1. Introduction
Non-alcoholic fatty liver disease (NAFLD) is a major cause of morbidity and a healthcare burden worldwide, affecting 25% of the global adult population and 34% of the Asian population [
1]. About 20% of NAFLD patients progress to non-alcoholic steatohepatitis (NASH), which is a severe stage of NAFLD with hepatocyte injury, liver inflammation, and/or pericellular fibrosis [
2]. Although several supplements and drugs such as vitamin E, peroxisome proliferator-activated receptor agonist, and pentoxifylline have borderline efficacy in NASH therapy, they are restricted in clinical use due to their potential side effects and toxicity [
3], [
4]. As a result, no drug has been approved to date by the US Food and Drug Administration or the National Medical Products Administration for NASH therapy. Identifying safe and effective therapeutics against NASH is an unmet clinical need.
Traditional Chinese medicine (TCM) is known for its advantages in terms of its multi-target and multi-channel mechanisms of action [
5]. Pien Tze Huang (PTH) is a TCM formula that is widely used in China and Southeast Asia for treating viral hepatitis and inflammatory diseases [
6]. It is composed of four TCM ingredients:
Radix et Rhizoma Notoginseng (85%),
Moschus (3%), Calculus Bovis (5%), and snake gallbladder (7%). The effect of PTH in the treatment of NASH has not been investigated.
Different dietary NASH models that recapitulate human NASH have been developed [
7]. The methionine- and choline-deficient (MCD) diet replicates NASH histological phenotypes (i.e., severe inflammation, steatosis, oxidative stress, and fibrosis) within a shorter duration than other dietary models of NASH. However, the MCD diet can cause weight loss and does not induce features of metabolic syndrome in NASH [
7]. In comparison, although a relatively long duration is needed for mice to develop NASH induced by a choline-deficient high-fat diet (CD-HFD), key features of human metabolic syndrome and NASH are mimicked in this model [
8]. However, the CD-HFD is a diet-deficient model that cannot mimic an actual human diet. The high-fat high-cholesterol (HFHC) diet model is a Western diet model that has been well established by our research team [
9]. It mirrors various phenotypes of NASH and their associated metabolic changes, including lipid accumulation, inflammation, oxidative stress, liver injury, obesity, and insulin resistance. Moreover, the HFHC diet induces gut microbiome dysbiosis and metabolites alteration, further contributing to NASH progression [
9], [
10].
Recent studies have demonstrated the crucial role of the gut microbiota in NASH progression [
9], [
11]. Gut microbial dysbiosis and its associated metabolites, such as bile acids, trimethylamine, and short-chain fatty acids (SCFAs), contribute to NASH initiation and progression [
12]. Given the importance of microbes and their metabolites, the therapeutic potential of modulating the gut microbiota to protect against liver diseases has been suggested. In particular, TCM has long been acknowledged for its ability to alter the gut microbiota to confer health benefits. TCM was reported to ameliorate metabolic syndrome by modulating the gut microbiota [
13] and improving lipid metabolism by regulating metabolites including SCFAs and bile acids [
14]. As NASH is a complex multifactorial disease, TCM may have therapeutic potential in NASH by targeting both the gut microbiota and host physiology. In this study, we investigated the role of PTH in the prevention and treatment of NASH using multiple dietary mouse models. The contribution of the gut microbiota and metabolites and the change of gut barrier function in experimental NASH with or without PTH treatment were also elucidated.
2. Materials and methods
2.1. Mice and treatments
Male C57BL/6 mice (6-7 weeks old) were randomly assigned to be fed with normal chow (NC), an HFHC diet with 23 g·L
-1 of fructose supplemented in drinking water [
15] for 12 weeks or a CD-HFD (Specialty Feeds Pty. Ltd., Australia) for 20 weeks (
n = 8-10 per group). NASH was also induced in another model by feeding male C57BL/6 mice (6-7 weeks old) with an MCD diet or the corresponding control diet (Specialty Feeds Pty. Ltd.) for 5 days, 10 days, or 4 weeks, respectively (
n = 8-10 per group). Mice were gavaged with PTH (540 mg·kg
−1; Zhangzhou Pien Tze Huang Pharmaceutical Co., Ltd., China) or phosphate-buffered saline (PBS) every day. At the experimental end points, the mice were fasted and serum/tissues were harvested. Body weights and visceral fat weights were recorded. Liver tissues were rapidly excised and weighed. Fecal samples and portal vein blood were collected for metagenomic sequencing and metabolomic profiling, respectively.
