Female C57BL/6 mice of specific pathogen-free (SPF) level, aged 6–8 weeks and weighing 19–21 g, were purchased from Jinan Pengyue Laboratory Animal Breeding Co., Ltd. (Ji nan, China). Each animal was bred under SPF conditions with a humidity of 60%-80% and an environmental temperature of 22 to 26 ℃ and had ad libitum access to food and water. All animals used in the study were treated in accordance with the guidelines set forth by the Institutional Animal Care and Use Committee of Binzhou Medical University Hospital.

Five milligrams of MOG35-55 polypeptide (GenScript, New Jersey, USA) was added to 2.5 ml of PBS at a final concentration of 2 mg/ml. Then, 17.5mg of heat-killed Mycobacterium tuberculosis ( Becton, Dickinson and Company Sparks, USA) was dissolved in 2.5 ml of complete Freund's adjuvant (CFA, Sigma-Aldrich, Missouri, USA), for a final concentration of 5 mg/ml. The two solutions were thoroughly mixed at a ratio of 1:1. Then, the antigen was fully emulsified on ice to form a water-in-oil emulsion. Each mice was immunized subcutaneously on the back with 0.2 ml of polypeptide. On Day 0 and Day 2 after immunization, the mice received intraperitoneal injections of pertussis toxin (PTX, Merck Millipore Corp, USA) at a dose of 300 ng per injection.

Clinical signs were evaluated daily using an established scoring system ranging from no clinical signs to tetraplegia or moribund by EAE: 0, no clinical signs; 0.5, limp distal tail; 1, paralyzed or staggering tail; 1.5, limp tail and hind limb weakness; 2, mild paresis of both hind limbs or severe paralysis of one hind limb; 3, complete paralysis of both hind limbs; 4, paralyzed limbs and forelimbs; and 5, tetraplegia or moribund by EAE. The intermediate clinical signs in between the major clinical signs were scored as 0.5 points.

All C57BL/6 mice were randomized assigned into 3 groups: (A) the control group, normal mice; (B) EAE group, saline solution-treated EAE mice; and (C) EAE + MP group, MP-treated EAE mice. Treatment with MP (Sinopharm Ronshyn Pharmaceutical Co., Ltd.) was administered once every day from the 9th day (early onset of EAE) to the end of the experiment. The dose was gradually reduced over 9 days, starting at 100 mg/kg for 3 days, followed by 50 mg/kg for another 3 days, and finally 25 mg/kg for the remaining 3 days. Intraperitoneal injections were stopped on the 18th day of the experiment. Control mice received equal volumes of saline solution.

Histopathological examination was performed on Day19 post-immunization). The spinal cord tissues were meticulously dissected from the mice, fixed in 4% paraformaldehyde, dehydrated, paraffin-embedded, and sectioned into 6 μm-thick sections. Subsequently, spinal cord sections were stained not only with hematoxylin and eosin (H&E) (Solebo Biotech, Beijing, China) to evaluate inflammation but also with luxol fast blue (LFB) (Solebo Biotech, Beijing, China) to assess demyelination. These sections were then scanned and imaged with a microscope (Olympus, Japan), and the inflammatory cell infiltration and demyelination of the spinal cord in the mice were analysed in a blinded fashion. The inflammation scale was determined based on the presence of inflammatory cells: 0 indicated no inflammatory cells, 1 denoted a few scattered inflammatory cells, 2 represented organized infiltration around blood vessels, and 3 signified extensive perivascular cuffing with extension into the surrounding tissue. Demyelination assessment in the spinal cord followed this scale: 0 indicated the absence of demyelination, 1 represented rare focal areas, 2 denoted multiple demyelinated regions, and 3 signified large areas exhibiting demyelination [32, 33].

For cell phenotype analysis, anti-CD4-FITC, anti-CXCR5-percp-cy5.5, anti-Foxp3-PE, anti-ICOS-APC, anti-PD-1-APC antibodies as well as the corresponding IgG isotypes were obtained from BioLegend (San Diego, CA, USA).

