Bile acid metabolic disorders and their role in hepatic encephalopathy of cirrhosis

Abstract

Liver cirrhosis is a progressive chronic liver disease characterized by metabolic disorders. Among these, bile acid (BA) metabolic disturbance serves as a key pathological link in disease progression and as a core factor driving development of hepatic encephalopathy (HE). Dysbiosis of the gut microbiota and BA metabolic disturbance mutually reinforce each other, forming a vicious cycle. Abnormally accumulated toxic hydrophobic BAs can directly damage the integrity of the intestinal mucosal barrier, activate local intestinal inflammatory signaling pathways, and induce systemic inflammation. These abnormal BAs can disrupt the structure and function of the blood-brain barrier through multiple BA-receptor-mediated signaling pathways, induce neuroinflammation in the central nervous system, and synergize with hyperammonemia to promote accumulation of neurotoxins in the central nervous system, ultimately triggering HE. BA metabolic profiles of liver cirrhosis caused by different causes show characteristic changes, which are expected to provide potential biomarkers for etiological identification and individualized intervention. Targeted therapeutic strategies aimed at the BA metabolic pathway have shown promising application, offering new targets and directions for prevention and treatment of HE. This review summarizes the core pathological mechanisms of BA metabolic disturbance in the context of liver cirrhosis and its role in the development and progression of HE.

Key Words: Liver cirrhosis; Bile acids; Hepatic encephalopathy; Gut microbiota; Blood-brain barrier; Gut-liver-brain axis

Core Tip: Cirrhosis disrupts bile acid (BA) homeostasis, creating a vicious cycle with gut dysbiosis and barrier injury. Accumulated toxic BAs cross the blood-brain barrier and modulate neuroinflammation via farnesoid X receptor, Takeda G protein-coupled receptor 5, and Sphingosine-1-phosphate receptor 2 signaling, synergizing with hyperammonemia to drive hepatic encephalopathy. BA profiles vary by cirrhosis etiology, offering diagnostic and prognostic potential. Targeting BA pathways represents a promising therapeutic strategy for hepatic encephalopathy.

INTRODUCTION

Liver cirrhosis represents the end-stage of chronic liver disease progression worldwide. Its pathological changes are characterized by liver fibrosis and nodular regeneration, accompanied by severe disturbances in systemic metabolic networks. Bile acid (BA) metabolic imbalance is one of the core drivers of disease progression[1,2]. Under physiological conditions, BA synthesis, secretion, and enterohepatic circulation (EHC) are finely regulated by several receptors. In liver cirrhosis, however, structural destruction of the liver, impaired liver function, and alterations in hepatic hemodynamics collectively lead to BA homeostasis disruption[1,3]. This disruption exhibits significant organ-interaction features. The levels of protective BAs decrease while toxic BAs accumulate, damaging the intestinal barrier and activating inflammatory signals to promote systemic inflammation[4]. Abnormally accumulated BAs can directly act on the central nervous system (CNS), destroy the blood-brain barrier (BBB), activate neuroinflammation, and have a synergistic effect with hyperammonemia, thereby aggravating the accumulation of neurotoxins in the brain[3,5].

BAs are amphipathic molecules derived from cholesterol. They are mainly synthesized in hepatocytes through two different pathways[6,7]. The classical pathway is started by the rate-limiting enzyme cholesterol 7α-hydroxylase (CYP7A1). It accounts for about 75% of total BA production in humans. This pathway produces primary BAs like cholic acid (CA) and chenodeoxycholic acid (CDCA). The alternative pathway is started by mitochondrial sterol 27-hydroxylase (CYP27A1), which mainly produces CDCA[6,7]. These primary BAs are conjugated with glycine or taurine in peroxisomes. This enhances their water solubility for bile secretion[7]. After release into the intestine, BAs help with lipid emulsification and absorption[6]. About 95% of BAs are reabsorbed in the terminal ileum and return to the liver through the portal vein circulation[7]. The remaining BAs go through microbial transformation in the colon. This produces secondary BAs like deoxycholic acid (DCA) and lithocholic acid (LCA)[6,8].

Hepatic encephalopathy (HE) is the most severe neuropsychiatric complication of liver cirrhosis, and has traditionally been attributed to mechanisms centered on ammonia toxicity, systemic inflammation, and oxidative stress[9,10]. Recent studies have revealed that BA metabolic disturbance is an important driving factor in HE development. Serum total BA levels are significantly elevated in patients with liver cirrhosis, and particularly in those with concomitant HE, they are positively correlated with blood ammonia levels[3]. Specific BA subtypes show distinct associations with the clinical phenotype and prognosis of HE.

The core mechanism by which BA metabolic disturbance drives HE lies in its systematic disruption of the gut-liver-brain axis. The gut microbiota participates in BA conversion through bile salt hydrolase (BSH) and 7α-dehydroxylase, while BAs in turn regulate the composition of the microbiota[11,12]. In liver cirrhosis, portal hypertension leads to intestinal barrier injury and dysbiosis, significantly altering the BA metabolic profile[13]. At the brain level, accumulated BAs can cross the BBB and enter brain tissue, activating relevant BA receptors to regulate microglial activation and mediate neuroinflammatory responses, thereby causing cognitive impairment.

Against this background, we focus on the gut-liver-brain axis to systematically outline the key processes by which BA metabolic disturbance, together with ammonia, inflammation, and other factors, collectively drives the development of HE in cirrhosis (Figure 1). This review elaborates the underlying pathological mechanisms and aims to provide new theoretical foundations and therapeutic targets for the precise prevention and treatment of HE[14,15].

Figure 1 Schematic diagram of the mechanism of hepatic encephalopathy induced by gut-liver-brain axis imbalance in cirrhosis. In healthy individuals, the liver maintains intestinal homeostasis via normal enterohepatic circulation of bile acids (BAs). In cirrhosis, destruction of liver structure accompanied by portosystemic shunting, together with disordered BA metabolism, intestinal mucosal edema and gut microbiota dysbiosis, leads to the entry of gut-derived toxins such as ammonia, lipopolysaccharide and toxic BAs into the systemic circulation. These toxins cross the blood-brain barrier, inducing neuroinflammation and neurotransmitter imbalance, and ultimately resulting in hepatic encephalopathy. CCL2: Chemokine ligand 2; GABA: Γ-Aminobutyric acid; IL: Interleukin; iNOS: Inducible nitric oxide synthase; LPS: Lipopolysaccharide; MMP-9: Matrix metalloproteinase-9; ROS: Reactive oxygen species; TNF-α: Tumor necrosis factor-α. Created in BioRender.

