New insights into the bile acid-based regulatory mechanisms and therapeutic perspectives in alcohol-related liver disease

Abstract

Cholestasis is a key causative factor in alcohol-related liver disease (ALD) and variable degrees of cholestasis occur in all stages of ALD. However, the pathogenetic mechanisms and biomarkers associated with cholestasis are not well characterized. Cholestatic disease is marked by the disruption of bile acids (BA) transport and homeostasis. Consequently, in both human and experimental ALD, the disease shows a direct correlation with an imbalance in BA equilibrium, which in turn may also affect the severity of the disease. Modulation of BA metabolism or signaling pathways is increasingly considered as a potential therapeutic strategy for ALD in humans. In this paper, we highlight the key advances made in the past two decades in characterizing the molecular regulatory mechanisms of BA synthesis, enterohepatic circulation, and BA homeostasis. We summarize recent insights into the nature of the linkage between BA dysregulation and ALD, including the abnormal expression of genes involved in BA metabolism, abnormal changes in receptors that regulate BA metabolism, and disturbance in the gut flora engaged in BA metabolism caused by alcohol consumption. Additionally, we provide novel perspectives on the changes in BAs in various stages of ALD. Finally, we propose potential pharmacological therapies for ALD targeting BA metabolism and signaling.

Introduction

Alcohol-related liver disease (ALD) comprises a wide range of hepatic disorders, including asymptomatic steatosis, alcoholic steatohepatitis (ASH), alcoholic liver fibrosis and cirrhosis, and alcohol-related hepatocellular carcinoma (HCC). ALD is one of the most prevalent global causes of chronic liver diseases. The mortality and morbidity rates associated with ALD have shown an increasing trend over successive years. According to the World Health Organization's latest Global Status Report on Alcohol, more than two-fifths of the world's population over the age of 15 years consumes alcohol, which accounts for approximately 2.3 billion people. Every year, approximately 35 billion liters of pure ethanol are consumed, which corresponds to three cups of pure ethanol per person per day (33 g of pure ethanol), demonstrating the growing popularity of alcohol [1]. According to epidemiological research, roughly 1,256,900 individuals perish annually due to chronic liver disease. About 27% of all deaths are attributed to alcohol abuse [2]. The pathogenetic mechanisms of ALD are not well characterized. At present, ethanol is believed to cause liver toxicity through modifying hepatocellular lipogenesis and boosting the secretion of endotoxins from the intestine, thereby promoting inflammation and oxidative stress. Furthermore, environmental and genetic variables, including sex, ethnicity, nutrition, liver disease complications, and oxidative stress, may have a fundamental impact on the progression of ALD [3]. While abstinence from alcohol remains the optimal option for ALD, this objective is frequently not met adequately or on time for a large number of patients. As a result, the search for effector chemicals that may be involved in ALD may help identify novel therapeutic targets. Cholestasis occurs in various phases of ALD, and disruption of transport and equilibrium of bile acids leads to the development of cholestatic disease. Alcohol consumption can induce abnormal changes in the levels and content of the BA pool in the human body, which further aggravates ALD. The intestinal microbiota (IM) plays an essential role in the development and progression of ALD. Alcohol intake can lead to changes in the composition and function of the gut microbiota. Microbial functions, especially those related to bile acid synthesis and biotransformation, can modulate alcohol-induced liver damage. Nearly all the proteins required for the synthesis, absorption, metabolism, and transport of BAs have the ability to modify the size and composition of the BA pool and BA signaling, making them desirable drug candidates for ALD. Here, we review the emerging evidence pertaining to the protective effect of BA signaling against ALD by altering several cellular and molecular pathways. Moreover, we summarize the potential mechanisms driving the alcohol-induced abnormal alterations in BAs, as well as the abnormal changes in the size and composition of BA pool observed in various stages of ALD. Finally, we discuss the prospects of use of agents targeting BA signaling in the therapy of ALD (Fig. 1).

Fig. 1.

The effect of alcohol abuse on bile acid metabolism and the therapeutic target for ALD based on the bile acid signaling pathway. Alcohol abuse increases CYP7A1 transcription while decreasing the expression of fibroblast growth factor receptor 4 (FGFR4). Upon alcohol exposure, the expression levels of the bile acid CoA: amino acid N-acyltransferase (BAAT) are dramatically decreased, whereas bile acid CoA synthetase (BACS) expression is marginally elevated. Chronic alcohol abuse upregulates the transcription of bile acid efflux transporters, such as the multidrug resistance-associated protein ¾ (MRP3/4) and the organic solute transporter α/β (OSTα/β) in the liver, and also upregulates the expression of the OSTα/β and the apical sodium-dependent BA transporter (ASBT) in the ileum. Furthermore, alcohol treatment inhibits the farnesoid X receptor (FXR) activity in the liver and intestine, which results in increased BA synthesis. Finally, chronic heavy drinking causes the dysbiosis of intestinal flora involved in bile acid metabolism. Both alcohol consumption and derivatives of lithocholic acid (LCA) can regulate the intestinal immune response. Beneficial targets for ALD along with the enterohepatic and cholehepatic BA circulation. Enterohepatic drugs acting within the gut–liver axis: FXR agonists and FGF19 analogs. Transport blockers: ASBT inhibitors and BA sequestrants. PPAR agonists. The red box indicates potential effective targets for ALD therapy based on bile acid metabolism. Bile acid transporters are indicated by the green box. Created with BioRender.com. CB1R cannabinoid receptor type 1, CREBH cAMP response element-binding protein hepatocyte-specific, DCA deoxycholic acid, FGF15/19 fibroblast growth factor 15/19, PPAR peroxisome proliferator-activated receptor, SHP small heterodimer partner, LRH-1 liver receptor homolog-1, CBA conjugated primary bile acids, UBA unconjugated bile acids, MRP2 multidrug resistance-associated protein2, BSEP bile salt export pump, IBABP ileum bile acid binding protein, NTCP sodium/taurocholate cotransporting polypeptide, OATP organic anion-transporting polypeptide 1, IM intestinal microbiota

Advances in the regulation of bile acids in ALD

Bile acid homeostasis

Bile acid is a lipophilic steroid having one or more hydroxylation sites and a carboxylic acid-terminated side chain. There are two main pathways for BA synthesis: the classical and alternative routes [4]. Bile acid is a byproduct of hepatic cholesterol metabolism. Primary BAs are synthesized in the hepatocytes and are largely delivered to the enteric cavity as amino acid conjugates. Approximately 95% of BAs are resorbed in the distal ileum and colon before being returned to the liver through the portal vein, a process known as enterohepatic circulation of BA. The remaining 3–5% of BAs excreted into the bile are not reabsorbed in the small intestine and pass through to the colon where they undergo deconjugation and multistep dehydroxylation processes carried out by intestinal microflora. The conjugated primary BAs are further converted to secondary BAs [5].

Under normal circumstances, the BA pool size remains constant. This is accomplished through the several "checkpoints" located across the enterohepatic circulation. Overexpression or underexpression of several transporter genes may result in an increase in BA reprocessing or a decrease in the BA reservoir through excretion. Pharmacological targeting of BA sites and carriers is a fast-emerging frontier in the therapeutics research for ALD.

Bile acid production and transport are controlled in hepatocytes and intestinal epithelial cells via a complex network of hormone-like signaling. Bile acids primarily sense the existing BA reserve in the liver and ileum by combining with the nuclear receptor FXR (farnesoid X receptor) and closely regulating de novo BA synthesis via negative feedback. Bile acid production and enterohepatic circulation are modulated by FXR, a transcription factor that is activated by ligands [6]. Two different pathways are used by FXR to regulate the production of BAs [7, 8]. Bindings of BAs with FXR inhibit the CYP7A1/Cyp7a1 and the CYP8B1/Cyp8b1 gene production in the liver by inducing small heterodimer partners (SHP) to engage with liver receptor homolog-1 (LRH-1) [9]. Bile acids also suppress their own synthesis in the gut via an SHP-independent route involving FGF15 (FGF19 in humans). Bile acids activate FXR, inducing the secretion of fibroblast growth factor 15/19 (FGF15/19) from the gut into the portal circulation. FGF15 phosphorylates fibroblast growth factor receptor 4 (FGR4), which activates the extracellular regulated protein kinases1/2 (ERK1/2) pathway and necessitates the presence of cofactors, β-klotho, to decrease the expression of CYP7A1/CYP7A1 and Cyp8b1/Cyp8b1 genes [10, 11].

