Ursodeoxycholic acid alleviates α-Casein-induced cow’s milk protein allergy via the TGR5/NF-κB signaling pathway

Abstract

Cow’s milk protein allergy (CMPA), one of the most common food allergies in infant, is primarily manifesting gastrointestinal symptoms and closely associated to the disorders of gut microbiota. Ursodeoxycholic acid (UDCA), a gut microbiota derived secondary bile acid, is significantly decreased in infant with CMPA. However, the effect and mechanism of UDCA in colitis of CMPA is unknown. Our analysis demonstrated that UDCA administration alleviated the systemic allergic symptoms, improved body weight gain, and mitigated histopathological damage in both liver and colon tissues in a mouse model of α-Casein-induced CMPA. UDCA restored intestinal barrier integrity by increasing goblet cell numbers and enhancing the expression of the tight junction gene in colonic tissue. In α-Casein-sensitized RAW264.7 cells, UDCA promoted the healing and migration. Additionally, UDCA significantly down-regulated the secretion and mRNA levels of key pro-inflammatory factors (IL-1β, TNF-α, IL-6) and the production of inflammatory mediators (NO, ROS). Mechanistically, UDCA increased the cAMP level and up-regulated G-protein coupled receptors (TGR5) expression and inhibited the NF-κB signaling pathway, as evidenced by reduced phosphorylation and nuclear translocation of p65 in vivo and in vitro. Pharmacological antagonism with SBI-115 confirmed that UDCA suppressed α-Casein-induced NF-κB p65 activation primarily by targeting TGR5. This study suggested that UDCA may alleviate CMPA by targeting TGR5 to enhance intestinal barrier function and modulate immune response via NF-κB signaling, highlighting its potential as a therapeutic strategy for allergic diseases.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-38293-1.

Introduction

Cow’s milk protein allergy (CMPA) is a common food allergy (FA) in infants and young children. In recent years, the prevalence of CMPA has been increasing, making it a significant global health concern¹. Infants and young children with CMPA primarily presenting with gastrointestinal symptoms often exhibit manifestations such as vomiting, diarrhea, mucous stools, and bloody stools². The most common type of CMPA is food protein-induced allergic proctocolitis (FPIAP), which is characterized by inflammatory cell infiltration triggered by food proteins that disrupt the intestinal barrier, leading to inflammation in the distal colon. This condition severely impacts the growth and development of infants and young children, impairs family quality of life, and increases the risk of other serious allergic diseases later in childhood³. The intestinal immune response in CMPA is largely attributed to cellular immunity, which participates in regulating the CMPA phenotype expression through the secretion of various cytokines, thereby affecting gastrointestinal mucosal function. Food antigens cross the intestinal epithelium via active and passive transport mechanisms to reach mucosal antigen-presenting cells⁴. Gastrointestinal symptoms in CMPA may results from interplay of multiple factors, including digestive tract inflammation and dysbiosis⁵. Early dysbiosis of the intestinal flora can increase the host’s sensitivity to allergies and inflammatory responses. Emerging evidence suggests that early gut dysbiosis is associated with increasing prevalence of allergies in children⁶.

Gut microbiota metabolites influence susceptibility to FA by interacting with pattern recognition receptors on immune cells⁷. Intestinal macrophages help alleviate inflammation, initiate epithelial damage repair, and maintain intestinal homeostasis by clearing apoptotic cellular debris and secreting growth factors and mediators, thereby providing nutritional support to the surrounding tissue environment⁸. Studies have shown that metabolites generated by the gut microbiota can modulate the function of intestinal macrophages and maintain intestinal immune homeostasis⁹. Metabolites can suppress inflammatory response to allergens or reduce intestinal permeability to inhibit allergen passage through the intestinal barrier¹⁰. According to the production pathways, gut microbial metabolites can be categorized into as follows: those derived from the breakdown of dietary components (e.g., short-chain fatty acids, SCFAs), those synthesized by the host and subsequently modified by microbes (e.g., secondary bile acids, SBAs), and those synthesized de novo by microbes (e.g., vitamin K)¹¹. SBAs, produced through microbial modification of primary bile acids (PBAs), are recognized as key regulators of various metabolic processes¹². Major SBAs include deoxycholic acid (DCA), lithocholic acid (LCA), and ursodeoxycholic acid (UDCA), which are generated by specific microbes and play significant roles in disease processes. Studies report that SBAs produced by symbiotic microbiota inhibit the differentiation of RORγt+ Treg cells and the secretion of IL-17 by Th17 cells in mouse colon tissues¹³.

Gut microbiota dysbiosis alters the composition of bile acids (including primary, secondary, conjugated, and deconjugated bile acids), which further participate in immune responses and affect the function of intestinal epithelial cells¹⁴. Lactobacillus rhamnosus GG has been shown to protect intestinal immunity and promote oral tolerance in CMPA infants¹⁵. Bile acids derived from the microbial modification dose-dependently promote M1 macrophage polarization and contribute to high-fat diet-induced colon inflammation¹⁶. Furthermore, Bile acids have been reported to modulate antigen exposure, thereby influencing immune tolerance¹⁷. SBAs regulate innate and adaptive immunity through specific receptors such as the nuclear receptor (farnesoid X receptor, FXR) and the membrane receptor (G-protein coupled receptors, TGR5/GPBAR1)¹⁸. Binding of SBAs to TGR5 up-regulates cyclic adenosine monophosphate (cAMP) expression, which in turn inhibits the secretion pro-inflammatory cytokines in LPS-stimulated macrophages, thereby maintaining mucosal barrier integrity and intestinal homeostasis^19,20. Our previous study found that children with CMPA had significantly lower levels of SBAs, particularly UDCA, which may be associated intestinal inflammation²¹. However, the effects of UDCA in the progression of CMPA is still not fully understood, and further research is needed to elucidate potential mechanisms.