In an additional experiment, male C57BL/6 mice (6-7 weeks old) were treated with an antibiotics cocktail to verify the role of the microbiota in PTH-suppressed NASH. Drinking water was supplemented with an antibiotics cocktail (0.2 g·L
-1 of ampicillin, 0.2 g·L
-1 of neomycin, 0.2 g·L
-1 of metronidazole, and 0.1 g·L
-1 of vancomycin) for 2 weeks, every other 2 weeks until the end of the experiment, to deplete the gut microbiota [
16]. Mice were harvested 12 weeks after HFHC diet feeding as well as 4 weeks after MCD diet treatment.
In order to explore the role of probiotics in PTH-ameliorated NASH, male C57BL/6 mice (6-7 weeks old) were fed with an MCD diet and gavaged with L. acidophilus (1 × 109 colony-forming unit (CFU) per 200 µL per day), Escherichia coli (E. coli) strain MG1655 (bacteria control), or PBS for 10 consecutive days. All the mice were maintained in a specific-pathogen-free environment with controlled conditions: namely, a 12-h light/dark cycle at 20-22 °C and (45 ± 5)% humidity. All procedures adhered to the guidelines approved by the Animal Experimentation Ethics Committee of The Hong Kong Polytechnic University Shenzhen Research Institution (IACUC No.: 181105).
2.2. Shotgun metagenomic sequencing and analysis
To determine PTH-induced gut microbiota alterations, DNA was extracted from fecal samples using a QIAamp DNA mini kit (QIAGEN, Germany). Shotgun metagenomic sequencing of mice fecal DNA was performed on an Illumina HiSeq 2000 platform (Illumina, USA). Raw sequence reads were first trimmed using Trimmomatic (version 0.39) to remove low-quality reads and adaptors. The trimmed reads were aligned with Bowtie2 (version 2.4.4) against the host (mouse mm10) genome to remove human reads with the “very sensitive” default parameter settings. Kraken2 (version 2.0.8) was then used for taxonomy classification of the unmapped reads (Table S1 in Appendix A). Bracken was used to produce accurate species-level abundance estimates with the results of Kraken2. Diversity richness (Chao1 index) was calculated by means of R package phyloseq and vegan. Visualization of the microbiota between different groups was based on the Bray-Curtis dissimilarity matrix. We used Ebseq for the differential analyses of metagenomic data. We used the ternary plot method to identify bacteria associated with PTH treatment at the genus level.
2.3. Metabolomic profiling and analysis
Liquid chromatography-mass spectrometry (LC-MS) analyses were performed using an ultra-high-performance liquid chromatography (UPLC) system (1290; Agilent Technologies, USA) with a Waters UPLC BEH C18 column (1.7 μm, 2.1 mm × 100 mm; USA) from Shanghai Biotree Biotech. The column temperature was set at 55 °C and the sample injection volume was set at 5 μL. The flow rate was set at 0.5 mL·min−1.
The mobile phase consisted of 0.1% formic acid in water and 0.1% formic acid in acetonitrile. The multistep linear elution gradient program was as follows: 0-11.0 min, 85%-25% A; 11.0-12.0 min, 25%-2% A; 12.0-14.0 min, 2%-2% A; 14.0-14.1 min, 2%-85% A; 14.1-15.0 min, 85%-85% A; 15-16 min, 85%-85% A. A Q Exactive Focus mass spectrometer (Thermo Fisher Scientific, Finland) coupled with Xcalibur software was employed to obtain the mass spectrometry (MS) and tandem mass spectrometry (MS/MS) data based on the information-dependent acquisition mode. During each acquisition cycle, the mass range was from 100 to 1500. The top three masses of every cycle were screened, and the corresponding MS/MS data were further acquired. The sheath gas flow rate was 45 arbitrary units (Arb), the aux gas flow rate was 15 Arb, the capillary temperature was 400 °C, the full MS resolution was 70 000, the MS/MS resolution was 17 500, the collision energy was 15/30/45 in normalized collision energy (NCE) mode, and the spray voltage was 4.0 kV (positive) or -3.6 kV (negative). An in-house MS2 database was applied for metabolites identification. Limma was used for the differential analyses of metabolomics profiling data. The Benjamini-Hochberg method was used to control the false discovery rate (FDR). FDR < 0.01 and abs(log(fold change)) > 2 were considered as significant different. Clustering analysis was performed based on the results of screening out different metabolites.