LIVER CIRRHOSIS CAUSES BA METABOLISM DISORDERS

EHC and the liver-gut regulation of BA metabolism

BAs are synthesized in the liver from cholesterol through a series of enzymatic reactions, primarily via two pathways. One is the classical pathway, also known as the neutral pathway. This pathway is initiated and rate-limited by CYP7A1 located in the endoplasmic reticulum. Sterol 12α-hydroxylase (CYP8B1) controls the ratio of CA to CDCA production. Mitochondrial CYP27A1 processes the side chain to complete the synthesis of primary BAs. In mice, CDCA can be further converted to α-muricholic acid and β-muricholic acid. Subsequently, primary BAs are conjugated with taurine or glycine in peroxisomes by BA-CoA synthetase and BA-CoA: Amino acid N-acyltransferase[6,7]. This conjugation reduces toxicity, increases amphiphilicity, and prevents premature reabsorption in the intestine. The classic pathway accounts for about 90% of BA synthesis and determines the hepatic BA pool size. The other pathway is the alternative pathway, also known as the acidic pathway. Due to the lack of branching enzymes, CDCA is synthesized mainly through CYP27A1 and oxysterol 7α-hydroxylase. In patients with liver disease and in neonates, the classical pathway may have impaired function or immature development, and the alternative pathway is important[16,17]. Brain neurons can convert cholesterol to 24-hydroxycholesterol via cytochrome P450 46A1, which can participate in the local synthesis of BA in the brain through the alternative pathway[18,19].

Under physiological conditions, most BAs exist in conjugated form. Due to their physicochemical properties, BAs cannot freely penetrate cell membranes and thus rely on specific transporters[20]. BAs synthesized in hepatocytes are secreted into the bile canaliculi in an ATP-dependent manner by the bile salt export pump (BSEP), which is primarily responsible for transporting conjugated BAs[21]. BAs are stored in the gallbladder with bile and are released into the intestinal tract after meals. Some primary BAs undergo bacterial 7α-dehydroxylation to form secondary BAs, such as DCA and LCA[22]. In the terminal ileum, BAs are actively taken up by enterocytes via the apical sodium-dependent BA transporter (ASBT). After binding to the ileal BA-binding protein inside the cell[23,24], BAs exit across the basolateral membrane into the portal circulation, mediated by the organic solute transporter α/β[23]. The BAs are then efficiently taken up from the portal blood by hepatocytes via the sodium taurocholate cotransporting polypeptide (NTCP) on the sinusoidal membrane and are resecreted into bile[25]. This entire process is known as the EHC of BAs. The EHC cycles 4-10 times per day, with approximately 95% of the BAs being reabsorbed and returned to the liver. Only 0.2-0.6 g per day are excreted in the feces, and this loss is replenished by de novo synthesis in the liver[7].

The synthesis and metabolism of BAs are precisely regulated by a negative feedback loop centered on the farnesoid X receptor (FXR). When the level of BAs in hepatocytes is increased, BAs act as endogenous ligands to activate FXR. Activated FXR then upregulates the expression of the small heterodimer partner (SHP)[26]. The SHP protein directly binds to and inhibits the activity of two key transcription factors: Liver receptor homolog-1 and hepatocyte nuclear factor (HNF)-4α[27,28], thereby blocking the transcriptional activation of CYP7A1 gene. Since CYP7A1 catalyzes the rate-limiting step in BA synthesis, its repression reduces de novo BA production. This mechanism is central to maintaining BA homeostasis.

Another negative feedback regulation of BA synthesis is primarily mediated by the gut-liver axis. BAs absorbed in the distal ileum activate FXR in intestinal epithelial cells, which in turn induces the expression and secretion of fibroblast growth factor (FGF) 19, which is known FGF15 in mice[29,30]. Subsequently, FGF15/19 reaches the liver via the portal vein and binds to the βKlotho-FGF receptor (FGFR) 4 complex on the cell membrane in hepatocytes. This binding activates the downstream mitogen-activated protein kinase signaling pathway[31], ultimately inhibiting CYP7A1 gene expression and reducing de novo BA synthesis. It should be noted that there are differences in the regulatory hierarchy between humans and mice. Mice primarily rely on the endocrine pathway involving intestinal FXR and FGF15 to act on the liver. In humans, besides the gut-liver axis, BAs can also activate FXR within hepatocytes themselves. This hepatic FXR activation induces the expression of liver-derived FGF19. FGF19 then acts locally via autocrine or paracrine mechanisms to directly inhibit CYP7A1, forming an independent and rapid feedback loop within the liver[29].

However, these two negative feedback pathways do not operate independently but function cooperatively. After hepatocyte-specific knockout of the cytoplasmic Shp2 gene, FGF15/19-FGFR4 signaling and BA-FXR-SHP pathway activation are impaired[32]. Activated FGFR4 drives downstream kinases to phosphorylate the SHP protein. This phosphorylation modification enhances the stability of the SHP protein and finely tunes its binding ability to the transcription factors liver receptor homolog-1 and HNF-4α. Ultimately, modified SHP can more effectively repress the promoter activity of the CYP7A1 gene, resulting in synergistic inhibition of BA synthesis (Figure 2)[33,34].

Figure 2 Bile acid synthesis, transformation, reabsorption, and negative feedback regulation. In the liver, cholesterol undergoes a series of enzymatic reactions to produce primary bile acids (BAs), which then enter the intestine and are converted into secondary BAs. Both primary and secondary BAs can conjugate with glycine or taurine. Most BAs are eventually reabsorbed into the bloodstream, with only a small portion excreted in feces. This process is negatively regulated by small heterodimer partner and intestine-derived fibroblast growth factor 15/19. FGFR4: Fibroblast growth factor receptor 4; GCA: Glycocholic acid; GCDCA: Glycochenodeoxycholic acid; BAs: Bile acids; CYP7A1: Cholesterol 7α-hydroxylase; CYP8B1: Sterol 12α-hydroxylase; CYP27A1: Sterol 27-hydroxylase; CYP7B1: Oxysterol 7α-hydroxylase; CA: Cholic acid; CDCA: Chenodeoxycholic acid; BACS: Bile acid-CoA synthetase; BAAT: Bile acid-CoA: Amino acid N-acyltransferase; BSEP: Bile salt export pump; T: Taurine; G: Glycine; BSH: Bile salt hydrolase; HSD: Hydroxysteroid dehydrogenase; DCA: Deoxycholic acid; LCA: Lithocholic acid; ASBT: Apical sodium-dependent bile acid transporter; IBABP: Ileal bile acid-binding protein; OSTα/β: Organic solute transporter α/β; NTCP: Sodium taurocholate cotransporting polypeptide; FXR: Farnesoid X receptor; SHP: Small heterodimer partner; LRH-1: Liver receptor homolog-1; HNF-4α: Hepatocyte nuclear factor 4 α; FGF15/19: Fibroblast growth factor 15/19. The double upward arrows (↑↑) indicate elevated levels or increased expression. Created in BioRender.