Casein, which constitutes about 80% of the total protein content in milk, is the primary allergen responsible for allergic reactions. Additionally, casein sensitivity has been reported in 61% of children with milk allergy, and are more likely to have severe reactions²². Studies have shown that casein can regulate macrophage function and participate in immune response. Macrophages may be involved in the immune response to dietary antigens by contributing to either oral tolerance or allergic sensitization²³. α-Casein induce the production of pro-inflammatory cytokines such as TNF-α, IL-10, and up-regulate mRNA levels of IL-1β, IL-6, TNF-α, IL-10 in macrophages, indicating a successful cellular model^24,25. Additionally, α-Casein has been used to establish a CMPA model to investigate allergic immune responses^26,27. In this study, we therefore investigated the protective role and underlying potential mechanism of UDCA in CMPA in α-Casein-induced CMPA mouse and cell models. Our results demonstrated that UDCA alleviated the allergic symptoms by suppressing the pro-inflammatory response via the NF-κB signaling pathway targeting TGR5 in vivo and in vitro. These findings provide new insights into the role of UDCA in the immune system of allergic diseases.

Materials and methods

Reagents

UDCA and α-Casein (≥ 70%) were purchased from Sigma-Aldrich (St. Louis, MO). Cell culture reagents, including Dulbecco’s modified Eagle’s medium (DMEM), penicillin/streptomycin, RIPA buffer, and protease and phosphatase inhibitors, were obtained from New Cell & Molecular Biotech Co., Ltd (NCM) (Zhengzhou, China). Fetal bovine serum (FBS) was sourced from SORFA (Zhejiang, China). Cell Counting Kit-8 (CCK-8) and Cholera toxin were purchased from GlpBio (Shanghai, China). The Reactive Oxygen Species Assay Kit and Nitric Oxide Assay Kit were purchased from Beyotime (Shanghai, China). cAMP, TNF-α, IL-6, IL-1β and IL-10 enzyme-linked immunosorbent assay (ELISA) kits were purchased from Elabscience (Wuhan, China). Primary antibodies against F4/80, p65 and p-p65 were purchased from Cell Signaling Technology (Danvers, MA). β-actin and Lamin B were from Abways (Shanghai, China), TGR5 was from Immunoway (Beijing, China), and ERK1/2, p-ERK1/2, Akt, p-Akt were from Proteintech (Beijing, China). Anti-rabbit secondary antibodies were purchased from Abways (Shanghai, China).

Mouse model establishment and experimental design

Female BALB/c mice (3 weeks, 15–20 g, n = 32) were obtained from Laboratory Animal Center of Zhengzhou University (Zhengzhou, China). Mice were housed in a standard conditions, with a temperature of 22 ± 2 °C, a humidity of (55.0 ± 5.0)%, and a 12-hour light/12 h dark cycle. The animals were allowed free access to food and trap water and acclimated for one week prior to the experiment. All animal experiments were performed in accordance with the guide the regulations of Chinese Laboratory Animal Management and ARRIVE guidelines 2.0²⁸. All experiments followed a protocol approved by the by the Ethics Committee of Zhengzhou University (2023-KS-ZD01).

The murine model of CMPA was established according to the previously described method with appropriate modification^26,27. All mice were randomly divided into four groups by a researcher not involved in the subsequent study, including control group (C), α-Casein-induced CMPA model group (A), α-Casein + 30 mg/kg UDCA group (UL), and α-Casein + 60 mg/kg UDCA group (UH). Each group contained 8 mice which marked with a unique, non-sequential code that concealed its group identity from the investigators. Mice in the A, UL, and UH groups were sensitized intragastrically (i.g.) with 0.2 mL of α-Casein (2 mg in 1 ml PBS) containing cholera toxin (CT) (4 µg/mouse) as an adjuvant on days 0 and 3. Then, mice were orally challenged with 0.2 mL of α-Casein (6 mg in 1 ml PBS) containing CT (4 µg/mouse) on days 5 and 7. Allergic symptoms were observed within 1 h after the α-Casein challenge. UDCA was dissolved in 0.5% carboxymethyl cellulose sodium (CMC) with a concentration of 3 mg/ml and 6 mg/ml. Then, from day 9 to 15, mice received daily oral gavage of the following: the UL and UH groups received UDCA at 30 and 60 mg/kg, respectively, while the CMPA model group received an equal volume of 0.5% CMC. In the control group, mice were given an equal volume of PBS and 0.5% CMC. The animal experimental design was shown in Fig. 1. After the last oral gavage, individual mice were observed for 30 min to assess the clinical allergic symptoms according to the standard symptom scores (Table 1) as previously described^26,29. Mice body weights were measured once a day throughout the experiment. The whole blood was collected from the retro-orbital venous plexus, and centrifuged at 3,000 rpm for 10 min to obtain serum. After being sacrificed by dislocation of the cervical vertebra, mice were fixed in the dissecting tray to withdrawn the colon tissues under sterile conditions.

Fig. 1.

The animal experimental design.

Table 1.

Description of clinical allergic symptoms score in mice.

Scores	Behavioral appearance	Stool consistency
0	No symptom	No symptoms
1	Scratching around nose and head (< 5 episodes)	Normal stool mixed with yellow mucus
2	Puffiness around eyes and mouth, scratching; rubbing of nose, head, or feet (> 5 to < 10 episodes)	Loose stool
3	Wheezing, labored respiration, retardation, irritability, piloerection, stiffness of tail; severe scratching (> 10 episodes)	Loose stool with yellow mucus
4	No activity after prodding, or tremor and convulsion	Diarrhea
5	Death

Hematoxylin-eosin (HE), immunohistochemistry (IHC) and Alcian blue-periodic acid-Schiff (AB/PAS) staining

Colon tissues were collected, fixed with 4% paraformaldehyde, embedded in paraffin and sectioned. 4 μm paraffin sections were stained with HE and AB/PAS for histological analysis, or incubated with TGR5 (1:200) and NF-κB p65 (1:200) and stained with a diaminobezidine (DAB) kit for immunohistochemistry. Histological analysis and immunohistochemical quantification was performed using image analysis software (Image Pro Plus).

