Background
Inflammatory bowel disease (IBD) is a chronic, relapsing disorder of the gastrointestinal tract (GI); Crohn’s disease and ulcerative colitis are common forms of IBD [
1]. Although the etiology of IBD remains unclear, it is believed to involve complex interactions among multiple genetic factors, altered gut microbiota, inappropriate host immune responses, and environmental factors [
2,
3]. Traditionally, IBD has been more common in individuals of the Western European descent [
4]. The majority of the > 215 genes with variants associated with an increased risk of IBD, including
NOD2 and
ATG16L, have been identified in cohorts of Western patients with IBD [
5], but not in Asian cohorts [
6,
7]. Moreover, immigrants to Western countries have an increased risk of IBD [
8,
9], suggesting that the Western environment and lifestyle contribute to the IBD pathogenesis. Both murine and human studies showed that intake of a Western diet (WD), characterized by a high-fat and sugar content, influences IBD pathogenesis [
10‐
12]. Dietary surveys revealed that some patients with IBD believe that WD intake triggers disease flare-ups and exacerbates symptom severity [
13,
14]. However, the precise molecular mechanisms underlying these phenomena are unknown.
Taste receptors (TRs) include two families of G protein-coupled receptors: taste receptor type 1 (T1R) and taste receptor type 2 [
15]. The T1R family senses sweet (TAS1R2, TAS1R3) and umami (TAS1R1, TAS1R3) stimuli, whereas the taste receptor type 2 family senses bitter stimuli. TRs previously characterized in the oral cavity have recently been reported in extra-oral tissues, with particularly high expression in enteroendocrine cells (EECs) of the gastrointestinal tract [
16]. TRs regulate the release of intestinal peptide hormones, including glucagon-like peptide 1 (GLP-1) [
17]. GLP-1, secreted by EECs, potentiates glucose-induced insulin response, promotes beta-cell survival, slows gastric emptying, and regulates energy expenditure [
18]. Although the mechanisms underlying TR function in the contexts of digestion and metabolism are well-characterized, their roles in intestinal inflammation and immunity are unclear.
Recent studies showed that taste chemosensory signaling proteins are expressed in intestinal tuft cells, where they play critical roles in detecting and eliminating intestinal parasites by initiating type II immune responses and regulating gut epithelial homeostasis [
19‐
21]. These studies suggest that taste or taste-like chemosensory pathways in the intestine play important roles in regulating immune responses to components of luminal stimuli. However, most studies have primarily focused on elucidating the immunomodulatory reactions of these pathways in response to microorganisms or microbial metabolites. To our knowledge, no studies have reported the role of TR responses to endogenous dietary ligands. Given the immense diversity of non-microbial signals (nutrients and non-nutritive chemicals) that stimulate immune responses at the barrier site, intestinal TRs, which are directly activated by dietary ligands, may modulate intestinal mucosal immunity and inflammation.
In this study, we developed a murine model of severe intestinal inflammation triggered by long-term WD intake to evaluate the influence of WD intake on intestinal inflammation and investigate the mechanisms by which WD intake influence the development of IBD. We hypothesized that modulation of the nutrient-induced gut-taste receptor TAS1R3 (TR type 1 member 3), is central to the regulation of intestinal inflammation. Our findings provide novel insights into the previously unknown effects of long-term WD consumption on intestinal inflammation and may lead to the development of preventive or therapeutic strategies for IBD.
Methods
Study design
The objective of the study was to evaluate the influence of long-term WD intake on intestinal inflammation and investigate possible mechanisms by which WD intake could affect IBD development. To this end, mice were fed normal diet (ND) or WD for 10 weeks, and bowel inflammation was evaluated through pathohistological and infiltrated inflammatory cell assessments. To identify the mechanisms by which intestinal inflammation is prompted by WD, RNA-seq was performed on the inflamed intestinal tissues. Because the results revealed increased expression of the nutrient-sensing taste receptor TAS1R3 in inflamed bowel tissues, we hypothesized that nutrient-induced TAS1R3 modulation is central to regulating intestinal inflammation. We first determined the dietary ligand(s) responsible for strongly activating TAS1R3 using in vitro assessment of TAS1R3-expressing enteroendocrine cells (EECs) to confirm whether TAS1R3 could trigger intestinal inflammation, and then assessed the downstream molecular changes. To understand the role of TAS1R3 in WD-induced intestinal inflammation, we investigated changes in intestinal gene expression profiles, inflammatory cell infiltration in intestinal tissues, and the gut microbiome of Tas1r3-deficient and littermate wild-type mice fed WD. Finally, we confirmed the expression of TAS1R3 and downstream players in intestinal biopsies of patients with IBD ex vivo to demonstrate their relevance to disease. All mice were assigned randomly to treatment groups, and animals demonstrating sickness or severe stress were euthanized and excluded. Otherwise, all data were included in the study. The experimental protocol was approved by the Seoul National University Institutional Animal Care and Usage Committee (approval SNU-181001-2).