2.4. Biochemical assay
Serum was obtained by centrifuging whole blood samples. The levels of serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), cholesterol, and triglyceride were determined using a Catalyst One Chemistry Analyzer (IDEXX Laboratories, USA). Total cholesterol (JL13847-48 T) and triglycerides (JL13528-48 T) in mice livers were measured by means of corresponding assay kits (Shanghai Jianglai Biotechnology Co., Ltd., China). Lipid peroxidation was quantified by measuring malondialdehyde using a thiobarbituric acid reactive substances (TBARs) assay (Shanghai Jianglai Biotechnology Co., Ltd.).
2.5. Histological analysis
Liver histology was assessed via hematoxylin and eosin (H&E) staining of paraffin-embedded sections. Two pathologists who were blinded to the treatment evaluated the stained slides independently and assigned scores for steatosis and inflammation. Steatosis was scored via a low-to-medium power evaluation of parenchymal involvement according to the following criteria: 0 (< 5%), 1 (5%-33%), 2 (33%-66%), or 3 (> 66%). Inflammation was scored by an overall assessment of all inflammatory foci according to the following criteria: 0 (no foci), 1 (< 2 foci per 200X field), 2 (2-4 foci per 200X field), or 3 (> 4 foci per 200X field). Oil Red O staining was performed on optimal cutting temperature compound-embedded frozen liver sections.
2.6. Measurement of pro-inflammatory cytokines
Serum tumor necrosis factor α (TNFα) and interleukin-6 (IL-6) levels were measured using a mouse TNFα enzyme-linked immunosorbent assay (ELISA) kit and mouse IL-6 ELISA kit (Shanghai Jianglai Biotechnology Co., Ltd.), according to the manufacturer’s instructions.
2.7. Quantitative real-time polymerase chain reaction (qRT-PCR)
The total RNA was extracted from mouse liver using TRIzol Reagent (Thermo Fisher Scientific). RNA quality was determined using a spectrophotometer and was reversely transcribed using a complementary DNA conversion kit (Thermo Fisher Scientific). Fecal bacterial DNA was extracted using a QIAamp DNA mini kit. Liver complementary DNA (cDNA) and bacterial DNA were used for a qRT-PCR using SYBR Green Master Mix (Roche, Switzerland) in a Light Cycler 480 real-time PCR system (Roche). The primers used are listed in Table S2 in Appendix A.
2.8. Intestinal permeability assay
A 100 mg·mL
−1 solution of fluorescein isothiocyanate (FITC)-dextran (Sigma-Aldrich, USA) was orally gavaged into each mouse (0.3 mg·g
−1 body weight). Retro-orbital blood was collected 5 h after FITC-dextran administration and was maintained at 4 °C overnight. The blood was subsequently centrifuged at 3000 rotations per minute (rpm) for 20 min. The FITC-dextran in the serum was determined at excitation wavelength of 490 nm and emission wavelength of 530 nm [
17].
2.9. Measurement of serum lipopolysaccharides
Serum was collected from portal vein of mice when harvesting. The serum lipopolysaccharide (LPS) level was measured using a mouse LPS enzyme-linked immune sorbent assay kit (Cusabio, China), according to the manufacturer’s instructions.
2.10. Western blot test
The total protein in 30-50 mg of tissues was obtained using CytoBuster protein extraction reagent (Merch Chemicals, UK). Protein concentration was measured using a detergent compatible protein assay (BIO-RAD, USA). Protein (20-40 mg) was first separated by means of 5% upper gel and 10% lower gel, and then transferred onto polyvinylidene difluoride membranes (GE Healthcare, USA). Membranes were incubated first with E-cadherin primary antibody (Catalog No.: 14472; Cell Signaling Technology, USA) overnight at 4 °C and then with secondary antibody at room temperature for 1 h. E-cadherin protein was visualized using ECL Plus Western Blot Detection Reagents (GE Healthcare). β-actin was used as the total protein loading control.
