Ileal Bile Acid Transporter Inhibitor Improves Hepatic Steatosis by Ameliorating Gut Microbiota Dysbiosis in NAFLD Model Mice

Six-week-old mice were fed either the control diet or high-fat diet (HFD) for 12 weeks. During the experiment, the average food intake of mice in the HFD plus IBATi and control diet groups did not differ. As shown inin the supplemental material, body weights were significantly higher in the HFD groups than in the control diet groups after 6 weeks of treatment. This difference was maintained until the end of the study. The average body weight gain of the HFD plus IBATi group was significantly suppressed during the 6- to 12-week period. The average liver weight/body weight ratio was significantly higher in the HFD groups than in the control diet groups, which was inhibited by treatment with IBATi. After 12 weeks, livers were excised from each group and examined by hematoxylin and eosin (H&E) and oil red O staining. Although the liver sections of the control diet and IBATi groups were free of lipid droplets, the HFD group exhibited increased accumulation of lipid droplets, indicating hepatic steatosis. Conversely, this increase was suppressed in the HFD plus IBATi group (). The hepatic crown-like structure, which is one of the unique histological features of NAFLD, was only found in liver sections of the HFD group (). The NAFLD activity score (NAS) increased significantly from 0.33 ± 0.58 (mean ± SD) in the control diet group to 7.67 ± 0.58 in the HFD group; this score was changed to 1.67 ± 1.55 in the HFD plus IBATi group at 12 weeks of treatment (). However, histopathological fibrosis was not observed in liver sections of any of the treatment groups at the end of the experiments (). These results indicated that IBATi prevented body and liver weight gain and improved hepatic steatosis in HFD mice. Fig. S1A and B Fig. S1C and D Fig. S1E Fig. S1F Fig. S1G

Serum samples obtained from each treatment group after 12 weeks were assayed using enzyme-linked immunosorbent assay (ELISA). Serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) concentrations were higher in the HFD group than in the control diet and IBATi groups. When HFD mice were treated with IBATi, these increases were prevented. Biochemical lipid measurements confirmed high levels of total cholesterol, low-density lipoproteins (LDL), high-density lipoproteins (HDL), and triglycerides in the sera of the HFD group compared to the control diet and IBATi groups. Serum levels of total cholesterol and LDL were significantly lower in the HFD plus IBATi group. Serum levels of total BAs in the HFD group did not differ from those in the HFD plus IBATi group (). Fig. S2A

Previous studies reported reduced hepatic cholesterol 7-a-monooxygenase (Cyp7a1) and increased ileal Fgf15 mRNA expression in NAFLD model mice (6, 15). Therefore, we analyzed hepatic and ileal mRNA expression in each treatment group. Although Cyp7a1 expression in livers of the HFD group was decreased, treatment with IBATi increased its expression up to 3-fold in livers of the HFD group. Cholesterol 8-a-monooxygenase (Cyp8a1) and farnesoid X receptor (Fxr) mRNA expression in livers did not differ among treatment groups. In the terminal ileum, Fgf15 mRNA expression was elevated in the HFD group, which was prevented with IBATi treatment (Fig. S2B and C). These results indicated that IBATi treatment improved hepatic steatosis and liver dysfunction. Hepatic Cyp7a1 mRNA expression was suppressed, and ileal Fgf15 mRNA expression was increased in the HFD group. However, treatment with IBATi ameliorated these mRNA expression changes in HFD mice. In Fig. S2D, amounts of the FXR agonists omega-muricholic acid (ωMCA) and deoxycholic acid (DCA) were high in the feces from HFD mice. Although the amount of DCA in feces from HFD plus IBATi mice was as same as from HFD mice, ωMCA was significantly reduced in feces from HFD plus IBATi mice. Additionally, β-muricholic acid (βMCA) and lithocholic acid (LCA) were also significantly reduced.