Mice
C57BL/6 (JAX 000664) and Tas1r3tm1Csz (JAX 013066) mice were obtained from the Jackson Laboratory (West Grove, PA, USA). Tas1r3-knockout (Tas1r3−/−) mice were backcrossed to C57BL/6 mice for at least seven generations. The animals were genotyped using standard PCR. The mice (5–6 weeks old; 23–25 g) were randomly assigned to groups. Age- and weight-matched male and female littermates were used as controls. Tas1r3−/− and wild-type littermate control (Tas1r3+/+) mice were housed under constant temperature (23 ± 2 °C) and humidity (55–60%) conditions in a specific pathogen-free animal facility. All mice were housed in the same room to minimize environmental effects.
Diets
To imitate the human WD, characterized by a high-fat content and sugary drinks, mice were fed high-fat diet and sucrose solution. The high-fat diet (60%; D12492) and matching normal diet (ND; D12450J) pellets were purchased from Research Diets (New Brunswick, NJ, USA). The sugar concentration in the sucrose solution was determined, as described in a previous rodent study [
22]. Two independent series of experiments were conducted, and mice were randomly assigned as follows:
In experiment 1, two groups of mice were used: (i) ND—mice administered ND with plain water (control group; n = 10 mice) and (ii) WD—mice administered the high-fat diet with sucrose solution (n = 10 mice).
In experiment 2, four groups of mice were used: (i) Tas1r3+/+ mice who were administered ND (n = 10 mice), (ii) Tas1r3−/− mice who were administered ND (n = 10 mice), (iii) Tas1r3+/+ mice who were administered WD (n = 10 mice), and (iv) Tas1r3−/− mice who were administered WD (n = 10 mice). Food and water were supplied ad libitum for 10 weeks. The body weight, as well as food and water intake, were monitored weekly.
All mice used in the experiments were randomly assigned to each group, and random numbers generated using SPSS software (version 18.0; SPSS Inc., Chicago, IL, USA) were assigned with a unique code linking to the individual animal.
Histology
After 10 weeks of diet induction or 7 days of 2% dextran sulfate sodium (molecular weight: 36,000–50,000 kDa; Cat# 02160110-CF; MP Biomedicals, Santa Ana, CA, USA) induction, all mice were anesthetized and euthanized by intraperitoneal injection of 20% urethane (U2500; Sigma-Aldrich, St. Louis, MO, USA), and the entire intestine was removed and opened longitudinally. Small and large intestinal tissues were fixed overnight in 10% formalin and embedded in paraffin. The tissue blocks were cut into Sects. (4–6 μm thick) that were mounted on glass slides, stained with hematoxylin and eosin, and photographed under a Nikon Eclipse TE2000-U microscope (Tokyo, Japan) equipped with a QImaging digital camera (Teledyne QImaging, Surrey, BC, Canada). H&E-stained intestinal sections were coded for blind microscopic assessment of inflammation. Sections coded for assessment of inflammation were scored by two blinded investigators as described previously (
n = 10 mice/group) [
23].
