Isoliquiritigenin ameliorates Parkinson’s disease via gut microbiota remodeling: potential mediating role of Parabacteroides distasonis Authors: Mengyun Li, Xiaobing Zhang, Jiafeng Lu, Junjie Xiang, Jie Wu, Matao Zheng, Siwei Yang, Lingyan He, Xian Shao Published: 2025 ABSTRACT AbstractBackgroundIncreasing evidence implicates the gut microbiota in the pathogenesis of Parkinson’s disease (PD). Microbiota-targeted interventions may be a promising therapeutic approach. Isoliquiritigenin (ISL), a natural flavonoid and primary pharmacological constituent of licorice (Glycyrrhiza), has neuroprotective effects. Whether this neuroprotection involves modulation of the gut microbiota is unclear.PurposeThe study assessed whether ISL mediates neuroprotective in PD by modulating the gut microbiota, and investigated the functional involvement of Parabacteroides distasonis (P. distasonis) in this process.MethodsMice with chemically-induced PD were first treated with ISL via oral gavage to assess its therapeutic effects. To investigate the microbiota-mediated mechanisms, fecal microbiota transplantation (FMT) from ISL-treated donors and direct gavage of live or heat-killed P. distasonis were performed in separate cohorts. Behavioral performance, dopaminergic neuron integrity, intestinal barrier function, neuroinflammation, gut microbiota composition, and serum metabolomic profiles were systematically evaluated.ResultsISL and FMT significantly improved motor deficits and protected dopaminergic neurons in the substantia nigra. ISL treatment compositionally reshaped the gut microbiota by suppressing pro-inflammatory genera and enriching beneficial bacteria, such as Parabacteroides. Oral administration of live P. distasonis fully reproduced the neuroprotective effects of ISL, including improved motor function, reduced neuroinflammation, restoration of blood–brain barrier integrity, and attenuation of intestinal mucosal damage. Heat-killed P. distasonis had no significant effects. Metabolomic analysis revealed that ISL and live P. distasonis jointly regulate neuroprotective pathways and metabolites related to the gut–brain axis.ConclusionISL alleviates PD-related pathology partly through gut microbiota remodeling, with P. distasonis as a potential mediator.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12967-025-07435-w.Keywords: Parkinson’s disease, Gut microbiota, Isoliquiritigenin, Parabacteroides distasonis, Neuroinflammation Introduction Parkinson’s disease (PD) is a chronic and progressive neurodegenerative disorder characterized by the selective degeneration of dopaminergic neurons in the substantia nigra. The neuron damage leads to motor symptoms, such as bradykinesia, rigidity, and resting tremor, as well as non-motor complications [1]. The exact etiology of PD remains elusive. Emerging evidence highlights the critical role of the gut–brain axis, wherein bidirectional interactions between the intestinal microbiota, immune signaling, and central neuroinflammation contribute to disease onset and progression [2]. Gut microbiome changes have been consistently observed in patients with PD and in experimental models. The changes are often accompanied by intestinal barrier dysfunction, elevated systemic inflammation, and enhanced microglial activation [2, 3]. Moreover, gastrointestinal disturbances frequently precede motor symptoms, underscoring the potential role of the gut in PD pathogenesis [4]. Accordingly, therapeutic strategies targeting gut microbiota may offer novel avenues for PD intervention. Isoliquiritigenin (ISL) is a natural flavonoid derived from licorice roots. ISL has demonstrated significant antioxidant and anti-inflammatory effects in various neurological disorder models [5–8]. Recent studies have shown that ISL suppresses microglia-mediated neuroinflammation and attenuates dopaminergic neurodegeneration in models of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced subacute PD by modulating c-Jun N-terminal kinase/protein kinase B/nuclear factor-kappa B (JNK/AKT/NF-κB) signaling pathways [9]. However, whether ISL exerts its neuroprotective effects by modulating the gut microbiota, a critical yet understudied axis in PD, remains unclear. Exploratory data described here indicate that ISL treatment not only improves motor function in mice with MPTP-induced PD, but also induces marked shifts in the gut microbial composition, including a significant enrichment of the genus Parabacteroides. Based on these findings, Parabacteroides distasonis (P. distasonis), a representative commensal species with anti-inflammatory and metabolic regulatory potential, was selected for further validation [10, 11]. Although P. distasonis has shown protective effects in models of colitis, obesity, and liver injury, its role in neurodegenerative diseases remains largely unexplored [12–14]. The present study aimed to determine whether the gut microbiota contributes to the neuroprotective actions of ISL, and evaluated whether P. distasonis plays a functional role in mediating ISL-induced benefits through the gut–brain axis. The findings provide novel mechanistic insights into how microbial modulation may enhance the therapeutic potential of natural compounds in the context of neurodegenerative diseases. Materials and methods Animals and treatment Eight-week-old male C57BL/6J mice were purchased from Qizhen Experimental Animal Technology Co., Ltd. (Hangzhou, China). All animals were housed under a 12-h light/dark cycle with food and water provided ad libitum in a standard environment (22 ± 2℃ and 50 ± 10% humidity). All experiments were conducted in accordance with ARRIVE guidelines and the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The protocols were approved by the Institutional Animal Care and Use Committee of Shaoxing People’s Hospital (ID number: 2025Z068). All mice were fed a standard laboratory chow diet provided by Suzhou Shuangshi Experimental Animal Feed Co., Ltd. (Suzhou, China). According to the manufacturer’s specifications, the diet contained ≥ 18% crude protein, ≥ 4% crude fat, ≤ 5% crude fiber, and ≤ 8% crude ash, with calcium (1.0–1.8%) and phosphorus (0.6–1.2%), maintaining a Ca: P ratio of 1.2:1–1.7:1. The remaining components consisted of carbohydrates, vitamins, and minerals. All groups were provided with the same batch of chow throughout the experiment to ensure consistent nutrient composition. Mice were divided into five groups: control, MPTP, MPTP + ISL-L, MPTP + ISL-M, and MPTP + ISL-H. Mice in the MPTP, MPTP + ISL-L, MPTP + ISL-M, and MPTP + ISL-H groups were intraperitoneally injected with 20 mg/kg/day MPTP, followed by 250 mg/kg/day probenecid dissolved in dimethylsulfoxide for 4 weeks [15]. Mice in the MPTP + ISL-L, MPTP + ISL-M, and MPTP + ISL-H groups were treated with 20, 40, and 80 mg/kg/day ISL (CAS#961-29-5, Aladdin, Shanghai, China) by gavage for 2 weeks [9]. Vehicle was given to control and MPTP groups. For fecal microbiota