Microbiota governs host chenodeoxycholic acid glucuronidation to ameliorate bile acid disorder induced diarrhea

Lin, Zishen; Feng, Yue; Wang, Jinping; Men, Zhaoyue; Ma, Xi

doi:10.1186/s40168-024-02011-8

Research
Open access
Published: 04 February 2025

Microbiota governs host chenodeoxycholic acid glucuronidation to ameliorate bile acid disorder induced diarrhea

Zishen Lin¹,
Yue Feng¹,
Jinping Wang¹,
Zhaoyue Men¹ &
…
Xi Ma¹

Microbiome volume 13, Article number: 36 (2025) Cite this article

3266 Accesses
6 Altmetric
Metrics details

Abstract

Background

Disorder in bile acid (BA) metabolism is known to be an important factor contributing to diarrhea. However, the pathogenesis of BA disorder-induced diarrhea remains unclear.

Methods

The colonic BA pool and microbiota between health piglets and BA disorder-induced diarrheal piglets were compared. Fecal microbiota transplantation and various cell experiments further indicated that chenodeoxycholic acid (CDCA) metabolic disorder produced CDCA-3β-glucuronide, which is the main cause of BA disorder diarrhea. Non-targeted metabolomics uncovered the inhibition of the BA glucuronidation by Lactobacillus reuteri (L. reuteri) is through deriving indole-3-carbinol (I3C). In vitro, important gene involved in the reduction of BA disorder induced-diarrhea were screened by RNA transcriptomics sequencing, and activation pathway of FXR-SIRT1-LKB1 to alleviate BA disorder diarrhea and P53-mediated apoptosis were proposed in vitro by multifarious siRNA interference, CO-IP, immunofluorescence, and so on, which mechanism was also verified in a variety of mouse models.

Results

Here, we reveal for the first time that core microbiota derived I3C represses gut epithelium glucuronidation, particularly 3β-glucuronic CDCA production, which reaction is mediated by host UDP glucuronosyltransferase family 1 member A4 (UGT1A4) and necessary of BA disorder induced diarrhea. Mechanistically, L. reuteri derived I3C activates aryl hydrocarbon receptor to decrease UGT1A4 transcription and CDCA-3β-glucuronide content, thereby upregulating FXR-SIRT1-LKB1 signal. LKB1 binds with P53 based on protein interaction, ultimately resists to apoptosis and diarrhea. Moreover, I3C assists CDCA to attain the ameliorative effects of FXR activation in BA disorder diarrhea, through reversion of abnormal metabolism pathway, improving the outcomes of CDCA supplement.

Conclusion

These findings uncover the crucial interplay between gut epithelial cells and microbes, highlighting UGT1A4-mediated conversion of CDCA-3β-glucuronide as a key target for ameliorating BA disorder-induced diarrhea.

Video Abstract

Introduction

Bile acid (BA) disorders are crucial associated with various gastrointestinal diseases in mammals, including human diarrhea-predominant irritable bowel syndrome (IBS-D), inflammatory bowel diseases (IBD), and gastrointestinal cancers, or pathogenic or weaning diarrhea in pigs [1,2,3]. Elevated levels of fecal total BAs serve as a common indicator of these disorders [4,5,6], and these disorders are characterized by excessive primary BA biosynthesis, cholestasis, and impaired BA absorption, all of which are associated with reduced expression of the farnesoid X receptor (FXR) in the intestinal epithelium [5, 7, 8]. Impairments in FXR and downstream fibroblast growth factor (FGF) signaling are key factors contributing to the severe intestinal inflammation, compromised mucosal barrier, and bile acid diarrhea observed in IBS-D patients [9]. Consequently, targeting FXR activation has become a crucial strategy for improving BA-related gastrointestinal diseases. Chenodeoxycholic acid (CDCA), an acknowledged FXR activator, has shown promise in reducing intestinal inflammation in mice [10]. Oral supplementation of CDCA alone has also demonstrated the potential to enhance growth performance and modulate the gut microbiome in piglets [11]. However, CDCA is not currently used in clinical practice or animal production for the treatment of gastrointestinal diarrhea conditions. Conversely, CDCA supplementation impairs the FGF19 expression when taken with a meal in cases of primary BA diarrhea [12], and increases bacterial mucosal uptake in the human mucosa, exacerbating mucosal barrier dysfunction in patients with collagenous colitis in remission [13]. In conclusion, although CDCA has the ability to activate FXR, its application appears to promote diarrhea and exacerbate injury, while the underlying mechanisms responsible for this phenomenon remain unclear.

UDP-glucuronide glycosyltransferases (UGTs) play a crucial role in detoxification by combining with endogenous and exogenous chemical to eliminate them from the host [14]. However, it has been observed that several UGTs mediating glucuronide metabolites can bind to proteins in gut epithelial cells and lead to toxic effects [15]. Specifically, UGT1 and UGT2 families are crucial enzymes involved in glucuronidation associated with toxic reactions [16]. Members of the UGT1A and UGT2B families mediate glucuronidation at the 3/6-hydroxyl or 24-carboxylic sites of BAs. The activity of these enzymes is regulated by ligand-activated transcription factors, particularly the nuclear receptor superfamily, and their natural or synthetic ligands, such as the aryl hydrocarbon receptor (AHR) [14, 17, 18]. Apart from UGT-mediated glucuronidation, BAs can be metabolized into tauro-/glyco-BAs, deconjugated secondary BAs, or other microbial BAs through the action of bile salt hydrolases produced by gut microbes [19]. Overwhelming evidence has revealed the crucial effects of gut microflora on BA metabolism. Pathogenic bacteria such as Escherichia coli, Brachyspira, and Ruminococcus gnavu have been shown to induce BA diarrhea [20,21,22], whereas probiotics have shown promise in modulating bile acid metabolism [23]. However, the specific roles of UGT and its glucuronidation in BA disorder-induced diarrhea, as well as the regulatory role of the gut microbiota in the process of glucuronidation, remains unknown.

Liver kinase B1 (LKB1) is an important upstream protein kinase that senses nutrients in multiple tissues, especially influencing the gut [24]. Additionally, LKB1 has been shown to regulate the fate of intestinal stem cells by restriction their differentiation into secretory lineages [25, 26]. Recently, studies have also exhibited the significance of LKB1 in inflammatory regulation and gut microflora alternation [27]. These findings indicate that LKB1 plays a vital role in gut development, cellular metabolism, and inflammatory resistance. However, the specific effects of LKB1 in enabling the host to resist BA disorder-induced diarrhea remains unknown.

In this study, we discovered that the development of BA disorder-reduced diarrhea is linked to a reduction in the colonization of Lactobacillus reuteri (L. reuteri), which leads to the toxic effects of CDCA glucuronidation. In the absence of L. reuteri-derived indolyl 3 carbinol (I3C), the host UGT1A4 enzyme converts CDCA into CDCA-3β-glucuronide at the third site, leading to the repression of FXR and the induction of gut epithelial apoptosis. In vivo experiments, mice that received the FXR agonist GW4064 orally exhibited a reduction in LKB1 acetylation and a mitigation of P53-mediated apoptosis in intestinal epithelial cells, ultimately leading to an improvement in diarrhea caused by BA dysregulation. Therefore, we have elucidated the underlying pathogenic mechanism of BA dysregulation-induced diarrhea and proposed a potential small molecule I3C that could be effective in ameliorating BA disorder-induced diarrhea in the improvement of mammal health.

Materials and methods

Animal models

The experimental procedures were approved by the Institutional Animal Care and Use Ethics Committee of China Agricultural University (AW82403202-1–1 and AW82403202-1–2).

Twenty-five-day, Duroc × (Landrace × Large), male, weaned piglets were selected; these piglets were picked from the three litters (8–10 piglets per litter), born and weaned on the same day during the same feeding environment. Then, at least eight healthy and at least eight diarrheal weaned piglets were randomly selected for the experiments, according to the diarrhea situation within 3 days, based on the following criteria [28,29,30]: A quantitative score of 0 to 3 points was performed based on the consistency of feces: A score of 0 indicates normal feces; A score of 1 indicates soft stool and mild diarrhea; A score of 2 indicates sticky loose stool and moderate diarrhea. A score of 3 means watery stool and severe diarrhea. The healthy weaned piglets chosen in our work had feces with moderate in hardness and normal in morphology (fecal score was 0), and we chose piglets with a fecal score of 3, indicating watery stool to represent the diarrheal group. Feces of these piglets were collected to measure the total bile acid content, and detect the abundance of E. coli through plate count. In addition, pigs were selected from the group with an average daily diarrhea of 3 to 5 times in 3 days, and total bile acid content were nearly approach to 30,000 ng/mg and abundance of E. coli in the fecal dilution coated plate were significantly higher than that in the healthy group, were constitute to bile acid disorder diarrhea (BADD) group. All pigs were supplied by Lianjiang Jinmei agricultural comprehensive company Co., Ltd. (Fujian, China) to collect colonic tissue and chyme.

Mice models

Three-week-old and 5-week-old male C57BL/6 mice were purchased from the SPF Biotechnology Co., Ltd. (Beijing, China). All mice gained daily care with a 12 h light/12 h dark cycle. Mice were established to different models as following. (a) Antibiotic mixture (ABX) mouse model: 3-week-old C57BL/6 mice drink antibiotic mixture as planned for 4 weeks. Antibiotic mixture was dispensed with sterile water. Antibiotic mixture was updated for mice once a week to ensure the effect on microbiota deletion. After 4 weeks, mouse feces were collected keep in ice box and coating to detect microbial content rapidly, and untreated mouse feces were used as control. The number of bacteria colonies on the medium was significantly less than that in the control group, or scarcely any colony formation could be regarded as successful construction. (b) FMT mouse model: ABX mice were received colonic microbiota transplantation from healthy donors or BADD donors, respectively. (c) UGT1A4 depletion mouse model: Each mice received 10 mg/kg body weight UGT1A4 inhibitor Gitogenin (HY-N2574, MCE, USA) intraperitoneal injection for 24 h, then the mRNA level of UGT1A4 measured in the gut is reduced by 10 to 100 times, which means model is successfully established. (d) LKB1 depletion mice model: Each mice received 40 mg/kg body weight LKB1 inhibitor (HY-10371, MCE, USA [31]) intraperitoneal injection once, then mRNA level of LKB1 measured in the gut is reduced than control group that means model is successfully established. (e) DSS model: In this study, DSS model used is to establish UGTs induce bile acid metabolism disorders model. Briefly, five-week-old mice began to receive 3% DSS (dissolved in sterile water) at the end of the 30 mg/kg FXR agonist GW4064 (MCE [32, 33], USA) supplement. The mice were planned to drink DSS water for 7 days. However, according to the actual situation, sample collection at immediately before blood appears in the feces. (f) L. reuteri oral gavage model: 10⁹ CFU/mL L. reuteri were oral gavage to mice and FMT mice for 14 days, respectively. (g) CDCA-3β-glucuronide diarrheal model: Mice were intraperitoneal injected 5 mg/kg CDCA-3β-glucuronide once, and collect samples 24 h later. (h) I3C cotreated with CDCA trial: Twelve WT mice were divided into 4 groups that FMT CON, FMT BADD, FMT I3C, and FMT 13C + CDCA, respectively. FMT 13C + CDCA group mice were supplied together to gavage with 100 mg/kg CDCA and 50 mg/kg I3C, which have received FMT from BADD donors, or giving 50 mg/kg I3C alone to FMT BADD mice. In the above experiments, we collect several samples as following in different trials. Body weight was recorded, besides colonic tissue samples, colonic fixed samples, and colonic chyme samples were collected during sampling, liver and spleen were weighed, and colonic length was measured. Serum was isolated after the blood was centrifuged at room temperature at 15 min 4000 g. Fecal samples were collected to measure fecal water index immediately, and the rest was used for bacterial count, metabolomics sequencing, and total bile acid concentration measure. All samples were frozen at − 80 ℃.

Colonic microbiota transplantation

Colonic chyme of pigs was collected. To prepare the FMT suspension, approximately 8 g of colonic chyme was homogenized in 32 ml of sterile saline, then passed through 15 mesh, 40 mesh, and 60 mesh filters to remove impurities. Subsequently, the filtered chyme was dissolved in twice the volume of a 25% sterile glycerol saline mixture. After mice received the antibiotic mixture treatment (a combination of 200 mg/kg neomycin sulfate and 200 mg/kg metronidazole or 200 mg/kg ampicillin and 100 mg/kg vancomycin) for 4 weeks, excrementitious bacterium were nearly eliminated. The treated chyme samples were thoroughly mixed, and 1 ml of the chyme liquid was taken for viable bacteria count. The viable bacteria count was adjusted to 10⁹ CFU/ml using sterile normal saline. For the FMT procedure, the chyme suspension of different groups was administered to the mice via intragastric administration. Specifically, the FMT CON mice received chyme from healthy piglets, while the FMT BADD mice received chyme from diarrheal piglets. Each mouse was administered 200 μl of the chyme suspension via gavage. After colonization for 5 days, microbiota transplantation was completed.

Primary gut epithelial cell isolation

Colon samples of pigs were collected in sterile container at 0 ~ 4 ℃, carefully peeled off excess tissue, then colonic mucosa was mechanically scraped and divided into small pieces. After cleaning with PBS, pieces were digested with 0.1% collagenase type II at 4 ℃ overnight, subsequently incubated the cells DMEF/12 medium (Gibico, USA) with 10% FBS (Mei5bio, China) and 1% penicillin/streptomycin (Solarbio, China) at 37 ℃ with 5% CO₂. After confirming the cell adherence, the medium is requested to change once a day. Primary gut epithelial cells normal grows for 24 h then collected in TRIzol and used for transcriptomic sequencing.

