Splenocytes were isolated from C57BL/6 mice and negatively selected by the MagniSort™ mouse B cell enrichment kit (Thermo Fisher Scientific) to acquire total mouse B cells. Separated B cells were stimulated by LPS from Escherichia coli (10 μg/ml, Sigma-Aldrich) for 3 days. B cells were cultured in RPMI 1640 (Welgene) with 10% fetal bovine serum (FBS), 1% antibiotics (Thermo Fisher Scientific), and 50 μM β-mercaptoethanol (Sigma-Aldrich). CD138⁺B220^low plasma cells were analyzed by flow cytometry. IgM produced during plasma cell differentiation was measured by ELISA (Thermo Fisher Scientific). For in vivo plasma cell differentiation by LPS, LPS (0.6 mg/kg) was intravenously injected into 8-12 week-old C57BL/6 mice. On day 4 after LPS injection, mice were euthanized, and the plasma cell population in the splenocytes was analyzed using flow cytometry.

Cells were blocked by anti-mouse CD16/32 antibodies (Invivomab) before surface staining. Surface proteins expressed on cells were stained with fluorescence conjugated antibodies diluted in FACs buffer (1×PBS with 0.5% bovine serum albumin) at 4°C for 30 min. Live cells were stained by fixable viability dye (Thermo Fisher Scientific). Intracellular protein staining was performed by using the Intracellular Fixation & Permeabilization Buffer Set (Thermo Fisher Scientific), following the manufacturer’s recommended protocol. Intracellular proteins such as ENPP2 and IRF4 were stained after permeabilization. Anti-mouse IRF4 APC, anti-mouse Blimp-1 APC and anti-mouse B220 PE were purchased from Thermo Fisher Scientific. Anti-mouse CD138 APC was purchased from BD Biosciences. After staining, cells were washed and flow cytometry data acquired on FACScanto II (BD Biosciences), and the data was analyzed using FlowJo software (Tree Star).

Cells were lysed by adding TRIzol reagent (Thermo Fisher Scientific) and chloroform was added to the lysed sample. RNA was isolated from the aqueous layer of chloroform and precipitated by using isopropyl alcohol. Extracted RNA was reversely transcribed into cDNA using Maxime RT premix (iNtRON Biotechnology). qPCR was performed on Qiagen real-time PCR instrument with 2X SYBR Green (BioFACT), and gene-specific primers: Pax5-forward, 5’-CCATCAGGACAGGACATGGAG-3’; Pax5-reverse, 5’-GGCAAGTTCCACTATCCTTTGG-3’; Xbp1-forward, 5’-AGCAAGTGGTGGATTTGGAAGA-3’; Xbp1-reverse, 5’-CACCAGCCTTACTCCACTCC-3’; Prdm1-forward, 5’-GCTGCTGGGCTGCCTTTGGA-3’; Prdm1-reverse, 5’-GGAGAGGAGGCCGTTCCCCA-3’; Bcl6-forward, 5’-ACCACCAGCCTCTTATCCCA-3’; and Bcl6-reverse, 5’-TGGGGACTCTGGGGACTAAC-3’, Irf4-forward, 5’-CTCTTCAAGGCTTGGGCATT-3’; Irf4-reverse, 5’-TGCTCCTTTTTTGGCTCCCT-3’, Aicda-forward, 5’-AAATGTCCGCTGGGCCAA-3’; Aicda-reverse, 5’-CATCGACTTCGTACAAGGG-3’, Gapdh-forward, 5’-TGCACCACCAACTGCTTAG-3’; Gapdh-reverse, 5’-GGATGCAGGGATGATGTTC-3’. Tbp-forward, 5’- AGAATAAGAGAGCCACGGACAA-3’; Tbp-reverse, 5’- CTTCACTCTTGGCTCCTGTG-3’. The qPCR data were quantified ΔCt values by using the 2-ΔΔCt method proportional to the expression of Gapdh and Tbp as internal controls.

