Microbiomics and metabolomics insights into the mechanism of egg yolk phosphatidylcholine enhancing intestinal immunity

Abstract

Egg yolk phosphatidylcholine (EYPC), the most vital functional lipid in egg yolk, has demonstrated neuroprotective, antibacterial, and cardioprotective effects. However, there is little information about its immunoregulation effect. Therefore, the aim of this study is to assess the effects and mechanisms of EYPC on intestinal immune homeostasis through an immunosuppressed mouse model induced by cyclophosphamide (CTX). The protective effect of intestinal immunity was evaluated on the basis of immune organ indices, intestinal tight junction (TJ) proteins, sIgA and cytokine secretion, nuclear transcription factor levels, and the equilibrium between Th1 and Th2 cells. It was shown that EYPC obviously inhibited thymus and spleen atrophy, enhanced the expression of TJ proteins, promoted the secretion of sIgA and cytokines, increased the levels of Th1 and Th2 cells, and also modulated the balance of Th1 and Th2 cells. The composition of the gut microbiota and metabolites were discussed to outline mechanisms. The results elucidated that EYPC could alleviate the gut microbiota dysbiosis caused by CTX via reducing the relative abundance of Akkermansia muciniphila and promoting the proliferation of Prevotella and Lactobacillus reuteri. Furthermore, EYPC improved the fecal metabolic profile, restoring the relative content of 13 metabolites and regulating bile secretion. Collectively, these findings suggested that EYPC may contribute to intestinal immune homeostasis through modulating gut microbiota and their metabolism, highlighting its potential as an immunomodulator.

---

1 Introduction

Intestinal immune homeostasis is crucial for body health. However, various factors, such as irregular diet, disruption of day and night rhythms, and chemotherapy, can induce immune dysfunction.^1,2 In addition, intestinal immune homeostasis is greatly dependent on the integrity of intestinal mucosal barriers, whose disruption will cause bacteria and other inflammation factors to come into close contact with epithelial cells, finally resulting in the development of inflammation. Intestinal mucosal barriers comprise the physical barrier, chemical barrier, immune barrier and microbial barrier.³ Tight junction (TJ) proteins between intestinal epithelial cells form the physical barrier; the chemical barrier consists of antimicrobial peptides and other substances, such as secretory immunoglobulin A (sIgA), which prevent bacterial adhesion. The immune barrier contains various immune cells, including neutrophils, macrophages and dendritic cells (DCs), B cells and T cells. Additionally, cytokines, such as interleukin (IL)-10, IL-6, and interferon-gamma (INF-γ), contribute an imperative portion to the immunologic barrier. The interaction between the intestinal microbiome and the intestinal mucosa constitutes a micro-ecosystem known as the gut microbial barrier.⁴

Growing evidence has shown that the intestinal micro-ecosystem is an integral part of the host and its composition heavily influences the balance of intestinal mucosal immunity. The communication between the gut microbiota and the immune system plays an indispensable part in training and educating immune cells, particularly T cells, to appropriately respond to antigens.⁵ Beneficial bacteria interact with the immune system, increasing immune cell activity and anti-inflammatory cytokine production, promoting the growth of immune cells and protecting gut barrier integrity. Moreover, the microbiota also exerts its effect through metabolic activities, such as carbohydrate metabolism to produce short-chain fatty acids and amino acid metabolism to produce tryptophan metabolites.⁶ Fucoidan from Apostichopus japonicus (Aj-FUC) can improve the intestinal barrier and enhance intestinal immunity by up-regulating the abundance of beneficial bacteria, such as Ligilactobacillus, Muribaculaceae, and Lactobacillaceae, while down-regulating the abundance of harmful bacteria. Notably, Aj-FUC significantly increased 5-hydroxyindole-3-acetic acid and indole-3-lactic acid.⁷ Bile acids (BAs), a major metabolites of the gut microbiota, regulate intestinal immunity by modulating the differentiation and proliferation of CD4⁺ T cells. BAs also interact with the Takeda G-protein receptor 5 (TGR5), leading to reduced expression of NOD-like receptor thermal protein domain-associated protein, activation of NLRP3, and inhibition of the NF-κB pathway, which contribute to lowering the levels of pro-inflammatory cytokines.⁸

Recently, egg yolk phosphatidylcholine (EYPC), one of the principal lipids in egg yolk, has received considerable attention due to its various bioactivities, such as anti-inflammatory, antioxidant and cognitive-enhancing effects.^9,10 In our previous report, EYPC was shown to exert anti-colitis effects.¹¹ However, its immunomodulation effects remain unknown. Cyclophosphamide (CTX), which is an effective immunosuppressive agent that is widely used for treating various immune disorders and malignancies, can, when administered long-term or at high doses, damage the immune system and intestinal mucosal barrier, leading to disorder of the gut microbiota.¹² Thus, this study investigated whether EYPC could prevent intestinal immunity dysfunction resulting from CTX treatment, and furthermore, explored the mechanisms based on intestinal microbiota and metabolites.

2 Materials and methods

2.1 Materials and chemical reagents

EYPC (purity ≥ 99%) was obtained from Sigma-Aldrich (St Louis, MI, USA), CTX was bought from Aladdin (Shanghai, China), levamisole hydrochloride (LH) was purchased from Shandong Renhetang (Linyi, China), and enzyme-linked immunosorbent assay (ELISA) kits for IL-2, IL-4, IL-10, tumor necrosis factor (TNF)-α, sIgA, T-bet and GATA-3 were obtained from Jiangsu Meimian (Yancheng, China). All other chemicals in this study were analytical grade and bought from Shanghai Chemicals and Reagents Co. (Shanghai, China).

2.2 Animals and experimental design

Forty female Balb/c mice (weighing 18–20 g, 6–8 weeks olds) were acquired from Hunan Slack Jingda Company (Changsha, China). All mice were housed at a temperature of 24 ± 1 °C, relatively humidity of 45 ± 5% with a 12 h light/dark cycle. Before the experiment, the mice were fed adaptively for a week, during which time the mice freely had access to food and water.

