Nonalcoholic fatty liver disease (NAFLD) covers different degrees of metabolic abnormalities, from simple steatosis (SFL) to steatohepatitis (NASH), liver fibrosis (NLF) and finally cirrhosis, which is the main cause of end-stage liver disease and hepatocellular carcinoma (Higarza et al., 2021). Nowadays, the prevalence rate of NAFLD is 25.24% worldwide (Lin et al., 2021), and the incidence rate is rising year over year along with the prevalence of obesity globally, and is predicted to reach 100.9 million in 2030 (Estes et al., 2018). Thus, NAFLD causes a huge social disease and economic burden to humans.

In recent years, more and more evidence has shown that intestinal microbiota are the source of hepatotoxic oxidative damage (Borrelli et al., 2018), and they play an important role in the occurrence and development of NAFLD (Liu et al., 2019). After Le Roy et al. first confirmed the decisive role of intestinal microbiota on the development of NAFLD through intestinal microbiota transplantation (Le Roy et al., 2013), an increasing number of studies have found that the alteration of intestinal microbiota promotes the occurrence and development of NAFLD by affecting liver gene expression, energy expenditure, lipid metabolism, insulin resistance and cytokine production. For example, altered intestinal microbiota affects the synthesis of triglycerides and the metabolism of fatty acids, increases toxic lipid metabolites, and then affects mitochondrial function and the oxidative stress response, and ultimately promotes the occurrence of liver inflammation (Leung et al., 2016). Then, altered intestinal microbiota leads to increase of intestinal mucosal permeability and bacterial translocation, which will aggravate the development of liver inflammation (Li et al., 2022). Moreover, NAFLD patients have an increased abundance of gram-negative bacteria in the gut microbiota, which means there are more lipopolysaccharides released into blood from the outer membrane after the death of these bacteria, and endotoxemia will promote inflammatory cytokines release and inflammatory pathway activation, and then promote the development of NAFLD (Khan et al., 2019; Fei et al., 2020). Also, altered intestinal microbiota can regulate bile acid metabolism and inhibit the farnesol-like X receptor (FXR) signaling pathway to promote the progression of NAFLD (Jiao et al., 2018). Additionally, altered intestinal microbiota increases the synthesis of endogenous ethanol, disrupts intestinal tight junctions and increase intestinal mucosal permeability, where endotoxin and ethanol cause direct damage to the liver (Yang et al., 2019; Huang et al., 2021). It is worth noting that intestinal microbiota and NAFLD may interact with each other. On the one hand, intestinal microbiota determine the occurrence and development of NAFLD (de Minicis et al., 2014), and on the other hand, the occurrence of NAFLD can also lead to changes in the intestinal microbiota of mice and humans (Henao-Mejia et al., 2012). However, it is unclear how different stages of NAFLD affect the composition of the intestinal microbiota.

NAFLD currently lacks definitive therapeutic agents (Loomba and Sanyal, 2013). However, based on current research, antioxidants, antidiabetic drugs, and lipid-lowering agents have demonstrated certain therapeutic effects on some patients with fatty livers (Cossiga et al., 2019). With further investigation, targeting the gut microbial composition with prebiotics, probiotics, and synbiotics has introduced a new therapeutic strategy for NAFLD (Safari and Gérard, 2019). Nevertheless, it should be noted that the aforementioned medications primarily target specific aspects of the disease, highlighting the need for a comprehensive drug that addresses the multifactorial nature of NAFLD.

Sheng-Jiang Powder (SJP) is a classical Traditional Chinese Medicine (TCM) prescription for the treatment of liver diseases (Chen et al., 2017). It contains four natural medicinal ingredients: Rhubarb (Rheum officinale Baill.), Curcuma (Curcuma longa L.), Stiff silkworm and Cicada slough. This prescription is derived from Ten Thousand Diseases Reviving Spring written by Gong Tingxian in the Ming Dynasty of China and has been used clinically for more than 400 years (Chen et al., 2017). Modern pharmacological studies have shown that SJP has good lipid reduction, insulin resistance improvement and anti-inflammatory effects (Miao et al., 2018). Our present studies have demonstrated that SJP significantly inhibited the body weight gain and attenuated liver pathological damage in mice with SFL induced by high-fat diet feeding, and potential mechanism may be related to the regulation of lipid metabolism related signal pathways and intestinal microbiota (Li et al., 2018). Thus, SJP has shown comprehensive effects that can target the complex physiopathology of NAFLD. However, our present results were limited in SFL mice. Whether SJP can regulate the lipid metabolism and intestinal microbiota and further attenuate liver damage in mice with NASH or NLF needs to be further verified.

Therefore, the purpose of this study was to investigate the effects of SJP on liver injury and intestinal microbiota in mice with varying severity of NAFLD, so as to clarify possible protective mechanisms for NAFLD and provide better evidence for clinical application.

