Cyanidin‐3‐O‐glucoside Alleviates Inflammatory Bowel Disease by Targeting the NF‐κB Pathway and is Associated With Alterations in 1‐DNJ

ABSTRACT

Inflammatory bowel disease (IBD) is characterized by immune dysregulation and an imbalance in the gut microbiota. Cyanidin‐3‐O‐glucoside (C3G), a natural anthocyanin with anti‐inflammatory properties, shows therapeutic potential, though its mechanisms remain incompletely defined. This study evaluated the effects of C3G on intestinal inflammation, microbiota, metabolites, and nuclear factor‐κB (NF‐κB) signaling in a dextran sulfate sodium (DSS)‐induced colitis model in BALB/c mice. In vitro RAW264.7 cells were used to assess pathway involvement. Molecular and histological analyses were performed using qRT‐PCR, Western blotting, and immunostaining, while microbiota and metabolite profiles were analyzed by 16S rDNA sequencing and UHPLC/Q‐TOF‐MS. C3G attenuated colitis severity, improved histological damage, increased interleukin‐10, and reduced IL‐1β, IL‐6, and TNF‐α levels. NF‐κB activation was inhibited, accompanied by enhanced anti‐inflammatory macrophage polarization. C3G also altered gut microbiota composition, reducing pathogenic taxa and enriching beneficial microbes. Metabolomic analysis identified 1‐deoxynojirimycin (1‐DNJ) as associated with C3G treatment. Overall, C3G modulates inflammatory responses, microbiota composition, and NF‐κB signaling in experimental colitis and is associated with changes in 1‐DNJ, warranting further mechanistic investigation.

Cyanidin‐3‐O‐glucoside (C3G) ameliorates DSS‐induced colitis through a gut microbiota–1‐deoxynojirimycin (1‐DNJ)–macrophage axis. Increased 1‐DNJ production is associated with suppression of NF‐κB signaling and inflammatory responses, resulting in improved intestinal integrity and reduced colonic inflammation.

Abbreviations

1‐DNJ

1‐Deoxynojirimycin

5‐ASA

5‐Aminosalicylates

Arg1

Arginase‐1

BALB/c

Bagg albino mouse strain

C3G

Cyanidin‐3‐O‐glucoside

CD

Crohn's disease

DAI

Disease activity index

DSS

Dextran sulfate sodium

H&E

Hematoxylin and eosin

IBD

Inflammatory bowel disease

IKKβ

Inhibitor of nuclear factor κB kinase β

LEfSe

Linear discriminant analysis effect size

NF‐Κb

Nuclear factor κB

OTU

Operational taxonomic unit

PICRUSt

Phylogenetic Investigation of Communities by Reconstruction of Unobserved States

SPF

Specific pathogen‐free

TNF‐α

Tumor necrosis factor‐α;

UC

Ulcerative colitis

UHPLC/Q‐TOF–MS

Ultra‐high–performance liquid chromatography coupled with quadrupole time‐of‐flight mass spectrometry

1. Introduction

Inflammatory bowel disease (IBD) encompasses chronic autoimmune disorders such as Crohn's disease (CD) and ulcerative colitis (UC), which primarily affect the gastrointestinal tract [1]. Although the precise etiology of IBD remains unclear, it is considered a multifactorial disease involving genetic predisposition, gut microbiota imbalance, environmental triggers, and abnormal immune responses [2]. The global incidence of IBD has increased in recent years, expanding from historically high rates in Western countries to emerging regions such as the Eastern Europe, Asia, and South America [3]. This growing burden highlights the urgent need for effective prevention and therapeutic strategies.

Intestinal macrophages play a critical role in maintaining immune homeostasis by distinguishing between commensal and pathogenic microbes. In patients and animal models of IBD, macrophage infiltration in the colonic mucosa increased. Depending on environmental cues, macrophages polarize into two major phenotypes: pro‐inflammatory M1 macrophages, activated by interferon‐gamma (IFN‐γ) and lipopolysaccharide (LPS) to secrete cytokines such as interleukin (IL)‐1β and IL‐6, and anti‐inflammatory M2 macrophages induced by IL‐4 and IL‐13 to express arginase‐1 (Arg1) and IL‐10 [4]. M2 macrophages promote mucosal repair and interact closely with intestinal microbes [5].

Current IBD therapies include 5‐aminosalicylates (5‐ASA) [6], corticosteroids [7], immunomodulators [8], and biologics [9], primarily aimed at suppressing inflammation and controlling symptoms. Nevertheless, these treatments often yield limited efficacy or undesirable side effects, motivating interest in alternative therapeutic strategies. Among natural compounds, cyanidin‐3‐O‐glucoside (C3G), a dietary anthocyanin abundant in colored fruits and vegetables, has attracted attention for its potent anti‐inflammatory, antioxidant, and microbiota‐modulating effects [10]. Previous studies have highlighted the therapeutic potential of C3G and related phytochemicals in IBD management [10, 11, 12, 13]. However, the precise mechanisms through which C3G influences intestinal immunity and gut microbiota composition remain to be fully elucidated.

Disruption of gut microbial homeostasis (dysbiosis) is now recognized as a key factor in IBD pathogenesis [14]. Restoring microbial equilibrium is therefore a promising therapeutic strategy for mitigating IBD symptoms. Despite evidence that C3G modulates gut microbiota, its direct effects on microbial diversity and functional metabolites in the context of IBD remain unclear. To investigate these effects, we employed a well‐established mouse model of IBD (DSS‐induced colitis), which recapitulates key features of human intestinal inflammation, including epithelial damage, immune dysregulation, and microbial imbalance. This model allows mechanistic evaluation of C3G's effects on macrophage polarization and gut microbial composition in vivo.

Importantly, the IκB kinase (IKK)/nuclear factor kappa B (NF‐κB) signaling pathway plays a central role in intestinal inflammation by regulating immune and cytokine responses [15]. Aberrant activation of this pathway in IBD leads to overproduction of pro‐inflammatory cytokines, including tumor necrosis factor‐alpha (TNF‐α), IL‐6, and IL‐1β, resulting in chronic inflammation and mucosal damage [16]. Thus, targeting NF‐κB signaling may offer a promising therapeutic approach.