RAW264.7 cells culture and treatment

The RAW264.7 cells, purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China), were cultured in DMEM high glucose culture medium supplemented with 10% FBS and 1% penicillin/streptomycin in a 5% CO₂ atmosphere incubator at 37 °C with. Adherent cells with a purity of > 95% were utilized for subsequent experiments. RAW264.7 cells were seeded in a 96-well plate at a density of 5 × 10⁴ cells/mL, and then treated with different concentrations of UDCA (5, 10, 15, 20, 50, 100, 150, 200 µM) for 24, 36 h and 48 h to screen optimal concentration. In addition, RAW264.7 cells were stimulated with α-Casein (0.1 mg/mL) for 4 h to induce CMPA cell model, followed by different concentrations of UDCA (5, 10, 20 µM) for 24 h, 36 h and 48 h. Cell viability was evaluated by CCK-8 cell viability assay kit. RAW264.7 cells were seeded in a 6-well plate and pre-treated with TGR5 antagonist SBI-115 (0.25 µM) for 2 h. Subsequently, cells were stimulated with α-Casein for 4 h, and then treated with UDCA (10 µM) for 24 h. Cells were collected for subsequent experiments.

Migration and invasion assays

Migration and invasion were detected using transwell migration and a wound healing assay. RAW264.7 cells were seeded in 24-well plates at a density of 2.5 × 10⁵ cells per well to form a confluent monolayer. The scratcher was used to make a wound. Cells were stimulated with α-Casein (0.1 mg/mL) for 4 h and followed by UDCA (5, 10, 20 µM) for 24 h, and photographed under microscope. Migration assays were performed with 24-well transwell chambers with 8.0 μm pore size. In general, RAW264.7 cells resuspended in 200 mL serum-free DMEM were seeded into the upper chamber at a density of 5 × 10⁴ cells per well which contained α-Casein (0.1 mg/mL) and/or UDCA (5, 10, 20 µM). DMEM with 10% fetal bovine serum was added to the lower chamber. After incubation for 24 h, cells on the upper surface of the membrane were carefully removed. The cells migrated through the membrane were fixed with 4% paraformaldehyde for 15 min, stained with 0.1% crystal violet for 20 min. The stained cells were imaged under an inverted microscope in 3 random fields per membrane.

Quantitative real-time PCR

Total RNA was extracted from cells and frozen colon tissue using Trizol reagent. The RNA was reverse transcribed into cDNA using the Hifair^® Ⅲ 1st Strand cDNA Synthesis SuperMix for RT-qPCR. For RT-qPCR amplification, 1 µg of cDNA was mixed with 0.25 µM of each primer and 10 µL of Hieff^® RT-qPCR SYBR Green Master Mix (No Rox) in a 20 µL reaction solution. The relative expression level of the target mRNA was determined using the 2^−ΔΔCt method, with GAPDH serving as an internal control. The primer sequences used for RT-qPCR were designed using Primer Premier 5 software (Premier, Canada) following retrieval of the target gene mRNA sequences from the NCBI Nucleotide database. Primer sequences are provided in Table 2.

Table 2.

Primer sequences.

Genes	Direction	Sequence 5’-3’
GAPDH	Forward	TTGATGGCAACAATCTCCAC
	Reverse	CGTCCCGTAGACAAAATGGT
IL-6	Forward	CTGCAAGAGACTTCCATCCAG
	Reverse	AGTGGTATAGACAGGTCTGTTGG
IL-10	Forward	GCTGGACAACATACTGCTAACC
	Reverse	ATTTCCGATAAGGCTTGGCAA
IL-1β	Forward	GCAACTGTTCCTGAACTCAACT
	Reverse	ATCTTTTGGGGTCCGTCAACT
TNF-α	Forward	CACGCTCTTCTGTCTACTGA
	Reverse	ATCTGAGTGTGAGGGTCTGG
iNOS	Forward	CCCTTCCGAAGTTTCTGGCAGCAGC
	Reverse	GGCTGTCAGAGCCTCGTGGCTTTGG
Occludin	Forward	CAGCCTCGGTACAGCAGCAAT
	Reverse	ATAGTGGTCAGGGTCCGTCCTC
TGR5	Forward	CCTGGCAAGCCTCATCGTC
	Reverse	AGCAGCCCGGCTAGTAGTAG
NF-κB p65	Forward	TGCGATTCCGCTATAAATGCG
	Reverse	ACAAGTTCATGTGGATGAGGC
CCL2	Forward	CTTCTGGGCCTGCTGTTCA
	Reverse	CCAGCCTACTCATTGGGATCA
CCL5	Forward	GTGCTCCAATCTTGCAGTCG
	Reverse	AGAGCAAGCAATGACAGGGA
CCL7	Forward	GCTGCTTTCAGCATCCAAGTG
	Reverse	CCAGGGACACCGACTACTG

Western blot

Cells or colon samples were lysed in a lysis buffer supplemented with protease and phosphatase inhibitors. Protein concentrations were measured using the BCA assay. Proteins were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with NcmBlot blocking buffer for 10 min and then incubated overnight at 4 °C with primary antibodies against β-actin (1:5000), Lamin B (1:10000), TGR5 (1:1000), p-p65 (1:1000), and p65 (1:1000). Then the membranes were incubated with a secondary antibody (diluted 1:20000) for 2 h at 4 °C. β-actin antibody and Lamin B were used as internal standards to normalize protein expression. Protein blots were detected using the ECL western blotting kit.

NO measurement

RAW264.7 cells were cultured in a 96-well plate at a density of 5 × 10⁴ cells/mL, and treated with or without α-Casein for 4 h, followed by treatment with UDCA (5, 10, 20 µM) for 24 h. Finally, 50 µL of cell supernatant was mixed with an equal volume of Griess reagent (Beyotime, China) and resting for 10 min. Absorbance was measured at 540 nm to calculate the concentration of NO by reference to the standard curve.

