Abstract Background Exosomes are involved in intercellular communication and regulation of the inflammatory microenvironment. In a previous study, we demonstrated that fresh ginseng exosomes (GEs) alleviated inflammatory bowel disease. However, the precise mechanism by which GEs activate the immune system and subsequently inhibit the formation of intestinal inflammatory microenvironment remains unknown. Methods Herein, we investigated the effects of GEs on autophagy, macrophage polarisation, intestinal inflammation, and the epithelial barrier by means of transcriptome sequencing, network pharmacology, transmission electron microscopy, immunoblotting, flow cytometry and small molecule inhibitors. Results GEs significantly activated autophagy and M2-like macrophage polarisation, which could be blocked by the autophagy inhibitor 3-methyladenine. In the co-culture system of macrophages and intestinal epithelial cells, macrophages treated with GEs secreted more interleukin-10 (IL-10) and significantly reduced Nitric oxide (NO) levels in intestinal epithelial cells in vitro. Furthermore, GEs acted directly on intestinal epithelial cells through the IKK/IкB/NF-кB signalling pathway to reduce inflammation and restore the intestinal barrier. Orally administered GEs could restore disrupted colonic barriers, alleviate inflammatory bowel responses, and regulate the polarisation of intestinal macrophages in vivo. Conclusion In summary, GEs may be a potential treatment for inflammatory bowel disease, and targeting autophagy and macrophage polarisation may help alleviate intestinal inflammation. Graphical Abstract [44]graphic file with name 12951_2025_3292_Figa_HTML.jpg Supplementary Information The online version contains supplementary material available at 10.1186/s12951-025-03292-3. Keywords: Ginseng exosomes, Autophagy, Intestinal barrier, M1/M2 polarisation Highlights * Using exosomes as nanoplatforms to facilitate intercellular communication. * sRNA is one of the important material bases in GEs. * GEs reduce intracellular inflammation by regulating macrophage phenotype through activation of autophagy. * Macrophage polarised secretions reduce inflammation in intestinal epithelial cells. * GEs can achieve repair of intestinal barrier to alleviate inflammatory bowel disease. Supplementary Information The online version contains supplementary material available at 10.1186/s12951-025-03292-3. Introduction Inflammatory bowel disease (IBD) is characterised by chronic and concomitant recurrent inflammation of the gastrointestinal tract, frequently accompanied by abdominal pain, weight loss, diarrhoea, and other symptoms [[45]1, [46]2]. The number of patients with IBD has been increasing in recent years. Despite the diversification of therapeutic agents, IBD remains a complex clinical problem. IBD patients at higher risk for colorectal cancer, posing significant challenges for both patients and the healthcare system [[47]3]. A more comprehensive and effective treatment strategy is urgently required, as common antibiotic therapies are highly susceptible to liver and kidney toxicity when used for long periods to treat chronic inflammation [[48]4, [49]5]. The causes of IBD are multifaceted, with intestinal inflammation and immune abnormalities being the main culprits. Gut inflammation is associated with impaired immune function, environmental stress, and genetic mutations [[50]6]. Macrophages, which are innate immune cells, could be involved in the remodelling of locally inflammatory-infiltrated tissues through a potential mechanism based on macrophage polarisation [[51]7, [52]8]. Diosgenin ameliorated ulcerative colitis in mice by modulating macrophage polarisation [[53]8]. Macrophages comprise two subpopulations: pro-inflammatory (M1-like) and anti-inflammatory (M2-like). M1-like macrophages induce inflammation and release pro-inflammatory cytokines, including Inducible nitric oxide synthase (iNOS), Tumor necrosis factor-α (TNF-α), Interleukin-1β (IL-1β), and Interleukin 6 (IL-6). Conversely, anti-inflammatory macrophages produce anti-inflammatory cytokines (Arginase 1 (Arg1), interleukin-4 and IL-10), and reduce the deleterious effects of pro-inflammatory cytokines by modulating macrophage phenotype [[54]9]. Therefore, modulation of macrophage polarisation may be an effective method for ameliorating IBD. Furthermore, autophagy can regulate intestinal immune homeostasis [[55]10]. It alleviates chronic diseases like intestinal inflammation by activating the immune system. Autophagy is a process whereby cells degrade and absorb their cytoplasmic proteins and organelles. After forming autophagic lysosomes, they degrade the encapsulated contents to maintain endostasis and metabolic homeostasis [[56]11, [57]12]. Decreased autophagy may lead to an increase in autoimmune and inflammatory diseases. Studies have shown that autophagy stimulators significantly ameliorate experimental colitis and oxidative stress injury and that autophagy is particularly important in intestinal inflammation [[58]13, [59]14]. Moreover, autophagy activation induces macrophage phenotype transition from the pro-inflammatory M1 to the anti-inflammatory M2 phenotype, which facilitates bone repair and regeneration [[60]15]. The intestinal barrier is maintained by a layer of epithelial cells, which separates the intestinal tissue from the large number of microorganisms and harmful substances in the intestinal tract. When the intestinal barrier is disrupted, a large number of bacteria or other microorganisms enter the lumen of the intestinal tract and exacerbate the inflammatory response. Therefore, repairing the damaged intestinal barrier is essential in IBD [[61]14, [62]16, [63]17]. Several studies have discovered that baicalein improves colitis by decreasing intestinal epithelial cell permeability, promoting the protein ZO-1, and regulating the AhR / IL-22 pathway to repair the intestinal epithelial barrier [[64]17]. Loganin attenuates lipopolysaccharide (LPS)-activated expression and release of intestinal epithelial inflammatory factors by modulating TLR4/NF-κB and JAK/STAT3 signalling pathways [[65]18]. Purple carrot anthocyanins inhibited LPS-induced inflammation in a macrophage and Caco-2 co-culture model [[66]19]. Recently, there has been an increase in the study of plant-derived