Indole Compounds from Gut Bacteria May Reduce Fatty Liver Disease by Improving Fat Metabolism and Inflammation

A prospective cohort study was conducted between April 2017 and March 2020 (ClinicalTrials.gov NCT04339725↗). This study involved liver disease patients who were followed up by the Department of Liver Diseases, University Hospital. The groups were 19 healthy controls, 16 MASLD patients without hepatitis, and 26 MASLD patients with hepatitis. This project followed the ethics of the 1975 Helsinki Declaration, as reflected by a prior approval by the institutional review board for human research in hospitals (2016–134). Informed consent was obtained from all participants. All authors had access to the study data and reviewed and approved the final manuscript.

MASLD is defined as hepatic steatosis and one or more of the five cardiometabolic risk factors.^17,18 MASLD was diagnosed using blood chemistry, clincal parameter, liver biopsy, or imaging study (ultrasound or computed tomography scan). Patients with elevated liver enzymes aminotransferase (AST) or aspartate aminotransferase (ALT) ≥ 50 IU/L or hepatitis on liver pathology were included in the hepatitis group. Enrolled patients with MASLD who did not drink excessive alcohol (male >60 and female >40 g/wk). Patients with autoimmune liver disease, alcohol use disorder, pancreatitis, hemochromatosis, viral liver disease, pregnancy, Wilson’s disease, drug-induced liver injury, and other cancers were excluded. At enrollment, patients taking medications affecting the gut microbiome were excluded. As a control group, normal people visiting the hospital for health examination were recruited. The patient’s disease was treated regardless of the study.

Animals were under human care, and all procedures were performed in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. All procedures were approved by the Animal Experiment Management Committee (2020–40, 2021–53) of Hallym University College of Medicine. Six-week-old specific pathogen-free male C57BL/6J mice were purchased from Dooyeol Biotech (Seoul, Korea). All mice were individually housed in steel microisolator cages at 22 ± 2°C with a 12/12 h light/dark cycle. Mice had ad libitum access to water and food throughout the experiment and were monitored daily. The experiment included an acclimatization period for all groups, during which mice in all groups had an acclimatization period to eating normal diet for 1 week. IPA and IAA were administered at 0.1 mg/ml diluted in water. As an additional animal experiment, Bifidobacterium bifidum was orally administered at a concentration of 10⁹ CFU/100ul daily for 12 weeks. WD (TD88137, Seoul, Korea) was purchased, and the nutritional composition was 42% fat, 42.7% carbohydrate, and 15% protein.

Specimens were fixed with 10% formalin and routinely embedded in paraffin, and the tissue sections were processed with hematoxylin and eosin. To grade fatty liver, nonalcoholic fatty liver disease activity score (NAS), an objective index consisting of steatosis, lobular inflammation, and ballooning, was evaluated.¹⁹ All biopsy specimens were analyzed by a single liver pathologist. Fatty liver was classified as grade 0 to 3 according to the Clinical Research Network scoring system (0: <5%, 1: 5%–33%, 2: 34%–66%, and 3: >66%). of steatosis). Inflammation was graded from 0 to 3 (0: none, 1: 1–2 foci per × 20 field, 2: 2–4 foci per × 20 field, 3: >4 foci per × 20 field). According to NAS-based guidelines, it may be helpful to recognize histological scoring of systems that cover the full spectrum of MASLD.