Mechanism of abnormal changes in bile acids in ALD

Effect of alcohol consumption on the expression of genes involved in the BA metabolism

First, chronic alcohol intake alters the size of BA pool by upregulating the expression of BA synthesis genes. Chronic heavy alcohol consumption was found to increase Cyp7a1 and Cyp8b1 transcription while decreasing the expression of FGFR4, a CYP7a1 transcriptional inhibitor [12, 13]. This may be attributable to alcohol-induced activation of cyclic AMP-responsive element-binding protein H (CREBH) via stimulation of the hepatic cannabinoid receptor type 1 (CBIR) [14], which is resistant to activated FXR-FGF19 or FXR-SHP pathways. However, a recent study in humanized mice.

revealed that gut microbiota can also trigger CREBH and BA biosynthesis genes via the production of a molecule that activates the hepatocellular CB1R [15]. Collectively, it is still debatable if the modulation of the key enzymes in the production of BA in the hepatocytes is caused by intestinal flora or alcohol consumption. Additionally, CYP7A1 synthesis was discovered to be decreased in patients with alcoholic hepatitis (AH) [16], which may be a compensatory response in these patients. Further studies are required to discern the changes in BA synthesis genes at various stages of ALD and to further illustrate the mechanism of alcohol-induced BA biosynthetic pathways.

Second, chronic alcohol intake affects the composition of BAs by changing the metabolic enzymes that facilitate BA conjugation. Upon alcohol exposure, the expression level of the bile acid CoA: amino acid N-acyltransferase (BAAT) was dramatically decreased, whereas the expression of bile acid CoA synthetase (BACS) was marginally elevated. In rodents, BAAT has a far greater specificity for taurine than it does for glycine, and the bulk of BAs in rodents tend to be taurine conjugated, whereas glycine-conjugated BAs are more common in humans. Therefore, this results in more unconjugated BAs, which are more hydrophobic and more damaging to the hepatocyte membrane. Unconjugated BAs can have a beneficial or detrimental effect on hepatic and gastrointestinal diseases depending on their concentrations [17]. Bile acids can also acquire epimerization on 3-, 7-, and 12-carbon hydroxyl groups by the gut microbiota, altering their hydrophobicity and hence potential cytotoxicity [18].

Third, chronic alcohol intake disrupts BA homeostasis by modulating bile acid transporters. Chronic alcohol abuse increases the transcription of BA efflux transporters, such as bile salt export pump (BSEP), multidrug resistance-associated protein 3/4 (MRP3/4), and organic solute transporter α/β (OSTα/β) in the liver, and downregulates the expression of BA uptake transporters, such as sodium/taurocholate cotransporting polypeptide (NTCP). In the ileum, it upregulates the expression of OSTα/β and apical sodium-dependent BA transporter (ASBT) [13, 19], which may be a compensatory response to cholestasis induced by chronic alcohol exposure. However, the changes in BSEP are highly controversial, and some previous investigations have demonstrated that heavy drinking can significantly increase the expression of BSEP, thereby affecting the basal metabolism and equilibrium of BAs in alcohol-treated mice [19, 20]. In contrast, in patients with alcohol-related hepatocellular carcinoma, alcohol consumption was found to significantly decrease BSEP expression compared with liver cancer patients who did not consume alcohol [23]. Additionally, previous studies have indicated a dramatic decrease in the expression of BSEP in patients with HCC [21]. In addition, the levels of BSEP showed an inverse correlation with the progression and prognosis of liver cancer [22]. All in all, the detailed mechanism of the effect of alcohol on BSEP may change with different stages of ALD and requires further exploration.

Effect of alcohol consumption on the gut microbiota involved in BA metabolism

The mammalian intestinal microbiome comprises germs, fungi, archaea, and viruses. Numerous recent studies have revealed the important role gut flora plays in ALD. For example, Duan et al. used bacteriophages to specifically alter the microbial community, lowering cytolytic E. faecalis and its secreted cytolysin levels in the liver which are associated with the severity of liver disease and mortality, thereby eradicating alcohol-induced liver injury in humanized rats transplanted with fecal germs from cytolysin-positive patients with AH [24]. Additionally, individuals with ALD showed decreased fungal diversity and significantly increased Candida albicans overgrowth, with an increased fraction of Candida in the fecal fungal community. Candida hemolysin is a peptide toxin produced by C. albicans, a commensal intestinal fungus, which can cause dose-dependent injury to primary hepatocytes in vitro and is closely linked with the severity of liver illness and mortality in people with AH. This suggests that Candida hemolysin may be a viable pharmacological target for ALD patients [25]. Patients with AH show considerably higher levels of the anti–Saccharomyces cerevisiae antibodies (ASCA), which act as a systemic immune response against fungal compounds or fungi, such as Candida. Moreover, ASCA levels in conjunction with the Model for  end-stage liver disease (MELD) score were found to effectively predict the severity of AH [26].

The numerous microorganisms that are found in the gastrointestinal system play an essential role in BA bioconversion and have a significant impact on the size and composition of the BA pool. Moreover, variations in the composition of the BA pool can adversely affect intestinal flora (Fig. 2). Recent evidence suggests that derivatives of secondary BAs metabolized by intestinal bacteria, such as 3-oxoLCA and isoallo-LCA, found in both humans and rodents directly affect CD4⁺ T cells [27]: in mice, 3-oxo-LCA was found to directly bind to retinoic acid receptor-related orphan receptor- γ t (RORγt) and inhibit the differentiation of Th17 cells, whereas isoallo-LCA induced the differentiation of T_reg cells by stimulating the production of mitochondrial reactive oxygen species (ROS), which resulted in an increase in the forkhead box protein 3 (FOXP3) expression (a major transcriptional modulator of T_reg cell differentiation). In addition, mice fed with 3-oxo-LCA or isoallo-LCA showed reduced Th17 cells and increased T_reg cell differentiation in the lamina propria of the ileum, suggesting that BA metabolites modify the host immunological responses partly by regulating the function of T_reg cells and Th17 cells [28]. Moreover, CD4⁺ T cells play different roles in the pathogenesis of ALD by producing a characteristic profile of cytokines [29]. For example, Th17 cells not only promote liver inflammation and fibrosis in ALD by producing IL-17 but may also help in liver repair by stimulating IL-22 production [30, 31]. Excessive alcohol drinking also increases T regulator cells, causing broad immunosuppression [32]. In conclusion, bile acid metabolites can control host immune responses by directly modulating the TH17 and Treg cell differentiation. This, in turn, affects the onset and progression of ALD. 3-oxoLCA inhibits IL-17 production by Th17 cells, thereby reducing liver inflammation and fibrosis. Isoallo-LCA enhances Treg cell differentiation, thereby resulting in an increased risk of bacterial infection in patients with ALD. Elderly individuals over the age of 100 years show higher levels of intestinal 3-oxoLCA and isoallo-LCA compared to young individuals. This is in line with the notion that these metabolites are beneficial for human health [33]. The ability of intestinal microflora to biotransform-conjugated primary BAs into unconjugated secondary BAs is at the core of gastrointestinal metabolic homeostasis. The digestive system and alcohol are inseparably related. Alcohol consumption alters the function of intestinal microflora [34], especially those related to BA metabolism in the following two aspects.

Fig. 2.