2.11. Statistical analyses
Differences between the two groups were compared by means of Student’s t test or a Mann-Whitney U test. Multiple group comparisons were assessed using a Kruskal-Wallis test or one-way analysis of variance (ANOVA). All statistical tests were performed using GraphPad Prism or R Software. Data were considered significant at P < 0.05.
3. Results
3.1. PTH prevents HFHC diet-induced steatohepatitis
To explore the effects of PTH on steatohepatitis, C57BL/6 mice were fed with an HFHC diet for 12 weeks to induce NASH development. Mice were administered with PTH or PBS control (
Fig. 1(a)). HFHC diet-fed mice supplemented with PTH exhibited a lower body weight compared with HFHC diet-fed mice supplemented with the PBS control (
P < 0.01) (
Fig. 1(b)). Histological examination of liver sections showed that PTH treatment significantly improved liver histology with reduced liver steatosis (
P < 0.01) and inflammatory cell infiltration (
P < 0.05), compared with HFHC diet-fed mice without PTH supplementation (
Fig. 1(c)). The decreased lipid accumulation in PTH-supplemented mice was confirmed by Oil Red O staining (
Fig. 1(c)). Consistent with the histological findings, PTH decreased blood and hepatic lipids in mice, as evidenced by significantly reduced levels of triglyceride and total cholesterol (
P < 0.05 for serum and hepatic triglyceride;
P < 0.001 for serum and hepatic total cholesterol) (
Fig. 1(d)). Measurements of hepatic lipid peroxide (
P < 0.01) and serum ALT and AST (both
P < 0.05) revealed that PTH significantly protected mice from HFHC diet-induced liver injury (
Figs. 1(e) and
(f)). Moreover, PTH significantly reduced the serum levels of key pro-inflammatory mediators in NASH including TNFα (
P < 0.01) and IL-6 (
P < 0.001) in HFHC diet-fed mice (
Fig. 1(g)). Collectively, these results suggest that PTH acts as a prophylaxis against the development of steatohepatitis.
3.2. PTH prevents CD-HFD or MCD diet-induced steatohepatitis
To confirm the preventive effects of PTH, we established a second NASH mouse model induced by a CD-HFD. Mice were fed with a CD-HFD supplemented with PTH or PBS control for 12 weeks. Mice developed steatohepatitis after 20 weeks feeding with a CD-HFD (
Fig. 2(a)). Consistent with the observations in HFHC diet-fed mice, CD-HFD-fed mice with PTH supplementation showed significantly decreased body weight, liver weight, and adipose tissue weight compared with CD-HFD-fed mice supplemented with the PBS control (all
P < 0.001) (
Fig. 2(b)). PTH administration also attenuated the development of steatohepatitis, with improved liver histology including significantly reduced steatosis (
P < 0.01) and inflammation (
P < 0.001) (
Fig. 2(c)). Attenuated steatohepatitis due to PTH was confirmed by decreased levels of serum cholesterol, hepatic triglyceride, hepatic total cholesterol (
Fig. 2(d)); liver lipid peroxidation (
Fig. 2(e)); serum ALT levels (
Fig. 2(f)); and pro-inflammatory cytokines TNFα and IL-6 (
Fig. 2(g)).