The vicious cycle of liver cirrhosis and BAs: From pathological alterations to disease progression

Liver fibrosis is the pathological repair response of the liver to chronic injury. It is characterized by fibrous connective tissue proliferation in the portal area and excessive deposition of diffuse extracellular matrix. If the disease progresses, it can develop into diffuse liver fibrosis with regenerative nodules of hepatocytes; a condition known as liver cirrhosis[35]. During this process, activated hepatic stellate cells (HSCs) and portal fibroblasts synthesize large amounts of type I and III fibrillar collagens. These collagens replace the normal basement membrane collagens in the liver, leading to increased liver stiffness and impaired hepatocyte function[36]. HNF-4α is a key transcription factor regulating the BA conjugation pathway. Dysregulation of HNF-4α expression or function disrupts BA metabolic homeostasis[37]. In liver cirrhosis, the prolonged exposure of hepatocytes to stiff matrix suppresses HNF-4α expression. This in turn downregulates the expression of BA-CoA: Amino acid N-acyltransferase, a critical enzyme in BA conjugation. Consequently, BA synthesis and conjugation functions are severely impaired[38].

During progression of cirrhosis, serum total BA concentrations significantly increase, and alterations in BA composition pattern accelerate the cirrhotic process[39]. Secondary BAs produced by intestinal microbiota metabolism are key pathogenic factors driving progression of liver fibrosis. Both conjugated and unconjugated forms of secondary BAs can induce activation of HSCs. Unconjugated BAs aggravate inflammation by activating the tumor necrosis factor/nuclear factor κB (NF-κB) signaling pathway. This promotes the transformation of HSCs into myofibroblasts, which express α-smooth muscle actin and secrete large amounts of collagen and factors like interleukin (IL)-6, leading to abnormal deposition of extracellular matrix and progression of liver fibrosis[40]. Conjugated 12α-hydroxylated BAs, such as taurodeoxycholic acid and glycodeoxycholate, are more water-soluble and stable, and can effectively bind to the membrane receptor Takeda G protein-coupled receptor (TGR) 5. This binding leads to sustained activation of the extracellular signal-regulated kinase 1/2 and p38 mitogen-activated protein kinase signaling pathways, thereby stimulating HSCs proliferation and exacerbating the fibrotic process[41].

During progression of liver cirrhosis, dysregulation of BA metabolism is a key factor driving disease progression and extrahepatic vascular dysfunction. As the ability of the liver to clear BAs declines, serum levels of toxic hydrophobic BAs increase, which are positively correlated with the severity of liver disease and portal pressure. In particular, these levels increase further with the onset of complications such as acute decompensation, acute-on-chronic liver failure, infection, or ascites, which may indicate clinical deterioration[42,43].

In liver cirrhosis, accumulated BAs can affect vascular function through multiple mechanisms. Hydrophobic BAs like CDCA and DCA can interfere with calcium channel function and activate potassium channels in vascular smooth muscle cells, thereby weakening vasoconstrictive responses[44]. Acting as endogenous agonists for the FXR and TGR5, they can upregulate expression of endothelial nitric oxide synthase and promote the release of nitric oxide from endothelial cells. The above effects together lead to systemic vasodilation and hyperdynamic circulation in patients with cirrhosis, increasing the risk of portal-hypertension-related decompensation[45]. It is worth noting that changes in BA composition may be used as potential biomarkers for the evaluation of portal hypertension. It has been shown that the serum concentration of primary conjugated BAs, such as taurochenodeoxycholic acid (TCDCA), is significantly and positively correlated with the hepatic venous pressure gradient. TCDCA has excellent predictive ability for clinically significant portal hypertension (defined as hepatic venous pressure gradient ≥ 10 mmHg), and its area under the receiver operating characteristic curve exceeds 0.95. This provides a new clinical approach for the noninvasive assessment of portal hypertension[46].

Gut microbiota dysbiosis in cirrhosis alters the composition of BAs

The liver produces BAs and secretes them into the intestine through bile. The gut microbiota can mediate the transformation and metabolism of BAs. Concurrently, the content and type of BAs also affect the composition and abundance of the gut microbiota. This bidirectional relationship is regulated by EHC of BAs. In this cycle, BAs are repeatedly transformed and utilized, which helps with nutrient absorption and act as important signaling molecules connecting the gut-liver and microbiota-host axes[16].

The precise regulation of BAs by the gut microbiota relies on a series of specific bacterial enzymes encoded by the microbiota. These enzymes work in concert to determine the composition of the BA pool. The reaction catalyzed by BSH is a key rate-limiting step that initiates subsequent transformations. This enzyme widely exists in various gut bacteria such as Bacteroidetes and Firmicutes. It catalyzes the deconjugation of conjugated BAs to generate free primary BAs[47]. Subsequently, enzymes like 7α-dehydroxylase, encoded by bacteria such as Clostridium, further convert free primary BAs into secondary BAs[8]. Additionally, the hydroxysteroid dehydrogenase (HSD) family is responsible for various oxidation, reduction, and epimerization reactions[48]. These subtle modifications can dynamically regulate the hydrophobicity, toxicity, and signaling capability of BAs as ligands for host receptors.

Portal hypertension caused by liver cirrhosis can lead to intestinal mucosal congestion and edema, which subsequently damages intestinal barrier function and alters the local microenvironment, ultimately resulting in significant gut microbiota dysbiosis. It is characterized by small intestinal bacterial overgrowth. Studies show that the incidence of small intestinal bacterial overgrowth in cirrhotic patients is significantly higher than that in healthy populations and is closely related to severity of liver disease[49]. Specifically, there is a marked reduction in gut microbiota diversity, with beneficial bacteria such as Bacteroidetes and Lactobacillus decreasing, while conditional pathogens and oral-origin bacteria like Streptococcus and Veillonella become abnormally enriched[50,51]. This disruption in the gut microbial profile directly leads to a significant accumulation of total BAs and primary BAs in the serum of cirrhotic patients, while the levels of total BAs and secondary BAs in the feces are notably decreased[52].

After entering the intestine, the primary BAs synthesized by the liver are converted to secondary BAs via 7α-dehydroxylation. The concentration of these secondary BAs, particularly DCA, is critical. At physiological concentrations, DCA exerts beneficial antibacterial effects by inhibiting the overgrowth of pathogenic bacteria such as Enterobacteriaceae[53]. However, when the level of DCA is abnormally elevated, it disrupts the intestinal barrier, induces leaky gut, and promotes endotoxin translocation into the bloodstream[54]. In patients with advanced cirrhosis, this balance is disrupted. As the total amount of BAs entering the intestine is reduced, DCA levels decrease, which weakens its natural inhibitory effect on pathogenic bacteria[53]. Studies show that in patients with hepatitis B-related cirrhosis, the abundance of Clostridium, which is involved in BA transformation, is significantly reduced. This reduction weakens the inhibitory effect on the ileal FXR-FGF19 signaling axis, leading to a feedback-driven overproduction of primary BAs by the liver via both the classical and alternative pathways. This subsequently exacerbates cholestasis and hepatocyte damage[55]. Insufficient production of secondary BAs in the gut weakens antimicrobial defense, while the compensatory overproduction of primary BAs creates a vicious cycle that worsens liver injury.