Given that the intestinal BA composition was likely to have altered the intestinal bacterial population, 16S rRNA sequencing was used to characterize the effect of IBATi on HFD-induced gut microbiota dysbiosis. In total, 166 operational taxonomic units (OTUs) were obtained from α-diversity among the control diet, HFD, IBATi, and HFD plus IBATi groups using different indices for the observed species and the Shannon index. The Shannon index was significantly lower in the HFD group than the control diet and IBATi groups, which was prevented by IBATi treatment (Fig. 1A).

Subsequently, the overall microbiota community structure of the control diet, HFD, and HFD plus IBATi groups were calculated using β-diversity indices for unweighted Bray-Curtis distances. Principal-coordinates analysis (PCoA) revealed significant differences in the microbiota structure of the control diet, HFD, and HFD plus IBATi groups (permutational multivariate analysis of variance [PERMANOVA], P = 0.001) (Fig. 1B). HFD was associated with a higher proportion of Streptococcaceae (Fig. 1C). Relative abundances at the family level by linear discriminant analysis effect size (LEfSe) are shown in Fig. 1D. The ratio of Firmicutes to Bacteroidetes was increased in HFD groups compared to control diet groups (Fig. 1E). Additionally, relative abundances at the genus and species levels are shown in Fig. S3A and B. These results suggested that IBATi treatment changed the reduced gut microbiota diversity and microbiota dysbiosis induced by HFD.

From all FASTA files in metagenomic analysis of normal mice, HFD mice, IBATi mice, and HFD plus IBATi mice, we extracted sequences that were determined as Streptococcaceae at the family level. Each sequence had a similarity of 99% or more in the sequence of V4 region using the Greengenes database version 13.8. Three sequences were extracted, but almost one sequence (246 bp) was determined to be Streptococcaceae in this study (Data Set S1). The percentage of this sequence in all OTUs of each sample was significantly increased in HFD mice and decreased in HFD plus IBATi group of mice. The bacteria whose sequence (246 bp) was 100% query covered were examined using Basic Local Alignment Search Tool (BLAST) in National Center for Biotechnology Information (NCBI). In the results, the bacteria whose all of the total 246 bp matched were Lactococcus lactis (taxonomy ID 1358), Lactococcus lactis subsp. cremoris (taxonomy ID 1359), Lactococcus taiwanesis (taxonomy ID 1151742), and Lactococcus kimuchii (taxonomy ID 2568007). However, most of the bacteria were Lactococcus lactis and Lactococcus lactis subsp. cremoris. Next, we investigated the influence of these strains on bile acid metabolisms using Kyoto Encyclopedia of Genes and Genomes (KEGG) organisms (https://www.genome.jp/kegg-bin/show_brite?br08601.keg+lgr↗). There were total of 13 strains of Lactococcus lactis and Lactococcus lactis subsp. cremoris registered in KEGG, including subspecies, and none of them showed promotion of the synthesis of secondary bile acids in KEGG pathway analysis. On the other hand, it has been reported that Lactococcus lactis and Lactococcus lactis subsp. cremoris may have bile acid resistance (16), and those with bile acid resistance express bile salt hydrolase (BSH) (17). It is known that primary bile acids are metabolized to secondary bile acids by bacteria having bile acid 7α-dehydroxylase after being deconjugated by BSH. Lactococcus lactis and Lactococcus lactis subsp. cremoris do not directly synthase secondary bile acids but may be involved in the increase in secondary bile acids in feces. To support this, the percentages of bacteria with bile acid dehydroxylase increased in the feces of HFD mice in this study. The Lachnospiraceae was increased in feces of HFD mice, and the sequences which were determined as Lachnospiraceae were extracted from FASTA file. This sequence was matched to Dorea by the Greengene database v13.8, and Dorea is known to have produced secondary bile acids by dehydroxylation (18). That is, these Lactococcus lactis and Lactococcus lactis subsp. cremoris may eventually be involved in the increase of secondary bile acids. In the future, it will be necessary to research the influence of the combination of these bacteria.

