Maresin-1 promotes neuroprotection and modulates metabolic and inflammatory responses in disease-associated cell types in preclinical models of multiple sclerosis

Abstract

Multiple sclerosis (MS) is a prevalent inflammatory neurodegenerative disease in young people, causing neurological abnormalities and impairment. To investigate a novel therapeutic agent for MS, we observed the impact of maresin 1 (MaR1) on disease progression in a well-known, relapsing-remitting experimental autoimmune encephalomyelitis mouse model. Treatment with MaR1 accelerated inflammation resolution, reduced neurological impairment, and delayed disease development by reducing immune cell infiltration (CD4+IL-17+ and CD4+IFNγ+) into the central nervous system. Furthermore, MaR1 administration enhanced IL-10 production, primarily in macrophages and CD4+ cells. However, neutralizing IL-10 with an anti-IL-10 antibody eliminated the protective impact by MaR1 in relapsing-remitting experimental autoimmune encephalomyelitis model, implying the significance of IL-10 in MaR1 treatment. Metabolism has been recognized as a critical mediator of effector activity in many types of immune cells. In our investigation, MaR1 administration significantly repaired metabolic dysregulation in CD4+ cells, macrophages, and microglia in EAE mice. Furthermore, MaR1 treatment restored defective efferocytosis in treated macrophages and microglia. MaR1 also preserved myelin in EAE mice and regulated O4+ oligodendrocyte metabolism by reversing metabolic dysregulation via increased mitochondrial activity and decreased glycolysis. Overall, in a preclinical MS animal model, MaR1 therapy has anti-inflammatory and neuroprotective properties. It also induced metabolic reprogramming in disease-associated cell types, increased efferocytosis, and maintained myelination. Moreover, our data on patient-derived peripheral blood mononuclear cells substantiated the protective role of MaR1, expanding the therapeutic spectrum of specialized proresolving lipid mediators. Altogether, these findings suggest the potential of MaR1 as a novel therapeutic agent for MS and other autoimmune diseases.

Keywords: DHA; EAE; IL-10; MS; Maresin1; Metabolism; SCENITH; SPM; inflammation; resolution; therapeutics.

Figure 1

Protective effect of MaR1 treatment on neurological deficits in the RR-EAE mouse model.A, clinical scores of SJL mice in the CFA, EAE, and EAE groups treated with MaR1 for 70 days after disease induction. EAE mice developed disease symptoms beginning on day 9 after immunization with PLP_139-151, whereas MaR1-treated mice showed symptoms on day 10 (n = 10). B, maximum score. C, cumulative score. D, incidence of clinical symptoms. E, disease severity scores of EAE- and MaR1-treated mice. ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 (as determined by the Mann‒Whitney test) versus CFA and the MaR1 EAE. The data are shown as the mean ± SEM. For spontaneous motor activity, before the end of the experiment, mouse locomotor activity was measured during the day and nighttime to evaluate disease severity. F, average diurnal horizontal and vertical activities were measured in CFA-, EAE-, and MaR1-treated mice. G, average nocturnal horizontal and vertical activity was measured in CFA-, EAE-, and MaR1-treated mice. H, average hourly horizontal and vertical activity (diurnal and nocturnal) in CFA-, EAE-, and MaR1-treated mice. The values are expressed as the means ± SDs (N = 10). Statistical analyses were performed with one-way ANOVA and two-way ANOVA. ∗p < 0.05, ∗∗p < 0.01 versus the CFA group; #p < 0.05, ##p < 0.01 versus the MaR1-treated group.

Figure 2

MaR1 modulates antigen-specific responses and abrogates the infiltration of IL-17- and IFNγ-producing CD4+ T cells.A, the antigen recall response of spleen/LN cells was examined on day 17 after immunization for 72 h in the presence of PLP_139-151 (n = 4). B–D, CNS tissues (brain and spinal cord) were processed, and the total number of leukocytes (CD45+) and CD4+ T cells and the frequencies of Th1 and Th17 cells in the CNS in treated and vehicle-treated RR-EAE mice (n = 4) were examined. E, infiltrating myeloid cells, including monocytic DCs, F4/80+ macrophages, and monocytes, were profiled in both treated and untreated EAE mice (N = 4). F and G, at the peak of the disease, the pro-inflammatory (class II and CD38) and anti-inflammatory (EGR2 and CD206) phenotypes of splenic macrophages (CD11b⁺F4/80⁺) were examined by flow cytometry, and the data are presented as a bar graph of the mean fluorescence intensity (MFI) (n = 4). H, the ratio of EGR2+/CD38+ macrophages was plotted to determine the macrophage phenotype (n = 4). I, the level of neurofilament-light chain (NFL) in the plasma of EAE model mice treated with or without MaR1 was examined via SIMOA (n = 5). J and K, spinal cord sections showing inflammatory infiltrates (H&E) and demyelination (LFB) (N = 5). Representative images showing histopathological changes in spinal cord tissue from EAE mice treated with MaR1 or vehicle. The circles indicate infiltration of inflammatory cells in (J) and demyelination of the nerves in (K). The percentage of demyelinated area was calculated as per previous publications. Scale bar represents 200 μm. The data are shown as the mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 3

