Immunomodulatory activity of a water-soluble polysaccharide extracted from mussel on cyclophosphamide-induced immunosuppressive mice models

Cyclophosphamide (Cy), a widely used anticancer alkylating agent, causes notable gastrointestinal toxicity, including damage to intestinal mucosa and dysbiosis, which can promote bacterial translocation and immunosuppression. Intestinal microbiota play a central role in gut health, barrier integrity, and immune regulation, partly via production of short-chain fatty acids (SCFAs). Natural polysaccharides exhibit diverse bioactivities (anti-inflammatory, antioxidant, antiviral, antitumor) and can modulate intestinal mucosal immunity and microbiota, functioning as potential prebiotics. Mussels, rich in bioactive polysaccharides, have reported immunoregulatory and metabolic benefits. The study hypothesized that a mussel-derived polysaccharide (MP) could alleviate Cy-induced immunosuppression and intestinal dysbiosis, restoring barrier function, immune homeostasis, and beneficial microbiota and their metabolites.

The paper contextualizes prior findings that polysaccharides from various sources (e.g., longan pulp, kidney beans, Tremella, Codonopsis pilosula, Ulva pertusa) can ameliorate Cy-induced mucosal injury, enhance SIgA secretion, improve immune organ indexes, and rebalance Th1/Th2 cytokines. Marine animal-derived polysaccharides have been shown to activate NF-κB signaling and favorably modulate gut microbiota composition and SCFAs. Dysbiosis in Cy-treated models typically includes increased Firmicutes/Bacteroidetes ratio and elevated Proteobacteria (e.g., Desulfovibrio), whereas beneficial genera such as Lactobacillus are reduced. These findings support evaluating MP for immunomodulation and microbiota remodeling in Cy-induced immunosuppression.

Materials: Cyclophosphamide (Sigma-Aldrich). ELISA kits for SIgA, LPS (Huamei), DAO (Nanjing SenBeiJia), IL-2, IFN-γ, IL-4, IL-10 (Boshide). MP (prepared previously by authors) had average MW 4.25 × 10^3 Da and was composed of glucose (0.995), galactose (0.003), glucosamine hydrochloride (0.001), and galactose hydrochloride (0.001). Animals and design: Forty male SPF BALB/c mice (5 weeks, 15–18 g) were acclimated and randomized (n=8/group) into: Control; Cy; Cy+LMP (low-dose MP 300 mg/kg BW/day); Cy+HMP (high-dose MP 600 mg/kg BW/day). Oral gavage days 1–21 per Table 1. Cy was injected i.p. 50 mg/kg BW/day on days 18–21 to induce immunosuppression (control received saline). On day 21, mice were weighed, euthanized, and samples collected: organs (liver, thymus, spleen) for organ index; blood for serum; ileum for ELISA, qRT-PCR, Western blot, SEM/TEM; cecal contents for 16S rRNA sequencing and SCFA quantification. All procedures were approved by the Ethics Committee of Zhejiang University of Technology. Histology/ultrastructure: Ileum fixed for SEM and TEM. SEM: dehydration, critical point drying, sputter coating, observed at 35 Kx. TEM: fixation, osmium tetroxide, dehydration, embedding, ultrathin sectioning, uranyl acetate/lead citrate staining; observed at 20 Kx. Biochemical assays: Serum SIgA, DAO, LPS quantified by ELISA per manufacturer protocols. Ileal cytokines (IFN-γ, IL-2, IL-4, IL-10) measured by ELISA on tissue homogenates. Gene expression (qRT-PCR): Ileal tissue homogenized for RNA extraction (Tiangen kits), cDNA synthesized, and qRT-PCR performed (SYBR Green) for IL-2, IFN-γ, IL-4, IL-10, mucin-2, occludin, claudin-1, ZO-1 with β-actin as reference. Primer sequences provided (Table 2). Relative expression analyzed. Western blot: Ileal proteins extracted (RIPA, PMSF, protease inhibitors), quantified (BCA), separated by SDS-PAGE, transferred to PVDF, blocked, probed with primary antibodies to NF-κB pathway components (p65, p-p65, IκBα, p-IκBα) and GAPDH, HRP-secondary, ECL detection, densitometry by ImageJ. Experiments in triplicate. Microbiota profiling: DNA extracted from cecal contents (six mice per Control, Cy, and Cy+HMP groups). V3–V4 16S rRNA region amplified (KAPA HiFi) and sequenced on Illumina MiSeq. Bioinformatics: quality control (QIIME, Fastp, Cutadapt), OTU clustering at 97% similarity (UPARSE, SILVA v132), taxonomy assignment (RDP Classifier), alpha diversity (Mothur), beta diversity (QIIME2), PCA, Venn diagrams, relative abundance at phylum/genus, correlation analyses (Spearman, network). SCFA quantification: Fecal SCFAs (acetic, propionic, isobutyric, butyric, isovaleric, valeric acids) extracted with internal standard (hexanoic acid-d3) and analyzed by GC-MS (Agilent 7890A/5975C). Identification by retention times against standards across 0.01–1000 µg/mL. Statistics: Data expressed as mean ± SD, analyzed by one-way ANOVA (SPSS 25.0, GraphPad Prism 9.1.1). p < 0.05 considered significant. Correlations assessed by Spearman.