In the present study, we established NAFLD mice models with different severity by special feeding and these models exhibited a series of clinical features that were consistent with different stages of NAFLD. Then, we found that the protective effect of SJP diminished with the NAFLD severity. Briefly, SJP decreased the body weight and serum TG level, regulated the expression of lipid synthesis related genes, improved liver function and hepatic steatosis in SFL mice. SJP reduced the liver enzyme level and hepatic lipid deposition but did not improve liver pathological damage in NASH mice. While in NLF mice, SJP appeared to only affect PPAR γ expression with no obvious protective effect on liver inflammation, steatosis or fibrosis. Despite the protective effect on liver pathology was limited, SJP did reduce hepatic lipid deposition in all NAFLD mice as we observed in oil red O staining. Similarly, we have detected obvious changes in intestinal microbiota in all NAFLD mice after SJP treatment.

Hepatic lipid synthesis is a rigorously regulated biological process that is extremely important for the synthesis of very low-density lipoproteins and tissue energy transfer. Excessive fat intake induces the massive production of triglycerides, the accumulation of lipid droplets in hepatocytes, and ultimately leads to the formation of fatty liver (Wu et al., 2020). The expression of lipid synthesis genes is regulated by a series of transcription factors, including Srebp1c, Lxr, PPARs, etc. (Sohrabi et al., 2020). Srebp1c is the main transcription factor regulating fatty acid synthesis in the liver. Over-expression of Srebp1c can affect lipid accumulation and oxidation in the liver and other non-adipose tissues. When there is excess energy, Srebp1c is activated, which leads to lipid accumulation and ultimately causes fatty liver, insulin resistance, insulin secretion deficiency, and dyslipidemia (Ju et al., 2020). In this experiment, hepatic expression of Srebp1c was significantly reduced in NASH and NLF mice, which seems to be contrary to the current generally increased Srebp1c expression in HFD induced fatty liver models. This may be related to the fact that CDAHFD feeding did not cause obesity in our mice models, as the same decreased Srebp1c expression was also found in the non-obese mouse model of NASH established with MCD feeding by Yamamoto et al. (2016). Pparγ also plays an important role in regulating lipid metabolism, especially in lipid production gene expression and liver lipid synthesis induced by HFD (Liu et al., 2019). Pparγ can improve insulin sensitivity, inhibit lipolysis of adipose tissue, reduce the release of free fatty acids from central adipose tissue, and prevent ectopic fat deposition in the liver and pancreas (Ryyti et al., 2021). Excess energy induces abnormal Pparγ activation in liver and then promotes the occurrence of hepatic fat deposition. In a HFD induced NAFLD animal model, the expression of Pparγ in the liver increased significantly and the liver gradually developed steatosis as time went on (Yu et al., 2003). A high level of Pparγ expression in the liver, however, can prevent fat deposition in other tissues from insulin resistance, while suppressing the expression of other nuclear transcription factors and their downstream inflammatory genes to reduce the inflammatory response, thus realizing self-protection of the body (Gross et al., 2017). In the present study, we found that the hepatic expression of Pparγ tended to increase in all NAFLD mice models, especially in NASH mice, while the expression was further promoted by SJP treatment. The protective effect of SJP on NAFLD might be related to the regulation of Pparγ expression. In addition, Fasn is a fatty acid synthase that is mainly involved in the synthesis of long-chain saturated fatty acids such as palmitic acid, and its expression is regulated by transcription factors such as Srebp1c and Pparγ (Kim et al., 2022). Further studies on the expression of lipid synthesis proteins at different stages of NAFLD may help us better understand how lipid metabolism disorders affect the pathophysiological process of NAFLD.