exosomes, which are nanoscale vesicles rich in lipids, proteins, nucleic acids, and other bioactive molecules [[67]20, [68]21]. The high biocompatibility of plant exosomes allows them to play an important role in intercellular communication, while their low toxicity and immunogenicity make them more suitable for the development of medical materials [[69]21]. Numerous studies have revealed that exosomes from fresh fruits, vegetables, and herbs can be used for the treatment and prevention of various diseases. Grape exosomes maintain intestinal homeostasis by activating the Wnt/TCF4 signalling pathway and promoting interspecies communication between plant and mouse intestinal cells [[70]22]. Yam-derived exosome-like nanovesicles promoted tibial growth and stimulated osteoblast proliferation and differentiation in osteoporotic mice [[71]23]. Exosome-like nanoparticles from mulberry bark protect against dextran sulphate sodium (DSS)-induced colitis via the AhR-COPS8-mediated anti-inflammatory pathway [[72]24]. Ginseng (Panax ginseng C. A. Mey.) is a natural chinese herbs that includes bioactivities such as anti-inflammatory, antioxidant and anti-cancer properties [[73]25, [74]26]. Moreover, the active ingredients of ginseng are effective in relieving colitis and intestinal inflammation [[75]27]. Therefore, ginseng and its products could become a new therapeutic strategy against IBD. Previous studies have shown that some components of ginseng can modulate inflammation [[76]28]. Ginsenoside Rb1 attenuates colitis in mice by activating key enzymes in the endoplasmic reticulum [[77]29]. However, the protective mechanism of ginseng-derived exosomes against IBD has been less studied. In a previous study, we demonstrated that ginseng-derived nanoparticles (GDNPs) alleviated IBD through the TLR4/MAPK and p62/Nrf2/Keap1 pathways [[78]30]. In this study, ultracentrifugation and sucrose gradient centrifugation were used to separate and purify nanoparticles of natural medicines and analyse the distribution of microRNAs (miRNAs) in GEs. An inflammation model was then established using LPS-induced macrophage RAW264.7, and the effects of GEs on macrophage autophagy level and polarisation were determined. The mechanism of action of GEs in reducing inflammation in enterocytes was investigated using macrophages and intestinal epithelial cells alone and in a co-culture system. Next, the role of GEs in the IBD mouse model was investigated. Mechanistically, GEs alleviate IBD by activating autophagy, promoting macrophage polarisation towards M2, repairing damaged epithelial tissues, and inhibiting the IKK/IκB/NF-κB signalling pathway. A new perspective on the development of ginseng-based IBD prevention drugs is provided by this study. Materials and methods Isolation and purification of GEs GEs were extracted and purified as previously described [[79]30]. Cell culture and LPS-induced inflammation model The RAW264.7 and Caco-2 cells were procured from Procell Life Science&Technology Co., Ltd. (Wuhan, China). The RAW264.7 cells were cultured in DMEM (Procell) containing 10% FBS at 37 °C with 5% CO[2]. The Caco-2 cells were cultured in MEM complete medium (with NEAA, 10% FBS, Procell) at 37 °C in humid air with 5% CO[2]. Cells were collected using IncPancreatic enzymes(Shandong Sparkjade Biotechnology Co.,Ltd) digestion. The RAW264.7 was modelled using 1 µg/mL LPS, and Caco-2 cells were modelled using 5 µg/mL LPS and co-activated with GEs for 24 h. Macrophages were co-cultured with Caco-2 cells using Transwell chambers; macrophages were inoculated in the lower chamber and Caco-2 cells in the upper chamber. Macrophages were cultured in LPS and different concentrations of GEs for 24 h. After macrophage activation, the solution was changed and then co-cultured with Caco-2 cells for 24 h. The volume of both was 2 mL. RNA sequencing RNA sequencing (RNA-seq) of GEs was conducted by Beijing Novogene Technology Co. The samples were first subjected to total RNA sample testing, library construction, and HiSeq/MiSeq sequencing after library testing. Animal model and treatment C57BL/6 J mice (6–8 weeks, Male, 18–22 g) were acquired from Changchun Yisi Laboratory Animal Technology Co. (Changchun, China). The mice were divided into five groups (n = 6): negative control, DSS-induced IBD, low-dose GEs-treated IBD, high-dose GEs-treated IBD, and sulfasalazine-treated IBD. All mice were pre-protected for three days before the establishment of the mouse IBD model. For the first three days, each mouse was given free access to drinking water and gavage of 0.2 mL of the corresponding therapeutic drug per day. The negative control and the model groups were given 20 mM Tris–HCl, and the treatment group was given GEs (5 mg/mL or 10 mg/mL protein of GEs) and sulfasalazine (0.3 g/kg). On day 4, the negative control group received normal drinking water, and the remaining experimental groups received 2.5% DSS solution in their drinking water while the therapeutic drugs were still administered. The mice were euthanised on day 10, and fluorescein isothiocyanate (FITC) was used to detect GEs-induced changes in intestinal permeability. Mice were gavaged with 3.5 mg FITC-dextran (Sigma-Aldrich 3–5 kDA)/10 g before execution. After 4 h, serum from mice was collected, and FITC-dextran content was measured using an enzyme marker. H&E staining The organs and intestinal tissues of mice were fixed with 4% paraformaldehyde(POM), then gradually dehydrated, paraffin-embedded, cut into 4 μm sections, and stained with H&E. Changes in the tissues of each group were observed and scanned using a pathology section scanner. qRT-PCR Total RNA was extracted with TRIzol (Beyotime). Fluorescence quantification was performed using the Reverse Transcription Kit and Fluorescence Quantification Kit (TransGen Biotech, China) and a quantitative PCR instrument. The relevant mRNA primers for the target genes are listed in Table [80]1. Data relative to GAPDH expression levels were calculated using the 2^−∆∆Ct method. Table 1. Primers used for the qRT-PCR study (RAW264.7) Gene Primer (5′ to 3′) GAPDH F:AAGGTCATCCCAGAGCTGAA R:CTGCTTCACCACCTTCTTGA CD86 F:CTGCACGTCTAAGCAAGGTC R:CAGAACACACACAACGGTCA CD206 F:CTCTGTTCAGCTATTGGACGC R:CGGAATTTCTGGGATTCAGCTTC INOS F:GGACCCAGTGCCCTGCTTT R:CACCAAGCTCATGCGGCCT TGF-β F:GCCATAACCGCACTGTCATT