Astragaloside II, природный садонин, облегчает ремимиэлирование в демиэлинговых неврологических заболеваниях с помощью рецептора P75NTR, опосредованного осью сигнала β-катенина/Id2/MBP в клетках-прекурсорах олигодендроцитов.

Introduction

Myelin ensures rapid axonal conduction and integrity through its multilamellar structure [ [1] , [2] , [3] ]. Demyelination disrupts this insulation, driving neurodegeneration in neuromyelitis optica [ 4 ], leukodystrophies, amyotrophic lateral sclerosis (ALS), and multiple sclerosis, leading to impaired motor and sensory functions [ 5 ]. Currently, the treatment of demyelinating diseases primarily relies on immunosuppressants, steroids, and immunomodulatory drugs, which help to regulate the autoimmune response and manage disease progression [ 6 ]. However, immunomodulators often provide only symptomatic relief and fail to promote effective remyelination or reverse neuronal damage. Emerging therapies, such as stem cell or gene therapy, and disease-modifying treatments promote neurogenesis  in vitro [ 7 ], yet ​clinical translation challenges remain, including limited remyelination and safety concerns.

In the central nervous system (CNS), oligodendrocytes produce and sustain myelin sheaths. During CNS development, oligodendrocyte precursor cell (OPC) proliferate, migrate to their target locations, and differentiate into mature oligodendrocytes (OLs), which myelinate axons to facilitate neural function and repair [ 8 ]. Nonetheless, the loss of OLs and myelin sheaths leads to severe neurological dysfunction, highlighting the importance of promoting remyelination as a therapeutic strategy for demyelinating diseases. Unfortunately, reliance on the CNS's remyelination is incomplete or fails altogether in many demyelinating diseases. Hence, current strategies aiming at OPC differentiation promotion and remyelination are being actively pursued as potential treatments for demyelinating diseases [ 9 ]. Despite advancements in disease management, effective therapies that promote myelin regeneration remain elusive, highlighting an urgent need for novel therapeutic agents. Ongoing research efforts [ 10 ] continue to shed light on the molecular mechanisms underlying OPC differentiation and myelin loss, paving the way for developing novel therapeutic approaches.

p75 neurotrophin receptor (p75NTR), a low-affinity receptor in the neurotrophin family, exhibits​ context-dependent signaling critical for neural repair [ [11] , [12] , [13] ]. Currently, small-molecule drugs targeting p75NTR are undergoing clinical trials for Alzheimer's disease (AD), highlighting their relevance in neurodegenerative treatment strategies [ 14 ]. Emerging studies have shown that p75NTR gene modification in OPC enhances their migration and differentiation into OLs, thereby promoting myelin repair in spinal cord injury models [ 15 ]. In the experimental autoimmune encephalomyelitis (EAE) mouse model, a well-established model of multiple sclerosis, p75NTR expression was elevated in spinal cord white matter [ 16 ], and embryonically ablated p75NTR worsened disease progression by increasing inflammatory infiltration [ 17 ]. Furthermore, in multiple sclerosis patients, OPC in the lateral ventricles highly express p75NTR, suggesting its essential role in oligodendrocyte lineage progression and myelin repair [ 18 ]. Collectively, these findings establish p75NTR as a crucial regulator of OPC differentiation and a promising therapeutic target for demyelinating disorders, probably addressing a key limitation of current immunomodulatory treatments [ 19 ].

The canonical Wnt/β-catenin signaling pathway plays a crucial role in myelin regeneration [ 20 ], orchestrating the complex processes of OLs lineage progression and myelin formation. A key mechanism in this process involves the regulation of inhibitors of DNA binding 2 (Id2) [ 21 ], a negative regulator of oligodendrocyte differentiation. Inactivation of β-catenin signaling suppresses Id2 expression, which is essential for OPC maturation into myelinating OLs. This suppression leads to increased myelin basic protein (MBP) expression, thereby accelerating OPC differentiation and myelin formation [ 22 ]. Recent studies highlighted that promoting OPC differentiation into OLs was a key prerequisite for effective myelin regeneration [ 9 , 10 ]. However, whether p75NTR and its mediated β-catenin/Id2/MBP signaling axis serve as a critical regulatory mechanism for OPC differentiation and myelin repair remains unclear. Given that Id2 is crucial for maintaining myelin integrity [ 23 ], targeting p75NTR signaling may offer a promising therapeutic strategy to accelerate myelin regeneration.

Natural products [ 24 , 25 ] have long been a valuable source for developing drugs to prevent and treat demyelinating diseases, offering high efficacy and safety. Among them, total astragalosides (TA), the key components of Astragalus membranaceus (Fisch.) Bunge had been shown to promote neurogenesis, modulate immune responses, and counteract oxidative stress [ 26 ]. Notably, within this class of compounds, astragaloside II (AS-II) is particularly notable for its anti-inflammatory, antioxidant, and anti-apoptosis properties [ 27 , 28 ], which contribute to neuroprotection and tissue repair [ 29 ]. Beyond these effects, emerging evidence suggests that AS-II may directly enhance remyelination by promoting oligodendrocyte differentiation, positioning it as a promising candidate for targeted therapy in demyelinating diseases. While existing studies confirmed AS-II's functional benefits [ 27 ], its specific molecular mechanism in regulating OPC differentiation has not been systematically characterized. The absence of confirmed direct targets for AS-II has fundamentally limited mechanistic exploration, particularly its regulatory effects on myelin-synthesis pathways like β-catenin/Id2. Consequently, the structure–activity relationship remains obscure, impeding rational drug design. To bridge these gaps, this study integrated pharmacological and molecular approaches to systematically validate AS-II's remyelination efficacy in cuprizone (CPZ)-induced demyelinated mice through longitudinal assessments. We employed drug affinity responsive target stability (DARTs) mass spectrometry and molecular docking to identify p75NTR as the primary binding target of AS-II, and mechanistically dissected how AS-II disrupted the β-catenin/Id2 transcriptional complex, thereby rescuing Id2-induced suppression of MBP expression in OPC. These findings not only established AS-II as the first natural product targeting p75NTR for CNS regeneration but also provided a mechanistically grounded framework for developing demyelination therapies through modulation of the β-catenin/Id2/MBP axis.

Materials and methods

Reagents and antibodies

Astragaloside I (Cat. No. DH0016), AS-II (Cat. No. DH0017), and astragaloside IV (Cat. No. DH0015) were purchased from Desite Biotechnology Co., Ltd. (Chengdu, China). 3,3′,5-Triiodo-L-thyronine (T3, Cat. No. HY-A0070), SKL2001 (Cat. No. HY-101085), PD90780 (Cat. No. HY-110166 ), and K252a (Cat. No. HY-N6732) were obtained from MedChemExpress (Monmouth Junction, NJ, USA). CPZ (Cat. No. C9012 ) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Neural Tissue Dissociation Kit (Cat. No. 130–092-628) and anti-A2B5 magnetic beads (Cat. No. 130–093-392) were provided by Miltenyi biotec (Germany). The BCA kit (Cat. No. 20201ES76) and polybrene (Cat. No. 40804ES76) were from YEASEN (Shanghai, China). Opti-MEM (Cat. No. 31985062 ) was purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA). Poly-D-lysine (PDL, Cat. No. ST508) was from Beyotime (Shanghai, China). Platelet-derived growth factor receptor α (PDGFRα, Cat. No. ab134123), MBP (Cat. No. ab40390), neural/glial antigen 2 (NG2, Cat. No. ab129051), adenomatous polyposis coli clone CC1 (CC1, Cat. No. ab16794), and GAPDH (Cat. No. ab186930) primary antibodies were obtained from Abcam Technology (Cambridge, UK). Primary antibodies against glycogen synthase kinase 3β (GSK3β, Cat. No. A11731), p-GSK3β Ser9 (Cat. No. AP1088), p-GSK3β Tyr216 (Cat. No. AP0261), p-β-catenin (Cat. No. A1076), β-catenin (Cat. No. A19657), T-cell factor 4 (TCF4, Cat. No. A19548), Id2 (Cat. No. A0996), and FLAG-Tag (Cat. No. AE063) were provided by ABclonal Technology Co., Ltd. (Wuhan, China).

