Critical Role of Cholic Acid in the Development of iHFC Diet-induced MASH in TSNO Mice

Abstract

A high-fat/cholesterol/cholate-based (iHFC) diet induces pathological changes in Tsumura-Suzuki non-obese (TSNO) mice, resembling human metabolic dysfunction associated steatohepatitis (MASH), along with advanced liver fibrosis. In this study, we investigated the role of cholic acid (CA) in the development of iHFC diet-induced MASH development. In mice receiving an iHFC diet without CA (CA(-) iHFC diet), both lobular inflammation and fibrosis progression in the liver were attenuated compared to those on the standard iHFC diet. Notably, hepatocyte ballooning was significantly improved in the CA(-) iHFC diet group. The expression levels of genes associated with inflammation and fibrosis were lower in the livers of CA(-) iHFC diet-fed mice compared to those fed the iHFC diet. Furthermore, there were no significant changes in the proportion and number of monocyte-derived macrophages in the livers of CA(-) iHFC diet-fed mice relative to those in the ND (normal diet)-fed group. The co-localization of CD11c⁺ macrophages with collagen fibers in the livers of CA(-) iHFC diet-fed mice did not significantly differ from that of the ND-fed group. Moreover, the CA(-) iHFC-fed mice exhibited a distinct microbial composition relative to both ND- and iHFC-fed mice. Finally, the increase in deoxycholic acid in fecal samples and the reduced hepatic expression of Cyp27a1 and Cyp7a1 induced by the iHFC diet were less in the CA(-) iHFC-fed group. These results suggest that CA modulates iHFC diet-induced MASH development by influencing the accumulation of monocyte-derived macrophages in the liver and shaping the gut microbiota composition and bile acid profile.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10753-025-02294-5.

Introduction

Metabolic dysfunction associated steatohepatitis (MASH), a severe manifestation of metabolic dysfunction associated steatotic liver disease (MASLD), is a major public health issue owing to its increasing prevalence and progression to cirrhosis and hepatocellular carcinoma [1]. MASH is characterized by hepatic steatosis, inflammation, hepatocellular injury, and fibrosis. The global prevalence of obesity and metabolic syndrome has contributed significantly to the increasing incidence of MASLD, with MASH being a key area of research and clinical evaluation [2]. Therefore, elucidating the mechanisms that drive inflammation and fibrosis is critical for the development of effective therapeutic strategies for MASH.

Macrophages play a key role in MASH [3]. These immune cells drive chronic inflammation, fibrosis, and hepatocellular damage in the MASH [3]. Kupffer cells (KCs), liver-resident macrophages, and infiltrating monocyte-derived macrophages (MdMs) exhibit diverse phenotypes and either promote or resolve inflammation [4, 5]. During MASH progression, the accumulation of pro-inflammatory macrophage subsets, such as CD11c⁺ macrophages, that interact with hepatic stellate cells is increased, driving fibrogenesis and the deposition of extracellular matrix proteins, including collagen [6, 7]. Therefore, understanding the specific roles of different macrophage subsets and their interactions with other hepatic cells is crucial for developing effective MASH treatment methods. Recent studies have revealed the dynamic nature of macrophage polarization in response to the metabolic and inflammatory milieu of the liver [5, 8, 9], suggesting complex roles of macrophages in MASH.

Bile acids (BA) are amphipathic molecules synthesized from cholesterol in the liver. They play fundamental roles in the emulsification and absorption of dietary fat and fat-soluble vitamins in the intestine [10]. In addition to performing digestive functions, BAs act as signaling molecules that modulate key physiological processes, such as lipid and glucose metabolism, energy expenditure, and immune response [11]. BA synthesis is tightly regulated by a feedback mechanism involving the farnesoid X receptor (FXR), which inhibits the rate-limiting enzyme involved in BA synthesis, cholesterol 7α-hydroxylase 1 (Cyp7a1), thereby maintaining BA homeostasis [12]. Patients with MASH exhibit altered BA profiles characterized by an increased hydrophobic-to-hydrophilic BA ratio, which exacerbates liver injury [13]. Hydrophobic BAs such as deoxycholic acid (DCA) and chenodeoxycholic acid are particularly hepatotoxic because they induce oxidative stress, mitochondrial dysfunction, and apoptosis [14–16]. In contrast, hydrophilic BAs, such as ursodeoxycholic acid, exert cytoprotective effects and mitigate BA-induced liver injury [17]. BA signaling via FXR and transmembrane G protein-coupled receptor 5 (TGR5) plays a critical role in regulating hepatic inflammation and fibrosis [10, 18]. FXR activation exerts anti-inflammatory and anti-fibrotic effects by inhibiting pro-inflammatory cytokine production and hepatic stellate cell activation, which are key events in liver fibrosis [18, 19]. TGR5 modulates immune responses by suppressing pro-inflammatory cytokine production and promoting the release of anti-inflammatory mediators [20, 21]. Dysregulation of these signaling pathways may be responsible for persistent inflammation and fibrosis in MASH.

The role of the gut microbiota in MASH progression has garnered increasing attention. The gut microbiota affect BA metabolism, immune responses, and hepatic inflammation, which are crucial for MASH development. Recent studies have reported dysbiosis of the gut microbiota, with an increased abundance of pro-inflammatory bacteria and a decreased abundance of beneficial bacteria, leading to enhanced intestinal permeability and endotoxemia in patients with MASH [22–24]. Gut dysbiosis may contribute to systemic inflammation and liver injury in patients in MASH. Furthermore, gut–liver axis plays a significant role in MASH pathogenesis, with BAs serving as a critical link between the gut microbiota and liver inflammation [25]. Gut microbiota composition affects BA metabolism, and microbiota alterations change the BA pool and promote liver injury [26]. Dysbiosis, characterized by overgrowth of BA hydrolase-producing bacteria, increases BA deconjugation, leading to the accumulation of toxic free BAs in the liver [27]. BAs regulate gut permeability and inflammation via FXR and TGR5 signaling in the intestine. FXR activation in the gut improves gut barrier integrity by upregulating the expression of tight junction proteins, thereby preventing the translocation of bacterial endotoxins that trigger hepatic inflammation [28]. TGR5 activation in intestinal immune cells modulates cytokine production, thereby affecting liver inflammation [21]. Therefore, targeting BA signaling in the gut–liver axis is a promising therapeutic strategy for MASH.