Detection of ROS (Reactive oxygen species) production

RAW264.7 cells were cultured in a 6-well plate at a density of 2 × 10⁵ cells/mL and treated with or without α-Casein for 4 h. Then, UDCA treatment groups were incubated with UDCA (5, 10, 20 µM) for 24 h. The level of ROS was detected by Reactive Oxygen Species Assay Kit (Beyotime, China). DCFH-DA in ROS Assay Kit freely cross the cell membrane and hydrolyze to DCFH which is oxidized to fluorescent DCF by intracellular ROS. DCF fluorescence can be used to assess the level of intracellular ROS by fluorescence microscope (Olympus) and ImageJ software.

Immunofluorescence

The NF-κB Activation, Nuclear Translocation Assay Kit was used to test the trans-activation of the NF-κB signaling. RAW264.7 cells in indifference groups were fixed in 4% paraformaldehyde for 5–15 min. After being washed 3 times with washing solution, the cells were blocked with immunostaining blocking solution for 1 h, and then incubated with anti-NF-κB p65 primary antibody overnight at 4 °C. Afterward, the cells were washed 3 times and incubated with IgG-Cy3 for 1 h. The nuclei were stained with he nucleus staining solution (DAPI) for 5 min. Images were captured using fluorescence microscope (Olympus).

Statistical analysis

Data were analyzed with Prism 9.5 (GraphPad Software, LLC, San Diego, CA, USA). Data were expressed as mean ± standard error of the mean (SEM). Statistical differences between two groups were calculated by Student’s t-test, while multiple comparisons were analyzed using one-way ANOVA followed by a post-hoc Tukey’s test. The P-values < 0.05 were considered statistically significant.

Results

UDCA improved the pathological changes in CMPA mice model

To investigate the effect of UDCA on CMPA, the α-Casein-induced CMPA mice was established. As assessed by the allergic symptom score, CMPA mice displayed severe allergic symptoms (e.g., wheezing, scratching, piloerection) and a higher clinical score compared to the control group (P < 0.0001). UDCA treatment dose-dependently reduced this score (P = 0.0125 at 30 mg/kg and P = 0.0001 at 60 mg/kg) (Fig. 2a) and promoted body weight recovery from day 9 onward to the CMPA model group (Fig. 2b). H&E staining of liver tissue from CMPA mice showed hepatocyte necrosis, degeneration (red arrow), and leukocyte infiltration (black arrow), all of which were mitigated by UDCA, leading to a lower pathology score than control group (P = 0.0313 at 30 mg/kg) (Fig. 2c, e). In the colon, the CMPA model induced mucosal ulcers, irregular glands (yellow arrow), and inflammation (black arrow). In the UDCA groups, especially the 30 mg/kg dose group, showed restored colonic morphology, clear glandular structure, and reduced pathological scoring compared with the CMPA model group (P = 0.0055 at 30 mg/kg and P = 0.0014 at 60 mg/kg) (Fig. 2c, e). Additionally, AB/PAS staining confirmed that there were more goblet cells in CMPA mice treated with UDCA in colonic tissue sections than that in CMPA model group (P = 0.0027 at 60 mg/kg) (Fig. 2d, f). Together, these results suggest that UDCA may exert a therapeutic effect against α-Casein-induced CMPA.

Fig. 2.

UDCA ameliorated allergic symptoms of CMPA mice model. (a) The allergic symptoms in CMPA mice sensitized with α-Casein (n = 8). (b) Body weight changes of mice in different groups (n = 8). (c) and (e) Representative H&E staining and Histology score of the liver and colonic sections (n = 3). Lymphocyte infiltration (black arrow); Hepatocyte structure (red arrow); Irregular shape of intestinal glands (yellow arrow). (d) Representative AB-PAS staining of colonic sections. (f) The number of goblet cells and goblet cells/length (n = 3). C, A, UL and UH mean control group, CMPA model group, α-Casein + 30 mg/kg UDCA group and α-Casein + 60 mg/kg UDCA group, respectively. Values are expressed as mean ± SEM. Statistical differences were evaluated using Student’s t-test. ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 compared to CMPA model group. ^#P < 0.05 and ^##P < 0.01 were performed between UDCA-30 mg/kg and UDCA-60 mg/kg group.

UDCA inhibited the inflammation in CMPA mice

The mRNA levels of pro-inflammatory cytokines (IL-1β and TNF-α) in colonic tissues were significantly elevated in the CMPA model group, whereas UDCA groups demonstrated an inhibitory effect on the expression of IL-1β (P < 0.0001) and TNF-α (P = 0.0339 at 30 mg/kg and P = 0.0191 at 60 mg/kg) compared to CMPA model group (Fig. 3a, b). Inflammatory chemokines have been implicated in allergic inflammation. We also investigated the effect of UDCA on the mRNA expression of CCL2, CCL5, and CCL7 in CMPA pathogenesis. The results showed that, compared with the control group, the mRNA levels of chemokines were notably up-regulated in CMPA model group. However, the gene expression levels of CCL2 (P = 0.0006 at 30 mg/kg and P < 0.0001 at 60 mg/kg), CCL5 (P = 0.0254 at 30 mg/kg and P = 0.0002 at 60 mg/kg), and CCL7 (P < 0.0001) were reduced in CMPA mice after treated with UDCA (Fig. 3c, d, e). Occludin, a critical component of the tight junctions involved in maintaining intestinal barrier function, was also found to be significantly increased in α-Casein-sensitized CMPA mice administrated with UDCA, especially in UDCA-30 mg/kg group, thereby reducing α-Casein-induced intestinal barrier injury (P = 0.0006 at 30 mg/kg and P = 0.4349 at 60 mg/kg) (Fig. 3f).

Fig. 3.