Gut epithelial cell line culture

Gut epithelial cell line IPEC-J2 was gained from State Key Laboratory of Animal Nutrition and Feeding (Beijing, China), and cultured in DMEF/12 medium (Gibico, USA) with 10% FBS (Mei5bio, China) and 1% penicillin/streptomycin (Solarbio, China) at 37 ℃ with 5% CO₂. All these cells were used for following experiments under the normal growth situation. In this study, 50 μM and 100 μM CDCA, 10 μM CDCA-3β-glucuronide, and 10 μM UGT1A4 recombinant enzyme (UGT1A4 RE) were used to treat IPEC-J2 for 12–24 h, 10³ CFU/mL L. reuteri and 50 μM I3C were used to treated IPEC-J2 for 12 h. AHR antagonist (MCE, USA) were used to prevent AHR moving from the cytoplasm to the nucleus.

Bacterial isolation and culture

Two grams of colonic chyme were weighed and dissolved in PBS containing 25% glycerol at 1:4, then sifted to remove impurities and mix upside down several times until thoroughly blended. The treated liquid of different groups was diluted 1000, 100,000, and 1,000000 times to find the suitable dilution concentration. Then treated samples were cultured at suitable dilution concentration in the MRS agar or PYX medium at the temperature of 37 ℃ and 45 ℃, under the anaerobic conditions. Single bacterial colony selected should be purified for 3 times, then DNA was extracted by DNA extraction kit (CWbio, China) to identify.

Bile acid metabolomics sequencing and analysis

Bile acid standard solution preparation, bile acid extraction LC -MS/MS detection, and quality control samples were supported favor by Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). In brief, the bile acid standard product was weighed and prepared into a single master product in methanol, diluted with 50% acetonitrile and prepared into a working standard solution. Similarly, the isotope standard products (CDCA-D4, CA-D4) are also prepared into a single master product with methanol, and the working standard solution is added to the internal standard solution with a volume ratio of 1:1, then fully mixed to obtain solutions of different concentrations. Mixing standard working solution with methanol then vortex. Subsequently, ultrasound samples at 40,000 Hz power for 30 min, then frozen at − 20 ℃ for 90 min and centrifuged at 4 ℃ for 15 min. The supernatant was discarded and dried with nitrogen for precipitation [23, 34], thereby adding acetonitrile and mixing for ultrasounding and centrifugating, then leaving precipitation for LC–MS/MS (Sciex, USA). Finally, the concentrations of bile acids and their metabolites were calculated by linear regression equation.

16S rRNA microbial gene sequencing analysis

CTAB method was used to extract the total genomic DNA of the colonic chyme samples, and the DNA concentration was detected and diluted to 1 ng/μl. Then 16S rRNA genes of V3–V4 regions were amplified with (341F (5′- CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACNNGGGTATCTAAT-3′) by Phusion® High-Fidelity PCR Master Mix (New England Biolabs). PCR products were obtained by thermal cycling to perform AGAR gel electrophoresis and purify PCR products product. First, a sequencing Library was generated according to the NEB Next®Ultra DNA Library Prep Kit (Illumina, USA). After quality assessment, the library was sequenced on Illumina NovaSeq platform, and paired terminal reads of 250 bp were obtained. According to the “Qiime2 Atacama soil microbiome tutorial” tutorial complete follow-up analysis (https://docs.qiime2.org/2019.1/). Briefly, the feature sequence table was obtained by converting the original sequence fastq into the QIIME2 format for subsequent processing, followed by quality control, pruning, de-noising, splicing, and chimaera removal. Subsequently, the representative sequences of ASVs were mapped to the pre-treated 13_8 version of the GREENGENES database with 99% similarity to generate a species taxonomic information table, and all polluting mitochondrial and chloroplast sequences were eliminated [35]. Next, we used α diversity analysis and β diversity to identify species richness and bacterial composition structure from phylum to genus, and LEFSE analysis to screen for distinctively diverse bacterial organisms. At last, PICRUST software predicts the possible functional composition of the microbiome.

RNA-seq transcriptomics

Transcriptomics method was used to analyze the differential genes between healthy pigs and diarrheal pigs to further screen target genes of FXR. Briefly, total RNA isolated from primary gut epithelial cells of pigs was extracted by TRIzol reagent. After concentration determination of the obtained RNA, high quality RNA is used for cDNA library construction and Illumina sequencing by Genedenovo Biotechnology Co., Ltd. (Guangzhou, China). The resulting P values are adjusted using Bonferroni method for controlling the error discovery rate. Finally, the KEGG set was obtained by functional annotation of differentially expressed genes (deg) with fold change (FC) greater than 1.3. Next, we used the cluster Profiler R package to test the statistical enrichment of differentially expressed genes in the GO term, KEGG, and GSEA pathways. The differentially expressed genes to GO term mapping database (http://www.geneontology.org/), and calculate the number of genes each term, and which has the function of a GO gene list and the number of statistics, reoccupy hypergeometric examination to find the significant GO pathway enriched by differential genes [36]. KEGG identifies the major biochemical metabolic pathways and signal transduction pathways involved in differentially expressed genes through significant enrichment of pathway, which is similar to GO term enrichment analysis. Gene set enrichment analysis (GSEA) effectively makes up for the lack of effective information mining of minor genes by traditional enrichment analysis, and explains the regulatory effects of a certain functional unit (pathway, GO term or other) more comprehensively. GSEA and MSigDB were used to identify GO terms/pathways that differed between two groups [30, 37]. The expression scale was input, the genes were sequenced by Signal2Noisez, the enrichment scores (ES) and P values were calculated by default parameters, and the normalization results were obtained to |NES|. It is commonly thought that |NES|> 1, NOM P value < 0.05 of the gene sets in the pathway were significant; The greater the absolute value of NES, the higher the reliability of the analysis results.

Metabolomics sequencing

Metabolomics sequencing in this work is provided by Wuhan Metware Biotechnology Co., Ltd. The sequencing method of untargeted mass spectrometry was the same as in the previous study [38]. Here, we analyzed the significant metabolites at the range of VIP > 1, P value < 0.05, and detective level of metabolites was included level 1a or 1b.

LPS and UGT1A4 ELISA and total bile acid content assays

According to the instructions provided by the manufacturer, the mouse LPS Elisa kit (CSB, China) was used to determine the LPS content in the serum of mice, the total bile acids assay kit (NJJCBIO, China) was used to measure total bile acid concentration in chyme of mice, the mouse UGT1A4 enzyme Elisa kit (SRABLE, China) had detected the expression of UGT1A4 enzyme in mice colon. Pig UGT1A4 Elisa kit (SRABLE, China) was applied to measure the expression of UGT1A4 enzyme in the colon of pigs and the expression in the IPEC-J2 cell.

Caspase-1 and caspase-3 activity detection

According to the instructions provided by the Beyotime, caspase-1 activity assays (C1101) and caspase-3 activity assays kits (C1115) were used to measure the Caspase-1/3 enzyme activity in cell or tissue lysates by spectrophotometry.

RNA interference

All types of siRNA and NC siRNA were purchased from Integrated Biotech Solutions Co. Ltd. (Shanghai, China) and used the Lipofectamine reagents (Invitrogen, Carlsbad, CA, USA) to transfect into cells. The sequences of siRNA targeting of FXR is 5′-GGAGUGUCGACUAAAGGAAAUG-3′ (sense); the sequences of siRNA targeting of SIRT1 is 5′-GAGAUGGCAUUUAUGCACGUU-3′; the sequences of siRNA targeting of LKB1 is 5′-GCCAGGGCAUCGUGCACAAUU-3′; the sequences of siRNA targeting of negative control is UUCUCCGAACGUGUCACGUTT.

Total RNA extraction and quantitative reverse transcription PCR

Broadly, total RNA was extracted by total RNA extraction kit for tissues and cells (CWBIO, China). After concentration determination for RNA samples, and reverse transcript RNA into cDNA steadily by the reverse transcription kit (Mei5bio, China) and then used to RT-qPCR by SYBR green mix (Mei5bio, China) in LightCycler® 96 SW. All primers were shown in Table S2.

ChIP PCR analysis

Since JASPER could not predict binding fragments of porcine cells for the time being, we used mouse IEC-6 instead. About 10⁷ pieces of cells per dish were added 0.75% formaldehyde to hinge fixation at room temperature for 30 min, then add 125 mM glycine to gently shake for 5 min. Cells were washed and scraped off in cold PBS, then centrifuged at 1000 × g 4 ℃ and added FA lysate buffer, following by ultrasonic processing to obtain 500–1000 bp fragments. After that, samples were centrifuged at 4 °C at 8000 g for 30 s to precipitate the cell fragments; the supernatant was taken for DNA purification and DNA concentration detection. About 25 μg of chromatin with RIPA as per 1:10 diluted and added 5 µg of antibodies and 20 µl of protein A/G IgG beads, rotated overnight at 4 ℃, then centrifuged at 2000 × g for 1 min, whereafter washed with PBS for 3 times. Next, 120 μl of eluting buffer was added to protein A/G IgG beads, rotated and mixed for 15 min, then centrifuged at 2000 × g for 1 min. The supernatant containing protein /DNA complex was obtained, DNA has been extracted and quantified by RT-qPCR.

Protein interaction prediction

STRING database (https://cn.string-db.org) was used to investigate the protein–protein interactions involving the target protein LKB1 and apoptosis-related proteins, including BAX and BCL2, then performed a search using the names of these proteins within the STRING database and selected the species for analysis. During the analysis, P53 signal was identified to presence in the protein–protein interaction network. To further explore the correlation among LKB1, P53, and apoptosis-related proteins, a re-analysis was performed and the protein names of LKB1, P53, and the apoptosis-related proteins were re-input into the STRING database. Through the re-analysis, the relationship among LKB1, P53, and apoptosis-related proteins in humans, pigs, and mice were obtained and showed as network.

IP and Co-IP analysis

About 10⁷ pieces of cells per dish was washed with cold PBS and scrape off completely to add RIPA lysate buffer. After ultrasound, then centrifuge at 12,000 × g for 10 min at 4 ℃, and remove the supernatant, pre-washed protein A/G magnetic beads were added into EP tubes about 1 h rotation at 4 ℃. Next, all samples were used to quantify protein concentration to at least l mg/ml. (Tissues: about 5 mg of colon tissue was dissolved in RIPA lysate (added protease inhibitor and deacetylation inhibitor), centrifuged at low temperature after grinding, and the supernatant was ultrasonically broken and centrifuged at 4 ℃ 12,000 × g for 5 min, then the total protein was quantified to 2 μg/μl). A part of protein samples was taken as INPUT samples and temporarily frozen at − 80 ℃. Other protein samples were added the antibody, according to the dosage provided in the instructions and slowly shake the antigen–antibody complex at 4 °C overnight, then add 50% protein A/G magnetic beads to capture the antigen and antibodies overnight on a rotating vibrating screen at 4 °C. Centrifuge the bead-antigen–antibody complex 12,000 × g was collected and left for 1 min. The particles were retained and cleaned with pre-cooled PBS for 3 times. Buffer suspension particles were loaded with 60 μl 2 × SDS, then mixed evenly to boil at 100 ℃ for 5 min; finally, 50 μl supernatant was used for electrophoresis. Here, IP analysis was used to pull down LKB1 protein by acetylated-lysine antibody. Co-IP analysis was used to bidirectional pull down the one of LKB1 protein or P53 protein, then detect another protein of them by WB.

Histopathological analysis

The colonic tissues were trimmed and placed in an embedding box and washed with water droplets overnight to remove 4% paraformaldehyde. The embedding boxes were dehydrated with 70%, 85%, 95%, and 100% alcohol, and xylene I and xylene II in sequence. After waxing, embedding, and slicing, HE staining was performed. Briefly, HE staining method was consistent with the previous reports. After dewaxing and rehydrating, the slices were stained with hematoxylin, then washed with water, and differentiated with alcohol in hydrochloric acid before washing again. Dehydrated with 85% and 95% ethanol, the slices were dyed with eosin dye, and then soaked in 100% ethanol for three times, finally soaked in xylene for 5 min, the slices were dried and sealed.

Immunohistochemistry

After thorough dewaxing and rehydration of the sections, antigen repair was performed. Subsequently, the slides were rinsed twice 5 min on PBS containing 0.025% Triton X-100. Then PBS containing 10% normal serum and 1% BSA was blocked at room temperature for 2 h, and then the slides were drained. At this time, the primary antibody can be incubated at 4 ℃ overnight. Similarly, the antibody was washed in PBS containing 0.025% Triton X-100 and incubated with a special secondary antibody at room temperature for 1 h. After washing, it was incubated with streptavidin-perosidase at room temperature for 30 min. When washing with PBS finished, color development was performed with DAB for 3–10 min, then color development was terminated with tap water. Behind hematoxylin re-staining, cell nucleus differentiation by running tap water, and dehydration and transparency, the slides can be sealed.

TUNEL assay

We used Tunel Cell Apoptosis Detection Kit (Servicebio, Wuhan, China) in this work. Briefly, the paraffin sections were dewaxed and washed in distilled water. After repair with protease, the film breaking working solution was added, and the tissue was covered with buffer and balanced at room temperature. Subsequently, appropriate amounts of TDT enzyme, dUTP, and buffer were mixed and incubated at 1:5:50 for 1 h. The paraffin sections were dewaxed and washed in distilled water. After repair with protease, the film breaking working solution was added, and the tissue was covered with buffer and balanced at room temperature. Subsequently, appropriate amounts of TDT enzyme, dUTP, and buffer were mixed and incubated at 1:5:50 for 1 h. Restaining the nucleus with DAPI, then seal the film and observe.