Human feces were stored at −20°C as soon as the patient received 2–3 g of feces using the kit (stool paper and stool box) and moved to −80°C within 1 day. For mice feces, whole intestine was collected during euthanizing and frozen at −80°C and stored. Genomic DNA for metagenomic sequencing was extracted with a QIAamp stool kit (Qiagen, Germany) and library was prepared with a NEBNext Ultra II FS DNA Library Prep Kit for Illumina (New England BioLabs, USA) according to the manufacturer’s directions. The quantification of libraries was checked using a Qubit dsDNA HS assay kit (ThermoFisher Scientific, USA) and confirmed by qPCR with KAPA SYBR FAST qPCR Master Mix kit (Kapa Biosystems, USA). The quality of libraries was assessed on a Bioanalyzer 2100 (Agilent, USA) using a DNA 12,000 chip. All libraries were sequenced on the NovaSeq 6000 platform (Illumina, USA) with a paired end (PE) 150 bp reads. Assays were performed according to previous methods.²⁰ In brief, DNA was extracted with a QIAamp stool kit, and amplification of the V3-V4 region of the bacterial 16S rRNA gene was conducted using barcoded fusion primers. The forward fusion primer contained p5 adapter, i5 index, and gene-specific primer 341F (5′-AATGATACGGCGACCACCG AGATCTACAC-XXXXXXXX-TCGTCGGCAGCGTCAGATGTGTATAAGAGAC AG – CCTACGGGNGGCWGCAG-3′; underlining indicates the target region primer and X indicates the barcode region), and the reverse fusion primer contained p7 adapter, i7 index, and gene-specific primer 805 R (5′-CAAGCAGAAGACGGCATACGAGAT-XXXXXXXXGTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHV GGGTATCTAATCC-3′), in which included sequencing adapters and dual-index barcodes of the Nextera XT kit (Illumina, San Diego, CA, USA). The amplification was performed in the C1000 touch thermal cycler polymerase chain reaction system (Bio-Rad Laboratories, Inc., USA) with the following conditions: initial denaturation of 3 min at 95°C; followed by 25 cycles of denaturation at 95°C for 30 s, annealing at 55°C for 30 s, extension at 72°C for 30 s and final extension at 72°C for of 5 min. Each amplified PCR product was confirmed with 1% agarose gel electrophoresis and visualized on a Gel Doc XR+ imaging system (Bio-Rad laboratories, Inc., USA). The amplified products were purified and size-selected by Agencourt AMPure XP beads (Beckman Coulter, USA). Library was constructed with pooled PCR products and quality of library was assessed on a Bioanalyzer 2100 (Agilent, USA) using a DNA 12,000 chip and quantified by qPCR with KAPA SYBR FAST qPCR Master Mix kit (Kapa Biosystems, USA). Sequencing was carried out according to the manufacturer’s instructions at CJ Bioscience, Inc. (Seoul, Republic of Korea) with the Illumina MiSeq platform using reagent kit V3 in PE 250 bp mode. Microbiome taxonomic profiling was conducted with the EZBioCloud platform (CJ Bioscience Inc., Republic of Korea) using the database version PKSSU4.0. After taxonomic profiling of each sample, comparative MTP analyzer of EZBioCloud was used for the comparative analysis of the samples.²¹ The number of operational taxonomic units (OTUs) picking was conducted with UCLUST and CDHIT with 97% of similarity cutoff.²² Subsequently, Good’s coverage, rarefaction, and alpha-diversity indices including Simpson, Jackknife, ACE, Chao1, Shannon, and NPShannon were calculated. Beta-diversity was shown by clustering using the and principal coordinate analysis (PCoA) unweighted pair group method with arithmetic mean (UPGMA).

Metabolic profiles in the mouse cecum were obtained using a combination of gas chromatography-mass spectrometry (GC-MS) and liquid chromatography-mass spectrometry (LC-MS) methods. Cecal samples were thawed at 4°C and mixed with 1.1 ml cold extraction solvent I (acetonitrile/water 1:1, v/v). The mixture was vortexed for 1 min, sonicated for 5 min under ice, and centrifuged at 13,200 rpm for 5 min at 4°C. Each supernatant (500 μl) was transferred to a new 2 ml tube for SCFA analysis. The remaining supernatant was mixed with 600 μl of cold extraction solvent II (acetonitrile/methanol, 1:3, v/v). For the second extraction step, the mixture was vortexed for 1 min and centrifuged at 13,200 rpm for 5 min at 4°C. The supernatant (500 μl) was aliquoted and transferred to new 1.5 ml tubes for gas chromatography time-of-flight mass spectrometry and liquid chromatography Orbitrap mass spectrometry. Aliquots were concentrated to complete dryness using a speed vacuum concentrator (SCANVAC, Korea). Detailed analyzes were performed according to previous research methods.