8-12-week-old C57BL/6 female mice were purchased from Orient Bio Inc. B cell-deficient μMT mice were purchased from the Jackson Laboratory. The mice were housed 4-6 per cage in a clear and ventilated environment maintained under laboratory conditions (Temperature 22 ± 1°C, relative humidity 50 to 70%, and 12 h light and dark cycle). Standard food and water were provided throughout the experiments. Mice were acclimated to their surroundings over 5 days to eliminate the effect of stress prior to initiation of the experiments. Mice were immunized by MOG^35-55/CFA emulsion, by injecting the emulsion subcutaneously (Hooke lab). At 2 h after immunization, pertussis toxin was intraperitoneally injected into mice at day 0 and 1 (100 ng/head). EAE mice were monitored for weight change and clinical scores were measured from day 7 after immunization. The weight of each mouse was measured by using an electronic scale and the clinical score was measured. After monitoring EAE, mice were sacrificed for analysis at day 19. Plasma cell populations in the spleen and lymph nodes were analyzed by flow cytometry.

Spinal cords were fixed and processed in paraffin blocks for histological analysis. Tissues were sectioned horizontally at 5 μm. Sectioned samples on the slides were deparaffinized and hydrated by incubating slides in xylene for 15 min, 3 times, and 100% and 95% ethanol for 5 min, 2 times each. Then, an antigen unmasking procedure was conducted by boiling in 10 mM sodium citrated buffer (pH 6) for 10 min. Mouse-/rabbit-specific HRP/DAB (ABC) detection IHC kit (Abcam) was used following the manufacturer’s recommended protocol to develop slides by the DAB substrate. Anti-mouse CD138 and anti-rabbit CD4 polyclonal antibodies for use as primary antibodies were purchased from Thermo Fisher Scientific. Infiltrated cells and myelinated area were measured by hematoxylin & eosin (H&E) staining and luxol fast blue staining, followed by using ImageJ. Luxol fast blue staining was conducted by using the Luxol Fast Blue Stain Kit (Abcam), following the manufacturer’s recommended protocol. Representative histological images were selected for each group of mice.

Cells were stained by 2 μM of CFSE (Thermo Fisher Scientific) in PBS supplemented with 3% FBS. The cells were incubated for 20 min at room temperature. Cells were washed by PBS with 5% FBS twice. Then, the cells were incubated for 1 h with RPMI 1640 supplemented with 10% FBS and 1% antibiotics for stabilization of CFSE. After administration of proper stimulation, the intensity of CFSE was measured by flow cytometry.

GraphPad Prism software was used to evaluate results. All results are expressed as the mean ± SEM for the data obtained from the indicated number of experiments. Statistical analysis was performed using Student’s t-test or one-way-ANOVA, or two way-ANOVA. A P value ≤ 0.05 was considered statistically significant.

This study aimed to elucidate the impact of SPC on MS pathology, with a particular focus on plasma cells as a crucial cell type in disease progression. Furthermore, we aimed to investigate the impact of SPC under conditions of elevated autotaxin levels, which mimics the inflammatory condition and MS. The expression of autotaxin is increased with proinflammatory stimulation including tumor necrosis factor, IL-6, and LPS (13). Moreover, previous reports demonstrated that the expression of autotaxin is augmented in several cell types including B cells in autoimmune disorders (13, 14). Therefore, we differentiated B cells with LPS stimulation and examined the level of autotaxin. Stimulation of isolated B cells with LPS for 3 days induced plasma cell differentiation, and the gene expression of Enpp2 encoding autotaxin was strongly increased during the differentiation induced by LPS ( Figure 1A ). Flow cytometric analysis also revealed that LPS significantly increased the expression of autotaxin ( Figure 1B ).

Autotaxin converts SPC or LPC into S1P or LPA with its lysophospholipase D activity (15). Our finding on the upregulation of autotaxin during plasma cell differentiation led us to examine whether plasma cell differentiation is affected by the substrates of autotaxin (SPC or LPC) or the products of autotaxin (S1P or LPA). We found that SPC can downregulate plasma cell differentiation but the other three lipids (S1P, LPC and LPA) did not affect LPS-induced plasma cell differentiation ( Figure 1C ). Stimulation of B cells with LPS for 3 days significantly increased the level of plasma cells which are B220^lowCD138⁺ ( Figure 1C ). Addition of 10 μM SPC significantly decreased the level of B220^lowCD138⁺ cells in the presence of LPS ( Figure 1C ), suggesting that SPC inhibits plasma cell differentiation induced by LPS. In separate experiments, we found that all the tested bioactive lipids had no effect on the viability of B cells ( Supplementary Figure 1 ). Therefore, we can rule out the possibility that SPC inhibits plasma cell differentiation by increasing cell death. We found that SPC inhibited plasma cell differentiation induced by LPS in a concentration-dependent manner, showing significant inhibitory effects at 5-10 μM ( Figure 1D ). Plasma cell differentiation is associated with increased secretion of IgM (16). We also found that LPS stimulation of B cells significantly increased secreted IgM level, which was markedly decreased by SPC ( Figure 1E ).