Cell culture and differentiation

Human oligodendroglial cell line MO3.13 (ATCC, USA) was selected based on its well-documented capacity to recapitulate OPC differentiation kinetics while enabling high-throughput pharmacological screening. MO3.13 cells were cultured in Dulbecco's modified Eagle medium (Meilun, Dalian, China), supplemented with 10 % FBS and 1 % penicillin/streptomycin (100 units/mL), in a humidified atmosphere of 5 % CO 2 at 37 °C. To dissect the active monomeric components in TA promoting the differentiation of OPC, MO3.13 cells were seeded into 24 wells of flat-bottom plates at a density of 1 × 10 4 cells/mL and then treated with three main astragalosides in TA, containing astragaloside I, AS-II, and astragaloside IV at 50 μM for 72 h. The 50 μM concentration was selected based on previous studies demonstrating the optimal bioactivity of astragalosides in modulating neuroprotection and cell differentiation while maintaining cell viability and minimizing cytotoxic effects [ 30 ]. For AS-II dose dependency assessment, MO3.13 cells were treated with the concentrations of AS-II (25, 50, and 100 μM), which were chosen to span both suboptimal and supraoptimal ranges based on dose–response curves from pilot studies, with T3 (100 nM) as a positive control for the assessment of OPC differentiation and maturation. In the wake of this, an EdU proliferation assay was performed using an EdU assay kit (C10310-3, Ribobio, Guangzhou, China) according to the manufacturer’s instructions. EdU-positive proliferating cells were quantified by calculating the ratio of EdU-positive cells to total DAPI-stained nuclei using ImageJ software (National Institutes of Health, USA), following previously described methods [ 31 ].

For further investigation of signal transduction of the Wnt/β-catenin pathway, MO3.13 cells were treated with both the Wnt/β-catenin activator SKL2001 (40 μM) and AS-II (100 μM) for 3 days. MO3.13 cells were then collected for Western blotting analysis.

Animals and drug administration

Experimental animals (C57BL/6J mice, 8 weeks, male) were sourced from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China). The mice were housed in a controlled, pathogen-free environment with a temperature maintained at 25 ± 1 ℃, 50 % ± 10 % humidity, and a 12 h light/dark cycle. All mice were accommodated at the Animal Experimental Center of Shanghai University of Traditional Chinese Medicine (SHUTCM; Shanghai, China). The animal experiments conformed with the guidelines approved by the University Animal Care and Use Committee of SHUTCM (Permission number: SZY201809002). Ethical approval was obtained before the initiation of the study, ensuring compliance with institutional and national regulations for the care and use of laboratory animals.

Following a 1-week acclimation period, the mice were randomly assigned to six groups: (1) control group: wild-type (WT) mice treated with vehicle; (2) model group: CPZ-induced mice; (3) positive drug group: CPZ-induced mice treated with 0.3 mg/kg T3; (4) 12.5 mg/kg AS-II group: CPZ-induced mice treated with 12.5 mg/kg AS-II; (5) 25 mg/kg AS-II group: CPZ-induced mice treated with 25 mg/kg AS-II; (6) 50 mg/kg AS-II group: CPZ-induced mice treated with 50 mg/kg AS-II, with 12 mice in each group. To induce the CPZ demyelination model, 400 mg/kg of CPZ suspension was administered daily via gastric lavage for a continuous period of 4 weeks. This consistent and prolonged administration of CPZ is known to effectively induce demyelination, thus establishing the CPZ demyelinated mouse model [ 32 ], as described previously [ 31 ]. Post-CPZ induced myelin damage, mice received daily oral administrations of AS-II (12.5, 25, and 50 mg/kg) for an additional week. The AS-II dosages were calculated proportionally according to its relative abundance in TA used in prior studies [ 31 ], adjusted through pilot dose-ranging experiments confirming equivalent pharmacological activity. The positive control T3 (0.3 mg/kg) dosage was chosen by following the established protocols [ 31 ]. Behavioral tests were conducted at the culmination of the drug administration phase. Upon the completion of the experiment, all mice were anesthetized with isoflurane and euthanized.

p75NTR embryonic knockout (KO) mice were introduced using the CRISPR /Cas9 technique from Cyagen Biosciences Inc. (Strain number: KOCMP-18053-Ngfr-B6J-VA, Suzhou, China). Eight-week-old WT mice and p75NTR KO mice, which originated from the same litter, were divided into six groups for the experiment. Subsequently, the CPZ demyelinated models were established in WT mice and p75NTR KO mice to trigger myelin loss according to the protocol previously described [ 31 ]. Afterward, they were given daily oral administration of AS-II (50 mg/kg) or vehicle. One week later, the mice were euthanized, and brain tissue specimens were collected.

Behavioral test

At the end of the 5th week, the open field test was carried out to detect the spontaneous activity and exploratory behavior of mice. Before the experiment, the mice needed to adapt to the experimental environment for 2–3 h. The experimental environment needed to be quiet, without any sound or light stimulation. The mice were placed in a square open field of 25 cm × 25 cm × 32 cm and allowed to move freely for 5 min. A camera was used to record their activity trajectories, and then real-time analysis software was used to analyze the total distance they moved.

The rotarod test was employed to evaluate the motor coordination and balance ability of mice, which was conducted at the end of the 5th week. The rotating rod was set at a speed of 32 rpm, rotating in a consistent direction. The number of times the mice completely fell off the rod within 5 min, as well as the latency period before falling, known as the time the mice could stay on the rod before falling, were both recorded and calculated.

Luxol fast blue (LFB) staining

After perfusion, mouse brain tissue was fixed in 4 % paraformaldehyde at 4 °C for 24 h, dehydrated in 10 % and 30 % sucrose at 4 °C for 48 h, and embedded in OCT embedding medium. Coronal frozen sections (20 μm) were cut using a Leica 1950 cryostat and stored at −20 °C. For LFB staining, sections were incubated overnight in 0.1 % LFB and 95 % ethanol, differentiated in 0.05 % lithium carbonate, and counterstained with hematoxylin. Sections were dehydrated, cleared with xylene, and mounted with neutral resin.