To establish the mechanism underlying improvement in NAFLD induced by IBATi, we recolonized antibiotic (ATB) solution-treated mice reared in specific-pathogen-free conditions by fecal microbiome transplantation (FMT) using stool from HFD or HFD plus IBATi groups. After FMT, both groups were fed HFD for 4 weeks. The body weights of both groups increased similarly, and the liver weight/body weight ratios of the groups did not differ at the end of the experiment (Fig. 2A). Fatty changes were observed in livers excised from mice with FMT from HFD mice, whereas this change was not detected in livers of mice with FMT from HFD plus IBATi mice (Fig. 2B). Lipid droplets were observed in histological liver sections of mice with FMT from HFD mice but not in mice with FMT from HFD plus IBATi mice (Fig. 2C and D). The average NAS of livers from mice with FMT from HFD mice (5.67 ± 2.08) was significantly higher than that from mice with FMT from HFD plus IBATi mice (2.67 ± 0.58) (P = 0.033) (Fig. 2E). As shown in Fig. S1G, fibrosis was not observed in liver sections of either FMT group (Fig. 2F). AST, ALT, total cholesterol, triglycerides, LDL, and HDL levels in sera from mice with FMT from HFD plus IBATi mice were lower than those from mice with FMT from HFD mice (Fig. 3A). In addition, mice with FMT from HFD plus IBATi mice exhibited elevated hepatic Cyp7a1 and reduced ileal Fgf15 mRNA expression (Fig. 3B and C). In Fig. 3D, ωMCA in feces from mice with FMT from HFD plus IBATi mice was significantly reduced compared to the feces from mice with FMT from HFD mice.

The α-diversity in mice with FMT from HFD mice was reduced to the same level as that in HFD mice, but the α-diversity in mice with FMT from HFD plus IBATi mice was maintained even when fed HFD (Fig. 4A). β-Diversity differed significantly between the two groups (PERMANOVA, P = 0.007) (Fig. 4B). The relative abundance of Bacteroidaceae at the family level was decreased in mice with FMT from HFD plus IBATi mice (P = 0.00069). The percentage of Streptococcaceae in total OTUs from FMT mice with FMT from HFD plus IBATi was decreased from that of FMT mice with FMT from HFD mice (Fig. 4C). However, based on LEfSe, Streptococcaceae was not associated with mice with FMT from HFD mice (Fig. 4D). The ratio of Firmicutes to Bacteroidetes was the same in both groups (Fig. 4E). We additionally examined the effect of elobixibat in mice receiving long-term antibiotics. Mice receiving long-term antibiotics (6 weeks) were transplanted with gut microbiota from HFD or HFD plus IBATi groups of mice, and both groups of mice were fed HFD for 4 weeks. The gut microbiota fecal transplant from HFD plus IBATi mice prevented hepatic steatosis, even if mice receiving long-term antibiotics were fed HFD (Fig. 5). These results indicated that the gut microbiota fecally transplanted from HFD plus IBATi mice prevented hepatic steatosis caused by HFD.

Next, we tested the therapeutic effect of IBATi in HFD mice. Six-week-old mice were fed HFD for 12 weeks and received IBATi treatment (IBATi-Tx group) between weeks 6 to 12. The control group was treated with the same dose of phosphate-buffered saline (PBS). The IBATi-Tx group did not exhibit the same increase in body weight as the control group; their weights remained stable with IBATi treatment. Liver weight/body weight ratios in the control group increased, but those in the IBATi-Tx group decreased after IBATi treatment (Fig. 6A). Fatty changes observed in the macroscopic photographs of liver sections from the control group were prevented in the IBATi-Tx group (Fig. 6B to D). The average NAS of livers from the IBATi-Tx group (1.66 ± 0.57) was improved compared with the control group, and fibrosis was not detected in either group (Fig. 6E and F). Lower serum AST, ALT, total cholesterol, and LDL levels were detected in the IBATi-Tx group. Other biological parameters in the sera did not differ between treatment groups (Fig. 7A). Elevated hepatic Cyp7a1 and reduced ileal Fgf15 mRNA expression were detected in mice after IBATi treatment (Fig. 7B and C). In Fig. 7D, ωMCA in feces from the IBATi-Tx group of mice was significantly reduced.