Lactobacilli used in the preliminary study were isolated from various sources, including sour milk, cheese, feces of healthy Korean adults, and feces of newborn babies. B. bifidum is a lactobacillus strain isolated from the feces of newborns. B. bifidum was inoculated into flasks containing de Man, Rogosa and Sharpe medium (BD/Difco). Strains were incubated under anaerobic conditions at 37°C for 24 h. Stocks of each strain were prepared by mixing the culture with an equivalent 20% skim milk solution and then storing the mixture at −80°C.

Cultures of each strain were previously grown in liquid culture anaerobic conditions at 35° with or without Trp. After 24 h, sample cultures were separated by centrifugation at 4000 rpm for 10 min at 4°C. Prepared by diluting 50:1 of Salkowski’s reagent, 35% HClO4 (perchloric acid) and 0.5 M FeCl3, reacted 1:1 with the bacterial culture medium and incubated in the dark at 30°C for 30 min. The absorbance is then measured at 530 nm. Calculate the concentration by comparing the optical density of the test sample to a standard IAA curve (10–100 μg ml).

To study how metabolites derived from the gut microbiota are altered as MASLD progresses, we analyzed the microbiota with 16S rRNA gene sequencing and the levels of specific microbiota-derived metabolites in stool samples from 19 healthy controls (HCs) and 42 MASLD patients (Figure 1a, Supplementary Figure S1 and Figure S2, and Supplementary Table S1). In the analysis of composition, phylum composition, and the Firmicutes/Bacteroidetes (F/B) ratio, statistically significant differences were observed among the groups (Figure 1b, c). The abundance of Bifidobacterium was increased as liver disease progressed (Figure 1d). At the species level, B. bifidum was significantly increased in MASLD. B. longum was increased in MASLD. However, statistically, there was no difference. The levels of B. adolescentis, B. catenulatum, and B. breve were increased as liver disease progressed (Supplementary Figure S1e)

We analyzed Trp metabolites including indole derivatives and kynurenine (Kyn), in the stool of patients.²³ Trp was significantly increased in MASLD patients. On the other hand, MASLD, IPA significantly decreased in MASLD patients. AA levels decreased in MASLD patients, but this was not statistically significant. (Figure 1e). However, there was a significant decrease in changes according to the disease progression (Supplementary Figure S2a). In the volcano plot, the IPA level was the most significantly decreased compound in MASLD patients (Figure 1f). Other indole derivatives did not reveal the difference between healthy controls and MASLD patients (Figure 1e). The level of Kyn was not significantly different among the groups (data not shown).

Fecal samples from MASLD patients were observed to have lower levels of major short-chain fatty acids (SCFAs) than stool samples from HCs (Supplementary Figure S2b). In contrast, the levels of cholic acid and chenodeoxycholic acid were significantly increased (Supplementary Figure S2c). In addition, the levels of the liver function markers AST and ALT were increased in the blood of MASLD patients (Supplementary Table S1). Overall, these clinical data indicate that IPA and IAA are candidate metabolites of preventive value and suggest a potential role in the pathogenesis of hepatic steatosis.

We first divided mice into four groups to study the effects of IPA and IAA on hepatic lipid metabolism in MASLD: normal control (NC), WD, and groups supplemented with IPA and IAA under WD conditions (Figure 2a). Treatment with IPA and IAA improved the liver/body ratio (L/B) compared to that in the WD group, and the L/B ratio was significantly improved with IPA treatment (Figure 2b). Hematoxylin and eosin staining upon liver histopathology showed abnormal lipid droplets and inflammation in WD-fed mice, which were reduced by IPA and IAA treatment; the inflammation grade and ballooning grade were greatly reduced. The NAS improved with IPA and IAA treatment, especially IAA treatment, which resulted in a significant improvement (Figure 2c).