R:AAGGTGGTGCCCTCTGAAAT [81]Open in a new tab Western blot analysis Proteins were extracted from cells and tissues using RIPA lysis buffer, and the protein concentration was measured using a BCA kit. Protein samples were separated using 12% sodium dodecyl sulphate–polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, and sealed with 5% skimmed milk powder. The membrane was washed with PBST and incubate with primary antibody overnight, then incubate with secondary antibody for 1 h at 25 °C. β-actin was obtained from Bioss Technology, Beijing, China. The protein bands were visualised using a chemical gel imaging system and ECL Kit. Image J (v1.8.0) was chosen for protein blotting quantification. Immunofluorescence (IF) The RAW264.7 and Caco-2 cells were fixed in 4% POM at 25 °C for 15 min, then washed thrice in PBS and incubated with 0.1% Triton X-100 for 15 min to punch the cells. The cells were blocked with 5% BSA solution for 30 min at 25 °C. Primary antibody incubated overnight, secondary antibody conjugated, nuclei stained with DAPI and then sealed with an anti-quencher. The colon tissue was sliced by paraffin, dewaxed, antigenically repaired, washed, and slightly shaken dry before being circled around with a histochemical pen, sealed with a drop of 3% BSA for 30 min, and operated in the same way as the cells. Photographs were taken using a laser confocal microscope protected from light. ELISA The quantities of diamine oxidase (DAO), 5-hydroxytryptamine (5-HT), substance P (SP, TNF-α), IL-10, and IL-1β were determined in the culture supernatants, serum, and tissue using ELISA kits according to the instructions. The sensitivity of the assay was < 5 pg/mL. GFP-LC3 transfection Before transfection with the GFP-LC3 plasmid, RAW264.7 cells were inoculated in a 24-well plate. After 24 h, the cells were transfected with Lipofectamine 2000 and plasmid (200 ng/well) diluted in Opti-MEM according to the manufacturer's instructions. The OPTI-MEM was replaced with a medium containing GEs and GEs + 3-methyladenine (3-MA) after 2 h. The expression of LC3 was observed using laser confocal microscopy after 24 h. The GFP-LC3 plasmid used in this study was provided by Jilin University. Transwell invasion assay The Caco-2 cells were inoculated in Transwell chambers (Corning). After reaching confluence, the cells were incubated for 21 days to fully differentiate and form tight junctions (TJs). The medium was changed three times per week. To induce inflammation, the differentiated Caco-2 cells were treated with LPS and co-cultured with LPS + drugs for 24 h. Then, 0.5 mg/mL FITC-dextran (Sigma-Aldrich 3–5 kDA) was added to the upper layer of the Transwell and incubated for 2 h, and samples were obtained from the lower layer of the liquid 96-well plate. The FITC-dextran concentration was measured using an enzyme marker. Flow cytometry to detect the polarisation of intestinal macrophages Intestinal tissues were collected, cut longitudinally, and washed with precooled DPBS to remove faeces. Then, it was cut into small pieces and incubated in 0.15% EDTA with shaking for 45 min. The supernatant was discarded, and the remaining tissue pieces were digested with a mixture of digestive enzymes in a shaker water bath at 37 °C for 1 h with blowing and mixing every 20 min. The digestive enzyme solution consisted of 2% collagenase IA, 2.5% collagenase type IV, 2.5% collagenase type II, and 0.25% hyaluronidase. Following sequentially filtering digested tissues through 100 mesh and 200 mesh sieves, the supernatant was removed by centrifugation at 1000 rpm for 5 min. After washing twice with PBS, the cells were resuspended in PBS and centrifuged at 2000 rpm for 20 min. The intermediate white membrane layer was aspirated, washed, and subsequently stained with CD68, CD80 and CD206 for 30 min, respectively, and analysed using flow cytometry. Statistical analysis Statistical analysis was carried out with GraphPad Prism software, using mean and standard deviation. Dunnett’s post-hoc test was used to compare multiple groups in a one-way analysis of variance. p < 0.05 was considered significant. Results Isolation and characterisation of GEs To isolate exosomal nanoparticles from fresh ginseng, a homogeniser was used for adequate homogenisation at 10 °C. The ginseng juice was sequentially centrifuged at low speed to remove impurities, followed by ultracentrifugation to isolate GEs and sucrose gradient density centrifugation for purification (Fig. [82]1A). GEs were extracted and purified for quantification using a BCA kit. The morphology and size of GEs, as well as their potential, were observed using transmission electron microscopy and Zetasizer Nano ZS. The results revealed that the GEs had double-membrane spherical structures (Fig. [83]1B) with an average zeta potential of -38.1 mV (Fig. [84]1C). The average particle size of the GEs was 253.61 nm (Fig. [85]1D). The lipid content, protein species, and ginsenoside content of the GEs have been determined in previous experiments [[86]30, [87]31]. Fig. 1. [88]Fig. 1 [89]Open in a new tab Isolation and characterisation of GEs. A Schematic diagram of the process of isolating GEs from ginseng. B Specific morphology of GEs observed using transmission electron microscopy (TEM). Scale bar: 200 nm. C Potential of GEs. D Particle size of GEs GEs transcriptome sequencing Transcriptomics can detect known miRNAs of GEs containing 404 mature and 397 hairpins (Table [90]2). The signature hairpin structure of miRNA precursors can be used to predict novel miRNAs. We predicted that the novel miRNAs for GEs contain 14 mature and 14 hairpins (Table [91]3). The sRNAs classification is shown in Table [92]4. The length of small RNAs in GEs was then determined (Fig. [93]2A). The secondary structures of known miRNAs (Fig. [94]2B) and the new miRNA (Fig. [95]2C) were also obtained. The process of miRNA development from hairpin to mature body is completed by Dicer cleavage. The specificity of the cleavage site makes the first base of the miRNA mature body sequence significantly biased; therefore, we investigated the first-site base distribution of miRNAs of different lengths, and statistics of the distribution of bases at each locus of miRNAs were also examined (Fig. [96]2D, [97]E). Detailed information about sRNAs in GEs was obtained, and the miRNAs