We established a third NASH mouse model to investigate whether PTH could suppress steatohepatitis development in different stages by feeding C57BL/6 mice with an MCD diet for 10 days or 4 weeks, with or without PTH (
Fig. 3(a)). The MCD diet significantly induced weight loss, while PTH administration further decreased body weight (Fig. S1 in Appendix A). PTH protected mice from MCD diet-induced steatohepatitis, with significantly reduced liver steatosis (
P < 0.05 for 10 days on an MCD diet;
P < 0.01 for 4 weeks on an MCD diet) and inflammation (
P < 0.05 for 10 days on an MCD diet;
P < 0.01 for 4 weeks on an MCD diet), regardless of the duration of MCD diet feeding (
Fig. 3(b)). Consistently, hepatic triglycerides (both
P < 0.001) and lipid peroxidation (
P < 0.001 for 10 days on an MCD diet;
P < 0.01 for 4 weeks on an MCD diet) were significantly decreased in PTH-supplemented mice with 10 days or 4 weeks of feeding on an MCD diet, compared with those without PTH (
Fig. 3(c)). Liver injury was also ameliorated in PTH-supplemented mice with 10 days or 4 weeks of feeding on an MCD diet, as evidenced by decreased serum levels of ALT and AST (all
P < 0.001) (
Fig. 3(d)). The increased serum TNFα and IL-6 levels were reduced by the PTH in the 4-week MCD diet-fed mice (Fig. S2 in Appendix A). Consistent with the findings from 10-day and 4-week MCD diet-fed mice, PTH administration for 5 days to MCD diet-fed mice significantly decreased hepatic triglycerides (
P < 0.001), lipid peroxidation (
P < 0.001), serum ALT (
P < 0.001), and AST levels (
P < 0.001) (Fig. S3 in Appendix A). Taken together, our results from multiple NASH mouse models revealed the preventive effect of PTH against the development of diet-induced steatohepatitis.
3.3. PTH has a therapeutic effect in established steatohepatitis
After confirming the preventive effect of PTH, we next explored whether PTH had therapeutic effects against steatohepatitis. Steatohepatitis was developed in mice after feeding with an MCD diet for 3 weeks; these MCD diet-fed mice then received either PTH treatment or vehicle controls for 10 days (
Fig. 4(a)). Histological examination of liver sections showed significantly reduced lipid accumulation and inflammatory cell infiltration in the MCD diet-fed mice with 10 days of PTH treatment, compared with the vehicle controls (both
P < 0.001) (
Fig. 4(b)). PTH treatment also significantly decreased hepatic total cholesterol (
P < 0.01), triglycerides (
P < 0.01), lipid peroxidation (
P < 0.001) (
Fig. 4(c)), and serum levels of ALT (
P < 0.01) and AST (
P < 0.001) (
Fig. 4(d)), indicating the marked alleviation of MCD diet-induced liver injury by PTH treatment. These results thus show the therapeutic potential of PTH on steatohepatitis.
3.4. PTH restores the gut microbiota in mice with steatohepatitis
We profiled the gut microbiota in fecal samples of HFHC diet-fed mice with or without PTH and in NC-fed mice by means of shotgun metagenomic sequencing. Principal component analysis (PCoA) revealed distinct microbial communities among the three groups of HFHC diet-fed mice with PTH, HFHC diet-fed with PBS, and NC-fed mice with PBS (
Fig. 5(a)). Significantly lower bacterial diversity (Chao1 index) was observed in the HFHC diet-fed mice compared with the control mice (
P < 0.0001), while PTH supplementation was found to restore bacterial diversity in the HFHC diet-fed mice (
P < 0.05) (
Fig. 5(b)). The compositions of the fecal bacteria communities of the three groups are shown in
Fig. 5(c).
We then used the ternary plot method to identify bacteria associated with PTH treatment at the genus level (
Fig. 5(d)). Lactobacillus—a genus including many probiotic species that ameliorate NASH progression [
18]—was found to be depleted in HFHC diet-fed mice, whereas PTH restored the abundance of
Lactobacillus back to a level similar to that found in the control mice (
Fig. 5(d)). Examination of the microbiota at the species level identified a consortium of bacteria that were significantly enriched by PTH (
Fig. 5(e)). The majority of PTH-enriched bacteria were probiotics, including
Lactobacillus (e.g.,
L. acidophilus and Lactobacillus plantarum (
L. plantarum)), Lactococcus (e.g., Lactococcus lactis), and Bacillus (e.g., Bacillus subtilis) (
Fig. 5(e)). On the other hand, the abundances of several pathogenic bacteria from Citrobacter and Pseudomonas [
19] were decreased by PTH (
Fig. 5(e)). Increased abundances of
L. acidophilus (
P < 0.0001) and
Lactococcus lactis (
P < 0.001) were also found in the stool of the NC-fed mice with PTH administration (
Fig. 5(f)). Collectively, our metagenomic analysis revealed that PTH restored the gut microbiota in mice with steatohepatitis by inducing the enrichment of probiotics.