This study aimed to investigate the protective effects of C3G in DSS‐induced colitis and to elucidate its mechanism by modulating macrophage‐associated gut microbiota via the IKK/NF‐κB signaling pathway. We explored the involvement of the metabolic product 1‐deoxynojirimycin (1‐DNJ) in mediating these effects. Using 16S ribosomal DNA (rDNA) sequencing and ultra‐high–performance liquid chromatography/quadrupole time‐of‐flight mass spectrometry (UHPLC/Q‐TOF–MS) metabolomics, we characterized the alterations in gut microbiota composition and metabolic profiles induced by C3G treatment.

2. Experimental Section

2.1. C3G

Cyanidin‐3‐O‐glucoside (C3G; chemical formula C₂₁H₂₁O₁₁), the predominant anthocyanin found in many edible fruits, was used in this study. Purified C3G (≥95% purity by HPLC, as per certificate of analysis) was purchased from Weikeqi Biotechnology Co., Ltd. (Sichuan, China).

2.2. Animal Model Establishment and C3G Treatment

Male BALB/c mice (6–8 weeks old, 20 g) were purchased from the Animal Research Center of Jiangsu University (Jiangsu, China). A total of 15 mice were randomly assigned into three groups (n = 5 per group): negative control (NC), dextran sulfate sodium (DSS)‐induced colitis (DSS), and C3G treatment group. A dose of 100 mg·kg^{− 1} C3G was selected based on previous studies [17, 18]. NC mice received sterilized water; DSS and C3G groups received 3% DSS in water for 10 days. C3G mice were gavaged daily with 100 mg·kg^{− 1} C3G. Fecal samples were collected for 16S rDNA sequencing and UHPLC/Q‐TOF‐MS analysis. Mice were sacrificed on day 10, and the colon and spleen tissues were collected. Outcome assessments were blinded. All in vivo experiments were performed using biological replicates (n = 5 per group), and measurements were conducted with appropriate technical replicates where applicable. Animals were randomly assigned to groups, and investigators were blinded to group allocation during data collection and analysis, where feasible.

2.3. Cell Line

The macrophage cell line (RAW264.7) was obtained from China Beiner Biotechnology Company.

2.4. Cell Culture

The macrophage cell line (RAW264.7) was cultured in Dulbecco's Modified Eagle Medium (DMEM; SH30243.01, HyClone, USA) supplemented with 10% fetal bovine serum (FBS; ESP500, Excell, Uruguay). Cellular inflammation was induced by treating cells with lipopolysaccharide (LPS) at a final concentration of 1 µg/mL.

2.5. Western Blot

The colon mucosa and RAW264.7 cells were lysed in RIPA buffer with protease inhibitors. Protein samples were then subjected to separation by 12% sodium dodecyl sulfate‐polyacrylamide gel (SDS‐PAGE) electrophoresis. The isolated proteins were transferred to polyvinylidene fluoride (PVDF) membranes and blocked in 5% skim milk powder (dissolved in TBST buffer) for 1 h. PVDF membranes were incubated with primary antibodies (Cell Signaling; p65‐1:800, Cell Signaling; pp65‐1:800, Affinity; p‐IKKβ‐1:800, Starter; β‐actin‐1:1000, Proteintech; IKKβ‐1:1000) at 4°C overnight. After incubation with the primary antibody overnight, the blots were exposed to the secondary antibody for 1 h at 37°C. Finally, image development and analysis were performed.

2.6. Real‐Time Quantitative Reverse Transcription PCR (qRT‐PCR)

Total RNA extracted from colon samples and RAW264.7 cells was analyzed by qRT‐PCR following purification with RNA total cleanup kits according to the manufacturer's protocol. The first‐strand cDNA was synthesized using the HiS‐cript II first Strand cDNA Synthesis Kit (Vazyme Biotech Co., Ltd.), followed by qRT‐PCR amplification with the AceQ Universal SYBR Green qPCR Master Mix (Vazyme Bio‐tech Co., Ltd.). β‐actin was used as an internal control to detect target gene expression. The primer sequences used are shown in Tables 1 and 2.

TABLE 1.

The sequence of primers used for cytokine expression level in qRT‐PCR analysis.

Gene	Primer name	Sequence (5′→3′)	GenBank Accession No.
IL‐1β	mouse‐IL‐1β‐F	CACTAAGGCTCCGAGATGAACAAC	NM_008361.4
	mouse‐IL‐1β‐R	TGTCGTTGCTTGGTTCTCCTTGTAC
IL‐6	mouse‐IL6‐F	CTCCCAACAGACCTGTCTATAC	NM_031168.2
	mouse‐IL6‐R	CCATTGCACAACTCTTTTCTCA
TNF‐α	mouse‐TNF‐α‐F	ATGTCTCAGCCTCTTCTCATTC	NM_013693.3
	mouse‐TNF‐α‐R	GCTTGTCACTCGAATTTTGAGA
IL‐10	mouse‐IL10‐F	TTCTTTCAAACAAAGGACCAGC	NM_010548.2
	mouse‐IL10‐R	GCAACCCAAGTAACCCTTAAAG

TABLE 2.

The sequence of primers related to NF‐Kβ genes used in qRT‐PCR analysis.

Gene	Primer name	Sequence (5′→3′)	GenBank Accession No.
NF‐κB1	mouse‐NF‐κB1‐F	ATGGCAGACGATGATCCCTAC	NM_008689.3
	mouse‐NF‐κB1‐R	CGGAATCGAAATCCCCTCTGTT
NF‐κB2	mouse‐NF‐κB2‐F	AATTCAGCCCCTCCATCGTG	NM_019408.3
	mouse‐NF‐κB2‐R	TCTGAAGCCTCGCTGTTTGG
cRel	mouse‐cRel‐F	ATTTATGACAACCGTGCCCC	NM_009044.3
	mouse‐cRel‐R	CTCTGGCTTCCCAGTCATTCA
RelB	mouse‐RelB‐F	CCTTGGGTTCCAGTGACCTC	NM_009046.3
	mouse‐RelB‐R	ATCTCACTCAGCTGTGTCCC

2.7. Immunohistochemistry (IHC)

Paraffin‐embedded mouse colon and spleen tissue sections (4 µm thick) were processed for hematoxylin and eosin (H&E) staining or deparaffinized for immunohistochemistry (IHC). Sections were blocked with 5% BSA, incubated with primary and secondary antibodies, and developed with DAB.

