Healthy female/male C57BL/6 mice (17–20 g, 6–8 weeks old) were purchased from Vital River Laboratory Animal Technology Co., Beijing. China (certificate number: 20210520Abzz 0619000101). The animals were maintained in a specific-pathogen-free room at a temperature of 22°C ± 3°C and humidity of 55% ± 5% under a 12 h light/dark cycle. All animal experiments and protocols were conducted in strict accordance with the relevant regulations. The study protocol was approved by the Ethics Committee of Animal Experiments of Shanghai University of Traditional Chinese Medicine (approval number: PZSHUTCM 201023015).

After 1 week of adaptive feeding, female C57BL/6 mice were immunized at three sites on the back (the middle of both shoulders and upper the groin on both sides) with 4 mg/mL heat-killed Mycobacterium tuberculosis H37Ra (#231141; BD, Franklin Lakes, NJ, United States) with 200 μg of myelin oligodendrocyte glycoprotein (MOG) 35-55 (#T510219-0001; Sangon Biotech, Shanghai, China) in 200 μL of emulsion containing 50% complete Freund’s adjuvant (#F5506; Sigma, St. Louis, MI, United States). The mice were then intraperitoneally injected with 400 ng pertussis toxin (#GC17532; GLPBIO, Montclair, CA, United States) in phosphate buffered saline (PBS) at 0 and 2 days post immunization (dpi). All animals were weighed and scored daily in a blinded manner using Weaver’s 15-point scoring system (27, 28). In terms of tail scores, a score of 0 indicated no clinical signs, a score of 1 indicated reduced tail tension, and a score of 2 indicated tail paralysis. In terms of limb scores, a score of 0 indicated no clinical signs, a score of 1 indicated waddling gait, a score of 2 indicated mild limb paralysis (limb dragging while walking), and a score of 3 indicated limb paralysis. The tail and limb scores were then accumulated, and a score of 15 represented death. For therapeutic administration experiments, mice were given individualized treatment, that is, mice were treated only after they developed clinical symptoms (onset, 10–14 dpi). Mice that did not onset until 15 dpi after modeling were not included in this study.

UAOS-Na, batch number: 20200706, prepared by the Shanghai University of Traditional Chinese Medicine, and the voucher specimen (No. UAOS-Na20200706) was deposited at the Department of TCM Chemistry, Pharmacy of Shanghai University of Traditional Chinese Medicine (Shanghai, China). In our previous work, UAOS-Na was synthesized by forming a C-3 hydroxyl ester of UA with succinic acid and then alkalizing to a salt, and its structure was confirmed by ¹H-NMR, ¹³C-NMR and LCMS (ESI) (Figure 1A). The water solubility of UAOS-Na (577.58 mg/L) was determined by using ultraviolet-visible spectrophotometry and was significantly higher than that of UA [<5.64 mg/L (29)]. In this study, UAOS-Na was dissolved in 0.5% sodium carboxymethyl cellulose (CMC-Na, #abs47051270; Absin Bioscience, Shanghai, China) in the comparative efficacy experiments with UA (#SLURA190502; Sanleng Biotech, Guilin, China). Otherwise, UAOS-Na was dissolved in ddH₂O.

For prophylactic administration experiments (Figure 1B), mice were randomly divided into a normal group (n = 6, 0.5% CMC-Na), an EAE group (n = 10, EAE induction, 0.5% CMC-Na), a UA group (n = 7, EAE induction, 23 mg/kg/d), and a UAOS-Na group (n = 7, EAE induction, 30 mg/kg/d, the molar concentration of UAOS-Na was consistent with UA). UA or UAOS-Na was prepared with 0.5% CMC-Na. Mice were treated with 0.5% CMC-Na, UA, or UAOS-Na by gavage daily from the beginning of modeling (0 dpi). Asymptomatic animals were included in the statistical analysis.