Immunohistochemistry and immunofluorescence
Serial sections (4–6 μm thick) were prepared from formalin-fixed paraffin-embedded specimens and mounted on silane-coated slides (Dako Japan Co., Ltd., Kyoto, Japan). Tissues were deparaffinized and rehydrated through a graded xylene and alcohol series, placed in citrate-buffered solution (pH 6.0) (C9999; Sigma-Aldrich), and heated in a microwave oven to 100 °C for 20 min for antigen retrieval. After washing with phosphate-buffered saline, the slides were incubated with phosphate-buffered saline-Tween with 1% bovine serum albumin (Sigma-Aldrich) for 1 h, incubated overnight at 4 °C with anti-TAS1R3 antibodies (1:100) (Cat#OSR00184W, RRID: AB_2271552; Invitrogen, Carlsbad, CA, USA) or pan-leukocytes (CD45; 1:100) (Cat#ab10558, RRID: AB_442810; Abcam, Cambridge, UK), and developed using a Mouse and Rabbit-Specific HRP/DAB Detection IHC Kit (Cat#ab64264; Abcam), according to the manufacturer’s instructions. Primary antibodies used for small or large intestine immunofluorescence staining included mouse anti-CD4 (Cat#14–0041-86, RRID: AB_467065; eBioscience, San Diego, CA, USA), rat anti-CD8 (Cat#ab22378, RRID: AB_447033; Abcam), and mouse anti-CD11b (Cat#14–0112-82, RRID: AB_467108; eBioscience), to label different types of immune cells in intestinal tissues. Secondary and secondary-conjugated primary antibodies used for small intestine and colon immunofluorescence staining included Alexa Fluor 488 anti-mouse/human CD11b (Cat#101219, RRID: AB_493545; BioLegend, San Diego, CA, USA), Alexa Fluor 594 donkey anti-rabbit (Cat#A21207, RRID: AB_141637; Thermo Fisher Scientific, Waltham, MA, USA), Alexa Fluor 488 donkey anti-mouse (Cat#A21202, RRID: AB_141607; Thermo Fisher Scientific), and Alexa Fluor 594 donkey anti-rat (Cat#A21209, RRID: AB_2535795; Thermo Fisher Scientific). Immunoreactivity was visualized using a confocal microscope system (LSM 510; Carl Zeiss, Oberkochen, Germany). Images were captured at 20X magnification using an EVOS FL Cell Imaging System (Thermo Fisher Scientific) and a BZ-X710 All-in-One Fluorescence Microscope (Keyence, Osaka, Japan), and analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Tissue-infiltrated cells are expressed as the ratio of the stained area to the total area of the measured tissue region (n = 10 mice/group).
Enterocyte cell culture and GLP-1 secretion assay
Human enteroendocrine NCI-H716 cells (CCL-251; American Type Culture Collection, Manassas, VA, USA) were maintained in suspension culture at 37 °C and 5% CO
2, according to the supplier’s protocol. The culture medium was RPMI-1640 (Invitrogen) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 IU/mL penicillin, and 100 μg/mL streptomycin. Two days prior to the experiments, 1 × 10
6 cells were seeded in 24-well culture plates precoated with Matrigel (BD Biosciences, Franklin Lakes, NJ, USA), as described previously [
24]. On the day of the experiments, the supernatants were replaced with medium containing 10 mM glucose (Cat# G7021), 10 mM fructose (Cat# F3510), 10 mM sucrose (glucose + fructose), and/or 10 μM palmitate (Cat# P5585). In the TAS1R3 antagonist experiment, NCI-H716 cells were pretreated with the TAS1R3 antagonist, lactisole (2.5 mM) (Cat# M6546) for 30 min and stimulated with medium containing fructose (10 mM), glucose (10 mM), and palmitate (10 μM) for 12 h. In the PPARγ antagonist experiment, NCI-H716 cells were transfected with
TAS1R3 siRNA or control siRNA for 48 h, and then stimulated with the PPARγ antagonist, GW9662 (Cat# M6191) (10 μM), in the presence of fructose (10 mM), glucose (10 mM), and palmitate (10 μM) for 12 h. The cells were incubated for 2, 4, 12, and 24 h at 37 °C with or without different test agents and inhibitors. Glucose, fructose, palmitate, lactisole, and GW9662 were purchased from Sigma-Aldrich. Following incubation, the medium was collected, centrifuged at 1,000 ×
g for 10 min at 4 °C, to remove any floating cells, and frozen at -20 °C for subsequent biochemical analysis. GLP-1 was measured using a commercial GLP-1 (active) Enzyme-Linked Immunosorbent Assay Kit (Cat#EGLP-35 K; Millipore, Billerica, MA, USA), according to the manufacturer’s protocol (
n = 3/group).
siRNA knockdown
siRNA duplexes for
TAS1R3 were synthesized by Bioneer (Daejeon, South Korea). Scrambled negative control siRNA was also purchased from Bioneer. For knockdown experiments, 5 × 10
5 endocrine differentiated NCI-H716 cells were plated into 6-well plates and cultured for 48 h.
TAS1R3 (10 nM) or control (10 nM) siRNA was transfected into the cells using Lipofectamine RNAiMAX Reagent (Invitrogen). After 48 h of transfection, the cells were induced with medium containing 10 mM glucose, 10 mM fructose, and 10 μM palmitate for 12 h (
n = 3/group). The target sequences of the siRNA are listed in Additional file
1 (Table S1).