2.8. Immunofluorescence (IF)

RAW264.7 cells were fixed with 4% paraformaldehyde for at least 30 min, permeabilized with 0.1% Triton X‐100 for 30 min, and blocked with 5% BSA for 30 min at room temperature. Cells were incubated overnight at 4°C with primary antibodies diluted in 5% BSA, followed by incubation with the corresponding fluorescent secondary antibody for 2 h at room temperature. The specimens were counter‐stained with DAPI for 10 min at room temperature, and coverslips were mounted using an anti‐fade mounting medium. Images were captured using a fluorescence microscope.

2.9. Microbiome And Metabolomic Analysis

16S rDNA gene sequencing and untargeted metabolomics were performed on fecal samples to characterize the gut microbial community and metabolome, respectively. For metagenomics, the V3–V4 hypervariable region of the 16S rRNA gene was amplified and sequenced on an Illumina MiSeq platform. Alpha and beta diversity analyses were conducted to assess microbial diversity within and between groups. Linear discriminant analysis effect size (LEfSe) was used to identify differentially abundant taxa (LDA score > 2, p < 0.05), and PICRUSt was employed for KEGG pathway‐based functional prediction of the microbiota. For metabolomics, metabolites were profiled using ultra‐high–performance liquid chromatography coupled to a quadrupole time‐of‐flight mass spectrometry (UHPLC/Q‐TOF‐MS).

2.10. LC–MS/MS Analysis

The analyses were conducted using an ultra‐high–performance liquid chromatography (UHPLC) system (Agilent Technologies, 1290 Infinity LC) connected to a quadrupole time‐of‐flight mass spectrometer (AB Sciex TripleTOF 6600) at Shanghai Applied Protein Technology Co., Ltd.

2.11. Statistics Analysis

All data are presented as mean ± standard deviation (SD). Due to the exploratory nature of this study, no a priori or retrospective power calculation was performed, and the sample size was determined based on commonly used group sizes in comparable preclinical studies. Statistical analyses were performed using GraphPad Prism software. For comparisons involving more than two groups, one‐way analysis of variance (ANOVA) was used to determine differences between groups. When the data did not meet the assumptions of normality, nonparametric tests, including the Kruskal–Wallis test were used. For two‐group comparisons where applicable, Student's t‐test was used. Microbiome diversity analyses, including α‐diversity and β‐diversity, were evaluated using non‐parametric statistical methods as appropriate and are reported accordingly in the figures. For metabolomics data, multivariate statistical analyses were combined with univariate testing, and false discovery rate (FDR) correction using the Benjamini–Hochberg method was applied to reduce the risk of multiple comparisons. A p‐value < 0.05 was considered statistically significant.

3. Results

3.1. C3G Alleviates the Features of Dextran Sulfate Sodium (DSS)‐Induced Colitis in Mice

To evaluate the protective effect of C3G on colitis, a DSS‐induced mouse model was established using a total of 15 mice, randomly assigned into three groups (n = 5 per group): negative control (NC), dextran sulfate sodium (DSS)‐induced colitis, and C3G treatment group. Following colitis induction, mice in the C3G group received 100 mg/kg C3G daily by oral gavage (Figure 1A). Compared with the DSS group, C3G‐treated mice exhibited attenuated body weight loss (Figure 1B) and a reduction in disease activity index (DAI) (Figure 1C). Moreover, C3G administration improved the gross morphology of the colon and spleen (Figure 1D,E). Histopathological examination by hematoxylin and eosin (H&E) staining revealed that C3G preserved intestinal villus structure and reduced epithelial damage compared with the DSS group (Figure 1F,G). At the molecular level, quantitative real‐time polymerase chain reaction (qRT‐PCR) analysis demonstrated that C3G upregulated the anti‐inflammatory cytokine IL‐10 and downregulated pro‐inflammatory cytokines IL‐1β, IL‐6, and TNF‐α (Figure 1H). These findings show that C3G attenuates intestinal inflammation and is associated with improved mucosal integrity, reducing both macroscopic and microscopic features of DSS‐induced colitis in mice.

FIGURE 1.

C3G alleviates the features of dextran sulfate sodium (DSS)‐induced colitis in mice. (A) Schematic diagram of animal model construction. (B) Percentage of weight loss in Mice. (C) Disease Activity Index (DAI) score. (D) Gross appearance of the spleen. (E) Gross appearance of the colorectum. (F) Histopathological examination by hematoxylin and eosin (HE) staining of colon tissue (200×, scale bar =  50 µm). (G) HE staining of spleen tissue (100×, scale bar =  50 µm). (H) qRT‐ PCR analysis of mRNA expression levels of inflammatory cytokines (TNF‐α, IL‐6, IL‐1β, and IL‐10). Data are presented as mean ± SD (n = 5 mice per group). *p < 0.05, **p < 0.01, and ***p < 0.001.

3.2. C3G Suppresses Canonical and Non‐Canonical NF‐κB Gene Expression in Macrophages

To determine whether C3G modulates the NF‐κB signaling pathway, the relative mRNA expression levels of four representative NF‐κB genes, RELB, REL, NF‐κB1, and NF‐κB2, were analyzed in lipopolysaccharide (LPS)‐induced macrophages by qRT‐PCR. The expression of RELB was elevated in the LPS group relative to NC, indicating activation of the non‐canonical NF‐κB pathway. C3G treatment lowered RELB expression toward baseline (Figure 2A). Similarly, REL expression, which was elevated in the LPS group, was notably reduced following C3G treatment (Figure 2B). In the canonical branch, NF‐κB1 expression was upregulated by LPS stimulation but suppressed after C3G treatment (Figure 2C). Likewise, NF‐κB2 expression was increased by LPS and partially reduced by C3G (Figure 2D). Collectively, these data demonstrate that C3G downregulates both canonical and non‐canonical NF‐κB gene expression, suggesting multitargeted inhibition of NF‐κB transcriptional activity.