In addition, compared to those in the NC group, the levels of the liver function indicators AST and ALT were significantly increased in the mice fed a WD, and liver function was significantly improved through IPA and IAA supplementation. On the other hand, there was no difference between triglyceride (TG) and cholesterol (CHOL) levels (Figure 2d). These data demonstrated improvement in both histological and serologic changes with IPA and IAA administration for 12 weeks, suggesting a protective role for IPA and IAA in WD-induced hepatic steatosis.

To evaluate liver inflammation and lipid metabolism in WD-fed mice, we detected changes in the mRNA expression of markers related to inflammation and lipid metabolism in liver tissue. First, the expression of tumor necrosis factor (Tnf)-α, a proinflammatory cytokine, was significantly increased in mice fed a WD, which was significantly improved through IPA and IAA administration. However, there was no significant difference in interleukin 1 beta (Il-1β) and Il-6 expression between the WD group and the IPA and IAA groups (Figure 3a). We also evaluated the mRNA expression of chemokines that activate inflamed tissues and play a pivotal role in the pathogenesis of MASLD.²⁴ The C-X-C motif chemokine ligand 10 (Cxcl10) level was significantly decreased in the IPA and IAA groups compared to that in the WD group (Figure 3b). Next, the gene expression of carnitine palmitoyltransferase-1a (Cpt1a), peroxisome proliferator-activated receptor-α (Ppar-α), and acyl-CoA oxidase 1 (Acox1), which are genes involved in the β-oxidation of fatty acids, was significantly increased by IPA and IAA administration (Figure 3c). Fatty acid synthase (Fasn), acetyl-CoA carboxylase 1 (Acc1), cluster of differentiation 36 (Cd36), and genes involved in adipogenesis and transport were induced in the WD group, whereas they were significantly suppressed in the IPA and IAA groups (Figure 3d).

Trp metabolites, indole compounds, are known as ligands of the aryl hydrocarbon receptor (AHR), and IAA is a representative AHR ligand.²³ Thus, the levels of cytochrome P450 family 1 subfamily B member 1 (Cyp1b1) and cytochrome P450 family 1 subfamily A member 2 (Cyp1a2), which are genes regulated by AHR in the ileum, were significantly increased by IAA administration, suggesting that AHR was activated in the ileum (Figure 3e). In contrast, the expression of AHR in liver tissue was not activated (Supplementary Figure S3a). Increased proinflammatory cytokine expression were decreased by IPA and IAA (Supplementary Figure S3b). Overall, these results suggest that the administration of IPA and IAA to WD-fed mice improves hepatic steatosis by regulating the inflammatory response and fatty acid synthesis and β-oxidation in the liver.

The treatment of Caco-2 cells with IPA and IAA increased the transepithelial electrical resistance (TEER) values by 1.2-fold and 1.6-fold, respectively, compared with those of untreated controls (Figure 3f).

We evaluated whether treatment with IPA and IAA improved liver inflammation by reducing intestinal-derived endotoxin, reducing the activity of liver macrophages, and finally inhibiting NF-κB signaling. First, we evaluated the mRNA level of Toll-like receptor 4 (TLR4), which is activated by recognizing lipopolysaccharide (LPS), in the livers of WD-induced mice. As a result, the level of Tlr4 increased in the WD-induced mouse group, while treatment with IPA and IAA showed a decrease its levels to levels similar to those of the NC group, but did not show statistical significance (Figure 4a). In the Western blot analysis, IAA treatment significantly decreased the expression of Tlr4 (Figure 4b). Elevated levels of LPS binding protein (LBP), a protein that binds serum endotoxin induced by a WD, were significantly reduced in the IAA-treated group (Figure 4c). Next, we evaluated the mRNA levels of C-C motif ligand 2 (Ccl2) and C-C motif ligand 5 (Ccl5), which are representative chemokines that attract monocytes and macrophages to the site of inflammation. The levels of Ccl2 and Ccl5 were significantly increased in WD-fed mice, but these levels were significantly decreased with IPA and IAA treatment (Figure 4d). Based on this finding, we investigated the effect of IPA and IAA treatment on the protein level of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), a protein complex involved in cytokine production and cell survival in macrophages. Interestingly, the phosphorylated protein levels of p65 and IκBα were increased in WD-fed mice but were restored by treatment with IPA and IAA, and the change with IAA treatment was significant (Figure 4e).