For therapeutic administration experiments (Figure 1B), mice were randomly divided into a normal group (n = 11, 0.5% CMC-Na), an EAE group (n = 10, EAE induction, 0.5% CMC-Na), a UA group (n = 7, EAE induction, 45 mg/kg/d), and UAOS-Na groups [EAE induction, 30 mg/kg/d (n = 9), 60 mg/kg/d (n = 10), 120 mg/kg/d (n = 8)]. UA or UAOS-Na was prepared with 0.5% CMC-Na. Mice were given individualized treatment, that is, mice were treated with 0.5% CMC-Na, UA, or UAOS-Na only after they developed clinical symptoms (onset, 10–14 dpi). Mice that did not onset until 15 dpi after modeling were considered asymptomatic animals and were not included in the statistical analysis.

Mice were anesthetized and sacrificed at 25 dpi and perfused with 4% paraformaldehyde (PFA; #G1101; Servicebio, Wuhan, China). Their serum, spinal cord, and spleen were harvested. The spleens were isolated to calculate the spleen index (SI) as follow: SI = spleen weight (mg)/body weight (g) × 10.

After 1 week of adaptive feeding, male C57BL/6 mice were randomly divided into a normal group (10 kcal% Fat Diets, Research Diets, United States) given ddH₂O, a cuprizone-induction group [CPZ group, 0.2% cuprizone (w/w), Research Diets] given ddH₂O, a UAOS-Na group [0.2% cuprizone (w/w), Research Diets] given UAOS-Na (60 mg/kg/d). These mice in CPZ and UAOS-Na groups were fed cuprizone for 6 weeks to achieve complete demyelination of corpus callosum. From the 2nd week, ddH₂O or UAOS-Na (60 mg/kg/d) was administered daily by gavage. At the 6th week, the mice were sacrificed, and brain were obtained.

Spleens were mechanically dissociated through a 70 μm cell strainer (#02036461; Titian, Shanghai, China) and incubated with red blood cell lysis buffer (#C3702; Beyotime, Shanghai, China) for 5 min. Harvested cells were washed with cold PBS and fractionated on lymphocyte separation reagent (#LSM01; Multisciences Biotech, Zhejiang, China) gradient by centrifugation at 400 × g for 30 min. Lymphocyte were obtained by collecting mononuclear cells (MNCs) from the middle layer and washing with PBS.

Bone marrow cells (BMC) were extracted from the femurs and tibiae of C57BL/6 mice under sterile conditions, lysed with red blood cell lysis buffer for 10 min, centrifuged at 400 × g at 4°C for 5 min, and washed twice with RPMI 1640 medium (#MA0215; Meilunbio, Dalian, China). Cells were resuspended in DCs complete culture medium consisting of granulocyte-macrophage colony-stimulating factor (#70343; CST, Massachusetts, United States, 20 ng/mL), interleukin-4 (#abs04697, Absin, 10 ng/mL), and RPMI 1640 complete medium, and adjusted to 1 × 10⁶ cells/mL. Cells were seeded into 6-well plate and cultured for 72 h. Adherent cells were retained, supplemented with fresh DCs complete medium, and cultured for 48 h. Half of the medium was replaced with DCs complete medium for an additional 48 h. On day 7, suspension cells, immature bone marrow-derived dendritic cells (imBMDC), were collected and reseeded in 6-well plate and stimulated with LPS (100 ng/mL) for 24 h to obtain mature mouse bone marrow-derived dendritic cells (mBMDC). In the UAOS-Na groups, different concentrations of UAOS-Na (0, 10, 20, and 40 μM) were added to the culture medium together with LPS (100 ng/mL) for 24 h.