Western blot trial

All tissues and cell samples were cracked in RIPA lysate buffer. The extraction, quantification, and analysis were operated as previously study [23]; the concentration of the sample for Western blot is 1 μg/μl. The detection of nuclear and cytoplasmic protein was performed according to the instructions of Nuclear and cytoplasmic protein extraction kit (Beyotime, China). Results were analyzed by Quantity One. All antibodies used are described in the next section.

Antibodies

The following antibodies were used in this article: FXR (72,105, Cell Signaling Technology, USA, WB: 1:1000 dilution, IF: 1:500 dilution, IHC:1:100 dilution, ChIP: 10 μl antibody/10 μg chromatin), LKB1 (10,746–1-AP, Proteintech Group, China, WB: 1:500 dilution, IF: 1:500, IHC:1:100 dilution, IP:4 μg), SIRT1 (60,303–1-Ig, Proteintech Group, China, WB: 1:500 dilution), BAX (50,599–2-Ig, Proteintech Group, China, WB: 1:1000 dilution), BCL2 (ab59348, Abcam, UK, WB: 1:1000 dilution), P53 (10,442–1-AP, Proteintech Group, China, WB: 1:1000 dilution, IF: 1:500, IP:4 μg), Phospho-MAPK3/MAPK1 (CSB-PA000749, Cusabio, China, 1:1000 dilution), MAPK3/MAPK1 (PA002419, Cusabio, China, 1:1000 dilution) NF-κB P65(8242, Cell Signaling Technology, USA, WB: 1:1000 dilution, IF: 1:500 dilution), AHR (67,785–1, Proteintech Group, China, IF: 1: 500 dilution), UGT1A4 (PA5-102,796, Invitrogen, USA, IF:1: 200 dilution), Laminb1 (12,987–1-AP, Proteintech Group, China, WB: 1:1000 dilution), β-actin (GB11001, Servicebio, China, 1:1000 dilution), GAPDH(GB11002, Servicebio, China, 1:1000 dilution), Tubulin α (2125, Cell Signaling Technology, USA, WB: 1:1000 dilution), Anti-mouse IgG (H + L) (5257, Cell Signaling Technology, USA, WB: 1:10,000 dilution), Anti-rabbit IgG (H + L) (515S, Cell Signaling Technology, USA, WB: 1:10,000 dilution), Cy3 Goat anti-mouse IgG antibody (GB21301, Servicebio, China, IF: 1:200 dilution), Alexa Fluor 488 labeled Goat Anti-mouse IgG antibody (GB25301, Servicebio, China, IF: 1:200 dilution), Immunohistochemical-goat anti-rabbit antibody (G1213, Servicebio, China, IHC: 1:200 dilution), and Immunohistochemical-goat anti-mouse antibody (G1214, Servicebio, China, IHC: 1:200 dilution).

Immunofluorescence analysis

Put the slides in the cell culture dishes in advance, and after the cells grew completely, the cells were fixed with 4% paraformaldehyde for 30 min, and then washed with cold PBS. Then, 0.25% Triton-100X was applied to permeable cells, and cleaned again with PBS. PBS containing 5% BSA was used as blocked buffer, now the cell slides were capable of incubating with primary antibody at 4 ℃ overnight. Next, secondary antibody for immunofluorescence (with green or red fluorescence) was added to incubation after washing with cold PBS. After staining the nucleus with the DAPI, the slices were sealed with a reagent containing an anti-fluorescence quencher.

Flow cytometry apoptosis detection

After cell cultured completely, the DMEF/12 medium was collected into the tubes. Subsequently, the cells are digested with pancreatic enzymes for 3 min to quickly terminate digestion. Cell precipitation was obtained by centrifugation at low speed (1000 × g). Here, the cell precipitate was cleaned with cold PBS to completely remove pancreatic enzymes. According to the flow detection apoptosis kit of Beyotime, the cells were stained with two channels of fluorescent dye (PI and FITC). In addition, the blank sample, the PI sample, and the FITC sample were set up separately. The cells were sieved and diluted into the flow tube, and the blank sample, separately stained FITC sample, and separately stained PI sample were first tested. About 10,000 cells were set to be collected in each tube and recorded immediately.

Cell viability assay (Cell Counting Kit-8)

Remove about 1/6 of 10⁷ pieces cell was used to culture into a 96-well plate. After waiting for the cells to attach to the plate, the cells were treated with drugs immediately for 10–12 h, then 10 μl of CCK8 reagent was added and incubated for 2–4 h. Next, the OD value was detected at 450 nm wavelength.

Loads of E. coli in the feces

Briefly, fresh fecal samples were collected, 0.5 g of fecal matter was weighed of each mouse, dissolved in sterile saline solution, mixed, and diluted in gradient (10³,10⁵,10⁷,10⁹, respectively), the E. coli in feces were cultured and counted on agar plates using selective media (MacConkey Agar), the red strain on the plate is E. coli [35], then count the red strain number in the ager plate and multiply by dilution was the CFU of E. coli.

Statistical analysis

Statistical analyses were used GraphPad Prism 8.0. Immunohistochemistry and immunofluorescence were analyzed by ImageJ. The data of this study, except from sequencing data, were shown as mean ± standard error of mean (SEM). The differences between groups were compared using either student’s t test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Unilateral spearman analysis with confidence interval of 90% was used for correlation analysis. Except for individual data significance marked with specific P values, the rest are marked with *P value < 0.05, **P value < 0.01, ***P value < 0.001expressed significance level.

Results

UGT1A4 mediated CDCA-3β-glucuronide overgeneration leads to BA disorder diarrhea

Abnormal BA metabolism during diarrhea has been reported without the research on the factors leading to BA disorder [27]. Additionally, pigs are considered a suitable biological model for studying gut diseases in humans [39]. Weaned piglets are prone to diarrhea; regarding the incidence of diarrhea in the population of weaned piglets, it is approximately 30–40% in general livestock production [40, 41]. However, in cases of bacterial infection, the percentage of piglets developing diarrhea can increase to nearly 80% [40]. Herein, we distinguished healthy and diarrheal pigs based on fecal diarrheal rates, pathogen abundance, and total BAs contents (Table. S1). Diarrheal pigs exhibited excessive total BAs, highly secondary BAs contents and conjugated BAs (Fig. 1A–E and S1A), accompanied with decreasing FXR expression, inflammation stimulation, and cell apoptosis in the colon (Fig. 1F and G). Further histomorphology analysis revealed colonic mucosal injury, inflammatory infiltration, and abnormal morphology of colonic glands in diarrheal pigs (Fig. 1H and I). These results are agreeing with previous studies on disorders of BAs, indicated diarrheal pig model establishment is appropriate for exploration of pathogenesis in BA disorder diarrhea; therefore, we labeled pigs with diarrhea as bile acid disorder diarrhea (BADD) pigs in the following discussion. Volcano plot showed a biomarker metabolite occurrence in diarrheal donor chyme, which is CDCA-3β-glucuronide (Fig. 1 K and L). According to previous studies on bile acid metabolism, CDCA is enable to be metabolized into various BAs by different enzymes (Fig. 1J), of which UGT1A4 is responsible for CDCA-3β-glucuronide biosynthesis and exists high content in the colon of BADD pigs (Fig. 1M). Other CDCA metabolites of this metabolic pathway were all no significant changes (Fig. S1B). The spearman analysis further indicated UGT1A4 and CDCA-3β-glucuronide contents were correlated with FXR expression, fecal total BAs level, BAX expression, and gut jury index (Fig. 1N-P and S1C-F). Similarly, we established a BA disorder mouse model by microbiota transplantation from diarrheal donors that also showed the less FXR level and highly UGT1A4 enzyme level in the colon (Fig. S1G–J). These data indicated the UGT1A4 mediated excessive CDCA-3β-glucuronide biosynthesis might be associated to bile acid disorder induced diarrhea.

To evaluate the effects of CDCA-3β-glucuronide and UGT1A4 in BA disorders induced diarrhea phenotype, we injected CDCA-3β-glucuronide to wide type (WT) mice and UGT1A4 inhibitor to FMT BADD mice, respectively (Fig. 2A and H). Mice received CDCA-3β-glucuronide that showed diarrheal situation by fecal water index detection (Fig. 2B). Moreover, CDCA-3β-glucuronide was seen to increase inflammation and apoptosis in mice, which histomorphology also indicated the disappearance of goblet cells and strange morphology of colonic glands (Fig. 2C–G). In vitro, small quantity of CDCA-3β-glucuronide and UGT1A4 recombinant enzyme were enough to cause cell viability to decrease, and which is no obvious different in different concentrations of CDCA supplement (Fig. S2A–C). Additionally, UGT1A4 is capable of increasing inflammatory factors by transforming CDCA into CDCA-3β-glucuronide in vitro (Fig. S2D–E). Both CDCA-3β-glucuronide and the co-treatment of CDCA and UGT1A4 RE increased the activities of caspase-1 and caspase-3 in gut epithelial cells (Fig. S2F). CDCA-3β-glucuronide significantly increased cell apoptosis rate compared to the control group and the CDCA group. Similarly, co-treatment of CDCA and UGT1A4 RE also induced cell apoptosis compared to the control group and the CDCA group (Fig.S2G). These results indicate that CDCA-3β-glucuronide induces cell apoptosis. UGT1A4 inhibitor significantly reduced the concentration of CDCA-3β-glucuronide in the colonic chyme, compared to FMT BADD mice (Fig. S2H). Besides, UGT1A4 depletion ameliorated the dramatically harmful changes in colonic morphology by reducing CDCA-3β-glucuronide synthesis (Fig. 2H–J). Furthermore, inflammatory factors and apoptosis level were largely reduced and tight junction protein Occludin was enhanced by UGT1A4 inhibitor in WT mice and FMT BADD mice, respectively (Fig. 2K and S2I–K). Additionally, immunofluorescence for nucleus location of FXR exhibited UGT1A4 which inhibited the FXR activation by producing CDCA-3β-glucuronide (Fig. 2L and M). Together, it demonstrated that UGT1A4 had an inescapable responsibility in development of BA disorder induced diarrhea by producing excessive CDCA-3β-glucuronide.

LKB1 regulated by FXR is associated with apoptosis in BA disorder diarrhea

To determine which is the target gene regulated by decreasing FXR during the BA disorder diarrhea process, we isolated the primary colonic epithelium cells from healthy and BA disorder induced diarrheal pigs to transcriptome sequencing (Fig. 3A). Importantly, the relative mRNA levels of TNF-α and IL-1β was increased in primary colonic epithelium cells of diarrheal weaned piglets (Fig. S3A); these findings were consistent with the NFκB-p65 activation observed in the colonic tissue of weaned piglets (Fig. 1F and G). Moreover, the UGT1A4 content was significantly enhanced in gut epithelium cells isolated from diarrheal piglets (Fig. S3B); this observation aligns with the UGT1A4 content in colon tissue as shown in Fig. 1M. These both reflected the same situation between primary colonic epithelium cells and tissue, indicating that the RNA-seq of primary gut epithelium cells could reflected the genes expression in colonic tissue of weaned piglets. The genes of primary colonic epithelium cells between the two group donors show a difference (Fig. S3C, D). Venn plot showed there are 11,201 genes that are shared between the control group and diarrhea group. Besides, 233 and 249 unique genes were detected in the healthy group and diarrheal group in primary gut epithelium cells between healthy and BADD piglets, respectively (Fig. S3E). We initially selected genes with significant changes in the front differential analysis based on fold change and P value. Our aim was to identify genes that are normally expressed in the healthy group but show reduced expression in the bile acid disorder diarrheal (BADD) group. Considering the potential role of FXR in alleviating cell apoptosis, the next step was to explore which of these chosen downregulated genes could be regulated by FXR. To investigate FXR regulation, we employed CDCA (an FXR activator (Fig. 3B)) and examined the expression change of upon FXR activation in the case of TNF-α challenge, the method of regulated genes chosen is based on previous researches [42, 43]. Of significantly distinguishing genes, which could be regulated by FXR in inflammatory situation that exactly is LKB1, which is deficiency in BA disorder diarrheal pigs (Fig. 3C, D, and S3F). There were not obvious and eligible changes in other six genes (Fig. S3G) [44,45,46]. In transcriptome sequencing of human study, which is important, IBS-D patients were also seen to lack LKB1 expression in gut epithelium, which deficiency was improved after therapy (Fig. 3E). Besides, several bacteria-related signaling and NF-κB signaling were enriched of BADD group shown in GSEA plots, which were all satisfied |Normalized-ES|> 1 and nominal P value < 0.05 (Fig.S3F).

To explore this causal role between LKB1 and apoptosis of gut epithelial cell, we used 20 μM TNF-α to simulate inflammatory environment of diarrhea (Fig. S4A). FXR activation (CDCA treated for 24 h) increased LKB1 protein expression to reduce apoptosis-associated protein BAX/BCL2 ratio and against TNF-α (treated for 24 h) caused apoptosis (Fig. 3F, G, and S4B, C). On the contrary, knocking down FXR decreased the expression of LKB1, increased the activities of caspase-1 and caspase-3, and caused cell apoptosis (Fig. 3H, I, and S4D-F). In addition, LKB1 siRNA treatment revealed that LKB1 is a negative regulator of cell apoptosis, exhibited by treating with LKB1 siRNA increased activities of caspase-1 and caspase-3 and number of apoptosis cells (Fig. 3J, K and Fig. S3, H). Therefore, the data indicated the important role of LKB1 regulated by FXR in the gut to fight apoptosis.