were pooled with IBD cross-target genes, resulting in 5021 target genes (Fig. S1A). A total of 50 core target genes, including NFкB1, TNF, IL-6, IL-10, and IL-1β, were obtained from the STRING database using Cytoscpace and CytoNCA plug-in, and a protein–protein interaction (PPI) map was obtained (Fig. S1B). Moreover, KEGG pathway enrichment analysis revealed that GEs may play an important role in pathways related to autophagy, cancer, lipids, and atherosclerosis (Fig. S1C) (Table [98]5). Table 2. Number of known miRNA mature and hairpin in GEs Types Total GEs Mapped mature 404 397 Mapped hairpin 404 397 [99]Open in a new tab Table 3. Number of novel miRNA mature and hairpin predicted in GEs Types Total GEs Mapped mature 14 14 Mapped hairpin 14 14 [100]Open in a new tab Table 4. Statistical table of sRNA classification in GEs Types GE-sRNA Percent Total 3,908,609 100% Known-miRNA 33,180 0.85% rRNA 43,934 1.12% tRNA 1476 0.04% snRNA 7268 0.19% snoRNA 26,780 0.69% repeat 61,585 1.58% novel-miRNA 163 0.00% Exon: +  58,406 1.49% Exon: − 174,458 4.46% Intron: +  514,655 13.17% Intron: − 157,369 4.03% Other 2,829,335 72.39% [101]Open in a new tab Fig. 2. [102]Fig. 2 [103]Open in a new tab Small RNA transcriptome sequencing of GEs. A Length distribution statistics of the total sRNAs fragments. B, C Secondary structures of known and unknown miRNAs, respectively. D First nucleotide bias of known miRNAs (18–30 nt) in GEs. The horizontal coordinate represents the length of the miRNA, and the vertical coordinate is the percentage of the first base in the miRNA of that length in which A/U/C/G occurs (the value at the top of the bar represents the total number of miRNAs of that length). E miRNA nucleotide bias at each position. The horizontal coordinate represents the base position of the miRNA, and the vertical coordinate is the percentage of bases A/U/C/G occurring in the miRNA at that position Table 5. Gene types for each group of samples Sample A1 A2 A3 B1 B2 B3 C1 C2 C3 lncRNA 136,773 136,200 131,962 160,787 122,392 162,456 128,938 168,366 139,995 TEC 12,362 13,027 12,444 14,136 10,346 14,751 11,731 15,335 12,381 Processed_pseudogene 223,519 206,000 202,515 256,393 208,516 257,389 201,684 257,401 225,615 IG_J_gene 133 135 139 1 0 1 0 1 1 IG_C_gene 3012 2670 2727 254 218 242 194 246 226 Mt_rRNA 14,532 13,948 13,603 13,956 11,105 13,674 12,669 15,937 13,789 Transcribed_processed_pseudogene 36,567 32,498 32,558 34,610 27,588 34,487 27,709 35,320 30,609 snoRNA 1001 1046 910 1143 782 1196 937 1209 961 scaRNA 41 54 56 34 26 55 37 42 35 Transcribed_unitary_pseudogene 739 690 683 729 620 728 602 764 685 Pseudogene 819 891 780 847 690 868 733 877 752 snRNA 407 488 444 387 265 372 315 372 328 Transcribed_unprocessed_pseudogene 547 692 589 1563 954 1594 1072 1430 1195 rRNA 5644 5989 5806 8638 6084 8153 6600 8211 7005 Unitary_pseudogene 49 44 41 68 30 67 36 63 68 misc_RNA 198 207 170 148 134 186 148 192 127 Mt_tRNA 600 503 530 984 888 1036 598 1084 924 Protein_coding 14,800,756 14,387,620 13,982,045 16,449,301 13,044,661 16,718,534 13,061,021 16,924,737 14,441,011 Unprocessed_pseudogene 4401 3940 3865 4093 3680 4153 3522 4325 3949 miRNA 590 581 537 692 557 793 621 800 611 [104]Open in a new tab GEs modulate M1/M2-like macrophage polarisation and enhance autophagy in LPS-stimulated macrophages We confirmed through protein immunoblotting experiments that GEs affected RAW264.7 macrophage polarisation (Fig. [105]3A), down-regulated the expression levels of CD86, an indicator of M1 polarisation, and increased CD206, an indicator of M2 polarisation in the inflammation model. In addition to the polarisation-associated surface markers, we also examined polarisation-associated factors and proteins (CD86, INOS, CD206, TGF-β) using QPCR (Fig. [106]3B). The results showed that the GEs reduced LPS-induced macrophage inflammatory responses by modulating macrophage polarisation ratios, consistent with previous studies [[107]30]. Fig. 3. [108]Fig. 3 [109]Open in a new tab Detection of macrophage polarisation modulation by GEs. A Protein expression of β-actin, CD86, and CD206 was measured using western blotting (WB). B Changes in intracellular CD86, INOS,CD206 and TGF-β mRNA levels in LPS-stimulated RAW264.7 cells treated with and without different doses of GEs. Values are shown as the mean ± SD. n = 3. ^#p < 0.05, ^###p < 0.05 vs. Control; **p < 0.01, ***p < 0.001 vs. Model Autophagy suppresses inflammation by inhibiting the activation of NF-κB and inflammasomes [[110]32, [111]33]. Therefore, we continued to investigate whether GEs could activate cellular autophagy to exert immunomodulatory effects. Several studies have demonstrated that autophagy plays a role in various aspects of innate and adaptive immunity [[112]34]. To assess whether GEs are involved in immune cell-macrophage autophagy, we first used confocal microscopy to determine the autophagy status of GFP-LC3 mono-fluorescently labelled macrophages. The results revealed that the green fluorescent spots of LC3 were significantly increased, and autophagy was induced in the presence of GEs. The addition of 3-MA inhibited the formation of autophagosomes, resulting in the weakening of the green spots (Fig. [113]4A). The expression level of the autophagy-related protein LC3 was subsequently determined in RAW264.7 cells. The results revealed that the levels of LC3 II, Atg7, and Beclin-1 proteins were increased in the RAW264.7 cells after GEs treatment compared to the LPS-treated group. AKT/mTOR pathway was associated with autophagy induction [[114]35].P-mTOR/mTOR and p-AKT/AKT were downregulated by GEs treatment (Fig. [115]4B). These data indicate that GEs can promote autophagy in RAW264.7 cells through the AKT/mTOR pathway. The detection of LC3 in cells at different time points showed that GEs were more effective at 24 and 48 h than at 12 h (Fig. [116]4C). LC3-I can enzymatically convert a small segment of polypeptide into LC3-II during autophagy. LC3-II is subsequently recruited to the autophagosome membrane to promote autophagic lysosome formation. The magnitude of the LC3-II/I ratio can be used to estimate the level of autophagy; however, it is insufficient to accurately reflect the level of autophagy due to the dynamic changes that LC3 undergoes during autophagy. Next, we investigated macrophage autophagy by