Changes of gut microbiota during silybin mediated treatment of high-fat diet-induced nonalcoholic fatty liver diseasein mouse

Xiuxia Li a ,b,*, Yanping Wanga. b, YilanXinga, b, Renxin Xinga, b, Yongsheng Liua, b, Yinsheng Xua, b

Abstract

Objective Gut microbiota involved in pathogenesis of nonalcoholic fatty liver disease (NAFLD). Silybin (Sil), a naturally occurring hepatoprotective agent, was widely used for treating NAFLD. Whether Sil effects gut microbiota during its actions in treating NAFLD, which is unknown. We aimed to examine the effect of Sil on intestinal flora dysbiosis induced by high-fat diet (HFD).
Methods After 10 weeks of feeding normal chow diet (NCD) or HFD, mice were given daily by
gavage for 8 weeks. Cecal contents were harvested for study of short chain fatty acids (SCFAs), bile acids and gut microbiota alteration. Results Sil showed protective effects against dietary-induced obesity and liver steatosis, accordingly gut microbiota composition changed. At phylum level, compared with HFD group, mice in the Sil-treated group had significant lower level of Firmicutes and the ratio of Firmicutes to Bacteroidetes
(P<0.05). At genus level, Sil-treated group have significant lower level of Lachnoclostridium, Lachnospiraceae_UCG-006 and Mollicutes_RF9, which were reported to be potentially related to diet-induced obesity, and increased levels of Blautia (P<0.05), Akkermansia (P<0.05), Bacteroides(P<0.05) which were known to have beneficial effect on improving NAFLD. Sil also showed inhibitory effect on well-known beneficial bacteria, such as Alloprevotella, Lactobacillus. Moreover, production of acetate, propionate and butyrate increased, while generation of formate and conversion of cytotoxic secondary metabolites (LCA, DCA) decreased in mice treated with Sil.
Conclusion Sil might have beneficial effects on ameliorating NAFLD and mediate HFD-induced change of gut microbiota composition, followed with major changes in secondary metabolites, such as SCFAs and BAs.

Keywords: Silybin, nonalcoholic fatty liver disease (NAFLD), Gut Microbiota, short chain fatty acids (SCFA), Bile acid (BA)

Introduction

Non-alcoholic fatty liver disease (NAFLD) is one of the most common chronic liver disease worldwide, affecting about 25% of the general population. It is increasingly prevalent, and has become a major global health burden1. Currently, no complete knowledge on the pathogenesis of NAFLD is clarified2,3, among these theories for mechanism, the “multiple parallel hits hyptothesis” is a new model. This model indicated that a number of hits may act in parallel, and lead to liver inflammation finally, factors derived from the gut and the adipose tissue may play a central role.
Accumulating evidence suggested that gut microbiota plays an important role in the development of NAFLD4,5,6,7, where several mechanisms may be involved: a). It promotes obesity and NASH though increasing energy yield from food, b). It regulates gut permeability, low-grade inflammation and immune balance; c). It modulates dietary choline and bile acid metabolism; d). It improves endogenous ethanol production. This suggests that regulation of gut flora may be a potential therapeutic strategy for such disease treatment.
To date, no specific therapy was performed for NAFLD. Several strategies were proposed, including lifestyle management and pharmacotherapy (insulin sensitizer agents, antioxidants, lipid lowering drugs). However, these strategies are difficult to achieve. It is needed to find new effective candidates for alleviatingNAFLD.With little toxicological effects, many compounds from natural products were identified with effective treatment in mouse or rat model of NAFLD. For instance, berberine alkaloid from Berberis Aristata, Dioscinfrom Dioscoreanipponica Makino, Silybin(Sil) isolated from milk thistle. As a clinically effective drug for liver protection, Sil limit the progression of a spectrum of liver diseases. Pharmacological studies have shown Sil could disturbing NF-kB and IRS-1/PI3K/Akt signaling pathway8,9, it also has anti-inflammatory, antioxidant, and anti-fibrotic activity. Could Sil improve gut microbiota, and exhibit liver protection, which was unknown. In this study, we fed C57BL/J mouse with high-fat diet (HFD) to induce gut microbial dysbiosis and develop NAFLD, when typical NAFLD symptoms were shown, Sil-treatment was given. We examined the structural changes of gut microbiota in response to Sil intended to alleviate HFD-induced NAFLD in mouse.