Histologically, liver fibrosis typically originates in zone 3, extends into the portal area, and eventually manifests as bridging fibrosis, connecting different portal areas and the central vein in patients with MASH [29]. However, few animal models have accurately replicated the histological features of human MASH, limiting our understanding of its pathogenesis and hindering the development of effective treatment methods. The Tsumura–Suzuki non-obese (TSNO) mouse MASH model, established using a high-fat/cholesterol/cholate-based (iHFC) diet, mimics the metabolic and histological characteristics of human MASH and is used to elucidate disease progression mechanisms [30]. We previously demonstrated two distinct macrophage subsets in the liver of iHFC diet-fed TSNO mice: CD11c⁺/Ly6C⁻ macrophages, which contribute to fibrosis and hepatic stellate cell activation, and CD11c⁻/Ly6C⁺ macrophages, which are associated with anti-inflammatory responses and tissue repair [31]. A previous study on iHFC diet-fed TSNO mice revealed the crucial role of alterations in the gut microbiota in the enhanced accumulation of CD11c⁺/Ly6C⁻ macrophages, thereby driving inflammation and advanced fibrosis in MASH [32]. Furthermore, Tsumura–Suzuki obese diabetes (TSOD) mice are less susceptible to MASH progression on an iHFC diet and exhibit lower gut microbiota diversity than TSNO mice [33]. Additionally, TSOD mice exhibited low levels of BAs associated with intestinal barrier disruption upon iHFC diet intake [33]. Therefore, BA metabolism plays a crucial role in MASH progression under an iHFC diet; however, the specific role of CA in MASH progression remains unclear.

In this study, we investigated the role of CA in the progression of iHFC diet-induced MASH, which mimics the pathology of human MASH. We established an iHFC diet without CA (CA(-) iHFC diet) and compared its effects with those of the standard iHFC diet. We also evaluated MASH pathology via histological analysis of the liver tissues. Additionally, we performed PCR analysis of inflammatory and fibrotic gene expression levels, flow cytometry to quantify macrophage accumulation in the liver, 16S rRNA sequencing analysis of the gut microbiota, and BA composition analysis of feces.

Materials and Methods

Mice

Animal experiments were performed in compliance with the ARRIVE reporting guidelines, which include detailed reporting of study design, statistical analysis, experimental procedures, animal care, and housing. Additionally, the experiments adhered to the guidelines for the Proper Conduct of Animal Experiments established by the Science Council of Japan. The animal protocols were approved by the Ethics Committee for Animal Experiments at Toyama Prefectural University (R1-3 and R4-1).

Six-week-old male TSNO mice were obtained from the Institute of Animal Reproduction (Ibaraki, Japan) and housed under specific pathogen-free conditions at the Toyama Prefectural University animal facility. The mice were provided with unrestricted access to food and water and maintained under a standard 12 h light/dark cycle. At seven weeks of age, the mice were fed one of the following diets: a high-fat, high-cholesterol, high-cholate (iHFC) diet (composed of 69.5% standard chow, 28.75% palm oil, 1.25% cholesterol, and 0.5% cholate) from Hayashi Kasei (Osaka, Japan); a cholate-free high-fat, high-cholesterol (CA(-) iHFC) diet (70.0% standard chow, 28.75% palm oil, and 1.25% cholesterol) from Hayashi Kasei or a normal diet (ND) from Oriental-Yeast (Tokyo, Japan). At the conclusion of the experiments, the mice were anesthetized with isoflurane, and samples were collected for subsequent analysis.

Plasma Biochemical Analysis

Blood was collected from the inferior vena cava, and plasma was separated. Plasma levels of alanine aminotransferase (ALT), total cholesterol (T-CHO), and triglycerides (TG) were measured using a DRI-CHEM NX700 analyzer (Fujifilm, Tokyo, Japan), following the manufacturer’s protocol.

Isolation of Non-Parenchymal Cells from the Liver

To separate non-parenchymal cells from the liver, mice were anesthetized with isoflurane and perfused with PBS. The isolation of non-parenchymal cells was performed using a Liver Dissociation Kit (Miltenyi Biotech, Bergisch Gladbach, Germany), following the manufacturer’s instructions. The resulting cell suspension was passed through a 100 μm cell strainer and subsequently prepared for flow cytometry analysis.

Flow Cytometry Analysis

Non-parenchymal cells (1.25 × 10⁵) were pre-incubated with anti-mouse FcγR (clone 2.4G2) to block the binding of fluorescence-labeled antibodies to FcγR. After a 20-min incubation, the cells were stained with the respective antibodies at their predetermined optimal concentrations. To exclude non-viable cells, 7-Amino-actinomycin D (7-AAD) from BD Biosciences (San Diego, CA, USA) was used. Flow cytometry analysis was conducted using a FACSCanto II flow cytometer (Becton Dickinson, Mountain View, CA, USA). The data were processed with FlowJo software (BD Biosciences). Details of the antibodies used for flow cytometry are provided in Supplementary Table 1.

Preparation of RNA and cDNA

Total RNA was extracted using the NucleoSpin RNA Mini Kit (Macherey–Nagel, Düren, Germany), following the manufacturer’s instructions. The isolated RNA was subsequently reverse-transcribed using the PrimeScript® RT Reagent Kit (Takara Bio, Shiga, Japan) following the provided protocol.

Quantitative Real-Time PCR

Quantitative real-time PCR (qRT-PCR) was performed using the FastStart Universal Probe Master (Roche Applied Science, Mannheim, Germany), following the manufacturer’s protocols. Reactions were analyzed on the CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Transcript levels were normalized to the Hprt mRNA as an internal control. The TaqMan primers and probes (Applied Biosystems, Carlsbad, CA, USA) used for the analysis are listed in Supplementary Table 2.

Histological and Immunohistochemistry Analysis

Liver tissues were harvested and immediately fixed in 4% formaldehyde. The fixed samples were embedded in paraffin, sectioned into 6-μm thick slices, and mounted on glass slides. The sections were stained with hematoxylin and eosin or Sirius red using standard protocols. Histopathological findings were scored using the NASH Clinical Research Network Scoring System based on four semi-quantitative factors: steatosis (0–3), lobular inflammation (0–3), hepatocyte ballooning (0–2), and fibrosis (0–4) as previously described [34]. The NAFLD activity score (NAS) was defined as the unweighted sum of the scores for steatosis, lobular inflammation, and hepatocyte ballooning; thus, scores ranged from 0 to 8. A NAS of 0 to 2 was considered not diagnostic of steatohepatitis, and scores of 5 or greater were diagnostic of steatohepatitis [34]. Histological assessments were performed by K.T. and M.I-S, who assigned scores and grades in a blinded manner.

Fluorescent Immunohistochemistry Analysis

Frozen tissue Sects. (7-μm thick) were incubated with anti-CD11c (Invitrogen), followed by treatment with a secondary antibody (anti-hamster IgG, Southern Biotech, Birmingham, AL, USA) and processed using the TSA Fluorescein System (Akoya Biosciences, Marlborough, MA, USA). Additionally, the sections were stained with anti-collagen type 1 (Novotec, Bron, France), followed by a secondary antibody (anti-rabbit IgG Alexa Fluor 594, Abcam). Finally, the sections were counterstained with DAPI (Invitrogen). Images were captured using a BX50 microscope and imaging system (Olympus, Tokyo, Japan). Positive signals within co-localized areas were identified and analyzed according to the method described by Tolivia et al. [35] using Adobe Photoshop CS software, version 8.

Metagenomic 16S rRNA Sequencing

Fecal DNA was extracted using the method described in a previous study [33]. Libraries for 16S rRNA gene sequencing were prepared following the Illumina protocol (San Diego, CA, USA), as outlined in prior research. The prepared libraries were pooled and sequenced on the MiSeq System using a 500-cycle kit (Illumina).