The mice were randomly assigned to five groups (n = 8): (1) the normal control (N) group, which received daily intragastric administered normal saline at 0.1 mL per 20 g body weight (BW); (2) the model (M) group, which received normal saline at 0.1 mL per 20 g BW daily; (3) the positive control (P) group, which received daily intragastric administered LH at 40 mg per kg BW; (4) the low-dose EYPC (LPC) group, which received daily EYPC at 50 mg per kg BW; and (5) the high-dose EYPC (HPC) group, which received daily EYPC at 100 mg per kg BW. The doses of EYPC in this study were determined based on a previous study.¹¹ The animal experiment lasted for 14 days. All mice, except those in the N group, were intraperitoneally injected with CTX at a dosage of 80 mg per kg BW on the 9^th, 10^th, and 11^th days to induce immunosuppression. The chart of the experimental design and period is depicted in Fig. 1A. Additionally, food intake and BW were recorded every day during the experiment. The weight change rate was calculated by comparing the final weight (W) of the mouse with the weight (W₀) of the mouse on the day before CTX treatment. The detailed formula is as follows: .

Fig. 1 Effects of EYPC on body weight and immune organ indices of mice (n = 8). (A) Schematic representation of the experimental design. (B) Changes in body weight of mice during feeding. (C) The change rate of the final body weight of mice compared with the body weight on the day before modeling. (D) Food consumption of mice in different treatment groups during the experiment. (E) Thymus index. (F) Spleen index. (G) Liver index. All data are presented as mean ± SEM, *p < 0.05 and **p < 0.01 versus the N group; ^#p < 0.05 and ^##p < 0.01 versus the M group.

On the 15^th day, all mice were sacrificed. The blood of each mouse was collected from the orbital cavity and the thymus, spleen, small intestine as well as colonic contents were collected for further analysis. All animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Jiangxi Agricultural University and approved by the Animal Ethics Committee of Jiangxi Agricultural University (JXAUA01).

2.3 Organ indices analysis

The thymus and spleen of each mouse were weighed to calculate the organ indices using the following formula: .

2.4 Histopathological evaluation

To assess histopathological changes, a 1 cm segment of the jejunum was fixed in 4% paraformaldehyde. All fixed tissues were then dehydrated, embedded in paraffin, and stained with hematoxylin and eosin (HE). For each sample, four fields were randomly selected, and three villi in each field were examined at 200× magnification using an optical microscope (Eclipse Ci-e, Nikon, Japan). The villus length (V) and crypt depth (C) were measured with Image J software, and the V/C value was subsequently calculated.

2.5 Alcian blue and periodic acid Schiff (AB-PAS) staining

AB-PAS staining was performed to determine mucin secretion and the number of goblet cells. After dewaxing, the paraffin-embedded sections were stained with AB dye and PAS dye. Stained tissues were observed under an optical microscope (Eclipse Ci-e, Nikon, Japan) at 100× magnification, and data collected from three villi in one field per sample. In detail, goblet cells appeared as obvious blue-purple aggregates, the alkaline mucus reacted with the dye to appear purplish-red, and the acidic mucus was stained blue.

2.6 ELISA

The supernatant of the small intestine was obtained after homogenizing in normal saline. The levels of ZO-1, Occludin, sIgA, TNF-α, IL-2, IL-4, IL-10, T-bet and GATA-3 were detected using ELISA kits according to the instructions of the manufacturer.

2.7 16s rRNA sequencing

Fecal microbial DNA was extracted according to the manufacturer's instructions and amplified using 16S V3–V4 hypervariable region primers. After amplification, the purity and specificity of the products were determined through agarose gel electrophoresis to ensure the reliability for subsequent experiments. Using the Illumina Novaseq platform (Illumina Inc., CA, USA), libraries were constructed by the paired-end method for sequencing. Raw reads were processed sequentially with Trimmomatic v0.33, Cutadapt 1.9.1, DADA2, and the merging functions in QIIME 2 (version 2020.6). Finally, non-chimeric reads were obtained for species annotation and abundance analysis. The BMKCloud platform (https://www.biocloud.net) was then used to analyze Alpha diversity, Beta diversity, and distinct bacteria at different levels.

2.8 Non-target metabolomic analysis

Total fecal metabolites were isolated based on the method reported by Turroni et al.¹³ In detail, a fecal sample was placed into a centrifuge tube, and to this 600 μL of methanol containing 4 ppm 2-chloro-L-phenylalanine was added. Then, the mixture was vortexed for 30 s, sonicated for 10 min, and centrifuged at 4 °C for 10 min at 12 000 rpm. The supernatant was collected and filtered with a 0.22 μm filter membrane, and finally analyzed using LC-MS. The MSConvert tool of ProteoWizard software was employed to convert the original mass spectrometry files into mzXML files. After peak detection, peak filtering and peak alignment by the R package of XCMS software, a list by quantity of all substances was obtained. Systematic errors were eliminated based on QC samples, and metabolites with coefficient of variation less than 30% in the QC samples were retained for analysis. The metabolites were identified and matched using public databases, including HMDB and KEGG as well as the databases built by various “omics” companies. By matching retention times, molecular masses (with a molecular mass error of less than 10 ppm), secondary fragmentation spectra, and collision energy values of metabolites contained in these databases, the structures of metabolites in the samples were identified, and the identification results strictly double-checked and confirmed manually. The appraisal level is at least Level 2.¹⁴ Additionally, all differential metabolites were filtered at FDR < 0.01. Sample fitting was evaluated using the R package of ropls software, and Metabo Analyst software was employed to predict the pathways of differentiated metabolites.

2.9 Statistical analysis

All data in this experiment are represented by means ± SEM. One-way analysis of variance was used for statistical analysis of the data, and SPSS 26 software (IBM, Chicago, USA) was employed for the Duncan test, when p < 0.05 was statistically significant. Graphs were plotted utilizing GraphPad Prism 8 software (GraphPad, San Diego, USA).