Next, we examined whether SPC affects plasma cell differentiation in vivo. Administration of LPS (0.6 mg/kg) for 4 days increased B220^lowCD138⁺ cells in the spleen, suggesting that LPS injection increased plasma cell differentiation in vivo ( Figure 1F ). Daily subcutaneous injection of SPC for 4 days significantly decreased the level of LPS-induced plasma cell differentiation in the spleen in the LPS-injection model ( Figure 1F ). This result supports our hypothesis that SPC is a new molecule that inhibits plasma cell differentiation.

Plasma cell differentiation is associated with changes of gene expression pattern in B cells (17–20). Addition of SPC during plasma cell differentiation induced by LPS significantly decreased the expression levels of several plasma cell-associated genes such as Prdm1, Irf4, Xbp1 ( Figure 2A ). Xbp-1 is a downstream signaling molecule of Blimp-1 encoded by Prdm1. Besides, Aicda is related to class switch recombination of B cells. Among the plasma cell-associated genes, Irf4 and Prdm1 are known as key transcription factors that mediate plasma cell differentiation (20). We found that stimulation of B cells with LPS for 3 days strongly upregulated the level of IRF4 in the cells, which was significantly decreased by SPC ( Figure 2B ). This result is consistent with our finding that SPC inhibits plasma cell differentiation. We further examined the effects of SPC on the expression of IRF4 during plasma cell differentiation. Stimulation of B cells with LPS generated two IRF4-positive populations (IRF4^int and IRF4^high) by decreasing the IRF4^negative population in a time-dependent manner. Addition of SPC significantly increased the IRF4^int population, while strongly decreasing the IRF4^high population ( Figure 2C ). A previous report demonstrated that the IRF4^high population differentiates into plasma cells in response to outside stimuli (20). Moreover, we found that Blimp-1 was also decreased by SPC, which is another key transcription factor of plasma cell differentiation ( Figure 2D ). In conclusion, SPC downregulates plasma cell differentiation, which is accompanied by decrease of IRF4 and Blimp-1 levels in B cells after activation. During the differentiation to plasma cells, B cells undergo multiple divisions to receive the necessary signals for regulating key transcription factors, such as IRF4 and Blimp-1 (21). We observed that the addition of SPC during LPS-induced plasma cell differentiation significantly reduced cell proliferation compared to the vehicle ( Figure 2E ). The inhibitory effect of SPC on B cell proliferation suggests its impact not only on suppressing plasma cell differentiation but also on modulating B cell activation.

Because of the similarity of the structure of sphingolipid, SPC acts on several S1PRs such as S1PR1, S1PR2, S1PR3, and so on (22–24). Thus, we focused on S1PRs to identify the target receptor(s) of SPC that mediates the inhibitory effects of SPC on plasma cell differentiation induced by LPS. First, we examined whether B cells express S1PRs during activation. We found that S1PR1, S1PR3, and S1PR4 expression are relatively high compared to S1PR2 and S1PR5 at the naïve state of B cells ( Figure 3A ). We next investigated whether SPC acts on S1PRs to block plasma cell differentiation using two different S1PR antagonists: VPC23019 as an S1PR1/3 antagonist, JTE-013 as an S1PR2 antagonist. We found that the SPC-induced inhibitory effects on plasma cell differentiation were significantly attenuated by VPC23019 but not by JTE-013 ( Figure 3B ). Addition of several different concentrations of VPC23019 significantly alleviated SPC-induced inhibition of plasma cell differentiation in response to LPS in a concentration-dependent manner ( Figure 3C ). Fingolimod (FTY720), an FDA-approved immunosuppressive drug for the treatment of MS, mainly acts on S1PR1 (25). It is known that S1PR1 specific modulators (S1P or FTY720) internalize S1PR1 with receptor binding. We found that stimulation of B cells with S1P elicited S1PR1 internalization with flow cytometric analysis ( Figure 3D ). However, SPC did not induce S1PR1 internalization ( Figure 3D ), suggesting that SPC may not act through S1PR1. Moreover, the S1PR1 specific antagonist, W146, did not block the suppressive effect of SPC on the plasma cell differentiation induced by LPS ( Figure 3E ). On the other hand, TY52156, another S1PR3 antagonist, significantly blocked the inhibitory actions of SPC on plasma cell differentiation ( Figure 3F ). Moreover, TY52156 reversed the downregulating effects of SPC on IRF4 ( Figure 3G ). Collectively, our results suggest that SPC may act on S1PR3 but not S1PR1 or S1PR2 to block plasma cell differentiation in response to LPS.