Bacterial Community Analysis

The bacterial composition was analyzed using the Quantitative Insight into Microbial Ecology 2 (QIIME2) pipeline, a platform for processing raw DNA sequencing data from microbiome studies [33]. Taxonomic assignment was performed employing the SILVA138 reference database. Taxonomic classification was performed using the SILVA138 reference database.

Bile Acid Analysis

Fecal BAs were extracted by homogenizing dried fecal samples (30 – 60 mg) in a solution containing 0.5 mL methanol, 0.8 mL acetonitrile, and 0.2 mL of 28% (w/v) ammonium hydroxide, with 100 nmol 23-dinor-deoxycholic acid included as an internal standard (Steraloids, Newport, RI, USA). The homogenates were centrifuged, and the resulting supernatants were processed through solid-phase extraction columns to isolate BA-containing fractions, as previously described [36]. Quantification of fecal BAs was performed using liquid chromatography-electrospray ionization-mass spectrometry (LC–ESI–MS), following established protocols [36].

Statistical Analysis

Statistical significance was evaluated using a two-way ANOVA followed by Tukey's post-hoc test for multiple comparisons, or an unpaired Student’s t-test for comparisons between groups. Analyses were conducted using GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). An F-test was performed in Prism 9 to confirm that the p-value from the F-test exceeded 0.05, indicating homogeneity of variance. A p-value of < 0.05 was considered statistically significant. Results are presented as the mean ± standard deviation (SD).

Results

CA is Essential for iHFC Diet-Induced Liver Damage

We examined the role of CA in the iHFC diet-induced changes in liver weight, hepatocellular injury, and abnormalities of lipid metabolism. Over 4–12 weeks of feeding, the iHFC diet significantly increased the body weight compared to the ND (Fig. 1A, left panel). Similarly, the CA-deficient iHFC diet (CA(-) iHFC) significantly increased body weight compared to the ND (Fig. 1A, left panel). A significant difference in food intake between the ND- and iHFC diet-fed groups was observed only after 24 weeks of feeding (Fig. 1A, right panel). Mice fed the CA(-) iHFC diet showed a significant increase in liver weight compared to those fed the ND (Fig. 1B); however, this increase was less pronounced than that observed in the iHFC diet-fed group (Fig. 1B). Macroscopically, livers of iHFC diet-fed mice were enlarged and pale, and these changes were less prominent in the livers of CA(-) iHFC diet-fed mice (Fig. 1C). Although iHFC diet significantly elevated the plasma ALT levels, no significant differences were observed between the CA(-) iHFC diet and ND (Fig. 1D, left panel). After 24 weeks of feeding, the iHFC diet increased plasma T-CHO levels, which was not observed in the CA(-) iHFC diet-fed group (Fig. 1D, middle panel). Furthermore, plasma TG levels in the CA(-) iHFC diet-fed group were significantly lower than those in the ND- and iHFC diet-fed groups after 24 weeks of feeding (Fig. 1D, right panel). These results highlighted the key role of CA in iHFC diet-induced liver enlargement, hepatocellular damage, and lipid metabolism dysregulation.

Fig. 1.

Impact of CA on liver enlargement, liver damage, and lipid metabolism abnormalities induced by HFC diet. A. Body weight (n = 6 or 9) and food intake (n = 3) were measured in TSNO mice fed either the ND, CA(-) iHFC diet or iHFC diet. B. Liver weights of TSNO mice fed either the ND, CA(-) iHFC diet or iHFC diet (n = 6). C. Representative photos of livers from TSNO mice fed either the ND, CA(-) iHFC diet or iHFC diet. D. Plasma ALT, T-CHO, and TG levels were measured for TSNO mice fed either the ND, CA(-) iHFC diet or iHFC diet (n = 6). *p < 0.05, **p < 0.01, ***p < 0.001

CA Plays Important Roles in iHFC Diet-Induced Lobular Inflammation, Hepatocyte Ballooning, and Fibrosis

We explored the role of CA in iHFC diet-induced MASH development via pathological and histological analyses. Notably, 12 weeks of iHFC diet feeding caused steatosis, lobular inflammation, and hepatocyte ballooning at levels comparable to those caused by 24 weeks of feeding (Fig. 2A and 2B), with a significant increase in the NAS score compared to that caused by the ND (Fig. 2B). Mice fed the CA(-) iHFC diet exhibited similar levels of steatosis, but showed significantly reduced lobular inflammation after 12 weeks of feeding (Fig. 2B). Moreover, hepatocyte ballooning was markedly alleviated in the CA(-) iHFC diet-fed group (Fig. 2B), resulting in a significantly lower NAS score (Fig. 2B) than that in the iHFC diet-fed group.

Fig. 2.

Impact of CA on the histopathological changes of MASH induced by iHFC diet. A. Representative histological images of hematoxylin and eosin-stained liver sections. Scale bars, 100 mm. B. Steatosis (0 to 3), lobular inflammation (0 to 3), hepatocyte ballooning (0 to 2), and NAS score were assessed according to the criteria proposed by Kleiner et al. as described in Materials and Methods (n = 3). C. Representative histological images of Sirius red-stained liver sections. Scale bars, 100 µm. D. Five locations were captured per three liver sections for each group. The positive areas for Sirius red were then quantified at 15 locations using ImageJ software. E. Liver fibrosis (0 to 4) was evaluated (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001

Advanced liver fibrosis was observed in the iHFC diet group after 12 weeks (Fig. 2C, D, and E). However, the fibrotic changes were significantly attenuated in the CA(-) iHFC diet-fed group (Fig. 2C, 2D, and 2E). At week 24, the fibrosis grade in the CA(-) iHFC diet-fed group was similar to that in the iHFC diet-fed group (Fig. 2E). These findings suggested that CA plays a critical role in MASH progression, particularly in hepatocyte ballooning, mid-stage lobular inflammation, and fibrosis. However, CA did not influence iHFC diet-induced steatosis development.

CA Plays Important Roles in iHFC Diet-Induced Inflammation- and Fibrosis-Related Gene Expression in the Liver

The expression levels of a pro-inflammatory gene (Tnf), a chemokine gene (Ccl2), M1 macrophage marker genes (Nos2 and Itgax), and a macrophage marker gene (Adgre1) were significantly upregulated in the livers of TSNO mice fed the iHFC diet for 12 or 24 weeks compared with those in the livers of TSNO mice fed the ND (Fig. 3A). The expression levels of collagen type 1 (Col1a1) and tissue inhibitor of metalloproteinase 1 (Timp1), which are associated with fibrosis, were also markedly elevated in the iHFC diet-fed mice compared with those in the ND-fed mice (Fig. 3A). In contrast, the CA(-) iHFC diet-fed group did not show a significant increase in the level of these genes, except for Tnf and Adgre1, after 24 weeks of feeding (Fig. 3A), compared with the ND-fed group. Moreover, the gene levels in the CA(-) iHFC diet-fed group were significantly lower than those in the iHFC diet-fed group, except for Nos2 levels after 12 weeks of feeding (Fig. 3A).