3 Results and discussion

3.1 EYPC alleviates body weight decrease and changes of organ indices induced by CTX treatment

The adverse effects of CTX include significant weight loss and loss of appetite. The body weight and food consumption for each mouse were recorded daily during the experiment. Fig. 1B–D clearly show that the body weight in the N group increased stably, without significant fluctuations, and the food intake was relatively stable during the experimental period, indicating that the physiological state of the mice was not affected by environmental factors during the feeding period. The body weight in the M group from day 1 to day 9 was similar to that in the N group. However, the mice, treated with 40 mg per kg BW LH, exhibited lower body weight and slightly reduced food intake than the other mice, which may be related to some side effects of LH.¹⁵ Similar results were also reported by Sun et al.¹⁶ After CTX treatment, the body weight of mice in the M group was significantly lower than that of mice in the N group, and food intake was also obviously reduced, implying that the immunosuppressed mouse model was successfully established. The body weight and food intake of mice supplemented with different doses of EYPC also decreased, but recovered to the level close to that in the N group on the 14^th day, which was significantly higher than that in the M group, suggesting that EYPC could inhibit weight loss caused by CTX treatment.

Organ indices are closely associated with immune regulatory activity, and the thymus, spleen, and liver are indispensable immune organs in the body. Han et al.¹⁷ reported that CTX treatment significantly decreases spleen and thymus indices in Balb/c mice. In this study, as shown in Fig. 1E–G, the thymus index (1.28) and spleen index (1.55) in the M group were markedly decreased, while the liver index (55.35) was dramatically increased when compared with those indices in the N group (3.62, 3.41 and 48.2, respectively), confirming that the immunosuppressed mouse model was successfully established. In comparison with the M group, the groups featuring LH and EYPC supplementation effectively suppressed those changes. Specifically, after EYPC gavage, the thymus, spleen and liver indices were 2.07, 1.92 and 47.49 in the LPC group, and 2.00, 1.89 and 48.3 in the HPC group. These results indicate that both of 50 and 100 mg per kg BW EYPC protected immune function by restoring immune organ indices.

3.2 EYPC protects small intestinal morphology from CTX-induced damage

HE staining was employed to elucidate the protective effect of EYPC on small intestinal histology. Moreover, the V/C value, a key indicator of intestinal structure, was analyzed and it was found that CTX-induced intestinal damage is often accompanied by a decrease in V/C value.¹⁸ As shown in Fig. 2, the tissue in the N group was normal and the villi were intact and appeared normal, while in the M group, the intestinal morphology was severely damaged, characterized by breakage of villi and increased crypt depth. Compared with the N group, the V/C value in the M group was significantly decreased (p < 0.01). In contrast to the M group, the intestinal structure was well repaired in the P, LPC and HPC groups, where the villi were longer and compact, and the value of V/C was dramatically improved. These results revealed that EYPC can prevent injury to the small intestine induced by CTX, and are consistent with results reported by Fang et al.¹⁹

Fig. 2 Effects of EYPC on the histology of the small intestine in mice. (A) Representative images of HE staining of the small intestine (magnification: 40× upper, and 100× lower). The yellow lines indicate villus length (V), and the red lines indicate crypt depth (C). (B) V/C values of the small intestine (n = 6). All data are presented as mean ± SEM.

3.3 EYPC improved the intestinal physical barrier of immunosuppressed mice

Goblet cells are a type of epithelial cell primarily responsible for synthesizing and secreting mucus as well as TJ proteins, and they play crucial roles in maintaining the integrity of the intestinal barrier.²⁰ Decreases of goblet cells and TJ proteins can lead to severe barrier dysfunction and this subsequently causes some forms of disease, such as inflammatory bowel disease.²¹ In the present study, AB-PAS staining was used to determine the number of goblet cells and associated mucus levels (Fig. 3A and B). When compared with the N group, the distribution of goblet cells was chaotic and their number was significantly reduced in the M group; in addition, the level of mucus was seriously decreased. Supplementation with EYPC significantly increased the number of goblet cells and mucus level. The levels of ZO-1 and Occludin were also detected via ELISA. Fig. 3C and D reveal that intraperitoneal injection of CTX dramatically reduced the levels of ZO-1 and Occludin in the small intestine. However, treatment with EYPC significantly restored the levels of these two TJ proteins. These results suggest that EYPC can enhance intestinal barrier integrity by restoring the levels of goblet cells and TJ proteins in immunosuppressed mice.

Fig. 3 Effects of EYPC on intestinal barrier function in mice (n = 6). (A) Representative images of AB-PAS staining of the small intestine (magnification: 40× upper, and 100× lower). (B) The number of goblet cells in the small intestinal villi. The levels of Occludin (C) and ZO-1 (D) in the small intestine tissues of mice. All data are presented as mean ± SEM.

3.4 EYPC alleviated intestine immune dysfunction in immunosuppressed mice

CTX has been shown to destroy the cellular immunity process, resulting in immune function suppression.²² SIgA is the most abundant antibody in intestinal mucosa and a key participant in intestinal immunity by directly combining with pathogenic microorganisms.²³Fig. 4A reveals that CTX treatment dramatically reduces sIgA secretion when compared with that in the N group. In contrast, high-dose EYPC significantly increases sIgA levels in the small intestine of immunosuppressed mice, while low-dose EYPC can restore sIgA levels in the small intestine of mice to a certain extent, although the effect is not statistically significant.

Fig. 4 Regulatory effects of EYPC on intestine immunity of immunosuppressed mice (n = 6). (A) The levels of sIgA in the small intestine; the levels of T-bet (B) and GATA-3 (C) in the small intestine; and (D) the values of T-bet/GATA-3. The levels of TNF-α (E), IL-2 (F), IL-4 (G) and IL-10 (H). All data are presented as mean ± SEM, *p < 0.05 and **p < 0.01 versus the N group; ^#p < 0.05 and ^##p < 0.01 versus the M group.

Moreover, helper T (Th) cells containing various subtypes, including Th1, Th2 and Th17 cells, are major contributors to cellular immunity and are abundant in the lamina propria of the gut.²⁴ A dynamic equilibrium of Th1 and Th2 is of importance to maintain immune homeostasis, and this balance is regulated by specific transcription factors. T-bet and GATA-3 are nuclear transcription factors of Th1 and Th2 cells, respectively, and their levels reflect the levels of Th1 and Th2 cells.²⁵ Their imbalance has been observed previously in CTX-induced immunosuppressed mice.²⁶ In this study, after CTX treatment, the levels of T-bet and GATA-3 in the small intestine of the M group were significantly decreased in comparison with their levels in normal mice. Compared with mice in the M group, EYPC noticeably improved the levels of T-bet and GATA-3 in the small intestine of immunosuppressed mice (Fig. 4B and C). Moreover, as shown in Fig. 4D, treatment with CTX significantly reduced the ratio of T-bet and GATA-3, while following administration with EYPC, the value of the T-bet/GATA-3 ratio was restored to a level close to that of the normal group, indicating that EYPC can reverse the damaged Th1/Th2 immune balance in the small intestine of immunosuppressed mice.