Lentiviral transduction

The sh-p75NTR (PGMLV-HU6-p75NTR-CMV-ZsGreen1-PGK-Puro) and sh-NC lentivirus were constructed by Genomeditech Co., Ltd. (Shanghai, China), as shown in Table S4 and Fig. S4. MO3.13 cells were seeded at an appropriate density to achieve 30–50 % confluence before transduction. For lentivirus transduction, the viral stock was diluted in a fresh complete culture medium to a final concentration of 1 × 10 8 TU/mL. To achieve an MOI (multiplicity of infection) of 100, an appropriate volume of diluted lentivirus was added to the culture medium, along with 2 % (v/v) FBS and 5 μg/mL polybrene to enhance infection efficiency. The transduction medium was replaced with a fresh complete medium after 24 h to allow for optimal gene expression. After 48 h of transduction, selection with 0.25 μg/mL puromycin (GM-040401-1, Genomeditech Co., Ltd.) was initiated, with medium changed every 3–4 days for a total of 7 days. The puromycin concentration was then reduced to 0.125 μg/mL, and selection was continued for another 7 days to establish stable knockdown cell lines. Transduction efficiency was assessed using multiple approaches. For instance, fluorescence microscopy was used to confirm ZsGreen1 expression in transfected cells, qPCR to evaluate knockdown efficiency at the mRNA level, and Western blot to confirm p75NTR protein expression reduction. To investigate the role of p75NTR in OPC differentiation and maturation promoted by AS-II, DMSO (control) or AS-II (100 μM) was added to both sh-NC control and sh-p75NTR stable cell lines for a 72-h differentiation induction before further analyses.

Cell transfection experiments

Cells of the sh-p75NTR stable knockdown cell line were inoculated into a 24-well plate at a density of 5 × 10 4 cells/mL, and were then transfected with the pEnCMV-p75NTR-3 × FLAG overexpression plasmid (Miaoling Co., Ltd., Wuhan, China) using Neofect TM DNA transfection reagent (TF201201, Neofect biotech Co., Ltd., Beijing, China) in Opti-MEM. Simultaneously, DMSO and the drug AS-II (100 μM) were added for a 72-h culture to induce differentiation. The analysis aimed to determine whether the function of AS-II in promoting OPC differentiation and maturation could be restored after the recovery of p75NTR expression.

cDNAs of P253A and/or S257A mutant of human p75NTR was cloned into pEnCMV-3 × FLAG vector. The plasmids were constructed and obtained from GenScript Biotech Corporation. The sequences of the gene-specific primer set utilized in this study are provided in supporting information Table S3. cDNA constructs were transfected into MO3.13 cells, as described in the manufacturer’s protocol. After successful transfection of the mutant plasmid for 24 h, MO3.13 cells were stimulated with AS-II (100 μM) and further cultured for another 24 h, and were then collected for analyzing the expression of downstream signaling pathway proteins such as GSK3β, β-catenin, TCF4, and Id2.

Isolation of rat primary OPC

To confirm translational relevance, the primary OPC were isolated and purified from the cerebral cortex of female SD rats aged 7–9 days after birth using the magnetic bead purification method via Neural Tissue Dissociation Kit [ 33 ]. The decision to use female SD rats was based on previous reports [ 33 ], which indicate that female neonatal rodents exhibit higher OPC yields and survival rates, improving the reproducibility and efficiency of cell isolation. In brief, the hippocampus, cerebellum, and brainstem were removed, and the cerebral cortex was collected. After papain digestion and lysis, the tissue was ground into a single-cell suspension. Anti-A2B5 magnetic beads were further used for column purification according to the manufacturer's instructions to obtain OPC. Then, OPC were resuspended and plated in PDL-coated dishes with OPC proliferation medium supplemented with 0.5 % B27 (17504–044, Invitrogen, USA) and PDGF-AA (20 ng/mL, 100-13A, Peprotech, USA) for 3 days to promote proliferation. The purity of the isolated OPC was assessed via immunofluorescence staining for PDGFRα, with purity levels exceeding 90 %, ensuring the reliability of subsequent differentiation experiments. Subsequently, drug-induced differentiation experiments with AS-II intervention were conducted.

Immunofluorescence

After 72 h of drug-induced differentiation, MO3.13 cells or primary OPC were fixed with 4 % paraformaldehyde for 30 min at room temperature and were then washed with PBS. Meanwhile, the cell membrane was permeabilized with 0.3 % Triton X-100. To block non-specific binding, cells were followed by blocking in 5 % donkey serum. Then, they were incubated at 4 °C with a primary MBP antibody. 12 h later, Alexa Fluor®488-conjugated secondary antibody (green, Invitrogen, A-21202) was applied to MO3.13 cells or primary OPC. Finally, fluorescence was examined using an inverted microscope (Olympus, BX61VS), capturing images for analysis of MBP fluorescence intensity in cells.

Immunohistochemistry

Frozen brain sections of the corpus callosum were selected, which was followed by permeabilization with 0.3 % Triton X-100 in PBS. After blocking with 5 % donkey serum for 1 h, antibodies against MBP, NG2, and the oligodendrocyte marker CC1 were applied to brain sections. The next day, the sections were incubated for 1 h with Alexa Fluor® 488- or 594-conjugated secondary antibodies (red, Invitrogen, A32754). Afterward, the sections were mounted on slides and sealed in a DAPI-containing anti-fade mounting medium. Fluorescence images were captured by a fluorescent microscope for analysis.

qPCR analysis

Following differentiation, MO3.13 cells were washed twice with PBS and lysed in Trizol reagent for total RNA extraction. Similarly, total RNA from mouse brain tissue was isolated using the same method. RNA concentration and purity were assessed using a Nanodrop spectrophotometer (Thermo Fisher Scientific), ensuring A260/A280 ratios between 1.8 and 2.0.

The extracted RNA was reverse-transcribed into cDNA using a cDNA synthesis kit (11141ES60, Yeasen) following the manufacturer’s instructions. qPCR was performed using a Hieff® qPCR SYBR kit (11202ES08, Yeasen) on a qPCR system (Applied Biosystems, Life Technologies) under the following thermal cycling conditions, including 95 °C for 10 min (initial denaturation), 40 cycles of 95 °C for 15 s, and 60 °C for 60 s.

Gene expression levels were normalized to GAPDH using the 2 −ΔΔCt method. Negative controls (without reverse transcriptase) were included to confirm the absence of genomic DNA contamination. Primer sequences are listed in Table S1.

Western blot analysis

Corpus callosum or cell samples were homogenized in pre-chilled RIPA buffer. Proteins were extracted after centrifugation at 4 °C, followed by quantification using a BCA assay. After denaturation at 95 °C, the protein samples were loaded on SDS-polyacrylamide gel for electrophoresis and then transferred onto PVDF membranes. The membranes were blocked with 5 % non-fat milk and incubated overnight with primary antibodies against MBP, p-GSK3β Ser9 , p-GSK3β Tyr216 , GSK3β, p-β-catenin, β-catenin, TCF4, Id2, and GAPDH. After incubation with secondary antibodies, the protein bands were detected using an ECL reagent and imaged for quantification analysis.

Molecular modeling analysis

For virtual binding analysis, the 3D structure of p75NTR protein (PDB: 5ZGG) was retrieved from the Protein Data Bank, referencing its interaction with the positive small-molecule ligand NSC49652. Five amino acid residues (Ile 252, Pro 253, Val 254, Cys 256, Ser 257) were chosen as flexible residues. The p75NTR protein was converted into pdbqt format using Autodock Tools, with a docking region defined by a center point (x = 40.105, y = 13.428, z = 47.085) and a grid box of 18 Å × 18 Å × 18 Å. Small molecules (AS-II, NSC49652, and K252a) were energy minimized and converted to 3D structures (mol2 format) using Chem 3D, and then further processed into pdbqt format via Autodock Tools. Full flexible molecular docking between these small molecules and p75NTR was performed using Autodock Vina (version 1.1.2), and binding affinities of the best conformations were scored, with higher absolute values indicating stronger binding affinities.