introducing inhibitors and observing autophagy lysosomes. The changes in autophagic lysosomes in RAW264.7 cells treated with GEs in the inflammatory state were observed using transmission electron microscopy (Fig. [117]4D), indicating that GEs increased the number of autophagic lysosomes in the inflammatory state. GEs alone elevated the cellular autophagy marker LC3. The presence of 3MA blocked autophagy and reduced Atg 7 expression (Fig. S4). These results demonstrate that GEs activated autophagy, thereby attenuating LPS-induced cellular inflammatory injury. Fig. 4. [118]Fig. 4 [119]Open in a new tab Determining the regulatory role of GEs in macrophage autophagy. A Confocal microscopy to detect the formation of GFP-LC3 spots. Scale bar: 10 µm. B Protein expression of β-actin, Beclin1, Atg7, LC3I/II, m-TOR, p-m-TOR, AKT, and p-AKT was measured using WB. C Protein expression of autophagy factors LC3 at different time points. D GEs increased the intracellular autophagic lysosomal activity. Scale bar: 5 µm or 2 µm. Values are shown as the mean ± SD. n = 3. ^##p < 0.01 vs. Control; *p < 0.05, **p < 0.01, ***p < 0.001 vs. Model GEs-induced autophagy leads to macrophage polarisation Autophagy is closely associated with inflammation and immunity. Enhanced autophagy in sepsis can play a protective role by modulating macrophage polarisation phenotypes, reducing the activation of inflammatory vesicles, and releasing inflammatory factors [[120]36]. GEs not only activate cellular autophagy but also modulate macrophage polarisation. We investigated the potential association between the two. The generation of autophagic vesicles was inhibited by the addition of 3-MA, an early inhibitor of autophagy. The expression levels of CD80 and CD206 were determined. The involvement of 3-MA increased the M1/M2 ratio and exacerbated the cellular inflammatory response (Fig. [121]5A, [122]B). These results demonstrated that ginseng exosomes promote the polarisation of M2-like macrophages by inducing autophagy. Fig. 5. [123]Fig. 5 [124]Open in a new tab Activation of autophagy leads to macrophage polarisation. A Expression of CD80 and CD206 in the presence of 3-MA. B Quantile of M1/M2 with 3-MA participation. C, D TNF-α and IL-10 concentrations in macrophage supernatants. E NO levels in the supernatant of Caco-2 cells. F Protein levels of NF-κB in Caco-2 cells. G Schematic diagram of the autophagy-regulated polarisation process. Values are shown as mean ± SD. n = 3. ^#p < 0.05, ^##p < 0.01 vs. Control; *p < 0.05, **p < 0.01, ***p < 0.001 vs. Model M2-like macrophages produce an anti-inflammatory response by secreting anti-inflammatory cytokines, like IL-10, which promotes inflammatory regression and new tissue formation [[125]37]. Dysregulation of intestinal epithelial cytokine secretion and signalling mechanisms associated with IBD pathogenesis, and NF-κB expression and activation are significantly induced in the inflamed intestine of IBD patients. Particularly, Macrophages and epithelial cells isolated from inflamed intestinal samples from IBD patients contain high levels of NF-κB [[126]38]. Therefore, we investigated the effects of GEs on macrophages and intestinal epithelial cells. In a co-culture system of macrophages and intestinal epithelial cells, we measured the levels of pro-inflammatory factor (TNF-α) and anti-inflammatory factor (IL-10) in macrophage supernatants using ELISA. The results revealed that GEs decreased TNF-α and elevated IL-10 levels in a concentration-dependent manner (Fig. [127]5C, [128]D). Macrophage-polarised secretions in the presence of GEs reduced the NO concentration in the supernatant of Caco-2 cells and Protein levels of NF-κB in Caco-2 cells (Fig. [129]5E, F). The flowchart is shown in Fig. [130]5G. Combined with a pharmacological screen identifying NF-κB as a potential target of GEs in the treatment of IBD, we went on to investigate the effects of GEs on Caco-2 cells. Effects of GEs on Caco-2 cells in inflammatory states To elucidate the mechanism of action of GEs in the inflammatory cascade response of enterocytes, WB was used to analyse the expression levels of key proteins involved in the activation of the NF-κB pathway. LPS significantly increased the expression of P-IKKα/β (p < 0.05), P-IĸBα, Cox-2, and TNF-α (p < 0.01) compared to that in the control group (Fig. [131]6A) and attenuated the expression level of inflammation-associated proteins under the action of GEs. Moreover, we analysed the nuclear entry of NF-кB. The results revealed that GEs effectively inhibit LPS-induced NF-кB nuclear translocation (Fig. [132]6B). Asperuloside intervention significantly reduced NF-кB protein levels. The effects of GEs were consistent with those of asperuloside (Fig. [133]6C). These findings indicate that GEs counteract LPS-induced activation of inflammatory cascades and exert anti-inflammatory effects through the IKK/IкB/NF-кB signalling pathway. Fig. 6. [134]Fig. 6 [135]Open in a new tab GEs attenuate cellular inflammation by targeting the IKK/IкB/NF-кB signalling pathway. A Cells were untreated, treated with 5 μg/mL LPS alone for 24 h, or co-treated with 5 μg/mL LPS and GEs (5, 10, or 20 μg/mL) for 24 h. WB analysis was conducted to determine protein expression. Relative expression levels of P-IKKα/β, P-IĸBα, NF-κB, Cox-2 and TNF-α . B Expression levels of the NF-кB nuclear translocator protein in intestinal epithelial cells. C After treatment with asperuloside, the expression levels of NF-кB and β-actin were measured using WB and respective antibodies. β-actin was used as a protein loading control. D FITC-dextran flux assay of Caco-2 cells after GEs treatment. Values are shown as the mean ± SD. n = 3. ^#p < 0.05, ^##p < 0.01 vs. Control; *p < 0.05, **p < 0.01, ***p < 0.001 vs. Model Disruption of the intestinal epithelial barrier underlies IBD [[136]39]. Intestinal permeability is a functional characteristic of the intestinal barrier. We used FITC-dextran to investigate the effects of GEs on intestinal epithelial cell permeability and elucidate the protective effect of GEs on the intestinal epithelial barrier. Analysis of intestinal cell permeability assays revealed that FITC-dextran flux in Caco-2 cells was increased following treatment with LPS (Fig. [137]6D); however, the increase was prevented by GEs treatment. This demonstrates