Furthermore, Th1 cells largely secrete pro-inflammatory cytokines, such as IFN-γ, TNF-α, and IL-2, while Th2 cells mainly release anti-inflammatory cytokines like IL-4, IL-5, and IL-10.²⁷ In order to explore further the effects of EYPC on the balance of Th1 and Th2 cells, the levels of TNF-α, IL-2, IL-4 and IL-10 in the small intestine were detected by using ELISA kits. As shown in Fig. 4E–H, the levels of TNF-α, IL-2, IL-4 and IL-10 in the M group were significantly decreased. After EYPC treatment, the levels of TNF-α, IL-2, IL-4 and IL-10 were dramatically higher than those of the M group, which further supported the conclusion that EYPC could maintain the balance between Th1 and Th2 cells.

3.5 EYPC modulated the gut microbiota in immunosuppressed mice

A dynamic equilibrium exists between the composition of the gut microbiota and the health of the host, with the stability of this balance significantly influencing physiological processes such as immune modulation, metabolism, and the synthesis of vital compounds.²⁸ To reveal the reconstruction effects of EYPC on the gut microbiota of immunosuppressed mice, 16S rRNA sequencing was leveraged. Rarefaction curves (Fig. S1) showed that the abundance of bacterial species leveled off as the sampling depth was increased, indicating that the sequencing effort was sufficient and that total diversity within the sample was captured. The richness of the gut microbiota was characterized by the Chao1 and Observed species indices, while its diversity was evaluated by the Shannon and Simpson indices. Fig. 5A shows that there is no statistical significance between these indices in the N, M and EYPC groups, indicating that EYPC did not influence the alpha diversity of the gut microbiota. Beta diversity is invariably employed to assess differences in microbiota composition among groups. In principal component analysis (PCA) and principal coordinates analysis (PCoA), greater distances between samples indicate larger differences in the composition of the gut microbiota.²⁹ As shown in Fig. 5B, the M group exhibited marked divergence in the community of gut microbiota compared to the N group, confirming that the intestinal microbiota was substantially changed after treatment with CTX. In contrast, samples from the EYPC group displayed coordinates closer to those from the N group and farther away from the M group, suggesting that EYPC can modulate the disorder of intestinal microbiota in immunosuppressed mice.

Fig. 5 Effects of EYPC on the intestinal microbiota of immunosuppressed mice (n = 6). (A) α-Diversity analysis; (B) β-diversity analysis; the abundance of intestinal microbiota at the phylum (C), genus (E) and species (H) levels. The relative abundance of Verrucomicrobia (D), Akkermansia (F), Prevotella (G), Akkermansia muciniphila (I), Lactobacillus reuteri (J) and Ruminococcus gnavus (K). All data are presented as mean ± SEM, *p < 0.05 and **p < 0.01 versus the N group; ^#p < 0.05 and ^##p < 0.01 versus the M group.

To further clarify modification in the intestinal microbiota, the major taxa were analyzed. Fig. 5C shows the relative abundance of the top 10 identified microbes at the phylum level. Firmicutes and Bacteroidota are the predominant bacteria across all groups, accounting for more than 90%. In particular, after CTX treatment, the relative abundance of Verrucomicrobia was significantly increased (p < 0.05), which is in accordance with the report of Yu et al.³⁰ However, EYPC administration suppressed this trend (Fig. 5D). At the genus level, the top 20 bacteria are shown in Fig. 5E–G, and when compared to the N group, Akkermansia is significantly enriched in the M group (p < 0.05), which is consistent with the results reported by Zhou et al.¹² The relative abundance of Akkermansia was reduced in mice supplemented with EYPC, but there was no significant difference when compared to the M group. The high-dosage EYPC group had the greatest abundance of Prevotella compared to all other groups (p < 0.05). Prevotella was reported to be able to up-regulate the immune response of Th17 cells in vitro, and some of its strains could stimulate DCs in vivo to induce immune activation.^31,32 In addition, a total of 9 bacteria were identified at the species level (Fig. 5H–K). Among them, Akkermansia muciniphila (A. muciniphila) and Ruminococcus gnavus (R. gnavus) were significantly enriched in the M group when compared with their presence in the N group (p < 0.05), while EYPC significantly reduced the enrichment of A. muciniphila (p < 0.05). It was also discovered that EYPC could significantly increase the relative abundance of Lactobacillus reuteri (L. reuteri) in immunosuppressed mice when compared to the M group (p < 0.01). A. muciniphila proliferates in the gut by consuming mucin and it has previously been reported to exert beneficial activities through anti-inflammatory and anti-obesity effects, but not all strains are capable of probiotic effects.^33,34 For example, one study examined the anti-inflammatory effects of four different A. muciniphila and found that only one of them had significant anti-inflammatory effects, suggesting that the probiotic effects of A. muciniphila are strain specific.³⁵ Furthermore, the abundance of A. muciniphila has been found to be significantly elevated in colorectal cancer patients, indicating a potential hazard to intestinal health.³⁶ Another study showed that some murine A. muciniphila strains may aggravate colitis in IL-10^−/− mice, although subsequent studies have shown that A. muciniphila can positively affect intestinal integrity and immunity, and is influenced by many factors, such as bacterial strain, and the physical condition of the host. Similar to A. muciniphila, R. gnavus also targets mucins in the intestine as a nutrient source for its proliferation.³⁷R. gnavus has been reported to have more adverse effects on intestinal health and is significantly enriched in the intestinal mucosa of patients with inflammatory bowel disease and irritable bowel syndrome.^38–40L. reuteri, a kind of probiotic in the gut, has been demonstrated to possess anti-inflammatory, immune regulation, and intestinal protection effects.⁴¹ According to the above results, it was deduced that EYPC regulates intestinal immunity by inhibiting A. muciniphila and enriching L. reuteri. Additionally, phosphatidylcholine can be digested and absorbed in the upper gastrointestinal tract so whether the immune-regulatory effects of EYPC rely on gut microbiota needs to be clarified using antibiotics treatment and fecal microbiota transplantation in the future.