Materials and methods

Chemicals and Diets

Sil, short chain fatty acids (SCFAs) and bile acids standard mentioned in this article were obtained from Sigma-Aldrich (St. Louis, MO, USA). High-fat diets (HFD) composed of 60% kcal fat (D12492) and normal chow diet (NCD) containing 10% kcal fat (D12450) were purchased from Hua Fukang (Beijing, China).

Animal experiments

Animal care and experimental procedures were approved by the Animal Care and Use Committee of Sichuan academy of medical science& Sichuan provincial people hospital (Chengdu, Sichuan, China) and performed strictly complied with the People’s Republic of China Legislation Regarding the Use and Care of Laboratory Animals. C57BL/6J male mice of 8 weeks of age were purchased from institute of laboratory animals of Sichuan academy of medical science & Sichuan provincial people hospital, Mice were housed in a controlled environment on12h light/12 h dark cycle, 50 ± 10 % humidity and 24 ± 2 °C temperature, with free access to water and food. After 1 week of acclimatization, the mice were randomly assigned to two groups. (1)NCD group(n = 10), mice were fed a normal chow diet(NCD) containing 10% kcal fat (D12450, Hua Fukang, Beijing, China). (2) HFD group (n = 30), mice were raised with high-fat diet (HFD) composed of 60% kcal fat (D12492; Hua Fukang, Beijing, China).After 10 weeks of feeding, the HFD group mice were randomly assigned to three groups(n =8) with continued HFD, and given the following treatments once every day (qd): (a) HFD group (n =8), fed with HFD diet and oral gavage of0.5% carboxymethylcellulose sodium (CMC-Na),(b) HFD +Sil_1 group (n =8), fed with HFD diet and oral gavage of Sil (100 mg/kg), (c) HFD+Sil_2group (n =8), fed with HFD diet and oral gavage of Sil (300 mg/kg). Sil was prepared in a solution of 0.5% CMC-Na, and the NCD and HFD mice were given an equal volume of 0.5%CMC-Na daily.
During the long-term trial, food intake in each group was recorded weekly, and body weight of each mouse was measured every five days. The trial was terminated after 8 weeks of treatment, after an overnight fast, the mice were sacrificed and serum was collected. serum levels of fasting glucose, total triglyceride (TG), total cholesterol (TC), aspartate aminotransferase (AST) and alanine aminotransferase (ALT)were detected by an automatic biochemistry analyzer. cecum and liver were excised and immediately frozen in liquid nitrogen for further analysis.

Hepatic pathological examination

After being sacrificed, mouse livers were collected and fixed in10% formalin for more than 24 h, with sequentially dehydrated, infiltrated and cut into 5-μm-thick sections according to the routine procedure. Each section was routinely stained with hematoxylin and eosin (H&E). Frozen pieces of formalin-fixed livers were stained with Sudan III and Oil Red O. Photomicrographs were captured using a Nikon Eclipse TE2000-U, NIKON, Japan.

DNA extraction

Total Genomic DNA was isolated from cecal contents using InviMag ® Stool DNA kits (Invitek, Germany) following the manufacturer’s instruction. DNA concentration and quality were examined with Nano Drop Spectrophotometer. Then extracted DNA was diluted into10 ng/μL and stored at -40°C for further processing.

Pyrosequencing

The extracted DNA was used as a template for amplifications of region 1 and 4(519F/907R) of 16S rRNA genes and sequenced on the MiSeq system (Illumina, San Diego, CA, USA). Each DNA sample was PCR amplified with the bacterial universal forward primer（5- CAGCMGCCGCGGTAATWC -3），the reverse primer（5-CCGTCAATTCMTTTRAGTTT-3）. PCR reaction, pyrosequencing of the PCR amplifications, and quality control of raw data were performed according to a previous description with some modification.