Fig. 3.

Impact of CA on the expression of inflammation- or fibrosis-associated genes in the livers induced by iHFC diet. A. RT-qPCR of TNF-α (Tnf), iNOS (Nos2), CD11c (Itgax), MCP-1 (Ccl2), F4/80 (Adgre1), collagen type 1 (Col1a1), and TIMP-1(Timp1) mRNA (n = 6). B. RT-qPCR of FAS (Fasn), SREBP1 (Srebf1), and Scarb3 (Cd36) mRNA. *p < 0.05, **p < 0.01, ***p < 0.001

Fasn levels in the liver were lower in CA(-) iHFC diet-fed mice than in ND-fed mice, similar to the reduction observed in iHFC diet-fed mice (Fig. 3B). This reduction is likely due to the intake of saturated fatty acids from the diet, as both the iHFC diet and the CA(-) iHFC diet contain the same amount of palm oil, most of which consists of saturated fatty acids. The levels of Srebf1, another fatty acid synthesis gene, at 12 weeks were also lower in the CA(-) iHFC diet-fed mice than in the ND-fed mice (Fig. 3B). However, the levels of the fatty acid uptake gene Cd36 at 24 weeks were significantly higher in the CA(-) iHFC diet-fed group than in the ND-fed group (Fig. 3B). Therefore, CA upregulated inflammation- and fibrosis-related gene levels but did not significantly affect lipid-regulating gene levels in the liver in response to the iHFC diet.

CA Plays an Important Role in iHFC Diet-Induced Leukocyte Accumulation in the Liver

Next, we examined the effects of CA on immune cell dynamics in the liver of TSNO mice. The numbers of viable non-parenchymal cells in the total liver (Fig. 4A, left panel) and per liver weight (Fig. 4A, right panel) were significantly increased by the iHFC diet compared to the ND, with a peak observed at 8 weeks of feeding. However, in the CA(-) iHFC diet-fed group, no significant increase in the number of viable cells was observed compared to the ND-fed group (Fig. 4A, left and right panels). After 4, 8, and 12 weeks of feeding, the CA(-) iHFC diet-fed group exhibited a marked decrease in the number of viable non-parenchymal cells compared with the iHFC diet-fed group (Fig. 4A, left and right panels). We further analyzed the accumulation of CD45⁺ leukocytes in the liver using flow cytometry. The percentage of CD45⁺ cells increased, whereas that of CD45⁻ cells decreased in the iHFC diet-fed group compared to the ND-fed group (Fig. 4B). However, these changes were significantly attenuated in the CA(-) iHFC diet-fed group compared to those in the iHFC diet-fed group (Fig. 4B). Statistical analysis confirmed these differences in CD45⁺ and CD45⁻ cell populations (Fig. 4C, left panel; and 4D, left panel). Number of CD45⁺ cells was significantly higher in the iHFC diet-fed group than in the ND-fed group (Fig. 4C, right panel), whereas that of CD45⁻ cells showed a significant reduction only after 24 weeks of feeding (Fig. 4D, right panel). In contrast, the CA(-) iHFC diet-fed group did not exhibit a significant increase in the number of CD45⁺ cells compared to the ND-fed group (Fig. 4C, right panel), but the cell number was significantly lower than that in the iHFC diet-fed group (Fig. 4C, right panel). A significant difference in the number of CD45⁻ cells between the iHFC- and CA(-) iHFC diet-fed groups was observed only after 24 weeks of feeding (Fig. 4D, right panel). These findings suggest that CA is essential for the iHFC diet-induced accumulation of leukocytes in the liver.

Fig. 4.

Impact of CA on the accumulation of CD45⁺ leukocytes in the liver induced by iHFC diet. A. Cell number of live non-parenchymal cells of the livers from TSNO mice on either the ND, CA(-) iHFC diet or iHFC diet (n = 3). B. Representative flow cytometry data of CD45 expression on non-parenchymal cells of the livers from TSNO on either the ND, CA(-) iHFC diet or iHFC diet. C. Percentage (left) and cell number (right) of CD45⁺ live non-parenchymal cells were determined by flow cytometry analysis conducted in Fig. 4B (n = 3). D. Percentages (left) and cell numbers (right) of CD45.⁻ live non-parenchymal cells were determined by flow cytometry analysis conducted in Fig. 4B (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001

CA Plays a Crucial Role in iHFC Diet-Induced MdM Accumulation in the Liver

We compared KCs and MdMs within the CD45⁺ leukocyte population by examining the F4/80 and CD11b expression levels. In the liver of TSNO mice, KCs were characterized as F4/80^Hi/CD11b^Int, whereas MdMs were identified as F4/80^Int/CD11b^Int−Hi (Fig. 5A) [31]. The percentage of F4/80^Hi/CD11b^Int KCs significantly decreased after 8, 12, and 24 weeks in the iHFC diet-fed group compared to that in the CA(-) iHFC diet-fed group (Fig. 5B, left panel). However, the number of F4/80^Hi/CD11b^Int KCs increased after 4 and 8 weeks in the iHFC diet-fed group compared to the ND-fed group (Fig. 5B, right panel). After 24 weeks of the iHFC diet, the number of F4/80^Hi/CD11b^Int KCs was significantly lower than that in the CA(-) iHFC diet-fed group (Fig. 5B, right panel). In contrast, the percentage of F4/80^Hi/CD11b^Int KCs in the CA(-) iHFC diet-fed group was comparable to that in the ND-fed group (Fig. 5B, left panel), and the number of KCs increased at 8 and 24 weeks compared to that in the ND-fed group (Fig. 5B, right panel). The F4/80^Int/CD11b^Int−Hi MdM percentage significantly increased after 12 and 24 weeks in the iHFC diet-fed group compared with that in the ND-fed group (Fig. 5C, left panel). However, this increase was not observed in the CA(-) iHFC-diet-fed group (Fig. 5C, left panel). The number of F4/80^Int/CD11b^Int−Hi MdMs was also significantly increased in the iHFC diet-fed group compared to that in the ND-fed group (Fig. 5C, right panel). In contrast, the CA(-) iHFC diet did not show such a significant increase, and a notable reduction in the number of these cells was observed at 4, 8, and 12 weeks compared to the iHFC diet-fed group (Fig. 5C, right panel). These findings highlighted the critical role of CA in MdM accumulation in the liver in response to an iHFC diet.

Fig. 5.