Alterations in BA profiles in liver cirrhosis of different etiologies

The main causes of cirrhosis include metabolic dysfunction, chronic hepatitis B or C virus infection, long-term alcohol abuse, and autoimmune abnormalities. These pathogenic factors act on the liver over a long period and are key drivers for the continuous progression of liver fibrosis to cirrhosis. BA metabolism disorders are commonly observed in cirrhosis from various causes, characterized by elevated serum total BA levels, imbalance in the composition of individual BAs[56], and alterations in the fecal BA profile[57]. Against the backdrop of liver inflammation and fibrosis, cirrhosis resulting from different etiologies exhibits both shared metabolic remodeling and certain differences in the specific patterns of BA profile changes, reflecting etiology-specific metabolic characteristics (Table 1)[58-84].

Table 1 Serum bile acid profiles and key metabolic alterations in liver diseases of different etiologies.

Etiology	Serum BA characteristics	Key metabolic enzyme/transporter alterations
MASLD	Total BAs elevated with fibrosis progression; conjugated primary BAs increased[77]. Cytotoxic DCA, LCA, and CDCA elevated, associated with advanced fibrosis[59,60,78]	CYP7A1 upregulated, FGF19 decreased, NTCP upregulated, BSEP unchanged[58]. CYP8B1 downregulated, HSD3B7 unchanged[61]
HBV-related	Total BAs elevated; increased ratio of glycine/taurine-conjugated BAs[65]. Lithocholate-3-O-glucuronide decreased in early stage[64]. GCA, GCDCA, TCA, TCDCA, and TUDCA progressively elevated (TCDCA most significant)[66,79]	HBV preS1 competitively binds to NTCP; NTCP downregulated in severe liver injury, reduced FXR activation relieves CYP7A1 inhibition[62,63,80]
HCV-related	Total BAs elevated, independently correlated with fibrosis severity[67]. TCA, GCA, GUDCA, TCDCA, and GCDCA elevated[68]	HSD3B7 and CYP8B1 characteristically downregulated, CA synthesis reduced; restored post-treatment with compensatory CYP27A1 upregulation[61,81]
Alcoholic-related	Total BAs significantly elevated[72]. Increased secondary BAs proportion with abnormal conjugated secondary BAs in duodenum[69,73]	CGH upregulation enhances deconjugation, disrupting gut-liver axis and driving CYP7A1 overexpression[70]. FXR inactivation via acetylation[71]. FGF19 elevated, correlated with MELD score[82]
PBC	Total BAs significantly elevated, TCA and GCA markedly elevated[71,74,83]. Taurine-conjugated BAs proportion significantly elevated[83]. GCDCA and TDCA levels elevated with Child-Pugh grade, correlating with disease severity[84]	Reduced BAs entering intestine, impaired BSH activity[74,75]. Weakened intestinal FXR-FGF19 negative feedback, CYP7A1 overexpression[76]

MASLD: Metabolic dysfunction-associated steatotic liver disease; BAs: Bile acids; DCA: Deoxycholic acid; LCA: Lithocholic acid; CDCA: Chenodeoxycholic acid; CYP7A1: Cholesterol 7α-hydroxylase; FGF19: Fibroblast growth factor 19; NTCP: Sodium taurocholate cotransporting polypeptide; BSEP: Bile salt export pump; CYP8B1: Sterol 12α-hydroxylase; HSD3B7: 3β-hydroxysteroid dehydrogenase type 7; HBV: Hepatitis B virus; GCA: Glycocholic acid; GCDCA: Glycochenodeoxycholic acid; TCA: Taurocholic acid; TCDCA: Taurochenodeoxycholic acid; TUDCA: Tauroursodeoxycholic acid; FXR: Farnesoid X receptor; HCV: Hepatitis C virus; GUDCA: Glycoursodeoxycholic acid; CGH: Choloylglycine hydrolase; MELD: Model for end-stage liver disease; PBC: Primary biliary cholangitis; BSH: Bile salt hydrolase; TDCA: Taurodeoxycholic acid; CYP27A1: Sterol 27-hydroxylase.

Metabolic dysfunction-associated steatotic liver disease

Metabolic dysfunction-associated steatotic liver disease (MASLD), formerly known as nonalcoholic fatty liver disease, is a metabolic condition characterized by excessive fat accumulation in the liver. It encompasses nonalcoholic fatty liver and metabolic dysfunction associated steatohepatitis, and has become the most common chronic liver disease worldwide[85,86].

In patients with MASLD, expression of the hepatic rate-limiting BA synthesis enzyme CYP7A1 is significantly upregulated, whereas the serum level of the negative feedback regulator FGF19 is decreased. This may be related to the more significant increase of the FXR antagonist DCA compared to the agonist CDCA in the intestine. Concurrently, the expression of the BA uptake transporter NTCP in the liver is increased, but the function of the BA export pump BSEP is not correspondingly enhanced. This uptake-export imbalance contributes to elevated peripheral blood BA concentrations[58]. Research indicates that in patients with MASLD, BA metabolism dysregulation occurs as early as the initial stages of liver fibrosis. It is specifically manifested by a significant rise in levels of particular BAs such as glycochenodeoxycholic acid (GCDCA) and 7-ketodeoxycholic acid, and these levels progressively increase with the advancement of fibrosis[87]. When MASLD progresses to liver fibrosis and cirrhosis, the BA profiles in serum and feces show distinct alterations. In serum, hydrophobic and cytotoxic BAs from the CDCA family, their conjugates, as well as LCA and DCA are significantly elevated[59]. A longitudinal study confirmed that when patients with MASLD-related cirrhosis experienced decompensation events, the serum levels of DCA and its conjugates, particularly taurodeoxycholic acid, markedly increased. These dynamic changes in BA levels are not only biomarkers for assessing disease severity but also potential key drivers of disease progression[60]. Fecal total BA concentration and the activity of BSH, reflecting gut microbiota deconjugation function, did not change significantly[59]. However, the proportion of secondary BAs increased relatively. This may be associated with downregulated expression of CYP8B1 while the expression of 3β-HSD type 7 was not downregulated[61]. Consequently, in the late stages of MASLD-related cirrhosis, the dysregulation of serum BA metabolism is more likely to originate from severe dysfunction in the liver’s own synthesis, conjugation, and transport functions, rather than from the action of the gut microbiota.

Hepatitis-B-related cirrhosis

In chronic hepatitis-B-related cirrhosis, as liver fibrosis progresses, expression of the BA transporter NTCP on the hepatocyte membrane is downregulated[62]. This may lead to reduced hepatic uptake of BA, resulting in elevated serum BA levels and aggravation of liver injury and fibrosis progression. NTCP is also a functional receptor for hepatitis B virus (HBV) entry into hepatocytes. The preS1 domain on the HBV particle competes with endogenous BA for binding to NTCP, leading to decreased intracellular BA levels. This in turn reduces FXR activation and SHP expression, relieving the inhibition of cholesterol CYP7A1 and promoting de novo BA synthesis[63].