Analysis of cecal SCFAs and bile acids

The cecal contents were diluted in 0.01 M PBS, and processed with ultrasound in ice-cold water bath for 30 min. After centrifugation (8000×g, 10 min at 4℃), the supernatant was transferred into another tube, and extracted in ethyl ether. The concentration of SCFAs was determined in organic phase using an Agilent 6890N gas chromatograph system equipped with a polar capillary column and a flame ionization detector (Agilent Technologie Inc., Palo Alto, USA). Cecal bile acids analysis was performed according to Liquid chromatography-electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) as described previously10.

Immunoblot analysis

The colonic mucosa (75mg) was washed twice with ice-cold PBS, and homogenized in Radio

Immunoprecipitation

Assay (RIPA) buffer for 20 min. After that, the crude extracts were centrifuged, the supernatants were harvested, with protein concentration being measured with the Bradford protein assay. Equal amounts of proteins were separated by SDS-PAGE and transferred to PVDF membranes (Millipore, USA). The membranes were blocked for 2 hours in Tris-buffered saline containing 5% skim milk, then was incubated with primary antibodies (including junctional adhesion molecule (JAM)-A, Occludin, zonula occludens (ZO)-1 and Claudin) overnight at 4 ℃. After being washed with TBS-T, the membrane was incubated with HRP conjugated secondary antibodies subsequently. Bands were visualized using Super-Signal Ultra chemiluminescent reagent (Pierce, U.S.A.) and X-ray film. Quantification was performed by densitometric analysis of each bands using Image J software.

Statistical analysis

Data analyses was performed on SPSS 16.0 (SPSS Inc., Chicago, IL) software, except microbial community analyses. Statistical comparisons were analyzed by student’s t-tests for each paired group. Data are presented as means ± standard deviation (SD). P <0.05 or P<0.01 is considered as a significant difference.

Bioinformatics Analysis

Raw sequences were processed with Quantitative Insights into Microbial Ecology (QIIME) software. Paired-end reads were merged using FLASH. Before clustering, reads with quality score <20, ambiguous bases and improper primers were all removed. The resultant high-quality sequences of each sample were subsampled to an equal sequencing depth, and then clustered into operational taxonomic units (OTUs) at 97% similarity, meanwhile chimeras were examined and eliminated. Taxonomical assignment of representative sequences was analyzed using RDP classifier (http://rdp.cme.msu.edu). Rarefaction analysis and Alpha diversity estimations including diversity (Shannon), richness (Sobs) were calculated using Mothur. In order to assess the variation between experimental groups, beta diversity including Principle coordinate analysis (PCoA) and Principal component analysis (PCA) based on OTUs was analyzed with R Project. Based on genus abundance, clustering was performed through partitioning around medoids (PAM) with Jensen-Shannon divergence (JSD). Community composition provides the classification information at phylum level and genus level, and differences among groups were determine using Kruskal-Wallis H test.

Results

Effect of Sil on body weight and food intake

In order to evaluate the effect of Sil on NAFLD under HFD, NAFLD in mice was induce by given a diet of HFD for 10-weeks, after the initial feeding period, the mice were continued for an additional HFD supplemented with Sil (100 or 300 mg/kg) simultaneously by gavage. As expected, at the end of the initial 11-week feeding period, compared to mice receiving normal chow, HFD mice gained more weight. Sil exhibited a dose-dependent efficacy in inhibition of body weight gain, Sil (300 mg/kg) treatment prevented the increase of body weight gain obviously (Figure1A). During the 60-days treatment, compared with HFD group (Figure1S), there was a slight decrease in food intake of Sil-treated group, but no statistical significance appeared, which may show that the mitigation effect on body weight occur on the Sil treated mice were not due to changes in food consumption Sil attenuated hepatic steatosis in HFD- induced NAFLD mice.
NAFLD is strongly associated with glucose tolerance and insulin resistance, we performed oral glucose tolerance test to examine effect of Sil on NAFLD. The kinetics of plasma glucose (Figure 1B) after the glucose challenge was different in the HFD-control and NCD-control groups, Sil supplementation ameliorated the impaired glucose tolerance dose-dependently, however no notable difference was observed in the areas under the curve for plasma glucose. Liver injury was confirmed by significant increase of serum AST, ALT, TC, TG. when compared to the NCD group, Serum levels of AST (P<0.01), ALT(P<0.05) and TG (P<0.05) (Figure1C, D) increased significantly in the HFD-control group, Administration of Sil reversed the increase of Liver injury indicator dose-dependently Long-term HFD resulted in steatosis, inflammation, hepatocyte swelling and necrosis in HFD-control group (Figure1E), while Sil (100 mg/kg and 300 mg/kg) improved the severe hepatosteatosis, with the lipid inclusions reduced strikingly in both size and number, steatohepatitis scores also lowered