Mass spectrometry-based DARTs assay

MO3.13 cells were cultured at a density of 1 × 10⁶ cells/mL in a 37 °C, 5 % CO 2 incubator until 80–90 % confluence. Cells were then collected, washed with PBS, digested using 0.25 % trypsin-EDTA (25200–72, Thermo Fisher Scientific), and the resulting cell pellets were lysed with M-PER reagent (78501, Thermo Fisher Scientific) on ice. After determining protein concentrations, AS-II was incubated with cell lysates at room temperature for 1 h. Protease (53702, Sigma-Aldrich) was added at varying enzyme-to-protein ratios (1:100, 1:300, 1:1000, 1:3000, 1:10000) for 30 min. The reaction was terminated with protease inhibitors. The protein samples were then prepared for SDS-PAGE, followed by Coomassie Blue staining (P1300, Solarbio Science & Technology Co., Ltd., Beijing, China). Finally, the protein bands were excised for mass spectrometry sequencing and data analysis [ 34 ].

DARTs validation

MO3.13 cells were transfected with the pEnCMV-p75NTR-3 × FLAG plasmid and cultured for 24 h. Afterward, MO3.13 cells were collected, and proteins were extracted and quantified. The protein samples were incubated with either DMSO (D2650, Thermo Fisher Scientific) or AS-II (100 μM) for 1 h at room temperature. Pronase digestion was carried out with varying enzyme-to-protein ratios (1:100 to 1:10,000), and the reaction was terminated using protease inhibitors. Western blot analysis was conducted to detect FLAG-tagged p75NTR protein expression using a FLAG-Tag antibody, assessing the interaction between AS-II and the p75NTR receptor. Quantification of FLAG-tagged expression levels was performed using ImageJ software (National Institutes of Health, USA) by measuring band intensity, which was normalized to total protein levels and compared across conditions. A dose-dependent analysis was conducted using different concentrations of AS-II (100, 1000, 10000, and 100000 nM) under the same conditions to confirm the interaction's dependency on drug concentration.

Cellular thermal shift assay (CETSA)

MO3.13 cells were transfected with pEnCMV-p75NTR-3 × FLAG plasmid using Neofect TM DNA transfection reagent and cultured for 24 h, and then the cells were harvested using trypsin digestion, washed in PBS, and split into two aliquots. Cell lysate was prepared to undergo freeze–thaw cycles in liquid nitrogen and warm water. After centrifugation at 300 × g at 4 °C for 15 min, the supernatant was divided into two portions, each incubated with DMSO or AS-II (100 μM) at room temperature for 1 h. Aliquots were then subjected to thermal gradients (42–62 °C) in a PCR machine [ 34 ]. The lysates were mixed with SDS buffer and heated at 95 °C for 15 min, followed by Western blot analysis for the detection of p75NTR protein.

Surface plasmon resonance (SPR) analysis

Small molecule solutions were prepared by dissolving AS-II, K252a, or PD90780 at different concentrations in 1 × PBS-EP (BR100669, GE Healthcare Life Sciences, Piscataway, NJ, USA) containing 5 % DMSO. SPR assay was conducted on a BIAcore T200 biosensor system (GE Healthcare Life Sciences) using a CM7 chip (BR100530, GE Healthcare Life Sciences), and protein coupling was initiated by flowing 1 × PBS-EP + 5 % DMSO at a rate of 10 μL/min. p75NTR receptor protein was immobilized on the chip via anti-histone antibody coupling, following the activation and coupling protocols using EDC/NHS, anti-histone antibody, and ethanolamine blocking reagent (28995056, GE Healthcare Life Sciences). Next, solvent calibration solutions were prepared using different concentrations of DMSO (5.8 % and 4.5 %). Small molecules were injected into the flow cell with a flow rate of 30 μL/min for 120 s, followed by a dissociation time of 120 s. Regeneration of the sensor surface was achieved using 50 % DMSO. The KD value was calculated by fitting the binding response data to the Steady State Affinity model using Affinity software [ 35 ].

Statistical analysis

Statistical analysis was performed using GraphPad Prism 7 software, with experimental data expressed as mean ± SD. Two-group comparisons were made using Student's t -test, and multiple-group comparisons were analyzed using one-way ANOVA followed by Dunnett's test. p -value < 0.05 was considered statistically significant.

Results

AS-II facilitated the differentiation and maturation of MO3.13 cells and primary OPC in a dose-dependent manner

Our previous in vitro screening identified AS-II as the key monomer in TA that promoted OPC differentiation (Fig. S1). To investigate further the dose-dependent effects of AS-II on OPC differentiation, MO3.13 cells treated with AS-II ( Fig. 1 A, at 25 μM, P < 0.05 or 50, 100 μM, P < 0.01) showed increases in MBP fluorescence intensity compared to the control group. Protein and mRNA analyses confirmed dose-responsive MBP upregulation ( Fig. 1 B and G, at 25, 50, 100 μM, P < 0.01 or P < 0.001), suggesting its role in facilitating OPC differentiation and maturation.

AS-II promotes the differentiation and maturation of OPC . (A) Representative immunofluorescence staining images and quantifications for MBP in MO3.13 cells treated with or without AS-II (n = 4). Scale bar: 100 μm. (B) Quantification for the mRNA expression level of MBP in MO3.13 cells in different groups (n = 5). (C, E) Representative MBP immunofluorescence staining images (C) and quantification (E) in primary OPCs (n = 4). Scale bar: 50 μm. (D, F) Representative EdU proliferation staining images (D) and quantification (F) in MO3.13 cells (n = 4). Scale bar: 50 μm. (G) The protein expression level of MBP was determined by Western blot analysis (n = 4). Data are presented as mean ± SD; # p  < 0.05, ## p  < 0.01, ### p  < 0.001, vs Ctrl group, ns, not significant. Statistical test: One-way ANOVA with Dunnett's test.

To further explore its role in primary OPC, primary OPC were isolated from neonatal SD rats to evaluate if AS-II could promote OPC differentiation. As shown in Fig. 1 C, AS-II (100 μM, P < 0.01) intuitively elevated MBP fluorescence intensity compared to the control group. In the positive control T3, MBP expression was also significantly raised ( P < 0.001). AS-II had comparable efficacy to positive control T3, further confirming that AS-II not only enhanced the differentiation of MO3.13 cells but also facilitated the differentiation of primary OPC into OLs ( Fig. 1 E, at 100 μM, P < 0.01).

In addition, EdU staining showed that there were no significant changes in proliferating cells across AS-II doses ( Fig. 1 D and F). Immunofluorescence, qPCR data, and EdU assay demonstrated that AS-II increased MBP expression without affecting OPC proliferation.

AS-II improved neurological dysfunction via promoting myelin regeneration in demyelinated mice caused by CPZ

To assess the potential of AS-II in promoting CNS remyelination, we employed the CPZ-induced demyelination model and evaluated its effects on motor impairments ( Fig. 2 A). AS-II administration (12.5, 25, and 50 mg/kg, P < 0.001) reversed body weight loss in CPZ mice compared to the CPZ-only group ( Fig. 2 B and C). AS-II treatment markedly increased total movement distance in the open field test ( Fig. 2 D and E, 12.5 mg/kg, P < 0.05, or 25, 50 mg/kg, P < 0.001) and extended rotarod retention time in the rotarod test ( Fig. 2 F, 12.5, 25 mg/kg, P < 0.05, or 50 mg/kg, P < 0.001). The positive control T3 showed comparable efficacy in body weight and motor coordination ( P < 0.05 and P < 0.001), supporting the effectiveness of AS-II in restoring motor function.