that GEs decrease the permeability of intestinal epithelial cells and repair intercellular TJs. GEs alleviate DSS-induced mouse inflammatory bowel disease To determine the protective effect of GEs in a mouse model of IBD, we first evaluated morphological changes in the intestinal tissue. GEs treatment significantly alleviated the pathological changes in the intestine, including inflammatory cell infiltration, reduction of cuprocytes, widening or disappearance of gland spacing, and epithelial rupture (Fig. [138]7A). GEs also increased intestinal length (Fig. [139]7B). The disease activity index (DAI) was used to assess the extent of IBD in the mice, and body weight, faecal consistency, and faecal bleeding were recorded. The three composite scores demonstrated that GEs improved the disease state of intestinal inflammation in mice. In addition, GEs improved the survival rate of mice (Fig. [140]7C). To determine the anti-inflammatory effects of GEs on the IBD model. The IL-1β and IL-10 were first detected in the intestinal tissues, and the results revealed that the administration of GEs to mice reduced the expression of IL-1β in the intestinal tissues, while the expression of IL-10 anti-inflammatory factor was increased compared to the model group (Fig. [141]7D, [142]E). Fig. 7. [143]Fig. 7 [144]Open in a new tab A Histological staining. Scale bar: 100 µm. B Mouse intestinal length. C DAI index and Survival rate in mice. D, E Levels of IL-1β and IL-10 in intestinal tissue. Values are shown as the mean ± SD. n = 3. ^##p < 0.01, ^###p < 0.001 vs. Control; *p < 0.05, **p < 0.01, ***p < 0.001 vs. Model GEs alleviate IBD in mice by protecting the intestinal tract epithelial barrier We further investigated the effects of GEs on DSS-induced IBD in mice. It has been found that production of inflammatory factors in the gut is mostly due to defects in the epithelial barrier and alterations in mucus production, leading to increased intestinal permeability and overexposure of the mucosal immune system to external luminal contents [[145]19]. We subsequently assessed intestinal permeability and barrier function in IBD mice. In DSS-induced IBD mice, GEs reduced the elevated serum levels of FITC-dextran (Fig. [146]8A). This showed that intestinal permeability was reduced, and intercellular TJs were repaired. Plasma levels of D-lactic acid were then measured using a biochemical kit. When the intestinal barrier is damaged, the intestinal flora produces a large amount of D-lactic acid into the bloodstream through the damaged intestinal mucosa, reflecting the changes in intestinal permeability and intestinal mucosal damage [[147]40]. The plasma levels of D-lactic acid in mice were significantly higher after DSS induction, which was attenuated by GEs (Fig. [148]8B). This indicates an improvement in intestinal mucosal damage and a reduction in intestinal permeability. Fig. 8. [149]Fig. 8 [150]Open in a new tab GEs protect the intestinal mucosa to repair the intestinal barrier. A FITC-dextran flux. B Plasma levels of D-lactate. C Colonic tissue and serum DAO levels. D 5-HT and (E) SP. (F) Immunofluorescence detection of mouse intestinal tight junction protein (ZO-1). Scale bar: 100 µm. G Protein expression levels of the tight junction protein occludin. Values are shown as the mean ± SD. n = 3. ^#p < 0.05, ^##p < 0.01 vs. Control; *p < 0.05, **p < 0.01, ***p < 0.001 vs. Model To determine whether GEs can repair the intestinal mucosa, we examined mouse intestinal tissue and serum DAO levels using ELISA. DAO is an intracellular enzyme in the villi of the upper mucosal layer of the small intestine in humans and mammals. It is involved in histamine and several polyamine metabolism. Its activity is closely related to nucleic acid and protein synthesis in mucosal cells, which can reflect the integrity of the intestinal barrier and the degree of damage sustained [[151]41]. Our results revealed that DAO content in the intestinal tissues of mice was significantly reduced, and serum DAO was elevated after DSS induction, which was improved by GEs treatment (Fig. [152]8C). This indicates that GEs alleviated the diffusion of DAO from the intestinal mucosa into the bloodstream and mitigated damage to the intestinal mucosa. Subsequently, ELISA was used to detect the neurotransmitters 5-HT and SP in intestinal tissue. 5-HT is produced in the gut and is synthesised by intestinal chromaffin cells and intestinal interosseous nerves, accounting for 95% of the total, with the remainder synthesised in central neurons [[153]42]. SP is a tachykinin neuropeptide expressed in the central and peripheral nervous system, as well as in the enteric nervous system and immune cells. SP can cause neurogenic inflammatory responses, including vasodilation, increased permeability, and extravasation of blood proteins in this innervated area [[154]43]. The results revealed that the levels of 5-HT and SP in the intestinal tissues of mice were significantly reduced after DSS induction, while their serum levels were elevated, which was improved after GEs treatment (Fig. [155]8D, E). This indicated that GEs reduced the extravasation of 5-HT and SP from the intestinal mucosa into the bloodstream during intestinal stimulation and reduced the sensitivity of the intestinal mucosa. Furthermore, the proteins (ZO-1 and occludin) in intestinal tissues was evaluated using immunofluorescence. The results revealed that ZO-1 and occludin in the model group appeared to have discontinuous cell boundaries, and a portion of the tight junction proteins disappeared from the cell boundaries (Fig. [156]8F and S5). GEs significantly rescued the deletion of tight junction proteins in the IBD model group. The same pattern was observed for occludin tight junction protein in the colon tissue (Fig. [157]8G). Autophagy plays a crucial role in the regulation of the intestinal epithelial barrier, and deletion of autophagy gene expression leads to changes in the expression or distribution of intestinal junctional proteins [[158]44]. We examined the autophagy level of intestinal epithelial cells and found that GEs increased the expression of green fluorescent dots in GFP-LC3 monofluorescent labelling, and protein blotting experiments showed that the protein levels of LC3 II and Atg 7 were significantly increased in the presence of GEs. GEs