Modulation of bacterial relative abundances after Sil treatment

To assess the structural changes of gut microbiota after Sil treatment, we sequenced V1-V4 amplifications of 16S rRNA genes. After trimming, assembly and quality filtering, a total of 662 604 high quality sequences were obtained from 20 samples, with an average of 33130±8861 per sample of 406 bp length. Each sample were subsampled to an equal sequencing depth and clustered, 2691 operational taxonomic units (OTUs) at 97% similarity were obtained. As shown in Figure 2A and C, the near saturated rarefaction curves indicates that the vast majority of the gut microbial diversity in all sample was captured with the current sequencing depth. As shown in Figure 2, both the diversity and richness of the gut microbiota decreased in HFD group, no obvious difference was observed in Comparison of diversity between NCD and HFD group, while significant difference of estimated OTU richness (Sobs) (P<0.05) was existed. In Sil-treated group, shannon diversity and OTU richness increased with no significant differences.

Response of gut microbiota structure to Sil in HFD-fed mouse.

Principal component analysis (PCA) and UniFrac distance-based principal coordinate analysis (PCoA) were used to analyze gut microbiota structural of each sample. Compared with NCD (Figure 3A), HFD diet resulted in structural shift along the first principal component (PC1), with main principal component (PC) scores: PC1 = 20.14%, PC2 = 32.1%, which formed a separate cluster. The treatment of Sil reverted HFD-induced variations significantly along PC1. Principal coordinates analysis (PCoA) also showed a similar pattern of changes of the gut microbiota between HFD and NCD groups, followed by Sil treatment resulted in reverted changes. Here, as shown in Figure 2 S, typing analysis on genera level revealed two distinct clusters. HFD induced changes in the composition of the gut microbiota, following Sil treatment altered the cecal community structure in HFD-fed mice, the treatment group were more resembled the NCD mice closely, thus suggesting that Sil might reverse the changes of gut microbiota in HFD-fed mice.

Changes of gut microbiota composition during Sil-treatment in HFD-fed mouse.

Representative sequences were listed in Figure4, at phylum level, Firmicutes, Bacteroidetes and proteobacteria were the most abundant divisions in all of the groups. Compared with NCD group, mice in the HFD group had significant higher level of Firmicutes (73.7% versus 45.6%; P<0.01), and lower level of Bacteroidetes (14.2% versus 42.5%; P<0.001), the ratio of Firmicutes to Bacteroidetes increased (Figure 3S), Sil (100mg/kg and 300mg/kg) obviously reversed these changes of bacterial abundance. At genus level, as shown in Figure 4B,Figure 4S, 18 dominant genera accounted for more than 70% of the total OTUs, during which, we observed higher level of Lachnoclostridium (8.7% versus 2.5%; P<0.05), Lachnospiraceae_UCG-006(6.5% versus 1.0%; P<0.01), norank_o_Mollicutes_RF9(4.0% versus 1.0%; P<0.05), Lactobacillus (17.38% versus 1.0%; P<0.05), high dose of Sil showed significant inhibitory to these changes, while low dose demonstrated no significant effect to Lachnoclostridium (P=0.151), norank_o_Mollicutes_RF9 (P=0.052). There were lower levels of Blautia (2.9% versus 11.1%; P<0.05) Bacteroides (0.9% versus 8.2%; P<0.05), Akkermansia (0.01% versus 0.06%; P=0.67) in the HF group than in the NCD group. High dose of Sil supplementation enrich the relative abundance of these change significantly, while no obvious change was observed in low-dose to Akkermansia. Sil supplementation showed inhibitory effect on well-known beneficial bacteria, such as Alloprevotella, although no significant difference was shown between NCD and HFD-fed groups.