Intestinal microbiota is a complex ecological community that performs many functions vital to health and survival, including digesting food, producing anti-inflammatory substances, promoting metabolism, and regulating the immune system (Yan et al., 2022). Microbiota affect the physiology and susceptibility to disease of the host through their collective metabolic activities and host interactions. Therefore, the composition of the microbiota at the community level varies with the physiological state of the host and is closely related to the health of the host (Liu et al., 2022). Bacteria are the main members of the intestinal microbiota, and the intestinal microbiota of mice mainly comes from Firmicutes, Bacteroidetes, Verrucomicrobia, Actinobacteria, and Proteobacteria. Among them, Firmicutes and Bacteroidetes constitute the largest components of intestinal microbiota (Duan et al., 2010). Factors such as age, disease, host genes, and dietary habits all affect the composition of the microbiota. Indeed, it is generally believed that diet has a substantial impact on regulating the structure of the intestinal microbiota, which in turn affects intestinal metabolism (Li et al., 2022). Integrative analysis of the microbiota and metabolome of the human intestinal mucosal surface reveals their exquisite interrelationships. Accumulated evidences have revealed that HFD reduced the intestinal microbial diversity and increased Firmicutes/Bacteroidetes flora ratio, thus increased energy acquisition in the gut which provided more substrates for the activation of the lipogenesis pathway while inducing upregulation of genes involved in lipid metabolism in the distal small intestine in both humans and animals (Do et al., 2018). And the idea that increased Firmicutes/Bacteroidetes flora ratio is correlated with the development of NAFLD is well accepted nowadays. Many studies believe that the importance of the Bacteroidetes phylum bacteria in the gut flora is based on their effects on the production of short-chain fatty acids (SCFAs) and the metabolism of the intestinal microbes. For example, a 20% decrease in the fecal Bacteroidetes community and a corresponding increase in Firmicutes bacteria will result in a 150 kcal increase in energy obtained from the diet, while the SCFAs produced in the gut can provide energy to the liver, and they may also reduce energy harvest and protect against fatty liver by increasing Bacteroidetes/Firmicutes ratio (Jumpertz et al., 2011). Our results also detected a significant increase in Firmicutes phylum bacteria in all of the NAFLD mice despite there was no significant changes in the proportion of Bacteroidetes phylum bacteria. In fact, many SCFAs-producing bacteria are from Firmicutes phylum, such as Roseburia, Blautia, Fusobacterium, Coprococcus, Ruminococcus and so on. The increased abundance of these bacteria will promote glucose and lipid metabolism, hence benefit for NAFLD.

At the genus level, we found that most of the microorganisms that contributed to the significant differences between groups in NAFLD mice with varying severity had the ability to produce SCFAs, including Faecalibaculum, Blautia, Lachnospiraceae NK4A136-group, unclassified-f-Lachnospiraceae, Roseburia, norank-f-Lachnospiraceae, Ruminiclostridium-9, Eubacterium_fissicatena-group, norank-f-Erysipelotrichaceae, Lactobacillus, norank-f-Bacteroidales-S24–7-group, Bacteroidales and Allobaculum et al. SCFAs are the main end products of the bacterial fermentation of soluble dietary fiber and indigestible carbohydrates, mainly including formic acid, acetic acid, propionate and butyric acid (Koh et al., 2016). SCFAs have multiple effects on maintaining human health, such as protecting the intestinal mucosal barrier, acting as the nutritional and energy component of the intestinal epithelium, reducing inflammation, and enhancing gastrointestinal function. They essentially represent the main carbon flow from the intestinal microbiota to the host, changing the energy harvest and regulating metabolic processes, including regulating appetite, promoting energy expenditure, stimulating insulin sensitivity, reducing energy harvest, and thus affecting the development of NAFLD (Morrison and Preston, 2016). It was found that the abundance of intestinal SCFAs-producing bacteria in patients with NAFLD became a characteristic change in gut microbial composition in NASH patients (Rau et al., 2018). Boursier et al. (2016) found that the severity of NAFLD was associated with gut dysbiosis and metabolic functional shifts in the gut microbiota, where increased Bacteroides abundance was independently associated with NASH, and increased Ruminococcus abundance was independently associated with liver fibrosis. In our study, the abundance of Lactobacillus and Bifidobacterium was significantly decreased in SFL mice, while the abundance of Desulfovibrio, unclassified-f-Lachnospiraceae and Lachnospiraceae NK4A136-group was significantly increased. In NASH mice, the abundance of intestinal Faecalibaculum, Akkermansia and Bifidobacterium was significantly decreased, while the abundance of Bacteroides, Lachnospiraceae NK4A136-group, unclassified-f-Lachnospiraceae, Blautia, norank-f-Lachnospiraceae, Unclassified-f-Peptostreptococeceae, Alistipes, Enterococcus and Ruminiclostridium-9 was significantly increased. In NLF mice, the abundance of norank-f-Bacteroidales-S24-7-group was significantly decreased, while the abundance of Bacteroides, norank-f-Erysipelotrichaceae, Eubacterium-fissiatena-group, Allobaculum, Unclassified-f-Peptostreptococeceae and Roseburia was significantly increased. According to current researches, the above apparently changing bacteria take part in a series of in vivo biological processes, such as metabolic and immune regulation. For example, Akkermansia has been demonstrated to be involved in regulating metabolic and immune function, protecting mice from HFD induced injuries and may also increase intestinal endocannabinoid levels, thus improving inflammation and intestinal barrier function (Derrien et al., 2017; Zhang et al., 2019). Allobaculum was reported to be positively correlated with the expression of angiogenin-like protein 4 (ANGPTL4), a key regulator of lipid metabolism and a circulating mediator of gut microbiota and fat deposition (Zheng et al., 2021). The abundance of intestinal Blautia and Bifidobacterium was found to be negatively correlated with human visceral fat (Li et al., 2022). Some researches revealed that the consumption of Blautia in obese children is related to the deterioration of intestinal inflammation and metabolic phenotype (Benítez-Páez et al., 2020). While other studies found the abundance of Blautia and Allobaculum was positively correlated with serum triglyceride, low density lipoprotein, IL-6, IL-1β, TNF-α, and body weight, but negatively correlated with serum high density lipoprotein level in HFD-induced fatty liver model (Tang et al., 2018). The norank-f-Erysipelotrichaceae is derived from the Erysipelotrichaceae family which has been shown to be associated with gastrointestinal inflammation. While the abundance of Erysipelotrichaceae was increased in mice with TNF-α-driven Crohn’s disease-like transmural inflammation, it significantly decreased in patients with recurrent Crohn’s disease. The above different performance may be due to host differences (Dey et al., 2013). Eubacterium_fissicatena-group, a butyrate producing bacteria, has been demonstrated to have a high correlation with host obesity and obesity-related metabolic disorder, and the increase of abundance can lead to an increase in the intestinal microbial choline metabolite trimlamine oxide (TMAO), which will affect lipid and hormone homeostasis, enhance the expression of adhesion proteins and induce vascular inflammation. Roseburia plays an important role in metabolism and inflammation regulation, and it was confirmed to be efficient in reducing liver steatosis and inflammatory response. In the present study, after analyzing the alteration in the gut microbiota of NAFLD mice with different severity with or without SJP treatment, we found a very interesting phenomenon: all the decreased bacteria were beneficial among the apparent changed SCFAs-producing bacterium; the abundance of bacterium with pro-inflammatory effects, such as Desulfovibrio, Enterococcus and Eubacterium_fissicatena-group was significantly increased; and the abundance of anti-inflammatory bacterium, such as Akkermansia, Lactobacillus, norank-f-Bacteroidales-S24-7-group and Bifidobacterium was significantly decreased. Moreover, SJP treatment can effectively increase the intestinal Roseburia abundance in SFL mice and the Akkermansia abundance in NASH and NLF mice. The beneficial effect of SJP on NAFLD may be associated with the regulation of the above bacterium.