The relationship between alcohol consumption, intestinal flora, and bile acids. AFLD alcoholic fatty liver disease, ASH alcoholic steatohepatitis. Created with BioRender.com

First, gut microflora responsible for the production of bile salt hydrolase (BSH) is out of balance. It is currently believed that functional BSH exists in all major bacterial and archaeal species in the human intestine, including Lactobacillus, Bifidobacterium, Clostridium, and Bacteroides [35]. BSH can convert taurine-conjugated and glycine-conjugated primary BAs to their corresponding free forms. Chronic alcohol intake is linked to qualitative and quantitative alterations in the intestinal flora. For example, compared with healthy humans, alcohol was found to augment the species richness of Proteus, Enterobacteriaceae, and Streptococcus, while reducing the richness of Bacteroides, Ackermann, and Coccobacteria [36]. The pathophysiology of this biological disorder is not well characterized. Alcohol-induced oxidative stress (which obligate anaerobes, like Bacteroides, do not handle well) and down-regulation of antimicrobial peptides, like defensin expression, are two potential mechanisms. Additionally, another metagenomic investigation revealed that the bacterial DNA coding glycocholic acid hydrolase was more abundant in the gut lumen of rodents administered ethanol than in the control group. According to the sequencing results, glycocholic acid hydrolase is principally synthesized by Lactobacillus, Lactococcus, Bacteroides, and Micrococcus [37]. Accordingly, an important reason is that reduced BA secretion in the intestine during cholestasis of ALD results in bacterial overgrowth in the small bowel because of the direct antibacterial effect of BAs; in addition, BAs also indirectly promote the generation of antimicrobial peptides via their signaling properties. Furthermore, another study of HCC patients and chemical-induced HCC mouse models [38] revealed that decreasing gut BSH-rich microorganisms may promote the progression of HCC by substantially reducing the proportion of serum secondary BAs in the overall BA pool of patients with HCC.

Second, the gut microflora responsible for the dehydroxylation reaction is out of balance. The intestinal flora engaged in the biotransformation of secondary BAs is mainly Clostridium and Eubacterium in thick-walled Helicobacter pylori, especially Clostridium. The primary BAs CA and CDCA can be converted into secondary BAs DCA and LCA, respectively, by 7-dehydroxylase-rich bacteria in the gut [39], which are the most plentiful secondary BAs in human beings, and are recognized to play a significant role in three metabolic pathways: Clostridium difficile outgrowth suppression [40], hepatocellular carcinogenesis promotion [41], and immunological response modulation [27]. For example, accumulation of DCA leads to liver injury. Obesity-induced lipoteichoic acid (LTA) (membrane phosphoteic acid, a common component of gram-positive bacterial cell wall) combines with the intestinal microbial metabolite DCA to enhance the expression of senescence-associated secretory phenotype (SASP) of the hepatic stellate cells (HSCs) and cyclooxygenase-2 (COX-2) via toll-like receptor 2 (TLR2). On one hand, SASP creates a microenvironment that promotes the occurrence of liver cancer. On the other hand, through binding to the prostaglandin E receptor 4 (PTGER4), COX-2-mediated prostaglandin E2 (PGE2)suppresses antitumor immunity, thus promoting the development of HCC [41]. Recent research in mouse models revealed that NKT cells whose expression is regulated by CXCL16 suppress carcinogenesis in the liver. CXCL16 expression was found to be positively correlated with the primary BA CDCA and negatively correlated with the secondary BA GLCA, suggesting that secondary BAs may contribute to the growth of liver tumors [42]. Furthermore, Jin et al. developed the gene-editing tool for the intestinal microbiota and determined the single-gene transfer method of intestinal Firmicutes and Clostridium in the complex gut microbes; they demonstrated that it is more favorable to investigate the role of a single microorganism related to BA metabolism in the complex gut flora [43]. Collectively, these findings indicate progressive advances in obtaining mechanistic insights into the relationships between alcohol consumption and BAs processed by intestinal flora as a result of the development of novel profiling approaches and techniques, including metagenomics and genetic manipulation.

Effect of alcohol consumption on the BA receptor involved in BA metabolism

The role of FXR in ALD

Following alcohol consumption, a small amount is digested promptly in the stomach, while the majority of leftover alcohol is absorbed via the gastrointestinal system and then transferred to the liver through the portal circulation. In the liver, alcohol is metabolized in three distinct ways. First, the majority of alcohol is converted in hepatocytes to acetaldehyde, mostly by alcohol dehydrogenase (ADH), and then to acetate via acetaldehyde dehydrogenase (ALDH). Acetate is quickly eliminated from the body as water and carbon dioxide. In addition to ADH, ethanol can be oxidized by the microsomal ethanol oxidation system (MEOS), which is primarily composed of the cytochrome P450 2E1 enzyme (CYP2E1). Finally, catalase is capable of eliminating alcohol. Due to the fact that it is a tetramer and can be expressed by colonic flora, it may result in the generation of acetaldehyde in the lower gastrointestinal tract [44]. The ADH genes, which are human alcohol metabolizing enzymes, have previously been identified as specific targets of the BA receptor FXR. FXR overexpression was found to increase ADH mRNA levels, whereas silencing of FXR prevented FXR agonists from affecting the ADH transcriptional activity [45]. Furthermore, FXR activation also reduced oxidative stress in hepatocytes by inhibiting ethanol-induced upregulation of the Cyp2e1 gene [12]. However, the effect of FXR on ALDH2 is largely unknown, and more basic and clinical experiments are required to further explore the potential mechanism.

Tissue specificity of FXR in ALD

Emerging evidence has revealed that pharmacological FXR agonists can attenuate liver steatosis and oxidative stress induced by chronic alcohol consumption [46], while whole-body FXR gene knockout was found to aggravate alcohol-induced liver injury [47]. These findings suggest an essential role of FXR in the occurrence and development of ALD. Nevertheless, systemic lack of FXR leads to an increased BA pool, making it hard to assess the separate impacts of FXR and BAs on the severity of ALD. Zhang et al. studied the tissue-specific function of FXR in the onset and development of ALD [48]. They found that hepatocyte-specific FXR knockout mice (FXRhep−/−) showed similar BA pool size as wild-type (WT) mice, which enabled them to explore the direct effect of hepatic FXR deficiency on the development of ALD independent of the increase in BA pool size. After alcohol treatment, hepatic FXR deficiency was not found to aggravate alcoholic liver injury in FXRhep−/− mice compared with WT mice. Specifically, they exhibited similar severity of alcohol-induced hepatic steatosis, inflammation, and fibrosis. Moreover, hepatic FXR deficiency did not alter the expression of genes engaged in alcohol metabolisms, such as ADHs, CYP2E1, and ALDHs. Therefore, hepatic FXR deficiency may only play a minor role in the occurrence of ALD. They then [49] compared the severity of the alcoholic fatty liver disease (AFLD) in WT and gut-specific FXR knockout (FXRInt−/−) mice after either a control or an ethanol administration. They discovered intestinal-specific FXR deficiency worsened the occurrence and development of ALD. In conclusion, FXR deficiency in the whole body and the intestines worsens alcohol-induced injury, but FXR deficiency in the hepatocyte has little effect on the occurrence and development of ALD.

Effect of alcohol consumption on FXR

Chronic alcohol abuse has been demonstrated to suppress FXR activation in previous investigations, and the activation of FXR by the specific agonist WAY-362450 saved FXR activity. Chronic alcohol consumption impairs the interaction between FXR and RXRα (retinoid X receptor-α) by increasing the acetylation of FXR, which leads to the inactivation of FXR. For the reason that chronic ethanol intake upregulates the ratio of [NADH:NAD+] and downregulates the level of NAD+, which may reduce the activity of the NAD + -dependent sirtuins1 (SIRT1) enzyme, resulting in the failure of deacetylation of FXR and an increase in acetylation [50]. Moreover, alcohol abuse may also reduce the activity of the NAD+-dependent (sirtuins5) SIRT5 enzyme in the same way. A recent study of human HCC samples and HCC mouse models found that SIRT5 deficiency can promote the progression of HCC. Lack of SIRT5 in hepatocytes was found to promote the abnormal accumulation of BAs in hepatocyte peroxisomes, forming an immunosuppressive tumor microenvironment [51]. This may also be one of the potential mechanisms by which alcohol consumption leads to hepatocellular carcinoma. Alcohol impairs FXR activity, aggravating the development of ALD from the following five pathophysiological aspects.