Therapeutic effects of AS-II treatment on myelin regeneration of CPZ-induced mice . (A) Timeline of the animal experiment. (B, C) Line and bar graphs of body weight changes in CPZ-induced mice (n = 11). (D) Tracking plot of mice in the open field test (n = 11). (E) Total distance traveled by mice in the open field test (n = 11). (F) Latency to fall in the rotarod test (n = 11). (G) Representative LFB staining images for myelin damage in the corpus callosum (white matter) of CPZ-induced mice (n = 4). Scale bar: 200 μm; scale bar for zoom in: 50 μm. (H) Representative MBP fluorescence images and intensity in the corpus callosum of CPZ-induced mice (n = 4). Scale bar: 500 μm. (I) MBP protein expression and quantification in the corpus callosum of CPZ-induced mice (n = 4). Data are presented as mean ± SD; * p  < 0.05, ** p  < 0.01, *** p  < 0.001, vs CPZ group. Statistical test: One-way ANOVA with Dunnett's test.

To further evaluate the effect of AS-II on myelin integrity, LFB staining revealed severe myelin loss in the corpus callosum of CPZ-treated mice compared to controls. AS-II treatment dose-dependently reduced myelin loss ( Fig. 2 G, at 12.5, 25, and 50 mg/kg, P < 0.001), showing the most pronounced protective effect. Furthermore, immunofluorescence staining ( Fig. 2 H, 12.5, 25, and 50 mg/kg, P < 0.001) and Western blot analysis confirmed a dose-dependent increase in MBP expression in the medial corpus callosum ( Fig. 2 I, 12.5 and 25 mg/kg, P < 0.05, or 50 mg/kg , P < 0.01). Similarly, positive control T3 also reduced myelin loss and upregulated MBP expression ( P < 0.01).

To confirm the remyelination effects of AS-II in inflammatory demyelination, we utilized the EAE mouse model. AS-II administration (50 mg/kg) significantly increased body weight in EAE mice (Fig. S3B, P < 0.05), and notably reduced clinical scores at 25 and 50 mg/kg doses (Fig. S3C, P < 0.05 and P < 0.01). In addition, AS-II treatment overtly reduced inflammatory infiltration in the spinal cord and mitigated myelin damage (Fig. S3D and E, 25 and 50 mg/kg, P < 0.001), reinforcing its therapeutic potential in both toxic and autoimmune demyelination models.

AS-II enhanced the generation of newly formed OLs in CPZ-induced demyelinated mice

To ascertain whether AS-II increased the number of newly generated OLs in demyelinated mice, we examined its effects on the number of OPC and newly formed OLs in the CNS of CPZ-induced demyelinated mice. As shown in Fig. 3 A-D, AS-II administration (12.5, 25, and 50 mg/kg, P < 0.001) dramatically decreased NG2 + OPC counts while markedly increasing the number of CC1 + mature OLs in the corpus callosum compared to the CPZ model group (25 and 50 mg/kg, P < 0.001). These results indicated that AS-II facilitated OPC differentiation rather than merely promoting OPC proliferation. Similarly, the positive control T3 also elevated CC1 + OLs counts. In the EAE mouse model, AS-II administration substantially boosted the number of CC1 + mature OLs in demyelinated white matter regions (Fig. S3F, 25 and 50 mg/kg, P < 0.01 or P < 0.001). These findings suggested that AS-II not only reduced OPC recruitment at injury sites but also actively promoted lesion-area remyelination.

AS-II promotes OLs regeneration in CPZ-induced demyelinated mice . (A, B) Representative images of NG2 and CC1 immunofluorescence staining of the corpus callosum in mice treated by CPZ. Scale bar: 100 μm; scale bar for zoom in: 50 μm. (C) Quantification for the number of NG2 + cells in the visual field (n = 4). (D) Quantification for the number of CC1 + cells in the corpus callosum of CPZ-induced mice (n = 4). Data are presented as mean ± SD; *** p  < 0.001, vs CPZ group. Statistical test: One-way ANOVA with Dunnett's test.

AS-II enhanced OPC differentiation and myelin repair via modulating the Wnt/β-catenin signaling in MO3.13 cells and CPZ demyelinated model mice

In vitro , we first explored the effect of AS-II on the Wnt/β-catenin signaling pathway, specifically during the differentiation of MO3.13 cells. As displayed in Fig. 4 A, AS-II treatment notably enhanced the phosphorylation of GSK3β Tyr216 and β-catenin (25, 50, and 100 μM, P < 0.01 or P < 0.001), while markedly reducing the phosphorylation of GSK3β Ser9 at concentrations of 50 and 100 μM ( P < 0.001). Additionally, 100 μM AS-II notably decreased TCF4 and Id2 protein expression ( Fig. 4 A, P < 0.05 or P < 0.01). AS-II intervention also notably repressed the mRNA levels of TCF4 and Id2 ( Fig. 4 B, 25, 50, and 100 μM, P < 0.001), while enhancing Olig2 expression ( Fig. 4 B, 25 μM, P < 0.05, or 50, 100 μM, P < 0.001), indicating the promotion of OPC differentiation through the Wnt/β-catenin signaling pathway in vitro .

AS-II inactivates the Wnt/β-catenin signaling pathway in MO3.13 cells and the corpus callosum of CPZ-induced mice . (A) Western blot analysis and quantification of p-GSK3β Ser9 , p-GSK3β Tyr216 , p-β-catenin, TCF4 and Id2 in MO3.13 cell lysates (n = 4). (B) qPCR analysis of mRNA expression levels of TCF4 , Id2 and Olig2 in MO3.13 cells after 72 h AS-II treatment (n = 5). (C) Protein expression levels of p-GSK3β Ser9 , p-GSK3β Tyr216 , p-β-catenin, TCF4, and Id2, and quantification after AS-II administration in mice induced by CPZ (n = 4). (D) Wnt/β-catenin activator SKL2001 counteracted the promoting effect of AS-II on the Wnt/β-catenin signaling pathway in MO3.13 cells (n = 4). Data are presented as mean ± SD; * p  < 0.05, ** p  < 0.01, *** p  < 0.001, vs CPZ group; # p  < 0.05, ## p  < 0.01, ### p  < 0.001, vs Ctrl group; & p  < 0.05, &&& p  < 0.001, vs AS-II-treated group. Statistical test: One-way ANOVA with Dunnett's test.

Next, in vivo , we examined the effect of AS-II on Wnt/β-catenin signaling in the CPZ-induced demyelination mouse model. As illustrated in Fig. 4 C, CPZ treatment resulted in the suppression of GSK3β activity, a reduction in β-catenin phosphorylation, and downregulation of TCF4 and Id2 expression ( P < 0.05). Conversely, treatment with AS-II during remyelination led to increased phosphorylation of p-GSK3β Tyr216 (25 and 50 mg/kg, P < 0.05, P < 0.01 or P < 0.001), but decreased phosphorylation of p-GSK3β Ser9 (12.5, 25, and 50 mg/kg, P < 0.05), indicating GSK3β activation. Meanwhile, AS-II notably elevated β-catenin phosphorylation (12.5, 25, and 50 mg/kg, P < 0.05, P < 0.01 or P < 0.001), and reduced TCF4 and Id2 protein expression (50 mg/kg, P  < 0.01), suggesting that AS-II modulated Wnt/β-catenin signaling to drive remyelination in vivo .