UDCA inhibits inflammation of CMPA mice model. The mRNA expression of IL-1β (a), TNF-α (b), CCL2 (c), CCL5 (d), CCL7 (e), OCLN (f) in mice colon. Data are mean ± SEM (n = 3). Statistical differences were evaluated using Student’s t-test. ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 compared with CMPA model group. ^#P < 0.05 and ^##P < 0.01 were performed between UDCA-30 mg/kg and UDCA-60 mg/kg group.

UCDA regulated the intestinal immune in CMPA mice by NF-κB signal pathway

To elucidate the immune mechanisms of UDCA in CMPA, the effects of UDCA on the protein expression of TGR5 and p65 in the colonic tissue of different groups were assessed using immunohistochemistry (IHC) (Fig. 4a). Compared to the control group, α-Casein induced the decrease of TGR5 (P = 0.0019) and the increase of p65 (P = 0.018) in the colon of CMPA model group. For CMPA mice administrated with UDCA, UDCA up-regulated TGR5 (P = 0.0141 at 30 mg/kg and P < 0.0001 at 60 mg/kg) and inhibited p65 (P = 0.0003 at 30 mg/kg and P = 0.0165 at 60 mg/kg) expression in the colonic tissues of α-Casein-induced CMPA model mice, particularly in the UDCA-60 mg/kg group (Fig. 4a). Western blot analysis revealed a concomitant reduction in TGR5 expression, and an elevation in p-p65/p65 ratio and nuclear p65 levels in the colon tissue of CMPA mice. Conversely, UDCA treatment ameliorated the allergic response by up-regulating TGR5 expression (P = 0.0003 at 30 mg/kg and P = 0.0094 at 60 mg/kg), and inhibiting the levels of p-p65/p65 (P = 0.0001 at 30 mg/kg and P = 0.0002 at 60 mg/kg) and nuclear p65 (P = 0.0030 at 30 mg/kg and P = 0.0024 at 60 mg/kg) (Fig. 4b). To investigate the expression change of TGR5 in macrophages in colon tissue, the co-expression of TGR5 (red) and F4/80 (green) was detected. As a result, TGR5 co-expression with F4/80 was increased in the CMPA model group compared to the control group, while UDCA further promoted the maturation of macrophages (Fig. 4c), indicating that UDCA may participate in the CMPA-induced intestinal immune response by regulating macrophage function.

Fig. 4.

Mechanism of UDCA on intestinal immune in mice with CMPA. (a) The protein expressions of TGR5 and p65 in colon tissues were observed using IHC. (b) The protein expression of TGR5, p-p65, p65 and nuclear p65 in mice colon analyzed by western blot. (c) The co-expression of TGR5 (red) and F4/80 (green) in colon tissue of mice with CMPA by Immunofluorescence. C, A, UL and UH mean control group, CMPA model group, α-Casein + 30 mg/kg UDCA group and α-Casein + 60 mg/kg UDCA group, respectively. Data are mean ± SEM (n = 3). Statistical differences were evaluated using Student’s t-test. ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 compared to CMPA model group. ^#P < 0.05 and ^##P < 0.01 were performed between UDCA-30 mg/kg and UDCA-60 mg/kg group.

UDCA reduced the α-Casein-sensitized inflammatory response in RAW264.7 cells

The viability of RAW264.7 cells treated with UDCA (0, 5, 10, 20, 50, 100, 150, 200 µM) showed low cell toxicity assessed by CCK-8 assay (Fig. 5a). The concentrations of UDCA (5, 10, 20 µM) were selected for further experiments. Additionally, RAW264.7 cells were stimulated with α-Casein and subsequently treated different concentrations of UDCA (5, 10, 20 µM) for 24 h, 36 h and 48 h. Results showed that α-Casein-induced cellular activation of RAW264.7 cells remained unaffected after treatment with UDCA for 24 h and 36 h. While α-Casein stimulation led to increased activation at 48 h, UDCA treatment modulated this response (Fig. 5b). Therefore, the processing time of UDCA was 24 h. Healing (Fig. 5c) and migration (Fig. 5d) of RAW264.7 cells sensitized with α-Casein were promoted by UDCA, providing functional evidence that UDCA enhances macrophage motility in CMPA. Additionally, in α-Casein-sensitized macrophages UDCA markedly suppressed the ratio of p-Akt to Akt (P = 0.0061, 0.0013, 0.0007 at 5 µM, 10 µM, 10 µM, respectively), while showed no effect on the p-ERK1/2 to ERK1/2 ratio (P > 0.05) (Fig. 5e). ELISA and RT-qPCR results showed that the protein and mRNA levels of pro-inflammatory cytokines (IL-1β (P = 0.0035 with ELISA, P < 0.0001 with RT-qPCR), IL-6 (IL-1β (P = 0.0008 with ELISA, P < 0.0001 with RT-qPCR), and TNF-α (IL-1β (P = 0.0005 with ELISA, P < 0.0001 with RT-qPCR)) were increased, while anti-inflammatory cytokine IL-10 (P = 0.0003 with ELISA, P = 0.0045 with RT-qPCR) decreased in α-Casein-sensitized RAW264.7 cells. After treatment with UDCA, the secretion and mRNA of pro-inflammatory cytokines significantly decreased, and IL-10 increased in a dose-dependent manner (Fig. 5f).

Fig. 5.

Effects of UDCA on the proliferation and inflammatory response in α-Casein-sensitized RAW264.7 cells. (a) RAW264.7 cells were treated with concentrations of UDCA (0–200 µM) for 24 h, 48 h and 72 h to select the optimal concentration and time by CCK8. (b) RAW264.7 cells induced by treated α-Casein (0.1 mg/mL) were treated with UDCA (5, 10, 20 µM) and measured the cell viability by CCK8. (c) Representative images acquired 24 h after wounding of RAW264.7 cells treated UDCA. (d) Representative photographs of the Transwell migration assay of RAW264.7 cells co-cultured with UDCA. (e) The effect of UDCA on the phosphorylation of Akt and ERK1/2 in α-Casein-sensitized RAW264.7 cells were measured by Western-blot. (f) The effect of UDCA on the secretion and mRNA levels of TNF-α, IL-10, IL-6 and IL-1β in α-Casein-sensitized RAW264.7 cells were measured by ELISA and qPCR, respectively. Data are mean ± SEM (n = 3). Statistical differences were evaluated using Student’s t-test. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 and ^****P < 0.0001 compared to CMPA group.