Enzyme-linked immunosorbent assay
Enzyme-Linked Immunosorbent Assay (ELISA) Kits specific for human TNF-α (Cat# ab181421) and IL-8 (Cat# ab46032) were purchased from Abcam. TNF-α and IL-8 in the cell-conditioned medium were quantified using Enzyme-Linked Immunosorbent Assay, according to the manufacturer’s instructions. Briefly, the samples were diluted with assay buffer and added to microwells precoated with anti-human TNF-α or IL-8 antibodies, followed by incubation with an antibody cocktail and 3,3′,5,5′-tetramethylbenzidine substrate. Conjugated enzyme activity was detected by measuring absorbance at 450 nm (n = 3/group).
RNA isolation and quantitative reverse transcription-PCR
Total RNA was extracted from the ileum tissue of WD- or DSS-induced wild-type and knockout mice and NCI-H716 cells. After harvest, the ileal and cell extracts were immediately snap-frozen by immersion in liquid nitrogen and stored at –80 °C until RNA extraction. Total RNA was isolated using RNAqueous (Cat# AM1914; Ambion, Austin, TX, USA), according to the manufacturer’s instructions. RNA extraction involved a DNase treatment step. RNA was quantified using a NanoDrop 2000/2000c Spectrophotometer (Thermo Fisher Scientific), and 1 μg of RNA from each sample was used for cDNA synthesis (Cat# 11917010; Invitrogen). Quantitative reverse transcription-PCR was performed using StepOnePlus (Real-time PCR System; Applied Biosystems, Foster City, CA, USA) and SYBRGreen (Cat# 4367659; Applied Biosystems). The relative quantification of gene expression was performed using the 2
−△△Ct method. The cycle threshold values of the genes of interest were normalized to that of
Gapdh (
n = 3–10 mice/group). The primers used for quantitative reverse-transcription-PCR are listed in Additional file
1 (Table S2).
RNA-sequencing (RNA-seq)
Isolated RNA (1000 ng) was used as input for library generation. Intact mRNA was isolated from total RNA using a Dynabeads mRNA DIRECT Micro Kit (Ambion). The total mRNA samples were depleted of up to 99.9% of 5S, 5.8S, 18S, and 28S rRNA using the RiboMinus Eukaryote System v2 (Life Technologies, Carlsbad, CA, USA). Barcoded cDNA libraries were prepared from the rRNA-depleted mRNA samples and constructed using the Ion Total RNA Seq Kit v2 (Life Technologies). Whole-transcriptome libraries were diluted to 100 pM and amplified with ion sphere particles by emulsion PCR using an Ion One Touch 2 System (Life Technologies) and Ion PI Hi-Q OT2 200 Kit (Cat# A26434; Life Technologies). Template-positive ion sphere particles were enriched using the Ion OneTouch Enrichment System (Life Technologies), in which biotinylated adaptor sequences were selected by binding to streptavidin-conjugated beads. The template-positive ion sphere particles were sequenced using the Ion PI Hi-Q Sequencing 200 Kit (Cat# A26433; Life Technologies). Sequencing primers were annealed to template fragments attached to the ion sphere particles, and the template-positive ion sphere particle samples were loaded onto a chip from the Ion PI Chip Kit v3 (Cat# A26771; Life Technologies) and incubated with polymerase. Finally, the chip was placed on an Ion Proton System (Life Technologies) for ion semiconductor sequencing, which is based on the principle of hydrogen ion release detection when nucleotides are incorporated into the growing DNA template (n = 6 mice/group). All procedures were performed according to the manufacturer’s instructions.
Raw RNA-seq reads were split into individual samples based on their barcodes and quality controlled using the FASTQC tool. The reads were analyzed using Partek Flow software (Partek, St. Louis, MO, USA) [
25]. Briefly, the reads were mapped to Genome Reference Consortium Mouse Build 38 (mm10) using STAR 2.5.3a aligner. Quantification was performed using the transcript model Ensembl Transcripts release 91. Differential expression analysis was conducted using R package DESeq2. Genes with an absolute fold-change of ≥ 2 and a false discovery rate (FDR) < 0.01 were considered as differentially expressed genes (DEGs). The DEGs were subjected to ingenuity pathway analysis using IPA software (Qiagen, Hilden, Germany) [
26]. Hierarchical clustering and biological classification analyses were also performed. Fisher’s exact test was used to determine the significance of the enrichment of specific biological processes among the DEGs. Hierarchical clustering analysis was performed using Genesis v1.7.545 based on a Pearson correlation distance matrix and average linkage algorithm.