transplantation (FMT) experiments, fresh fecal pellets (200 mg) were collected from 10 healthy mice treated with ISL by oral gavage and were processed as described previously [16]. In brief, fresh fecal samples were weighed and homogenized in 1 mL of pre-cooled PBS using sterile silica beads at 45 Hz for 1 min. The homogenate was filtered through a 70 μm cell strainer and centrifuged at 6,000 ×g for 15 min at 4 °C. The resulting pellet was resuspended in 30% (v/v) glycerol and stored at − 80 °C. Mice were divided into three groups: control, MPTP, and MPTP + FMT. Mice in the MPTP and MPTP + FMT groups were intraperitoneally injected with MPTP (20 mg/kg/day) and probenecid (250 mg/kg/day) for 4 weeks. The mice in FMT group were treated with 100 µL of filtered stool homogenates via gavage once daily for 2 weeks. Vehicle was given to control and MPTP groups. In the P. distasonis intervention experiment, the mice were divided into six groups: control, MPTP, MPTP + Pd-L, MPTP + Pd-M, MPTP + Pd-H, and heat-killed P. distasonis-treated MPTP group (MPTP + Hk-Pd). Mice in the MPTP, MPTP + Pd-L, MPTP + Pd-M, MPTP + Pd-H, and MPTP + Hk-Pd groups were intraperitoneally injected with MPTP (20 mg/kg/day) or probenecid (250 mg/kg/day) for 4 weeks. Mice in the MPTP + Pd-L, MPTP + Pd-M, and MPTP + Pd-H groups were treated with the live P. distasonis at a concentration of 107, 108, and 109 colony forming units (cfu) by gavage, once daily for 2 weeks. Mice in the MPTP + HK-Pd group were treated with the heat-killed P. distasonis.Vehicle was given to control and MPTP groups. Culture of P. distasonis P. distasonis purchased from BeNa Culture Collection (BNCC354946, Xinyang, China) was cultured on Columbia blood agar (BNCC352241, BNCC, Beijing, China) at 37℃ for 48 h in an anaerobic environment using anaerobie glove box (HYQX-Ⅲ, Hengyue, Shanghai, China). The concentration of P. distasonis was estimated based on the optical density of the culture at 600 nm (OD600). After the bacteria completely resuspended, the suspension was aspirated. P. distasonis was inactivated by pasteurization (70℃ for 30 min). Open field test The open field test was used to measure spontaneous locomotion. Each mouse was gently placed in the center of a 25 cm × 25 cm dark box and allowed to freely explore the arena for 3 min. The movement trajectories of the mice were recorded using a camera and analyzed using behavioral analysis software (Shanghai Jiliang Software Technology Co.Ltd., China). At the end of this period, each mouse was immediately returned to its cage. After removing feces and urine left by the previous mouse, the apparatus was cleaned with a 75% alcohol solution. The total distance traveled and speed were analyzed. Grip strength test The grip strength test was used to assess forelimb strength. The test measures the peak force exerted by a rodent grasping a specially designed mesh assembly. The exerted peak force is displayed on a digital display using a precision force gauge. Immunohistochemistry Mouse brain tissues were fixed with a pre-cooled 4% paraformaldehyde solution and paraffin sections were prepared. After antigen retrieval, the sections were blocked with 3% bovine serum albumin (BSA) and incubated with a 1:500 dilution of mouse anti-tyrosine hydroxylase (TH) antibody (sc-25269, Santa Cruz Biotechnology, Dallas, TX, USA) at 4℃ overnight. The sections were incubated with a 1:5000 dilution of horseradish peroxidase-labeled secondary antibody (GB23301, Servicebio, Wuhan, China) at room temperature for 50 min, and finally the sections were photographed under a microscope. ImageJ software (v1.54d, NIH, Bethesda, MD, USA) was used to quantify TH-positive dopaminergic neurons. Western blot The striatum and colon were lysed using a tissue grinder in RIPA lysis buffer containing 1% protease inhibitors. Tissue lysates were centrifuged at 12,000 rpm for 15 min at 4℃. After high-temperature denaturation, the resulting supernatants were subjected to western blot analysis. The following antibodies and dilutions were used: mouse anti-TH (sc-25269, Santa Cruz Biotechnology, 1:5000), rabbit anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (GB11002-100, Servicebio, 1:5000), rabbit anti-Occludin (27260-1-AP, Proteintech, Rosemont, IL, USA, 1:10000), rabbit anti-Claudin1 (13050-1-AP, Proteintech, 1:4000), horseradish peroxidase (HRP)-labeled goat anti-mouse IgG (Cat#A0216, Beyotime, Shanghai, China, 1:5000) and HRP-labeled goat anti-rabbit IgG (Cat#A0208, Beyotime, 1:5000). Enzyme-linked immunosorbent assay (ELISA) To evaluate systemic inflammatory cytokine levels, whole blood samples were collected from mice via cardiac puncture and allowed to clot at room temperature. The samples were centrifuged at 3,000 rpm for 10 min at 4 °C to isolate serum, which was then stored at − 80 °C until further analysis. Serum concentrations of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and interleukin-6 (IL-6) were measured using commercial ELISA kits (Byabscience Biotechnology Co., Ltd., China) according to the manufacturer’s instructions. Absorbance was recorded at 450 nm using a microplate reader, and cytokine concentrations were calculated based on standard calibration curves. Hematoxylin-eosin (HE) staining For HE staining, mouse colon tissues were fixed with pre-cooled 4% paraformaldehyde solution and cut into paraffin sections. After dewaxing, paraffin sections were stained with HE (G1076, Servicebio). Finally, the sections were examined by microscopy and photographed. Alcian blue/periodic acid-schiff (AB–PAS) staining For AB–PAS staining, mouse colon tissues were fixed with a pre-cooled 4% paraformaldehyde solution and paraffin sections were prepared. After dewaxing, the paraffin sections were stained with AB–PAS dye solutions C, B, and A (G1049, Servicebio), respectively. Finally, the sections were photographed under a microscope. Immunofluorescence staining To evaluate astrocytic and microglial activation in the substantia nigra, double immunofluorescence staining for glial fibrillary acidic protein (GFAP) and ionized calcium binding adaptor molecule 1 (Iba1) was performed on paraffin-embedded mouse brain sections. Paraffin-embedded brain sections were deparaffinized and rehydrated, followed by antigen retrieval in citrate buffer by microwave heating. After cooling, the sections were permeabilized with 0.3% Triton X-100 for 15 min and blocked with 5% BSA for 1 h at room temperature. The sections were then incubated overnight at 4 °C with rabbit anti-GFAP (60190-1-Ig; Proteintech, 1:200) and rabbit anti-Iba1 (019-19741, Wako, Gloostrup, Denmark, 1:400) antibodies. After washing with PBS, the appropriate Alexa Fluor-conjugated secondary antibodies (goat anti-mouse 488, BA1126 or goat anti-rabbit Cy3, BA1032; both from Boster and both at 1:200 dilution) were applied for 1 h at room temperature in the dark. Nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI). After mounting with an anti-fade medium, fluorescence images were captured using a fluorescence microscope (80I, Nikon, Japan). 