Plasma cells have been regarded to be important players in the development and progression of autoimmune disorders including MS (26, 27). Since we found that SPC strongly blocks plasma cell differentiation, we investigated whether SPC shows beneficial effects against an EAE model, an animal model of MS. Establishment of the EAE model caused the decrease of gradual body weight and the increase of clinical score. However, daily subcutaneous injection of SPC significantly decreased the clinical score and led to a recovery of body weight ( Figure 4A ). Luxol fast blue staining showed that EAE establishment markedly increased demyelination in the spinal cord, which was almost completely blocked by SPC administration ( Figure 4B top). Immune cell infiltration, analyzed by H&E staining, was induced by EAE establishment. SPC strongly blocked immune cell infiltration into the spinal cord ( Figure 4B bottom). Previous reports demonstrated that EAE pathogenesis is associated with CD138⁺ plasma cells in the CNS (26, 28). We also found that CD138⁺ cells were highly increased in the meningeal area of the spinal cord in EAE model mice. SPC administration markedly blocked the accumulation of CD138⁺ cells ( Figure 4C ). In addition to reducing CD138⁺ cell population, the administration of SPC also led to a decrease in the infiltration of CD4 lymphocytes into the spinal cord ( Supplementary Figure 2 ). These results suggest that SPC effectively inhibits plasma cell infiltration into the CNS.

Next, we examined the effects of SPC on plasma cell differentiation in the EAE model. Establishment of the EAE model significantly increased B220^lowCD138⁺ cell levels in the inguinal lymph node but not in the spleen. Administration of SPC significantly decreased the levels of the B220^lowCD138⁺ cell population to the basal level ( Figure 4D ). In addition, when we induced the EAE model in μMT B cell-deficient mice and injected SPC, SPC did not show therapeutic effects ( Supplementary Figure 3 ). These results suggest that B cells are important for the SPC-induced therapeutic effects on the EAE model. In conclusion, the results suggest that SPC may show beneficial effects against the EAE model with decreased plasma cell differentiation in peripheral lymph nodes and the infiltration into the CNS.

Since plasma cells have an important role in the production of autoantibodies (29, 30) and SPC downregulates plasma cell differentiation via S1PR3 ( Figure 3C ), we subsequently investigated whether an S1PR3 antagonist affects the therapeutic effects of SPC on the EAE model. Administration of TY52156 (an S1PR3 antagonist) significantly blocked the therapeutic effects of SPC in the EAE model ( Figure 5A ). TY52156 administration blocked the beneficial effects of SPC on the EAE model in terms of infiltration of immune cells in the spinal cord, demyelination, and plasma cell population ( Figures 5B, C ) and peripheral plasma cell population ( Figure 5D ). In separate experiments, we compared the effects of SPC and FTY720 (an FDA-approved MS therapeutic drug, Fingolimod) on the generation of plasma cells in the EAE model. Unlike SPC, FTY720 failed to decrease plasma cell differentiation in vitro ( Supplementary Figure 4A ). Administration of FTY720 also did not attenuate plasma cell differentiation in draining lymph nodes of EAE model mice ( Supplementary Figure 4B ). Collectively, SPC seems to exhibit therapeutic effects to treat EAE through mechanisms different from FTY720 in terms of target receptor and effect on plasma cell differentiation.