FIGURE 2.

C3G suppresses canonical and non‐canonical NF‐κB gene expression in macrophages. (A)–(D) Relative mRNA expression levels of RELB, REL, NF‐κB1, and NF‐κB2 measured by qRT‐ PCR. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

3.3. C3G Attenuates Colitis and is Associated With Selective Inhibition of NF‐κB Phosphorylation

To investigate whether C3G modulates colitis through inhibition of the NF‐κB signaling pathway, the expression of pathway‐associated proteins was examined in RAW264.7 macrophages and colonic tissues from mice using western blotting. The analyzed proteins included phosphorylated p65 (p‐p65), total p65 (t‐p65), phosphorylated IKKβ (p‐IKKβ), and total IKKβ (IKKβ). In RAW264.7 cells, p‐p65 levels were increased in the LPS‐treated group relative to NC, reflecting strong activation of the NF‐κB pathway. Treatment with C3G reduced p65 phosphorylation, as evidenced by a decreased p‐p65/t‐p65 ratio, indicating suppression of NF‐κB activation (Figure 3A). In contrast, total p65 (t‐p65) protein levels remained unchanged among the groups, suggesting that C3G primarily regulates the phosphorylation status of p65 rather than altering its overall expression. Treatment with C3G reduced p65 phosphorylation, as reflected by a decreased p‐p65/t‐p65 ratio, demonstrating that C3G suppressed NF‐κB activation (Figure 3A). This pattern indicates selective targeting of phosphorylated signaling proteins. These results were confirmed in mouse tissues by Western blotting. In the DSS group, p‐p65 was increased relative to NC, supporting NF‐κB activation, while C3G reduced p‐p65 without affecting t‐p65 (Figure 3C). DSS exposure increased p‐p65 and p‐IKKβ expression compared with NC, consistent with NF‐κB pathway activation. C3G treatment reduced both p‐p65 and p‐IKKβ to near‐control levels without affecting total protein expression, indicating selective inhibition of phosphorylation events (Figures 3E).

FIGURE 3.

C3G attenuates colitis via selective inhibition of NF‐κB phosphorylation. (A) Western blot analysis of PP65 and Tp65 in RAW264.7 cells among the three groups. (B) Grayscale quantification of PP65 and Tp65, as shown in A. (C) Western blot analysis of PP65 and tP65 in mice colon tissues. (D) Grayscale quantification of PP65, tP65, as shown in C. (E) Western blot analysis of P‐IKKB, and IKKB in mice colon tissues. (F) Grayscale quantification of P‐IKKB, IKKB, as shown in E. (G) Immunofluorescence (IF) staining of the DAPI, FITC, and MERGE in LPS‐induced RAW264.7 cells detected by IF (600×, scale bar =  50 µm). (H) Nuclear‐to‐cytoplasmic ratio of NF‐κB p65 fluorescence intensity quantified in LPS‐induced RAW264.7 cells, corresponding to the immunofluorescence staining in G. Western blot quantification was performed using five independent biological replicates per experimental group (n = 5 per group). Representative blot images from the three experimental groups are shown. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

To further verify macrophage activation and subcellular localization, immunofluorescence staining (IF) was performed using DAPI, FITC, and merged channels (Figure 3G). C3G treatment decreased FITC fluorescence and altered its nuclear localization, suggesting that C3G mitigates LPS‐induced activation and nuclear translocation. Quantitative analysis of the nuclear‐to‐cytoplasmic ratio of NF‐κB p65 fluorescence intensity confirmed that LPS stimulation significantly increased p65 nuclear entry, whereas C3G treatment reduced this ratio, further supporting the inhibitory effect of C3G on NF‐κB nuclear translocation (Figure 3H).

3.4. C3G and IKK‐16 Exhibit Enhanced Inhibitory Effects on Inflammation by Suppressing the NF‐κB Pathway

To evaluate the anti‐inflammatory effects of C3G and the IKK‐16 inhibitor, the relative mRNA expression levels of key pro‐inflammatory cytokines (IL‐1β, IL‐6, and TNF‐α) were analyzed using qRT‐PCR. LPS stimulation elevated IL‐1β expression relative to NC, confirming inflammatory activation. C3G treatment lowered IL‐1β levels, while co‐treatment with IKK‐16 and C3G produced the greatest reduction, indicating an enhanced inhibitory effect (Figure 4A). Similar trends were observed for IL‐6 and TNF‐α, where the combination produced stronger suppression than either treatment alone (Figures 4B,C). To further elucidate the mechanism, we examined the expression of NF‐κB signaling genes (RELB, REL, NF‐κB1, and NF‐κB2). LPS upregulated these genes relative to the NC group, whereas both C3G and IKK‐16 treatments reduced their expression. Notably, the combined C3G + IKK‐16 treatment produced the most pronounced suppression, significantly lower than in the LPS, C3G, or IKK‐16 + LPS groups (Figure 4D–G), suggesting enhanced inhibition of NF‐κB pathway activation. Consistently, Western blot analysis revealed that LPS elevated p‐p65, indicative of NF‐κB activation. C3G or IKK‐16 alone reduced p‐p65/t‐p65 ratios, whereas combined treatment further reduced phosphorylation levels and enhanced inhibition of NF‐κB pathway activation (Figure 4H,I).

FIGURE 4.

C3G and IKK‐16 synergistically inhibit inflammation by suppressing the NF‐κB pathway. (A‐C) The relative mRNA expression levels of IL‐1β, IL‐6, and TNF‐α in LPS‐induced cells across different treatment groups were measured by qRT‐ PCR. (D‐G) The relative expression level of NF‐κB pathway‐related genes (RELB, REL, NF‐κB1, and NF‐κB2) was measured by qRT‐ PCR. (H) Western blot analysis of p‐p65 and t‐P65 in LPS‐induced RAW264.7 cells. (I) Grayscale quantification of p‐P65/t‐P65, as shown in (H). Western blot quantification was performed using five independent biological replicates per experimental group (n = 5). Representative blot images are shown. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, and ***p < 0.001.