3.6 Potential biomarkers of the gut microbiota

LEfSe analysis was used to identify representative microorganisms in each murine group with an LDA value >3.0. Based on results at the genus level (Fig. 6A and B), biomarkers in the N group were Megamonas and Arthrobacter, whereas in the M group, the biomarkers were Megasphaera and Leuconostoc mesenteroides, revealing that these two microorganisms may be related to intestinal immunosuppression. In the EYPC group, Thermoactinomyces, Prevotella, Bacteroides, L. reuteri and Bacteroides ovatus were identified as biomarkers. Moreover, Fig. 6C presents a separate comparison of the biomarkers in the M and EYPC groups. It was observed that Anaerotruncus was enriched in the M group, while L. reuteri and Prevotella were enriched in the EYPC group, which supported the results shown in Fig. 5.

Fig. 6 LEfSe analysis of the gut microbiota. (A) Evolutionary tree diagram; (B) biomarkers from three different treatment groups; (C) biomarkers in the M and HPC groups; (D) Spearman correlation analysis of the gut microbiota and intestinal immunity (n = 6). All data are presented as mean ± SEM, *p < 0.05 and **p < 0.01 versus the M group.

3.7 Correlation analysis between gut microbiota and immunity markers

In order to explore the relationship between intestinal microbiota and intestinal immunity as influenced by EYPC, Spearman correlation analysis was performed to examine the associations between specific bacteria and intestinal cytokines, sIgA, and TJ proteins (Fig. 6D). The results showed that Akkermansia and Ruminococcus were significantly negatively correlated with these indicators (p < 0.05), while Adlercreutzia was significantly negatively correlated with T-bet and the levels of two TJ proteins (p < 0.05). These results indicate that the three strains are closely correlated with the degree of intestinal immune injury in mice. Roseburia was significantly positively correlated with intestinal immune indices, suggesting that Roseburia exerts a beneficial effect on the improvement of intestinal immune homeostasis in response to EYPC treatment.

3.8 The effect of EYPC on fecal metabolites of immunosuppressed mice

Untargeted metabolomic analysis was performed to analyze fecal metabolites. PCA clustering analysis (Fig. 7A) under different ion modes was used to preliminarily evaluate differences among murine groups. QC samples overlapped well, indicating that data are strongly reliable (Fig. S2). Samples from the N, M and EYPC groups were clearly separated, indicating distinct metabolite composition and abundance. Simultaneously, OPLS-DA (Fig. 7B) revealed differences between samples from the N and M groups, and between the M and high-dosage EYPC groups, under positive and negative ion modes, supporting the suitability for subsequent analysis. Differential metabolite quantity statistics in positive and negative ion modes were recorded and FC ≥ 2.5 or ≤0.4, VIP > 1.2, and p ≤ 0.05 were used to screen metabolite differences between different groups. In Fig. 7C, in the positive ion mode, there were 24 up-regulated differential metabolites and 68 down-regulated differential metabolites in the M group compared with the N group, and 8 up-regulated differential metabolites and 19 down-regulated differential metabolites in the HPC group compared with the M group. In the negative ion mode, 9 different metabolites were up-regulated and 25 different metabolites were down-regulated in the M group compared with the N group, while 4 differential metabolites were up-regulated and 11 differential metabolites were down-regulated in the HPC group compared with the M group.

Fig. 7 Effects of EYPC on fecal metabolites of immunosuppressed mice (n = 6). (A) PCA clustering analysis of the overall sample. Pos: ESI⁺, Neg: ESI⁻. (B) Metabolite OPLS-DA clustering analysis. Pos: ESI⁺, Neg: ESI⁻. (C) Numbers of differential metabolites. (D) The main differential metabolites in positive ion mode. (E) The main differential metabolites in negative ion mode. All data are presented as mean ± SEM, *p < 0.05 and **p < 0.01 versus the N group; ^#p < 0.05 and ^##p < 0.01 versus the M group.

In the positive ion mode, 106 different metabolites were screened (Fig. 7D). Among them, the relative contents of four differential metabolites (β-carotene, S-adenosine methionine, prostaglandin I2, myristic acid) were down-regulated after CTX treatment, but up-regulated after EYPC supplementation. Two differential metabolites (sphingosine and L-homophenylalanine) were up-regulated after CTX treatment and down-regulated after EYPC supplementation. Under the negative ion mode (Fig. 7E), a total of 43 differential metabolites were identified. Among these, five differential metabolites, including docosapentaenoic acid, 5,7-dihydroxyflavone, thymine, naringin and guanidinosuccinic acid, were down-regulated after CTX treatment, but their levels were restored by treatment with EYPC. L-Aspartic acid and folic acid were up-regulated after CTX treatment and down-regulated after EYPC supplementation. These results indicate that supplementation with EYPC can significantly modulate the fecal metabolic profile of immunosuppressed mice.