Impact of CA on the dynamics of KCs and MdMs in the liver induced by iHFC diet. A. Representative flow cytometry data of F4/80 and CD11b expression on CD45⁺ non-parenchymal cells including KCs of the liver from TSNO mice on either the ND, CA(-) iHFC diet,or iHFC diet. B. Percentage (left) and cell number (right) of F4/80^Hi/CD11b^Int KCs were calculated (n = 3). C. Percentage (left) and cell number (right) of F4/80^Int/CD11b.^Int−Hi MdMs were calculated (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001

CA Plays Critical Roles in iHFC Diet-Induced Accumulation of CD11c⁺/Ly6C⁻ and CD11c⁻/Ly6C⁺MdM Subsets in the Liver

Next, we analyzed F4/80⁺ cells, excluding dead cells and KCs with high autofluorescence (Supplementary Fig. 1A) [31]. Compared with the ND-fed group, the iHFC diet-fed group exhibited a significantly increased percentage of F4/80⁺ cells, excluding KCs (Supplementary Fig. 1A and B, left panel). However, this increase was not observed in the CA(-) iHFC-diet-fed group (Supplementary Fig. 1A and B, left panel). Moreover, the percentage of F4/80⁺ cells in the CA(-) iHFC-diet-fed group was significantly lower than that in the iHFC-diet-fed group (Supplementary Fig. 1B, left panel). A similar trend was observed for the total number of F4/80⁺ cells, reflecting the differences in percentages among the groups (Supplementary Fig. 1B, right panel).

In the livers of iHFC diet-fed TSNO mice, MdMs were categorized into three subsets based on CD11c and Ly6C expression levels [31]. iHFC diet-fed mice showed significantly increased proportions of CD11c⁺/Ly6C⁻ and CD11c⁻/Ly6C⁺ cells (Fig. 6A). In CA(-) iHFC diet-fed mice, the proportion of CD11c⁺/Ly6C⁻ cells was significantly reduced after 4 and 8 weeks of feeding compared to that in iHFC diet-fed mice; however, at 12 and 24 weeks, their proportions were comparable to or higher than those in iHFC diet-fed mice (Fig. 6A and 6B, left panel). Additionally, the number of CD11c⁺/Ly6C⁻ cells was not significantly increased at 4, 8, and 12 weeks in CA(-) iHFC diet-fed mice compared to that in ND-fed mice and was significantly lower than that in iHFC diet-fed mice (Fig. 6B, right panel). The percentage of CD11c⁻/Ly6C⁺ cells was significantly lower in CA(-) iHFC diet-fed mice than in iHFC diet-fed mice but was significantly increased at 12 and 24 weeks compared to that in ND-fed mice (Fig. 6A and 6C, left panel). A significant increase in the number of CD11c⁻/Ly6C⁺ cells was observed in the iHFC diet-fed group compared to the ND-fed group, but this change was not observed in the CA(-) iHFC diet-fed group (Fig. 6C, right panel). These findings suggest that CA is essential for iHFC diet-induced increases in the proportions of CD11c⁺/Ly6C⁻ and CD11c⁻/Ly6C⁺ MdM subsets.

Fig. 6.

Impact of CA on the dynamics of CD11c⁺/Ly6C⁻ and CD11c⁻/Ly6C⁺ MdM subsets in the liver induced by iHFC diet. A. Representative flow cytometry data of CD11c and Ly6C expression was on F4/80⁺ recruited macrophages of the liver from TSNO mice on either the ND, CA(-) iHFC diet or iHFC diet. B. Percentage (left) and cell number (right) of CD11c⁺/Ly6C⁻ cells were calculated (n = 3). C. Percentage (left) and cell number (right) of CD11c⁻/Ly6C.⁺ cells were calculated (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001

CA is Essential for iHFC Diet-Induced Colocalization of CD11c⁺MdMs with Collagen Fibers in the Liver

We previously performed RNA sequencing and revealed the critical role of CD11c⁺/Ly6C⁻ MdMs in fibrosis and hepatic stellate cell activation [31]. This macrophage subset is also involved in hepatic crown-like structure formation and colocalizes with collagen fibers [31]. Fluorescent immunostaining analysis revealed a significantly higher colocalization of CD11c-positive areas with collagen-positive areas in mice fed the iHFC diet than in those fed the ND for 12 weeks (Fig. 7A and 7B). However, this effect was not observed in the CA(-) iHFC-diet-fed group (Fig. 7A and 7B). These findings suggest that CA is critical for iHFC diet-induced interactions between CD11c⁺/Ly6C⁻ MdMs and collagen fibers, which explains the decreased fibrosis observed in the CA(-) iHFC diet-fed group (Fig. 2C, 2D, and 2E).

Fig. 7.

Impact of CA on iHFC diet-induced co-localization of CD11c.⁺ cells and collagen fibers in the livers. A. Representative histological images (10 × , 20 × , or 40 × magnification) of fluorescent immunohistochemistry for CD11c (green), collagen type 1 (red), and DAPI (blue) after 12 weeks of diet feeding. White scale bars, 250 µm (10 ×), 100 µm (20 ×), 50 µm (40 ×). B. Colocalization areas were calculated as described previously [31]. ****p < 0.0001

Gut Microbiota Composition of CA(-) iHFC-Fed Mice Differs from that of iHFC-Fed Mice

Next, we investigated the effects of CA on the iHFC diet-induced changes in the gut microbiota using DNA samples from fecal-derived microbiota. Notably, CA had no impact on the $\alpha$-diversity of the gut microbiota (Fig. 8A). In contrast, clear differences in the microbial community composition were observed among the three diet groups (Fig. 8B). Although no significant differences were observed, the iHFC diet-fed group showed a lower proportion of gram-positive bacteria and a higher proportion of gram-negative bacteria than the ND-fed group (Fig. 8C and 8D). In contrast, the CA(-) iHFC diet-fed group exhibited proportions intermediate between those of the ND- and iHFC diet-fed groups (Fig. 8C and 8D). ND- and iHFC diet-fed TSNO mice showed a high abundance of Firmicutes and Bacteroidota phyla (Supplementary Fig. 2A and 2B). However, the CA(-) iHFC diet-fed group exhibited no significant differences in the relative abundance of these phyla compared to the ND- and iHFC diet-fed groups (Supplementary Fig. 2A and 2B). The CA(-) iHFC diet-fed group exhibited a high abundance of the phylum Actinobacteria and a decreased abundance of the phylum Verrucomicrobiota; however, no significant differences were observed compared with the other two groups (Supplementary Fig. 2A and 2B). At the family level, the relative abundance of the families Tannerellaceae and Oscillospiraceae, both belonging to the phylum Firmicutes, was significantly higher in the iHFC diet-fed group than in the ND-fed group (Fig. 8E and 8F). However, this increase was not observed in the CA(-) iHFC-diet-fed group (Fig. 8E and 8F). Conversely, the abundance of the family Lactobacillaceae was significantly decreased in the iHFC diet-fed group compared to that in the ND-fed group; however, this change was not detected in the CA(-) iHFC diet-fed group (Fig. 8E and F). The relative abundance of some bacterial families either increased or decreased in the CA(-) iHFC diet-fed group compared to that in the other two groups. Specifically, the abundance of the Muribaculaceae family in the phyla Bacteroidota and the Erysipelotrichaceae family in the phylum Firmicutes was significantly increased, whereas that of the Rikenellaceae family in the phylum Firmicutes was significantly decreased (Fig. 8F) in the CA(-) iHFC diet-fed group compared to those in the other groups. Several differences in the relative abundance of bacterial species were observed among the three diet groups (Supplementary Fig. 3). Our results suggest that the gut microbiota composition in CA(-) iHFC diet-fed mice differs from that in iHFC diet-fed mice, possibly contributing to decreased MASH progression in TSNO mice.