16S rDNA sequencing To evaluate the regulatory effects of ISL on gut microbiota, fresh fecal samples from mice were collected, immediately snap-frozen in liquid nitrogen, then stored at − 80 °C until use. After extracting genomic DNA, PCR amplification was performed. A library was constructed according to the standard procedure for the ALFA-SEQ DNA Library Prep Kit. The stool DNA was then amplified using Phusion High-Fidelity PCR Master Mix (New England Biolabs) by PCR targeting the variable regions 3 and 4 (V3–V4) of the 16 S rRNA gene (forward primer, ACTCCTACGGGAGGCAGCA; reverse primer, GGACTACHVGGGTWTCTAAT). The constructed amplicon library was sequenced using a PE250 device on an Illumina platform (Guangdong Magigene Biotechnology Co., Ltd. Guangzhou, China). Raw data were processed using QIIME2, which included quality filtering, chimera removal, amplicon sequence variant (ASV) clustering, and taxonomic assignment based on the Silva reference database. Before statistical analysis, ASVs annotated as chloroplasts, mitochondria, or potential contaminants were removed, and singleton ASVs and low-depth samples were excluded. Alpha diversity (e.g., Simpson index) was calculated from rarefied tables normalized to the same sequencing depth, and groups were compared using the Kruskal–Wallis test. Beta diversity was based on relative-abundance tables (total-sum normalization) to compute Bray–Curtis distances, visualized by principal coordinate analysis (PCoA) and tested with PERMANOVA. Statistical differences in microbiota composition among the groups were identified using linear discriminant analysis (LDA) effect size (LEfSe), with a logarithmic LDA score threshold of 2.0. The differential abundances of key genera are visualized using bar plots. Additionally, hierarchical clustering heatmaps were generated to present the genus level relative abundance profiles across groups. Untargeted metabolomics To investigate the differences in serum metabolites among the control, MPTP, MPTP + ISL, and MPTP + Pd groups, untargeted metabolomic analysis was performed. Serum metabolites were extracted using organic solvent-based protein precipitation. Quality control (QC) samples were prepared by pooling equal volumes of all serum samples. All samples, including QC samples, were analyzed in a randomized order. QC samples were inserted at regular intervals (beginning, middle, and end) throughout the analytical run to monitor instrument stability and evaluate technical reproducibility. Samples were analyzed in both positive and negative ion modes by liquid chromatography–mass spectrometry (LC-MS). Peak detection, alignment, and QC were performed using XCMS software. Metabolites were identified using the MetaX platform, followed by functional annotation using public databases. Subsequently, quantitative sample correlation and differential metabolite screening were performed. Partial least squares discriminant analysis (PLS-DA) was used to visualize group separation in the metabolomic profiles. Prior to PLS-DA, metabolomic data were median-centered, log-transformed, and Pareto-scaled after missing-value imputation (half of the minimum positive value). All variables were mean-centered before model construction, and model robustness was assessed through cross-validation and permutation testing. Data QC was performed, followed by PLS-DA. Venn diagrams and heat maps were generated to visualize and compare differential metabolites. Differentially expressed metabolites were subjected to Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis to explore their potential biological functions and pathway associations. Spearman’s correlation analysis was performed to explore the relationships among behavioral outcomes, dopaminergic neuronal markers, gut bacterial abundance, and serum metabolites. The parameters included open-field test performance, grip strength, striatal TH expression, fecal Parabacteroides abundance, and serum levels of the metabolites theophylline, inosine, hypoxanthine, and xanthine. Correlation coefficients (r) were calculated using the corrplot package in R, and nonparametric p-values were adjusted for multiple comparisons. The resulting correlation matrix was visualized as a clustered heatmap. Untargeted metabolomics analyses were conducted by Hangzhou LC-Bio Technology Co., Ltd, and analyzed using the OmicStudio cloud platform (https://www.omicstudio.cn). BV2 cell culture BV2 cells were obtained from Servicebio Biotechnology Co., Ltd. (Wuhan, China). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Servicebio, China) supplemented with 10% fetal bovine serum (FBS, Servicebio, China) and 1% penicillin–streptomycin (Servicebio, China). Cultures were maintained at 37 °C in a humidified atmosphere containing 5% CO₂. For viability analysis, BV2 cells were seeded into 96-well plates at a density of 5 × 10⁴ cells/mL and allowed to adhere overnight. To establish the injury model, cells were exposed to MPP⁺ (0–4 mM) for 24 h. For intervention assays, cells were pretreated with bacteria-free P. distasonis culture supernatant (Pd-CS; 2–10%) or individual metabolites (hypoxanthine, xanthine, inosine, theophylline) for 6 h, followed by incubation with 2 mM MPP⁺ for another 24 h. Cell Counting Kit-8 viability assay Cell viability was determined using a CCK8 (Servicebio, China) according to the manufacturer’s instructions. After treatment, 10 µL of CCK-8 solution was added to each well, and cells were incubated for 2 h at 37 °C. The absorbance at 450 nm was measured using a microplate reader (BioTek, USA). Results were expressed as a percentage of the control group. Statistical analyses Statistical analyses were performed using the GraphPad Prism 9.0 software (GraphPad Software, Inc., San Diego, CA, USA). Data are presented as the mean ± standard error of the mean (SEM). One-way ANOVA followed by Tukey’s post-hoc test was used for multiple group comparisons. For non-normally distributed data, the Kruskal–Wallis test followed by Dunn’s post-hoc test was used. Statistical significance was set at p < 0.05. Depending on the experiment the significance levels were p < 0.05, *p < 0.01, **p < 0.001, and ***p < 0.0001. Results ISL improves behavioral dysfunction and dopaminergic neuron loss in MPTP-treated mice To evaluate the therapeutic effects of ISL in PD, we used a mouse model of PD induced by chronic application of MPTP. As illustrated in the experimental design (Fig. 1A). ISL improves behavioral dysfunction and dopaminergic neuron loss in MPTP-treated mice. (A) Experimental timeline. (B) Representative movement tracks in the open field test. Quantification of total distance (C) and average velocity (D) in the open field test; (n = 6). (E) Forelimb grip strength test; (n = 6). (F) Representative TH immunostaining in the substantia nigra. Scale bar = 100 μm. (G) Quantification of TH+ optical density; (n = 5). (H) Western blot of TH in the striatum. (I) Quantification of TH protein expression normalized to GAPDH; (n = 6). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns: not significant To assess the effects of ISL on motor function in the PD model mice, behavioral tests were performed following drug administration. The mice underwent standard behavioral training prior to testing to minimize variability due to stress or unfamiliarity. The open field test was used to evaluate spontaneous locomotor activity by measuring the total distance traveled and average movement speed, which reflect general activity levels and exploratory behavior [17]. Mice exposed to MPTP exhibited marked hypoactivity, with evident reductions in the total travel distance and average velocity compared to control animals. These findings suggest successful induction of motor impairment by MPTP. Interestingly, oral administration of ISL at medium and high doses markedly improved these locomotor parameters in MPTP-treated mice, indicating that ISL may counteract MPTP-induced motor deficits (Fig. 1B–D). The grip strength test was performed to assess the neuromuscular function by quantifying the forelimb force exerted when the animal instinctively grasped a horizontal bar [18]. The grip strength test further supported the protective effects of ISL. Mice in the MPTP group showed a pronounced loss of forelimb muscle strength, whereas those receiving ISL displayed a clear improvement, suggesting the preservation of neuromuscular function (Fig. 1E). To investigate whether ISL also protected dopaminergic neurons, we examined the expression of TH, a rate-limiting enzyme in dopamine synthesis and a reliable marker for dopaminergic neurons [19]. TH is abundantly expressed in the substantia nigra and striatum, and its reduction reflects neuronal loss in PD models. TH immunostaining revealed fewer TH-positive neurons in the substantia nigra of the MPTP-treated mice. In contrast, the MPTP + ISL groups retained more TH-positive cells, suggesting a protective effect (Fig. 1F, G). Western blot analysis of striatal tissue confirmed a marked decrease in TH protein levels in the MPTP group, which was partially restored by ISL administration (Fig. 1H, I). Taken together, these findings indicate that ISL treatment is associated with improvements in motor function and partial preservation of dopaminergic neurons in MPTP-induced PD mice in a dose-dependent manner. To ensure that ISL itself did not affect normal locomotor or neuronal function, additional experiments were conducted in healthy mice. ISL administration alone did not significantly alter open-field performance, grip strength, or TH expression, indicating that the observed neuroprotective effects are specific to the pathological context of MPTP-induced PD (Additional file 1: Fig. S1). ISL attenuates peripheral inflammation and improves gastrointestinal function in MPTP-treated mice Given that ISL alleviated motor dysfunction and supported dopaminergic neuron integrity in the central nervous system, we investigated whether it could also modulate peripheral pathological features commonly associated with PD, such as systemic inflammation and gastrointestinal disturbances. ELISA analysis revealed that MPTP treatment significantly elevated serum levels of the pro-inflammatory cytokines TNF-α, IL-1β, and IL-6 compared to control group. ISL administration effectively suppressed these inflammatory markers in MPTP-treated mice, indicating that ISL exerted anti-inflammatory effects beyond the central nervous system (Fig. 2A–C). ISL attenuates systemic inflammation and improves colonic pathology in MPTP-treated mice. (A–C) ELISA results for TNF-α (A), IL-1β (B), and IL-6 (C) in serum; (n = 5–6). (D,E) Representative colons (D) and colon length quantification (E); (n = 5–6). (F) Number of fecal pellets produced within 15 min. (n = 6). (G) Representative HE and AB–PAS staining of colon sections. Scale bar = 100 μm. (H, I) Histological score (H) and AB–PAS positive goblet cell quantification (I); (n = 5). Data are shown as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns: not significant Gross anatomical inspection of the colon showed that MPTP led to a reduction in colon length, a commonly observed indicator of colonic inflammation [20]. This shortening was markedly ameliorated in the MPTP + ISL groups, suggesting that ISL helped maintain colonic structural integrity (Fig. 2D, E). To assess intestinal motility, we recorded the number of fecal pellets excreted over a 15-min period. MPTP-treated mice produced fewer fecal pellets, whereas ISL intervention increased the output, indicating an improved gastrointestinal transit function (Fig. 2F). Histological assessments further supported these findings. HE staining of colon sections revealed epithelial damage and mucosal disruption in the MPTP group, whereas AB–PAS staining indicated a reduction in goblet cells and mucus secretion. These pathological changes were alleviated by ISL treatment, as evidenced by restored tissue architecture and goblet cell abundance (Fig. 2G). Quantitative analysis showed that ISL significantly reduced the histological damage scores and increased AB–PAS positive goblet cell area (Fig. 2H, I). To further evaluate the intestinal epithelial barrier integrity, Western blot analysis was performed to detect the tight-junction proteins Occludin and Claudin-1 (Additional file 1: Fig. S2). The protein levels of Occludin and Claudin-1 in the colon were markedly reduced in the MPTP group. In contrast, ISL treatment partially restored the expression of these tight junction proteins, approaching the levels observed in the control group. These observations suggest that ISL may help preserve the epithelial barrier function disrupted by MPTP treatment. Collectively, these results indicate that ISL alleviates peripheral inflammation and intestinal dysfunction in MPTP-treated PD mice. FMT improves motor deficits and reduces neuroinflammation and peripheral inflammation in MPTP-treated mice To explore whether the gut microbiota contributes to the neuroprotective effects in PD, we performed FMT using fecal samples from healthy donor mice pretreated with ISL. As shown in the experimental timeline (Fig. 3A), mice were randomly assigned to three groups: control, MPTP, and MPTP + FMT. After chronic MPTP exposure, the FMT group received oral gavage of fecal suspensions from healthy mice for 2 weeks. FMT improves behavioral dysfunction, dopaminergic neuron loss, systemic inflammation, and peripheral inflammation in MPTP-treated mice. (A) Experimental timeline. (B) Representative movement tracks in the open field test. (C, D) Quantification of total distance (C) and average velocity (D) in the open field test; (n = 6). (E) Forelimb grip strength test; (n = 6). (F) Western blot of TH in the striatum. (G) Quantification of TH protein expression normalized to GAPDH; (n = 6). (H) Representative TH immunostaining in the substantia nigra. Scale bar = 100 μm. (I) Quantification of TH+ optical density; (n = 5). (J–L) ELISA results for TNF-α (J), IL-1β (K), and IL-6 (L) in serum; (n = 6). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 In the open field test, MPTP administration resulted in a substantial reduction in the total distance traveled and average velocity, indicating impaired spontaneous locomotion. FMT treatment significantly improved these parameters, suggesting that gut microbial modulation could partially restore motor activity (Fig. 3B–D). Likewise, grip strength measurements showed decreased neuromuscular function in the MPTP group, which was ameliorated by the FMT intervention (Fig. 3E). Western blot analysis of striatal tissue confirmed that TH protein levels were significantly reduced by MPTP, but were restored upon FMT treatment (Fig. 3F, G). TH immunohistochemistry revealed marked neuronal loss in the substantia nigra of MPTP-treated mice, which was partially preserved in the FMT group (Fig. 3H, I). In addition, FMT exerted notable peripheral protective effects. MPTP administration elevated serum TNF-α, IL-1β, and IL-6 levels, which were significantly reduced after FMT intervention (Fig. 3J–L). Collectively, these findings indicate that FMT mitigates MPTP-induced motor deficits, neuroinflammation, and peripheral inflammation, suggesting a key role of gut microbiota in mediating the neuroprotective effects observed in PD. ISL modulates gut microbiota composition in MPTP-treated mice To investigate whether the therapeutic effects of ISL were associated with gut microbial regulation, 16 S rDNA sequencing of fecal samples from the control, MPTP, and MPTP + ISL groups was performed. Alpha diversity analysis was performed to assess the within-sample microbial complexity. As shown in Fig. 4A, the overall comparison using one-way ANOVA revealed a significant difference in species richness among the three groups. The Simpson index, which reflects species evenness, remained relatively stable across groups. Beta diversity was evaluated using PCoA based on the Bray–Curtis distance. As shown in Fig. 4B, although partial overlaps in sample distribution were observed among the three groups, the overall microbial community composition exhibited group-dependent clustering tendencies. In particular, MPTP-treated mice tended to deviate from the control group, whereas ISL treatment partially shifted the microbial profile toward that of the control group. PERMANOVA confirmed the significant intergroup differences (R = 0.369, p = 0.001). To identify the key microbial taxa contributing to group differences, LEfSe analysis was performed across the three groups (Fig. 4C). The MPTP group showed an increased abundance of certain bacterial genera, such as Faecalibaculum and Turicibacter, indicating marked gut microbiome changes. In contrast, the control group was enriched with several commensal and potentially beneficial genera, including Muribaculum and Parabacteroides. The ISL-treated group demonstrated microbial features that partially overlapped with those of both the control and MPTP groups, with enrichment of genera such as Staphylococcus and Corynebacterium, which have been implicated in mucosal balance and anti-inflammatory activities. 16S rDNA sequencing analysis of gut microbiota in mice. (A) Alpha diversity analysis using Richness and Simpson indices; (n = 6). (B) PCoA based on Bray–Curtis distance showing β-diversity among groups. (C) Linear discriminant analysis effect size identifying differentially abundant taxa among control, MPTP, and MPTP + ISL groups. (D) Relative abundance of representative genera with significant intergroup differences; (n = 6). (E) Heatmap of genus level microbial abundance. Each column represents an individual sample and each row corresponds to a bacterial genus. Color intensity reflects relative abundance. Blue: control group; Red: MPTP group; Green: MPTP + ISL group. Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns: not significant At the genus level, Parabacteroides was notably enriched in the control group and markedly reduced in the MPTP-treated mice, consistent with gut barrier disruption and inflammatory activation. ISL intervention was associated with a mild restoration of Parabacteroides abundance, as supported by bar plots and heatmap visualization (Fig. 4D, E). This pattern, along with the modulation of other host-beneficial genera, such as Bacteroides and Muribaculum, suggests that ISL may partially rebalance microbial communities toward a healthier state. In contrast, the MPTP group exhibited increased abundance of Turicibacter, Faecalibaculum and Bifidobacterium. Collectively, these results indicate that ISL intervention can partially reverse MPTP-induced gut microbial alterations, thereby supporting its potential role in gut–brain axis modulation in PD. Intervention with viable P. distasonis alleviates motor deficits and dopaminergic neuron loss in MPTP-induced PD mice Given that ISL restored gut microbial balance and alleviated PD phenotypes in previous experiments, we aimed to determine whether direct microbiota modulation using a specific bacterial strain could recapitulate these effects. Based on the 16S rDNA sequencing results, P. distasonis was selected as a probiotic candidate for intervention. To investigate the neuroprotective effects of P. distasonis, mice were divided into four groups: control, MPTP, MPTP + Pd, and MPTP + HK-Pd. As shown in the experimental timeline (Fig. 5A), MPTP was administered intraperitoneally to induce Parkinsonian pathology. Subsequently, live or heat-killed P. distasonis was delivered orally for 2 weeks. P. distasonis treatment alleviates motor deficits and dopaminergic neuron loss in MPTP-induced PD mice. (A) Experimental timeline. (B) Representative movement tracks in the open field test. (C, D) Quantification of total distance (C) and average velocity (D) in the open field test; (n = 6). (E) Forelimb grip strength test; (n = 6). (F) Representative TH immunostaining in the substantia nigra. Scale bar = 100 μm. (G) Quantification of TH+ optical density; (n = 5). (H) Western blot of TH in the striatum. (I) Quantification of TH protein expression normalized to GAPDH; (n = 6). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns: not significant Open field test results demonstrated that MPTP administration significantly reduced locomotor activity, whereas live P. distasonis intervention markedly improved total distance traveled and average velocity (Fig. 5B–D). Similarly, grip strength analysis showed a significant improvement in neuromuscular function in the MPTP + Pd group compared with that in the MPTP group (Fig. 5E). Immunohistochemistry and western blot analyses further revealed that live P. distasonis preserved TH-positive dopaminergic neurons in the substantia nigra and striatum (Fig. 5F–I), whereas heat-killed P. distasonis did not provide comparable protection. To further confirm the dose-dependent neuroprotective effect, additional experiments were performed using low (107 CFU/mL), medium (108 CFU/mL), and high (109 CFU/mL) doses of live P. distasonis. Western blot analysis showed that TH protein expression increased in a dose-dependent manner, with the highest level observed in the MPTP + Pd-H group (Additional file 1: Fig. S3). These findings suggest that live P. distasonis supplementation effectively alleviates motor dysfunction and dopaminergic neurodegeneration in a mouse model of PD, highlighting its therapeutic potential as a gut-derived intervention. P. distasonis alleviates neuroinflammation and restores blood–brain barrier (BBB) integrity in MPTP-treated mice Building on the behavioral and neuroprotective benefits of P. distasonis, we examined its effects on neuroinflammation and BBB integrity, which are closely linked to PD pathology. Immunofluorescence analysis of the substantia nigra revealed that MPTP treatment induced significant astrocyte and microglial activation, as indicated by enhanced GFAP and Iba1 staining (Fig. 6A–C). Glial activation was markedly attenuated in the MPTP + Pd group that received live P. distasonis, whereas heat-killed P. distasonis failed to elicit a similar effect. P. distasonis attenuates neuroinflammation and restores BBB integrity. (A) Representative double immunofluorescence images of GFAP (green) and Iba1 (red) in the substantia nigra. Scale bar = 50 μm. (B, C) Quantification of GFAP+ (B) and Iba1+ (C) optical density; (n = 5) (D) Western Blot of Occludin, Claudin1, and GAPDH. (E, F) Quantification of Occludin (E) and Claudin1 (F) normalized to GAPDH; (n = 6). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001; ns: not significant To assess BBB status, we evaluated the expression of the tight junction proteins Occludin and Claudin-1 in the midbrain. MPTP administration resulted in the downregulation of both proteins, reflecting barrier disruption. Notably, only live P. distasonis supplementation reversed these reductions (Fig. 6D–F), suggesting that its anti-inflammatory effect also contributes to the preservation of BBB integrity. These findings indicate that viable P. distasonis mitigates neuroinflammatory responses and supports the structural preservation of the BBB in PD mice. P. distasonis improves intestinal integrity and motility in MPTP-treated mice To explore whether P. distasonis affects gastrointestinal health, we assessed the colon structure and motility in MPTP-treated mice. MPTP induces a clear shortening of the colon, which is often associated with intestinal inflammation. This was significantly improved in the MPTP + Pd group, but not in the heat-killed bacteria group (Fig. 7A, B). Similarly, the number of fecal pellets produced in 15 min was reduced in MPTP mice and restored by live P. distasonis treatment (Fig. 7C), indicating enhanced gut motility. P. distasonis improves intestinal integrity and motility in MPTP-treated mice. (A, B) Representative colons (A) and colon length quantification (B); n = 6. (C) Number of fecal pellets produced within 15 min; (n = 6). (D) Representative HE and AB–PAS staining of colon sections. Scale bar = 100 μm. (E, F) Histological injury score (E) and AB–PAS positive goblet cell quantification (F); (n = 5). Data are presented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns: not significant Histological analysis confirmed the protective effects of P. distasonis on the intestinal structure. HE staining showed epithelial damage and AB–PAS staining revealed a loss of goblet cells after MPTP treatment, both of which were alleviated by live bacteria (Fig. 7D–F). We used 16S rRNA gene sequencing to evaluate fecal microbiota in control, MPTP, and MPTP + Pd mice and assess community changes after bacterial intervention. Alpha diversity measured by the Simpson index was similar across groups; the Kruskal–Wallis test indicated no significant difference (Additional file 1: Fig. S4A). PCoA showed group-dependent separation, with P. distasonis partially shifting the community toward the control profile (Additional file 1: Fig. S4B). A genus-level clustered heatmap of the top taxa, displayed as row-scaled Z-scores, highlighted distributed compositional adjustments following P. distasonis treatment (Additional file 1: Fig. S4C). These findings indicate that P. distasonis supplementation effectively restores intestinal structure and motility impaired by MPTP, accompanied by partial normalization of gut microbial composition. Together, these results suggest that P. distasonis plays a protective role in maintaining gut homeostasis and alleviating gastrointestinal dysfunction associated with Parkinson’s pathology. ISL and P. distasonis reshape serum metabolomic profiles and modulate metabolic pathways in MPTP-treated mice To elucidate the systemic metabolic changes underlying the therapeutic effects of ISL and P. distasonis, we performed untargeted metabolomic profiling of mouse serum in the control, MPTP, MPTP + ISL, and MPTP + Pd groups. PCoA using PLS-DA revealed distinct clustering among the four groups, indicating marked alterations in the serum metabolite composition induced by MPTP, which were partially reversed by either ISL or P. distasonis treatment (Fig. 8A and Additional file 1: Fig. S5A). Untargeted metabolomic profiling reveals distinct serum metabolic signatures among groups. (A) PLS-DA score plot showing distinct separation of the four groups based on serum metabolic profiles. (B) Venn diagram showing overlapping and unique differential metabolites between the ISL vs. MPTP and P. distasonis vs. MPTP comparisons. (C) Heatmap of representative differential metabolites. The color gradient indicates relative abundance (red: high, blue: low). (D) KEGG pathway enrichment analysis of differential metabolites, categorized by biological function. Bars indicate the number of metabolites involved in each pathway A comparative analysis identified 107 overlapping differential metabolites between the ISL vs. MPTP and P. distasonis vs. MPTP comparisons, whereas 125 and 112 metabolites were uniquely altered by ISL and P. distasonis, respectively (Fig. 8B). These results indicate that although both treatments share some regulatory targets, they also influence distinct metabolic pathways. Representative metabolite heat maps demonstrated that MPTP disrupted a range of bioactive molecules, including inosine, xanthine, indole-3-propionic acid, and formononetin, many of which were restored by either intervention (Fig. 8C). To further characterize the specific metabolic effects of P. distasonis, we directly compared the treated group with the MPTP group. A total of 215 significantly altered metabolites were identified, including 124 upregulated and 91 downregulated metabolites (Additional file 1: Fig. S5B). Heatmap visualization highlighted the unique metabolic shifts induced by