MOG-specific mBMDC (MOG-mBMDC) was obtained by stimulating imBMDC with MOG35-55 (20 μg/mL) for 4 h followed by LPS (100 ng/mL) for 24 h and centrifuged at 400 × g at 4°C for 5 min. MOG-specific lymphocytes (MOG-lymphocytes) were obtained from EAE mouse spleen according to the splenic lymphocyte preparation protocol. MOG-mBMDC and MOG-lymphocytes were co-cultured at a concentration of 1:10 in DCs complete culture medium containing IL-2 (#abs04100; Absin, 10 ng/mL) for 48 h. Then, cell counting kit-8 (CCK-8) and flow cytometry were performed.

In the UAOS-Na groups, different concentrations of UAOS-Na (0, 10, 20, and 40 μM) were added to the culture medium together with LPS (100 ng/mL) for 24 h.

After anesthesia, mouse blood was collected by removing eyeballs. Serum was separated by centrifugation at 1,000 × g for 10 min. To evaluate the peripheral inflammation in mice, serum levels of interferon-γ (IFN-γ; #EK280; Multisciences Biotech), tumor necrosis factor α (TNF-α; #EK282; Multisciences Biotech), interleukin-17A (IL-17A, #EK217; Multisciences Biotech), interleukin-6 (IL-6, #EK206; Multisciences Biotech), and interleukin-10 (IL-10, #EK210; Multisciences Biotech) were determined using commercial enzyme-linked immunosorbent assay (ELISA) kits.

The spinal cords were harvested, fixed with 4% PFA, embedded in paraffin, and sliced into 4 μm-thick sections for pathological analysis. To determine the degree of inflammatory infiltration (MNCs infiltration), spinal cord sections were stained with hematoxylin and eosin (H&E, #G1005; Servicebio). Images were obtained with a 10× objective of Lecia microscope at a 100× magnification. The white matter in transverse spinal cord tissue sections was divided into the following parts, dorsal columns, spinocerebellar, spinothalamic, ventral corticospinal tract, and vestibulospinal, all parts were assessed and scored in a blind manner. Finally, the scores of each part were averaged. The scoring criteria for tissue inflammation were as follow: 0 = none, 1 = some scattered MNCs, 2 = MNCs infiltrating around blood vessels and forming “cuff-like changes” of blood vessel, 3 = numerous “cuff-like changes” were formed and MNCs infiltrated to adjacent tissues (24, 30). To determine the degree of demyelination, spinal cord sections were stained with luxol fast blue (LFB, #G1030; Servicebio). For semiquantitative methods of demyelination (LFB staining), images were obtained with a 10× objective of Lecia microscope at a 100× magnification, the spinal cord transection was divided into four parts, front, back, left, and right. These four visions were selected to calculate degree of LFB staining (myelination area) in each group and the average was taken as the final count. The degree of LFB staining was analyzed automatically using Image-Pro Plus 6.0 software.

The brains of CPZ mouse were harvested, fixed with 4% PFA, embedded in paraffin, and sliced into 4 μm-thick sections for pathological analysis. To determine the degree of corpus callosum demyelination, brain sections were stained with LFB, images were then obtained using a Lecia microscope and quantified using Image-Pro Plus 6.0 software.

Spinal cord of EAE or brain sections of CPZ mice were immunostaining with glial fibrillary acidic protein (GFAP, #80788; CST; 1:200) to evaluate astrocytic lesions. Images were then obtained using a Lecia microscope and quantified using Image-Pro Plus 6.0 software.

Transverse sections of spinal cord were cut. After antigen repair and endogenous peroxidase removal, slices were blocked in 3% bovine serum albumin (BSA, #MB4219; Meilunbio) for 1 h. Primary Abs, major histocompatibility complex class I (MHC-I, #NBP3-09017; NOVUS, Beijing, China; 1: 300) and CD8a (#GB114196; Servicebio; 1: 5000), incubations were conducted overnight at 4°C. HRP labeled secondary Abs incubations were conducted 50 min at room temperature, followed by washing with PBS 3 times. To amplify signal, slices were stained with TSA (#G1223; Servicebio) for 10 min at room temperature. After washing with PBS 3 times, nuclei were stained with DAPI (#G1012; Servicebio). Slides were covered with anti-fluorescence quenching tablet seal (#G1401; Servicebio). Images were then obtained using multispectral tissue slice scanning system and quantified using Image-Pro Plus 6.0 software.