MaR1 treatment impedes the encephalitogenic property of CD4 effector cells.A, EAE was induced in SJL mice, and MaR1 treatment began when the mice in the randomly divided groups had recovered from the disease on day 22 after immunization. B, at the end of the study (∼38 days), isolated LN cells from both groups were cultured with PLP_139-151, anti-IFNy (10 μg/ml), IL-12p70, and IL-23 (10 ng/ml). After 3 days, the enriched CD4+ cells were injected into SJL mice (n = 5), after which the clinical score was monitored. Note: The Adt-EAE-MaR1 group was not treated with MaR1. ∗∗p < 0.01.

Figure 4

Effect of IL-10 neutralization on the protective effect of MaR1 on EAE.A and B, spleen cells from EAE mice treated with or without MaR1 (day 18) were stimulated with PMA/ionomycin in the presence of GolgiPlug for 4 h and subjected to surface staining for CD4 (CD3+CD4+), CD8 (CD3+CD4+), B cells (CD3-CD19+), mDCs (CD11b + CD11c + Class II+) and macrophages (F4/80+), and intracellular staining for IL-10. CD45+IL-10+ cells were gated from live cells; on the basis of surface markers, all cell types expressing IL-10 were profiled (n = 4). C, plasma levels of IL-10 were measured via the SIMOA on day 70 in EAE mice treated with MaR1 or vehicle (n = 4). D, disease severity plot showing abrogation of the protective effect of MaR1 by an IL-10 neutralizing antibody compared with that of an IgG neutralizing antibody (N = 8). E i-ii, the number of CNS-infiltrating Th17 cells in all groups was determined via flow cytometry; the data are presented as a bar graph (n = 5). ∗p < 0.05 compared with IgG group; ∗∗∗∗p < 0.00001 compared with MaR1 IgG group. F, adoptive transfer of antigen-specific CD4+ T cells derived from the IL-10 neutralization experimental batch and monitoring of clinical scores (N = 5). ∗p < 0.05 versus vehicle EAE. The data are shown as the mean ± SEM. ∗∗p < 0.01, ∗∗∗p < 0.001 compared with the EAE group.

Figure 5

MaR1 induces metabolic reprogramming in CD4+ T cells and macrophages of EAE mice.A, an XF mitochondrial stress test was performed on CD4+CD25 cells (purity ∼95%) isolated from CFA-, EAE-, and MaR1-treated EAE mice on day 18 postimmunization. The maximal respiration is presented as a bar graph (N = 6). B, compensatory glycolysis was examined using an XF Seahorse bioanalyzer and is presented as a bar graph (N = 6). C, the bioenergetic profile depicts the metabolic state of CD4+ cells isolated from the various groups described in (A). D, ATP levels detected in CD4+ cells from the various groups in (A) using an ATP assay kit (N = 4). NS, not significant compared with the CFA and EAE groups; Student’s t test was used. E, at the peak of the disease, brain infiltrating cells (BILs) were isolated from all groups using a Percoll gradient and processed for SCENITH. MFI of puromycin across samples treated with different inhibitors of CD4+ T cells (CD45⁺CD4⁺), including deoxyglucose (DG), oligomycin (OM), or deoxyglucose + oligomycin (DGO) (N = 3). Metabolic perturbations in infiltrating CD4+ cells from all mouse groups are shown as a bar graph. F–I, an XF mitochondrial stress test was performed on splenic F4/80+ macrophages (purity ∼95%) isolated from CFA-, EAE-, and MaR1-treated EAE mice. The maximal respiration, basal glycolysis, bioenergetic profile, and total ATP levels were detected in macrophages from all the mouse groups, as described above in detail. J, using SCENITH, metabolic perturbation of infiltrating macrophages (CD45⁺CD11b⁺F4/80⁺) from all the mouse groups was detected as described above. The data are shown as the mean ± SEM (N = 3). NS, not significant; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001 compared with CFA-treated or EAE mice, as determined via Student’s t test.