Fig. 8.

Impact of CA on the changes in gut microbiota composition induced by the iHFC diet. Fecal samples were collected after 4 weeks of diet feeding. A. Effect of diet on the α-diversity (Observed features) of gut bacteria (n = 8). B. Principal component analysis of β-diversity values (n = 8). C. Ratio of Gram-positive bacteria, Gram-negative bacteria, and other bacteria. D. Comparison of the percentages of Gram-positive bacteria and Gram-negative bacteria (n = 8). E. Representation of relative abundance of bacterial family in the fecal microbiota (n = 8). A mean abundance of more than 1% was extracted. F. Relative abundance of bacterial families in the fecal microbiota (n = 8). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

BA Composition in the CA(-) iHFC Diet-Fed Mice Differs From that in the iHFC Diet-Fed Mice

To elucidate the mechanism underlying CA-induced MASH progression, we examined BA metabolism. Consistent with a previous report [32], the iHFC diet increased the relative abundance of DCA and decreased that of muricholic acid (MCA) (Fig. 9A). In the CA(-) iHFC diet-fed group, the relative abundance of DCA was comparable to that in the ND-fed group, but was lower than that in the iHFC diet-fed group (Fig. 9A). Conversely, the relative abundance of MCA was higher in the CA(-) iHFC diet-fed group than in the iHFC diet-fed group (Fig. 9A). The total concentration of unconjugated BAs in the CA(-) iHFC diet-fed group was comparable to that in the iHFC diet-fed group, but with reduced DCA and increased MCA levels (Fig. 9B, left panel). CA(-) iHFC diet increased the total concentration of conjugated BAs, particularly tauromuricholic acid (TMCA), compared to the ND and iHFC diets (Fig. 9B, right panel). Individual BA analyses confirmed that DCA levels were significantly lower in the CA(-) iHFC diet-fed group than in the iHFC diet-fed group (Fig. 9C). Among the three types of MCA, αMCA levels were significantly higher in the CA(-) iHFC diet-fed group than in the iHFC diet-fed group (Fig. 9C). Among the conjugated BAs, taurocholic acid (TCA) and taurodeoxycholic acid (TDCA) levels were significantly higher in the CA(-) iHFC diet-fed group than in the ND-fed group (Fig. 9D). Among the TCA species, TαCA levels were higher in the CA(-) iHFC diet-fed group than in the ND- and iHFC diet-fed groups (Fig. 9D), and TβCA levels were significantly higher in the CA(-) iHFC diet-fed group than in the iHFC diet-fed group (Fig. 9D). We also examined BA-associated gene levels in the liver. The expression levels of Cyp27a1 and Cyp7a1 were lower in the iHFC diet-fed group than those in the ND-fed group (Fig. 9E). However, this reduction was not observed in the CA(-) iHFC-diet-fed group (Fig. 9E). The effects of BAs are exerted through the nuclear receptor FXR and G protein-coupled receptor TGR5 [37]. Here, the expression levels of FXR (Nr1h4) in the liver were significantly reduced in the CA(-) iHFC- and iHFC diet-fed groups compared to those in the ND-fed group after 12 weeks of feeding (Fig. 9F, left panel). This reduction was also observed in the iHFC diet group after 24 weeks (Fig. 9F, left panel). The expression levels of TGR5 (Gpbar1) in the liver were significantly decreased in the CA(-) iHFC diet-fed group compared to those in the ND-fed group only after 24 weeks of feeding (Fig. 9F, right panel). Furthermore, the expression of Ntcp (Slc10a1), a BA transporter in the liver, was reduced by the iHFC diet (Fig. 9G, left panel) [33]. Although this reduction was milder in the CA(-) iHFC diet-fed group, it was still significantly lower than that in the ND-fed group (Fig. 9G, left panel). Liver expression levels of Bsep (Abcb11), another BA transporter, increased more in the iHFC diet group than in the ND group (Fig. 9G, right panel), and a significant difference was observed between the iHFC and CA(-) iHFC diet-fed groups after 12 weeks of feeding (Fig. 9G, right panel).

Fig. 9.

Influence of CA on the changes in fecal BA composition induced by the iHFC diet. A. Total BA composition in the feces of ND-, CA(-) iHFC diet- or iHFC diet-fed TSNO mice after 4 weeks of feeding (n = 3 or 6). B. Concentration of unconjugated BAs (left) and conjugated BAs (right) in the feces after 4 weeks of diet feeding (n = 3 or 6). C, D. Changes in individual unconjugated BAs (C) and conjugated BAs (D) in the feces after 4 weeks of diet feeding (n = 3 or 6). E, F, G. RT-qPCR of Cyp27a (Cyp27a1), Cyp7a1 (Cyp7a1), Fxr (Nr1h4), TGR5 (Gpbar1), Ntcp (Slc10a1), and Bsep (Abcb11) mRNA in the liver (n = 6). *p < 0.05, ***p < 0.01, ***p < 0.001, ****p < 0.0001

Overall, alterations in BA-associated gene levels influenced BA composition, thereby inhibiting MASH progression in the CA(-) iHFC diet-fed group compared to that in the iHFC diet-fed group.

Discussion

Previous studies have demonstrated the importance of BAs, including primary BAs such as CA, in the pathogenesis of MASH and MASLD. CA, a primary BA synthesized in the liver, influences gut microbiota composition, immune responses, and lipid metabolism, which are closely associated with MASH pathophysiology [38, 39]. CA promotes hepatic inflammation and fibrosis upon conversion to secondary BAs such as DCA by gut microbiota [40]. This study revealed the role of CA in the exacerbation of liver injury, inflammation, and fibrosis in iHFC diet-induced MASH. Specifically, the removal of CA from the iHFC diet decreased lobular inflammation, hepatocyte ballooning, and fibrosis progression (Fig. 2). These findings highlight a more direct role of CA in promoting histopathological changes in MASH than previously suggested. Unlike previous studies that mostly focused on BA dysregulation, this study elucidated the specific roles of CA in MASH development and revealed its mechanistic effects on macrophage accumulation, immune responses, gut microbiota, and BA metabolism.

Compared to the ND-fed group, we observed a significant increase in liver weight in mice fed the iHFC diet (Fig. 1B). In contrast, this change was significantly suppressed in mice fed the CA(-) iHFC diet compared to those fed the iHFC diet (Fig. 1B). Up to 24 weeks of feeding, no liver cancer was observed in the iHFC diet-fed group (data not shown) [30], indicating that the difference in liver weight is not due to tumor mass. As shown in Fig. 2B, steatosis was approximately grade 1, with no significant differences between the iHFC diet- and CA(-) iHFC diet-fed groups. However, compared to the CA(-) iHFC diet-fed group, hepatocytes in the iHFC diet-fed group appeared brighter and more hypertrophic (Fig. 2A), suggesting the accumulation of microvesicular lipid droplets, which may contribute to the difference in liver weight. Future studies should quantify hepatic TG content or perform Oil Red O staining to clarify this issue.