The mRNA level of iNOS in RAW264.7 cells was further quantified by qPCR. Stimulation with α-Casein significantly up-regulated iNOS expression compared to the control group (P < 0.0001), and this increase was suppressed by UDCA in a dose-dependent manner (Fig. 6a). Consistent with this finding, UDCA treatment also significantly attenuated the α-Casein-induced production of NO compared with the CMPA model group (5 µM, P = 0.0963; 10 µM, P = 0.0065; 20 µM, P = 0.0001) (Fig. 6b). Furthermore, ROS Assay Kit was used to measured the level of intracellular ROS. The results confirmed that UDCA markedly reduced the intracellular levels of ROS which was elevated by α-Casein stimulation (5 µM, P = 0.0175; 10 µM, P = 0.0025; 20 µM, P = 0.0001) (Fig. 6c).

Fig. 6.

Effects of UDCA on the release of NO and ROS in α-Casein-sensitized RAW264.7 cells. (a) The mRNA expressions of iNOS in RAW264.7 cells. (b) Effect of UDCA on α-Casein-induced NO expression in RAW264.7 cells. (c) Effect of UDCA on α-Casein-induced ROS accumulation. Data are mean ± SEM (n = 3). Statistical differences were evaluated using Student’s t-test. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 and ^****P < 0.0001 compared to CMPA group.

UDCA inhibited the activation of NF-κB pathway via TGR5 in vitro

In vitro analyses revealed that UDCA up-regulated TGR5 at both mRNA (P = 0.0069, 0.0003, 0.0001 at 5 µM, 10 µM, 10 µM, respectively) and protein (P = 0.1296, 0.0487, 0.0001 at 5 µM, 10 µM, 10 µM, respectively) levels compared to the CMPA group in RAW264.7 cells (Fig. 7a, c). ELISA results showed that α-Casein inhibited the level of cAMP in RAW264.7 cells, whereas UDCA markedly restored the cAMP levels at the concentration of 10 µM (P = 0.095) and 20 µM (P = 0.008) (Fig. 7b). Additionally, UDCA suppressed the α-Casein-induced activation of the NF-κB pathway, as evidenced by a reduction in p65 mRNA (P = 0.0009, 0.0001, 0.0001 at 5 µM, 10 µM, 10 µM, respectively), a decreased p-p65/p65 ratio (P = 0.0001, 0.0001, 0.0014 at 5 µM, 10 µM, 10 µM, respectively), and inhibited nuclear translocation of p65 (P = 0.0762, 0.0001, 0.0003 at 5 µM, 10 µM, 10 µM, respectively) in RAW264.7 cells (Fig. 7a, c and d). A nuclear translocation assay confirmed that α-Casein induced robust nuclear import of p65, which was inhibited by UDCA in α-Casein-sensitized RAW264.7 cells. (Fig. 7c).

Fig. 7.

The mechanism of UDCA in α-Casein-sensitized RAW264.7 cells. (a) The mRNA levels of TGR5 and p65 in RAW264.7 cell. (b) The effect of UDCA on the cAMP level in α-Casein-sensitized RAW264.7 cells. (c) The protein expression of TGR5, p-p65 and p65 in α-Casein-sensitized RAW264.7 cells treated with UDCA (5, 10, 20 µM). (d) The nuclear translocation of p65 in α-Casein-sensitized RAW264.7 cells with or without UDCA. Data are mean ± SEM (n = 3). Statistical differences were evaluated using ANOVA. ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 compared to CMPA group.

SBI-115, pharmacological antagonism to selectively block TGR5, was used to confirm the pathway involved in RAW264.7 cells. Results showed that in α-Casein-sensitized macrophages SBI-115 had no effect on the ratios of p-Akt/Akt and p-ERK1/2/ERK1/2 ratio induced by UDCA group (P > 0.05) (Fig. 8a). ELISA results UDCA significantly suppressed the secretion of the pro-inflammatory cytokines TNF-α (P = 0.0019), IL-6 (P = 0.0165), and IL-1β (P = 0.0010) induced by SBI-115 in macrophages. Furthermore, in α-Casein-sensitized macrophages, SBI-115 enhanced TNF-α production (P = 0.0002) but inhibited IL-10 secretion (P < 0.0001) (Fig. 8b). Further analysis showed that UDCA markedly up-regulated the cAMP level which inhibited by SBI-115 in macrophages (P = 0.0011), and similar results were observed in α-Casein-sensitized macrophages (Fig. 8c). Additionally, SBI-115 inhibited the expression of TGR5 in RAW264.7 cells compared with the control group (P < 0.0001), which was abolished by UDCA. In α-Casein-sensitized RAW264.7 cells, UDCA significantly induced the expression of TGR5 than that in SBI-115 (P < 0.0001). UDCA dramatically dampened nuclear accumulation (P = 0.0002) in α-Casein-sensitized RAW264.7 cells, SBI-115 further augmented this inhibition (P = 0.0005) (Fig. 8d). Thus, UDCA likely acts through TGR5 to inhibit p65 phosphorylation and activation, mitigating the allergic response.

Fig. 8.