To evaluate the anti-inflammatory effects of IPA and IAA, we measured the levels of cytokines and chemokines in LPS-stimulated macrophages. RAW264.7 cells were pretreated with IPA and IAA in a dose-dependent manner and then treated with LPS. IPA and IAA significantly reduced Tnf-α, and Ccl2 in LPS-stimulated Raw264.7 cells (Figure 4f, g). mRNA levels of Ahr with IPA and IAA treatment on Raw264.7 cell and CaCo2 cells were increased (Figure 4h, i). These results suggest that IPA and IAA treatment increase Ahr expression and suppress the expression of endotoxin-stimulated Tlr4, thereby inhibiting macrophage activation and NF-κB signaling and ultimately improving inflammation.

IPA and IAA increased the expression of Acox1 and Cpt1, fatty acid oxidation markers, in AML12 cells (Supplementary Figure S4a and S4b). Inflammation-related cytokines were reduced by IPA and IAA treatment in RAW264.7 cells (Supplementary Figure S4c and S4d). These results suggest that IPA and IAA play a role in controlling fat metabolism and inflammation.

To investigate how IPA and IAA contribute to liver protection from reduced inflammation and lipid accumulation in WD-induced mice, we performed RNA-seq of livers harvested from NC, WD, IPA and IAA mice and profiled liver transcripts All significant findings had a fold change > 2, normalized data (log2) >1, p < .05, and WD/NC, IPA/WD, and IAA/WD were compared. p-values were adjusted for FDR (False Discovery Rate). Differentially expressed gene (DEG) clustering heatmap analysis showed significant upregulation and downregulation in IPA and IAA treatment compared to WD-induced mice, and dendrograms indicating expression similarity between samples were relatively similar in the IPA and IAA groups, unlike in the WD group (Figure 5a). Clustering analysis with significant DEGs revealed a unique expression profile for tissues in each group.

We analyzed the WD group vs. IPA and IAA in the Gene Ontology (GO) biological process category. First, in the IPA group, regulation of Il-6 production and leukocyte activation, which contribute to inflammation, were downregulated, and biological oxidation, an energy-generating response in living cells, was upregulated. Next, in the IAA group, the negative regulation of protein phosphorylation was downregulated, and the metabolism of lipids and lipid catabolic processes involved in lipid metabolism were upregulated (Figure 5b). Principal component analysis (PCA) showed a distinct separation between the four groups according to group type (Figure 5c). Next, we performed gene set enrichment analysis (GSEA) to identify pathways regulated by IPA and IAA in WD-induced mice. Not surprisingly, we found significant upregulation of the Trp metabolism pathway in mice administered IPA and IAA, which are metabolites of Trp (Figure 5d). Genes differentially expressed in the liver were identified by comparing the IPA and IAA-administered mouse groups with the WD group.

IPA administration upregulated 34 genes and downregulated 55 genes, and IAA administration resulted in 37 upregulated genes and 48 downregulated genes. Genes with common changes in IPA/WD and IAA/WD were 26 upregulated genes and 27 downregulated genes. Interestingly, consistent with previous studies, the gene expression of Ccl2, a chemokine that attracts monocytes and macrophages to the site of inflammation, was significantly downregulated, and the gene expression of Cyp1a2, a gene associated with AHR, was significantly increased.