Enrichment of pathways for different metabolites between the N and M groups and between the M and HPC groups were established through KEGG database comparison. The results (Table 1) revealed that the top 10 significantly enriched pathways between the N and M groups include adenosine triphosphate binding box (ABC) transporters, linoleic acid metabolism, amino acid biosynthesis, cancer central carbon metabolism, cysteine and methionine metabolism, protein digestion and absorption, bile secretion, aminoacyl-tRNA biosynthesis, PPAR signaling pathway, and cofactor biosynthesis. Between the M and the HPC groups, the top 10 enriched pathways included cholesterol metabolism, vitamin digestion and absorption, bile secretion, vascular smooth muscle contraction, cyclic adenosine phosphate signaling pathway, sphingolipid metabolism, cell apoptosis, vascular endothelial growth factor signaling pathway, primary bile acid biosynthesis, and aldosterone-regulated sodium reabsorption (Table 2). Moreover, it was discovered that the common differential metabolic pathway between the two comparison groups is “bile secretion”. BAs, the main components of bile, are metabolites produced from the breakdown of cholesterol and are reused in the body through the intestine–liver cycle. Under healthy conditions, 95% of bile acids in the gut are reabsorbed by the intestine and processed by specific enzymes in the liver before being stored in the gallbladder.⁴² However, when intestinal epithelial function has been damaged and the reabsorption capacity of the intestine is impaired, bile acid malabsorption (BAM) will occur.⁴³ In addition, the accumulation of bile acids in the intestinal lumen can further cause epithelial damage.⁴⁴ BAM may be responsible for the increase of bile acid-related metabolites in feces of immunosuppressed mice in this study. BAs are also associated with intestinal immune homeostasis. Activation of bile acid receptors, TGR5 and farnesoid X receptor (FXR), reduces cytokines associated with macrophages and DCs, thereby disrupting intestinal immune homeostasis.^45,46 Thus, we further analyzed the metabolites in the pathway. It was discovered that three metabolites (L-carnitine, bilirubin and taurocholic acid) in this pathway were significantly up-regulated in the M group compared with the N group, while allocholic acid was significantly down-regulated (Fig. 8). Four metabolites (folic acid, glycocholic acid, glycinodeoxycholic acid and lamivudine) were down-regulated in the HPC group when compared with the M group. These results suggested that EYPC may reduce enrichment of BAs in the intestine by restoring intestinal reabsorption of BAs in immunosuppressed mice, thereby maintaining intestinal barrier integrity and intestinal immune homeostasis. Furthermore, bile acids could become signaling molecules acting on immune cells by activating a variety of receptors involving TGR-5, FXR and ROR-γt, thus enhancing intestinal immunity disrupted by CTX, a process that will be explored in future research.

Fig. 8 The difference in bile acid levels between the N and M groups as well as that between the M and the HPC groups (n = 6). All data are presented as mean ± SEM, *p < 0.05 and **p < 0.01 versus the N group; ^#p < 0.05 and ^##p < 0.01 versus the M group.

Table 1 The top ten differential metabolite KEGG enrichment pathways between the N and M groups

Pathway_ID	Pathway_name	Up	Down	P value
Up/down: enrichment of the number of up-regulated/down-regulated differential metabolites in this pathway.
mmu02010	ABC transporters	6	6	4.63 × 10^−6
mmu00591	Linoleic acid metabolism	0	6	7.56 × 10^−6
mmu01230	Biosynthesis of amino acids	5	4	0.000409
mmu05230	Central carbon metabolism in cancer	4	1	0.000455
mmu00270	Cysteine and methionine metabolism	5	1	0.001149
mmu04974	Protein digestion and absorption	4	1	0.001392
mmu04976	Bile secretion	3	1	0.001604
mmu00970	Aminoacyl-tRNA biosynthesis	4	1	0.002199
mmu03320	PPAR signaling pathway	0	2	0.003072
mmu01240	Biosynthesis of cofactors	4	8	0.012947

---

Table 2 The top ten differential metabolite KEGG enrichment pathways between the M and HPC groups

Pathway_ID	Pathway_name	Up	Down	P value
Up/down: enrichment of the number of up-regulated/down-regulated differential metabolites in this pathway.
mmu04979	Cholesterol metabolism	0	2	0.001971
mmu04977	Vitamin digestion and absorption	2	1	0.002239
mmu04976	Bile secretion	0	4	0.003965
mmu04270	Vascular smooth muscle contraction	1	1	0.005125
mmu04024	cAMP signaling pathway	1	1	0.012333
mmu00600	Sphingolipid metabolism	0	2	0.02345
mmu04210	Apoptosis	0	1	0.027133
mmu04370	VEGF signaling pathway	1	0	0.040431
mmu00120	Primary bile acid biosynthesis	0	2	0.040508
mmu04960	Aldosterone-regulated sodium reabsorption	0	1	0.053554

---

Previous studies elucidated that EYPC can inhibit the colitis caused by dextran sulfate sodium through decreasing the levels of TNF-α, IL-1β, IL-6 and MPO, restoring the number of goblet cells and levels of TJ proteins, modulating the composition of the gut microbiota, lowering the relative abundance of Parabacteroides, and increasing the content of Alistipes and Lachnospiraceae_NK4A136_group. In immunosuppressed mice, EYPC similarly regulates the secretion of cytokines and enhances intestinal integrity. Moreover, the results of the gut microbiota analysis in this study revealed a new mechanism that differs from the anti-colitis mechanism of EYPC, where EYPC exerts an immune-enhancing effect. Specifically, EYPC regulates the equilibrium between Th1 and Th2 cells, thereby reducing the relative abundance of A. muciniphila, and promoting the enrichment of Prevotella and L. reuteri, thus restoring intestinal reabsorption of BAs in immunosuppressed mice. The impact of EYPC on the composition of the gut microbiota was significantly influenced by the mouse model employed.

4 Conclusions

EYPC showed immunomodulatory effects on CTX-induced immunosuppressed mice by maintaining body weight and organ indices, alleviating pathological injury to the small intestine, enhancing the intestinal barrier, stimulating the secretion of cytokines and expression of nuclear transcription factors, and regulating Th1/Th2 homeostasis. Furthermore, EYPC significantly reduced the relative abundance of A. muciniphila, and promoted the enrichment of Prevotella and L. reuteri. EYPC also attenuated metabolic disturbances through restoring intestinal reabsorption of BAs in immunosuppressed mice. These findings reveal novel insights into the immunoregulatory effects of EYPC, which supports the development of EYPC-based functional foods.

Conflicts of interest

The authors declare no conflicts of interest.

Data availability

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5fo02959g.

Acknowledgements

The authors gratefully acknowledge the support provided by the Key Research and Development Plan of Jiangxi Province, China (20232BBF60025), the National Natural Science Foundation of China (82260642), and the Key Program of the Natural Science Foundation of Jiangxi Province, China (20224ACB205015).