Comparing the gut microbiota in both treatment groups, the α-diversity in the IBATi-Tx group was changed to the same levels as the control diet group, and β-diversity differed significantly between treatment groups (PERMANOVA, P = 0.036) (Fig. 8A and B). The relative abundance at the family level of Streptococcaceae was increased in the IBATi-Tx group (Fig. 8C and D). Moreover, the ratio of Firmicutes to Bacteroidetes was significantly decreased after IBATi-Tx treatment (Fig. 8E). Thus, the therapeutic effect of IBATi against hepatic steatosis was reproduced in HFD-fed mice.

NAFLD is highly prevalent worldwide, but effective therapeutic agents for its treatment are lacking. HFD mice treated with IBATi exhibited significantly suppressed body weight gain, liver dysfunction, and serum LDL levels, and their NAFLD activity scores were significantly decreased. The reduced α-diversity in the gut microbiota induced by HFD was recovered by IBATi treatment. IBATi treatment also ameliorated the increased ileal Fgf15 and decreased hepatic Cyp7a1 mRNA levels in HFD mice. Fecally transported gut microbiota from HFD plus IBATi mice prevented hepatic steatosis caused by HFD, indicating that IBATi protected against hepatic steatosis by ameliorating gut microbiota dysbiosis in NAFLD model mice.

All animal experiments were carried out in compliance with the International Council for Laboratory Animal Science (ICLAS) Ethical Guideline for Researchers and Japanese regulations. The local institutional animal ethics board of Osaka Medical College approved all mouse experiments (approval number 2019-138). Five-week-old male C57BL/6 mice were obtained from Japan SLC, Inc. (Shizuoka, Japan). All animal experiments were performed at the animal facility in Osaka Medical College, where mice were housed under specific-pathogen-free conditions. After acclimation for 1 week, mice were divided into four groups (n = 6): two groups were fed the control diet (CE-2; CLEA Japan Inc., Tokyo, Japan), and two groups were fed HFD (HFD32; CLEA Japan, Inc.) for 12 weeks (34). For the IBATi and HFD plus IBATi groups, mice were treated with 2.5 μmol/kg of the oral IBATi elobixibat (EA Pharma Co. Ltd., Tokyo, Japan) in 0.2 ml PBS by daily oral gavage from 6 to 12 weeks (5). Mice in the control diet groups were subjected to the same oral dose of PBS. Body weight and food consumption were monitored weekly throughout the experiments.

Elobixibat has only 2% bioavailability and is excreted from the body within 48 h after oral administration. When absorbed, elobixibat is protein bound in plasma (>99%), with a half-life of less than 4 h (35, 36). For fecal microbiota transplantation (FMT), feces were collected from the 12 week-HFD and HFD plus IBATi groups 48 h after the last oral administration of elobixibat and stored at −80°C. Six-week-old male C57/B6 mice (n = 10) were fed HFD for 3 weeks, after which they were treated with ATB solution containing ampicillin (1 mg/ml), streptomycin (5 mg/ml), and colistin (1 mg/ml) added to their sterile drinking water. Solutions and bottles were changed 3× per week. ATB treatment was discontinued 48 h before the first FMT treatment (37). FMT was performed by thawing fecal materials, and 500 μl of fecal suspension (50 mg/mouse) was then transferred by oral gavage into each ATB-treated recipient 3× every 2 days for 1 week. Mice were then fed HFD for an additional 4 weeks.

In some experiments, mice treated with ATB solution containing ampicillin (1 mg/ml), streptomycin (5 mg/ml), and colistin (1 mg/ml) added to their sterile drinking water for 6 weeks were transplanted with gut microbiota from HFD or HFD plus IBATi groups of mice (n = 5 each), and both groups of mice were fed HFD for 4 weeks.