Huang et al[64] found that in patients with early liver inflammation and fibrotic stages of chronic hepatitis B, the serum level of lithocholate-3-O-glucuronide was significantly decreased. This metabolite is the glucuronidated product of LCA and has cytotoxic effects. Its early decrease may reflect that the liver has already developed metabolic dysfunction at this stage. As the disease progresses, the serum BA profile undergoes significant changes. The proportion of glycine- and taurine-conjugated BAs increases, while the proportion of unconjugated BAs decreases. The elevated conjugated BAs are mainly primary BAs such as GCDCA and TCDCA[65]. Among them, taurocholic acid (TCA) is noteworthy. Its level can distinguish liver disease patients from healthy individuals and predict the severity of liver disease[66].

Hepatitis-C-related cirrhosis

In patients with chronic hepatitis C virus (HCV) infection, serum total BA levels are significantly higher in those with cirrhosis than in those with mild fibrosis, and can serve as an independent predictor of liver fibrosis severity[67]. The ABCB11 gene 1331T>C variant is associated with elevated serum BA levels and progression of cirrhosis, suggesting that reduced BSEP function leads to impaired BA excretion and cholestasis, thereby promoting fibrosis[88]. In terms of BA synthesis, HCV-related cirrhosis also exhibits characteristic changes. The expression of 3β-HSD type 7 and CYP8B1, a key enzyme determining CA synthesis, is significantly downregulated in hepatocytes, leading to a severe reduction in CA synthesis. This specific co-downregulation pattern is an important feature that distinguishes it from MASLD[61].

Similar to hepatitis B cirrhosis, TCA is also the most significantly changed BA in hepatitis C cirrhosis. It can accelerate the process of cirrhosis by activating HSCs and aggravating inflammation[89]. Khalil et al[68] found that serum conjugated BAs, such as TCA, glycocholic acid (GCA), GUDCA, TCDCA, and GCDCA, were significantly increased in the serum of patients with HCV cirrhosis through targeted metabolomics analysis. These BAs could effectively distinguish cirrhotic patients from healthy or noncirrhotic individuals, with an area under the receiver operating characteristic curve ranging from 0.85 to 0.96[68]. Serum total BA level serves as a marker of disease severity, and its reduction after effective anti-HCV treatment can also reflect the degree of liver histological improvement more sensitively than conventional liver function tests[90].

Alcoholic cirrhosis

In alcoholic cirrhosis, BA metabolism and regulatory pathways are significantly disrupted. Compared with patients who abstain from alcohol, those with cirrhosis who continue to consume alcohol have elevated levels of total BA and an increased proportion of secondary BAs, and abnormal conjugated secondary BAs appear in the duodenum. In compensation, the ileum upregulates BA receptor expression, but this still cannot prevent the continuous accumulation of toxic BAs[69]. Studies in mouse models of alcoholic liver disease have found that alcohol-related gut microbiota dysbiosis increases the abundance of bacterial genes encoding choloylglycine hydrolase, which enhances BA deconjugation and leads to an increased proportion of unconjugated BAs and a corresponding decrease in conjugated BAs in the ileum[70]. As conjugated BAs are potent agonists of FXR, their reduction inhibits intestinal FXR activity, which in turn decreases FGF15 secretion and relieves the inhibition of CYP7A1, ultimately driving excessive BA synthesis in the liver[70]. Alcohol can inactivate FXR by inhibiting sirtuin 1 and activating acetyltransferase p300, leading to FXR hyperacetylation. Together with microbiota dysbiosis, this creates a vicious cycle of uncontrolled BA synthesis and an expanded total BA pool[71].

The failure of this dual regulatory mechanism makes patients with alcohol-related cirrhosis, similar to those with primary biliary cholangitis (PBC) (formerly known as primary biliary cirrhosis), exhibit significantly higher serum total BA levels compared to patients with cirrhosis with other etiologies[72]. It has been shown that alcohol consumption significantly alters the fecal BA profile, leading to overproduction of total and secondary BAs in feces accompanied by the proliferation of specific proinflammatory flora such as Veillonellaceae. These changes increase intestinal permeability and aggravate alcohol-induced liver injury[73].

PBC

PBC is an autoimmune cholestatic liver disease characterized by nonsuppurative destruction of intrahepatic bile ducts and cholestasis[91]. This leads to a significant increase in serum total BA levels of patients, while the fecal total BAs concentration is not significantly different from that of healthy individuals[72]. However, the composition of fecal BAs is markedly altered: The proportion of primary BAs in feces increases whereas that of secondary BAs decreases, with conjugated BAs predominating[74]. These changes mainly result from reduced bile flow, which decreases the total amount of BAs entering the intestine, and from gut microbiota dysbiosis, particularly a low abundance of Clostridium. BSH activity may also be impaired, leading to reduced deconjugation of conjugated BAs[74,75]. More importantly, gut microbiota dysbiosis is not merely a concomitant phenomenon but a key factor driving the progression of PBC. Due to the reduction in natural FXR agonists such as secondary BAs, the intestinal FXR-FGF19 negative feedback axis is weakened. This attenuates inhibition of hepatic CYP7A1, leading to excessive BA synthesis in the liver and exacerbation of liver injury[76].

In PBC, BA metabolism disorders caused by gut microbiota dysbiosis can affect the CNS and are closely related to the early cognitive impairment seen in patients. This cognitive impairment is not related to the severity of liver disease, and its pathogenesis may be different from that of typical HE. Although neuropsychological test results are often normal, the pattern of white matter lesions shown on brain magnetic resonance imaging is distinctly different from that in HE, suggesting that its etiology is more likely related to autonomic dysfunction, hypotension, and subsequent structural brain changes associated with PBC[92]. A study by Hennenberg et al[93] using a bile duct ligation-induced rat model of biliary cirrhosis found that BAs can inhibit the activation of RhoA in vascular smooth muscle and downregulate Rho-kinase protein expression by inducing oxidative stress and inflammation, leading to vasodilation and hypotension. In addition, cholestasis and impaired liver function associated with PBC can reduce the ability of the liver to synthesize neuroactive steroids, leading to an enhanced inhibitory effect of γ-aminobutyric acid (GABA) in the brain, thereby exacerbating cognitive dysfunction at the neurochemical level[94].

BA METABOLISM DISORDERS AND HE

Epidemiological evidence of BA levels and HE

HE is a neuropsychiatric syndrome caused by severe acute or chronic liver dysfunction or portosystemic shunting. Its clinical manifestations can range from cognitive deficits to altered consciousness and impaired neuromuscular function[95]. The incidence of HE is closely related to the type and severity of liver disease. Epidemiological data show that approximately 40% of patients with acute liver failure and up to 50% of patients with chronic liver disease can develop HE[96]. The pathogenesis of HE has long centered on the synergistic effects of ammonia toxicity, systemic inflammation, and oxidative stress. Recent studies indicate that BAs play an important role in HE. In addition to their physiological functions, in the setting of cirrhosis, BAs can participate in the development and progression of HE through various pathways, including altering BBB permeability, inducing neuroinflammation, and modulating specific receptors expression[97].