The correlation of genes was statistically anal-yzed and visualized (Figure 5e–g). A total of 53 genes commonly regulated by IPA/IAA (Figure 5e). The comparative analysis revealed 43 regulated genes that crossed WD/NC, IPA/WD, and IAA/WD, satisfying the three conditions of fold change > 2, normalized data (log2) >1, and p < .05 (Figure 5f and Supplementary Table S3). Figure 5g shows a total of 43 regulated genes statistically analyzed and visualized for correlation based on the protein-protein interaction database. Ccl2, inflammation marker, is decreased in IPA and IAA groups. Mup (major urinary protein), which inhibits insulin resistance and glucose tolerance, and cytochrome P450 relating genes were closely related and increased IPA and IAA groups (Figure 5g). These RNA-seq studies indicate that IPA and IAA administration activate the Trp metabolic pathway and that this activation is attributable to the regulation of transcription related to inflammation and lipid metabolism. Esm1 (endothelial cell-specific molecule 1) and lymphocyte antigens 6 (Ly6d), inflammation markers, were decreased in IPA and IAA groups. In addition, the level of Cyp1a2 was increased in IPA and IAA groups as shown in Figure 3e.

We performed 16S rRNA-based intestinal microbial taxonomic profiling to investigate the potential relevance of IPA and IAA administration to changes in intestinal microbial composition and to specific bacteria that produce indole compounds. Among the ɑ-diversity indicators, Simpson and Shannon indices showed no significant difference across all groups (Figure 6a). On the other hand, β-diversity, which indicates the difference in the composition of the gut microbiota between samples and groups, showed a distinct difference across all groups based on principal coordinate analysis (PCoA) and showed a large change in the microbial composition due to WD supply, and the IPA and IAA groups were partially separated (Figure 6b). Cluster analysis also showed that there were differences between WD group and indole groups that were administered concurrently with WD (Figure 6c). Several studies have shown that the F/B ratio is associated with obesity and various diseases.²⁵ The F/B ratio was decreased in the IPA and IAA groups compared to that in the WD group but was not statistically significant (Figure 6d). The abundance of Bifidobacterium was significantly increased in the IPA and IAA groups (Figure 6e). Each group had a different genus composition (Supplementary Figure S5).

We tested the ability to produce IAA by adding Trp, a precursor of IAA, through a Salkowski assay using 13 strains, which are representative strains that are approved for use as a probiotics in South Korea. The control strains E. coli and B. bifidum showed positive results, suggesting that B. bifidum can metabolize IAA using Trp (Figure 6f). Therefore, measurements were made from 0 h to 72 h to determine the correlation between the growth of B. bifidum and IAA production. B. bifidum growth peaked at 24 h with and without Trp, but the IAA concentration increased up to 24 h only in the medium containing Trp (Figure 6g).

Our previous findings revealed that B. bifidum is a novel bacterium capable of producing IAA. Therefore, we hypothesized that IAA-producing B. bifidum could protect against liver dysfunction in a WD-induced MASLD mouse model. To prove this hypothesis, we divided mice into the NC group and the WD group. To evaluate whether B. bifidum metabolizes Trp, an IAA precursor, under WD conditions, B. bifidum with or without TRP was administered by gavage. In addition, L. plantarum Q180 and Trp were included as controls (Figure 7a). Since Trp administration is known to induce hepatic steatosis, we evaluated whether hepatic steatosis was attenuated through B. bifidum metabolic ability.²⁶ The L/B ratio increased in the WD group and the TRP group, but supplementation with B. bifidum and L. plantarum improved the L/B ratio, and in particular, the group treated with TRP and B. bifidum had a significantly decreased L/B ratio (Figure 7b).