Fresh feces were collected, placed into tubes, and stored at −80°C until further use. Bacterial genomic DNA was extracted from fecal samples as previously described (38), using a PureLink microbiome DNA purification kit (Thermo Fisher Scientific, Tokyo, Japan). Preparation of the library for DNA sequencing of the 16S RNA gene was conducted as previously described (39) using an Ion 16S Metagenomics kit (Thermo Fisher Scientific) to amplify hypervariable regions from the polybacterial samples according to the manufacturer’s protocol. Library preparation was performed using 50 ng pooled amplicons for each sample with the Ion Plus fragment library kit (Thermo Fisher Scientific). The StepOnePlus real-time PCR system (Thermo Fisher Scientific) was used to perform quantitative PCR amplification.

Sequence data analysis was performed using QIIME v1.9.0 as previously described (39). Statistical differences (P < 0.05) in the relative abundance of bacterial phyla and genera between groups were evaluated using Student’s paired t tests. Shannon phylogenetic diversity indices were calculated using the R phyloseq package and statistically analyzed using Kruskal-Wallis tests followed by Steel-Dwass post hoc tests. β-Diversity was estimated using Bray-Curtis distances between samples, visualized by PCoA, and statistically examined using permutational analysis of variance (PERMANOVA). The final figures were generated using QIIME v1.9.0. LEfSe was determined using default values (a value of 0.05 for both the factorial Kruskal-Wallis test among classes and the pairwise Wilcoxon test between subclasses; threshold value of 2.0 for the logarithmic LDA score for discriminative features).

Mice were euthanized before tissue sampling, and liver was obtained from these mice. Liver tissue samples were fixed in 20% formalin and embedded in paraffin. Sections (4 μm thick) were cut and stained with H&E. Sections (8 μm thick) from frozen liver tissue samples were stained with oil red O. Histological features and the fat area in liver tissue sections were determined using a BZ-X710 fluorescence microscope (Keyence, Osaka, Japan). Blinded histological assessment of the slides was performed by an experienced hepatopathologist. Hepatic steatosis was evaluated by NAS (40), and hepatic fibrosis was evaluated by fibrosis stage (41).

Blood was obtained from mice that were euthanized before tissue sampling, and serum was obtained from each mouse. Serum concentrations of AST, ALT, total BAs, total cholesterol, LDL, HDL, and triglycerides were measured using ELISA kits from Elabscience (Houston, TX, USA) and MyBioSource, Inc. (San Diego, CA, USA) following the manufacturers’ instructions. Fecal bile acid pool composition profiling of each group of mice was assayed by mass spectrometry (TechnoSuruga Laboratory Co., Ltd., Shizuoka, Japan).

Liver and terminal ileum samples were obtained from each group of mice after the anesthetization and prepared for analysis using an RNeasy minikit (Qiagen, Tokyo, Japan). The StepOnePlus real-time PCR system (Thermo Fisher Scientific) was used to perform quantitative PCR amplification. Hepatic mRNA expression (cholesterol 7-a-monooxygenase [Cyp7a1], cholesterol 8-a-monooxygenase [Cyp8a1], and farnesoid X receptor [Fxr]), and ileal mRNA expression (Fxr and fibroblast growth factor 15 [Fgf15]) were measured by real-time PCR using CYBR Green real-time PCR master mix (Thermo Fisher Scientific). The thermal cycling conditions included an initial denaturation step 95°C for 3 min, followed by 40 cycles at 95°C for 10 s and 56°C for 30 s.

All data are presented as the mean ± SEM. Body weight gain curves were analyzed using ANOVA, followed by Tukey’s multiple-comparison tests. Pearson’s rank correlation coefficient was used to analyze the correlation between gut microbiota. Independent samples t tests or Wilcoxon-Mann-Whitney U tests were used for differences between continuous variables. Statistical analyses were performed using Statistics Analysis Software (JMP Institute, Inc., Cary, NC, USA). All results were considered statistically significant at a P value of <0.05.

The data sets used and/or analyzed in the study are available at NCBI BioProject (accession numbers PRJNA714695↗ and PRJNA714835↗).