3.5. C3G Improves the General Community Structure and Diversity of Gut Microbiota

To assess the effects of C3G on gut microbial composition and diversity, 16S rDNA gene sequencing was performed on fecal samples from mice. The Venn diagram analysis of operational taxonomic units (OTUs) showed that the DSS group exhibited an increase in unique OTUs (490) compared with the normal control (NC) group (249). This DSS‐induced alteration was attenuated by C3G treatment, which reduced the number of unique OTUs to 226. Moreover, the C3G group shared more OTUs with the NC group (86) than with the DSS group (63), indicating that C3G shifted the microbial community structure toward a profile resembling the NC group (Figure 5A). At the phylum level, the gut microbiota was predominantly composed of Firmicutes, Bacteroidetes, Proteobacteria, and Actinobacteria, accounting for approximately 99% of total bacteria, consistent with previous studies [19, 20]. DSS exposure decreased the relative abundance of Bacteroidetes and Proteobacteria while increasing Firmicutes, whereas C3G reversed these trends (Figure 5B). Furthermore, C3G restored the relative abundance of additional phyla, including Actinobacteria, Verrucomicrobiota, Bacteroidota, Firmicutes, Proteobacteria, and Cyanobacteria (Figure 5C). At the family level, cluster heatmap analysis demonstrated that the microbial community structure in the C3G group was more similar to that of the NC group than to that of the DSS group (Figure 5D). Alpha‐diversity analysis revealed a reduction in the DSS group, whereas no difference was observed between the C3G and NC groups (Figure 5E). Likewise, beta‐diversity analysis revealed a substantial difference between the DSS and NC groups, but not between the C3G and NC groups (Figure 5F). Collectively, these findings suggest that C3G ameliorates DSS‐induced dysbiosis and contributes to the restoration of microbial community structure, while maintaining microbial alpha and beta diversity.

FIGURE 5.

C3G improves gut microbial general community structure and diversity. (A) Venn diagram showing shared and unique OTUs among the DSS, C3G, and NC groups. (B) Relative abundance of the top 10 phyla across groups. (C) Cluster heatmap of species abundance at the phylum level. (D) Cluster heatmap of species abundance at the family level. (E) Goods coverage boxplot representing α‐diversity among groups. (F) Boxplot of β‐diversity distances based on unweighted UniFrac metrics. Microbiome analyses were performed using n = 5 mice per group.

3.6. C3G Modulates Specific Bacterial Species and Restores Functional Dysregulation

Differential analysis of significantly abundant bacterial taxa revealed distinct microbial profiles among the groups. C3G administration attenuated DSS‐induced bacterial overgrowth and modulated the relative abundances, normalizing the relative abundances of Empedobacter, Enterococcus, Lachnospiraceae_UCG, and Rhizobiaceae (Figure 6A–E). In addition, C3G reduced the abundance of other overrepresented bacterial families in the DSS group, including Prevotellaceae, Erysipelotrichaceae, Enterococcaceae, and Clostridiaceae (Figure 6A). At the genus level, the DSS group showed elevated levels of potentially pathogenic taxa, including Stenotrophomonas, Erysipelatoclostridium, and Rikenella (Figure 6F). Notably, the C3G‐treated group exhibited microbial profiles comparable to the NC group (Figure 6H). Previous studies have demonstrated that Enterococcus faecium strains isolated from UC patients can induce colitis in mice, whereas strains from healthy individuals do not [21]. Conversely, Lachnospiraceae abundance has been reported with conflicting roles, being linked to both precancerous lesions [22] and reduced UC symptoms [23]. Furthermore, C3G treatment increased the abundance of beneficial genera, including Turtlebacter, Roseburia, Candidatus_Arthromitus, [Eubacterium]_brachy, and Monoglobus (Figure 6H). Thus, C3G appears to exert a dual modulatory effect, suppressing overgrown pathogenic taxa while enriching beneficial commensals.

FIGURE 6.

C3G modulates specific bacterial species and restores functional dysregulation. (A) Cladogram (evolutionary branch diagram) of statistically different microbiota within each group. (B–E) Relative abundances of representative bacterial taxa (Empedobacter, Enterococcus, Lachnospiraceae_UCG, and Rhizobiaceae) showing significant differences among NC, DSS, and C3G groups. (F) STAMP t‐test of genus‐level differences between DSS and NC groups. (G) KEGG heatmap prediction of functional differences between C3G and DSS groups. (H) STAMP t‐test of genus‐level differences between C3G and NC groups. (I) KEGG LDA score of predicted functional pathways between DSS and NC groups. Metabolomic analyses were performed using n = 5 mice per group.

Functional prediction using KEGG analysis revealed that DSS administration significantly altered six major pathways, including downregulation of carbohydrate and amino acid metabolism and upregulation of xenobiotic biodegradation, bacterial infectious disease pathways, and the metabolism of terpenoids and polyketides (Figure 6I). Notably, no functional differences were detected between the C3G and NC groups, suggesting that C3G may modulate gut microbial function toward a profile similar to that of the normal NC group (Figure 6G,I).

3.7. C3G Reduces Dysregulation of Gut Metabolites in the Mitigation of Colitis

The activity of gut microbiome‐derived metabolites is closely associated with the pathogenesis of IBD [24, 25, 26]. Consistent with previous studies [27, 28], we observed significant alterations in gut metabolites in DSS‐induced colitic mice compared with the normal NC group. The UHPLC‐Q‐TOF‐MS‐based untargeted metabolomics analysis was performed following standard data processing and validation procedures. Differential analysis identified 252 significantly altered metabolites between the DSS and NC groups, including 94 in negative ion mode and 158 in positive ion mode. DSS treatment resulted in substantial metabolite dysregulation, whereas C3G supplementation attenuated these alterations (Figure 7A,B). In positive‐ion mode, the C3G group showed fewer differentially abundant metabolites compared with untreated DSS mice, suggesting a partial restoration of metabolic balance. Principal component analysis (PCA) revealed clear separation between groups. The DSS group was distinct from the NC mice, while the C3G group occupied an intermediate position between them (Figure 7C,D). Volcano plot analysis further showed fewer dysregulated metabolites in the C3G group than in the DSS group relative to controls (Figure 7E,F). Differential metabolites grouped by superclass are summarized in Figure 7G,H, illustrating the major metabolic categories influenced by DSS and modulated by C3G.