FXR transcriptional controls SIRT1 is necessary for LKB1 deacetylation

To investigate how FXR regulates LKB1 in BA disorder diarrhea, we further explored the reasons of several deacetylated-associated reactions which were enriched between healthy and BADD pigs (Fig. 4A). Acetylated-lysine pull down revealed that deacetylated-associated reaction was lesser occurred in LKB1 protein under the inflammation, compared with FXR activation (Fig. 4B). In human study, SIRT1 mRNA has significant changes in the gut between IBS-D patients and IBS-D-recovered subjects, which is also shown in gut epithelial cells of pigs (Fig. 4C–G). There were no prominent changes in other deacetylases (Fig. S4I and J). Expectedly, FXR knocking down has altered SIRT1 expression, and SIRT1 siRNA treatment is capable of regulating LKB1 expression (Fig. 4H–K, S4K and L). Further ChIP assay showed that FXR is binding to the DNA promoter of SIRT1 mRNA in response to fight with TNF-α (Fig. 4Land S4M). Therefore, it uncovered the suppression of FXR-SIRT1-LKB1 signal under the inflammation during the processes of BA disorder induced diarrhea on host.

LKB1 binds with P53 in the nucleus is necessary to fight with inflammatory induced apoptosis

Notably, we are still unable to explain the mechanism of influence on gut epithelial cell apoptosis regulated by LKB1 in BA disorder induced diarrhea. Therefore, we predicted the LKB1 with the several apoptosis associated protein based on protein–protein interaction networks (PPI) (https://cn.string-db.org), and found that the P53 (called TRP53 in mice, and called TP53 in humans and pigs) seem to be the bridge to connect the LKB1 with other apoptosis protein in humans, pigs, and mice (Fig. 5A). GSEA analysis of LKB1 and GO analysis reinforced this conjecture that decrease of P53 activity has played a role in BADD pigs (Fig. S5A–C). The docking of protein-to-protein simulated the interactions between LKB1 and P53 in mice and pigs, which docking pLDDT scores are nearly approximate 70, and the pTM scores are 0.446 and 0.452, respectively (Fig. S5D and E). Meanwhile, we speculated LKB1 regulated the transcriptional control ability of P53, affecting downstream apoptosis genes, then we measured the P53 associated apoptosis genes APAF-1, PUMA, and NOXA [47, 48], in the inflammation or treating alongside with FXR agonist, and observed the increasing gene levels of APAF-1, PUMA, and NOXA in the inflammatory situation but reduced after FXR activation (Fig. 5B). Based on above results and prediction, we further pull down the LKB1 protein or P53 protein to measure the binding intensity of both. CO-IP results showed the LKB1 had a lesser binding with P53 under the inflammation induced apoptosis than which after activation of FXR, suggesting that LKB1 bound with P53 to regulate the activity of P53 so that affecting the gene levels of APAF-1, PUMA, and NOXA (Fig. 5C). Co-immunofluorescence (Fig. S6A) and respective-immunofluorescence (Fig. 5D and E) localization further emphasized the location of LKB1 had effects on P53 activity in the nucleus after being received the activation of FXR, and protein intensity of nucleus. In the control group, LKB1 was observed to be expressed both inside and outside the nucleus. Similarly, P53 was detected in both the nucleus and cytoplasm. Notably, the expression of P53 in the nucleus was found to be weaker in the control group compared to the TNFα-treated group. Upon TNFα treatment, the expression of LKB1 decreased, and it localized predominantly outside the nucleus. However, when FXR was activated (CDCA), LKB1 translocated into the nucleus and exhibited expression therein. Regarding P53, TNFα treatment led to its activation and subsequent translocation into the nucleus. However, upon FXR activation, the expression of P53 in the nucleus decreased, and it was predominantly observed in the cytoplasm, these results were similar to the P53 protein expression inside and outside nucleus (Fig. 5F). In the GSEA of LKB1, |Normalized-ES| of regulation of P53 activity by phosphorylation and total regulation of P53 were greater than 1, and nominal p value of which were < 0.050 and 0.077, both reinforced this conclusion (Fig. 5G and H). More in more, LKB1 was eliminated induced upregulation of P53 activity in the nucleus, and LKB1 siRNA co-treated with TNF-α could induce the more server influence of apoptosis by increasing P53 activity than that in TNF-α treating alone (Fig. 5I–K, S6B). These data suggested that LKB1 bound with P53 in the nucleus to against cell apoptosis on the host. Additionally, we verified that this host mechanism was enhanced and strengthened the gut barrier function in healthy pigs than those in BA disorders diarrheal pigs (Fig. S6C–F).

FXR agonist alleviates BA disorders diarrhea through FXR-SIRT1-LKB1 axis in mice

We attempted to identify whether activation of FXR could ameliorate BA disorder induced diarrhea through FXR-SIRT1-LKB1 signal pathway in mice; we had established different kinds of mice model. Firstly, fecal microbiota transplantation of BADD pigs to antibiotic mixture (ABX)-treated mice decreased the body weight and colonic length, increased the diarrheal index, total bile acid concentrations, and gut UGT1A4 enzyme content (Fig. S7A–G). Histomorphology and immunohistochemistry indicated that the location and expression of FXR were changed by microbiota from BADD donors than those in healthy donors, and length of mucosa of FMT BADD mice was thickened, and the morphology of gut glands was mutated, accompanied by inflammatory infiltration (Fig. S7H, I). Besides, the intensity of FXR-SIRT1-LKB1 signal pathway we had proposed was stimulated after FXR agonist GW4064 supplement in FMT BADD mice (Fig. S7J and K). Especially, FMT model from donors with BADD, we observed an increase in acetylation of LKB1 compared to the FMT control group. However, treatment with an FXR agonist reduced the enhanced acetylation of LKB1 induced by FMT from BADD donors (Fig. S7L). Moreover, interleukin 6 (IL-6), BAX, APAF-1, PUMA, and NOXA gene levels were measured to enhance in FMT BADD mice; interleukin 4 (IL-4) and BCL2 were differently reduced; however, those phenotypes were all ameliorated by FXR agonist (Fig. S7M-N). More than that, activation of FXR intensified gut barrier function, and decreased the abundance of pathogenic bacteria, latter which was positively correlated with the UGT1A4 enzyme contents in the gut (Fig. S7O and P).

Dextran sulfate sodium salt (DSS) is a classic model for inducing acute or chronic colonic inflammation in mice [49]. Recently, reports proposed that DSS induced colitis was exacerbated through activating UGT enzymes and changes of BA composition [18], suggesting that DSS challenge could be a suitable model for research on UGT enzymes mediated BA disorders. To explore the important role of LKB1 on UGTs induced BA dysbiosis, we established the LKB1 depletion model, with the phenotypes of colonic epithelial LKB1 gene depletion, cell apoptosis, tissues inflammatory infiltration, and increasing abundance of E. coli (CFU) in feces (Fig. S8A–D). In DSS mice, we firstly examined the degree of reduction of body weight, colonic length, injuring degree of tissue morphology, colonic stasis level of total bile acids, and gut UGT1A4 enzyme contents, finding that DSS mice had an enhancing level of UGT1A4 enzyme and dysbiosis of bile acid homeostasis with serious colitis (Fig. 6A–E and S8E, F). Moreover, LKB1 depletion exacerbated the DSS induced increasing UGT1A4 level and excessive bile acids in the colon, accompanied with the degree of weight loss (from 15 to 21 days) and the inflammation significantly severer than DSS-treated mice (Fig. 6A–E and S8G–K). Lipopolysaccharide (LPS) is the membrane of Gram-negative bacteria like E. coli that could entranced to serum and other visceral organs to cause inflammation and damage [50]. Here, we found amount of E. coli colonization in the excrement of DSS mice, which were much larger abundance in the situation of LKB1 depletion, serum level of LPS exhibited an analogous tendency (Fig. S8L and M). Correspondingly, mice received DSS challenge showed a grievous disease activity index (DAI) (criteria of DAI showed in Table. S3) and high proportion of organ weight ratio (Fig. S8N and O). Besides, we also investigated the restoration extent of FXR agonist for intensity of FXR-SIRT1-LKB1-P53 signaling resist to BA disorder induced inflammation. The nucleus location of FXR and LKB1 is detected by immunohistochemical score, indicating that the FXR agonist were able to restore the translocation and expression of FXR and LKB1 induced by DSS (Fig. 6F and G). Furthermore, LKB1 deficiency with DSS challenge induced intensify/area ratio of LKB1 reduced, but no additional alternation of FXR, compared to DSS treating alone (Fig. 6F and G). Moreover, DSS challenge increased the acetylation level of LKB1 compared to control mice. Treatment with an FXR agonist reduced the enhanced acetylation of LKB1 induced by DSS. Additionally, co-challenging with DSS and an LKB1 inhibitor increased the acetylation level of LKB1, but there was no significant difference compared to the DSS challenge alone (Fig. S8P). Under the inflammation caused by DSS, mice received challenge happened to be cell apoptosis in the gut that measured by TUNEL assay, BAX/BCL2 protein expression, and p-P38/P38 signaling pathway intensity (Fig. 6H, I, M,andN), which also reflected in P53 signaling activation and increased of P53 associated gene levels (Fig. 6O and S8Q). These injuries were ameliorated by FXR agonist; however, FXR agonist had no significant improvement in changes of body weight decrease induced by DSS challenge; it indicated that improvement of gut injury by FXR activation might not be enough to cure the decrease of body weight induced by DSS challenge (Fig. S8I). The DSS induced-gut injury became more severer, in LKB1 depletion case, demonstrating the important role of stimulation of FXR-SIRT1-LKB1-P53 signaling to against BA metabolism disorders (Fig. 6I–L), which could improve gut barrier function by increasing the relative gene levels of ZO-1, Claudin-1, and Occludin (Fig. S8R). These data exhibited the crucial role of LKB1 activation in improvement of UGTs mediated BA metabolism disorder. Collectively, activating FXR in vivo could improve the UGT1A4 mediated BA metabolism dysbiosis, inflammatory-induced apoptosis, and gut injury through the FXR-SIRT-LKB1 axis on the host.

Suppression of UGT1A4 enzyme by Lactobacillus reuteri contributes to alleviate BA disorder diarrhea

Gut microbiota plays a crucial role in the occurrence and treatment of BA disorder induced gut diseases through its structure, metabolites, probiotic/pathogen ratio, and ability of regulation in BA composition [23, 51]. However, we were still unable to explain the effects and mechanism of gut flora on UGT1A4-mediated bile acid disorder induced diarrhea. Herein, we analyzed the gut flora structure and composition between healthy and BADD pigs. Shannon index and observed feature of alpha diversity analysis showed the different richness between different group (Fig. 7A and S9A). NMDS of beta diversity and Venn plot showed that there was a distinguish microbiome in healthy and BADD group (Fig. 7B and S9B). Further bacterial proportion analysis at genus level revealed the Lactobacillus and Pretovella had a relatively high proportion in healthy group, and these microbes were low proportion in BADD group. Conspicuously, Lactobacillus was in a main proportion in healthy group. Instead, Sharpe, Gemmiger, and Escherichia were in the high proportion of BADD group, whereas the proportion of Lactobacillus is unimpressive (Fig. 7C). Moreover, KEGG enrichment uncovered that the ability of bile acid metabolism in the BADD group is inferior to healthy group (Fig. S9C). Lefse exploration further analyzed 17 strains of differential flora, of which Lactobacillus reuteri (L. reuteri) and Lactobacillus helveticus (L. helveticus) were the most domainer and distinguished probiotic (Fig. 7D–F). Pathogens like E. coli, Clostridia formicllis, Pseudomonas aerogen (P. aerogens), and Lactobacillus salivarius were significantly negatively related to UGT1A4 enzyme and glucuronidation-associated enzyme levels (Fig. 7E and S9D); several bacterial infective epitheliums signaling and biofilm biosynthesis of pathogenic bacteria is enriched synchronously in microbiota communities of BADD colon, which were positively related to the changes of UGT1A4 level (Fi. S9E and H). It demonstrated the pathogenic flora might be responsible for causing highly UGT1A4 enzyme content in the gut. Differently, the L. reuteri, L. helveticus, Pretovella copri (P. copri), and Limosilactobacillus mucosae (L. mucosae) were both negatively related to UGT1A4 enzyme contents and concentration of CDCA-3β-glucuronide (Fig. 7E and G). To find the core probiotic to targeted inhibit the function and level of UGT1A4 enzyme, gut epithelial cell was cocultured with L. reuteri, L. helveticus, P. copri, and L. mucosae respectively, to measure the level of UGT1A4 (Fig. 7H, I, S9F and S9G). The results indicated that L. reuteri is able to reduce the UGT1A4 enzyme content in vitro; L. helveticus, P. copri, and L. mucosae had no effects on UGT1A4 expression (Fig. S9F and G). Similarly, 10⁹ CFU/mL oral gavage of L. reuteri in mice showed decreased gene level of UGT1A4 (mice UGT1A5 gene is homology to human UGT1A4), which alternation of UGT1A4 is negatively correlated to L. reuteri colonized abundance in the gut (Fig. 7J–L). Meanwhile, the relative abundance of L. reuteri between healthy and BADD group is negatively related to the signaling pathway of bacterial infection or pathogenic biofilm biosynthesis (Fig. S9E). Above data revealed the L. reuteri has ability to fight with pathogens induced UGT1A4 enhancement.

To investigate the role of L. reuteri in improvement of UGT1A4-induced inflammation and apoptosis, UGT1A4 recombinant enzyme (UGT1A4 RE) was used to injured gut epithelial cells and cultured alongside with L. reuteri. The data showed UGT1A4 RE resulted in inflammation by changing the nucleus location of NF-κB-P65, and finally caused TNF-α and BAX expression intensity enhanced. These inflammation and apoptosis phenotypes were observed to ameliorate after appending L. reuteri, which further increases the level of Occludin (Fig. 7M-O). Similarly, microbiota-enriched pathways were also found apoptosis signaling even P53 signaling (Fig. 7P). Collectively, it demonstrated that L. reuteri has the beneficial ability to recover the inflammation and apoptosis affected by UGT1A4 enzyme.