In the present study, the functional prediction of the top 30 highly abundant species in the gut microbiota of NAFLD mice showed that the gut microbiota of SFL and NASH mice was significantly enriched in a series of level3 signal pathways and participated in various biological activities including metabolism, genetic information processing, environmental information processing, cell progression, human disease and organism system. There was no significant difference in the enrichment of gut microbiota in above signal pathways between the NLF mice and the controls. While SJP treatment not only increased the enrichment of gut microbiota in certain pathways in SFL mice, but also increased the enrichment in almost all the level3 pathways involved in NLF mice. Thus, we concluded that SJP may be involved in regulating multiple pathophysiological processes in HFD induced NAFLD mice model by regulating the gut microbiota. However, due to the limited understanding of the function of intestinal microflora and their specific role in vivo, the specific mechanism of how their changes affect the occurrence and development of NAFLD, and how the SJP regulates the intestinal microbiota still needs further study.

Current researches has demonstrated that SJP is benefit for NAFLD, but it is not clear whether it has a role in different severity and how it affects the severity of the disease. So we designed this study to further explore the effect of SJP on NASH and NLF based on our previous finding that SJP can improve HFD induced SFL through regulating the intestinal microbiota. In our study, unlike SFL mice models induced by HFD feeding, NASH and NLF mice models are induced by a special HFD feeding CDAHFD because even prolonged HFD feeding is difficult to establish the characteristic inflammation and fibrosis manifestations of the NASH and NLF models. Fortunately, CDAHFD feeding is currently a very mature method to establish NASH and NLF models, which can well simulate the histopathological changes of NASH and NLF. And future models that can better mimic NAFLD may allow for better validation of our current findings.

In summary, we found significant differences in the intestinal microbiota of NAFLD mice with different severity, and SJP could regulate the gut microbiota composition and thereby affect the related biological activities. SJP was much suitable for early treatment of NAFLD as the protective effect diminished with the disease severity and the potential mechanism might be related to the regulation of intestinal microbiota.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

The animal study was approved by Experimental Animal Ethics Committee, West China Hospital, Sichuan University. The study was conducted in accordance with the local legislation and institutional requirements.

PZ: Writing – original draft, Data curation, Formal analysis. JL: Writing – review & editing, Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology. YM: Writing – original draft, Data curation, Investigation. XZ: Writing – original draft, Formal analysis, Software. LZ: Writing – original draft, Investigation, Methodology. JY: Writing – original draft, Data curation, Methodology. MW: Writing – review & editing, Conceptualization, Supervision. WT: Conceptualization, Funding acquisition, Methodology, Supervision, Writing – review & editing.