Liver-nervous system axis: pathways, dysregulation, and translational perspectives

Abstract

The liver and the nervous system form a dynamic bidirectional regulatory network through the liver-nervous system axis, which is crucial for maintaining metabolic homeostasis, immune balance, and the integrity of neural function in the body. However, the specific regulatory pathways between the liver and the nervous system, as well as their mechanisms in the development and progression of diseases, remain incompletely understood. Moreover, the dynamic regulatory mechanisms of the blood–brain barrier in the transport of liver metabolites are yet to be elucidated. Therefore, we have integrated the latest research advances in recent years in the fields of liver innervation, the neuroregulatory effects of liver-derived substances, and the trans-organ conduction of inflammatory mediators. Additionally, we incorporate clinical translational evidence from the application of β-blockers, vagus nerve stimulation, liver transplantation, and other interventions to systematically elucidate the coordinated regulatory patterns of the liver and the nervous system under physiological conditions, as well as the mechanisms of interactive disorders under pathological conditions. The aim is to clarify the interactions between the liver and the nervous system and their significance in physiological and pathological processes, thus providing a theoretical basis and potential therapeutic targets for future related research.

Introduction

As the central hub for maintaining metabolic homeostasis in the body, the liver not only undertakes key physiological functions such as nutrient conversion, toxin elimination, and immune regulation but also forms a highly coordinated bidirectional regulatory network with the nervous system, known as the liver-nervous system axis [1, 2]. The functional integrity of this axis is crucial for sustaining metabolic balance, immune homeostasis, and normal neurological function; conversely, its interactive dysfunction is closely associated with the onset and progression of various major diseases [3, 4]. However, the specific regulatory pathways between liver and nervous system, as well as their mechanisms in the development and progression of diseases, remain incompletely understood, and there is a lack of targeted therapeutic strategies.

Anatomically, the liver is intricately innervated by the sympathetic nerves, parasympathetic nerves (vagus nerve), and sensory nerves. These nerve fibers penetrate deep into the liver through the porta hepatis, directly regulating metabolic, inflammatory, fibrogenic, and secretory activities of key cells such as hepatocytes, hepatic stellate cells (HSCs), and Kupffer cells (KCs), liver sinusoidal endothelial cells (LSECs), and cholangiocytes [2, 4, 5]. Specifically, sympathetic stimulation regulates hepatocyte glycogenolysis and gluconeogenesis via norepinephrine (NE), while parasympathetic input modulates glycogen synthesis via Acetylcholine (ACh) [6]. In HSCs, sympathetic neurotransmitters (NE, serotonin) promote activation and fibrogenesis, whereas ACh reduces HSC activation [7]. Regarding KCs, catecholamines induce ADRB2-mediated apoptosis via the GDF15 pathway, limiting excessive inflammation in alcohol-associated liver disease [8]. For LSECs, foundational studies have established that nerve endings in the Disse spaces directly contact these cells, providing an anatomical basis for neural regulation of sinusoidal microcirculation [9]. A recent comprehensive review by Trucas et al. further elaborated on this neurovascular interface, highlighting how neural inputs influence hepatic blood flow, sinusoidal endothelial function, and portal hypertension [10]. For cholangiocytes, sympathetic nerves form direct synaptic contacts on these cells, particularly on small intrahepatic bile ducts, and β-adrenergic receptor activation promotes ductular remodeling during chronic liver injury [11]. Furthermore, beyond direct cellular regulation, hepatic nerves play critical roles in broader physiological processes. For instance, Tanaka et al. demonstrated that sympathetic nerves promote hepatic lymphangiogenesis by stimulating Schwann cells to secrete vascular endothelial growth factor-C (VEGF-C), and this process is protective against liver fibrosis [12]. Collectively, these findings establish that the nervous system exerts broad regulatory control over diverse liver cell populations and pathophysiological processes, with significant implications for understanding liver pathophysiology and developing neuromodulatory therapeutic strategies.

At the molecular level, the blood–brain barrier (BBB) serves as a crucial structure for material exchange between blood and central nervous tissue, precisely regulating the transmembrane transport of hepatic metabolites, hormones, and inflammatory factors through specific transport systems [1, 2]. For instance, fibroblast growth factor 21 (FGF21), synthesized in the liver, can cross the BBB to act on hypothalamus. Additionally, neurotoxins such as ammonia and manganese, which accumulate during liver dysfunction, can enter the central nervous system (CNS) through the BBB [13, 14]. Furthermore, emerging mediators, including extracellular vesicles, utilize the BBB to facilitate information exchange between the liver and the CNS [15].