3.5. L. acidophilus protects against steatohepatitis
We next evaluated the effect of enriched probiotic
L. acidophilus in PTH-prevented NASH. MCD diet-fed mice were gavaged with
L. acidophilus,
E. coli, or PBS for 10 consecutive days (
Fig. 5(g)).
L. acidophilus-gavaged mice showed greatly improved liver histology, with reduced steatosis and inflammatory cell infiltration compared with mice gavaged with
E. coli or PBS (
Fig. 5(g)). Accordingly, the liver-to-body weight ratio was significantly decreased by
L. acidophilus (
Fig. 5(g)). These results clearly show that
L. acidophilus suppressed steatohepatitis development in mice.
3.6. PTH restores gut barrier function in mice
Gut barrier dysfunction driven by microbial dysbiosis is a prerequisite for steatohepatitis development [
20]. To test whether the gut barrier contributes to the anti-steatohepatitis effects of PTH, we examined the paracellular permeability in mice colon by measuring the serum LPS level. Serum LPS concentration was elevated in mice fed with an HFHC diet, a CD-HFD, and an MCD diet, compared with NC or corresponding control diet-fed mice (all
P < 0.001) (
Fig. 6(a)), confirming the impaired gut barrier function in multiple NASH mouse models. PTH significantly decreased serum LPS levels in all diet-induced NASH mouse models, compared with mice without PTH (
P < 0.05 for CD-HFD-fed mice;
P < 0.001 for HFHC diet-fed and MCD diet-fed mice) (
Fig. 6(a)). MCD diet-fed mice supplemented with PTH showed a significant reduction in intestinal permeability according to FITC-dextran assay (
P < 0.05) (
Fig. 6(b)), implying that gut barrier function was restored upon PTH administration. Consistently, the expression of E-cadherin—a cell adhesion molecule and marker of gut barrier integrity—was significantly increased in HFHC diet-fed (
P < 0.01), MCD diet-fed (
P < 0.05), and CD-HFD-fed (
P < 0.05) mice with PTH treatment (
Fig. 6(c)). Moreover, 10 days of PTH treatment after the onset of MCD diet-induced steatohepatitis reduced the serum LPS level in mice, compared with the PBS controls (
P < 0.001) (
Fig. 6(d)). Taken together, these results indicate that PTH alleviates NASH development at least in part by restoring a functional gut barrier.
3.7. PTH influences the bile acid metabolism in the portal vein of mice
Gut microbial dysbiosis impairs the gut-liver axis, leading to unrestrained transfer of metabolites from the intestines into the liver. Some of these translocated metabolites can be harmful, which may cause liver injury and contribute to NASH development [
12]. To reveal the metabolomic phenotypes associated with PTH treatment, we performed untargeted metabolomics on the portal vein blood of mice fed with NC, an HFHC diet, or an HFHC diet with PTH. PCoA showed a marked disparity in the metabolites in the portal vein among the three groups of mice (
Fig. 7(a)). Nineteen metabolites were found to be significantly enriched in the portal vein of HFHC diet-fed mice with PTH, compared with those without PTH (
Fig. 7(b)). In contrast, 21 metabolites were depleted in the portal vein of PTH-supplemented HFHC diet-fed mice. To understand the functional characteristics of the enriched metabolites, we conducted a pathway enrichment analysis and found that bile acid biosynthesis was one of the top altered pathways among the portal vein metabolites in PTH-treated mice (
Fig. 7(c)). Further examination of the portal vein metabolites showed that serum levels of conjugated bile acids including tauroursodeoxycholic acid (TUDCA;
P < 0.001), taurodeoxycholic acid (TDCA;
P < 0.01), glycocholic acid (GCA;
P < 0.01), taurocholic acid (TCA;
P < 0.001), and chenodeoxycholic acid (CDCA;
P < 0.05) were significantly upregulated in PTH-supplemented HFHC diet-fed mice, compared with those without PTH (
Fig. 7(d) and Fig. S4 in Appendix A). Moreover, the ratio of deoxycholic acid (DCA) to CDCA in the portal vein—which is a marker of human NASH [
12]—was significantly decreased in HFHC diet-fed mice administrated with PTH (
P < 0.05) (
Fig. 7(e)).