Protein extraction and western blotting
Approximately 0.1 g of terminal ileum tissue isolated from Tas1r3+/+ or Tas1r3−/− mice was transferred to 1 mL Tissue Extraction Reagent 1 (Cat# FNN0071; Invitrogen) with 1/1,000 protease inhibitor (Cat# P-2714; Sigma-Aldrich, St. Louis, MO, USA) and homogenized. NCI-H716 cells were also lysed using RIPA buffer and homogenized. Protein concentrations were determined using the Pierce BCA Protein Assay Kit (Cat# 23227; Thermo Fisher Scientific) with bovine serum albumin as the standard. Aliquots of each protein lysate (20 μg) were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred to a polyvinylidene fluoride membrane, blocked for 1 h at 24 °C with 5% fat-free milk in Tris-buffered saline containing 0.1% Tween 20, and incubated with monoclonal rabbit anti-mTOR (1:1,000) (Cat# 2972S, RRID: AB_330978), rabbit anti-phospho-mTOR (1:1,000) (Cat# 2971S, RRID: AB_330970), and rabbit anti-PPARgamma (1:1,000) (Cat# 2443S, RRID: AB_823598) primary antibodies from Cell Signaling Technology (Danvers, MA, USA), and rabbit anti-Occludin (1:2,000) (Cat# ab216327, RRID:AB_2737295) and rabbit anti-Claudin-1 (1:2,000) (Cat# ab180158) primary antibodies from Abcam. Mouse anti-α-tubulin (1:10,000) (Cat# T5168, RRID: AB_477579) from Sigma was used as a control. Horseradish peroxidase-conjugated goat anti-rabbit secondary antibodies (1:5,000) (Cat# 7074S, RRID: AB_2099233) from Cell Signaling Technology and goat anti-mouse secondary antibodies (1:10,000) (Cat# G21040, RRID: AB_2536527) from Invitrogen were used for detection. The target proteins were detected using enhanced chemiluminescence western blot detection reagents (Amersham Pharmacia Biotech, Piscataway, NJ, USA) (n = 6 mice/group).
16S rRNA gene sequencing
Fecal samples were frozen immediately after collection and stored at –80 °C. Total bacterial DNA was isolated from fecal samples using a QIAamp Fast DNA Stool Mini Kit (Cat# 51604; Qiagen), according to the manufacturer’s protocol for pathogen detection, with slight modifications. Briefly, approximately 200 mg of fecal material was homogenized for 1 min at 30 Hz with 5-mm sterilized steel beads in ASL buffer using a TissueLyser bead mill (Qiagen). The suspension was heated at 95 °C to lyse gram-positive bacterial cells. In the final incubation step, we extended the elution time from 1 to 5 min to increase the DNA yield. A 16S rRNA sequencing library was constructed, according to the 16S metagenomics sequencing library preparation protocol (Illumina, San Diego, CA, USA), targeting the V3 and V4 hypervariable regions of the 16S rRNA gene. KAPA HiFi HotStart ReadyMix (KAPA Biosystems, Wilmington, MA, USA) and the Agencourt AMPure XP System (Beckman Coulter Genomics, Brea, CA, USA) were used to amplify and purify the PCR products, respectively. The amplicons were sequenced in paired-end mode (PE275) using a MiSeq System (Illumina) (n = 10–14 mice/group).
16S rRNA-seq analysis
Paired sequences were dereplicated using the QIIME pipeline [
27], and de novo and reference-based chimeras were removed using UCHIME software. Sequences from all samples were merged and sorted based on their relative abundances, and then closed operational taxonomic unit selection was performed using a 99% similarity threshold, followed by stringent taxonomic assignment using Silva v1.19. Based on the operational taxonomic unit abundance matrix and respective taxonomic classifications, feature abundance matrices were calculated at different taxonomic levels (from genus to phylum). To estimate the species diversity within samples (α-diversity), Shannon’s and Pielou’s indices were calculated using the R package phyloseq, after rarefaction. For comparisons among samples (β-diversity), Bray–Curtis and weighted principal coordinate analysis (PCoA) UniFrac dissimilarities were determined using phyloseq, based on profiles normalized to sample depth. Linear discriminant analysis effect size analysis was conducted online using the Galaxy workflow framework [
28] to identify differentially abundant bacterial genera.