Under pathological conditions, the bidirectional regulation balance of the liver-nervous system axis is often disrupted. Excessive activation of the sympathetic nervous system (SNS) can promote the activation of HSCs through α1-adrenergic receptors (α1-AR) [16]. Mechanistically, α-adrenergic signaling triggers downstream Ca²⁺ mobilization, ERK1/2 and PI3K/Akt phosphorylation cascades, which further upregulate the expression of transforming growth factor β (TGFβ) and enhance the transcriptional activity of profibrotic genes, thereby stimulating collagen deposition and α-smooth muscle actin (α-SMA) expression to accelerate liver fibrosis [16–19]. Inflammatory factors such as TNF-α released by the liver can damage the BBB, inducing CNS inflammation and insulin resistance [20, 21]. Conversely, the hyperactivity of the hypothalamic–pituitary–adrenal (HPA) axis caused by chronic stress, through the secretion of excessive cortisol, promotes lipid synthesis in hepatocytes [22]. The loss of cholinergic neurons in the dorsal motor nucleus of the vagus (DMV) nerve in patients with Parkinson's disease (PD) weakens hepatic anti-inflammatory signaling and impairs the liver’s capacity to restrict pro-inflammatory cytokine production and limit excessive immune cell infiltration, thereby aggravating intrahepatic inflammation [23, 24].

Despite intensive study of the liver-nervous system axis in recent years, several critical scientific questions remain unresolved. The precise roles of neurotransmitters, including NE and ACh, in liver regeneration and fibrosis are still not fully understood. Additionally, the molecular targets and signaling pathways associated with hepatogenic substances require further elucidation. The disparity between preclinical research and clinical application remains considerable. Current interventions, such as vagus nerve stimulation and β-blockers, exhibit insufficient specificity, long-term safety, and personalized efficacy.

Therefore, this review systematically integrates the anatomical foundations, molecular mechanisms, and bidirectional regulatory patterns of the liver-nervous system axis. It emphasizes the pathological significance of dysregulation within this axis in the context of metabolic and chronic liver disease (CLD), while summarizing therapeutic strategies aimed at modulating liver-nervous system communication pathways. By synthesizing existing research advancements, this review aims to establish a theoretical framework for understanding the interactions between the liver and the nervous system. Additionally, it provides new targets and insights for the collaborative prevention and treatment of complex diseases, including metabolic liver diseases and neurodegenerative disorders.

Anatomy of the liver and nervous system

Innervation of the liver: distribution of the sympathetic nerves and vagus nerve

The liver forms close connections with the CNS and peripheral nervous system (PNS) through a complex innervation network [25]. This neural innervation is not only an anatomical connection but also directly regulates the core physiological functions of the liver, such as metabolism, inflammation, and repair [25]. Anatomical studies have demonstrated that the liver is mainly innervated by the autonomic nerves, including sympathetic and parasympathetic nerves (vagus nerve), as well as sensory nerves [2, 4]. The distribution and function of different nerve types constitute a coordinated regulatory network [2, 4].

The preganglionic fibers of the sympathetic nerves innervating the liver originate from the 7th to the 12th thoracic vertebra (T7 ~ T12) of the intermediolateral column (IML) of the spinal cord, while the postganglionic fibers arise from the celiac ganglion and the superior mesenteric ganglion [26]. NE released by the sympathetic nerves can directly regulate intrahepatic vasoconstriction, gluconeogenesis, and lipid metabolism [2], and also participate in the regulation of HSCs activation, which is closely related to liver dysfunction [27].

The parasympathetic nerves of the liver originates from cholinergic neurons (ChAT +) in the DMV in the dorsal brainstem [5, 28]. And they regulate the anti-inflammatory function of KCs through α−7 nicotinic ACh receptors (α7nAChR), to maintain liver homeostasis [29–31].

The sensory nerves, as the ascending pathway of the liver-nervous system axis, can sense glucose concentration, osmotic pressure and inflammatory state within the liver in real time. They transmit transmit signals to the CNS through the nucleus tractus solitarius (NTS), to provide a basis for integrating metabolic signals and regulating feeding and energy expenditure in hypothalamus and other brain regions [32].