activated autophagy in intestinal epithelial cells (Fig. S6). GEs regulate macrophage polarisation and intestinal autophagy levels in vivo Based on these results, we further investigated the effects of GEs on intestinal macrophage polarisation in an IBD model. First, the levels of M1-like and M2-like macrophage markers, including CD86 and CD206, were determined. The results revealed a significant increase in CD86 expression in the colonic mucosa and lamina propria in the DSS-induced model group; however, GEs treatment significantly decreased macrophage infiltration into the intestinal mucosa and lamina propria, while the opposite was true for CD206. The positive control group (sulfasalazine) exhibited no effect on CD86 expression compared to DSS group, although it increased CD206 expression (Fig. [159]9A and C). Subsequently, flow cytometry was used to examine macrophage polarisation in the mouse colon. The results revealed that GEs down-regulated the ratio of M1/M2-like intestinal macrophages compared to the mouse intestinal inflammation model group (DSS group) (Fig. [160]9B and D), thereby reducing intestinal inflammation in mice. These data indicate that GEs exert anti-inflammatory effects by modulating macrophage polarisation in vivo. Moreover, LC3 and Atg7 protein levels increased in intestinal tissues after GEs treatment compared to those in the DSS group (Fig. [161]9E), suggesting that GEs induced autophagy in mice in vivo. Fig. 9. [162]Fig. 9 [163]Open in a new tab Effects of GEs on intestinal cell polarisation and intestinal autophagy. A Expression of CD86 and CD206 in intestinal tissue. Scale bar: 100 µm. B Expression of CD80 and CD206 in the intestinal macrophages. C Quantitative immunohistochemistry of the intestinal tissue. D Quantification of macrophage polarisation detected using flow cytometry. E WB analysis of protein expression and relative expression levels of Atg7 and LC3. Values are shown as the mean ± SD. n = 3. ^##p < 0.01, ^##p < 0.001 vs. Control; **p < 0.01, ***p < 0.001 vs. Model Molecular mechanisms by which GEs attenuate DSS-induced inflammatory bowel disease in mice We evaluated the expression levels of P-IKKα/β, P-IĸBα, and NF-κB in intestinal tissues using WB. The results obtained using the pharmacological model were similar to those obtained at the cellular level. P-IKKα/β, P-IĸBα, and NF-κB levels were significantly elevated in the model group, while the levels of inflammatory proteins were reduced in both the GEs and positive control groups (Fig. [164]10A). In addition, we also assessed the expression of NF-κB in mouse intestinal tissue nuclear proteins and showed that GEs effectively inhibited DSS-induced NF-κB nuclear translocation (Fig. [165]10B). These data indicate that GEs can exert anti-inflammatory effects in a mouse model of IBD by targeting the IKK/IкB/NF-кB signalling pathway. Our results indicate that GEs alleviate IBD in mice by activating intestinal autophagy, regulating intestinal macrophage polarisation, reducing intestinal inflammation, and restoring intestinal epithelial barrier function. Fig. 10. [166]Fig. 10 [167]Open in a new tab Ginseng exosomes alleviate IBD through the IKK/IкB/NF-кB signalling pathway. A WB analysis of protein expression and relative expression levels of P-IKKα/β, P-IĸBα, and NF-κB. B Expression levels of NF-кB nuclear transporter protein in intestinal tissues. Values are shown as the mean ± SD. n = 3. ^##p < 0.01, ^###p < 0.001 vs. Control; *p < 0.05, **p < 0.01, ***p < 0.001 vs. Model Discussion Nanotechnology is currently being used in the treatment of various diseases, including plant-derived exosomes. Plant exosomes are rich in lipids, proteins, and nucleic acids while being less cytotoxic. As nanoparticles are more readily absorbed, they effectively target specific tissues [[168]23, [169]45]. GDEN-rich 6-gingerol activates Nrf2 by regulating the TLR2/TRIF pathway and prevents alcohol-induced liver injury via the anti-inflammatory effects of this pathway [[170]46]. Oral administration of plant exosome complexes restored redox levels in mice stimulated with external stimuli. Modern scientific studies have demonstrated that at the cellular and model animal experimental levels, ginseng and its main components exhibit a therapeutic effect on IBD. Red ginseng and Coix lacryma can improve intestinal microbial structure and alleviate the symptoms of ulcerative colitis [[171]47]. Fermented wild ginseng reduces the severity of colitis and the level of macrophage infiltration in colonic tissues, primarily through attenuation of the NF-κB signalling pathway [[172]48]. Ginseng saponin Rf reduces the production of inflammatory mediators in TNF-α-stimulated intestinal epithelial cells and mouse macrophages and inhibits TNF-α/lipopolysaccharide-induced NF-κB transcriptional activity [[173]49]. These data demonstrate the therapeutic effects of ginseng on IBD. In this study, we elucidated the mechanism of action of GEs in the treatment of IBD by investigating the effects of GEs on macrophages and alleviating inflammation of intestinal epithelial cells. We first used differential centrifugation and sucrose density gradient centrifugation to isolate and purify GEs (Fig. [174]1) and sequenced their transcriptomes to determine the sRNAs information (Fig. [175]2 and Table [176]2). Next, we conducted a PPI analysis of the target genes of miRNAs and IBD to determine significant associations with TNF, IL-6, NF-κB, and autophagy (Fig. S1). Therefore, we focused on changes in inflammation and autophagy levels in gut-associated cells. Macrophages as guardians of immune homeostasis in intestinal tissues, leading to chronic recurrent immune activation and gastrointestinal pathologies like IBD during impaired regulation of intestinal immunity [[177]50]. Macrophages play a crucial role in promoting and maintaining the anti-inflammatory state of the intestinal tract; therefore, they are considered potential targets for the development of new therapeutic approaches [[178]51]. GEs modulated macrophage polarisation ratios in inflammatory states in vitro (Fig. [179]3). Autophagy, with its key role in inflammation and immunity, is an essential homeostatic process [[180]52, [181]53]. Moreover, we evaluated the changes in autophagy levels in RAW cells by examining the expression of the