Mice were deeply anesthetized and perfused with 4% PFA, 1.5% glutaraldehyde and 1 mM CaCl₂ in 0.1 M cacodylate buffer. Spinal cord sections were harvested and fixed in the same solution at 4°C for 24 h. Sections were washed, post-fixed with 1% OsO₄ in 0.1 M PBS (pH 7.4) for 2 h at room temperature, and subsequently dehydrated in graded ethanol series. Embedding was performed in epoxypropane and embedding agent. Sections, 1.0 μm thick, were cut, stained in toluidine blue (1%), and examined by light microscopy (CKX41, Olympus, Japan) to locate the spinothalamic tract. Ultrathin sections (60 nm) were cut, dyed in 2% uranium acetate saturated for 8 min in dark, and in 2.6% lead citrate solution for 8 min without CO₂. Then, sections were viewed and photographed with a transmission electron microscope (FEI Tecnai G2 spirit, Leica, Germany) operated at 80 kV. For each sample at 8200× magnification, four fields (up, down, left, and right) in the spinothalamic tract were obtained for statistical analysis. Quantification of the thickness (G-ratio = axon diameter/fiber diameter) and the number of myelinated fibers in the spinothalamic plexus were analyzed by Image-Pro Plus 6.0 software.

Total RNA was extracted from spinal cords or spleens using RNA isolater Total RNA Extraction Reagent (#R401-01, Vazyme, Nanjing, China) according to the manufacturer’s instructions. Reverse transcription was conducted using 5 × HiScript II qRT SuperMix II (#R223-01; Vazyme). Real-Time PCR was performed using the ChamQ Universal SYBR qPCR Master Mix (#Q711; Vazyme) according to the manufacturer’s instructions, and detection was performed using the ABI Prism^® 7500 Sequence Detection System (Applied Biosystems, Foster City, CA, United States). All data were normalized to an average of a housekeeping genes Gapdh. Gene relative expression was calculated by log2 of −ΔΔCt values from triplicate of PCR.

Genes	Primers (5′-3′)
M-GAPDH-F	TCTCCTGCGACTTCAACA
M-GAPDH-R	TGTAGCCGTATTCATTGTCA
M-H2-T23-F	GCCTCACAGATCTCTAAGCACA
M-H2-T23-R	CAGCACCTCAGGGTGACTTC
M-Rorc-F	CCTACTGAGGAGGACAGGGA
M-Rorc-R	GCAGGAGTAGGCCACATTACA
M-Tapbp-F	GTCTCCCTCGCGGACTAAAG
M-Tapbp-R	CCACACGGGGACATTCGTAA
M-T-bet-F	CCTTCTCACCTCTTCTATCC
M-T-bet-R	AACTCCGCTTCATAACTGT

The spinal cord was collected in accordance with standard procedures. Protein lysates were then obtained in RIPA lysis buffer (#P0013B; Beyotime) containing protease and phosphatase inhibitors. The concentration of these protein lysates was then measured using a Bicinchoninic Acid Protein Assay Kit (#P0012S; Beyotime). Equal amounts (3 μg/μL, 10 μL) of proteins in each sample were loaded onto 10% SDS-PAGE gels (#P0690; Beyotime) and transferred to polyvinylidene fluoride membranes (Millipore, Burlington, MA, United States). The membranes were then blocked with 5% BSA and probed with different primary antibodies: MHC-I (#ab281904, Abcam, Waltham, MA, United States), Tapbp (#66382; CST), GAPDH (#abs132004; Absin), Tubulin (#2148; CST), and anti-rabbit secondary antibody (#ab205718; Abcam). Blots were then exposed to an enhanced chemiluminescence detection reagent (#P0018S; Beyotime). Finally, signals were visualized using a Tanon imaging system and quantified using Image-Pro Plus 6.0 software.