Figure 6

MaR1 induces efferocytosis in macrophages.Ai, bone marrow–derived macrophages were treated with MaR1 (100 nM) or vehicle (0.1% EtOH) for 1 h, after which CFSE-labeled 70 to 75% apoptotic splenic cells devoid of monocytes/macrophages were added at a ratio of 1:5. After 18 h, the cells were washed and stained for F4/80, after which the number of F4/80 cells engulfing CFSE-labeled cells was quantified (n = 3). The data are presented as a bar graph and a representative flow plot. Aii, another set of cells was seeded on coverslips, and an in vitro efferocytosis assay was performed as described above. After 18 h, the cells were washed and fixed, and Z-stack images were taken with an Olympus FV1000 confocal microscope with a 40 × objective. Four consecutive images with an interspace of 1 μm were captured. The mean intensity of apoptotic cells was quantified with ImageJ analysis software (version 1.49; NIH) (N = 6). B, under efferocytosis conditions, the expression of IL-10 and TGFβ was examined by qPCR, and the data were normalized to those of the housekeeping gene L27 (N = 3). C, for efferocytosis in the spinal cords of the EAE- and MaR1-treated groups, macrophages (CD68+) were stained with a mouse anti-CD68 antibody and polyclonal degraded myelin (Millipore). The images were captured by LSCM at a 1-ary unit aperture at 60x resolution (N = 5). The data are shown as the mean ± SEM. ∗p < 0.001, ∗∗p < 0.05, ∗∗∗p < 0.01.

Figure 7

MaR1 promotes an anti-inflammatory phenotype in microglia.A, to examine microglial morphology in response to MaR1 treatment, lumbar spinal cord sections from the EAE- and MaR1-treated groups were stained with Iba1, and immunofluorescence images were threshold filtered, binarized, and analyzed via ImageJ/FIJI software for skeletal analysis (n = 5). Scale bar represents 100 μm. B, quantification of Iba1-immunoreactivity–positive cells is presented in a bar graph (n = 5). C, at the peak of the disease, after ∼18 days, brain infiltrating leukocytes (BILs) from all groups (EAE- and MaR1-treated) were processed for pro- and anti-inflammatory marker detection via flow cytometry (n = 3). D, bar graph of the EGR2/CD38 and CD206/class II ratios. E, primary microglia were treated with MaR1 (100 nM) or vehicle (0.1% EtOH) for 1 h, after which CFSE-labeled 70 to 75% apoptotic splenic cells devoid of monocytes/macrophages were added at a ratio of 1:5. After 18 h, the cells were washed and stained for CD11b, after which the number of CD11b + microglia engulfing CFSE-labeled cells was quantified (n = 3). The data are presented as a bar graph. F, changes in microglial metabolism in all groups were evaluated via SCENITH, and the data are presented in a bar graph as the mean ± SEM (n = 3). NS, not significant; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with EAE- versus MaR1-treated EAE.

Figure 8

MaR1 protects myelin possibly by regulating the metabolic fitness of oligodendrocytes during EAE.A, serial coronal brain sections from CFA-, EAE-, and MaR1-treated EAE mice on day 18 postimmunization were stained with fluoromyelin, and imaging was performed via laser scanning confocal microscopy. The intensity of myelin in the corpus callosum (CC) above the lateral ventricle was measured with ImageJ (labeled LV) (N = 3). B, at the peak of the disease, single suspensions of CNS tissues were processed via a Percoll gradient from all groups (CFA, EAE, and MaR1-treated), and the isolated cells were processed for the detection of intracellular reactive oxygen and nitrogen species (ROS and RNS) levels in O4+ oligodendrocytes (CD45^-O4⁺), which were evaluated via the Cellular ROS/RNS Detection Assay Kit via flow cytometry (n = 5). C and D, puromycin incorporation (MFI values indicated) in O4+ oligodendrocytes (CD45^-O4⁺) in EAE- and MaR1-treated EAE mice. Metabolic changes, including changes in mitochondria, glycolysis, and fatty acid oxidation (FAO), in O4⁺ oligodendrocytes in all groups were calculated via SCENITH (N = 3) and are shown as a bar graph. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with CFA or EAE.

Figure 9

MaR1 rescues EAE-induced spinal cord transcriptomic alterations.A, Multidimensional scaling (MDS) plot showing the segregation of each group: CFA, EAE, and MaR1. B, heatmap representing significant gene expression changes across the groups. Padj>0.05. C, Venn diagram representing the shared number of upregulated and downregulated genes in MaR1-treated EAE mice. D, gene ontology (GO) enrichment of common genes upregulated by EAE and downregulated by MaR1 treatment. FDR<0.05. E, gene ontology (GO) enrichment of common genes downregulated by EAE and upregulated by MaR1 treatment. FDR<0.05.

Figure 10

MaR1 reduces cytokine responses in activated human T-cell subsets in patients with relapsing-remitting (RR)-MS.A and D, peripheral blood mononuclear cells (1 × 10⁶ cells per well) were left untreated or treated with vehicle or MaR1 (10 nM) for 30 min (N = 8). The cells were then stimulated with anti-CD3/CD28 for 8 h, stained both at the cell surface and intracellularly, and analyzed by flow cytometry. B and C, cytofluorimetric plots and percentages of intracellular IFN-γ and IL-17 produced by CD4⁺ T cells and of TNF-α and IFN-γ produced by CD8⁺ T cells. E, cytofluorometric plots and percentages of intracellular FoxP3 and IL-10 in Tregs. NS not significant, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.