To further validate the essential role of the Wnt/β-catenin pathway in AS-II-mediated OPC differentiation, we employed SKL2001, a Wnt/β-catenin activator, in MO3.13 cells. As depicted in Fig. 4 D, 100 μM AS-II treatment significantly elevated β-catenin phosphorylation, decreased TCF4 and Id2 protein levels, and increased MBP protein expression compared to the control ( P < 0.001). However, co-treatment with both AS-II (100 μM) and SKL2001 resulted in a marked reduction in phosphorylated β-catenin levels and an increase in TCF4 and Id2 expression, along with a significant decrease in MBP expression ( P < 0.05 or P < 0.001). These results further confirmed that AS-II facilitated OPC differentiation by suppressing the Wnt/β-catenin signaling pathway, evidenced by pathway reactivation (SKL2001), reversing its effects.

AS-II interacted with the p75NTR via Pro253 and Ser257 residues

To identify the direct protein target of AS-II in promoting OPC differentiation, a combination of DARTs and qualitative proteomic analysis was used. As shown in Fig. 5 A, AS-II treatment caused differentially digested proteins by protease within 43–55 kDa. After mass spectrometric sequencing of these proteins, ratios of peptide spectrum matches (PSMs) were compared between AS-II and DMSO groups with a set threshold of ≥ 1.5-fold difference. DARTs screening identified five potential membrane protein targets ( Fig. 5 B), among which two proteins, p75NTR and PRKAR1A, had well-defined 3D structures. Compared with PRKAR1A, p75NTR showed the largest PSMs difference between the AS-II and DMSO groups. Considering p75NTR's strong association with myelin repair and the Wnt/β-catenin signaling pathway, subsequent experiments mainly focused on validating whether p75NTR was a potential target through which AS-II promoted OPC differentiation.

Affinity studies between AS-II and its target p75NTR . (A, B) Coomassie Blue staining and MS-based DARTs sequencing were conducted to identify the target proteins of AS-II. (C) Molecular docking visualization of AS-II binding to p75NTR complex structure. (D) Molecular docking results for p75NTR (PDB: 5ZGG) with AS-II, NSC49652, and K252a. (E) Antiproteolytic effect of p75NTR protein in combination with AS-II (n = 4). (F) Effects of varying concentrations of AS-II on its binding and protease resistance to p75NTR (n = 4). (G) Thermal stability of p75NTR with or without AS-II (n = 4). (H) Binding affinities of AS-II and K252a (at concentrations from 1.5625-100 μM), PD90780 (at concentrations from 1.5625-200 μM) to p75NTR were determined using SPR assay. (I) Predicted binding sites of AS-II to p75NTR. (J) Effects of AS-II (100 μM) on the expression of GSK3β Tyr216 , β-catenin, and MBP in MO3.13 cells transfected with WT or mutant p75NTR (n = 4). Data are presented as mean ± SD; & p  < 0.05, && p  < 0.01, vs GAPDH band; # p  < 0.05, ### p  < 0.001, vs Ctrl group; * p  < 0.05, ** p  < 0.01, vs DMSO group. ns, not significant. Statistical test: Two-group comparisons used Student's t -test (E, G, J), multi-group comparisons used one-way ANOVA with Dunnett's test (F). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

To confirm whether p75NTR is a direct target of AS-II in promoting OPC differentiation, a molecular docking simulation was used to evaluate the binding affinity of AS-II to p75NTR. The docking scores showed that AS-II could form strong hydrogen bonds with p75NTR, suggesting a high affinity with the latter ( Fig. 5 C and D). Additionally, the binding affinity of AS-II to p75NTR was comparable to positive small molecules, including NSC49652 and a known antagonist, K252a ( Fig. 5 D).

To further validate the interaction between AS-II and p75NTR, a DARTs assay was conducted. As depicted in Fig. 5 E, at a set ratio of 1:1000 and 1:300 for protease to protein, AS-II (100 μM) stabilized p75NTR against protease digestion ( P < 0.05 or P < 0.01). Meanwhile, with the ratio of protease to protein fixed at 1:1000, increasing concentrations of AS-II showed a dose-dependent protection against protease digestion ( Fig. 5 F, 10 and 100 μM, P < 0.05 or P < 0.001), confirming the strong and concentration-dependent interaction between AS-II and p75NTR.

Intriguingly, the CETSA experiment also demonstrated that the gray intensity of p75NTR protein bands in the DMSO group began to reduce intuitively at 46 °C, while the 100 μM AS-II-treated group manifested a notable decrease only starting at 50 °C. Within the temperature range of 50–54 °C, there was a significant difference in gray intensity between the AS-II-treated group and the DMSO group ( Fig. 5 G, P < 0.05 or P < 0.01). These data implied that AS-II binding stabilized p75NTR, enhancing its resistance to thermal denaturation.

In light of the previous findings, an SPR assay was utilized to detect the binding interactions of AS-II with the p75NTR receptor protein. As illustrated in Fig. 5 H, our results elaborated that AS-II (at a concentration from 1.5625-100 μM), along with the positive controls K252a and PD90780, exhibited significant binding to the p75NTR receptor protein. Of note, a comparison of the equilibrium dissociation constants (K D values) suggested that the binding affinities of AS-II were similar to those of p75NTR’s antagonists K252a and PD90780.

Aiming at inspecting the possible binding sites of AS-II on p75NTR, subsequent molecular docking analysis was conducted to unveil the binding mode with an established structure of p75NTR (PDB:5ZGG, Fig. 5 C and I). Given the results in Fig. 5 I, AS-II could form stable hydrogen bonds with the p75NTR receptor at the residues Pro253 and Ser257. Consequently, Pro253 and Ser257 were identified as the candidate key residues for this binding interaction.

All residues in the sequence were respectively converted into alanine (A) that cannot easily interact with its substrate. To confirm these precise binding sites for AS-II and their role in mediating downstream effects, various mutant plasmids of p75NTR were constructed and employed in MO3.13 cells and treated with AS-II to identify the binding sites of p75NTR. As depicted in Fig. 5 J, following individual mutations at Pro253 or Ser257, AS-II (100 μM, P < 0.05) overtly elevated the phosphorylation of GSK3β at Tyr216 and β-catenin protein levels, along with a marked increase in MBP protein expression. This suggested that P253A and S257A mutated plasmids failed to reverse the effects of facilitating OPC differentiation stimulated by AS-II. Conversely, the p75NTR plasmids with P253A and S257A double mutant abolished AS-II’s effects ( Fig. 5 J). These results implicated that both Pro253 and Ser257 residues formed essential dual binding sites ​for AS-II-p75NTR interaction and downstream signaling.

AS-II promoted the differentiation and maturation of OPC depending on p75NTR

To further dissect the role of AS-II in promoting OPC differentiation and maturation via the p75NTR, we established a stable MO3.13 cell line using lentiviral infection with sh-p75NTR. As shown in Fig. 6 A, 100 μM AS-II treatment raised MBP fluorescence intensity in control vector cells ( P < 0.001). In contrast, in the p75NTR knockdown cells, AS-II failed to elevate MBP fluorescence, highlighting that p75NTR was essential for the effect of AS-II on OPC differentiation and maturation. In addition, in control vector cells, AS-II (100 μM) distinctly increased phosphorylation of GSK3β at Tyr216 residue and β-catenin at the protein level, while reducing the expression of TCF4 and Id2 at protein and mRNA levels ( Fig. 6 B and C, P < 0.05, P < 0.01 or P < 0.001). These alterations ultimately resulted in elevated MBP protein and mRNA expression levels ( P < 0.01 or P < 0.001). Hence, the knockdown of p75NTR abolished AS-II’s modulation on the downstream β-catenin/Id2/MBP axis, further implicating that AS-II drove OPC differentiation and maturation by modulating this signaling axis through p75NTR.