Aliquots (200 mg) of fecal samples were thawed, to which 4 volumes of distilled water (800 μL) were added, followed by vortexing at room temperature for 5 min until the samples were homogenized. Then, 15 μL of 95% sulfuric acid (Sigma-Aldrich) was added to 300 μL of fecal supernatant for acidification (final 5% (v/v)), followed by stabilization for 5 min. After centrifugation at 14,000 ×
g for 5 min, the supernatants were transferred to a new tube. To extract volatile materials, 30 μL of 1% (v/v) internal standard (2-methylpentanoic acid; Sigma-Aldrich) and 300 μL of anhydrous ethyl ether (Sigma-Aldrich) were added to acidified fecal supernatant. The samples were vortexed for 1 min, and then centrifuged at 14,000 ×
g for 5 min. The upper layer was carefully transferred to a GC vial and stored at -80 °C before analysis. The 10 mM butyrate solution (Cat# T8626; Sigma-Aldrich) was used as the standard for butyrate analysis. Stool butyrate was measured using the Agilent Technologies 7890A GC System (Santa Clara, CA, USA), as described by David et al. [
29] (
n = 20 mice/group).
IBD gene expression profiling
To analyze the DEGs in IBD (
n = 204)
versus non-IBD (
n = 74), mRNA microarray expression profiles were retrieved and downloaded from the Gene Expression Omnibus database [
30] by searching the following keywords: “IBD,” “active ulcerative colitis,” “active Crohn’s disease,” and “
Homo sapiens” (organism). The inclusion criteria were as follows: (i) intestinal tissues (not cells) from adult patients with active IBD and (ii) samples from patients with IBD who had not received any interventions or treatments. After screening, five mRNA expression datasets (GSE160804, GSE126124, GSE95095, GSE75214, and GSE53306) were selected for analysis (see Additional file
1: Table S3). The raw microarray data were downloaded from the Gene Expression Omnibus database and preprocessed using Partek Genomics Suite version 6.6 (Partek) with the robust multichip analysis algorithm, which performs background adjustment, quantile normalization, and probe summarization. GC-content correction was used, as suggested by the default pipeline of Partek Genomics Suite. To estimate the effect of the normalization procedure, expression data without normalization and with standard robust multichip analysis normalization (without GC-content correction) were also generated. Differential gene expression analysis was performed using R/Bioconductor [
31,
32]. For DEG selection, an FDR
P < 0.001 was considered the cutoff value.
Murine colitis model
Colitis was induced by administering 2% DSS dissolved in drinking water for 7 days. Control mice were provided with drinking water without DSS. Colitis development was evaluated by monitoring daily weight changes. Colitis severity was also scored by evaluating the clinical disease activity through daily observation of the following parameters: weight loss (0 points = no weight loss or weight gain, 1 point = 5–10% weight loss, 2 points = 11–15% weight loss, 3 points = 16–20% weight loss, 4 points = > 21% weight loss); stool consistency (0 points = normal and well-formed, 2 points = very soft and unformed, 4 points = watery stool); and bleeding stool score (0 points = normal color stool, 2 points = reddish color stool, 4 points = bloody stool). The disease activity index was calculated based on the combined scores of weight loss, stool consistency, and bleeding, and ranged from 0 to 12. All parameters were scored from Days 0 to 7. On Day 7 after DSS-colitis induction, the mice were sacrificed, and the entire intestine was quickly removed. After determining the colon length as a marker of inflammation, the entire colon was cut open lengthways and gently flushed with sterile phosphate-buffered saline to remove any traces of feces. Small and large intestinal segments were immediately frozen in liquid nitrogen and stored at − 80 °C for subsequent extraction of total RNA. For histological analysis, intestinal segments were fixed in 10% neutral buffered formalin phosphate and stored at room temperature until inflammation was analyzed (n = 7 mice/group).
Statistical analysis
All data were analyzed using GraphPad Prism software (version 9.0; GraphPad Inc., San Diego, CA, USA). The Kolmogorov–Smirnov test was used to assess data normality. All data were found to be normally distributed. SigmaPlot 11.0 was used to estimate all sample sizes using data from previous experiments and preliminary data with α = 0.05 and β = 0.2. Unless otherwise specified, all data are expressed as means ± standard errors of the mean. Descriptive statistics are listed in Additional file
1 (Table S4). The mean values of two groups were compared using Student’s
t-test, and the means of multiple groups were compared using one-way analysis of variance, followed by Bonferroni post-hoc test. All statistical analyses were two-sided, and
P < 0.05 was considered significant. In gene expression analyses, significant enrichment of specific genes was determined using a right-tailed Fisher’s exact test. Correlations between the relative abundance of butyrate-producing bacteria and transcript expression of PPAR-associated molecules were analyzed using Spearman’s correlation test.
Discussion
In this study, we extensively analyzed the role of taste receptor TAS1R3 in bowel tissue inflamed by prolonged ingestion of WD. Our results showed that the nutrient-sensing intestinal receptor, TAS1R3, is a key modulator of host-gut microbial interactions, ultimately regulating intestinal inflammation (see Additional file
5: Fig. S4).