First, it worsens alcohol-related liver steatosis Steatosis is the initial response of the liver to chronic alcohol consumption, characterized by the buildup of lipids in hepatocytes. Chronic alcohol consumption increases hepatic lipogenesis and decreases hepatic lipolysis, resulting in hepatic steatosis. In wild-type and intestinal-FXR-deficient mice with normal liver histopathology, ethanol feeding only led to mild liver steatosis in wild-type mice; however, significant vesicular steatosis and bullous steatosis, as well as hepatocyte balloon degeneration and Mallory-Denk bodies, were observed in the latter, which were characterized by spherical lipid droplets of different sizes in the cytoplasm of hepatocytes [49]. This indicated that ethanol-induced impairment of FXR activity further exacerbates alcohol-induced liver steatosis. Deng et al. recently discovered that lack of FXR promotes liver adipogenesis via regulation of the pyruvate dehydrogenase kinase 4 (PDK4). PDK4 is a mitochondrial enzyme that inhibits the conversion of pyruvate to acetyl CoA by preventing the phosphorylation of pyruvate dehydrogenase complex (PDC), resulting in an increase in liver fat. Additionally, inhibiting PDK4 expression was found to alleviate hepatocyte lipid accumulation induced by FXR deficiency, providing additional knowledge of the potential mechanism of action of FXR agonists in the therapy of ALD [52].

Second, it worsens alcoholic steatohepatitis Inflammation in the liver has a profound impact on scarring, fibrosis, and eventually cirrhosis. Chronic ethanol consumption leads to impaired FXR activity, which mainly aggravates liver inflammation from three aspects. First of all, decreased FXR activity leads to increased intestinal permeability [47], leading to the entry of intestinal-derived pathogen-associated molecular patterns (PAMPs) into the liver. PAMPs in collaboration with damage-associated molecular patterns (DAMPs) activate innate immune cell receptors, such as toll-like receptors (TLRs) and nod-like receptors (NLRs). The signals sent by these receptors cause an increase in the expression of pro-inflammatory transcription factors (including NF-κB), which promotes the release of a large number of inflammatory factors and cytokines, resulting in liver inflammation, also known as the "liver inflammatory factor storm" [29]. For example, when lipopolysaccharides cross the intestinal barrier and reach the liver via the portal circulation, these bind to the TLR4 on liver Kupffer cells, causing the production of superoxide and tumor necrosis factor-α, which triggers liver injury [53]. In addition, the decrease in FXR activity upregulates liver lipid peroxidation. The level of malondialdehyde, a byproduct of lipid peroxidation, was found to increase after ethanol feeding in FXR−/− mice [47]. Finally, decreased FXR activity increases alcohol-mediated production of reactive oxygen species, further exacerbating oxidative stress.

Third, it worsens alcoholic liver fibrosis and cirrhosis Hepatic fibrosis is the process of repair of damaged cells and tissues in the liver caused by a variety of factors. Extracellular matrix deposition induced by activation of HSCs is a critical element in the onset and progression of hepatic fibrosis [54]. HSCs can be stimulated not only by chemokines but also by alcohol and its byproducts. The inflammatory reaction and the toxic reaction induced by the increase in BAs caused by the decrease in FXR activity activate HSCs and promote alcoholic liver fibrosis and alcoholic cirrhosis. Garrido et al. reported reduced levels of the microspherule protein 1(MCRS1) and NTCP in individuals with liver cirrhosis. More specifically, deletion of MCRS1 in hepatic cells increased the level of BAs in hepatic sinusoids, thus activating FXR on HSCs and promoting the progression of liver cirrhosis. This was accomplished by enhancing histone lysine acetylation of BA transporter genes, such as NTCP. These findings implied that modulation of the BA-FXR axis in hepatic fibrosis is critical for the progression of liver cirrhosis and is a viable therapeutic target for cirrhosis [55].

Fourth, it worsens the gut mucosal damage The gut mucosal barrier isolates the host from intestinal microorganisms and prevents the flow of bacteria and metabolites out of the enteric cavity. This barrier comprises three components: (1) physical barrier, which includes mucins connected by tight junction proteins and possibly intestinal epithelial cells; (2) biochemical barrier, which includes antimicrobial proteins; and (3) immune barrier, which includes immune cells. Previous experiments have shown that intestinal FXR also contributes to maintaining the integrity of the gastrointestinal wall and inhibiting pathogen overgrowth and translocation [56]. Huang et al. [49] examined the histology of the distal ileum in WT and FXRInt−/− mice fed a normal diet. They found that the ileum of intestinal-specific FXR gene knockout rodents displayed a deformed epithelial cell lining, expanded subepithelial space, degenerative changes, and perforated mucosal lining. Both WT and FXRInt−/− mice exhibited morphological alterations in their villi following alcohol feeding, including broadening and shortening of villi, as well as submucosa and muscular shrinkage. However, these histological abnormalities were more severe in FXRInt−/− mice. These findings suggest that an alcohol-induced decrease in FXR activity exacerbates the gut mucosal damage.

Fifth, it increases intestinal permeability Chronic ethanol intake increases the permeability of the gut, resulting in the impairment of gut barrier function and translocation of microorganisms [57]. E-cadherin is an intercellular adhesion molecule that is essential for the establishment of tight junctions between epithelial cells in the intestine. In the whole-body FXR knockout rats, the expression of E-cadherin is down-regulated, which may increase the likelihood of hepatocellular carcinoma in mice [58]. In a recent study [49], E-cadherin staining was performed on the small intestine of WT and FXRInt−/− mice on control diets. The basic protein level of E-cadherin in the ileal recess in FXRInt−/− mice was found to be markedly lower than that in WT mice. Furthermore, E-cadherin expression was also reduced by alcohol feeding in both strains of mice, but it was significantly reduced in FXRInt−/− mice. The intestinal permeability in vivo was evaluated by measuring the leakage of FITC-dextran in serum after oral administration. The serum FITC-dextran level in FXRInt−/− mice fed on a control diet was dramatically higher than that of WT mice. These findings indicate that decreased FXR activity caused by alcohol abuse can increase intestinal permeability and aggravate the occurrence and development of ALD by reducing the expression of E-cadherin.

Effect of alcohol consumption on the FGF15/19 involved in the bile acid metabolism

Under normal physiological conditions, FGF19 is primarily generated and released in the terminal ileum and shuttled into the portal vein circulation to modulate BA synthesis in hepatic cells. In an experiment conducted in 2013, ethanol administration to non-cirrhotic rats substantially reduced FGF15 expression, presumably explaining the increase in CYP7a1 and BA production [13]. In a study of patients with liver cirrhosis conducted in 2014, on the contrary, FGF19 levels in cirrhosis patients who were heavy drinkers were higher than those in cirrhosis patients who had ceased drinking and the healthy controls. This may be potentially attributable to a distinct modulation of the FXR-FGF19 axis in the setting of cirrhosis with chronic alcohol intake wherein CYP7A1 is not negatively regulated by FGF19 [59]. In a study of patients with alcoholic hepatitis conducted in 2018 [16], consistent with previous studies, the expression of FGF19 in patients with severe AH (SAH) was significantly higher than that in patients with alcohol use disorder and controls. Furthermore, serum FGF19 expression in patients with AH was found to be 100 times higher than that in patients with non-alcoholic steatohepatitis (NASH) or non-alcoholic fatty liver (NAFL), with serum levels of the latter being equal to those seen in normal control. There are two possible explanations for the rise in FGF19 levels in AH. On one hand, AH is characterized by a strong ductal response, and recent studies have shown that FGF19 can also be derived from biliary epithelial cells and ductal cells. On the other hand, systemic and hepatic FGF19 expression was shown to be increased in patients with extrahepatic cholestasis, which decreased after biliary drainage [60]. Moreover, patients with primary biliary cirrhosis (PBC) also have increased FGF19 levels [61], indicating that an increase in FGF19 levels is strongly associated with intrahepatic and extrahepatic cholestasis.