Effect of p75NTR knockdown on the promotion of OPC differentiation and maturation by AS-II. MO3 . 13 cells were infected with lentivirus sh-NC or sh-p75NTR to establish a stable transfected cell line and treated with or without AS-II (100 μM) for 72 h. (A) Impact of p75NTR knockdown on AS-II-regulated MBP expression (n = 4). Scale bar: 20 μm. (B) mRNA expression levels of p75NTR , TCF4 , Id2 , and MBP by qPCR assay (n = 5). (C) Western blot analysis and quantification of p75NTR, p-GSK3β Tyr216 , p-β-catenin, TCF4, Id2, and MBP in sh-NC or sh-p75NTR group upon AS-II treatment (n = 4). Data are presented as mean ± SD; * p  < 0.05, ** p  < 0.01, *** p  < 0.001, vs sh-NC DMSO-treated group； ### p  < 0.001, vs sh-NC AS-II-treated group; ns, not significant. Statistical test: Student's t -test.

To ascertain whether p75NTR overexpression could restore the downstream signaling changes induced by AS-II in p75NTR knockdown cells, a p75NTR overexpression plasmid was intended to be transfected. As exhibited in Fig. 7 , it was observed that the regulatory effects of AS-II on the p75NTR/β-catenin/Id2/MBP signaling were hindered in sh-p75NTR knockdown cells. Strikingly, following the overexpression of p75NTR, AS-II (100 μM) treatment significantly elevated the phosphorylation of GSK3β at Tyr216 and β-catenin, suppressed TCF4 and Id2 protein expression, and greatly enhanced MBP protein levels ( P < 0.05, P < 0.01, or P < 0.001). Ultimately, all of these data confirmed that AS-II promoted OPC differentiation exclusively via p75NTR-dependent β-catenin/Id2/MBP signaling.

Effects of AS-II on Wnt/β-catenin/Id2/MBP axis in stable p75NTR knock down MO3.13 cells overexpressed with p75NTR . Western blot analysis of the indicated protein expression level. sh-p75NTR stable strain from MO3.13 cells was transfected with p75NTR overexpression plasmid for 24 h, then cells were stimulated by AS-II (100 μM) for 72 h (n = 4). Data are presented as mean ± SD; * p  < 0.05, ** p  < 0.01, *** p  < 0.001, vs sh-p75NTR + OE DMSO-treated group; ns, not significant. Statistical test: Student's t -test.

AS-II failed to accelerate myelin regeneration in CPZ-induced p75NTR KO mice

To confirm AS-II accelerated remyelination in vivo via p75NTR, p75NTR KO mice were employed. As shown in Fig. 8 B-D, AS-II treatment (50 mg/kg) enhanced motor performance, demonstrated by a notable increase in total movement distance during the open field test and extended time on the rotarod test ( P < 0.05 or P < 0.01), compared with the CPZ model in WT mice. However, AS-II failed to improve motor performance in the p75NTR KO mice ( P < 0.05 or P < 0.001). Likewise, as manifested in Fig. 8 E and G, LFB staining supported the alleviation of myelin damage in the corpus callosum by AS-II (50 mg/kg, P < 0.001) was weakened in the p75NTR KO mice. Also, we have observed consistent results of MBP fluorescence intensity expression in the medial corpus callosum ( P < 0.001), while p75NTR was knocked out in mice. As such, p75NTR KO overtly neutralized the increasing expression of MBP mediated by AS-II (50 mg/kg, P < 0.01), compared with WT demyelinated mice ( Fig. 8 F and H). Collectively, these results further indicated that AS-II required p75NTR binding to execute its pro-myelination effects in vivo .

AS-II fails to accelerate myelin regeneration in CPZ-induced p 75 NTR KO mice. (A) Overview of the experimental timeline. (B) Movement tracking of mice during the open field test. (C) Total distance covered by mice in the open field test (n = 8). (D) Time to fall of mice in the rotarod performance test (n = 8). (E) Representative LFB-staining images illustrating myelin damage in the corpus callosum (white matter) of CPZ-treated mice. Scale bar: 200 μm. (F) MBP fluorescence images in the medial corpus callosum. Scale bar: 200 μm; inset scale bar: 100 μm. (G, H) Quantification of LFB staining and MBP intensity in the medial corpus callosum (n = 4). Data are presented as mean ± SD; * p  < 0.05, ** p  < 0.01, *** p  < 0.001, vs CPZ model group of WT mice; & p  < 0.05, &&& p  < 0.001, vs CPZ model group of p75NTR KO mice; ns, not significant. Statistical test: One-way ANOVA with Dunnett's test.

Discussion

Myelin regeneration is critical for neurological recovery in demyelinating diseases [ 36 , 37 ]. However, clinical observations reveal three persistent barriers, including inadequate OPC recruitment, differentiation arrest, and hostile inflammatory niche [ [38] , [39] , [40] , [41] ]. Current therapies primarily target inflammation without addressing remyelination failure [ 42 , 43 ].

While small-molecule therapeutics have emerged as promising remyelination enhancers [ 24 , 25 , 44 ], such as curcumin [ 45 ], ginsenosides [ 10 , 46 ], and resveratrol [ 47 ], their clinical translation has been hampered by insufficient targeting specificity. Here, we identified that AS-II, a natural saponin from A. membranaceus [ 27 , 31 , 48 ], directly bound p75NTR via hydrogen bonds with Pro253 and Ser257. It stabilized a receptor conformation that drove OPC differentiation via the β-catenin/Id2 axis, evidenced by increased MBP without proliferative effects. In CPZ and EAE models, AS-II improved motor function and restored myelin integrity, paralleled by histological confirmation of reduced demyelination lesions and increased mature oligodendrocytes. The concordant improvement in EAE neurological scores underscored its therapeutic potential for demyelinating neurodegeneration, positioning AS-II beyond immunomodulation as a precision remyelination therapy. Unlike therapies that primarily focus on inflammation control, AS-II's ability to enhance remyelination could provide a complementary or alternative approach to addressing the multifaceted challenges of progressive neurodegeneration.

p75NTR receptor, a member of the tumor necrosis factor receptor superfamily [ 13 ], mediates both oligodendrocyte differentiation and myelin repair through context-dependent mechanisms [ 16 , 49 ]. Structural docking revealed that AS-II bound p75NTR via Pro253 and Ser257 hydrogen bonds, stabilizing a conformation that activated regenerative signaling. Functionally, AS-II-mediated activation of p75NTR coordinately suppressed transcriptional repressors Id2 and TCF4 [ 50 ], and upregulated myelinogenic MBP, thereby removing OPC differentiation barriers to drive oligodendrocyte maturation [ 51 ]. β-catenin degradation and MBP transcriptional activation jointly drive myelin synthesis. Our data revealed that AS-II’s binding reprogrammed p75NTR toward regenerative outputs​ through structural stabilization, β-catenin degradation, Id2 suppression, and MBP  transcriptional elevation. p75NTR KO mice confirmed its necessity for AS-II-mediated myelin restoration, establishing its role in balancing oligodendrogenesis and myelin homeostasis. While our shRNA rescue experiments confirmed p75NTR's necessity for AS-II's canonical β-catenin/Id2 effects, we cannot definitively exclude low-efficiency engagement of parallel pathways under chronic p75NTR suppression. However, we have not detected the compensatory upregulation of related neurotrophin receptors or alternative signaling pathways in p75NTR KO mice under AS-II treatment. We cannot fully exclude latent adaptive mechanisms emerging in chronic deficiency models. Future lineage tracing studies tracking oligodendrocyte precursor dynamics in conditional p75NTR KO systems could further validate pathway specificity.