Conjugated bile acids play an important role in NASH by binding with farnesoid X receptor (FXR) in the intestine to regulate the production of hormone fibroblast growth factor 15 (FGF15), which is secreted into the liver to reduce bile acid synthesis [
21]. Given the upregulation of conjugated bile acids after PTH administration, we chose to examine the expression of
Fxr and
Fgf15 in the intestine of HFHC diet-fed mice. Our results showed that intestine
Fxr and
Fgf15 mRNA levels were significantly decreased by PTH supplementation (both
P < 0.01) (
Fig. 7(f)). In contrast, hepatic mRNA expression of bile acids biosynthesis including cytochrome P450 (
Cyp7a1,
Cyp7b1, and
Cyp27a1) was significantly increased in PTH-supplemented mice, compared with mice without PTH (
P < 0.01) (
Fig. 7(f)). Taken together, our results suggest that PTH mainly upregulates conjugated bile acids and reduces the secretion of intestinal Fgf15, thereby at least partially suppressing NASH development.
We also investigated the chemical identities of PTH by means of LC-MS. The main compounds in the formula of PTH include ginsenoside (ginsenoside Rg1, Rg2, Rg5, Rf, Rb1, Rb3, and F2) and bile acids (CA, CDCA, hyodeoxycholic acid, gentisic acid, and lithocholic acid). The relevant MS/MS spectra of the main compounds are provided in Fig. S5 in Appendix A.
3.8. An intact gut microbiota is essential in PTH’s amelioration of steatohepatitis
To evaluate the role of the gut microbiota in PTH-induced NASH prevention, an antibiotics cocktail was used to deplete the gut microbiota in HFHC diet-fed or MCD diet-fed mice with or without PTH supplementation (Figs. S6 and S7 in Appendix A). We found that PTH administration did not ameliorate steatohepatitis development in microbiota-depleted HFHC diet-fed mice, since these mice exhibited levels of steatosis, inflammation, and hepatic lipid peroxidation similar to those of microbiota-depleted HFHC diet-fed mice without PTH (Fig. S6). PTH also consistently failed to suppress steatohepatitis development in microbiota-depleted MCD diet-fed mice, since these mice exhibited levels of serum ALT, AST, hepatic triglyceride, and lipid peroxidation similar to those of mice without PTH treatment (Fig. S7). The probiotics enriched by PTH were eliminated by antibiotics administration (Fig. S8 in Appendix A). These findings clearly illustrate that the gut microbiota is essential for PTH to exhibit its anti-steatohepatitis effects against NASH development.
4. Discussion
PTH is known for its anti-inflammatory effects, yet its efficacy against NASH has not been investigated. In this study, our results showed that PTH significantly prevented the development of diet-induced steatohepatitis in multiple mouse models. Moreover, PTH reversed the established NASH, indicating its therapeutic role in NASH. PTH supplementation improved liver histology, reduced hepatic triglyceride, cholesterol, and lipid peroxide, and decreased serum ALT and AST levels, thereby contributing to the prevention and regression of mice steatohepatitis. PTH has been widely used to treat liver disease for 600 years [
22]. Preclinical studies have shown that PTH attenuates carbon tetrachloride-induced liver injury, inflammation, and fibrosis [
23], [
24], and inhibits the proliferation of liver cancer cell lines [
25]. Our study therefore extends the notion that PTH can act as a novel prophylactic and therapeutic drug for NASH, a highly prevalent and progressive disease with unmet clinical need.
The association of the gut microbiota with liver diseases has been increasingly acknowledged. Gut microbiome dysbiosis with lower microbial diversity has been reported in both human NASH [
26], [
27] and mouse NASH models [
9]. However, human NASH patients showed increased abundance of Bacteroides [
28] and Proteobacteria [
29], while Enterobacteriales [
30] was enriched in mouse NASH models. Although the altered bacteria are different between human and mouse NASH, modulation of the gut microbiome by probiotics (Lactobacillus and Bifidobacterium) or fecal microbiome transplantation was found to ameliorate NASH in both humans and mice [
11], [
31]. Our metagenomic analysis revealed enrichment of the beneficial probiotics
Lactobacillus, Alistipes, and Bacillus after PTH treatment. In particular,
L. acidophilus and
L. plantarum—two
Lactobacillus species that have been widely commercialized—were found to be enriched in PTH-treated mice with ameliorated NASH. A pilot clinical study performed in our institution showed that a probiotic formula containing