In contrast to chemosensory tuft cells, which sense parasite- or metabolite-derived ligands and modulate the immune system in response [
19,
20,
47,
48], EECs are key chemosensory cells that directly sense nutrient ligands and exhibit spatial heterogeneity with tuft cells. Herein, we found that EECs express TAS1R3 and directly recognize endogenous nutrient ligands to regulate the secretion of pro-inflammatory molecules, suggesting that they serve as key agents in the pathogenesis of IBD. Indeed, one of our primary findings is that nutrients, rather than microorganisms, directly prompt EECs to secrete cytokines via nutrient-sensing TRs. Moreover, although TAS1R3 ligands in the oral cavity were previously considered to only be sweet molecules [
33], herein we provide evidence that intestinal TAS1R3 responds to dietary fat as well. Furthermore, we demonstrate that robust pro-inflammatory cytokines secreted in EECs via TAS1R3 activation by nutrient binding can ultimately attract multiple inflammatory cells, thereby worsening intestinal inflammation. Consistent with this, pro-inflammatory cytokines, such as TNF, are key targets for IBD therapy [
49]. New agents that target cytokines, or their signaling cascades, are currently being tested in clinical trials, suggesting that cytokine blockade is a promising approach for IBD therapy. Thus, our findings may have important applications for IBD management.
WD intake promotes the expansion of pro-inflammatory pathogenic gut bacteria, ultimately leading to gut dysbiosis [
50]. Here, we found that at the phylum level, the relative abundance of Proteobacteria, which is expanded in mouse and human IBD [
42], was significantly high in WD-fed wild-type (
Tas1r3+/+) mice compared to that in ND-fed
Tas1r3+/+ mice. Moreover, the abundance of major discriminators (well-known pathobionts associated with irritable bowel syndrome and IBD) [
51,
52], significantly increased. Hence, long-term WD intake enhances the expansion of IBD-associated pathobionts, consistent with our model of WD-induced intestinal inflammation. In contrast, WD-induced gut dysbiosis did not occur in
Tas1r3-deficient mice with increased PPARγ as a downstream target. Elevated PPARγ expression induced by TAS1R3 deficiency may protect against gut dysbiosis through two main mechanisms: (1) driving β-oxidation, leading to the expansion of obligate anaerobic butyrate-producing bacteria, by maintaining a hypoxic state, with obligate anaerobic butyrate-producing bacteria suppressing the dysbiotic expansion of facultative anaerobic pathogenic bacteria by competing for oxygen; and (2) directly regulating the mTOR–PPARγ axis in the intestine, by an intrinsic mechanism, reflected by the markedly increased abundance of intestinal PPARγ in ND-fed
Tas1r3-deficient mice, which may protect against dysbiosis by strengthening gut barrier function. Overall, our results indicate that gut TAS1R3 is an intrinsic regulator of intestinal inflammation and constitutes a central mechanism for controlling intestinal crosstalk with the gut microbiota.
PPARγ regulates intestinal inflammation [
53‐
55]. Another novel finding of this study is the identification of PPARγ as a downstream target regulated by TAS1R3 deficiency. Genetic ablation of PPARγ results in increased susceptibility to experimental colitis in mice, supporting its anti-inflammatory properties [
54]. In contrast, PPARγ signaling agonists, such as rosiglitazone, attenuate the severity of inflammatory lesions in both experimental and spontaneous models of colitis [
55]. Although the anti-inflammatory mechanism of PPARγ remains unclear, findings from previous studies suggest the following: (i) PPARγ activation inhibits the production of numerous inflammatory cytokines, through its action on kinases and transcription factors, such as NF-κB, c-Jun, and c-Fos [
37,
38,
56]; (ii) PPARγ agonists enhance barrier function by upregulating tight-junction molecules in gastrointestinal epithelial cells and maintaining intestinal mucosal integrity [
39,
57,
58]; and (iii) PPARγ may exert key antimicrobial effects, by activating AMP production, including intestinal α- and β-defensins [
40,
41]. Consistent with these mechanisms, we found that
Tas1r3-deficient mice exhibited increased expression of intestinal PPARγ, leading to significantly increased expression of TJPs, and several AMPs, along with decreased expression of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6, and IL-8). Furthermore, PPARγ inhibition via GW9662 in TAS1R3 knockdown cells rescued the secretion of pro-inflammatory cytokines induced by the blockade of TAS1R3. Thus,
TAS1R3 deficiency plays an important role in regulating the inflammatory response through PPARγ. In turn, PPARγ-dependent processes, through which TAS1R3 mediates its effects, may alleviate intestinal inflammation.