Beyond demyelination, p75NTR signaling plays a crucial role in neurodegenerative diseases [ 52 , 53 ] such as AD, Parkinson’s disease (PD), and ALS, where its activation is linked to neuronal apoptosis, synaptic dysfunction, and inflammation. In AD, p75NTR mediates Aβ-induced synaptic loss [ 54 ], while in PD, its dysregulation contributes to dopaminergic neuron loss and neuroinflammation [ 55 ]. Given that AS-II enhanced p75NTR signaling in a regenerative manner, it might have broader neuroprotective potential beyond remyelination. Future studies should investigate whether AS-II attenuates p75NTR-mediated neurotoxicity, reduces neuroinflammation, and protects against synaptic loss in AD and PD models. Additionally, since p75NTR interacts with Trk receptors, further research should explore whether AS-II modulates Trk-mediated neurotrophic signaling, which is crucial for neuronal survival and plasticity.

Wnt/β-catenin pathway [ 56 ] exhibits stage-dependent regulation in oligodendrogenesis, essential for lineage initiation but inhibitory to differentiation during chronic activation [ 56 ]. In support of this, Boccazzi et al. found that inhibition of the Wnt/β-catenin pathway could facilitate oligodendrocyte differentiation and accelerate remyelination in murine models [ 57 ]. Thus, pharmacological modulation of this duality represents a strategic therapeutic window for remyelination. In our study, AS-II activated GSK3β, as evidenced by increased phosphorylation at Tyr216, driving the destabilization of β-catenin and its subsequent degradation to relieve differentiation blockade. This controlled Wnt suppression aligns with established mechanisms for overcoming OPC differentiation arrest [ 58 ]. A critical discovery in our study was the role of p75NTR in modulating this pathway. Lentiviral studies validated p75NTR's central role in orchestrating the β-catenin/Id2/MBP signaling axis. AS-II engagement of p75NTR potentiated GSK3β-mediated β-catenin destabilization, enabling myelinating oligodendrocyte generation. The p75NTR-GSK3β-β-catenin cascade positioned AS-II as a precision modulator of Wnt signaling dynamics, offering a targeted resolution of differentiation impedance in myelin disorders.

Despite the progress made in understanding the role of AS-II in promoting endogenous OPC differentiation and myelin repair, several limitations and future perspectives remain in this study as below.

Mechanistic specificity

First, while the interaction between AS-II and p75NTR was demonstrated, the precise molecular mechanisms through which this interaction regulates the β-catenin/Id2/MBP signaling axis in vivo require deeper investigation. While our study primarily focused on p75NTR as the key target of AS-II, we have acknowledged that AS-II may have additional off-target effects, warranting future validation. To resolve residual off-target possibilities, future studies will incorporate RNA-seq to exclude major pathway crosstalk, chemical proteomics, or​ kinome-wide CRISPRi strategies to systematically assess other potential signaling pathways influenced by AS-II in OPC differentiation.

Long-term efficacy and safety

Secondly, this study primarily focused on in vitro and short-term in vivo models. While AS-II demonstrated a significant myelin-protective effect in EAE mice, the long-term efficacy and potential side effects of AS-II in chronic demyelinating diseases remain to be explored. While we utilized HPLC-MS to detect AS-II’s concentration in mouse serum and brain tissue to address concerns regarding the oral bioavailability of AS-II, oral bioavailability remains suboptimal, and potential brain accumulation warrants further toxicology profiling. Long-term assessments will be crucial for evaluating not only the sustainability of its myelin-protective effects but also its impact on chronic neuroinflammation, remyelination, and functional recovery. Additionally, it will be important to assess the safety profile of AS-II, its potential neurodegenerative sequelae, and its ability to offer sustained protection against myelin loss in diverse patient populations. Future studies should investigate the sustainability of AS-II’s therapeutic effects in chronic models, particularly in more complex and heterogeneous pathological environments, such as progressive multiple sclerosis or other severe demyelinating diseases.

Clinical translation and regulatory strategy

To facilitate the clinical translation of AS-II, three key aspects integrating mechanistic validation, regulatory strategy, and patient-centric development must be addressed. First, preclinical validation in CPZ-chronic and aged lysophosphatidylcholine models will assess sustained efficacy in gliotic microenvironments. Additionally, AS-II’s brain penetrance should be further enhanced by nanoemulsion-based delivery while improving oral bioavailability through lymphatic uptake, bypassing first-pass metabolism. For regulatory-translational integration, AS-II must undergo Good Manufacturing Practice (GMP)-compliant nanoformulation, followed by phased, randomized, double-blind trials to evaluate safety, efficacy, and chronic treatment tolerability in human subjects. Meanwhile, AS-II’s effectiveness should be compared to existing cell-based and small-molecule therapies. While OPC transplantation enhances myelin repair, it faces challenges like cell survival, integration, and immune rejection. Its ability to enhance endogenous OPC differentiation without exogenous cell transplantation offers a potential advantage in clinical application. Compared to existing small-molecule OPC differentiation agents, AS-II could exhibit unique therapeutic advantages via its novel p75NTR targeting through Pro253/Ser257 binding. While benztropine (muscarinic antagonist) [ 59 ] and clemastine (M1 receptor antagonist) [ 60 ] have promoted OPC differentiation via broad neurotransmitter receptor modulation, their clinical utility is limited by off-target burden. Given its natural origin, AS-II may qualify for fast-track designation as a botanical drug. Future studies should also explore combination strategies with existing remyelination therapies to enhance clinical applicability and therapeutic outcomes.

Conclusion

In conclusion, this study highlights AS-II's pivotal role in promoting myelin regeneration by targeting the p75NTR receptor to suppress the β-catenin/Id2/MBP signaling axis, offering a novel therapeutic strategy for neurodegenerative diseases. Future studies should not only aim to elucidate AS-II's synergy with immunomodulatory agents but also to explore its long-term safety and translational feasibility in clinical settings.

Compliance with ethics requirements

This research work does not contain any studies with human subjects. All experiments involving animals were conducted according to the guidelines approved by the University Animal Care and Use Committee of Shanghai University of Traditional Chinese Medicine (animal housing no. SYXK(HU)2020-2009).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

Funding for this study was supported by National Key Research and Development Program of China (2022YFC3501701), National Natural Science Foundation of China (82374065, 82074043), Natural Science Foundation of Shanghai (24ZR1466500), Eastern Talent Plan Leading Project (BJKJ2024040), and Organizational Key Research and Development Program of Shanghai University of Traditional Chinese Medicine (2023YZZ02).

Footnotes

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jare.2025.04.028 .

Contributor Information

Hui Wu, Email: zgykdxwuhui@foxmail.com.

Xiaojun Wu, Email: xiaojunwu320@126.com.

Supplementary data

The following are the Supplementary data to this article:

Supplementary Data 1

Data availability

Data will be made available on request.

References

Associated Data

Supplementary Materials

Supplementary Data 1

Data Availability Statement

Data will be made available on request.