altogether, our findings suggest that the HGF/c-Met pathway could be exploited to modulate CD8⁺ T cell-mediated neuroinflammation.

Keywords: CD8+ T cell; EAE; HGF; MS; Neuroinflammation; c-Met.

Fig. 1

A subpopulation of effector CD8⁺ T cells expresses the c-Met receptor in EAE. MOG_35–55 gating strategy for the identification of c-Met-expressing CD8⁺ T cells. To characterize the frequency of CD8⁺ T cells that express c-Met, live lymphocyte gated in forward (FS Lin) versus side scatter (SS Lin) dot plots were selected, followed by a subsequent gate on live (7AAD^-) cells (not shown). CD8⁺ T lymphocytes were subsequently gated (CD45⁺CD3⁺CD8⁺ T cells). Numbers in the quadrants of the density plots (CD8 vs. c-Met) at acute phase (d14) indicate the percentage of the population CD8⁺c-Met⁺. Frequency values reflect results above the fluorescence values obtained by isotype control antibodies (a). Splenocytes, lymph node cells, or CNS mononuclear cells from MOG_35–55-immunized mice were harvested and stained with CD8-APC and c-Met-FITC and analyzed by flow cytometry. Data are gated on live CD8⁺ T cells. The frequency of c-Met⁺CD8⁺ cells in each organ is depicted (b top panel), as is the absolute number of c-Met⁺CD8⁺ cells (b lower panel), based on the counts of the total number of cells obtained. Results are the averages of at least n = 5 mice per group per time point (representative of 3 independent experiments). c CD8⁺ T cells from CNS were examined by flow cytometry for expression of the activation markers CD44 and CD62L (CD44^lowCD62L^high naive CD8⁺ T cells; CD44^highCD62L^low activated memory CD8⁺ T cells). Bars represent the mean ± standard error of values for n = 5 mice/group; *p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001 by Student’s t test. d Representative double immunofluorescent staining images for CD8 (red) and c-Met (green) of cytospin of spleen cells at peak disease. Cell nuclei were counterstained with DAPI (blue). Scale bar, 20 μm. Insets are magnified × 4.5 from the original images. The arrows indicate the area where we observed co-expression of both c-Met and CD8 markers. Comparable results were obtained with three other mice

Fig. 2

Phenotypic characterization of murine c-Met⁺ CD8⁺ T cells ex vivo in MOG-induced EAE. Splenocytes were isolated from MOG peptide-immunized mice (mean score 2.5–3.5). Cells were cultured in the presence of MOG_35–55 or MOG_37–50 or with gp100_25–33 peptide control for 72 h. CD8⁺ T cells were examined by flow cytometry to quantify the frequency of Ki67-positive CD8⁺ T cells, which was determined by intracellular staining. Representative histograms are shown including quantification (a, b). Bars represent the mean ± standard error of values for n = 5 mice/group. ****p ≤ 0.0001 by two-way ANOVAs followed by Tukey’s post hoc test

Fig. 3

Comparison of phenotype of CTLs located in the spleen and CNS in EAE. At day 14 post-EAE induction, CD8 T cells from the spleen and from CNS-infiltrating cells were isolated and incubated with irradiated syngeneic splenocytes loaded with 1 μM of peptide (MOG_35–55, MOG_37–50, or gp100_25–33 control peptide) for 48 h in the presence of 20 pg/ml IL-2. Granzyme B (a), IFNγ (b), and TNF (c) levels were quantified by ELISA in the cell supernatant of stimulated inflammatory T cells. n = 5 mice/group; data shown as mean ± standard error. *p ≤ 0.05, **p ≤ 0.01, ****p ≤ 0.0001 (two-way ANOVAs followed by Tukey’s post hoc test)

Fig. 4

Ex vivo gene and protein characterization of c-Met⁺ versus c-Met^-CD8⁺ T cells. a Relative expression of EOMES, RUNX3, FOXP3, GATA3, and T-bet mRNA in FACS-sorted spleen cells of peak disease c-Met⁺ CD3⁺ CD8⁺ vs. c-Met⁻ CD3⁺ CD8⁺ T from MOG_35–55-induced EAE mice, as measured by quantitative RT-PCR, representative of n = 3 mice; ∗p ≤ 0.05 (unpaired, 2-tailed Student’s t test). b Representative histogram and average MFI (bar graph) of Runx3 and Eomes in EAE c-Met⁺ vs. c-Met⁻ CD3⁺CD8⁺ T cells, representative of n = 5 mice. c Eomes and Runx3 protein levels in spleen cells at peak disease (d14) c-Met⁺CD8⁺ vs. c-Met⁻ CD8⁺ T cells from MOG-induced EAE mice treated with either vehicle or HGF (30 ng/ml for 48 h) in the presence of IL-2; representative histograms are shown. Measurement of Eomes and Runx3 gMFI compared to isotype control of each condition (b, c). Data are representative of n = 5 mice. *p ≤ 0.05, **p ≤ 0.01 (unpaired, 2-tailed Student’s t test for b). Data from c are presented as mean ± SEM and were analyzed using the two-way ANOVA followed by Tukey post hoc comparisons. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001

Fig. 5

HGF modulates the phenotype and function of c-Met⁺CD8⁺ T cells. a Cytolytic activity of CD8⁺ obtained from MOG_35–55-induced EAE (score 2.5–3.5) stimulated ex vivo with MOG_37–50 peptide for 48 h. Various effector/target (E/T) ratios were tested for killing of syngeneic B cells pulsed with MOG_37–50 peptide (squares) or control gp100_25–35 peptide (circles). Values represent the mean ± standard error of n = 3 mice per group, carried out in triplicate. Representative histogram and average MFI (bar graph) of granzyme B (b) and IFNγ (c) of c-Met⁺ vs. c-Met⁻ CD3⁺CD8⁺ T cells from EAE mice (d14) treated ex vivo with HGF (30 ng/ml for 48 h) or untreated as vehicle; n = 5 mice/group. Data are presented as mean ± SEM and were analyzed using the two-way ANOVA followed by Tukey's post hoc comparisons. *p ≤ 0.05, **p ≤ 0.01