FIGURE 7.

C3G reduces dysregulation of gut metabolites in the mitigation of colitis. (A) Venn diagram of differential metabolites within different groups in negative ion mode. (B) Venn diagram in positive ion mode. (C) PCA score map in negative ion mode. (D) PCA score plot in positive ion mode. (E) Volcano plot of differential metabolites between the DSS and NC groups in positive ion mode. (F) Volcano plot between C3G and NC groups in positive ion mode. (G) Multiple analyses of significant differences in metabolite expression between the DSS and NC groups. (H) Multiple analyses were conducted between the C3G and NC groups. Metabolomic analyses were performed using n = 5 mice per group.

3.8. C3G Modulates Key Metabolic Pathways and Gut Microbiota in Association With 1‐DNJ in DSS‐Induced Colitis

To provide a systematic context for the 252 identified differential metabolites, KEGG pathway enrichment analysis was performed. KEGG pathway enrichment analysis identified eight significantly enriched pathways (Figure 8A). Pyrimidine and nucleotide metabolism were the most affected, followed by non‐alcoholic fatty liver disease, ABC transporters, the insulin signaling pathway, the pentose phosphate pathway, the FoxO signaling pathway, and type II diabetes mellitus. Differential abundance score analysis revealed that the pentose phosphate pathway, insulin signaling, FoxO signaling, and type II diabetes pathways were positively associated with C3G treatment (scores: 0.5–1.0), suggesting partial normalization of metabolic alterations (Figure 8B). Based on this pathway‐level analysis, 1‐DNJ was identified as a prominently altered metabolite within the significantly enriched nucleotide metabolism pathway and was therefore selected for further correlation analysis with gut microbiota. The differential metabolite analysis revealed significant changes in metabolic profiles following C3G treatment (Figure 8C). Among the upregulated metabolites, 1‐DNJ levels were higher in the C3G‐treated group than in DSS mice, suggesting an association with the metabolic response to C3G. This observation suggests that 1‐DNJ may be associated with the metabolic response to C3G. Other metabolites also exhibited notable changes, indicating that C3G is associated with broader metabolic modulation. The metabolite interaction network further demonstrated that 1‐DNJ occupies a central position among differential metabolites, particularly those involved in nucleotide, amino acid, and lipid metabolism (Figure 8D). These interactions reflect metabolic differences between the C3G‐treated and DSS groups. Spearman correlation analysis combined with hierarchical clustering revealed significant associations between differential microbiota and metabolites (Figure 8E). The results showed that 1‐DNJ levels were positively correlated with beneficial bacterial genera such as Alloprevotella and Lachnoclostridium, while negatively correlated with pro‐inflammatory genera, including Corynebacterium and Pseudomonas. Moreover, the scatter plot indicated a strong negative correlation between 1‐DNJ and UCG_010 levels (Figure 8F). The C3G‐treated group exhibited elevated 1‐DNJ concentrations and reduced UCG_010 abundance, whereas the DSS group showed the opposite trend. These correlational findings suggest that 1‐DNJ is associated with changes in gut microbial composition, though further experimental validation (e.g., supplementation or inhibition studies) is needed to determine causality.

FIGURE 8.

C3G is associated with gut microbiota and metabolite alterations during the alleviation of IBD. (A) KEGG enrichment of 252 differential metabolites identified eight key pathways, with pyrimidine and nucleotide metabolism most affected. (B) Differential abundance analysis showed that the pentose phosphate, insulin, FoxO, and type II diabetes pathways were positively associated with C3G treatment, suggesting partial metabolic normalization. (C) Differential metabolite analysis showing significant metabolic profile changes following C3G treatment. (D) Network diagram illustrating interactions between metabolites in C3G‐treated and DSS groups. (E) Spearman correlation heatmap showing associations between differential microbiota and metabolites. (F) Scatter plot showing a negative correlation between 1‐DNJ and UCG_010. Analyses were performed using n = 5 mice per group.

4. Discussion

IBD is characterized by persistent intestinal inflammation caused by immune dysregulation and microbial imbalance. In agreement with prior studies, the present work demonstrates that C3G, a natural anthocyanin with potent anti‐inflammatory and antioxidant properties, significantly ameliorates DSS‐induced colitis in mice. This protective effect was evident through suppression of pro‐inflammatory cytokines (IL‐1β, IL‐6, TNF‐α), improved DAI scores, enhanced IL‐10 expression, and repair of colon and spleen tissue damage. Previous preclinical studies have similarly reported that C3G and other anthocyanins exert potent anti‐inflammatory effects by modulating cytokine expression and improving DAI in animal models [17, 29]. These findings collectively highlight C3G's immunomodulatory capacity, characterized by reduced TNF‐α, IFN‐γ, IL‐1β, and IL‐6 expression and upregulated IL‐10, while our study further elucidates the molecular pathways underlying this response. At the molecular level, C3G exerted a broad anti‐inflammatory effect by suppressing both canonical and non‐canonical NF‐κB signaling. Specifically, C3G downregulated LPS‐induced expression of REL, NF‐κB1, RELB, and NF‐κB2, indicating multitargeted inhibition of this central inflammatory axis. Moreover, C3G suppressed the phosphorylation of IKKβ and p65, thereby inhibiting NF‐κB activation without altering total protein expression. These findings align with earlier studies showing that anthocyanins inhibit NF‐κB by blocking IκBα phosphorylation and preventing nuclear translocation in macrophages, endothelial cells, and lung tissues [30, 31, 32, 33]. Such multi‐level inhibition positions C3G as a potential therapeutic candidate for NF‐κB–driven inflammatory diseases, including IBD. However, the translational relevance of the 100 mg/kg/day C3G dose used in this study warrants careful consideration. Using allometric scaling based on body surface area (Km ratio mouse/human = 0.081) as described by Jacob et al. [34], the 100 mg/kg/day dose in mice corresponds to a human equivalent dose of approximately 8.1 mg/kg/day (486 mg/day for a 60 kg adult). While this dose is achievable through anthocyanin‐rich supplements, it is substantially higher than the average dietary intake of anthocyanins (12–40 mg/day) [35]. Importantly, the systemic bioavailability of oral C3G is very low. Pharmacokinetic studies in mice have shown that after oral administration, the systemic bioavailability of parent C3G is only approximately 1.7%, with total anthocyanin bioavailability reaching 3.3% [36, 37]. The half‐life of C3G in mouse plasma ranges from 0.7 h to 1.8 h, with peak concentrations occurring within 30 min of administration [36]. Similar low bioavailability (0.5%–1.5%) has been reported in rats [38]. In healthy human volunteers, oral administration of black bean seed coat extract containing C3G resulted in a plasma half‐life of approximately 1.5 h, with no drug accumulation after 2 weeks of daily dosing [39]. These bioavailability data indicate that the concentrations of intact C3G reaching systemic circulation are very low, raising the question of whether the observed anti‐inflammatory effects are mediated by the parent compound, its metabolites, or both. Indeed, recent evidence suggests that the gut microbiota plays a critical role in metabolizing C3G into phenolic compounds such as protocatechuic acid and cyanidin, which may be the primary bioactive species responsible for the pharmacological effects [40].