In contrast to the difference in liver weight between the iHFC diet- and CA(-) iHFC diet-fed groups, body weight at 24 weeks of feeding was significantly higher in the CA(-) iHFC diet group compared to the iHFC diet group (Fig. 1A, left panel). This may be attributed to fat deposition in organs other than the liver due to the cholesterol and fatty acid content of the CA(-) iHFC diet. Since we did not measure the weight of visceral and subcutaneous adipose tissues in this study, addressing this issue remains a subject for future investigation.

Plasma TG levels did not increase in the iHFC and CA(-) iHFC diet group compared to the ND-fed group (Fig. 1D, right panel). A previous study reported that the addition of CA to a chow or high-fat diet suppresses the elevation of plasma TG levels [41]. This effect is mediated by FXR signaling, the receptor for CA, which leads to a reduction in hepatic TG and very-low-density lipoprotein levels [41]. Indeed, Fxr-/- mice exhibit decreased TG levels in both the liver and blood [42]. Therefore, a similar mechanism may be occurring in the iHFC diet group. On the other hand, plasma TG levels were lower in the CA(-) iHFC diet group compared to both the ND and iHFC diet groups (Fig. 1D, right panel). A high-fat diet has been reported to reduce hepatic FXR expression [43]. Therefore, the cholesterol and fatty acids contained in the CA(-) iHFC diet may have downregulated hepatic FXR expression, leading to a decrease in plasma TG levels. However, the precise mechanism remains unclear, and we will elucidate this issue in future studies.

KCs are liver-resident macrophages that play key roles in maintaining immune homeostasis in the liver and initiating immune responses under pathological conditions [44]. Furthermore, KCs are involved in the progression or suppression of MASH/MASLD by regulating polarization into M1 or M2 macrophages through lipid-mediated mechanisms [45, 46]. In the present study, the iHFC diet significantly decreased the percentage of F4/80^Hi/CD11b^Int KCs after prolonged feeding (12 and 24 weeks; Fig. 5A and 5B), suggesting that CA promotes KC depletion in the liver over time. This decline in KC proportion is consistent with previous reports that a high-fat diet and chronic inflammation lead to the depletion or functional impairment of KCs [47, 48]. In contrast, the CA(-) iHFC diet maintained a KC proportion comparable to that of the ND-fed group (Fig. 5A and 5B, left panel), thereby preserving homeostatic function in the liver. Increased numbers of KCs in the CA(-) iHFC diet-fed group at 8 and 24 weeks suggested that the absence of CA prevented the pathological depletion of KCs (Fig. 5B, right panel), potentially mitigating liver inflammation and fibrosis. Therefore, KC preservation sustains the immune surveillance and tissue repair mechanisms, thereby protecting against MASH [49].

MdMs play pro-inflammatory roles in liver diseases by exacerbating inflammation and fibrosis progression [50]. In this study, the iHFC diet significantly increased the percentage and number of F4/80^Int/CD11b^Int−Hi MdMs after 12 and 24 weeks (Fig. 5C). However, it reduced the proportion of KCs (Fig. 5B, left panel), which is consistent with the macrophage population shift often observed in steatohepatitis [51]. The CA(-) iHFC diet significantly inhibited MdM accumulation compared with the iHFC diet. The absence of CA reduced the number of MdMs at multiple time points (4, 8, and 12 weeks; Fig. 5C, right panel), suggesting that CA played a critical role in recruiting and activating MdMs in response to iHFC diet-induced liver injury. These findings align with previous reports that secondary BAs, particularly DCA, promote hepatic macrophage activation and recruitment via Toll-like receptor signaling and other inflammatory pathways [52, 53]. Low DCA levels may be responsible for the decreased MdM accumulation in the CA(-) iHFC diet-fed group, as secondary BA drives inflammatory macrophage activation (Fig. 9C).

The proportion of CD11c⁺/Ly6C⁻ cells, which are involved in fibrosis induction [31], significantly increased at 4 weeks in the iHFC diet group (Fig. 6B, left panel). In contrast, in the CA(-) iHFC diet group, this cell population gradually increased over time, reaching approximately twice the level observed in the iHFC diet group at 24 weeks (Fig. 6B, left panel). These findings suggest that CA or DCA promotes the accumulation of CD11c⁺/Ly6C⁻ cells in the liver, leading to increased expression of fibrosis-related genes and histological changes. On the other hand, the gradual increase in the proportion of CD11c⁺/Ly6C⁻ cells and mild fibrotic changes observed in the CA(-) iHFC diet group suggest that palm oil and cholesterol in the diet may also contribute to these effects.

Gut microbiota of the CA(-) iHFC diet-fed mice showed a unique composition compared to that of the iHFC diet-fed mice, despite no significant differences in α-diversity (Fig. 8A), which is consistent with previous reports that dietary BAs influence the microbial composition rather than the overall microbial diversity [40, 54]. β-diversity analysis revealed clear differences in the microbial community structures of the three diet-fed groups (Fig. 8B), indicating that CA directly or indirectly drives specific changes in microbial composition. At the phylum level, the CA(-) iHFC diet-fed group exhibited an increased Actinobacteria abundance and decreased Verrucomicrobiota abundance; however, these differences were not statistically significant (Supplementary Fig. 2A and 2B). Actinobacteria, including members of the genus Bifidobacterium, exert anti-inflammatory effects and improve gut health [55]. An increased abundance of Verrucomicrobiota species, particularly Akkermansia muciniphila, is associated with the disruption of the gut barrier integrity and metabolic disorders [56]. Although the observed trends in these phyla in the CA(-) iHFC diet-fed group suggest their roles in gut health, further investigation is necessary to confirm their functional relevance.

At the family level, iHFC diet significantly altered the relative abundance of key bacterial families, including Tannerellaceae and Oscillospiraceae; specifically, it decreased Lactobacillaceae abundance (Fig. 8F), consistent with dysbiosis previously reported in liver disease models [52, 57]. Notably, these changes were not observed in the CA(-) iHFC diet-fed group (Fig. 8F), suggesting that CA was the major driver of these alterations. The abundance of Lactobacillaceae, which exerts anti-inflammatory and gut-protective effects, was significantly decreased in the iHFC diet-fed group but was maintained in the CA(-) iHFC diet-fed group (Fig. 8F). Lactobacillaceae family produces short-chain fatty acids and modulates immune responses, potentially mitigating liver inflammation [58]. Preservation of Lactobacillaceae possibly contributed to the reduced hepatic inflammation and fibrosis in the CA(-) iHFC diet-fed group. Interestingly, abundances of Muribaculaceae and Erysipelotrichaceae families were significantly increased in the CA(-) iHFC diet-fed group compared to those in the ND- and iHFC diet-fed groups (Fig. 8F). Muribaculaceae are associated with carbohydrate fermentation and short-chain fatty acid production, which are beneficial for the gut and liver health [59]. Erysipelotrichaceae perform context-dependent roles and support gut homeostasis under specific conditions [60]. These findings highlight the shift toward a gut microbiota composition that is less conducive to MASH progression in the absence of CA. Conversely, the relative abundance of Rikenellaceae was significantly decreased in the CA(-) iHFC diet-fed group (Fig. 8F). Rikenellaceae are associated with pro-inflammatory activity and metabolic disorders [61]. Therefore, a reduction in Rikenellaceae abundance may have contributed to the improvement in liver pathology in the CA(-) iHFC diet-fed group.