P. distasonis, particularly in flavonoids, nucleoside derivatives, and amino acid intermediates (Additional file 1: Fig. S5C). Pathway enrichment analysis based on KEGG annotation revealed that the differentially abundant metabolites were primarily involved in lipid metabolism (e.g., bile secretion), ABC transporter systems, ferroptosis, efferocytosis, and microbial and amino acid biosynthetic processes. These pathways are linked to cellular stress, detoxification, inflammation, and host-microbiota interactions, suggesting that the therapeutic effects of P. distasonis may extend beyond the gut and brain, and include systemic metabolic reprogramming (Fig. 8D and Additional file 1: Fig. S5D). Taken together, these findings suggest that ISL and P. distasonis ameliorate MPTP-induced systemic metabolic disturbances through both shared and distinct biochemical pathways, contributing to their protective effects against PD. Functional validation of metabolites in BV2 microglia Spearman correlation analysis revealed that Parabacteroides abundance and its related metabolites were positively correlated with behavioral performance and striatal TH expression, suggesting a potential functional link between gut-derived metabolites and neuroprotection (Fig. 9A). Effects of P. distasonis culture supernatant and metabolites on BV2 viability, and correlations with systemic/behavioral readouts. (A) Spearman correlations between fecal Parabacteroides relative abundance and key serum metabolites (hypoxanthine, xanthine, inosine, theophylline), as well as behavioral/pathological outcomes (open-field test, grip-strength test, TH). Heatmap cells show correlation coefficients with hierarchical clustering; *p < 0.05, **p < 0.01. (B) MPP+ dose-response (0–4 mM, 24 h) in BV2 cells. (C) Effect of bacteria-free culture supernatant of P. distasonis (Pd-CS) on BV2 cell viability under 2 mM MPP+. (D) Effects of hypoxanthine, xanthine, inosine, and theophylline under 2 mM MPP+. Data are presented as mean ± SEM. One-way ANOVA with Dunnett’s multiple comparisons test: vs. 0 mM in (B), vs. MPP+ in (C) and (D).*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 BV2 cells were exposed to MPP+ (0–4 mM, 24 h) to establish an PD injury model; viability declined in a dose-dependent manner, and 2 mM MPP+ was used thereafter (Fig. 9B). Under this condition, treatment with bacteria-free culture supernatant of P. distasonis (Pd-CS) increased CCK-8 viability in a concentration-dependent fashion (Fig. 9C). Metabolomics-guided candidates including hypoxanthine, xanthine, inosine, and theophylline similarly mitigated MPP+-induced cytotoxicity (Fig. 9D). These data support the conclusion that ISL and P. distasonis-derived metabolites contribute to the protective effects. Discussion This study systematically investigated the neuroprotective effects of ISL in a mouse model of chronic MPTP-induced PD and uncovered the gut microbiota-mediated mechanism underlying its efficacy (Fig. 10). By combining direct ISL administration, FMT from ISL-treated donors, and targeted supplementation with P. distasonis, we established a functional causal link between microbial remodeling and neurological improvement. Our data demonstrate that ISL ameliorates motor deficits, preserves dopaminergic neurons, and also restores intestinal barrier integrity, which is recapitulated, at least in part, by microbiota-based interventions. These findings indicate that the gut microbiota is an essential intermediary in the therapeutic action of ISL. Isoliquiritigenin alleviates MPTP-induced PD pathology via the gut–brain axis. In MPTP-induced PD mice, gut microbiome changes (notably reduced Parabacteroides), epithelial barrier disruption, and systemic inflammation (elevated TNF-α, IL-1β, IL-6) contribute to dopaminergic neuronal degeneration in the brain. Oral administration of ISL reverses these pathological changes by enriching Parabacteroides, restoring intestinal integrity, suppressing peripheral inflammation, and preserving nigrostriatal dopaminergic neurons, supporting a mechanism involving the gut–metabolism–neuroinflammation axis While ISL has been traditionally recognized for its antioxidant and anti-inflammatory activities, our findings suggest that its neuroprotective effects may also involve modulation of the gut microbiota. 16S rDNA sequencing revealed that ISL significantly altered the microbial community structure in MPTP-treated mice and partially restored the features of a healthy gut profile. Specifically, ISL promoted the enrichment of potentially beneficial taxa, including Parabacteroides, Bacteroides and Muribaculum. These genera have been associated with the production of short-chain fatty acids, immune regulation, and mucosal barrier maintenance, suggesting that ISL may exert protective effects through microbial-derived homeostatic mechanisms [21–23]. Although several beneficial genera were modulated by ISL, Parabacteroides was prioritized for functional validation because of its potential immunometabolic functions and relevance as a human commensal [24]. Notably, Parabacteroides was consistently abundant in healthy control mice, but was reduced following MPTP administration, indicating its role in maintaining microbial and mucosal homeostasis. ISL intervention partially preserved its abundance, suggesting its involvement in the therapeutic effects. Based on this pattern, we selected P. distasonis, the dominant commensal species within the genus, as a representative candidate for subsequent validation. P. distasonis has been previously demonstrated to attenuate inflammation and regulate bile acid and short-chain fatty acid metabolism in various diseases [10, 14, 25, 26]. These features, together with ISL-associated recovery, support its functional relevance in gut–brain axis modulation in PD. The functional validation in our study further supports the involvement of P. distasonis. Oral supplementation with viable P. distasonis recapitulated the protective effects of ISL in PD mice, including improvements in motor function, reduction of gliosis, preservation of tight junction proteins in the midbrain, and partial restoration of serum metabolic balance. In contrast, heat-killed bacteria had a negligible effect, suggesting that live bacteria and their active metabolites are necessary for therapeutic effects. Additionally, all three interventions (ISL, FMT, and P. distasonis) consistently demonstrated protective effects on the intestinal barrier, as evidenced by histological improvements in colonic tissue. ISL treatment further confirmed this effect by upregulating colonic tight junction proteins, whereas P. distasonis treatment restored BBB integrity. Disruption of both the intestinal barrier and BBB has been increasingly recognized as a hallmark of PD pathogenesis [27, 28]. Impaired gut barrier function facilitates the translocation of microbial components, such as lipopolysaccharide, which can trigger systemic inflammation