Changes of bacterial-produced SCFAs after Sil treatment in HFD-fed mouse.

As one of the most important metabolites of microbiota, SCFA involved in NAFLD pathogenesis through influencing metabolism, immunity and so on. In this study, we examined cecal concentration of SCFAs, including formate, acetate, propionate, butyrate. Compared with mice from NCD group, levels of acetate, propionate, butyrate increased by 29.0% (P<0.05), 20.7% and 48.2% (P<0.05) in HFD-fed mice, Sil supplementation dose dependently increased the cecal concentration of these SCFAs, particularly, high dose resulted in a significant difference. Being treated with Sil for 10 weeks, the risen formate in HFD-fed mice decreased with concentration increased, a significant difference was shown in Sil (300mg/kg)- treated groups (Figure 5).

Response of cecal bile acid concentrations to Sil treatment

Intestinal microbial plays an important role in regulating bile acid metabolism, in this study, we determined effects of Sil on cholic acid (CA), muricholic acid (MCA), deoxy-cholic acid (DCA), and lithocholic acid (LCA). As shown in Table 1, being compared with NCD group，levels of CA, MCA from HFD group increased significant, Sil resulted in lower concentrations of these bile acids, CA reduced by 61.7% in HFD mice treated with Sil (300mg/kg) (P = 0.008), but no other significant differences were shown. As compared with NCD groups, there was a tendency toward higher levels of DCA and LCA in mice from HFD group (P<0.01), Sil caused improvement to these changes with the dose increased, levels reduced by 74.8% ( P<0.05) and 56.2% (P<0.05) for HFD+ Sil (300mg/kg) group, respectively.

Intestinal tight junction protein expression

Intercellular tight junction protein played crucial role in intestinal permeability. Western blot was used to examine effect of Sil on expression of JAM-A, Occludin, ZO-1 and Claudin. As expected (Figure6), Occludin, ZO-1 and Claudin expression in the colon of mice in HFD groups was significant lower than that in NCD group mice, Sil treatment caused a significant increase in the level of Occludin and ZO-1.