L. acidophilus and
L. plantarum reduced liver fat and ameliorated liver injury in NASH patients [
32]. We confirmed that the enriched
L. acidophilus prevented steatohepatitis in MCD diet-fed mice.
Lactococcus lactis, a beneficial microbe that suppresses hepatis steatosis and inflammation [
33], was also found to be enriched in the mice administered with PTH. The probiotic Bacillus subtilis ameliorates liver inflammation, improves gut barrier function, and modulates gut microbiota in mice with NAFLD [
34]. We previously reported that a high intake of dietary cholesterol induces microbial dysbiosis in mice, and that the dysbiotic microbiota then contributes to the pathogenesis of NASH and its related hepatocellular carcinoma [
9]. Several TCMs have been shown to ameliorate human NAFLD effectively by modulating the gut microbiota [
35]. Collectively, our study suggests that PTH suppresses NASH by alleviating dysbiosis and enriching probiotics.
Gut dysbiosis promotes liver pathologies in many different ways, particularly by damaging the gut barrier, with markedly increased intestinal permeability [
36]. In general, dysbiotic microbiota impair the gut barrier, thereby allowing the translocation of more pathogenic bacteria and their derived toxic metabolites to the liver through portal vein circulation [
36]. Our findings illustrate the occurrence of gut barrier dysfunction in mice with diet-induced steatohepatitis. Notably, PTH treatment greatly restored the impaired gut barrier, as indicated by increased intestinal permeability, reduced serum endotoxin LPS level, and increased expression of tight-junction protein E-cadherin. The gut barrier restored by PTH may prevent the translocation of pathogenic bacteria and toxic metabolites into the liver, thereby protecting mice from liver injury. Indeed, pharmacological drugs that improve the gut barrier can prevent and treat NASH [
20]. Our results suggest that PTH can act as a drug to protect against gut barrier disruption and suppress NASH development.
The metabolites produced by gut microbes are important in the development of steatohepatitis, as toxic metabolites can induce liver damage through the gut-liver axis [
11]. In this study, our metabolomic analysis revealed the enrichment of conjugated bile acids, including TUDCA, TCA, TDCA, and GCA in PTH-treated mice. Bile acids are crucial metabolites that modulate lipid metabolism [
37]. TUDCA attenuates diet-induced hepatic steatosis, inflammation, obesity, and insulin resistance by modulating the gut microbiota and improving gut barrier function [
38]. TUDCA also suppresses intestinal FXR-FGF15 signaling, which mediates activation of the cytochrome P450 family enzyme in hepatocytes and the metabolism of bile acids [
39]. In our experiments, intestinal
Fxr-
Fgf15 was suppressed, whereas hepatic genes for bile acid synthesis, including
Cyp7a1,
Cyp7b1, and
Cyp27a1, were upregulated. Therefore, our results suggest that PTH inhibits the intestinal FXR signaling pathway to promote bile acid synthesis in the liver, thereby suppressing NASH.
5. Conclusions
This study demonstrates, for the first time, that PTH could be an effective prophylactic and therapeutic against steatohepatitis. Mechanistically, PTH modulates the gut microbiota and restores the functions of the gut barrier and the altered gut metabolites, thereby alleviating the development and progression of NASH through the gut-liver axis.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (82103355 and 82222901); Research Grants Council-General Research Fund (RGC-GRF; 14117422); Health and Medical Research Fund, Hong Kong (08191336); and Vice-Chancellor's Discretionary Fund CUHK.
Authors’ contribution
Xianyi Zeng and Xiang Zhang performed experiments, analyzed the data, and drafted the manuscript; Hao Su, Hongyan Gou, and Yan Li performed experiments; Harry Cheuk-Hay Lau revised the manuscript; Xiaoxu Hu and Ziheng Huang performed the bioinformatic analysis; Jun Yu designed and supervised the study and wrote the paper.
Compliance with ethics guidelines
Xianyi Zeng, Xiang Zhang, Hao Su, Hongyan Gou, Harry Cheuk-Hay Lau, Xiaoxu Hu, Ziheng Huang, Yan Li, and Jun Yu declare that they have no conflict of interest or financial conflicts to disclose.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.eng.2022.10.010.
PDF
(6144KB)