The regulation of intestinal PPARγ expression and host intestinal-microbial interactions are also closely linked [
59]. PPARγ drives the energy metabolism of intestinal epithelial cells towards β-oxidation and enhances oxygen consumption while suppressing iNOS synthesis. Therefore, PPARγ signaling leads to decreased oxygen bioavailability in the gut lumen [
46]. In a hypoxic environment, the gut microbiota is dominated by obligate anaerobic bacteria, further limiting the generation of host-derived nitrogen and oxygen, preventing dysbiotic expansion of potentially pathogenic facultative anaerobic bacteria by competing for oxygen, resulting in the maintenance of healthy gut homeostasis [
47]. Conversely, inhibition of epithelial PPARγ signaling leads to metabolic reorientation of enterocytes towards aerobic glycolysis rather than β-oxidation, increasing epithelial oxygenation and elevating oxygen bioavailability to promote the expansion of facultative anaerobic bacteria [
47]. Accordingly, we confirmed that prolonged exposure to a WD resulted in the expansion of facultative pathogenic anaerobic bacteria, such as
Enterobacteriaceae and
Prevotellaceae, in low PPARγ-expressing
Tas1r3+/+ mice. However,
Tas1r3-deficient mice, with high PPARγ expression, harbored predominantly obligate anaerobic bacteria, even with the consumption of WD.
The drive towards β-oxidation, reinforced through PPARγ signaling under TAS1R3 deficiency, was positively correlated with the abundance of obligate anaerobic bacteria, such as butyrate-producing bacteria. Thus, epithelial hypoxia occurring consequent to host PPARγ expression maintains anaerobiosis in the gut to support the dominance of butyrate-producing bacteria. Moreover, luminal butyrate functions as a major driver of the PPARγ-dependent transcriptional response in intestinal epithelial cells [
46], constituting a positive feedback loop that substantially amplifies the protective effects against intestinal inflammation [
60,
61]. Our findings show that TAS1R3 acts as a coordinating hub for host enteric defense through crosstalk between bacteria and intestinal epithelial cells.
In humans, reduced PPARγ expression in intestinal epithelial cells is a hallmark of IBD [
62,
63]. However, the mechanism underlying this reduction remains poorly understood. Here, we report the co-occurrence of increased TAS1R3 and reduced PPARG expression in intestinal biopsies of patients with IBD. These clinical data are consistent with our animal model findings. In addition to the WD-induced inflammation model,
Tas1r3 deficiency suppressed intestinal inflammation in a DSS-induced colitis model. This is a widely used model exhibiting several characteristics of human IBD [
23]. Therefore, our findings provide strong evidence that TAS1R3 contributes to IBD in humans and mice.
We also found that
Tas1r3-deficient mice had significantly lower mTOR activity in the intestinal tract than wild-type mice, following the consumption of WD. Notably, low nutrient conditions suppress mTORC1, leading to increased rates of catabolic processes [
64,
65]. The absence of the nutrient-sensing TR, TAS1R3, may induce a nutrient-starvation-causing scenario. Thus, despite nutrient replete conditions (i.e., prolonged WD),
TAS1R3-knockout enhanced catabolism, because of the perceived starvation. This assumption is supported by the upregulation of mRNAs encoding the catabolic markers, Lipin1 and PPARγ, within the repressed mTOR signaling pathway. That is, inhibition of mTOR induces Lipin1 activation, thereby activating PPARγ, which inhibits lipogenesis and drives intracellular metabolism towards β-oxidation [
66,
67]. This was consistently observed in
Tas1r3-deficient mice; thus, TAS1R3 likely acts as a nutrient sensor that conveys early signals regarding nutrient availability to the energy-sensing component of mTOR.
A focus of this study was to assess the role of environmental factors in etiology of IBD. Our animal model findings confirmed that the consumption of a WD is a major trigger of intestinal inflammation, with gut-expressed TAS1R3 representing a key target of WD-induced inflammation. However, although intestinal TAS1R3 expression was markedly increased in patients with IBD, we did not determine whether WD intake directly increased TAS1R3 expression, due to the absence of dietary records. Further studies are needed to analyze habitual dietary patterns to better assess the associations between WD components and TAS1R3 overactivation in patients with IBD. Particularly, the expression of TAS1R3 as a function of WD intake should be monitored in patients during the pre-IBD to IBD transition. Furthermore, it might be needed for future work to look at how microbial changes in Tas1r3-deficient mice can affect colitis susceptibility by transferring altered microbial contents to germ-free or antibiotic mice.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.