References

1. R. Voigt, C. Forsyth and A. Keshavarzian, Circadian rhythms: a regulator of gastrointestinal health and dysfunction, Expert Rev. Gastroenterol., 2019, 13(5), 411–424.
2. L. Zhang, W. Cao, Y. Gao, R. Yang, X. Zhang and J. Xu, et al., Astaxanthin (ATX) enhances the intestinal mucosal functions in immunodeficient mice, Food Funct., 2020, 11, 3371–3381.
3. Y. Ma, H. Xie, N. Xu, M. Li, L. Wang and H. Ge, et al., Large yellow tea polysaccharide alleviates HFD-induced intestinal homeostasis dysbiosis via modulating gut barrier integrity, immune responses, and the gut microbiome, J. Agric. Food Chem., 2024, 72(13), 7230–7243.
4. Y. Yang, Y. Zhang, W. Zhang, K. Lu, L. Wang and Y. Liu, et al., Flammulina velutipes residue polysaccharide alleviates immunosuppression and intestinal injury by modulating gut microbiota and associated metabolites, J. Agric. Food Chem., 2025, 72(13), 7788–7806.
5. P. Zhou, C. Chen, S. Patil and S. Dong, Unveiling the therapeutic symphony of probiotics, prebiotics, and postbiotics in gut-immune harmony, Front. Nutr., 2024, 11, 1355542.
6. A. Lavelle and H. Sokol, Gut microbiota-derived metabolites as key actors in inflammatory bowel disease, Nat. Rev. Gastroenterol. Hepatol., 2020, 17, 223–237.
7. W. Qin, B. Wei, P. Ren, Y. Chang, C. Xue and Q. Tang, Fucoidan from Apostichopus japonicus enhances intestinal barrier function and promotes intestinal immunity via regulating the gut microbiota and tryptophan metabolism, Int. J. Biol. Macromol., 2025, 301, 139929.
8. J. Cong, P. Liu, Z. Han, W. Ying, C. Li and Y. Yang, et al., Bile acids modified by the intestinal microbiota promote colorectal cancer growth by suppressing CD8+ T cell effector functions, Immunity, 2024, 57(4), 876–889.
9. J. Chen, S. Lin, N. Sun, S. Bao, J. Shen and X. Lu, Egg yolk phosphatidylcholine: Extraction, purification and its potential neuroprotective effect on PC12 cells, J. Funct. Foods, 2019, 56, 372–383.
10. W. Tan, Q. Zhang, Z. Dong, Y. Yan, Y. Fu and X. Liu, et al., Phosphatidylcholine ameliorates LPS-induced systemic inflammation and cognitive impairments via mediating the gut-brain axis balance, J. Agric. Food Chem., 2020, 68(50), 148840–114895.
11. Y. Wen, L. Tan, S. Chen, N. Wu, Y. Yao and L. Xu, et al., Egg yolk phosphatidylcholine alleviates DSS-induced colitis in BALB/c mice, Food Funct., 2023, 14, 9309–9323.
12. Q. Zhou, J. Gao, X. Sun, J. Du, Z. Wu and D. Liang, et al., Immunomodulatory mechanisms of tea leaf polysaccharide in mice with cyclophosphamide-induced immunosuppression based on gut flora and metabolomics, Foods, 2024, 13(18), 2994.
13. S. Turroni, J. Fiori, S. Rampelli, S. Schnorr, C. Consolandi and M. Barone, et al., Fecal metabolome of the Hadza hunter-gatherers: a host-microbiome integrative view, Sci. Rep., 2016, 6(1), 32826.
14. I. Blaženović, T. Kind, J. Ji and O. Fiehn, Software tools and approaches for compound identification of LC-MS/MS data in metabolomics, Metabolites, 2018, 8(2), 31.
15. A. Dartevel, B. Chaigne, L. Moachon, F. Grenier, N. Dupin and L. Guillevin, et al., Levamisole-induced vasculopathy: A systematic review, Semin. Arthritis Rheum., 2019, 48(5), 921–926.
16. Y. Sun, F. Wang, Y. Liu, S. Liu, Y. An and H. Xue, et al., Microbiome-metabolome responses of Fuzhuan brick tea crude polysaccharides with immune-protective benefit in cyclophosphamide-induced immunosuppressive mice, Food Res. Int., 2022, 157, 111370.
17. L. Han, M. Meng, M. Guo, D. Cheng, L. Shi and X. Wang, et al., Immunomodulatory activity of a water-soluble polysaccharide obtained from highland barley on immunosuppressive mice models, Food Funct., 2019, 10(1), 304–314.
18. J. Wang, M. Li, Y. Gao, H. Li, L. Fang and C. Liu, et al., Effects of exopolysaccharides from Lactiplantibacillus plantarum, JLAU103 on intestinal immune response, oxidative stress, and microbial communities in cyclophosphamide-induced immunosuppressed mice, J. Agric. Food Chem., 2022, 70(7), 2197–2210.
19. Q. Fang, S. Chen, J. Wang, X. Huang, Q. Nie and G. Phillips, et al., Fractions from natural Cordyceps sinensis, alleviated intestinal injury in cyclophosphamide-induced mice, Bioact. Carbohydr. Diet. Fibre, 2021, 26, 100271.
20. B. Zheng, M. Ying, J. Xie, Y. Chen, Y. Wang and X. Ding, et al., A Ganoderma atrum polysaccharide alleviated DSS-induced ulcerative colitis by protecting the apoptosis/autophagy-regulated physical barrier and the DC-related immune barrier, Food Funct., 2020, 11(12), 10690–10699.
21. W. Kuo, L. Zuo, M. Odenwald, S. Madha, G. Singh and C. Gurniak, et al., The tight junction protein ZO-1 is dispensable for barrier function but critical for effective mucosal repair, Gastroenterology, 2021, 161(6), 1924–1939.
22. G. Cai, Y. Wu, A. Wusiman, P. Gu, N. Mao and S. Xu, et al., Alhagi honey polysaccharides attenuate intestinal injury and immune suppression in cyclophosphamide-induced mice, Food Funct., 2021, 12(15), 6863–6877.
23. G. Cai, Y. Yang, P. Gu, K. Li, W. Adelijiang and T. Zhu, et al., The secretion of sIgA and dendritic cells activation in the intestinal of cyclophosphamide-induced immunosuppressed mice are regulated by Alhagi honey polysaccharides, Phytomedicine, 2022, 103, 154232.