Discussion

With obvious clinical therapy for NAFLD, Sil affords pharmacological protection against liver diseases. Ni et al11. reported that silymarin (30 mg/kg) could attenuate hepatic steatosis, with no effect on body weight, food intake. Our results demonstrated that Sil (100 mg/kg, 300 mg/kg) administration could exhibited protective effects against dietary- induced obesity, insulin resistance and liver steatosis in mice, it could also prevent weight gain with the dose increased, significant inhibition to body weight appeared at the dose of 300 mg/kg. The mechanisms of Sil in its protective effects to liver received much attention. Accumulating evidence indicates that gut microbiota correlated with the pathogenesis of obesity-related metabolic diseases like nonalcoholic fatty liver disease (NAFLD) 12,13. The therapeutic role of Sil in NAFLD may partially attribute to alteration of gut microbes. In this report, after oral administration with Sil for 10 weeks, we examined changes of the gut microbiota and metabolites produced by gut microbiota in HFD-fed mouse. It was found that relative abundances and structure of gut microbiota in HFD-fed mice changed, Sil treatment resulted in no significant effect to diversity and richness in obese mice. Consistent with amelioration of NAFLD, we observed an altered cecal microbiota composition.
Accumulating evidence consistently showed that HFD could induce an increase in Firmicutes and a decrease in Bacterioidetes14-16, when interventions were performed to decrease body weight, the ratio of Firmicutes to Bacteroidetes (F/B) decreased. In this study we also found similar phylotype-specific changes of the gut microbiota in HFD-fed mice, compared with NCD-fed mice，Sil exhibited invert the influence on the F/B ratio in HFD-fed mice. As one of the microbial metabolites, short-chain fatty acid (SCFA) could protect against inflammation, pathogen-induced damage of mucosa, and improve pathogenesis of obesity-related metabolic diseases17-19.Previous report showed that SCFA-producing bacteria benefit the host, there was a decreased of such bacteria in metabolic diseases(such as NAFLD).SCFA producers distributed among a few bacterial genera relatively, such as most members in Akkermensia, Allobaculum, Bacteriodesand Blautia19-21. In this study, we showed that, Sil markedly enriched the SCFA-producing Blautia, Bacteriodesand Akkermensia, while with no obvious effect to Allobaculum in mice,
We examined concentration of SCFAs, which included formate, acetate, propionate and butyrate. The result confirms earlier study that cecal content in lean mice contained more SCFA, compared with obese ones19,22, Sil induces increased levels of acetate, propionate and butyrate. Meanwhile Sil treatment inhibited the elevated level of formate, it was known that formate rises in inflammatory conditions. Acetate promotes antilipolytic activity and participates in inflammation control, propionate inhibits production of hepatic glucose and cholesterol. Thus, Sil might reverse the reduced abundance of SCFA-producing bacteria, and promote production of the intestinal SCFAs accordingly, exhibited beneficial effect to liver protection, eventually.
Intestinal microbiota–bile acid axis plays important role in mediating cholesterol homeostasis and HFD-induce NFLD. Primary bile acids, including cholic (CA) acid and β-MCA, are synthesized in the liver and further transformed by intestinal flora into secondary bile acids, such as deoxycholic acid (DCA) and lithocholic acid (LCA)23. It is known that LCA is cytotoxic to hepatocytes, while DCA could promote cholestasis. Some intestinal flora, including some member of Lactobacillus and Bacteroides, had Bile salt hydrolase (BSH) activity to mediate the generation of LCA and DCA7,24-26. In this study, we found Sil could inhibit the abundances of Lactobacillus spp and increase level of Bacteroides. The reduced level of Lactobacillus might contribute to decreased conversion of cytotoxic secondary metabolites. Therefore, Sil may sequester potentially cytotoxic bile acids in cecal and used for treating NAFLD.
Inconclusion, our study indicated that alterations of microflora composition, induced by Sil, are associated with prevention and treatment of NAFLD. In particular, enrichment of the SCFA-producing bacteria elevated SCFA levels(including acetate, propionate and butyrate), accordantly increased level of Lactobacillus mediate transformation of cytotoxic secondary metabolites (such as LCA and DCA), which helps to alleviate systemic inflammation and improve metabolic diseases. The decrease of Firmicutes in HFD-fed mice, induced by Sil, might result in reduced capacity of harvesting energy from the diet. What's more, through the protection of Occluding and ZO-1, which mediated colonic barrier integrity in mice, Sil might restore colonic barrier integrity and gut microbiota composition effectively.