For surface-marker staining, cells were incubated with fluorochrome-conjugated Abs to CD4 (#100405; Biolegend, California, United States), CD8a (#162304; Biolegend), CD3ε (#100312; Biolegend), CD11c (#117309, Biolegend), CD86 (#159203; Biolegend), and MHC-I (#114606; Biolegend) at the recommended dilution 30 min on ice. Flow cytometry analysis was performed on Beckman CytoFlex S (Beckman, California, United States) and data were analyzed with FlowJo software (Treestar, Ashland, Kentucky, United States).

Statistical analyses were performed using GraphPad Prism 8 software (GraphPad, La Jolla, CA, United States). The results are presented as mean ± SEM. Group comparisons were established by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test. Student’s t-test was used for small sample sizes, and the normality of data was analyzed using the Kolmogorov–Smirnov test. A p-values less than 0.05 was considered statistically significant.

For prophylactic administration, EAE mice were given UA (23 mg/kg/d) and UAOS-Na (30 mg/kg/d) for prophylactic until 25 dpi (Figure 1B). In EAE group, mice began to develop the disease (onset) at 10 dpi, with an average onset time of 11.78 ± 0.57 dpi and an incidence of 90.0% (9/10). UA and UAOS-Na prophylactic administration significantly delayed the average onset time to 13.50 ± 0.29 dpi and 14.33 ± 1.45 dpi and reduced the incidence to 57.1% (4/7) and 42.9% (3/7), respectively (Figure 1C). In addition, UAOS-Na significantly reduced the cumulative score and 25 dpi score and improved the symptoms of EAE mice comparing with normal group, but UA just had a weakening trend in the same experiment (Figure 1D). UAOS-Na may be more effective than UA in preventing EAE.

For therapeutic administration, UA (45 mg/kg/d) and different doses of UAOS-Na (30, 60, 120 mg/kg/d) were administered to EAE mice. UAOS-Na 60 mg/kg/d significantly reduced both the cumulative score and the 25-dpi score of EAE mice. Just like the result of prophylactic administration in Figure 1D, there was only significantly change between UAOS-Na-60 mg/kg/d group and normal group (Figure 1E). Moreover, the effects of UAOS-Na (60 mg/kg/d) and UA (45 mg/kg/d) on inflammation (H&E) and demyelination (LFB) of EAE spinal cords were compared. The results showed that UAOS-Na was more effective than UA in alleviating spinal cord inflammation and demyelination in EAE mice (Figure 2), and UAOS-Na 60 mg/kg/d was used in subsequent in vivo experiments.

The imbalance of serum pro-inflammatory and anti-inflammatory cytokines is closely related to EAE inflammation. Literatures have reported that inhibiting the expression of pro-inflammatory cytokines such as IFN-γ, TNF-α, IL-6, and IL-17 (31, 32) and promoting the expression of anti-inflammatory cytokines such as IL-10 (33, 34) can alleviate the inflammation of EAE. Moreover, small molecules targeting T-bet and RORγt also inhibit EAE (35). Therefore, the serum levels of cytokines were measured by ELISA and the spleen index of mice was recorded. The results showed that UAOS-Na (60 mg/kg/d) significantly reduced serum levels of the pro-inflammatory cytokines IFN-γ, IL-17A, TNF-α and IL-6, increased the serum level of anti-inflammatory cytokine IL-10 (Figure 3A), and reduced the spleen index (Figure 3B). In parallel, real-time PCR was used to detect the mRNA levels of T-bet and ROR-γt. UAOS-Na obviously reduced the mRNA levels of T-bet and ROR-γt of EAE spinal cord tissues (Figure 3C). Combined with the results in Figures 2A,B, UAOS-Na reduced peripheral and CNS inflammation in the acute stage of EAE mice.