I3C derived by Lactobacillus reuteri improved UGT1A4-mediatedepithelialinjury by altering location of AHR

Previous research reported that AHR had ability to transcriptional control UGT1A families [52], and UGT1A4 could be transcriptionally regulated by AHR [14]. The results of transcriptome sequencing indicated that AHR was significantly reduced in gut epithelial cells of BADD group, while PXR, PPAR, GR, or other nuclear receptors that regulate UGTs were not significantly different (Fig. 8A). Gut epithelial cells received AHR antagonist significant decreased UGT1A4 content (Fig. 8B). To investigate the mitigating mechanism of L. reuteri in inflammation and apoptosis induced by UGT1A4 enzyme. We further measured the metabolites of L. reuteri to find a certain molecule could ameliorate BA disorder induced diarrhea in mice (Fig. 8C). Oral gavage with could decreased the fecal water index and total bile acid content induced by microbiota transplantation (Fig. S10A and B). PCOA of metabolomics showed that the metabolites were obviously different among FMT CON, FMT BADD, and FMT BADD + L. reuteri groups (Fig. 8D). Venn plot analyzed the diverse metabolites at the range of VIP > 1 and P value < 0.05, and found that I3C is the most significant metabolite derived by L. reuteri and enable to be the agonist of AHR (Fig. 8E). Subsequently, we used I3C to against AHR antagonist and UGT1A4 RE induced changes of location and expression of AHR (red) and UGT1A4 (green) in gut epithelial cells, and found I3C were capable of improving AHR inhibition and UGT1A4 upregulation (Fig. 8F–H). Therefore, we considered the L. reuteri reduced the content of UGT1A4 in the host by deriving I3C. Further oral administration of I3C alone or I3C co-supplement with CDCA to FMT mice indicated that the inflammatory infiltration decreased, goblet cell morphology integrity, intestinal gland structure intact (Fig. 8K and S10C), accompanied with decreased of fecal total BA concentration, fecal water index, and UGT1A4 content, compared to FMT BADD mice (Fig. S10D-F). Collectively, I3C supplement in vivo could ameliorate BA disorder induced diarrhea and gut injury, and with the existence of I3C derived by L. reuteri that CDCA supplement.

Discussion

Our study has demonstrated the primary pathogenesis of BA disorders-induced diarrhea. This condition arises due to dysbiosis of CDCA metabolism, with CDCA being excessively glucuronidated in the colon in the absence of L. reuteri, the majority of CDCA is converted to CDCA-3β-glucuronide by the UGT1A4 enzyme in colonic epithelial cells. CDCA-3β-glucuronide plays a crucial role in repressing FXR activity, which inhibits the FXR-SIRT1-LKB1 signal pathway in gut epithelial cells. Consequently, these damage lead to the activation of P53 and subsequent apoptosis of gut epithelial cells, resulting in impaired gut barrier function and BA disorder-induced diarrhea. Importantly, L. reuteri produces the metabolite I3C that facilitates crosstalk between L. reuteri and gut epithelial cells by altering AHR location to enter the nucleus. This, in turn, transcriptionally suppresses UGT1A4 mRNA expression, reducing the synthesis of CDCA-3β-glucuronide in the host. Therefore, the presence of L. reuteri and its derived metabolite I3C are essential for preventing CDCA-3β-glucuronide-induced BA metabolic disorders diarrhea. In conclusion, our study has revealed a previously unknown pathogenic mechanism and a potential strategy for mitigating diarrhea induced by BA disorders (Fig. 9).

Overwhelming studies have demonstrated a reduction in Lactobacillus levels in the gut of patients with diarrhea, IBD, and other gastrointestinal diseases [23, 53, 54]. Previous research has linked the absence of Lactobacillus gasseri and Lactobacillus rumen, along with their bacteriocin secretion, to early weaned diarrhea [55]. In mice with colitis and diarrhea, the absence of Lactobacillus johnsonii led to decreased expression of goblet cells and sulfomucin, exacerbating inflammation [56]. Furthermore, an adjuvant trial suggested that Lactobacillus supplementation had a potential alleviating effect on antibiotic-associated diarrhea [57]. Our findings highlight the crucial role of L. reuteri deficiency in causing disturbances in BA metabolism and subsequent diarrhea, leading to various intestinal damages, such as inflammation and apoptosis. Specially, the decrease in L. reuteri resulted in a shift toward glucuronidation in the BA metabolic pathway, converting CDCA into CDCA-3β-glucuronide in the colon and triggering apoptosis of gut epithelial cells. CDCA is a potential beneficial BA known to activate FXR and provide resistance against pathogenic bacteria infections [58,59,60]. Moreover, oral administration of CDCA alone has been shown to improve gut development, reduce gut barrier permeability, and enhance animal growth performance [11]. However, when BA metabolism is disrupted, CDCA interferes with FGF signaling and its ability to improve colon inflammation becomes unstable, for reasons that are not yet fully understood. Interestingly, we observed that the beneficial role of CDCA in colitis mitigation is generally associated with the presence of Lactobacillus in the gut [10], highlighting the importance of considering the contribution of Lactobacillus in CDCA influence and application. In our study, we discovered that L. reuteri produces a metabolite I3C, which acts as an activator of AHR [61, 62]. The level of I3C was significantly reduced in the case of diarrhea caused by the disturbance in BA metabolism, accompanied by decreased expression of AHR in gut epithelial cells and inhibition of their entry into the nucleus. Mice receiving either L. reuteri or I3C orally demonstrated a remission phenotype for BA disorder-induced diarrhea. Overall, these findings uncover that the pathogenesis of BA disorder-induced diarrhea, highlighting the important effects of L. reuteri and it derived I3C on activating AHR in gut epithelial cells.

The gut microbiota produces various small molecules that interact with or enter gut epithelial cells, either through their own production or by metabolizing other substances. These molecules can activate nuclear receptors or directly bind to specific proteins, thereby regulating the physiological functions of the host [63]. One such important receptor is AHR, which is a ligand-activated transcription factor found widely in the mammalian gut. AHR integrates signals from the gut microbiota and metabolism, initiating the transcription of target genes that are dependent on ligands, cells, and the environment. These genes include drug metabolism-related enzymes, thereby regulating many physiological and pathological processes [64]. In our study, we found that AHR received a signal from gut microbes in the form of I3C derived from L. reuteri, facilitating the crosstalk between gut microbes and gut epithelial cells. This microbial signal caused AHR to repress the transcription of the UGT1A4 enzyme, a member of the UGT family responsible for glucuronidation of CDCA to generate CDCA-3β-glucuronide. CDCA-3β-glucuronide, in turn, reduced cell viability, promoted the production of pro-inflammatory factors, and contributed to BA disorder-induced diarrhea. Interestingly, UGT1A4 also directly impaired cell viability and reversed the anti-inflammatory effects of CDCA and FXR activation by converting CDCA into CDCA-3β-glucuronide. Conversely, mice injected with a UGT1A4 inhibitor in the colon exhibited decreased expression of pro-inflammatory factors IL-6 and interleukin-1 (IL-1β), leading to the improvement of BA disorder-induced diarrhea. Additionally, L. reuteri reduced the expression of recombinant UGT1A4 and the activation of the inflammatory signal NFκB-P65 in vitro. Furthermore, L. reuteri directly alleviated BA disorder-induced diarrhea caused by fecal microbiota transplantation in mice. Notably, I3C derived from L. reuteri decreased the fluorescence intensity and expression of UGT1A4 in gut epithelial cells. Previous study has indicated that UGT family members are regulated by nuclear receptors [14], and similarly, our findings demonstrated that AHR activation transcriptionally inhibited UGT1A4 in gut epithelial cells. We showed that an AHR antagonist suppressed the nuclear localization of AHR, thereby promoting an increase in UGT1A4 contents. In contrast, I3C stimulated AHR to counteract the enhancement of colonic UGT1A4 in mice and UGT1A4-induced BA disorder-associated diarrhea. Importantly, the presence of I3C in the colon protected against excessive glucuronidation-induced gut injuries by diverting CDCA supplementation away from abnormal metabolic pathways. These findings reveal the pivotal role of L. reuteri in maintaining the homeostasis of CDCA metabolism and provide new insights into the crosstalk between L. reuteri and the gut epithelium. This crosstalk occurs through the production of I3C, which enters enterocytes and activates AHR, leading to the transcriptional repression of UGT1A4.

The occurrence of BA metabolism disorder-induced diarrhea is generally associated with the suppression of FXR signals [65]. Our findings provide insight into the relationship between BA disorder-induced diarrhea and the inhibition of FXR in gut epithelial cells, which leads to a decrease in the expression of LKB1 protein. LKB1 plays a crucial role in regulating various cellular processes, including inflammation, apoptosis, pyroptosis, and intestinal stem cell development [27, 66, 67]. Here, we demonstrate that gut epithelial LKB1 is essential to resist inflammation and apoptosis caused by BA disorder-induced diarrhea. Direct downregulation of LKB1 expression or inhibition of FXR, resulting in decreased levels of LKB1, led to apoptosis in gut epithelial cells. Furthermore, mice injected with an LKB1 inhibitor exhibited suppressed colonic epithelial LKB1, which aggravated inflammation, apoptosis, and the colonization of pathogenic bacteria. These results highlight the important effects of LKB1 on determining the fate of gut epithelial cells. In addition to FXR, LKB1 is also subject to regulation through epigenetic modifications, such as de-/phosphorylation and de-/acetylation, as there are modification sites present on the LKB1 protein [68, 69]. Notably, several de-/acetylation related reactions were found to be enriched in the gut during BA disorder-induced diarrhea. Through the detection of proteins pulled down by a pan-acetylation antibody, we observed increased levels of acetylation in LKB1 protein during inflammation, which may be attributed to a decrease in the deacetylase SIRT1. SIRT1 is an NAD⁺-dependent deacetylase involved in various biological processes, including metabolism, the immune response, and inflammation [70]. Analysis of previous transcriptome sequencing data from patients with IBS-D, a condition associated with BA diarrhea, revealed a significant reduction in SIRT1 mRNA levels in the mucosal epithelium of IBS-D patients compared to those who had recovered. Additionally, exposure to pro-inflammatory factors decreased SIRT1 levels in gut epithelial cells. Similarly, both pro-inflammatory factors and knockdown of SIRT1 downregulated LKB1 expression, leading to apoptosis, indicating that the deacetylation level of LKB1 is closely linked to its ability to resist apoptosis. Whether the deacetylation was also regulated by FXR remains unclear. Previous studies have shown that FXR regulates SIRT1 through miR-34a to improve nonalcoholic fatty liver disease (NAFLD) [71]. Here, we provide the first evidence that the abundance of potential FXR binding sites on SIRT1 DNA promoters is significantly reduced during inflammation. However, the utilization of FXR agonists increases the binding enrichment between FXR and SIRT1 DNA promoters, thereby counteracting the decrease in LKB1 expression and apoptosis. Obviously, the FXR/SIRT1/LKB1 axis is downregulated in the gut epithelium of mice or pigs suffering from UGT1A4-induced BA disorder-induced diarrhea. Overall, the mechanism by which CDCA-3β-glucuronide impacts gut epithelium involves the repression of FXR activity, resulting in a reduction in the FXR/SIRT1/LKB1 signaling pathway and subsequent apoptosis.

Inflammation is a crucial trigger for cell apoptosis, and P53 is a well-known nuclear transcriptional regulator that plays a key role in regulating cell cycle, apoptosis, and genome stability [72]. For apoptosis, P53 controls downstream BCL family genes that are associated with apoptosis [73, 74]. Our findings indicate that P53 signal is recruited in the host experiencing BA disorder-induced diarrhea but inhibited in the gut with an abundance of L. reuteri. Previous research has shown that P53 protein has multiple modification sites, such as acetylation, methylation, ubiquitination, and phosphorylation. Different modifications at these sites alter the target genes of P53 and affect transcription levels, thereby leading to different cell fates in terms of apoptosis, cell cycle arrest, and DNA damage [75], whereas we observed that P53 regulation of apoptosis occurs at the protein level through interaction with LKB1 protein in the nucleus rather than modification of P53, in the case of BA disorder-induced diarrhea. During inflammation, P53 was activated and localized in the nucleus, while LKB1 remained in the cytoplasm, resulting in fewer interactions between LKB1 and P53. While we have identified the binding relationship between LKB1 and P53, the direct relationship with LKB1 acetylation remains complex, there are three possible scenarios regarding the interplay between LKB1 acetylation, its binding to P53, and the TNFα-induced inflammatory situation: (1) In the TNFα-induced inflammatory condition, we observed a decrease in LKB1 expression and an increase in LKB1 acetylation. This suggests that the binding between LKB1 and P53 may be independent of LKB1 acetylation status. (2) Previous studies have indicated that increased acetylation of a protein can potentially prevent its binding to another protein [76]. This could provide a possible explanation for LKB1 acetylation inhibiting its binding to P53. (3) Our results also showed that SIRT1 siRNA, which reduces LKB1 protein expression and inhibits LKB1 deacetylation. Binding levels of LKB1 and P53, SIRT1 expression, and LKB1 expression were reduced in TNF-α challenge. Additionally, previous research has suggested that LKB1 acetylation can promote protein expression decrease and degradation [77]. Other studies have demonstrated that acetylation of certain proteins can also lead to expression of protein decreased and protein degradation [78, 79]. Thus, a third possibility arises where acetylation modification induces a decrease in LKB1 protein expression in the TNFα-induced inflammatory situation, which in turn reduces the protein binding between LKB1 and P53.