References

[1] Maurice J, Manousou P. Non-alcoholic fatty liver disease.[J]. Clinical Medicine, 2018, 18(3):245-250.
[2] Cazanave S, Podtelezhnikov A, Jensen K, et al. The Transcriptomic Signature Of Disease Development And Progression Of Nonalcoholic Fatty Liver Disease[J]. Scientific Reports, 2017, 7(1):17193.
[3] Anna A, Rita C, Valerio N. Pathogen- or damage-associated molecular patterns during nonalcoholic fatty liver disease development [J]. Hepatology, 2011, 54(5):1500-1502.
[4] Suk K T, Kim D J. Gut microbiota: novel therapeutic target for nonalcoholic fatty liver disease[J]. Expert Review of Gastroenterology and Hepatology, 2019, 13(3).
[5] Miura, Kouichi. Role of gut microbiota and Toll-like receptors in nonalcoholic fatty liver disease[J]. World Journal of Gastroenterology, 2014, 20(23):7381.
[6] Nicholson J K, Holmes E, Kinross J, et al. Host-Gut Microbiota Metabolic Interactions[J]. Science, 2012, 336(6086):1262-1267.
[7] Fiorucci, Stefano, Distrutti, et al. Bile Acid-Activated Receptors, Intestinal Microbiota, and the Treatment of Metabolic Disorders[J]. Trends in Molecular Medicine, 2015, 21(11):702-714.
[8] Zhang Y, Hai J, Cao M , et al. Silibinin ameliorates steatosis and insulin resistance during non-alcoholic fatty liver disease development partly through targeting IRS-1/PI3K/Akt pathway[J]. International Immunopharmacology, 2013, 17(3):714-720.
[9] Aghazadeh S, Amini R, Yazdanparast R, et al. Anti-apoptotic and anti-inflammatory effects of Silybummarianum in treatment of experimental steatohepatitis[J]. Experimental & Toxicologic Pathology, 2011, 63(6):569-574.
[10] Muto A, Takei H, Unno A, et al. Detection of Δ4-3-oxo-steroid 5β-reductase deficiency by LC–ESI-MS/MS measurement of urinary bile acids[J]. Journal of Chromatography B, 2012, 900(900):24-31.
[11] Ni X, Wang H. Silymarin attenuated hepatic steatosis through regulation of lipid metabolism and oxidative stress in a mouse model of nonalcoholic fatty liver disease (NAFLD)[J]. Am J Transl Res, 2016, 8(2):1073-1081.
[12] Mariana M, Helena C P. Diet, Microbiota, Obesity, and NAFLD: A Dangerous Quartet[J]. International Journal of Molecular Sciences, 2016, 17(4):481-.
[13] Leung C, Rivera L, Furness J B, et al. The role of the gut microbiota in NAFLD[J]. Nature Reviews Gastroenterology & Hepatology, 2016.
[14] Turnbaugh P J, Ridaura V K, Faith J J, et al. The Effect of Diet on the Human Gut Microbiome: A Metagenomic Analysis in Humanized Gnotobiotic Mice[J]. Science Translational Medicine, 2009, 1(6):6-6.
[15] Hildebrandt M A, Hoffmann C, Scott A. Sherrill–Mix, et al. High-Fat Diet Determines the Composition of the Murine Gut Microbiome Independently of Obesity[J]. Gastroenterology, 2009, 137(5):1716-172400.
[16] Yokota A, Fukiya S, Islam K B M S, et al. Is bile acid a determinant of the gut microbiota on a high-fat diet? - Gut Microbes - Volume 3, Issue 5[J]. Gut Microbes, 2012, 3(5):455-459.
[17] Zhao Y, Wu J, Li J V, et al. Gut Microbiota Composition Modifies Fecal Metabolic Profiles in Mice[J]. Journal of Proteome Research, 2013, 12(6):2987-2999.
[18] Bashiardes S, Shapiro H, Rozin S, et al. Non-alcoholic fatty liver Lithocholic acid and the gut microbiota[J]. Molecular Metabolism, 2016, 5(9):782-794.
[19] Morrison D J, Preston T. Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism[J]. Gut Microbes, 2016:1-12.
[20] Lukovac S, Belzer C, Pellis L, et al. Differential Modulation by Akkermansiamuciniphila and Faecalibacteriumprausnitzii of Host Peripheral Lipid Metabolism and Histone Acetylation in Mouse Gut Organoids[J]. Mbio, 2014, 5(4):91-7.
[21] Zhang X, Zhao Y, Xu J, et al. Modulation of gut microbiota by berberine and metformin during the treatment of high-fat diet-induced obesity in rats[J]. Scientific Reports, 2015, 5(6):14405.
[22] Turnbaugh P J, Ley R E, Mahowald M A, et al. An obesity-associated gut microbiome with increased capacity for energy harvest[J]. Nature (London), 2006, 444(7122):1027-131.
[23] Chiang, JYL. Bile acids: regulation of synthesis[J]. The Journal of Lipid Research, 2009, 50(10):1955-1966.
[24] Gérard P. Metabolism of Cholesterol and Bile Acids by the Gut Microbiota[J]. Pathogens, 2013, 3(1):14-24.
[25] Li F, Jiang C, Krausz K W, et al. Microbiome remodelling leads to inhibition of intestinal farnesoid X receptor signalling and decreased obesity[J]. Nature Communications, 2013, 4(3):2384.
[26] Muhammad N A, Bassis C M, Li Z, et al. Calcium Reduces Liver Injury in Mice on a High-Fat Diet: Alterations in Microbial and Bile Acid Profiles[J]. PLOS ONE, 2016, 11(11):e0166178-.