For cell surface staining, single cells from the spleens of EAE mice treated with saline solution or MP were incubated with fluorescent monoclonal antibodies at 4 ℃ for 30 min at the optimal dilutions. After cell surface staining, a Foxp3 Staining Buffer Set (Invitrogen) for intracellular staining of cells was used to fix and permeabilize the cells at room temperature for 1h. After the incubation period, the cells were washed twice with PBS supplemented with 2% foetal bovine serum. We determined the panels according to isotype control (Supplementary Fig. 1) and simple staining (Supplementary Fig. 2) of ICOS, PD-1, and Foxp3. Flow cytometry analysis was conducted using a FACS Calibur flow cytometer (Beckman Coulter). The obtained data were analysed utilizing FlowJo 10.0 software.

Mice were anaesthetized and sequentially transcardially perfused with cold PBS and 4% paraformaldehyde (PFA) after which the spinal cord and spleen were removed. The tissues were fixed in 4% paraformaldehyde for 24 h, and then dehydrated in 30% sucrose for 2 days.The sections were obtained using a freezing microtome (Leica, CM1950). For fluorescence staining, the sections were permeabilized with PBS containing 0.4% Triton X-100 (BOSTER, China). Subsequently, the spinal cord/spleen sections were blocked with 5% bovine serum albumin (BSA, BOSTER; China) at room temperature for one hour. Next, the sections were incubated overnight at 4 °C with primary antibodies against GFAP (Cell Signaling Technology), Iba-1 (Cell Signaling Technology), CD4 (Santa Cruz), CXCR5 (BioLegend), and Foxp3 (ABclonal). The sections were then incubated with Alexa Fluor 488-conjugated secondary antibodies (Cell Signaling Technology), Goat anti-Rabbit; Alexa Fluor 546, goat anti-rat (Cell Signaling Technology); Alexa Fluor 647, goat anti-mouse (Santa Cruz) at room temperature for one hour. Next, the nuclei were stained with DAPI before being imaged. Images were captured with an Olympus FV10 confocal microscope (Olympus, Japan).

Spinal cord tissues were dissected from each group. For the EAE group, the spinal cord samples were carefully dissected on Day 9 post-immunization (preimmune to EAE, Pre-EAE group) and Day19 postimmunization (peak of EAE, P-EAE group) and frozen immediately in liquid nitrogen. The spinal cord was subjected to protein extraction using RIPA lysis buffer supplemented with phosphatase and protease inhibitors. The concentration of proteins was determined by employing the BCA assay (BOSTER, China). Protein samples were added to each well for electrophoresis separation via SDS-PAGE (80 V, 0.5 h and then 120 V, 1.5 h). The separated protein bands were transferred to 0.45 μm PVDF membranes (Servicebio, China). The membranes were blocked in a solution containing TBST with 5% skim milk for one hour and then incubated respectively with anti-PI3K (Cell Signaling Technology, 34,050), anti-phospho-AKT (Affinity, AF0016), anti-AKT (Affinity, AF6261), anti-phospho-FoxO1 (Cell Signaling Technology, 9464), anti-FoxO1(Affinity, AF3417), anti-phospho-mTOR (Cell Signaling Technology, 5536), anti-mTOR (Cell Signaling Technology, 2983), CXCR5 (ABclonal, A8950), Foxp3 (Affinity, AF6544), ICOS (ABclonal, A1811), PD-1 (Cell Signaling Technology, 84,651), TGF-β1 (Proteintech, 21,898–1-AP), IL-21 (Affinity, DF4818), GFAP (Cell Signaling Technology, 3670), Iba-1 (Cell Signaling Technology, 17,198), Arg-1 (Proteintech,16,001–1-AP), iNOS (Proteintech, 18985–1-AP) and GAPDH (BOSTER, BM3874) antibodies overnight at 4 ℃. The membranes were incubated with an anti-rabbit antibody conjugated with HRP (Boster, BA1054) for one hour. Finally, ECL reagent (Boster, China) was added, and the protein bands were detected by a chemiluminescence imaging system (LI-COR Biosciences). The band density was quantified using image J.

The data for each group are presented as the mean ± standard error of the mean(SEM). Statistical analyses were conducted using GraphPad Prism, V8.0 (GraphPad Software Inc., San Diego, USA). To calculate differences between two groups, the statistical Student’s t test was used for statistical analysis, while for comparisons among multiple groups, one-way analysis of variance (ANOVA) was used, followed by Bonferroni multiple comparison tests. In the MOG-EAE experiments, the statistical significance of differences in neurological scores was determined using two-way ANOVA. Significant differences between or among groups were assessed accordingly: * P < 0.05, ** P < 0.01, and *** P < 0.001.