Effects of UDCA on the expression of relative proteins in α-Casein-sensitized RAW264.7 cell treated with or without SBI-115. (a) The effect of UDCA (10 µM) on expression of Akt, p-Akt, ERK1/2 and p-ERK1/2 in α-Casein-sensitized RAW264.7 cell treated with SBI-115. (b) and (c) The levels of cAMP, TNF-α, IL-10, IL-6 and IL-1β in α-Casein-sensitized RAW264.7 cells treated with UDCA and SBI-115. (d) The expression of TGR5, p-p65, p65 and Akt in α-Casein-sensitized RAW264.7 cell treated with UDCA and SBI-115. Data are mean ± SEM (n = 3). Statistical differences were evaluated using ANOVA. ^*P < 0.05, ^**P < 0.01 and ^***P < 0.001 compared to CMPA group; ^#P < 0.05, ^##P < 0.01and ^###P < 0.001 compared to SBI-115 group.

Discussion

CMPA has become a major public health issue affecting infants and young children. In infants, CMPA presenting with gastrointestinal symptoms is primarily bloody stools, vomiting, abdominal distension, diarrhoea, etc³. Food protein-induced allergic proctocolitis is characterized by inflammation of the distal colon and disruption of the intestinal epithelium due to inflammatory cell infiltration². These disorders typically present in infancy and are often triggered by cow’s milk protein^30,31. Gut microbiota-derived bile acids participate in gastrointestinal food allergies via regulating the function of intestinal immune cells³². Interestingly, our previous studies have shown that children with CMPA exhibit reduced levels of gut flora-derived SBAs, particularly UDCA²¹. This study demonstrates that UDCA is mechanistically linked to TGR5 and the suppression of NF-κB signaling pathway, ultimately leading to the restoration of intestinal immune homeostasis.

Previous studies and our observations indicate that cow’s milk protein disrupts the intestinal barrier^11,28. The α-Casein-sensitized mice exhibited classic signs of allergic symptoms, including wheezing, scratching, and piloerection, alongside impaired weight gain. The observed hepatocyte necrosis and leukocyte infiltration in the liver with emerging evidence suggesting that CMPA can have systemic effects beyond the gastrointestinal tract, potentially due to the circulation of inflammatory mediators or immune complexes. Low doses of UDCA (30 mg/kg) significantly improved liver histology, while the high-dose group (60 mg/kg) showed a pathology score comparable to the CMPA model. This may indicate a plateau effect on histologic repair at higher doses, a possibility that warrants further investigation. More critically, in the colon, the hallmark site of CMPA pathology, we observed mucosal ulcers, irregular glandular architecture, and significant inflammatory cell infiltration. The administration of UDCA resulted in a dose-dependent reversal of these pathological changes. The reduction in clinical allergy scores and the improvement in body weight trajectory indicate a systemic alleviation of the allergic state. Histologically, UDCA, particularly at the 30 mg/kg dose, markedly restored colonic architecture, reduced inflammation, and lowered pathological scores. A particularly noteworthy finding was the depletion of goblet cells in CMPA mice, which was vital for producing the protective mucus layer and serves as the first line of defense against luminal antigens. UDCA administration points to a potent restoration of intestinal barrier integrity evidenced by increased goblet cells and tight junction protein expression, which may in turn suppress allergen-induced immune responses. The pathophysiology of CMPA is driven by a dysregulated inflammatory response. Willart et al.³³ found that UDCA treatment promoted IL-12 production and enhanced dendritic cells migration, thereby reducing in eosinophilic airway inflammation in ovalbumin-sensitized mice. Our data confirm a significant up-regulation the secretion and mRNA levels of pro-inflammatory cytokines (IL-6, IL-1β and TNF-α) in vitro and in vivo. The up-regulation of cytokines s highlights a robust recruitment signal for monocytes, macrophages and T cells. Administration with UDCA down-regulate these cytokines to suppress the inflammatory cell infiltration, thereby alleviating the local intestinal inflammation to limite the allergic effects. These results suggest that UDCA may alleviate the symptoms of CMPA by bi-directional immune regulation, with simultaneous suppression of pro-inflammatory responses, while enhancing anti-inflammatory and reparative responses. Future investigations measuring systemic cytokine levels could provide a more comprehensive view on the effect of UDCA on the systemic immune response in CMPA.

Macrophages play a dual role in the immune response to food antigens by participating in either the induction of oral tolerance or the initiation of allergic sensitization^23,34. For instance, β-lactoglobulin (β-LG) stimulation activated the TLR4/NF-κB signaling pathway in macrophages, thereby promoting allergic inflammatory responses characterized by elevated iNOS expression, ROS production, and proinflammatory cytokine release³⁵. Similarly, Casein, known to have a stronger allergenic potential than β-LG, also modulates macrophage function and regulates immune responses^25,36. Consistent with these findings, our study demonstrated α-Casein alters macrophage function both in vivo and in vitro. In α-Casein-induced RAW264.7 cells, UDCA suppressed the release of inflammatory cytokines, reduced iNOS expression, and decreased the overproduction of NO and ROS. The increased co-expression of TGR5 with the macrophage marker F4/80 in the CMPA mice, which was further enhanced by UDCA, indicates that macrophages in the colon are actively engaged in the pathology and are a primary cell type through which UDCA signals. Therefore, the anti-allergic effect of UDCA against milk protein-induced inflammation is likely mediated, at least in part, through the regulation of macrophage immune responses.