In univariate regression analysis, serum FGF19 demonstrated a significant positive correlation with serum levels of bilirubin and gamma-glutamyl-transferase (GGT) in AH. FGF19 demonstrated a significant negative correlation with the stage of fibrosis and the degree of neutrophil infiltration, suggesting that FGF19 may serve as a potential diagnostic and prognostic marker for AH and also facilitate the differential diagnosis of clinical diseases. In addition, in patients with SAH, FGF19 levels were substantially linked with 30-day mortality [16], and FGF19 may predict mortality in AH similar to the MELD score. Additionally, the expression of FGF19 increased with an increase in the severity of hepatic pathology in AH. Whether this phenomenon is a manifestation of disease activity is not clear. However, considering the advantageous anti-fibrosis and anti-cholestasis effects of FGF19 [62], beneficial metabolic effects [37], and liver regeneration, the compensatory upregulation of FGF19 seems more likely to be a response to the deterioration of liver health.

Abnormal changes in the size and composition of the BA pool in ALD (Table 1)

Table 1.

Abnormal changes in the BA pool in ALD

	TBA
	UBA					CBA
	PBA		SBA			PBA		SBA
Normal	CA	CDCA	DCA	LCA	UDCA	GCA	GCDCA	GDCA	GLCA	GUDCA
						TCA	TCDCA	TDCA	TLCA	TUDCA
ALD
AH	TBA CBA TUDCA LCA UDCA ↑						DCA ↓
Cirrhosis							↓	↑
ACLF	TBA β-MCA ↑						CA CDCA ↓
HCC		↑				↑	TCDCA ↑	TDCA ↓	TLCA ↓

CA cholic acid, CBA conjugated bile acid, CDCA chenodeoxycholic acid, DCA deoxycholic acid, GCA glycocholic acid, GCDCA glycochenodeoxycholic acid, GDCA glycodeoxycholic acid, GLCA glycolithocholic acid, GUDCA glycoursodeoxycholic acid, LCA lithocholic acid, PBA primary bile acid, SBA secondary bile acid, TBA total bile acid, TCA taurocholic acid, TCDCA taurochenodeoxycholic acid, TDCA taurodeoxycholic acid, TLCA taurolithocholic acid, TUDCA tauroursodeoxycholic acid, UBA unconjugated bile acid, UDCA ursodeoxycholic acid, ALD alcoholic liver disease, AH alcoholic hepatitis, ACLF acute-on-chronic liver failure, HCC hepatocellular carcinoma, β-MCA β-muricholic acid

Alcoholic hepatitis

Severe alcoholic hepatitis, identified by Maddrey scoring ≥ 32, has a bad prognosis and is the most severe form of ALD with mortality of 20 to 30% within 1 month after a presentation and 30 to 40% within 6 months after presentation [63, 64]. In a clinical study, patients with alcoholic hepatitis (ABIC score ≥ 9.0) had a higher rate of mortality up to 75% in the first 90 days [65]. The BA profile in AH has been analyzed in five recent investigations [66]. Brandl et al. [16] examined the serum BAs levels of 132 patients with AH and found that total and conjugated BAs levels were considerably higher in patients with AH compared with healthy controls and alcohol use disorders (AUDs). This is due to the fact that during cholestasis, primary BAs mostly enter the bloodstream via MRP3/4. They are secreted far less into the intestines. In addition, although the expression of CYP7A1 genes and the C4 levels, a marker of de novo synthesis of BAs, were notably reduced in patients with AH, the serum total BAs were still increased in AH. However, the limitations of the study included failure to address anomalies in BA levels and their interaction with the gut flora, as well as neglecting the effects of cirrhosis, which was seen in 75% of AH patients. Previous research [67] has demonstrated that patients with SAH have a special dysbiosis, making their hepatocytes more vulnerable to alcohol-related injury. This susceptibility was found by transplanting intestinal microflora from AH patients to mice. Therefore, further exploration of the structure of intestinal microflora in AH patients and its role in BA homeostasis is a key imperative. Ciocan et al. [68] assessed plasma and fecal BA profiles and gut microflora constituents in patients with AH. They found that plasma total BA and total UDCA levels in cirrhotic patients with SAH were higher than in cirrhotic patients without AH(Cir_noAH), and the fecal total BAs and secondary BAs were considerably lower than in Cir_noAH. Moreover, patients with SAH had a higher proportion of Actinobacteria phylum in the gut. Conversely, the proportion of Bacteroidetes phylum was smaller. The increase in the BA pool and its transformation to more hydrophobic and toxic BAs may be the cause of specific changes in intestinal microflora. At present, it is also not clear whether the changes in intestinal microorganisms are caused by the liver disease itself or the etiology of liver disease. However, a limitation of this study was the inclusion of patients with both alcoholic hepatitis and alcoholic cirrhosis. In addition, the study lacked a control group consisting of individuals who consumed excessive amounts of alcohol but did not have an obvious liver injury. In a study in January 2022 by Muthiah et al.^.[69], aligned with previous research, it was found that plasma-conjugated BAs and secondary BAs UDCA and LCA levels were increased in patients with AH, while DCA was decreased. This study, however, observed a significant rise in 7-HOCA levels (a marker of de novo synthesis of BAs) in patients with AH, consistent with the elevated serum BAs, which suggests that the synthesis of primary BAs in the hepatocytes may increase despite the evidence of cholestasis. This is contrary to the previously reported reduced level of C4 in AH, which needs further research to explore the mechanism. In a recent study of patients with AH [70], the changes in urinary BAs were consistent with those of serum BAs, indicating that urine BAs may be valuable non-invasive markers for early diagnosis of AH and may have a role in predicting the severity of AH.

Alcohol-related fibrosis and cirrhosis

Approximately 10–20% of individuals with ALD develop cirrhosis. Alcohol was implicated in the causation of over 2 million cases of liver cirrhosis in 2017 in the US [71]. The size and composition of BAs in the serum, duodenum, and feces in patients with alcohol-related cirrhosis were markedly different from those in healthy individuals. Kakiyama et al. [59] discovered that serum-conjugated DCA and serum-conjugated CDCA considerably increased and decreased, respectively, in chronic alcoholics (both cirrhotic and non-cirrhotic) compared with the control groups. In feces, secondary BAs and the ratio of secondary BAs to primary BAs were higher in cirrhotic patients without alcohol abstinence than in cirrhotic patients who had abstained from alcohol and did not consume alcohol. Elevated biotransformation of primary BAs to secondary BAs due to changes in the gut microbes, a reduction in colonic transfer time due to heavy alcohol consumption, or a raised "overspill" of secondary BAs into the colonic cavity due to elevated total BA synthesis in the duodenum and ileum as a result of increased alcohol consumption are all possible explanations for the increase in secondary BA levels. Bajaj et al. [72] discovered that the conjugated bile acids in the duodenal fluid and feces of cirrhotic patients who were current alcohol users transformed from relatively harmless taurine-conjugated bile acids to comparatively toxic glycine-conjugated bile acids. It is believed that this is largely due to the dysbiosis reducing the bioavailability of taurine and causing an elevated enterohepatic cycling rate. Moreover, the latest evidence has shown that the shift in the serum BA profile toward a "healthy" composition may not only prevent but also restore already impaired neutrophil dysfunction in cirrhotic patients (alcohol-associated cirrhosis accounts for 50%), which will result in a decrease in bacterial infections and related morbidity and mortality in people with liver cirrhosis [73].

ALD-ACLF

Acute-on-chronic liver failure (ACLF) refers to the development of acute hepatic decompensation in the setting of chronic liver disease. The condition is often accompanied by multiple organ failures and is associated with a high death rate in a short period. Alcohol-related cirrhosis is the background of ALD-ACLF. In a prospective cohort, 65% of patients with severe AH [74] (sAH, Maddrey score ≥ 32) were found to have developed EASL-CLIF ACLF either at the time of diagnosis or within a 6-month follow-up period. However, whether ALD-ACLF is a specific form of ALD or simply a clinical progression of SAH is still uncertain. Previous studies have shown that approximately 33% of hospitalized cirrhotic patients develop ACLF, which is closely linked to an increased mortality rate [75].