To evaluate the effects of UAOS-Na on myelin sheath at the acute stage of EAE mice, LFB staining and transmission electron microscope for EAE spinal cord were performed. As shown in Figures 2C,D, the LFB staining in EAE group showed obvious demyelinating lesions in the cervical, thoracic, and lumbar spinal cords. In UAOS-Na group, demyelinating lesions were significantly reduced. Moreover, spinothalamic nerve cluster ultrastructure in the transverse section of lumbar spinal cord was measured by transmission electron microscope. As shown in Figures 4A–C, compared with normal group, the EAE group showed discrete stratification and edema of axon’s myelin sheath, as well as a reduction in the number of myelinated axons. The above data supported that UAOS-Na reduced axonal myelin injury in the acute stage of EAE mice.

UAOS-Na alleviated demyelination in EAE mice, which might be secondary to its immunomodulatory effect. To test whether UAOS-Na has a direct effect on myelin, we used the cuprizone-induced demyelination mouse model with minimal inflammation (Figure 4D). At 6th week after modeling, mouse brains were harvested for LFB staining to analyze the degree of demyelination in the corpus callosum. The results showed that CPZ group showed obvious demyelination compared with normal group. However, compared with CPZ group, demyelination was significantly alleviated in UAOS-Na group (Figures 4E,F). Therefore, UAOS-Na could directly protect the myelin sheath in cuprizone-induced mice.

The lesions of AST may be an important reason for the failure of remyelination. Studies have shown that reactive AST forms the physical barrier and secretes soluble factors to inhibit remyelination (36–38). So, we determined the effects of UAOS-Na on CNS astrogliosis in EAE mice and CPZ mice. Immunohistochemical staining of the lumbar spinal cord of EAE mice and the corpus callosum of CPZ mice showed deepened GFAP staining, massive reactive astrogliosis, and glial scar formation, which were reversed by UAOS-Na treatment (Figures 5A,C). And the data measured by Image-Pro showed that UAOS-Na treatment significantly reduced reactive astrogliosis and glial scar formation of EAE and CPZ mice (Figures 5B,D). Therefore, UAOS-Na ameliorated astrogliosis in the acute stage of EAE mice and prevented astrogliosis in CPZ mice.

Proteomic technique based on mass spectrometry has been widely used in searching for therapeutic targets of diseases (39, 40). In this study, tandem mass tag (TMT) technique was performed for proteomic analysis of the mouse spinal cord tissues (Figure 6A), and the data were presented in Supplementary Tables S1–S3. DEPs were defined as p < 0.05 and |fold change (FC)| >1.5. Compared with normal mice, 96 DEPs were enriched in EAE mice, of which 94 DEPs were upregulated. After UAOS-Na treatment, 16 DEPs were enriched with 15 DEPs downregulated compared with EAE group, and these DEPs were markedly enriched in phagosome, MAPK, autoimmune thyroid disease, and antigen processing and presentation (APP) signaling pathway, etc. (Figure 6D). Among them, there were few studies on APP signaling pathway, so we focus on it. Meanwhile, GO functional annotation analysis of DEPs in UAOS-Na vs. EAE group showed that the results were closely related to APP signaling pathway (Figure 6E). DEPs were enriched in vesicle membranes and participate in biological processes such as antigen processing and presentation, membrane localization and so on. And in the APP signaling pathway, DEPs in the MHC-I antigen presentation pathway were significantly up-regulated in EAE, while DEPs such as H2-T23 was significantly down-regulated after UAOS-Na treatment (Figure 6F). MHC-I antigen presentation pathway is related to the uptake and processing of antigens and the activation and proliferation of CD8 T cells. Overall, downregulation of H2-T23, the key proteins in MHC-I antigen presentation pathway, may be an important molecular mechanism of UAOS-Na in the treatment of EAE.