24. A. Hertweck, M. de mucha, P. Barber, R. Dagil, H. Porter and A. Ramos, et al., The TH1 cell lineage-determining transcription factor T-bet suppresses TH2 gene expression by redistributing GATA3 away from TH2 genes, Nucleic Acids Res., 2022, 50(8), 4557–4573.
25. I. Raphael, S. Nalawade, T. Eagar and T. Forsthuber, T cell subsets and their signature cytokines in autoimmune and inflammatory diseases, Cytokine, 2015, 74(1), 5–17.
26. M. Ying, Q. Yu, B. Zheng, H. Wang, J. Wang and S. Chen, et al., Cultured Cordyceps sinensis polysaccharides modulate intestinal mucosal immunity and gut microbiota in cyclophosphamide-treated mice, Carbohydr. Polym., 2020, 235(1), 115957.
27. F. Min, J. Hu, T. Huang, Y. Huang, S. Nie and T. Xiong, et al., Effects of Lactobacillus casei NCU011054 on immune response and gut microbiota of cyclophosphamide induced immunosuppression mice, Food Chem. Toxicol., 2023, 174, 113662.
28. X. Li, Y. Zhao, M. Liu, M. Zhang, Z. Zhu and A. Liao, et al., Fermented wheat germ ameliorates high-fat diet-induced maternal obesity in rats: Insights from microbiome and metabolomics, J. Agric. Food Chem., 2025, 73(21), 12666–12678.
29. H. Li, Y. Zhang, Y. Zheng, X. Li, Z. Li and C. Man, et al., Structural characterization of the exopolysaccharide produced by Bacillus amyloliquefaciens JM033 and evaluation of its ability to regulate immunity and intestinal flora, Int. J. Biol. Macromol., 2025, 306, 141052.
30. Y. Yu, H. Zhu, M. Shen, Q. Yu, Y. Chen and J. Xie, Sulfation modification enhances the intestinal regulation of Cyclocarya paliurus polysaccharides in cyclophosphamide-treated mice via restoring intestinal mucosal barrier function and modulating gut microbiota, Food Funct., 2021, 12(24), 12278–12290.
31. J. Larsen, The immune response to Prevotella bacteria in chronic inflammatory disease, Immunology, 2017, 151(4), 363–374.
32. N. Van Teijlingen, L. Helgers, E. Zijlstra-Willems, J. Van Hamme, C. Ribeiro and K. Strijbis, et al., Vaginal dysbiosis associated-bacteria Megasphaera elsdenii and Prevotella timonensis induce immune activation via dendritic cells, J. Reprod. Immunol., 2020, 138, 103085.
33. Q. Zhou, Y. Zhang, X. Wang, R. Yang, X. Zhu and Y. Zhang, et al., Gut bacteria Akkermansia is associated with reduced risk of obesity: evidence from the American Gut Project, Nutr. Metab., 2020, 17(1), 90.
34. L. Xue, Y. Zhao, H. Wang, Z. Li, T. Wu and R. Liu, et al., The effects of live and pasteurized Akkermansia muciniphila on DSS-induced ulcerative colitis, gut microbiota, and metabolomics in mice, Food Funct., 2023, 14(10), 4632–4646.
35. Q. Liu, W. Lu, F. Tian, J. Zhao, H. Zhang and K. Hong, et al., Akkermansia muciniphila exerts strain-specific effects on DSS-induced ulcerative colitis in mice, Front. Cell. Infect. Microbiol., 2021, 11, 698914.
36. T. Weir, D. Manter, A. Sheflin, B. Barnett, A. Heuberger and E. Ryan, Stool microbiome and metabolome differences between colorectal cancer patients and healthy adults, PLoS One, 2013, 8(8), e70803.
37. E. Crost, L. Tailford, M. Monestier, D. Swarbreck, B. Henrissat and L. Crossman, et al., The mucin-degradation strategy of Ruminococcus gnavus: The importance of intramolecular trans-sialidases, Gut Microbes, 2016, 7(4), 302–312.
38. E. Crost, E. Coletto, A. Bell and N. Juge, Ruminococcus gnavus: friend or foe for human health, FEMS Microbiol. Rev., 2023, 47(2), fuad014.
39. C. Png, S. Lindén, K. Gilshenan, E. Zoetendal, C. Mcsweeney and L. Sly, et al., Mucolytic bacteria with increased prevalence in IBD mucosa augment in vitro utilization of mucin by other bacteria, Am. J. Gastroenterol., 2010, 105(11), 2420–2428.
40. L. Zhai, C. Huang, Z. Ning, Y. Zhang, M. Zhuang and W. Yang, et al., Ruminococcus gnavus plays a pathogenic role in diarrhea-predominant irritable bowel syndrome by increasing serotonin biosynthesis, Cell Host Microbe, 2023, 31(1), 33–44.e5.
41. M. Michels, G. Jesus, M. Abatti, E. Córneo, L. Cucker and B. Medeiros, et al., of different probiotic strains B. lactis, L. rhamnosus and L. reuteri on brain-intestinal axis immunomodulation in an endotoxin-induced inflammation, Mol. Neurobiol., 2022, 59(8), 5168–5178.
42. R. Sun, C. Xu, B. Feng, X. Gao and Z. Liu, Critical roles of bile acids in regulating intestinal mucosal immune responses, Ther. Adv. Gastroenterol., 2021, 14, 17562848211018098.
43. M. Lenicek, D. Duricova, V. Komarek, B. Gabrysova, M. Lukas and Z. Smerhovsky, et al., Bile acid malabsorption in inflammatory bowel disease: Assessment by serum markers, Inflamm. Bowel Dis., 2011, 17(6), 1322–1327.
44. S. Zeissig, T. Bergann, A. Fromm, C. Bojarski, F. Heller and U. Guenther, et al., Altered ENaC expression leads to impaired sodium absorption in the noninflamed intestine in crohn's disease, Gastroenterology, 2008, 134(5), 1436–1447.
45. R. Gadaleta, K. Van Erpecum, B. Oldenburg, E. Willemsen, W. Renooij and S. Murzilli, et al., Farnesoid X receptor activation inhibits inflammation and preserves the intestinal barrier in inflammatory bowel disease, Gut, 2011, 60(4), 463–472.
46. O. Garcia-Irigoyen and A. Moschetta, A novel protective role for FXR against inflammasome activation and endotoxemia, Cell Metab., 2017, 25(4), 763–764.

---

Footnote

† These authors contributed equally to this work.

---