These nerve fibers enter the liver through the porta hepatis, forming anterior and posterior hepatic plexuses along the portal vein, hepatic artery, and bile ducts, which further extend into the hepatic lobules to directly innervate hepatocytes, HSCs, and LSECs [25, 27, 33, 34]. Sympathetic and parasympathetic nerve fibers terminate in the space of Disse.Sympathetic nerve fibers release NE, which binds to α1-AR and β2-adrenergic receptors (β2-AR) on hepatocytes, HSCs, and LSEC). NE binding on HSCs activates Gq-PLC-IP3/DAG-Ca2+-MAPK signaling, promoting HSCs activation, α-SMA expression, and collagen deposition. Parasympathetic nerve fibers release ACh, which binds to α7nAChR on Kupffer cells, suppressing pro-inflammatory cytokine (TNF-α, IL-6) release via the cholinergic anti-inflammatory pathway. LSECs and hepatocytes also receive direct cholinergic inputs that regulate vascular tone, permeability, and bile secretion (Fig. 1).

Fig. 1.

Anatomical and cellular basis of liver innervation. A Anatomical localization of the liver and nervous system in the human body. B Origin and distribution of sympathetic and parasympathetic nerves innervating the liver. C Gross anatomy of the hepatic plexus and intrahepatic nerve fiber distribution. D High-magnification schematic of nerve-liver cell interactions

Recent studies using immunofluorescence labeling combined with tissue-clearing three-dimensional (3D) imaging technology have confirmed significant species differences in hepatic innervation. In the livers of humans, non-human primates, mice, and other higher mammals, sympathetic innervation predominates [20, 35]. Notably, in animals such as rats, hepatic nerve fibers are primarily confined to the portal vein area, whereas in humans, liver nerves can penetrate deeply into the hepatic lobules and terminate directly on hepatocytes [28]. Additionally, neurons in the liver can produce a variety of neurotransmitters, including NE, Ach, neuropeptide Y (NPY), vasoactive intestinal peptide (VIP), calcitonin gene-related peptide (CGRP), substance P (SP), glucagon-like peptide (GLP), somatostatin, neurotensin, and serotonin. Hepatocytes express corresponding receptors for these neurotransmitters, collectively forming a complex signaling regulatory network [36] (Table 1).

Table 1.

Innervation and functional characteristics of the liver

Type	Sympathetic nerves	Parasympathetic nerves (Vagus nerve)	Sensory nerves
Origin	Preganglionic fibers arise from the IML of the spinal cord at T7–T12 levels [37]	DMV in the medulla oblongata [28]	Nodose ganglion of the parasympathetic nerve, and thoracic dorsal root ganglia
Pathway	Celiac ganglion → porta hepatis → surrounding the hepatic artery, portal vein, bile ducts [25, 34]	Vagus nerve → porta hepatis → portal vein region [5, 28]	Porta hepatis → CNS
Direction	Efferent (CNS → liver)	Efferent (CNS → liver)	Afferent (liver → CNS)
Functions	Regulate vasoconstriction, gluconeogenesis, lipid metabolism [2], and promote HSCs activation, accelerate liver fibrosis [16–19]	Promote bile secretion, intrahepatic vasodilation, and regulate the contraction and relaxation of the gallbladder and bile ducts [5, 38]	Transmit pain sensation, sense changes in portal vein glucose, lipids, and osmotic pressure, and convey signals to the nucleus tractus solitarius (NTS)
Core innervated sites	Intrahepatic blood vessels, HSCs, hepatocytes, biliary epithelium [27]	Hepatocytes, KCs, biliary system, intrahepatic blood vessels [29–31]	Liver capsule, intrahepatic blood vessel walls, biliary system
Markers	NE [18]	ACh	SP, CGRP [36]
Receptors	β-adrenergic receptors [18], α-adrenergic receptors [16–19]	α7nAChR [29–31]	No (mainly signal transmission)

Beyond hepatocytes, other hepatic cell types also express a diverse array of neurotransmitter receptors. HSCs express α1-AR, particularly the α1A subtype, which is markedly upregulated during liver fibrosis. NE-mediated activation of α1-AR on HSCs triggers intracellular Ca²⁺ oscillations, nuclear factor kappa-B (NF-κB) activation, and secretion of pro-inflammatory chemokines (IL-8), promoting both portal hypertension and fibrogenesis [16–19, 39]. In contrast, KCs uniquely express α7nAChR, which mediate the cholinergic anti-inflammatory pathway. Activation of α7nAChR on KCs suppresses the release of TNF-α, IL-6, and HMGB-1, effectively limiting liver inflammation [29–31]. This cell-type-specific distribution of neurotransmitter receptors, which common receptors (β-adrenergic receptors, mAChRs) ensuring broad regulatory coverage and liver cell-type-specific receptors (α1-AR on HSCs, α7nAChR on KCs) enabling precise targeting, allows the autonomic nervous system to precisely regulate distinct liver functions, ranging from fibrogenesis and inflammation to metabolism and hemodynamics.