Consequently, P53 had the opportunity to transcriptionally control the gene levels of NOXA, PUMA, BAX, and APAF-1. However, upon activation by FXR agonists, LKB1 translocated into the nucleus, competes for binding with P53, and exported a portion of P53 proteins outside the nucleus, thereby weakening the ability of P53 to regulate apoptosis associated genes (APAF-1, NOXA, PUMA, BAX, BCL2 etc. [47, 48]). Moreover, knockdown of LKB1 and LKB1 inhibitor injection in mice exacerbated P53-induced apoptosis in the context of inflammation. In summary, we propose that deacetylation of LKB1 and the interaction between LKB1 and P53 reduce cell apoptosis and alter the fate of gut epithelial cells, due to the regulatory role of LKB1 through FXR activation in the occurrence of BA disorder-induced diarrhea.

Conclusion

Collectively, we provided a crucial pathogenesis of BA disorder-induced diarrhea, which involves the deficiency of L. reuteri and its derived I3C. This deficiency prevents the activation of gut epithelial AHR, leading to the accumulation of UGT1A4 enzyme in enterocytes. This accumulation results in the toxic glucuronidation of exogenous and endogenous CDCA, leading to the biosynthesis of the secondary metabolite CDCA-3β-glucuronide. This metabolite, in turn, represses FXR/SIRT1/LKB1 signaling pathway, ultimately leading to apoptosis. Furthermore, we proposed an intervention strategy to prevent BA disorder-induced diarrhea, which involves the supplementation of L. reuteri and its derived I3C. This intervention serves as a novel approach to improve the application of CDCA and mitigate the development of diarrhea.