It has been shown that the TAP-related protein (Tapbp), plays a crucial role in MHC-I antigen presentation pathway. Tapbp is involved in stabilizing MHC-I molecules (41, 42), and increasing the assembly speed and number of MHC-I (43–45). Therefore, both H2-T23 and Tapbp were investigated in subsequent experiments. As shown in Figure 7A, Tapbp and H2-T23 mRNA expression levels in mouse spinal cord were assayed by Real-Time PCR. The results showed that Tapbp and H2-T23 mRNA levels were significantly increased in EAE group compared with normal group. However, UAOS-Na treatment markedly reduced these mRNA expression levels.

Previous studies have reported that restriction of Tapbp and H2-T23 expression resulted in decreased MHC-I expression on the cell membrane and inhibited CD8 T cell activation (46, 47). To analyze Tapbp and MHC-I protein expression levels in spinal cord tissues, western blotting was performed. The results revealed that Tapbp and MHC-I protein expression levels in EAE mouse spinal cords were higher than those in normal mouse spinal cords, and those protein expression levels were obviously decreased after UAOS-Na treatment (Figure 7B). In addition, to examine whether UAOS-Na treatment could reduce the CNS infiltration of CD8 T cells and the expression of MHC-I in EAE mouse spinal cord, we analyzed the MHC-I expression and the density of CD8⁺ T cells in mouse lumbar spinal cord by immunofluorescence staining. As shown in Figure 7C, compared with normal group, the density of CD8⁺ T cells and high MHC-I expression cells were higher in EAE group. UAOS-Na treatment reversed the results compared with EAE group (Figures 7D,E). Thus, UAOS-Na could inhibit the MHC-I expression and reduce the CD8 T cell infiltration in EAE mouse spinal cords. Furthermore, the proportion of CD8⁺ T cells in mouse spleen was detected by flow cytometry, and the results showed that UAOS-Na significantly reduced the proportion of CD8 T cells (CD3ε⁺, CD8a⁺) in splenocytes of EAE mice (Figures 7F,G). Overall, UAOS-Na reduced MHC-I expression in the spinal cord by inhibiting Tapbp and H2-T23 and reduced peripheral proportion and CNS infiltration of CD8 T cells.

We investigated the effect of UAOS-Na on mBMDC, an in vitro model of antigen-presenting cells, induced by C57BL/6 mouse bone marrow cells (Figure 8A). CCK-8 results showed that UAOS-Na did not affect the viability of mBMDC (Figure 8B). To evaluate the effect of UAOS-Na on the BMDC maturation, flow cytometry was performed. UAOS-Na did not affect the numbers of mBMDC (CD11c⁺, CD86⁺ and CD11c⁺, MHC-I⁺) (Figures 8C,D). However, UAOS-Na reduced the mRNA levels of Tapbp and H2-T23 in mBMDC (Figure 8E), which indicated that UAOS-Na could reduce the antigen presentation ability of mBMDC.

To investigate the effect of UAOS-Na on the antigen presentation ability of mBMDC, MOG-mBMDC treated by UAOS-Na were co-cultured with MOG-lymphocytes derived from EAE mouse spleen (Figure 9A). As shown in Figure 9B, UAOS-Na treatment inhibited MOG-lymphocytes proliferation induced by MOG-mBMDC which was analyzed by CCK-8 commercial kits. Furthermore, in Figures 9C,D, flow cytometry was used to determine the effect of UAOS-Na on the proportion of lymphocytes and CD8 T cells (CD3ε⁺, CD8a⁺) in the co-cultured system. UAOS-Na (40 μM) significantly inhibited the anumber of lymphocytes and CD8 T cells (CD3ε⁺, CD8a⁺; Figures 9B–D).

In conclusion, UAOS-Na inhibited mBMDC antigen presentation ability by reduced Tapbp and H2-T23 expression thereby suppressing CD8 T cell proliferation and differentiation.