Furthermore, intragastric administration of C3G has been shown to be more effective than intravenous administration in reducing inflammation and oxidative stress, despite achieving lower plasma concentrations of intact C3G, highlighting the importance of local gut effects and microbial metabolism [40]. Thus, the 100 mg/kg dose used here is consistent with the range (50–800 mg/kg) employed in prior preclinical studies [17, 18, 29], and its efficacy likely depends on gut microbiota‐mediated metabolism rather than systemic bioavailability of the parent compound. Nevertheless, direct extrapolation of this dose to humans must account for interspecies differences in gut microbiota composition, metabolism, and anthocyanin absorption kinetics.

However, it is important to note that the DSS‐induced colitis model predominantly reflects acute epithelial barrier damage accompanied by activation of the innate immune system, rather than the chronic, relapsing, and T cell‐mediated pathology that defines human IBD [41, 42]. As NF‐κB signaling has been more extensively characterized in chronic inflammatory and autoimmune settings, the translational applicability of the NF‐κB‐related findings in this study to human IBD, particularly with respect to sustained immune dysregulation, may be limited [43, 44]. Consequently, the DSS model represents only a portion of the complex immunological and tissue remodeling processes observed in patients, and the inhibitory effects on NF‐κB reported here should be interpreted primarily within the context of acute inflammatory injury.

A key observation from this study was the enhanced anti‐inflammatory effect of the combined treatment with C3G and the IKK‐16 inhibitor. The combined treatment resulted in greater suppression of pro‐inflammatory cytokines (IL‐1β, IL‐6, TNF‐α) and enhanced inhibition of NF‐κB signaling, as evidenced by both mRNA and protein phosphorylation levels. These findings suggest that C3G may potentiate the efficacy of pharmacological NF‐κB inhibitors through complementary mechanisms, offering a more comprehensive therapeutic strategy for managing IBD. However, it should be noted that formal synergy was not evaluated using established methods such as the Chou–Talalay combination index; therefore, the observed effects should be interpreted as additive or enhanced rather than strictly synergistic. Similar reports indicate that natural polyphenols can enhance the activity of IKK inhibitors by targeting multiple molecular pathways [45].

The gut microbiota plays a critical role in host immune regulation, metabolism, and disease progression. Dysbiosis, or alterations in the gut microbial community structure, is a hallmark of IBD pathogenesis. Using 16S rDNA sequencing, we observed that C3G treatment modulated colitis‐associated shifts in microbial diversity and composition. C3G improved both α‐ and β‐diversity and corrected the imbalance in the three dominant phyla (Bacteroidetes, Proteobacteria, and Firmicutes), thereby shifting the Bacteroidetes/Firmicutes ratio toward that of controls [19, 20]. Additionally, C3G reduced the relative abundance of potentially pathogenic genera such as Stenotrophomonas, Erysipelatoclostridium, and Rikenella, while promoting beneficial bacterial groups, including Turtlebacter, Roseburia, Candidatus Arthromitus, Eubacterium brachy, and Monoglobus. KEGG functional prediction analysis revealed that six microbiota‐associated pathways, including carbohydrate metabolism, amino acid metabolism, and xenobiotic degradation, were significantly dysregulated in colitis but were partially normalized following C3G treatment. Notably, the metabolic functions of the microbiota in the C3G‐treated group closely resembled those of healthy controls, suggesting that C3G may help modulate microbial balance and function. These results suggest that C3G may act in a prebiotic‐like manner, promoting microbial and metabolic homeostasis. Moreover, it should be noted that KEGG pathway enrichment derived from untargeted metabolomics provides primarily associative rather than definitive mechanistic evidence, and pathway interpretation may be influenced by metabolite annotation confidence and database coverage. Therefore, further multi‐omics integration, coupled with targeted validation, is required to confirm these pathway‐level alterations and strengthen mechanistic inference. In this study, microbial functional profiles were inferred using PICRUSt based on 16S rRNA gene sequencing data. While this approach provides useful insights into potential functional shifts, it relies on reference genomes and may be less accurate for communities containing uncharacterized or novel taxa. Therefore, the predicted functional pathways should be interpreted as indicative rather than definitive. Notably, the consistency between PICRUSt‐based predictions and metabolomics findings in key pathways, including amino acid, carbohydrate, and nucleotide metabolism, provides supportive evidence for the observed functional trends. However, future studies using shotgun metagenomics or meta‐transcriptomics are warranted to validate these results and achieve a more comprehensive functional characterization.