• General/organ indices: Cy caused body weight loss and reduced food intake; MP (300 or 600 mg/kg/day) attenuated weight loss and increased intake versus Cy. Liver, thymus, and spleen indices decreased significantly after Cy (p < 0.05); MP improved liver and thymus indices significantly, with a non-significant upward trend in spleen index.
• Intestinal barrier: SEM/TEM showed Cy-disrupted microvilli and loss of desmosomes; MP, especially high-dose, partially restored brush border architecture. Serum DAO and LPS increased with Cy and decreased dose-dependently with MP (p < 0.05). Ileal mRNA of tight junction and mucosal barrier markers (ZO-1, occludin, claudin-1, mucin-2) was significantly suppressed by Cy and significantly upregulated by MP; mucin-2 showed marked enrichment in Cy+HMP.
• Mucosal immunity: SIgA secretion decreased with Cy and was significantly increased by MP (both doses; p < 0.05, no significant difference between doses).
• Cytokines/Th1–Th2 balance: Ileal IL-2, IFN-γ (Th1) and IL-4, IL-10 (Th2) were significantly reduced by Cy and significantly increased by MP in a dose-dependent manner at both protein (ELISA) and mRNA levels; in Cy+HMP, mRNA levels approached control.
• NF-κB signaling: Cy inhibited NF-κB pathway, lowering p-IκBα and p-p65; MP significantly increased phosphorylation of IκBα and p65 dose-dependently (p < 0.05), indicating pathway activation.
• Microbiota diversity/composition: Alpha diversity (Chao1, Shannon) decreased with Cy and improved with MP; Simpson index showed non-significant positive change. Venn diagram: 808 OTUs (Control), 764 (Cy), 772 (Cy+LMP), 774 (Cy+HMP), with 664 shared across groups. PCA showed distinct separation of Control and Cy; Cy+HMP partially overlapped Control, indicating compositional correction.
• Taxa shifts: At phylum level, Cy increased Firmicutes and decreased Bacteroidetes, elevating Firmicutes/Bacteroidetes; MP partially reversed trends. At genus level, Cy decreased Lactobacillus and increased Desulfovibrio; MP significantly increased Lactobacillus and decreased Desulfovibrio. Lachnospiraceae_NK4A136_group and Rikenella decreased with Cy and were restored by high-dose MP. Co-occurrence analysis showed negative correlation between Lactobacillus and Desulfovibrio; positive correlation between Lachnospiraceae_NK4A136_group and Lactobacillus.
• SCFAs: Total and individual SCFAs (acetic, propionic, isobutyric, butyric, isovaleric, valeric acids) decreased after Cy and increased in Cy+HMP. Correlations: Lachnospiraceae_NK4A136_group positively associated with multiple SCFAs; Rikenella positively correlated with all six SCFAs; Desulfovibrio negatively associated with branched-chain SCFAs; Lactobacillus positively associated with five SCFAs (except butyric acid).
• Immune–microbiota correlations: Alistipes, Lactobacillus, and Rikenella negatively correlated with DAO and LPS, and positively with SIgA and mRNA levels of IL-2, IFN-γ, IL-4, ZO-1, occludin, claudin-1, mucin-2; Desulfovibrio positively correlated with DAO. Overall, MP exhibited dose-dependent immunoprotection at 300 and 600 mg/kg/day in a Cy model induced by 50 mg/kg/day i.p. on days 18–21, with multiple outcomes achieving p < 0.05.

The study demonstrates that mussel-derived polysaccharide (MP) mitigates Cy-induced intestinal injury and immunosuppression by reinforcing barrier integrity, enhancing mucosal immunity, restoring Th1/Th2 cytokine balance, and activating NF-κB signaling. Upregulation of tight junction proteins (ZO-1, occludin, claudin-1) and mucin-2 is consistent with improved barrier function and reduced permeability (lower DAO and LPS). Enhanced SIgA supports mucosal defense against pathogens. Restoration of IL-2, IFN-γ, IL-4, and IL-10 at both protein and mRNA levels indicates rebalanced Th1/Th2 responses. Western blot data suggest MP activates NF-κB signaling (increased p-IκBα and p-p65), which may contribute to cytokine production and immune modulation. MP also corrected Cy-induced dysbiosis: increasing beneficial Lactobacillus and reducing potentially pathogenic Desulfovibrio, while partially normalizing Firmicutes/Bacteroidetes ratios and improving alpha/beta diversity. Increased SCFAs likely contribute to epithelial energy supply, anti-inflammatory effects, and barrier maintenance. Correlations between key genera and immune/SCFA parameters support a microbiota-mediated mechanism whereby MP acts as a prebiotic to enhance beneficial taxa and their metabolites, reinforcing intestinal and systemic immune functions.

MP exhibits significant immunomodulatory and gut-protective effects in Cy-induced immunosuppressed mice. It improves immune organ indices, restores intestinal barrier markers and SIgA, rebalances Th1/Th2 cytokines, activates NF-κB signaling, corrects dysbiosis (increasing Lactobacillus, decreasing Desulfovibrio), and elevates SCFAs. These findings suggest MP has potential as a prebiotic or functional ingredient to prevent or ameliorate chemotherapy-associated immunosuppression and gut dysbiosis. Future research should: (1) elucidate detailed molecular mechanisms and causal pathways (e.g., receptor-mediated activation, NF-κB dependence via inhibitors/knockouts); (2) conduct dose-response and long-term safety/efficacy studies; (3) perform fecal microbiota transplantation or gnotobiotic experiments to establish microbiota causality; and (4) evaluate translational potential in other disease models and clinical trials.

• Preclinical mouse model (male BALB/c) limits generalizability to humans; sex-specific effects not assessed.
• Preventive design with short Cy exposure (days 18–21) and short intervention window; long-term outcomes not evaluated.
• Sample size modest (n=8/group; n=6 for sequencing) and some outcomes (e.g., spleen index) lacked significant change.
• 16S rRNA sequencing offers genus-level resolution; strain-level functions and absolute abundances were not determined.
• NF-κB activation inferred from phosphorylation; causal dependence not tested with pathway inhibitors or genetic models.
• SCFA measurements in feces do not capture mucosal concentrations or host absorption/kinetics.