In addition to influencing BA metabolism, the gut microbiota regulates immune responses via Toll-like receptor signaling. The abundance of gram-negative bacteria that produce lipopolysaccharides is high in MASH models, contributing to hepatic inflammation via KC and macrophage activation [62]. Although the iHFC diet increased the proportion of gram-negative bacteria, this trend was less pronounced in the CA(-) iHFC diet-fed group (Fig. 8C and 8D). These findings suggested that CA removal mitigated gut-derived endotoxemia and inhibited inflammatory signaling in the liver.

BAs act as signaling molecules and regulate metabolism, inflammation, and composition of the gut microbiota. This study demonstrated that CA is a critical driver of BA dysregulation in MASH. DCA induces hepatocyte apoptosis, oxidative stress, and macrophage activation, thereby contributing to liver injury and fibrosis [20]. In the CA(-) iHFC diet-fed group, decreased DCA and increased MCA levels were associated with improved liver histopathology, including reduced inflammation and fibrosis (Fig. 9B, left panel, and 9C). These findings highlight the mechanistic link between CA and BA metabolism, and MASH progression. Notably, CA removal decreased the production of harmful secondary BA, thereby improving liver pathology and alleviating inflammation.

FXR and the G protein-coupled receptor TGR5 are critical regulators of BA metabolism, inflammation, and lipid homeostasis in the gut–liver axis [20]. In this study, the FXR levels were lower in both the CA(-) iHFC and iHFC diet-fed groups than in the ND-fed group (Fig. 9F, left panel). This is consistent with previous reports showing that high-fat diet inhibits hepatic FXR expression, thereby disrupting BA metabolism and inflammation [43]. However, elevated levels of TMCA in the CA(-) iHFC diet-fed group counteracted FXR activation, limiting pro-inflammatory signaling and liver fibrosis progression (Fig. 9C). The TGR5 levels were lower in the CA(-) iHFC diet-fed group than in the ND-fed group, particularly after 24 weeks of feeding (Fig. 9F, right panel). TGR5 activation regulates macrophage polarization and reduces inflammation via AMP signaling [20]. Decreased TGR5 expression is a possible compensatory response to altered BA composition, particularly reduced DCA levels because DCA is a potent TGR5 agonist. The combined effects of altered FXR and TGR5 signaling in the CA(-) iHFC diet-fed group may have contributed to the overall reduction in hepatic inflammation and fibrosis.

The expression levels of the genes involved in BA synthesis and transport further confirmed the mechanistic basis of the observed changes in BA composition. In the iHFC diet-fed group, the downregulation of Cyp27a1 and Cyp7a1, key enzymes in BA synthesis (Fig. 9E), indicated the suppression of classical and alternative BA synthesis pathways, which are common in liver diseases characterized by BA dysregulation [54]. Interestingly, this suppression was not observed in the CA(-) iHFC diet-fed group (Fig. 9E), suggesting that the absence of CA preserved the normal BA synthetic pathways, thereby maintaining BA homeostasis. BA transporter expression in the liver also exhibits diet-dependent changes. The reduction in Slc10a1 expression levels in both the iHFC and CA(-) iHFC diet-fed groups indicated impaired BA uptake from the bloodstream (Fig. 9G, left panel), which is often associated with liver inflammation and cholestasis [63]. Although this reduction was less pronounced in the CA(-) iHFC diet-fed group (Fig. 9G, left panel), the absence of CA partially mitigated BA transporter dysregulation. The upregulation of Abcb11 expression in the iHFC diet-fed group compared to that in the CA(-) iHFC diet-fed group is a possible adaptive response for BA secretion to prevent intrahepatic BA accumulation and subsequent liver injury (Fig. 9G, right panel).

In conclusion, this study highlights the crucial role of CA in iHFC diet-induced MASH progression through its effects on macrophage accumulation, gut microbiota composition, and BA metabolism. The removal of CA from the diet significantly improved liver pathology, inhibited macrophage activation, and altered the gut microbiota and BA profiles. Our findings are consistent with a recent report that BA metabolism plays a central role in the gut–liver axis and MASH progression [64]. CA exerts a profound impact on liver pathology by influencing BA pool composition and associated signaling pathways. Therefore, therapeutic strategies targeting BA metabolism, such as those using FXR modulators or altering the gut microbiota to reduce harmful secondary BA production, show potential for MASH treatment. However, future studies should explore the specific mechanisms by which CA interacts with gut bacteria and modulates BA signaling pathways and immune responses to further understand its role in MASH pathogenesis. Furthermore, it is also important to investigate the role of CA in the development of iHFC diet-induced MASH using a ND supplemented with CA.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

7-AAD

7-Amino-actinomycin D

ALT

Alanine aminotransferase

BA

Bile acid

CA

Cholic acid

DCA

Deoxycholic acid

FXR

Farnesoid X receptor

KC

Kupffer cell

MASH

Metabolic dysfunction associated steatohepatitis

MASLD

Metabolic dysfunction associated steatotic liver disease

MCA

Muricholic acid

MdM

Monocyte-derived macrophage

ND

Normal diet

qRT-PCR

Quantitative real-time PCR

TCA

Taurocholic acid

T-CHO

Total cholesterol

TDCA

Taurodeoxycholic acid

TG

Triglyceride

TGR5

Transmembrane G protein-coupled receptor 5

TMCA

Tauromuricholic acid

TSNO

Tsumura-Suzuki non-obese

TSOD

Tsumura-Suzuki obese diabetes

Author Contributions

Conceptualization: Y.N.; Methodology: Y.N., M.I-S., S.W., K.T., and Y.F.; Investigation: K.G., K. Kani, M.K., N.I., Y.T., K. Kasai, S.W., M.I-S., K.T., and Y.F.; Writing-original draft preparation: Y.N.; Writing-review and editing: Y.N. and K.G.; Supervision: K.T. and Y.N.; Project administration: Y.N.; Funding acquisition: Y.N. All the authors have read and agreed to the published version of this manuscript.

Funding

This work was supported by the Japan Society for the Promotion of Science (JSPS) through JSPS KAKENHI (JP22K07005 to Y.N.) and the Tamura Science and Technology Foundation (Y.N.).

Data Availability

Data will be made available upon reasonable request from the principal investigator.

Declarations

Ethics Statement

This study did not involve human participants. All animal experiments were performed in compliance with the ARRIVE reporting guidelines, which include detailed reporting of study design, statistical analysis, experimental procedures, animal care, and housing. Additionally, the experiments adhered to the guidelines for the Proper Conduct of Animal Experiments established by the Science Council of Japan. The animal protocols were approved by the Ethics Committee for Animal Experiments at Toyama Prefectural University (R1-3 and R4-1).

Competing Interests

The authors declare no competing interests.