Data availability

Data is provided within the manuscript or supplementary information files

References

Fu T, Coulter S, Yoshihara E, Oh TG, Fang S, Cayabyab F, et al. FXR regulates intestinal cancer stem cell proliferation. Cell. 2019;176:1098-112.e18. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cell.2019.01.036.
Article CAS PubMed PubMed Central Google Scholar
Sinha SR, Haileselassie Y, Nguyen LP, Tropini C, Wang M, Becker LS, et al. Dysbiosis-induced secondary bile acid deficiency promotes intestinal inflammation. Cell Host Microbe. 2020;27:659-70.e5. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.chom.2020.01.021.
Article CAS PubMed PubMed Central Google Scholar
Carrasco-Labra A, Lytvyn L, Falck-Ytter Y, Surawicz CM, Chey WD. AGA technical review on the evaluation of functional diarrhea and diarrhea-predominant irritable bowel syndrome in adults (IBS-D). Gastroenterology. 2019;157:859–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1053/j.gastro.2019.06.014.
Article CAS PubMed Google Scholar
Vich Vila A, Hu S, Andreu-Sánchez S, Collij V, Jansen BH, Augustijn HE, et al. Faecal metabolome and its determinants in inflammatory bowel disease. Gut. 2023;72:1472–85. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/gutjnl-2022-328048.
Article CAS PubMed Google Scholar
Zhao L, Yang W, Chen Y, Huang F, Lu L, Lin C, et al. A Clostridia-rich microbiota enhances bile acid excretion in diarrhea-predominant irritable bowel syndrome. J Clin Invest. 2020;130:438–50. https://doiorg.publicaciones.saludcastillayleon.es/10.1172/jci130976.
Article CAS PubMed Google Scholar
Ocvirk S, O’Keefe SJD. Dietary fat, bile acid metabolism and colorectal cancer. Semin Cancer Biol. 2021;73:347–55. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.semcancer.2020.10.003.
Article CAS PubMed Google Scholar
Pushpass RG, Alzoufairi S, Jackson KG, Lovegrove JA. Circulating bile acids as a link between the gut microbiota and cardiovascular health: impact of prebiotics, probiotics and polyphenol-rich foods. Nutr Res Rev. 2022;35:161–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1017/s0954422421000081.
Article CAS PubMed Google Scholar
Camilleri M, Vijayvargiya P. The role of bile acids in chronic diarrhea. Am J Gastroenterol. 2020;115:1596–603. https://doiorg.publicaciones.saludcastillayleon.es/10.14309/ajg.0000000000000696.
Article PubMed PubMed Central Google Scholar
Walters JR. Bile acid diarrhoea and FGF19: new views on diagnosis, pathogenesis and therapy. Nat Rev Gastroenterol Hepatol. 2014;11:426–34. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrgastro.2014.32.
Article CAS PubMed Google Scholar
Dong S, Zhu M, Wang K, Zhao X, Hu L, Jing W, et al. Dihydromyricetin improves DSS-induced colitis in mice via modulation of fecal-bacteria-related bile acid metabolism. Pharmacol Res. 2021;171: 105767. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.phrs.2021.105767.
Article CAS PubMed Google Scholar
Song M, Zhang F, Chen L, Yang Q, Su H, Yang X, et al. Dietary chenodeoxycholic acid improves growth performance and intestinal health by altering serum metabolic profiles and gut bacteria in weaned piglets. Animal Nutrition. 2021;7:365–75. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.aninu.2020.07.011.
Article CAS PubMed PubMed Central Google Scholar
Borup C, Wildt S, Rumessen JJ, Bouchelouche PN, Graff J, Damgaard M, et al. Chenodeoxycholic acid stimulated fibroblast growth factor 19 response—a potential biochemical test for bile acid diarrhoea. Aliment Pharmacol Ther. 2017;45:1433–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/apt.14056.
Article CAS PubMed Google Scholar
Münch A, Söderholm JD, Ost A, Carlsson AH, Magnusson KE, Ström M. Low levels of bile acids increase bacterial uptake in colonic biopsies from patients with collagenous colitis in remission. Aliment Pharmacol Ther. 2011;33:954–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/j.1365-2036.2011.04611.x.
Article PubMed Google Scholar
Meech R, Hu DG, McKinnon RA, Mubarokah SN, Haines AZ, Nair PC, et al. The UDP-glycosyltransferase (UGT) superfamily: new members, new functions, and novel paradigms. Physiol Rev. 2019;99:1153–222. https://doiorg.publicaciones.saludcastillayleon.es/10.1152/physrev.00058.2017.
Article CAS PubMed Google Scholar
Milne GL, Nogueira MS, Gao B, Sanchez SC, Amin W, Thomas S, et al. Identification of novel F(2)-isoprostane metabolites by specific UDP-glucuronosyltransferases. Redox Biol. 2024;70: 103020. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.redox.2023.103020.
Article CAS PubMed PubMed Central Google Scholar
Yang G, Ge S, Singh R, Basu S, Shatzer K, Zen M, et al. Glucuronidation: driving factors and their impact on glucuronide disposition. Drug Metab Rev. 2017;49:105–38. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/03602532.2017.1293682.
Article CAS PubMed PubMed Central Google Scholar
Erichsen TJ, Aehlen A, Ehmer U, Kalthoff S, Manns MP, Strassburg CP. Regulation of the human bile acid UDP-glucuronosyltransferase 1A3 by the farnesoid X receptor and bile acids. J Hepatol. 2010;52:570–8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jhep.2010.01.010.
Article CAS PubMed Google Scholar
Zhou X, Cao L, Jiang C, Xie Y, Cheng X, Krausz KW, et al. PPARα-UGT axis activation represses intestinal FXR-FGF15 feedback signalling and exacerbates experimental colitis. Nat Commun. 2014;5:4573. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/ncomms5573.
Article CAS PubMed Google Scholar
Jia W, Xie G, Jia W. Bile acid-microbiota crosstalk in gastrointestinal inflammation and carcinogenesis. Nat Rev Gastroenterol Hepatol. 2018;15:111–28. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/nrgastro.2017.119.
Article CAS PubMed Google Scholar
Jabbar KS, Dolan B, Eklund L, Wising C, Ermund A, Johansson Å, et al. Association between Brachyspira and irritable bowel syndrome with diarrhoea. Gut. 2021;70:1117–29. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/gutjnl-2020-321466.
Article CAS PubMed Google Scholar
Zhai L, Huang C, Ning Z, Zhang Y, Zhuang M, Yang W, et al. Ruminococcus gnavus plays a pathogenic role in diarrhea-predominant irritable bowel syndrome by increasing serotonin biosynthesis. Cell Host Microbe. 2023;31:33-44.e5. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.chom.2022.11.006.
Article CAS PubMed Google Scholar
Liebregts T, Adam B, Bredack C, Röth A, Heinzel S, Lester S, et al. Immune activation in patients with irritable bowel syndrome. Gastroenterology. 2007;132:913–20. https://doiorg.publicaciones.saludcastillayleon.es/10.1053/j.gastro.2007.01.046.
Article CAS PubMed Google Scholar
Lin Z, Wu J, Wang J, Levesque CL, Ma X. Dietary Lactobacillus reuteri prevent from inflammation mediated apoptosis of liver via improving intestinal microbiota and bile acid metabolism. Food Chem. 2023;404: 134643. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.foodchem.2022.134643.
Article CAS PubMed Google Scholar
Jiang P, Ren L, Zhi L, Yu Z, Lv F, Xu F, et al. Negative regulation of AMPK signaling by high glucose via E3 ubiquitin ligase MG53. Mol Cell. 2021;81:629-37.e5. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.molcel.2020.12.008.
Article CAS PubMed Google Scholar
Gao Y, Yan Y, Tripathi S, Pentinmikko N, Amaral A, Päivinen P, et al. LKB1 represses ATOH1 via PDK4 and energy metabolism and regulates intestinal stem cell fate. Gastroenterology. 2020;158:1389-401.e10. https://doiorg.publicaciones.saludcastillayleon.es/10.1053/j.gastro.2019.12.033.
Article CAS PubMed Google Scholar
Ma N, Chen X, Johnston LJ, Ma X. Gut microbiota-stem cell niche crosstalk: a new territory for maintaining intestinal homeostasis. iMeta. 2022;1:e54. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/imt2.54.
Article PubMed PubMed Central Google Scholar
Compton SE, Kitchen-Goosen SM, DeCamp LM, Lau KH, Mabvakure B, Vos M, et al. LKB1 controls inflammatory potential through CRTC2-dependent histone acetylation. Mol Cell. 2023;83:1872-86.e5. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.molcel.2023.04.017.
Article CAS PubMed Google Scholar
Panah FM, Lauridsen C, Højberg O, Jensen HE, Nielsen TS. Composition of mucus- and digesta-associated bacteria in growing pigs with and without diarrhea differed according to the presence of colonic inflammation. BMC Microbiol. 2023;23:145. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12866-023-02874-1.
Article CAS PubMed PubMed Central Google Scholar
Zhang S, Wang J, Liu X, Kan Z, Zhang Y, Niu Z, et al. Pemetrexed alleviates piglet diarrhea by blocking the interaction between porcine epidemic diarrhea virus nucleocapsid protein and Ezrin. J Virol. 2024;98: e0162523. https://doiorg.publicaciones.saludcastillayleon.es/10.1128/jvi.01625-23.
Article CAS PubMed Google Scholar
Zhang Z, Zhang G, Zhang S, Zhao J. Fructooligosaccharide reduces weanling pig diarrhea in conjunction with improving intestinal antioxidase activity and tight junction protein expression. Nutrients. 2022;14:512. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/imt2.54.
Article CAS PubMed PubMed Central Google Scholar
Zuo A, Zhou Y, Li Y, Zhang Y, Yi Z, Xiao Y, et al. Molecular characterization of LKB1 of triploid crucian carp and its regulation on muscle growth and quality. Animals. 2022;12:12. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ani12182474.
Article Google Scholar
Singh N, Yadav M, Singh AK, Kumar H, Dwivedi SK, Mishra JS, et al. Synthetic FXR agonist GW4064 is a modulator of multiple G protein-coupled receptors. Mol Endocrinol. 2014;28:659–73. https://doiorg.publicaciones.saludcastillayleon.es/10.1210/me.2013-1353.
Article CAS PubMed PubMed Central Google Scholar
Lien F, Berthier A, Bouchaert E, Gheeraert C, Alexandre J, Porez G, et al. Metformin interferes with bile acid homeostasis through AMPK-FXR crosstalk. J Clin Invest. 2014;124:1037–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1172/jci68815.
Article CAS PubMed PubMed Central Google Scholar
de Groot P, Scheithauer T, Bakker GJ, Prodan A, Levin E, Khan MT, et al. Donor metabolic characteristics drive effects of faecal microbiota transplantation on recipient insulin sensitivity, energy expenditure and intestinal transit time. Gut. 2020;69:502–12. https://doiorg.publicaciones.saludcastillayleon.es/10.1136/gutjnl-2019-318320.
Article CAS PubMed Google Scholar
Wu J, Wang J, Lin Z, Liu C, Zhang Y, Zhang S, et al. Clostridium butyricum alleviates weaned stress of piglets by improving intestinal immune function and gut microbiota. Food Chem. 2023;405: 135014. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.foodchem.2022.135014.
Article CAS PubMed Google Scholar
Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34:i884–90. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/bioinformatics/bty560.
Article CAS PubMed PubMed Central Google Scholar
Owens NDL, De Domenico E, Gilchrist MJ. An RNA-Seq protocol for differential expression analysis. Cold Spring Harbor Protoc. 2019; 2019. https://doiorg.publicaciones.saludcastillayleon.es/10.1101/pdb.prot098368.
Bauermeister A, Mannochio-Russo H, Costa-Lotufo LV, Jarmusch AK, Dorrestein PC. Mass spectrometry-based metabolomics in microbiome investigations. Nat Rev Microbiol. 2022;20:143–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41579-021-00621-9.
Article CAS PubMed Google Scholar
Lunney JK, Van Goor A, Walker KE, Hailstock T, Franklin J, Dai C. Importance of the pig as a human biomedical model. Sci Transl Med. 2021;13: eabd5758. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/scitranslmed.abd5758.
Article CAS PubMed Google Scholar
Liu W, Guo K. Tannic acid alleviates ETEC K88-induced intestinal damage through regulating the p62-keap1-Nrf2 and TLR4-NF-κB-NLRP3 pathway in IPEC-J2 cells. J Sci Food Agric. 2024;104:5186–96. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/jsfa.13343.
Article CAS PubMed Google Scholar
Wang C, Zhong Y, Liu H, Wang H, Li Y, Wang Q, et al. Effects of dietary supplementation with tea residue on growth performance, digestibility, and diarrhea in piglets. Animals. 2024;14:14. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ani14040584.
Article CAS Google Scholar
Feng Y, Shen J, Lin Z, Chen Z, Zhou M, Ma X. PXR activation relieves deoxynivalenol-induced liver oxidative stress via Malat1 LncRNA m(6)A demethylation. Advanced Science. 2024;11: e2308742. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/advs.202308742.
Article CAS PubMed Google Scholar
Zeng Y, Wu R, Wang F, Li S, Li L, Li Y, et al. Liberation of daidzein by gut microbial β-galactosidase suppresses acetaminophen-induced hepatotoxicity in mice. Cell Host Microbe. 2023;31:766-80.e7. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.chom.2023.04.002.
Article CAS PubMed Google Scholar
Lee J, Hong EM, Jung JH, Park SW, Lee SP, Koh DH, et al. Atorvastatin induces FXR and CYP7A1 activation as a result of the sequential action of PPARγ/PGC-1α/HNF-4α in Hep3B cells. Korean J Gastroenterol. 2021;77:123–31. https://doiorg.publicaciones.saludcastillayleon.es/10.4166/kjg.2020.156.
Article PubMed Google Scholar
Chu WC, Hong WF, Huang MC, Chen FY, Lin SC, Liao PJ, et al. 1-Nitropyrene stabilizes the mRNA of cytochrome P450 1a1, a carcinogen-metabolizing enzyme, via the Akt pathway. Chem Res Toxicol. 2009;22:1938–47. https://doiorg.publicaciones.saludcastillayleon.es/10.1021/tx900241g.
Article CAS PubMed Google Scholar
Liu H, Wei M, Tan B, Dong X, Xie S. The supplementation of berberine in high-carbohydrate diets improves glucose metabolism of tilapia (Oreochromis niloticus) via transcriptome, bile acid synthesis gene expression and intestinal flora. Animals. 2024;14:14. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ani14081239.
Article CAS Google Scholar
Aubrey BJ, Kelly GL, Janic A, Herold MJ, Strasser A. How does p53 induce apoptosis and how does this relate to p53-mediated tumour suppression? Cell Death Differ. 2018;25:104–13. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/cdd.2017.169.
Article CAS PubMed Google Scholar
Wei H, Qu L, Dai S, Li Y, Wang H, Feng Y, et al. Structural insight into the molecular mechanism of p53-mediated mitochondrial apoptosis. Nat Commun. 2021;12:2280. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-021-22655-6.
Article CAS PubMed PubMed Central Google Scholar
Dong L, Xie J, Wang Y, Jiang H, Chen K, Li D, et al. Mannose ameliorates experimental colitis by protecting intestinal barrier integrity. Nat Commun. 2022;13:4804. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41467-022-32505-8.
Article CAS PubMed PubMed Central Google Scholar
Törk L, Moffatt CB, Bernhardt TG, Garner EC, Kahne D. Single-molecule dynamics show a transient lipopolysaccharide transport bridge. Nature. 2023;623:814–9. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-023-06709-x.
Article CAS PubMed PubMed Central Google Scholar
Collins SL, Stine JG, Bisanz JE, Okafor CD, Patterson AD. Bile acids and the gut microbiota: metabolic interactions and impacts on disease. Nat Rev Microbiol. 2023;21:236–47. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41579-022-00805-x.
Article CAS PubMed Google Scholar
Kalthoff S, Winkler A, Freiberg N, Manns MP, Strassburg CP. Gender matters: estrogen receptor alpha (ERα) and histone deacetylase (HDAC) 1 and 2 control the gender-specific transcriptional regulation of human uridine diphosphate glucuronosyltransferases genes (UGT1A). J Hepatol. 2013;59:797–804. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jhep.2013.05.028.
Article CAS PubMed Google Scholar
Huang QX, Liang JL, Yang CH, Li K, Niu MT, Fan JX, et al. Stimulation-responsive mucoadhesive probiotics for inflammatory bowel disease treatment by scavenging reactive oxygen species and regulating gut microbiota. Biomaterials. 2023;301: 122274. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.biomaterials.2023.122274.
Article CAS PubMed Google Scholar
Liu C, Ma N, Feng Y, Zhou M, Li H, Zhang X, et al. From probiotics to postbiotics: concepts and applications. AROH. 2023;1:92–114. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/aro2.7.
Article Google Scholar
Hu J, Ma L, Nie Y, Chen J, Zheng W, Wang X, et al. A microbiota-derived bacteriocin targets the host to confer diarrhea resistance in early-weaned piglets. Cell Host Microbe. 2018;24:817-32.e8. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.chom.2018.11.006.
Article CAS PubMed Google Scholar
Ma N, Guo P, Chen J, Qi Z, Liu C, Shen J, et al. Poly-β-hydroxybutyrate alleviated diarrhea and colitis via Lactobacillus johnsonii biofilm-mediated maturation of sulfomucin. Science China Life Sciences. 2023;66:1569–88. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11427-022-2213-6.
Article CAS PubMed Google Scholar
Lukasik J, Dierikx T, Besseling-van der Vaart I, de Meij T, Szajewska H. Multispecies probiotic for the prevention of antibiotic-associated diarrhea in children: a randomized clinical trial. JAMA Pediatrics. 2022;176:860–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jamapediatrics.2022.1973.
Article PubMed PubMed Central Google Scholar
Xu H, Fang F, Wu K, Song J, Li Y, Lu X, et al. Gut microbiota-bile acid crosstalk regulates murine lipid metabolism via the intestinal FXR-FGF19 axis in diet-induced humanized dyslipidemia. Microbiome. 2023;11:262. https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-023-01709-5.
Article CAS PubMed PubMed Central Google Scholar
Qin D, Liu S, Lu Y, Yan Y, Zhang J, Cao S, et al. Lgr5 (+) cell fate regulation by coordination of metabolic nuclear receptors during liver repair. Theranostics. 2022;12:6130–42. https://doiorg.publicaciones.saludcastillayleon.es/10.7150/thno.74194.
Article CAS PubMed PubMed Central Google Scholar
Han B, Lv X, Liu G, Li S, Fan J, Chen L, et al. Gut microbiota-related bile acid metabolism-FXR/TGR5 axis impacts the response to anti-α4β7-integrin therapy in humanized mice with colitis. Gut Microbes. 2023;15: 2232143. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/19490976.2023.2232143.
Article CAS PubMed PubMed Central Google Scholar
Weng JR, Tsai CH, Kulp SK, Chen CS. Indole-3-carbinol as a chemopreventive and anti-cancer agent. Cancer Lett. 2008;262:153–63. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.canlet.2008.01.033.
Article CAS PubMed PubMed Central Google Scholar
Ociepa-Zawal M, Rubiś B, Laciński M, Trzeciak WH. The effect of indole-3-carbinol on the expression of CYP1A1, CYP1B1 and AhR genes and proliferation of MCF-7 cells. Acta Biochim Pol. 2007;54:113–7. https://doiorg.publicaciones.saludcastillayleon.es/10.18388/ABP.2007_3276.
Article CAS PubMed Google Scholar
Agus A, Planchais J, Sokol H. Gut microbiota regulation of tryptophan metabolism in health and disease. Cell Host Microbe. 2018;23:716–24. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.chom.2018.05.003.
Article CAS PubMed Google Scholar
Qian M, Liu J, Zhao D, Cai P, Pan C, Jia W, et al. Aryl hydrocarbon receptor deficiency in intestinal epithelial cells aggravates alcohol-related liver disease. Cell Mol Gastroenterol Hepatol. 2022;13:233–56. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jcmgh.2021.08.014.
Article CAS PubMed Google Scholar
Li WT, Luo QQ, Wang B, Chen X, Yan XJ, Qiu HY, et al. Bile acids induce visceral hypersensitivity via mucosal mast cell-to-nociceptor signaling that involves the farnesoid X receptor/nerve growth factor/transient receptor potential vanilloid 1 axis. FASEB J. 2019;33:2435–50. https://doiorg.publicaciones.saludcastillayleon.es/10.1096/fj.201800935RR.
Article CAS PubMed Google Scholar
Ai YL, Wang WJ, Liu FJ, Fang W, Chen HZ, Wu LZ, et al. Mannose antagonizes GSDME-mediated pyroptosis through AMPK activated by metabolite GlcNAc-6P. Cell Res. 2023;33:904–22. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41422-023-00848-6.
Article CAS PubMed PubMed Central Google Scholar
Baixauli F, Piletic K, Puleston DJ, Villa M, Field CS, Flachsmann LJ, et al. An LKB1-mitochondria axis controls T(H)17 effector function. Nature. 2022;610:555–61. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41586-022-05264-1.
Article CAS PubMed PubMed Central Google Scholar
Zu Y, Liu L, Lee MY, Xu C, Liang Y, Man RY, et al. SIRT1 promotes proliferation and prevents senescence through targeting LKB1 in primary porcine aortic endothelial cells. Circ Res. 2010;106:1384–93. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/circresaha.109.215483.
Article CAS PubMed Google Scholar
Bi L, Ren Y, Feng M, Meng P, Wang Q, Chen W, et al. HDAC11 regulates glycolysis through the LKB1/AMPK signaling pathway to maintain hepatocellular carcinoma stemness. Can Res. 2021;81:2015–28. https://doiorg.publicaciones.saludcastillayleon.es/10.1158/0008-5472.Can-20-3044.
Article CAS Google Scholar
Wang Y, Wang M, Chen J, Li Y, Kuang Z, Dende C, et al. The gut microbiota reprograms intestinal lipid metabolism through long noncoding RNA Snhg9. Science. 2023;381:851–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1126/science.ade0522.
Article CAS PubMed PubMed Central Google Scholar
Kong M, Peng Y, Qiu L. Oligochitosan-based nanovesicles for nonalcoholic fatty liver disease treatment via the FXR/miR-34a/SIRT1 regulatory loop. Acta Biomater. 2023;164:435–46. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.actbio.2023.04.002.
Article CAS PubMed Google Scholar
Levine AJ. p53: 800 million years of evolution and 40 years of discovery. Nat Rev Cancer. 2020;20:471–80. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41568-020-0262-1.
Article CAS PubMed Google Scholar
Hao Q, Chen J, Lu H, Zhou X. The ARTS of p53-dependent mitochondrial apoptosis. Journal of Molecular Cell Biology. 2023;14:14. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/jmcb/mjac074.
Article CAS Google Scholar
Tan S, Li L, Chen T, Chen X, Tao L, Lin X, et al. β-Arrestin-1 protects against endoplasmic reticulum stress/p53-upregulated modulator of apoptosis-mediated apoptosis via repressing p-p65/inducible nitric oxide synthase in portal hypertensive gastropathy. Free Radical Biol Med. 2015;87:69–83. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.freeradbiomed.2015.06.004.
Article CAS Google Scholar
Hafner A, Bulyk ML, Jambhekar A, Lahav G. The multiple mechanisms that regulate p53 activity and cell fate. Nat Rev Mol Cell Biol. 2019;20:199–210. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41580-019-0110-x.
Article CAS PubMed Google Scholar
Liu C, Yang Q, Zhu Q, Lu X, Li M, Hou T, et al. CBP mediated DOT1L acetylation confers DOT1L stability and promotes cancer metastasis. Theranostics. 2020;10:1758–76. https://doiorg.publicaciones.saludcastillayleon.es/10.7150/thno.39013.
Article CAS PubMed PubMed Central Google Scholar
Li T, Zhu H. LKB1 and cancer: The dual role of metabolic regulation. Biomed Pharmacother. 2020;132: 110872. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.biopha.2020.110872.
Article CAS PubMed Google Scholar
Choi H, Kim HJ, Yang J, Chae S, Lee W, Chung S, et al. Acetylation changes tau interactome to degrade tau in Alzheimer’s disease animal and organoid models. Aging Cell. 2020;19:e13081. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/acel.13081.
Article CAS PubMed Google Scholar
Jiang N, Li W, Jiang S, Xie M, Liu R. Acetylation in pathogenesis: revealing emerging mechanisms and therapeutic prospects. Biomed Pharmacother. 2023;167: 115519. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.biopha.2023.115519.
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We appreciate assistance of all crew members during experiments in this study.

Funding

This work was supported by the National Key R&D Program of the National Natural Science Foundation of China (31930106, U22A20514, and U23A20232), China (2022YFD1300404), the National Natural Science Foundation of China (31930106, U22A20514, and U23A20232), the 2115 Talent Development Program of China Agricultural University (1041–00109019), and the 111 Project (B16044). This research is supported by Pinduoduo-China Agricultural University Research Fund, Grant No. PC2023A0100.