In a 1-year prospective cohort study of 143 patients with established cirrhosis (50% of patients with alcohol-related cirrhosis) [76], the more advanced was the organ failure, the higher was the total BA(TBA) level, suggesting that TBA may be a potential indicator of disease severity in patients with ACLF. Second, on receiver operating characteristic (ROC) curve analysis, TBA ≥ 36.9 µmol/L during follow-up was found to be the optimal cut-off level for predicting the new onset of AD (acute decompensation) or ACLF. These findings suggested that total BAs and individual BAs can help predict the new onset of AD or ACLF. Although this was the first research to evaluate the role of serum BAs in the development of AD and ACLF in individuals with liver cirrhosis, this study had some limitations. On one hand, patients with liver cirrhosis caused by cholestatic diseases (such as PBC, PSC, gallstones, and tumors) were excluded from this study. More investigations are required to ascertain whether the BAs profile is different in people with and without basic cholestatic liver disease. Secondly, this study did not assess the serum levels of BAs according to the etiology of liver cirrhosis. The changes in serum BAs in AD/ACLF caused by liver cirrhosis of different etiologies may be different. Additional research should be conducted to determine whether BA levels can be used as non-invasive markers to predict the onset of AD or ACLF in cirrhotic patients. Additionally, future experiments should clarify whether modulating serum BAs with resin or specific BA uptake blockers has an effect on the disease course in individuals with decompensated liver cirrhosis.

In a recent first experimental research on the relationship between the dynamic balance of BAs and acute-on-chronic liver injury (ACLI) [77], the researchers modeled an alcohol-associated ACLI animal model that resembles ACLF based on the Gao-binge model (chronic-plus-binge alcohol feeding) [78] in Abcb4 gene knockout mice (called the Ko-EtOH mouse group) in order to better understand the acute inflammatory response in chronic liver disease. The Ko-EtOH mouse group showed worsened hepatic injury [79]. Compared with the Wt-Cont group, the Ko-EtOH group presented higher β-MCA, and lower CA and CDCA levels. Moreover, the levels of CYP8b1 and CYP7a1 in the liver were significantly inhibited in this model, which is compatible with the transformation of the BA synthesis pathway to an alternative pathway. In addition, alcohol exposure significantly increased FGF21 levels in the livers of WT and KO mice. Some studies have shown that FGF21 is a novel modulator of alcohol desire in rodents and humans, by demonstrating a significant increase in the expression level of FGF21 after acute and chronic drinking [80]. However, there were no discernible differences in the expression levels of FGFR1, FGFR4, FXR, and SHP in the liver and plasma FGF15/FGF19 after alcohol stimulation [77]. This indicates that FGF21 induced by chronic alcohol abuse inhibits the CYP7a1 gene, independent of the FXR-SHP or FXR-FGF19-FGFR4 pathway. Additional research is necessary to identify the relationship between FGF21 and CYP7A1 in different stages of ALD.

Alcohol-related hepatocellular carcinoma

A cohort study showed that approximately 250,000 patients died of alcohol-related liver cancer in 2016, accounting for 30% of all hepatic tumor deaths [81]. Alcohol-induced BA disorder may be one of the essential and direct factors in the progression of hepatocellular carcinoma caused by alcohol abuse. In a recent study [23], the researchers collected tissue samples from HCC patients, including alcoholics and non-alcoholics, and discovered that alcohol abuse significantly disrupted the equilibrium of BAs. The levels of 4 primary BAs (CDCA, GCA, TCA, and TCDCA) were significantly higher, while the levels of 3 secondary BAs (TLCA, TDCA, and HCA) were significantly lower (decrease by 76%) in HCC patients who consumed alcohol; in addition, the level of GDCA (secondary bile acid) was 3.3 times higher in HCC patients who consumed alcohol. Furthermore, a previous study revealed that BA imbalance may lead to the formation and development of HCC through the downregulation of a variety of tumor suppressor genes [21]. In addition, only BSEP expression was dramatically decreased in patients with liver cancer who abused alcohol. Patients with BSEP deficiency were found to be at a higher risk of developing significant hepatobiliary disorders, with approximately 15% of cases progressing to HCC or cholangiocarcinoma [82]. These findings collectively suggest that alcohol can trigger BA dyshomeostasis by reducing the expression of BSEP to promote the progression of HCC.

Other liver diseases

Abnormal changes in BA levels and composition are also observed in other chronic liver diseases. Numerous studies have discovered significant differences in the size and composition of BA pool in mouse models and biological samples from patients with metabolic associated fatty liver disease (MAFLD) or NASH. For example, mice who were fed a methionine-choline-deficient diet (MCD) showed a significant increase in serum levels of Tβ-MCA and TCA [83], and their levels returned to normal after methionine or choline supplementation. In a recent study of patients with MAFLD [84], plasma TCA levels and GCA levels showed a positive correlation with the increase in inflammation and fibrosis, respectively. In patients with HBV infection, all HBV phases showed downregulation of metabolic pathways, such as BA metabolism [85]. In cholestatic diseases, such as PBC and primary sclerosing cholangitis (PSC), the strong inflammatory reaction in the bile duct leads to progressive bile duct fibrosis and cholestasis. Cholestasis leads to increased levels of BAs in the hepatic tissue and circulation [86]. Highly toxic hydrophobic bile acids, such as CDCA, induce Rubicon, a protein that inhibits the normal execution of autophagy. BAs induce Rubicon production and inhibit autophagy through FXR dependence, which weakens the clearance of injured hepatocytes [87]. In a study [21], levels of TCA, GCA, TCDCA, DCA, and TLCA in the liver of the HCC animal model fed with a high-fat diet were considerably higher than those in the healthy control group. In addition, GCDCA has been shown to induce activation of autophagy by targeting the AMPK/mTOR pathway in hepatoma cells to promote the invasion and metastasis of HCC cells. Therefore, it may be a potentially useful prognostic marker in patients with HCC [88].

Advances in the BA-based therapies for ALD

So far, no drug has been found to reverse ALD. Therefore, characterization of the pathogenetic mechanisms of ALD is a key imperative to developing novel drugs. The regulatory roles of BAs and their signaling molecules are increasingly acknowledged as prospective targets for ALD management. Here, various drugs under research are reviewed (Table 2).

Table 2.

BA-based therapies for ALD

Compound		Therapeutic mechanism	Potential clinical effects
FXR agonist	OCA	Inhibiting BA synthesis and up-regulating BA transporters	OCA alleviated steatosis, fibrosis, and portal hypertension [90] OCA reduced bilirubin and ALP [91, 92] OCA improved ALT, AST, FIB-4 levels [93]
	Fexaramine	FGF15 ↑ CYP7a1 ↓	Fexaramine improved insulin sensitivity and stabilized the intestinal barrier [98]
FGF19 analogue	NGM282	Suppression of BA synthesis	NGM282 improved noninvasive markers of hepatic fibrosis but not ALP in patients with PSC [101] NGM282 mildly improved liver enzyme levels in patients with PBC [103]
ASBT inhibitor	GSK2330672	Inhibition of luminal BA uptake	GSK2330672 reduced pruritus in patients with PBC [107]
PPAR agonist	Seladelpar (MBX-8025)	Maintenance of BA balance	MBX-8025 improved cholestasis and reduced ALP [110]
Intestinal BA sequestrant	Pectin	Increasing BA secretion	Pectin stabilized the intestinal barrier and reduced liver damage [115]

ALP alkaline phosphatase, ALT alanine aminotransferase, ASBT apical sodium-dependent bile acid transporter, AST aspartate aminotransferase, BA bile acid, CYP7A1 cholesterol 7α-hydroxylase, FIB-4 fibrosis4Score, FGF15/19 fibroblast growth factor 15/19, FXR farnesoid X receptor, OCA obeticholic acid, PBC primary biliary cholangitis, PPAR peroxisome proliferator-activated receptor, PSC primary sclerosing cholangitis