Microbial metabolites link gut microbiota composition with host immune regulation. Our metabolomic analysis (UHPLC/Q‐TOF‐MS) revealed that C3G treatment altered fecal metabolic profiles, with a notable increase in 1‐DNJ, a metabolite known to inhibit α‐glucosidase and modulate macrophage activity. These findings indicate that 1‐DNJ is associated with C3G treatment and may be involved in the observed anti‐inflammatory effects. Correlation analysis further demonstrated that 1‐DNJ levels were positively associated with beneficial genera, including Alloprevotella and Lachnoclostridium, which have been implicated in anti‐inflammatory processes. These associations suggest a potential link between microbial composition, metabolite production, and host immune responses. Additionally, interactions between 1‐DNJ and microbial taxa such as UCG_010 further support the involvement of gut microbiota–metabolite–immune crosstalk. However, it is important to note that the present study does not establish a causal role for 1‐DNJ in mediating the therapeutic effects of C3G. The identification of 1‐DNJ is based on correlative metabolomics analysis, and no direct functional validation (e.g., supplementation or inhibition studies) was performed. Therefore, 1‐DNJ should be considered a candidate metabolite associated with C3G treatment. Future studies are required to determine whether 1‐DNJ plays a mechanistic role in C3G‐mediated protection against colitis. Moreover, previous studies have shown that intestinal bacteria metabolize dietary C3G into phenolic compounds such as protocatechuic acid, which contribute to its bioactivity [46]. Therefore, C3G's efficacy likely depends on host–microbe co‐metabolism, with microbial metabolites amplifying its anti‐inflammatory effects. Several limitations should be noted. First, although all in vivo and quantitative analyses were performed using n = 5 biological replicates per group, the overall sample size remains relatively small, and no a priori or retrospective power analysis was conducted. This limited sample size may reduce statistical power and increase the risk of type II error, particularly for analyses with high inter‐individual variability, such as microbiome and metabolomics profiling. As a result, the robustness of effect size estimation and the generalizability of these findings may be constrained, and the results should therefore be interpreted with appropriate caution. Second, this study was conducted in a mouse model, and interspecies differences in C3G metabolism and gut microbiota composition may limit direct clinical translation. As discussed above, the low and variable oral bioavailability of C3G in both rodents and humans (approximately 0.5%–3.3%) further complicates dose extrapolation. The absence of pharmacokinetic data in our study (e.g., plasma C3G or metabolite levels) precludes a definitive understanding of the relationship between administered dose, tissue exposure, and therapeutic effect [36, 38, 39].

The DSS‐induced colitis model primarily reflects acute intestinal injury and epithelial disruption rather than the chronic, immune‐driven pathology of human IBD; consequently, the NF‐κB findings, which are more established in chronic relapsing disease, may not fully translate to the human setting [41, 44]. Moreover, because DSS colitis is largely driven by epithelial barrier damage rather than adaptive immune dysregulation, the model does not recapitulate key features of human IBD, such as T‐cell‐mediated relapses or granuloma formation [42]. This distinction is particularly relevant when interpreting the anti‐inflammatory effects of C3G on NF‐κB, as immune‐cell‐derived NF‐κB activation in chronic disease may respond differently than epithelial‐derived signals in acute injury. Additionally, potential sources of bias include cage and batch effects that may influence microbiota composition. Although investigators were blinded during histological and molecular analyses, residual confounding cannot be entirely excluded. Furthermore, while 1‐DNJ was identified as a key metabolite associated with therapeutic effects, causal validation through targeted intervention (e.g., supplementation or inhibition studies) was not performed. Despite these limitations, our findings demonstrate that C3G modulates NF‐κB signaling, cytokine expression, and gut microbiota composition in vivo. However, given the limitations in statistical power and model system, validation in larger, adequately powered studies and humanized or clinical settings will be essential to confirm the reproducibility, mechanistic causality, and translational potential of these results. Future studies should use germ‐free or humanized microbiota models to confirm causal roles of microbial metabolites, as well as chronic or relapsing colitis models (e.g., IL‐10⁻/⁻ mice or T‐cell transfer models) to more closely reflect how IBD occurs in humans and assess the durability of C3G's effects on NF‐κB signaling. Additionally, future pharmacokinetic studies in IBD patients are needed to determine the optimal dosing regimen, assess bioavailability in the context of intestinal inflammation (which may alter absorption), and evaluate the contribution of microbial metabolites to therapeutic efficacy.

Clinical trials evaluating the bioavailability, safety, and therapeutic efficacy of C3G in IBD patients are also necessary. In summary, C3G attenuates DSS‐induced colitis by modulating NF‐κB signaling, inflammatory responses, and gut microbiota composition. Additionally, C3G treatment is associated with alterations in microbial metabolites, including 1‐DNJ, which may contribute to its therapeutic effects. Further studies are required to elucidate the mechanistic role of 1‐DNJ. These findings provide new insights into the molecular and microbial pathways underlying C3G's therapeutic effects and support its potential development as a natural, microbial‐targeted therapy for IBD. 

Author Contributions

Z.H.Z., X.J.Z, and P.M. did conception and design, collection and/or assembly of data. H.J. performed data analysis and interpretation. H.K. did software. F.M. performed the visualization. Z.H.Z., X.J.Z., and P.M. did the manuscript writing – original draft. S.H.W., Y.Z., and D.M.Y did the writing – review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

The research was funded by the Special Fund Project for Basic Research of Zhenjiang (No: JC2024035), Innovation Special Fund of Danyang (No: SSF202410), and Zhenjiang Science and Technology Plan (No: SH2024047).

Ethics Statement

The ethics of animal research were approved by the Ethical Committee of Jiangsu University [Approval number: 2018‐0053] (Approval date: 2020.3.31).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: mnfr70539‐sup‐0001‐DataFile.zip.

Data Availability Statement

The 16S rRNA sequencing data generated in this study are publicly available in the NCBI Sequence Read Archive (SRA) under BioProject accession number PRJNA1471306. The BioProject record and associated SRA metadata are available at: https://dataview.ncbi.nlm.nih.gov/object/PRJNA1471306?reviewer=ul99ft104g9qjphbmnhfr121ot.