To elucidate the mechanism by which UDCA protects against α-Casein-induced allergy, we focused on the TGR5 and the central inflammatory transcription factor NF-κB. Studies have shown that bile acids are known to regulate inflammation, energy metabolism and cell proliferation through TGR5 signaling pathway³⁷. Meanwhile, NF-κB, one of the most studied transcription factors, has been activated in allergic airway inflammation to promote the release of downstream inflammatory mediators³⁸. IHC and Western blot analyses consistently showed down-regulation of TGR5 and activation NF-κB in the CMPA group. The administration of UDCA promoted the expression of TGR5 and inhibited the NF-κB pathway by reduced p-p65/p65 ratios and nuclear p65 levels in the colon of CMPA mice. UDCA initiates pro-inflammatory NF-κB cascade by TGR5, which is consistent with established literature where TGR5 expression in immune cells leads to the inhibition of NF-κB nuclear translocation. In vivo research found that SBI-115 inhibited TGR5 expression and up-regulated NF-κB phosphorylation, however this effect was abolished by UDCA in α-Casein-sensitized RAW264.7 cells. Additionally, cAMP acts as a second messenger produced within the cell upon TGR5 receptor activation, responsible for transmitting signals and triggering a series of downstream biological effects. Research indicates that deoxycholic acid contributes to the TGR5-mediated inhibition of NF-κB and NLRP3 via the cAMP-PKA signaling pathway³³. In this research, UDCA markedly up-regulated the cAMP level which inhibited by α-Casein and SBI-115in macrophages. It demonstrates that TGR5 is functionally requisite for UDCA to exert its inhibitory effect on NF-κB phosphorylation and nuclear translocation in macrophages. Further evidence suggests that UDCA inhibited the nuclear translocation of NF-κB p65 in α-Casein-sensitized cells. He et al.³⁹ proved that UDCA blocked bacterial invasion and alleviated commensal bacterial dysbiosis in neonatal mouse models of sepsis and colitis through the TGR5-NF-κB axis. Given that the inhibition effect of UDCA on NF-κB activation was blocked by SBI-115, and that NF-κB was the principal transcriptional regulator of the cytokines and chemokines suppressed by UDCA, we conclude that the anti-inflammatory effects in allergic enteritis of UDCA are mediated through the NF-κB pathway targeting TGR5.

Interestingly, our study found that UDCA promoted the healing and migration of RAW264.7 cells sensitized with α-Casein, while suppressed the ratio of p-Akt/Akt. Based on our findings of UDCA on the expressions of TGR5, cAMP, and NF-κB, we propose an integrated model wherein UDCA reprograms macrophage function through the TGR5/NF-κB signaling. This reprogramming is characterized by the concurrent suppression of pro-inflammatory signaling (NF-κB phosphorylation and cytokine production), inhibition of Akt phosphorylation, and restoration of migration and invasion capabilities. The observed dissociation between reduced Akt activity and enhanced motility can be explained by the dominant role of cAMP and its downstream effector IL-10, which are established drivers of macrophage migration independent of Akt. Biologically, this unique phenotype-low inflammation, high motility, and reduced Akt signaling-may reflect a macrophage state conducive to inflammation resolution and tissue repair, rather than proliferative growth. Our study thus delineates a sophisticated immunomodulatory mechanism of UDCA and highlights the potential of targeting the TGR5 to therapeutically reshape macrophage behavior in inflammatory diseases. Metabolomics analysis of intestinal flora in CMPA children has been shown a reduce, which was correlated with the changes in the abundance of a variety of intestinal flora²¹. He et al.⁴⁰ revealed the potential mechanism of UDCA in colitis by which the altered microbiota was associated with NF-κB signaling pathway in macrophages. UDCA may be a risk of advanced CMPA through modulating homeostasis of the intestinal immune microenvironment. However, the causal relationship between UDCA and the alteration of gut microbiota in allergic enteritis remains to be well characterized. Further experiments with measurement of serum UDCA levels or germ-free mice are encouraged to validate the role of UDCA in CMPA.

In conclusion, our study demonstrated that the potential mechanism of UDCA in allergy may target TGR5, characterized by an increase of cAMP levels, to reduce pro-inflammatory factors and restore intestinal barrier integrity by NF-κB pathway in macrophages, ultimately alleviating both systemic allergic symptoms and local intestinal pathology. While our study establishes a critical role for the TGR5 receptor in mediating the anti-inflammatory effects of UDCA, the molecular mechanism of the cell reprogramming within macrophages warrant further elaboration. Moreover, future studies should explore the relationship between UDCA and gut microbiome in allergic colitis and investigate its potential synergistic effects with other therapeutic strategies, such as probiotics or oral immunotherapy. Therefore, targeting UDCA or UDCA related gut microbiota may provide novel insights for prevention or mitigation of allergic diseases.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CMPA

Cow’s milk protein allergy

UDCA

Ursodeoxycholic acid

SBAs

Secondary bile acids

DCA

Deoxycholic acid

LCA

Lithocholic acid

FXR

Farnesoid X receptor

cAMP

cyclic adenosine monophosphate

DMEM

Dulbecco’s modified Eagle’s medium

CCK-8

Cell Counting Kit-8

IHC

Immunohistochemistry

DAB

Diaminobezidine

PVDF

Polyvinylidene difluoride

FPIES

Food protein-induced enterocolitis syndrome

FPIAP

Food protein-induced allergic proctocolitis

β-LG

β-lactoglobulin

Author contributions

Conceptualization: Xiaoqin Li, Wancun Zhang; Methodology: Zhidan Yu, Zihui Wang, Lingling Yue, Yuesheng Wang; Formal analysis and investigation: Zhidan Yu, Zihui Wang, Lingling Yue; Software: Bo Sun, Ruifeng Wang, Xiangzhan Zhu; Writing - original draft preparation: Zhidan Yu, Zihui Wang; Writing - review and editing: Xiaoqin Li; Funding acquisition: Zhidan Yu; Supervision: Lifeng Li, Wancun Zhang. All authors reviewed the manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No. 81903330), the Scientific and Technological Projects of Henan Province (252102311161), the Medical Science and Technology Project of Henan Province (SBGJ202303047), and Research Project for Young Medical Researchers of Henan Academy of Medical Sciences (QNYJ2023009).

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethical statement

In our study, mice were euthanized using the cervical dislocation method. The procedure was performed by trained personnel to ensure it was carried out quickly and efficiently, minimizing any potential distress to the animals. All animal procedures, including euthanasia, were approved by the relevant animal ethics committee of Zhengzhou University and conducted in accordance with established ethical guidelines.

Ethics approval

The animal experiments were approved by the The Ethics Committee of Zhengzhou University and all methods are reported in accordance with ARRIVE guidelines.