Hematoxylin and eosin staining showed abnormal lipid droplets and inflammation in WD-fed mice, which were alleviated by administration of B. bifidum, and showed an effect of relieving hepatic steatosis in the group administered Trp coadministered with B. bifidum. As a result, the inflammation grade and balloon grade were significantly reduced, and the NAS was significantly improved with B. bifidum treatment (Figure 7c). In addition, the levels of AST and ALT, liver function indicators, were significantly increased in the mice of the WD and Trp groups, and liver function was significantly improved through the administration of B. bifidum. On the other hand, there was no significant difference in TG levels, and the CHOL levels were significantly reduced only in the group treated with Trp and B. bifidum (Figure 7d). These data demonstrated improvement in both histological and serological changes with B. bifidum administration for 12 weeks, suggesting that B. bifidum may play a protective role by metabolizing Trp, especially in Trp-induced hepatic steatosis.

To determine whether B. bifidum mediates an inhibitory effect on hepatic steatosis, we evaluated WD-induced mice in groups treated with B. bifidum alone or in combination with Trp and B. bifidum. First, the levels of Tnf-α, a proinflammatory cytokine, were significantly increased in the WD group, but significantly improved in the strain-treated group. However, as in previous studies, there was no difference between Il-6 and Il-1β levels (Figure 8a and Supplementary Figure S6a). Levels of the chemokine Cxcl10 were decreased in the group treated with B. bifidum alone, and levels of Ccl2 and Ccl5, which recruit macrophages, were greatly reduced in the group treated with B. bifidum and cultured medium (Figure 8b). In the TEER test with L, plantarum and B. bifidum, B. bifidum increased tight junction function (Figure 8c). Next, the expression of genes involved in the β-oxidation of fatty acids, Ppar-α and Acox1, increased in the group treated with the strain, especially in the group administered Trp and B. bifidum together (Figure 8d). The expression of Ppar-γ, which is involved in fat synthesis, was increased in the WD group and greatly improved by B. bifidum treatment. However, unlike previous studies, B. bifidum treatment did not significantly alter the expression of genes related to fat synthesis (Supplementary Figure S6b). In Raw264.7 cell analysis, B. bifidum and culture medium effectively improved the expression of inflammation (Supplementary Figure S7). In the microbiota analysis, each group showed typical findings of beta-diversity, phyrum and genus composition, and proportion of bifidobacterium (Supplementary Figu-re S8).

We also evaluated genes associated with AHR, which binds IAA. AHR expression was almost absent in the WD group but increased in the Trp and strain groups. In particular, AHR expression was significantly increased in the group administered Trp and B. bifidum together, suggesting that B. bifidum can metabolize Trp to generate IAA to activate AHR in the ileum (Figure 8e and Supplementary Figure S9a). Conversely, consistent with previous studies, the expression level of AHR in liver tissue was not found to be activated by B. bifidum treatment (Supplementary Figure S6c).

Our results suggest that B. bifidum can reduce translocation and inhibit the activation of macrophages, as in previous studies. Additionally, these results were not improved in the Trp group, suggesting that B. bifidum attenuated inflammation in the liver by metabolizing Trp. Serum IAA levels were increased in the B. bifidum groups (alone and with Trp)(Figure 8f). In the human data alone, decreased cecal IAA and IPA levels in WD were recovered in B. bifidum group (Supplementary Figure S9b and Supplementary Table S2). In liver tissue, changed IAA and IPA levels were recovered in the B. bifidum group (Supplementary Figure S9c).

Next, we evaluated the role of AHR by using AHR antagonist, CH-223191 (Sigma Korea, Seoul)(Figure 8g, h). level of inflammatory cytokine, NAS score, and fat regulation factors, which were improved in the B. bifidum with Trp group, were worsened (Supplementary Figure S9c, S9d, and S9f). This result suggests that indole producing microbiota increased AHR expression on ileum and inhibited MASLD progression.

Overall, treatment with B. bifidum alone could improve hepatic inflammation and steatosis via AHR. In addition, although the Trp group experienced exacerbation of hepatic inflammation and hepatic steatosis, the administration of B. bifidum had an anti-MASLD effect, which validated our hypothesis.

All data are available in manuscript and supplementary file.