In patients with cirrhosis, serum total BA levels are significantly elevated, especially in those with HE, and are positively correlated with blood ammonia concentrations[3]. In-depth analysis of the BA profile revealed specific associations between different BA subtypes and the type and clinical outcome of HE. Alterations in the serum BA profile are an independent risk factor for the development of covert or overt HE and even death in patients with cirrhosis. Specifically, high serum levels of conjugated primary BAs, particularly TCDCA, are strong independent predictors of overt HE and all-cause mortality, with a predictive ability not influenced by traditional factors such as the model for end-stage liver disease score or blood ammonia. In contrast, high levels of unconjugated primary BAs are independently associated with covert HE[39]. Prospective metabolomics studies have expanded the clinical value of BAs as predictive biomarkers for HE. Dantas Machado et al[98] performed paired pre-operative and postoperative untargeted metabolomics analysis of plasma from 22 patients with cirrhosis who underwent transjugular intrahepatic portosystemic shunt (TIPS). They found that the abundance of three conjugated di-hydroxylated and tri-hydroxylated BAs (m/z 450.3, 432.3, and 414.3) in peripheral blood after surgery was negatively correlated with HE severity within 1 year after the procedure. Patients with pre-existing intrahepatic shunts had smaller metabolomic changes after TIPS and a lower risk of HE. This suggests that specific BA profiles can be used for postoperative HE risk stratification[98]. Bajaj et al[99] studied a multicenter cohort of 602 hospitalized patients with cirrhosis. They found that higher serum levels of 7α-hydroxy-3-oxo-4-cholestenoate on admission were associated with development of advanced HE[99]. These studies extend the role of BA testing from a simple diagnostic marker to a tool for prognosis prediction. This provides a new strategy for early identification of high-risk patients in clinical practice.

In addition to peripheral blood, BAs in the CNS are also significantly altered. BAs such as GCA and TCA are significantly enriched in the cerebrospinal fluid of patients with HE, and their concentrations are positively correlated with the severity of liver disease. The cerebrospinal fluid BA profile is a specific diagnostic marker that distinguishes patients with HE from those with cirrhosis alone, and its diagnostic value is superior to changes in plasma BAs[100]. Animal studies have also confirmed that in both acute and chronic liver failure models, BA levels in the serum and cerebral cortex increase in parallel, and the degree of this increase is consistent with the severity of neurobehavioral impairment. Therefore, BA metabolic disorder is a key pathogenic factor driving the development of HE, rather than a simple epiphenomenon of liver failure[3,101].

A precise retrieval was performed in the Web of Science Core Collection based on the MeSH terms “bile acids” and “hepatic encephalopathy”, with a total of 326 relevant articles screened out. On this basis, we conducted bibliometric statistics on the trends of annual and cumulative publication quantity from 2010 to 2025 (Figure 3A), and carried out keyword co-occurrence analysis on these 326 literatures to explore the research hotspots and thematic correlations in this field (Figure 3B).

Figure 3 Bibliometric analysis and keyword co-occurrence analysis. A: Bibliometric analysis of annual and cumulative publications on bile acid and hepatic encephalopathy based on MeSH terms in Web of Science (2010-2025); B: Keyword co-occurrence analysis of “bile acid” and “hepatic encephalopathy” based on VOSviewer Software.

BA-mediated alterations in BBB permeability

The BBB is a highly selective physiological barrier composed of endothelial cells, tight junction proteins, pericytes, and astrocytes. Its core function is to protect the CNS by regulating the permeability of substances[102,103]. BAs have received considerable attention as a class of endogenous molecules that can cross and potentially modulate this barrier. Under physiological conditions, hydrophobic unconjugated BAs primarily cross the BBB via passive diffusion due to their lipophilicity. In contrast, hydrophilic conjugated BAs rely on active transport mediated by specific transporters, including organic anion transporting polypeptides, organic anion transporters, and the ASBT[104]. In addition to their own transport, some BAs can be used as components of liposomes to significantly enhance the efficiency of drug delivery across the BBB. For example, liposomes modified with GCA improve drug penetration primarily through the transcellular pathway, while those modified with TCA are ineffective[105]. The effect of BAs on BBB permeability differs by brain region. In the choroid plexus, the expression of key transporters such as NTCP, BSEP, and ileal BA transporter is low. Therefore, under physiological conditions, BAs cannot easily enter the CNS through active transport mechanisms at this site[106].

BAs with different properties have distinct regulatory effects on the integrity and function of the BBB. Hydrophobic BAs, such as CDCA and DCA, can activate the small GTPase Rac1, leading to phosphorylation of tight junction proteins, disruption of intercellular tight junctions, and impairment of BBB integrity, thereby facilitating their own entry into the brain[107]. DCA directly induces apoptosis in BBB endothelial cells and pericytes and activates inflammatory signaling pathways, exacerbating BBB damage[108]. In contrast, hydrophilic BAs, such as UDCA and GUDCA, exhibit protective effects. They significantly inhibit unconjugated bilirubin-induced apoptosis in human brain microvascular endothelial cells and effectively maintain BBB integrity through their anti-inflammatory and cell-stabilizing properties[109].

BA-receptor-mediated regulation of neuroinflammation via the gut-liver-brain axis

BAs participate in the absorption of lipids and act as signaling molecules. They regulate various pathophysiological processes, including glucose metabolic homeostasis, apoptosis, and inflammatory responses, by acting on nuclear receptors such as FXR and pregnane X receptor (PXR), as well as membrane receptors such as TGR5 and sphingosine-1-phosphate receptor (S1PR) 2[110-112]. In cirrhosis, the dynamic imbalance in the expression of these BA receptors regulates the process of neuroinflammation. This constitutes a key bridge in the gut-liver-brain axis that connects liver injury to brain pathology. We summarize the different roles of BA receptors in the gut-liver-brain axis (Figure 4).

Figure 4 Bile acid receptor signaling in the gut-liver-brain axis. Bile acids act as signaling molecules to regulate the gut-liver-brain axis through nuclear receptors (farnesoid X receptor and pregnane X receptor) and membrane receptors (Takeda G protein-coupled receptor 5 and sphingosine-1-phosphate receptor 2). Farnesoid X receptor and pregnane X receptor maintain Bile acid homeostasis, enhance intestinal and blood-brain barrier (BBB) integrity, and suppress inflammation. Takeda G protein-coupled receptor 5 exerts anti-neuroinflammatory effects by inhibiting microglial activation and NACHT, LRR and PYD domains-containing protein 3 inflammasome. In contrast, sphingosine-1-phosphate receptor 2 disrupts the BBB, promotes chemokine ligand 2 release, and triggers neuroinflammation. These receptors jointly mediate intestinal barrier damage, liver fibrosis, BBB dysfunction, and central neuroinflammation, thereby driving the pathogenesis of hepatic encephalopathy. BA: Bile acid; S1PR2: Sphingosine-1-phosphate receptor 2; TGR5: Takeda G protein-coupled receptor 5; PXR: Pregnane X receptor; FXR: Farnesoid X receptor; SHP: Small heterodimer partner; BBB: Blood-brain barrier; CCL2: Chemokine ligand 2; P-gp: P-glycoprotein; IL-1β: Interleukin-1β; NF-κB: Nuclear factor κB; iNOS: Inducible nitric oxide synthase; COX2: Cyclooxygenase-2; TNF-α: Tumor necrosis factor-α; NLRP3: NACHT, LRR and PYD domains-containing protein 3; AP-1: Activator protein-1. Created in BioRender.