FXR agonists

Obeticholic acid

Obeticholic acid (also known as INT-747) is a selective FXR agonist and a semi-synthetic derivative of CDCA. Compared with CDCA, OCA (6a-ECDCA) exhibits approximately 100-fold higher FXR agonistic activity. It protects hepatocytes from the toxicity of BAs by inhibiting BA synthesis and up-regulating BA transporters. OCA has been approved by the FDA as a monotherapy or supplemental therapy for PBC in patients who cannot tolerate UDCA or who do not respond adequately to UDCA. OCA has been shown to improve biochemical indicators of cholestasis [89]. In preclinical studies, OCA was found to alleviate steatosis, fibrosis, and portal hypertension [90]. In a clinical trial of PBC, OCA administered in combination with bezafibrate reduced the levels of cholestasis markers, such as bilirubin and alkaline phosphatase (ALP) [91]. In a phase II clinical trial for PSC [NCT02177136], OCA effectively reduced ALP levels [92]. In a multicenter study [NCT02548351], OCA improved ALT, AST, FIB-4 levels, and activity scores in patients with NASH [93]. OCA has now entered phase III clinical trials and is considered as the most effective drug for the treatment of MAFLD and NASH [94]. However, its effect on moderate and severe AH still needs to be confirmed in phase II clinical trial [NCT02039219], as the study was temporarily terminated due to safety concerns. The side effects of OCA include pruritus and gastrointestinal side effects, such as constipation and nausea; elevated total cholesterol and low-density lipoprotein levels; and reduced high-density cholesterol levels [95].

Fexaramine

Fexaramine is an intestinal-restricted FXR agonist. Firstly, fexaramine can increase the expression of FGF15 [96] and transfer it to the liver, thereby reducing the expression of BA synthase CYP7a1 in the liver and protecting the hepatocytes from the toxic effect of BAs. Secondly, fexaramine can alter the composition of intestinal flora by increasing Bacteroides and Acetatifactor [97], which are the dominant bacteria that convert CDCA into LCA, which is a secondary BA identified as a takeda G protein-coupled receptor 5 (TGR5) agonist. The increase in LCA activates TGR5 by inducing the expression of the gene encoding TGR5 to increase the excretion of the glucagon-like peptide-1(GLP1), which can improve insulin sensitivity [98]. Fexaramine has the advantage of intestinal restriction, which potentially avoids the systemic effects of FXR agonists, such as pruritus. In a recent study, administration of fexaramine was found to reduce alcohol-induced liver injury by steadying the gut barrier, which may be another reason for the reduction of liver inflammation observed after ethanol treatment [37]. However, further experimental and clinical studies are required to determine the role of intestinal-specific FXR agonists and antagonists in ALD.

FGF19 analog

NGM282

NGM282 is an analog of FGF19. Previous studies have demonstrated the carcinogenic effect of FGF15/19. Rats with overexpression of the human orthologue FGF19 were found more likely to develop HCC. Moreover, in mice lacking the FGFR4 receptor, the pro-mitogenic and pro-cancerogenic properties of FGF19 were weakened, indicating its role as a potential pharmacological target for the therapy of HCC [99]. NGM282, a N-terminal amino acid modified FGF19 analog, does not promote HCC but completely retains the regulatory activity of BAs, indicating that the effect of FGF19 on BA modulation could be independent of its tumor-promoting activity [100]. In a phase II clinical trial of PSC [NCT02704364], NGM282 effectively inhibited BA synthesis and reduced fibrosis markers without significantly affecting ALP levels [101]. In a phase II clinical trial of NASH [NCT02443116], NGM282 reduced liver adipogenesis and improved fibrosis [102]. Moreover, in a multicenter study of patients with PBC [NCT02026401] [103], NGM282 slightly improved liver enzyme levels. The main adverse reaction of NGM282 was diarrhea. One of the questions that have to be answered is whether high serum FGF19 levels can contribute to downstream signaling resistance. This will be crucial in determining whether FGF19 analogs can be used as a treatment option in ALD patients.

ASBT inhibitor

GSK2330672

GSK2330672 (GSK672) is an ASBT inhibitor. In a recent experiment [104], in a chronic-plus-binge ALD mouse model, GSK2330672 improved hepatocellular steatosis and oxidative stress by blocking the reabsorption of BAs in the gut. These observations not only suggest that inhibiting BAs recycling in the intestine is a potential treatment for ALD, but they also reveal novel mechanistic details about the relationship between BA dyshomeostasis and ALD. GSK672 treatment will, however, increase the amount of BAs in the large bowel, which may have damaging consequences for enterohepatic circulation and intestinal bacteria. Several other ASBT inhibitors are being studied in clinical research on cholestatic diseases [105]. In a NASH preclinical trial, ASBT inhibitors combined with FGF19 analogs minimized BA synthesis and reduced liver inflammation in mice [106]. In a phase II clinical trial of PBC [NCT01899703], the application of GSK2330672 effectively reduced the degree of pruritus [107]. Whether ASBT inhibitors have clinical benefits in patients with ALD remains to be determined.

PPAR agonist

Seladelpar (MBX-8025)

PPAR-α can promote the transport and oxidation of fatty acids. Ethanol inactivates the peroxisome proliferator activated receptor (PPAR), leading to reduced hepatic lipolysis [108]. On one hand, acetaldehyde suppresses the transcriptional activation and DNA binding activities of PPAR. Alcohol, on the other hand, lowers PPAR activity indirectly through oxidative stress mediated by CYP2E1. A recent study [109] demonstrated that seladelpar (MBX-8025), a selective PPAR-δ agonist, protects mice from hepatic injury caused by ethanol administration by recovering intestinal barrier function and BA balance. A phase II clinical trial of seladelpar in the treatment of PBC [NCT02955602] demonstrated improvement in cholestasis status and ALP level of patients [110]. However, in a previous clinical trial [NCT02609048], treatment of PBC with seladelpar was associated with an increased risk of abnormal increase in transaminases [111]. Further clinical trials are required to verify the role of PPAR agonists in the treatment of patients with ALD.

Intestinal bile acid sequestrant

Pectin

Pectin is a kind of fiber that can alter the intestinal microflora and alleviate ALD. Excessive ethanol consumption can result in ALD in humans. However, only a minority of alcoholics develop severe liver damage, indicating that, in addition to the amount of alcohol consumed, there are additional factors that influence the development and progression of ALD. In an experiment [112] to study the relationship between IM (intestinal microflora) and alcohol-induced liver injury in alcohol-sensitive (SENS) or alcohol-resistant (RES) mice, the researchers used two different methods to prevent alcohol-induced changes in IM. The first was pectin treatment, which can lead to significant changes in IM. The pectin comes from apple (Sigma, Saint Quentin Fallavier, France); the second was transplantation of fecal microflora of alcohol-resistant (RES) mice into alcohol-sensitive (SENS) mice. Both methods were found to prevent steatosis, and liver inflammation, and to restore the balance of the intestinal environment. This demonstrates the potential benefit of pectin in the treatment of ALD. A recent study demonstrated the relationship between pectin and BAs [113]. First, due to its physical features, pectin functions as a BA sequestrant because it can form a viscous matrix with cholesterol and BAs in the colon cavity [114], resulting in a reduction of cholesterol and BA levels in the body. In addition, pectin increases the excretion of BAs and alters the overall composition of the BA pool. It is also conducive to the increase of hydrophilic BAs (less toxic) in the liver, plasma, and intestine. This may be due to the abundance of intestinal bacteria containing genes encoding bile acid metabolic enzymes. Finally, patients with SAH have low levels of bacterial tryptophan. Pectin [115] can restore intestinal barrier function by increasing the production of tryptophan metabolites to attenuate alcohol-induced hepatic damage. In conclusion, ingestion of pectin and other fibers can improve alcohol-induced hepatic damage in mice by reshaping the gut microflora and modulating bacterial metabolites, including indoles and BAs. Because patients with ALD often consume minimal fiber, particularly pectin, dietary fiber supplementation as a prebiotic could be a prospective treatment approach for ALD management.