and prime microglial activation in the brain [29]. Concurrently, BBB breakdown allows peripheral cytokines and immune cells to enter the central nervous system, exacerbating neuroinflammatory responses and neuronal degeneration [27]. The observed reinforcement of mucosal and neurovascular barriers in this study suggests that ISL and P. distasonis act locally within the gut and may also contribute to systemic immune containment and central protection. These findings highlight the significance of epithelial–endothelial barrier maintenance as a mechanistic bridge linking gut microbiota modulation and neurological outcomes in PD. In addition to reshaping the gut microbiota and restoring barrier integrity, P. distasonis reduced the expression of GFAP and Iba1 in the substantia nigra, indicating suppressed astrocytic and microglial activation. Glial activation, particularly sustained microglial and astrocytic reactivity, is a well-established contributor to dopaminergic neuron degeneration in PD [30, 31]. When chronically activated, microglia release pro-inflammatory cytokines, reactive oxygen species, and nitric oxide, all of which can exacerbate neuronal injury [32]. In turn, astrocytes modulate neuroinflammation through cytokine signaling and affect BBB function [33]. The observed attenuation of glial markers suggests that P. distasonis exerts neuroprotective effects by modulating the central inflammatory milieu. These results highlight that microbial interventions can influence the gut environment and downstream neuroimmune responses, supporting a role for gut–brain communication in shaping PD pathogenesis and therapy. Notably, both ISL and P. distasonis altered host serum metabolite profiles, particularly in pathways related to bile acid metabolism, purine derivatives (e.g., inosine), and indole-containing compounds. These metabolites are associated with microbes and have been increasingly recognized as key mediators in gut–brain signaling, immune regulation, and neuroinflammation. For instance, bile acids can act as signaling molecules modulating microglial activation and intestinal integrity [34–36]. Inosine promotes neuroregeneration and modulates T cell activity [37, 38]. Indole derivatives alleviate intestinal inflammation and modulate gut microbiota [39, 40]. Although this study did not directly confirm the causality of the individual metabolites, their consistent modulation across both interventions supports the hypothesis that ISL may partially exert neuroprotective effects through microbiota-derived metabolic reprogramming. These findings highlight that systemic metabolic remodeling is a potentially crucial intermediary between gut microbial modulation and central nervous system outcomes in patients with PD. Although 16S rRNA sequencing revealed that ISL treatment increased the relative abundance of Parabacteroides in MPTP-treated mice, this approach cannot determine causality or species-level specificity. Whether ISL selectively promotes the growth of P. distasonis remains to be verified. Future studies using in vitro co-culture assays and gnotobiotic mouse models colonized with P. distasonis will be essential to clarify this direct relationship and further elucidate the mechanistic link between ISL and gut microbial modulation. Taken together, our findings suggest a mechanistic framework wherein ISL improves PD phenotypes, partly through microbiota remodeling, with P. distasonis functioning as a critical effector. The findings expands the understanding of how natural compounds may exert host benefits beyond direct receptor interactions, highlighting microbiota composition and function as integral components of therapeutic efficacy. Future studies should validate the specific bacterial metabolites and host receptors involved in this pathway, and assess the translational potential of microbiota-targeted strategies in clinical PD populations. Conclusion ISL ameliorates Parkinsonian symptoms by modulating the gut microbiota, with P. distasonis identified as a key mediator. ISL reshapes gut microbial ecology, enriches Parabacteroides, and leads to downstream improvements in neuroinflammation, barrier integrity, and systemic metabolism.These findings provide mechanistic insights into how dietary flavonoids exert neuroprotection via the gut–brain axis and suggest that microbial enrichment, particularly of P. distasonis, may serve as a novel therapeutic strategy for PD and related neurodegenerative disorders. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Abbreviations Alcian Blue/Periodic Acid–Schiff Amplicon sequence variant Blood–brain barrier Bovine serum albumin Colony-Forming Unit Central nervous system 4′,6-diamidino-2-phenylindole Enzyme-linked immunosorbent assay Fecal microbiota transplantation Glial fibrillary acidic protein Hematoxylin–eosin Horseradish peroxidase Ionized calcium-binding adapter molecule 1 Interleukin-1 beta Interleukin-6 Isoliquiritigenin Kyoto Encyclopedia of Genes and Genomes Liquid chromatography–mass spectrometry Linear discriminant analysis Linear discriminant analysis effect size 1-Methyl-4-phenylpyridinium ion 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine Phosphate-buffered saline Principal coordinate analysis Parkinson’s disease Partial least squares discriminant analysis Quality control Standard error of the mean Tyrosine hydroxylase Tumor necrosis factor alpha Author contributions Conceptualization: ML, XZ, LH and XS. Data curation: ML, XZ, JL. Formal analysis: MZ and JX. Funding acquisition: ML, XZ, LH and XS. Investigation: ML, JL and XS. Methodology: ML, LH, XZ, XS. Project administration: LH, JL and XS. Resources: ML and XS. Supervision: JW and SY. Validation: ML, JW and MZ. Visualization: JX and XS. Writing-original draft: ML, XS and LH. Writing–review & editing: ML, XS. Funding This research was supported by The Traditional Chinese Medicine Science and Technology Project of Zhejiang Provincial under Grant No. 2024ZF031 and 2024ZF165, The Medical Health Science and Technology Project of Zhejiang Provincial under Grant No. 2023RC287 and 2023KY345, Shaoxing Health Science and Technology Plan (2023SKY015 and 2023A14011). Data availability The original contributions presented in the study are included in the article/ Additional files, further inquiries can be directed to the corresponding author/s. Declarations Ethics approval and consent to participate All animal protocols were approved by the Animal Care and Use Committee of Shaoxing Peoples’ Hospital (2025Z068). Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Mengyun Li and Xiaobing Zhang contributed equally to this work. Contributor Information Lingyan He, Email: Lingyanh_1223@usx.edu.cn. Xian Shao, Email: 18315318073@163.com. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The original contributions presented in the study are included in the article/ Additional files, further inquiries can be directed to the corresponding author/s.