FXR

FXR is a member of the nuclear receptor superfamily, primarily distributed in the liver, intestine, kidneys, adrenal glands, and adipose tissue, and widely involved in regulating BA, glucose, and lipid metabolism and homeostasis[113]. As a core regulator of BA metabolism, FXR activation inhibits BA synthesis, enhances intestinal barrier function, and improves dysbiosis[114]. Recent studies have found that the anti-inflammatory effect of FXR is closely related to its regulation of the NACHT, LRR and PYD domains-containing protein 3 (NLRP3) inflammasome. The NLRP3 inflammasome is a key molecular complex in innate immunity, widely expressed in myeloid cells, epithelial cells, and the CNS[115]. At the molecular level, after FXR activation, it directly binds to NLRP3. While promoting NLRP3 transcription, FXR inhibits its phosphorylation at the Ser295 site. This unique dual regulation ultimately blocks the activation of the NLRP3 inflammasome, reduces the release of inflammatory cytokines such as IL-1β, and alleviates BA-induced liver fibrosis[2].

FXR plays a distinct role in the CNS compared to its function in the liver. In animal models of liver failure, BAs that cross the BBB are transported into neurons via the ASBT. This activates the FXR signaling pathway and induces the expression of the downstream SHP, ultimately worsening neurological decline[116,117]. Specific blockade of brain FXR significantly delays this neurological decline. Research shows that in the azoxymethane-induced acute HE model, activation of brain FXR suppresses cytochrome P450 46A1; a key enzyme for cholesterol clearance. This leads to abnormal cholesterol accumulation in neurons, disrupting neurotransmitter release and membrane receptor signaling, and ultimately aggravating HE[118]. However, this FXR-dependent neuropathological mechanism is not universally present. In the thioacetamide-induced HE model, although plasma BA levels increase, they do not cause BA accumulation in the brain or activation of the FXR signaling pathway[119]. Therefore, the role of the FXR signaling pathway in HE may depend on the type and severity of liver injury and permeability of the BBB.

TGR5

TGR5 is a G protein-coupled plasma membrane BA receptor. It is widely distributed in the liver, intestine, and immune cells, and is also expressed in the nervous system, such as in astrocytes and neurons, where it performs important functions[5].

During different stages of HE, expression and function of TGR5 show dynamic changes. In the azoxymethane-induced HE model, neuronal chemokine ligand (CCL) 2 is significantly increased. This leads to microglia activation and promotes neuroinflammation[120]. Treatment with a TGR5 agonist can inhibit the release of neuronal CCL2. This reduces the interaction between CCL2 and the microglial chemokine receptor 2 and chemokine receptor 4. Consequently, microglial activation is suppressed. Direct activation of TGR5 on microglia also reduces their phagocytic activity and proinflammatory cytokine production. These effects work together to exert anti-neuroinflammatory actions[14,120]. However, in the HE pathological environment, continuous activation by ammonia or neurosteroids causes TGR5 receptor downregulation and functional exhaustion. This is considered a receptor desensitization mechanism. It may be a self-protective response of the body against overactivation[5].

In the bile duct ligation-induced HE model, activation of the NLRP3 inflammasome promotes the maturation and release of proinflammatory cytokines such as IL-1β, induces hippocampal inflammation and microglial death, and exacerbates neurological damage[121]. Inhibiting NLRP3 activation reduces inflammatory responses and glial cell activation and helps restore BBB function and neuronal structure during HE[122]. Studies in models such as ischemic brain injury have confirmed that TGR5 activation suppresses NLRP3 inflammasome assembly and activation via the cAMP/PKA signaling pathway, thereby reducing IL-1β release, inhibiting M1 polarization of microglia, and exerting neuroprotective effects[123]. Although TGR5 activation has been demonstrated to confer neuroprotection by blocking NLRP3-dependent inflammation in various neurological disease models, it remains to be confirmed whether this mechanism holds true in HE.

Sphingosine-1-phosphate receptor 2

Sphingosine-1-phosphate receptor 2 (S1PR2) is a G-protein-coupled receptor widely expressed throughout the body, particularly in the immune, cardiovascular, and CNSs, as well as in organs such as the intestine and liver[112]. In the intestine, S1PR2 has a dual role. Under physiological conditions, it can maintain the intestinal barrier function by promoting the proliferation and migration of epithelial cells. However, in the inflammatory state, it can promote the polarization of macrophages to the proinflammatory M1 phenotype, aggravate the inflammatory response, and promote immune cell infiltration[124,125]. In the liver, this proinflammatory effect is particularly prominent. In cholestatic liver injury, accumulation of TCA activates S1PR2 on hepatocytes, which in turn triggers the NLRP3 inflammasome signaling axis. This leads to the release of proinflammatory factors and induces the formation of neutrophil extracellular traps, ultimately worsening liver injury. Inhibition of S1PR2 can alleviate this process[126].

Similar destructive effects are also observed in the brain. Research shows that after brain injury, S1PR2 expression is significantly upregulated in cerebral endothelial cells. Its activation disrupts BBB integrity, while inhibiting S1PR2 has a protective effect[127]. S1PR2 can mediate extension of liver disease pathology to the brain. In an HE model, elevated circulating BAs, such as TCA, cross the BBB and activate S1PR2 on neurons. This activation induces the production of chemokines like CCL2, which subsequently activates microglia and triggers neuroinflammation[15]. Therefore, existing evidence indicates that S1PR2, as a key BA sensor, plays complex roles in the gut, liver, and brain. Its abnormal activation demonstrates significant damaging effects in various diseases. However, the complete mechanism by which S1PR2 acts as a core molecule to systematically integrate gut-liver-brain axis signals and ultimately drive diseases like HE requires further clarification.

PXR

PXR is a nuclear receptor that plays important roles in the liver, intestine, kidney, brain, and some immune cells[128]. Its core function is to regulate the detoxification, synthesis, and transport of BAs[129]. In the intestine, activation of PXR can alleviate inflammation by inhibiting the NF-κB signaling pathway and maintain intestinal barrier integrity[130]. In the liver, PXR is particularly critical for detoxifying hydrophobic secondary BAs such as LCA. As a natural ligand of PXR, LCA can inhibit CYP7A1 expression to reduce BA synthesis and induce organic anion transporting polypeptide 2 expression to promote its uptake. At the same time, PXR can also upregulate the metabolic enzyme cytochrome P450 3A11 to accelerate the hydroxylation detoxification of LCA, thereby alleviating liver injury[131]. PXR is also expressed at the BBB. Its activation can upregulate expression and function of P-glycoprotein on endothelial cells, enhancing the protective effect of the BBB[132]. PXR deficiency can lead to anxiety-like behavior and cognitive memory impairment in mice. This effect is mainly related to its regulation of neurosteroid metabolism and electroencephalographic activity, rather than changes in BBB permeability, suggesting that it has a more subtle function in the brain[133]. In summary, PXR plays a protective role at key points of the gut-liver-brain axis. However, its specific role in HE models has not yet been clarified, and whether it can serve as a potential therapeutic target remains to be proven.