Author information

Authors and Affiliations

State Key Laboratory of Animal Nutrition and Feeding, Department of Animal Nutrition and Feed Science, China Agricultural University, Beijing, China
Zishen Lin, Yue Feng, Jinping Wang, Zhaoyue Men & Xi Ma

Authors

Zishen Lin
View author publications
You can also search for this author inPubMed Google Scholar
Yue Feng
View author publications
You can also search for this author inPubMed Google Scholar
Jinping Wang
View author publications
You can also search for this author inPubMed Google Scholar
Zhaoyue Men
View author publications
You can also search for this author inPubMed Google Scholar
Xi Ma
View author publications
You can also search for this author inPubMed Google Scholar

Contributions

Z. S. L performed investigation, analysis, and original draft. Y. F. performed investigation and analysis. J. P. W performed investigation. Z. Y. M performed investigation. X. M. performed conceptualization, funding acquisition, writing – original draft, and writing – review and editing. All authors reviewed the manuscript.

Corresponding author

Correspondence to Xi Ma.

Ethics declarations

Ethics approval and consent to participate

The experimental procedures were approved by the Institutional Animal Care and Use Ethics Committee of China Agricultural University.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

40168_2024_2011_MOESM1_ESM.pdf

Supplementary Material 1: Figure. S1. UGT1A4 promotes CDCA-3β-glucuronide synthesis to cause BA disorder diarrhea, related to Fig.1. (A) Changes of bile acid metabolism signals were enriched by PIRCUST between healthy pigs and BADD pigs (n = 8). (B) Changes in the concentration of primary bile acids and secondary metabolites produced by CDCA metabolism (n=8). (C-F) Correlation between CDCA-3β-glucuronide and FXR level, total bile acids level, BAX level, and histology score. (G) Establishment of ABX mice and that were received microbiota transplantation from healthy donors or BADD donors. (H and I) Relative protein expressions of FXR and UGT1A4 in mice(n=3). (J) Relative gene level of UGT1A4 in the colon of mice (n=3). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P < 0.01, ***P < 0.001. Correlation played on spearman two-tailed test in 99% confidence interval. Figure. S2. Effects of UGT1A4 and CDCA-3β-glucuronide on inflammation, cell viability, and gut barrier function, related to Fig. 2. (A)Cell viability of gut epithelial cells after cells received treatment of CDCA-3β-glucuronide (n=4-5). (B) Cell viability of gut epithelial cells after cells received treatment of UGT1A4 recombinant enzyme (n=3-4). (C) Cell viability of gut epithelial cells after cells received treatment of CDCA (n=4-6). (D) Gut epithelial cells were treated by CDCA or CDCA coculture with UGT1A4 recombinant enzyme, and measured the contents of CDCA-3β-glucuronide. (E) Relative gene levels ofIL1-β and IL-6 among control, CDCA, CDCA+UGT1A4 RE groups (n=3). (F) Casepase-1 activity and casepase-3 activity of gut epithelium cells were detected in control, CDCA-3β-glucuronide group, CDCA group, and co-treated CDCA and UGT1A4 RE group, respectively (n=3), results showed fold of control. (G) Apoptosis rates were measured by flow cytometry (n=3). FITC-A means the area of the pulse signal, which is an integral measurement of the fluorescence flux to reflect the sum of the cell fluorescence. (H) Concentration of CDCA-3β-glucuronide was measured in the colonic chyme (n=3) (I) A UGT1A4 depletion mouse model. (J) Relative gene levels of UGT1A4, IL1-β, and IL-6 in the colon of mice (n=3). (K) Relative protein expression of Occludin (n=3). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P< 0.01, ***P < 0.001. Figure. S3. Differential genes and GO signals are enriched in healthy and BA disorder diarrheal pigs, related to Fig. 3. (A)Relative mRNA of TNFα and IL-1β in primary gut epithelium cells isolated from healthy piglets and BADD piglets (n=3-7) (B) Content of UGT1A4 in primary gut epithelium cells isolated from healthy piglets and BADD piglets (n=3-7) (C) PCOA plot of genes on bray-curtis dissimilarity matrices in pigs (n=3-7). (D) Heatmap plot of genes (n=3-7). (E) Venn plot of genes (n=3-7). (F) Volcano plot of genes, compared to control group, purple represented up-regulated genes, orange down-regulated genes, and gray represented undifferentiated genes (n=3-7). (G) Expect from LKB1, rest differential gene were verified after FXR activation in the inflammation (n=3-6). (H) Changes of LPS mediated signaling, NF-κB signaling, signals of response to bacterium, LPS, and molecule of bacterial origin were showed by GO plots (n=3-7). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P< 0.05, **P < 0.01, ***P < 0.001. Figure. S4. Effects of FXR activation on changes of Sirtuins, HDACs, LKB1, and apoptosis gene expression, related to Fig. 4. (A) Cell viability of gut epithelial cells after cells received treatment of TNF-α (n=3-11). (B) Relative gene level of LKB1 in different concentrations of CDCA (n=3). (C) Apoptosis level was detected by flow cytometry (n=3), FITC-A means an integral measurement of the fluorescence flux to reflect the sum of the cell fluorescence. (D) Relative gene levels of LKB1, BAX, and BCL2 in treatment of FXR siRNA (n=3). (E) Apoptosis level was detected by flow cytometry in the siCON group and siFXR group (n=3). (F) Activities of caspase-1 and caspase-3 in the siCON group and siFXR group (n=3), results showed fold of control. (G) Apoptosis level was detected by flow cytometry in the siCON group and siLKB1 group (n=3). (H) Activities of caspase-1 and caspase-3 in the siCON group and siLKB1 group (n=3), results showed fold of control. (I and J) Relative gene levels of Sirtuins and HDACs families associated with bile acids (n=3-4). (K) Relative gene level of SIRT1 in treatment of FXR siRNA (n=3). (L) Relative gene levels of SIRT1, LKB1, and BAX in treatment of SIRT1 siRNA (n=3). (M) FXR and SIRT1 DNA promoter binding region was predicted by JASPER. The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P < 0.01, ***P < 0.001. Figure. S5. Interactions between proteins of LKB1 and P53 are predicted, related to Fig. 5. (A and B) GSEA enriched signals of LKB1 (n=3-7). (C) Enrichment of GO signal pathways (n=3-7). (D) Docking of LKB1 and TP53 was predicted in pigs. (E) Docking of LKB1 and TPR53 was predicted in mice. Data were analyzed by transcriptome sequencing (A to C). Figure. S6. Apoptosis associated genes regulated by P53 are measured, related to Fig. 5. (A) Co-immunofluorescence stain of LKB1 and P53 showed LKB1 and P53 both location in the cells. Scale bars, 25 μm. Green fluorescence is LKB1. Red fluorescence is P53. (B) Relative gene levels of NOXA and APAF-1 (n=3). (C and D) Relative protein expressions of SIRT1, LKB1, P53 of nucleus in the colon (n=4). (E) Relative gene levels of ZO-1 (n=3). (F) Relative gene levels of PUMA, NOXA, APAF-1, and BCL2 in the colon (n=4-8). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P < 0.01, ***P < 0.001. Figure. S7. Microbiota transplantation induced BA metabolism disorders are ameliorated by FXR activation, related to Fig. 6. (A) ABX mice after microbiota transplantation from healthy and BADD donors, were received oral gavage by FXR agonist. (B) Changes of body weight of mice in 14 days (n=6). (C) Faecal water index (n=6). (D and E) Colonic lengthen statistics (n=6). (F) Total bile acid contents in the faeces (n=6). (G) UGT1A4 enzyme content in the colon (n=6). (H) Histological staining and score (n=3). Scale bars, 100 μm. (I) Location and expression of FXR were measured by immunohistochemistry (n=3). Scale bars, 50 μm. (J and K) Relative protein expressions of FXR, SIRT1, LKB1, BAX/BCL2, and NFκB-P65 in the colon (n=3). (L) Acetylation-IP was used to measure acetylated LKB1 protein in colonic tissue of mice. (M) Relative gene levels of interleukin 4(IL-4) and IL-6 (n=3). (N) Relative gene levels of BAX, BCL2,BAX/BCL2, PUMA, APAF-1, NOXA (n=3). (O) Relative gene levels of ZO-1, ZO-2, Claudin-1, Clauin-2,Occludin (n=3). (P) Correlation between counts of faecal E. coli and UGT1A4 level (n=6). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P< 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P < 0.01, ***P< 0.001. Correlation played on spearman two-tailed test in 99% confidence interval. Figure. S8. DSS induced UGT1A4-BA disorders is alleviated by FXR-SIRT1-LKB1 axis, related to Fig. 6. (A)LKB1 depletion mouse model. (B)Relative gene levels of LKB1, FXR, SIRT1, BCL2, and BAX/BCL2 (n=3). (C) Histological staining and score (n=3). (D) Abundance of E. coli in the faeces. (E) DSS mouse model in treatment of FXR agonist and LKB1 inhibitor. (F) Colonic lengthen count (n=6). (G-J) Changes of body weight after DSS challenge (n=6). (K) Body weight change from 15d to 21d (n=6). (L) Abundance of E. coli in the faeces. (M) LPS concentration in the serum (n=6). (N) Disease activity index (n=6). DAI of DSS-treated mice and DSS+LKB1 inhibitor mice occurred significant differences compared to control group from 18d-21d. DAI of DSS+FXR agonist mice occurred significant decrease compared to DSS group from 19d-21d. (O) Organ weight ratio of liver and spleen (n=6). (P) Acetylation level of LKB1 protein in vivo. (Q) P53 protein expression in vivo. (R) Relative gene levels of Occludin,ZO-1, Claudin-1 in the colon (n=3). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P < 0.01, ***P < 0.001. Figure. S9. Relationship between microbiota and UGT1A4 content in BA disorder induced diarrhea, related to Fig. 7. (A) Observed feature index of α diversity (n=8). (B) Venn plot (n=8). (C) Enrichment of metabolic signaling pathway (n=8). (D) Relative abundance of P. aerogenes, E. coli, L. salivarius,C. formicilis, and E. cylindroides (n=8). (E) Heatmap of correlation, red * represents negative relation, blue* represents positive relation (n=8). (F and G) L. helveticus, P. copri, and L. mucosae in treatment of cell to measure UGT1A4 level, respectively (n=4). (H) Correlation between bacterial signaling pathways and level of UGT1A4 enzyme (n=8). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05, **P < 0.01, ***P < 0.001. Correlation played on spearman two-tailed test in 90 % confidence interval. Figure. S10. L.reuteri and it derived I3C prevent CDCA glucuronidation induced diarrhea, related to Fig. 8. (A) Faecal water content in FMT mice received L. reuteri orally gavage (n=6). (B) Total bile acid content in FMT mice received L. reuteri orally gavage (n=6). (C) Histological score of colonic tissues (n=4-6). (D) Faecal water content in FMT mice received I3C and CDCA orally gavage (n=3). (E) Total bile acid content in FMT mice received I3C and CDCA orally gavage (n=3). (F) Level of UGT1A4 enzyme in FMT mice received I3C and CDCA orally gavage (n=3). The differences between groups were compared using either student’s t-test or two-way analysis of variance (ANOVA) followed by Tukey’s test for multiple comparisons, and statistical significance was considered at P < 0.05. Data were represented as mean ± SEM, and exhibited *P < 0.05,**P < 0.01, ***P < 0.001. Table S1 Fecal total bile acid content and E. coli number of pigs. Table S2 Primers for qRT-PCR. Table S3 Criteria of disease activity index (DAI) [56].

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Lin, Z., Feng, Y., Wang, J. et al. Microbiota governs host chenodeoxycholic acid glucuronidation to ameliorate bile acid disorder induced diarrhea. Microbiome 13, 36 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-024-02011-8

Download citation

Received: 08 March 2024
Accepted: 17 December 2024
Published: 04 February 2025
DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40168-024-02011-8

Microbiota governs host chenodeoxycholic acid glucuronidation to ameliorate bile acid disorder induced diarrhea

Abstract

Background

Methods

Results

Conclusion

Introduction

Materials and methods

Animal models

Mice models

Colonic microbiota transplantation

Primary gut epithelial cell isolation

Gut epithelial cell line culture

Bacterial isolation and culture

Bile acid metabolomics sequencing and analysis

16S rRNA microbial gene sequencing analysis

RNA-seq transcriptomics

Metabolomics sequencing

LPS and UGT1A4 ELISA and total bile acid content assays

Caspase-1 and caspase-3 activity detection

RNA interference

Total RNA extraction and quantitative reverse transcription PCR

ChIP PCR analysis

Protein interaction prediction

IP and Co-IP analysis

Histopathological analysis

Immunohistochemistry

TUNEL assay

Western blot trial

Antibodies

Immunofluorescence analysis

Flow cytometry apoptosis detection

Cell viability assay (Cell Counting Kit-8)

Loads of E. coli in the feces

Statistical analysis

Results

UGT1A4 mediated CDCA-3β-glucuronide overgeneration leads to BA disorder diarrhea

LKB1 regulated by FXR is associated with apoptosis in BA disorder diarrhea

FXR transcriptional controls SIRT1 is necessary for LKB1 deacetylation

LKB1 binds with P53 in the nucleus is necessary to fight with inflammatory induced apoptosis

FXR agonist alleviates BA disorders diarrhea through FXR-SIRT1-LKB1 axis in mice

Suppression of UGT1A4 enzyme by Lactobacillus reuteri contributes to alleviate BA disorder diarrhea

I3C derived by Lactobacillus reuteri improved UGT1A4-mediatedepithelialinjury by altering location of AHR

Discussion

Conclusion

Data availability

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Additional information

Publisher’s Note

Supplementary Information

40168_2024_2011_MOESM1_ESM.pdf

Rights and permissions

About this article

Cite this article

Share this article

Keywords

Microbiome

Contact us