autophagy marker LC3 and other autophagy-related factors like ATG7 and Beclin1 at different time points. The results indicated that LC3 expression levels increased in response to GEs (Fig. [182]4). After 48 h of action, the high cell density and nutrient depletion in the culture medium caused an increase in autophagy levels in the blank group, which peaked in the GEs group at 24 h, and there was no sufficient substrate for transformation in the GEs group at 48 h. Next, TEM was used to observe the changes in autophagic lysosomes in ginseng exosome-intervened RAW264.7 cells in the inflammatory state. The results revealed that GEs increased the number of autophagic lysosomes in the inflammatory state. Therefore, we concluded that GEs induced autophagy in macrophages. Furthermore, the results demonstrated that autophagy induced macrophage polarisation (Fig. [183]5). The intestinal epithelial barrier consists of a layer of epithelial cells in the intestinal lining and the TJs between cells. It plays a key role in the mucosal immune response and acts as a physical barrier that prevents the excessive entry of bacteria and other antigens from the intestinal lumen into the circulation [[184]54]. In IBD, paracellular gap permeability is increased, and TJ regulation is impaired [[185]55]. We initially assessed the intestinal barrier function by measuring the intestinal permeability of Caco-2 cells in the inflammatory state and discovered that GEs reduced the LPS-induced increase in intestinal permeability (Fig. [186]6D). Secreted factors after macrophage polarisation effectively reduced NO levels in intestinal epithelial cells in a co-culture system (Fig. [187]5E). Disruption of the intestinal barrier accelerates the intestinal inflammatory response, resulting in sustained epithelial damage, cachexia, ulceration, and reduced defensins [[188]55]. Exosomes from human umbilical cord mesenchymal stem cells (MSCs) repair the intestinal barrier and alleviate IBD via TSG-6 [[189]56]. Extracellular vesicles of citrus origin can alter the expression of tight junction-related genes and restore the functional barrier [[190]57]. TLR4 is the key membrane signalling receptor that induces inflammation in IBD, and several canonical signalling pathways, including the NF-κB pathway, play a potential role in this process [[191]58]. The MSC-derived exosomes reduce TLR4 expression in epithelial cells and cellular infiltration and alleviate acetic acid-induced ulcerative colitis [[192]59]. Next, we evaluated the level of NF-κB protein expression and discovered that GEs reduced the nuclear translocation of NF-κB protein. Additionally, it activated the IKK/IкB/NF-кB signalling pathway and attenuated inflammation in intestinal epithelial cells (Fig. [193]6). Furthermore, the favourable in vitro results prompted us to investigate the effects of GEs in vivo further. The results demonstrated that GEs similarly modulated intestinal macrophage polarisation and induced autophagy in vivo (Fig. [194]9). We focused on the restorative effects of GEs on intestinal barrier function. First, intestinal permeability and plasma D-lactate levels were measured, and it was confirmed that GEs reduced intestinal permeability. Second, DAO measurements indicated that intestinal mucosal damage was alleviated. Although it is widely believed that 5-HT research originated in the central nervous system, most of the early research on 5-HT was conducted in the gastrointestinal tract; therefore, it is known as an intestinal stimulating factor-enteropeptide, which is involved in many physiological and pathophysiological processes in the CNS and periphery. Since then, there has been an increasing number of studies on enteropeptides, including 5-HT and SP [[195]60]. The current study demonstrated that 5-HT and SP were released from the intestinal tissues into the bloodstream after DSS stimulation in mice, which was ameliorated by GEs treatment. However, in previous studies, 5-HT and SP levels often increased or decreased in both tissues and blood after drug treatment. This controversy is not currently reported to explain the specific reasons which may be linked to the inflammatory stage and the complex dynamic balance of mice. Intestinal epithelial cells are connected by TJs, composed of several transmembrane molecules, including proteins like claudins, occludin, ZO-1, and others, which are important for maintaining TJ function [[196]61, [197]62]. Subsequently, we measured the expression of tight junction proteins ZO-1 and occludin using immunofluorescence. The results demonstrated that ginseng exocytosis increased the expression of tight junction proteins in the intestinal cells (Fig. [198]8 and S5). In a DSS-induced mouse model of IBD, GEs increased the number of cup cells in the epithelium, decreased cellular infiltration, and inhibited the expression of inflammatory factors (Fig. [199]9A). The specific mechanisms through which GEs regulate IBD were revealed further by measuring the expression of the IKK/IкB/NF-кB pathway and intercellular tight junction proteins. Conclusions DSS causes colitis for the following reasons: 1. direct damage to intestinal epithelial cells and disruption of the intestinal barrier [[200]63]; 2. bacterial stimulation of macrophages in the lamina propria and activation of the immune response [[201]64]; 3. proinflammatory factors promote the inflammatory cascade response [[202]65]. In this paper, IKK/IкB/NF-кB is the main cause of inflammation in the intestinal epithelium, and the secretion of pro-inflammatory factors by M1-type macrophages will continuously amplify the inflammatory response in the intestinal epithelium, especially modulating NF-кB. Therefore, we believe that the inhibition of IKK/IкB/NF-кB is the main factor in repairing the intestinal epithelium, and the activation of autophagy to cause the macrophage polarisation switch has a major role in preventing this signalling pathway cascade amplification. In summary, our results revealed that GEs could regulate the M1/M2 polarisation ratio by activating macrophage autophagy in the inflammatory state, reducing the inflammatory level of intestinal epithelial cells, and targeting the IKK/IкB/NF-кB signalling pathway to alleviate IBD in mice. Supplementary Information [203]Additional file 1^ (11.2MB, docx) Acknowledgements