Conclusion

ALD is associated with a high morbidity and mortality worldwide. Firstly, alcohol consumption can alter the composition and size of BA pool by inducing abnormal changes in genes engaged in BAs metabolism, such as up-regulating BAs synthesis genes, changing BAs metabolic enzymes, and regulating BA transporters. Second, alcohol consumption can disrupt BA homeostasis by inducing dysbiosis of intestinal flora involved in BA metabolism. Finally, alcohol consumption causes abnormal alteration of BAs by regulating BA receptors. FXR agonists can alleviate liver steatosis and oxidative stress caused by chronic alcohol consumption, while systemic FXR gene knockout aggravates alcohol-induced hepatic damage. Available evidence suggests a key role of FXR in the occurrence and progression of ALD. Consumption of alcohol suppresses the activity of FXR, which affects the development of ALD by aggravating alcohol-related steatosis, causing mucosal damage, and increasing intestinal permeability, leading to endotoxin translocation and aggravation of liver inflammation and fibrosis. Alcohol abuse will lead to abnormal changes in the size and composition of the BA pool in serum, duodenum, and feces, and the changes in the BAs are markedly distinct in various stages of ALD and other chronic liver diseases. BA signaling may serve as a potential indicator for early diagnosis and prediction of the course of ALD in the foreseeable future, and also in the differential diagnosis of clinical diseases. The current treatment strategy for clinical ALD mainly involves abstinence from alcohol, glucocorticoids, and nutritional support. Recently, FXR agonists, FGF15 analogs, ASBT inhibitors, PPAR agonists, and intestinal BA sequestrants have been shown to alleviate liver injury caused by chronic alcohol consumption. Further clinical studies are required to investigate whether combined or tissue-specific administration can improve the treatment efficacy and reduce treatment-related adverse events in humans. Both BAs and alcohol recycle to the hepatocytes through the portal vein and affect the liver. Firstly, the independent and dual effects of BA dyshomeostasis and alcohol intake on the liver are not clear. More basic and clinical experiments are required to further explore the potential mechanism. Nevertheless, considering the considerable disparities in the makeup of the human and mouse BA profiles, recently, a novel mouse model with human-like BA pools was established [116]. Because Cyp2c70 can convert CDCA to MCA in mice but not in humans, these CYP2c70-deficient mice may be a more appropriate model for exploring BA function in ALD. Secondly, future research should further explore the impact of different types of alcoholic beverages and drinking patterns on the BAs. Finally, the limited patient sample size in the clinical studies does not allow further stratification of patient groups by race, sex, and BMI. Future studies need to evaluate the relationship of demographic characteristics with BA alteration in patients with ALD. Moreover, longitudinal studies are required to accurately evaluate the changes in BAs during the development or regression of ALD. Identification of novel serological biomarkers and therapeutic options for ALD are key research imperatives. Future research should investigate the mechanism between BA metabolism and ALD. This could help us better understand the regulatory role of BAs in the pathogenesis of ALD and the possibility of targeting BA metabolism and signaling as a therapy for ALD.

Abbreviations

ABCB4

ATP-binding cassette subfamily B member 4

ACLF

Acute-on-chronic liver failure

ACLI

Acute-on-chronic liver injury

AD

Acute decompensation

ADH

Alcohol dehydrogenase

AFLD

Alcoholic fatty liver disease

AH

Alcoholic hepatitis

ALD

Alcohol-related liver disease

ALDH

Acetaldehyde dehydrogenase

ALP

Alkaline phosphatase

ALT

Alanine aminotransferase

ASBT

Apical sodium-dependent BA transporter

ASCA

Anti-Saccharomyces cerevisiae antibodies

ASH

Alcoholic steatohepatitis

AST

Aspartate aminotransferase

AUDs

Alcohol use disorders

BA

Bile acids

BAAT

Bile acid-CoA: amino acid N-acyltransferase

BACS

Bile acid-CoA synthetase

BSEP

Bile salt export pump

BSH

Bile salt hydrolase

CA

Cholic acid

CB1R

Cannabinoid receptor type 1

CDCA

Chenodeoxycholic acid

COX-2

Cyclooxygenase-2

CREBH

CAMP response element-binding protein hepatocyte-specific

CYP2E1

Cytochrome P450 2E1 enzyme

CYP7A1

Cholesterol 7α-hydroxylase

CYP8B1

Sterol 12α-hydroxylase

DAMPs

Damage-associated molecular patterns

DCA

Deoxycholic acid

ERK1/2

Extracellular regulated protein kinases 1/2

FGF15/19

Fibroblast growth factor 15/19

FGF21

Fibroblast growth factor 21

FGR4

Fibroblast growth factor receptor 4

FIB-4

Fibrosis4Score

FOXP3

Forkhead box protein 3

FXR

Farnesoid X receptor

GCA

Glycocholic acid

GCDCA

Glycochenodeoxycholic acid

GDCA

Glycodeoxycholic acid

GGT

Gamma-glutamyl-transferase

GLCA

Glycolithocholic acid

GLP1

Glucagon-like peptide-1

HCA

Hyocholic acid

HCC

Hepatocellular carcinoma

HSC

Hepatic stellate cells

IM

Intestinal microbiota

LCA

Lithocholic acid

LRH-1

Liver receptor homolog-1

LTA

Lipoteichoic acid

MAFLD

Metabolic-associated fatty liver disease

MCD

Methionine-choline-deficient diet

MCRS1

Microspherule protein 1

MELD

Model for end-stage liver disease

MEOS

Microsomal ethanol oxidation system

MRP3/4

Multidrug resistance-associated protein 3/4

NAFL

Non-alcoholic fatty liver

NASH

Non-alcoholic steatohepatitis

NF-κB

Nuclear transcription factor-κB

NLRs

Nod-like receptors

NTCP

Sodium/taurocholate cotransporting polypeptide

OSTα/β

Organic solute transporter α/β

PAMPs

Pathogen-associated molecular patterns

PBC

Primary biliary cirrhosis

PDC

Pyruvate dehydrogenase complex

PDK4

Pyruvate dehydrogenase kinase 4

PGE2

Prostaglandin E2

PPAR

Peroxisome proliferator-activated receptor

PSC

Primary sclerosing cholangitis

PTGER4

Prostaglandin E receptor 4

ROC

Receiver operating characteristic

RORγt

Retinoic acid receptor-related orphan receptor- γ t

ROS

Reactive oxygen species

RXRα

Retinoid X receptor-α

SASP

Senescence-associated secretory phenotype

SHP

Small heterodimer partner

SIRT1

Sirtuins 1

SIRT5

Sirtuins 5

TBA

Total BA

TCA

Taurocholic acid

TCDCA

Taurochenodeoxycholic acid

TDCA

Taurodeoxycholic acid

TGR5

Takeda G protein-coupled receptor 5

TLCA

Taurolithocholic acid

TLR4

Toll-like receptor 4

TLRs

Toll-like receptors

Treg

Regulatory T cell

UDCA

Ursodeoxycholic acid

β-MCA

β-Muricholic acid

Author contributions

All authors contributed to the study conception and design. YG designed the review outline; YG and YL drafted the manuscript; YG, YL, TL, and XZ revised and edited the paper. All the authors read and approved the final manuscript.

Funding

This work was sponsored by the National Natural Science Foundation of China (Grants nos. 81972265, 82170602), the National Natural Science Foundation of Jilin Province (20200201324JC), and the Project for Health Talents of Jilin Province (JLSWSRCZX 2021-079).

Availability of data and material

Not applicable.

Declarations

Conflict of interest

The authors declare no conflict of interest.

Consent for publication

All authors agreed on the publication of the current version of manuscript.

Ethical approval and consent to participate

Not applicable.