Bidirectional regulation of BA receptors and blood ammonia in HE

Hyperammonemia has long been considered the core pathogenic factor in the occurrence and development of HE. Elevated blood ammonia is the initiating factor of HE, and the subsequent neuroinflammation acts as an accelerator for disease progression. Together, they form a synergistically amplified pathogenic network by disrupting the BBB, inducing glial cell pathology, and causing neuronal damage.

As the core receptor of BA, FXR plays a key role in regulating ammonia metabolism. Activation of FXR upregulates the gene expression of key enzymes in the hepatic urea cycle and induces glutamine synthetase. This converts toxic ammonia into nontoxic glutamine for storage, effectively reducing blood ammonia levels and alleviating liver injury[134,135]. FXR activation reduces blood ammonia at the source and directly acts on the CNS. It alleviates astrocyte swelling and downregulates inflammatory factors such as IL-1β and oxidative stress levels, directly inhibiting neuroinflammation[136].

In HE, ammonia acts as a neurotoxin and exhibits an anti-inflammatory effect mediated by astrocytes. In cocultures of astrocytes and microglia, ammonia significantly suppresses the lipopolysaccharide-induced upregulation of proinflammatory cytokines such as tumor necrosis factor-α and IL-6. This effect does not occur in microglia monocultures[137]. The mechanism involves neurosteroids produced by astrocytes under ammonia stimulation, which activate the TGR5 receptor on microglia, thereby triggering anti-inflammatory signaling pathways[5,137]. This finding also explains why activated microglia in the brains of HE patients often display an anti-inflammatory M2 phenotype[5,137].

The clinical manifestations of HE are complex and varied. Beyond cognitive impairment, late-stage patients often present with motor coordination deficits such as ataxia and bradykinesia, which severely reduce quality of life. These motor symptoms are primarily attributed to neuroinflammation within the cerebellum and abnormal enhancement of GABAergic neurotransmission[138]. S1PR2 has been identified as a key molecular node linking hyperammonemia to the neurological symptoms of HE. Under hyperammonemic conditions, expression of S1PR2 is upregulated in cerebellar Purkinje neurons and hippocampal neurons[138,139]. Its activation triggers a cascade of downstream events: Neuronal production of chemokines like CCL2 activates microglia, leading to substantial release of brain-derived neurotrophic factor. Brain-derived neurotrophic factor then activates the tropomyosin receptor kinase B receptor on Purkinje neurons. This increases the membrane expression of the K⁺-Cl⁻ cotransporter 2 in neurons. This enhances GABAergic neurotransmission, resulting in motor incoordination and cognitive impairment. It also promotes NF-κB nuclear translocation via the phosphoinositide 3-kinase–protein kinase B pathway, exacerbating neuroinflammation[138-140]. Blocking S1PR2 reverses this entire neuroinflammatory pathway and restores cognitive function. This suggests that S1PR2 is a crucial molecular link between hyperammonemia and neurological symptoms of HE, and provides a promising new therapeutic target for the disease.

RESEARCH PROSPECTS OF BA IN HE

The development and progression of HE are not driven by a single factor but represent a complex network resulting from the synergistic effects of various metabolic disturbances. In addition to the classic ammonia toxicity, mechanisms such as inflammatory responses, oxidative stress, accumulation of false neurotransmitters, and amino acid imbalances are interwoven and mutually amplified. BA metabolism disorders and dysregulation of the gut-liver-brain axis mediated by BAs are increasingly recognized as key for these mechanisms.

At present, the treatment of HE mainly relies on ammonia-lowering drugs, but their efficacy is limited and the recurrence rate is high. In recent years, multidimensional intervention strategies targeting BA metabolism have provided new directions for the treatment of HE. They focus on the regulation of BA-related receptors, flora intervention, and BA replacement therapy. BAs can mediate bidirectional pathological regulation of HE through three core pathways. FXR agonists obeticholic acid and tropifexor can enhance the ability of liver to clear ammonia and improve nerve function damage[135,136]. The selective TGR5 agonist betulinic acid can protect the CNS through anti-inflammatory pathways[14]. The TCA-S1PR2 axis can drive early neuroinflammation in HE, and blocking this pathway can independently improve neurological function[15]. Traditional Chinese medicine, fecal microbiota transplant, and gut-targeted antibiotics can regulate BA metabolism by remodeling intestinal flora. Rhubarb decoction can reshape intestinal flora, increase the levels of conjugated protective BAs (TCDCA, TUDCA, etc.), and indirectly regulate the FXR pathway to improve HE[141]. Fecal microbiota transplant can remodel BA metabolism, restore secondary BA synthesis, repair intestinal barrier, effectively improve cognitive function in patients with HE, reduce recurrence, and evaluate the efficacy of transplantation. However, its clinical promotion is limited by issues such as donor standardization and infection risk[142,143]. Rifamycin SV MMX, a new colon-targeted antibiotic, can reshape the BA detoxification spectrum and improve the gut-liver-brain axis and physical function[144]. Clinical trials of UDCA in the preventive treatment of HE after TIPS are ongoing (NCT07273734), which is expected to expand the clinical application of BA replacement therapy. However, at present, all kinds of targeting strategies have shortcomings in clinical transformation and the efficacy of drugs needs to be verified.

Future research on the role of BAs in HE should be more in-depth and systematic. It should establish a more comprehensive database of serum and fecal BA profiles. The research focus should shift from simple correlation analysis to revealing the specific mechanisms of BAs as core signaling molecules in the gut-liver-brain axis. It is necessary to clarify the precise effects of different BA species on BBB permeability, astrocyte function, neuroinflammation, and the neurotransmitter system in HE caused by cirrhosis of various etiologies. This will systematically elucidate the intrinsic mechanisms by which BA metabolic disorders drive the occurrence and development of HE. Interventions targeting BA metabolic pathways have become a research hotspot in therapeutics. The development of specific probiotics or drugs that can target the regulation of BA metabolism of intestinal flora and thus improve HE is an important research direction in the future.

CONCLUSION

BA metabolic disorder induced by liver cirrhosis is a consequence of liver injury and an important mechanism driving the onset and progression of HE. The imbalance of BAs in the liver and intestine leads to gut microbiota dysbiosis, intestinal barrier damage, and inflammation, collectively increasing the influx of neurotoxins such as blood ammonia into the bloodstream and brain, thereby inducing neuroinflammation. Therefore, it is essential to tailor combination therapies including drugs, diet, and microbiota transplantation for individual patients based on their gut microbiota characteristics, liver function status, and BA profiles. Alleviating neuropsychiatric symptoms may slow the progression of liver disease, improve long-term quality of life, and open new strategies and approaches for the prevention and treatment of HE.