The BBB and transport of hepatic metabolites

Structure and function of BBB

The BBB is a dynamic interface composed of brain microvascular endothelial cells (BMECs), astrocyte endfeet, pericytes, and the basement membrane. Its tight junction proteins, such as claudin-5 and occludin, form a physical barrier that restricts the free entry of peripheral substances into the CNS [40]. The molecular selectivity of the BBB is co-regulated by transport proteins such as P-glycoprotein and GLUT1, and enzyme systems such as cytochrome P450 (CYP), which is predominantly expressed in brain microvascular endothelial cells (BMECs) and astrocytes [41–43], ensuring the passage of essential substances such as glucose and amino acids while preventing the invasion of toxins and pathogens [44]. Recent studies have further revealed that the functional state of the BBB is closely related to systemic metabolic levels, and its structural integrity is also significantly affected by factors such as neuroinflammation and oxidative stress [44].

Mechanism of hepatic metabolites crossing the BBB

Hepatic metabolites rely on specific molecular mechanisms to cross the BBB. For instance, 24S-hydroxycholesterol, generated by the neuronal enzyme CYP46A1, can passively diffuse through the BBB due to its lipophilicity and is subsequently metabolized and cleared by the liver [45]. Under pathological conditions such as metabolic dysfunction-associated steatotic liver disease (MASLD), liver dysfunction often leads to abnormal accumulation of toxic lipids like ceramides. These molecules can enter the brain via active transport or membrane diffusion across the BBB, where they inhibit insulin signaling pathways and trigger neuroinflammation [46]. In addition, the ABC transporter family on the BBB, such as ABCB1, can efflux liver-derived exogenous toxins such as neonicotinoid pesticides. However, long-term exposure may still result in their accumulation in the brain [47].

In summary, the BBB plays a crucial role as a dynamic interface and signaling hub in the liver-nervous system axis. It not only serves as a selective barrier for the transport of liver metabolites and signaling molecules to the CNS, but also acts as a defensive barrier to maintain the homeostasis of the brain microenvironment and block the invasion of hepatogenic toxic substances. The functional integrity of the BBB is key to maintaining bidirectional communication between the liver and the CNS, particularly the brain, and its dysfunction represents a core pathological link between liver diseases and neurological complications.

Impact of liver disease and neuroinflammation on BBB integrity

Liver disease and neuroinflammation compromise BBB integrity through various molecular pathways, with lipopolysaccharide (LPS) serving as a central trigger and amplifier.

In CLD, intestinal barrier dysfunction and impaired hepatic detoxification facilitate the entry of gut-derived LPS into the portal circulation, resulting in endotoxemia [48, 49]. Circulating LPS binds to Toll-like receptor 4 (TLR4), triggering downstream signaling cascades [50]. The engagement of LPS with TLR4 activates myeloid differentiation primary response 88 (MyD88)-dependent signaling, leading to the activation of NF-κB [51]. Activated NF-κB promotes the transcriptional upregulation of pro-inflammatory cytokines, including TNF-α, IL-1β, and IL-6 [51, 52]. These pro-inflammatory factors further activate matrix metalloproteinase-9 (MMP-9), an endopeptidase that degrades basement membrane components and tight junction proteins, directly contributing to BBB leakage [52]. Concurrently, NF-κB-mediated signaling downregulates and relocalizes tight junction proteins (claudin-5, occludin, ZO-1) via activation of Rho kinase, MMPs, and ubiquitin–proteasome pathways. Studies have demonstrated that IL-17 reduces occludin and ZO-1 expression by activating NADPH oxidase, increasing superoxide production and MLCK activity, thereby inducing BBB permeabilization [53]. Notably, LPS and hyperammonemia act synergistically in a "two-hit" mechanism. In acute liver failure, infection (LPS exposure) precipitates hepatic encephalopathy (HE) by amplifying inflammation in a BBB already compromised by ammonia toxicity [52]. Furthermore, hyperammonemia itself induces a transient increase in peripheral IL-17, which disrupts BBB integrity through the mechanisms described above [53].

In addition, neuroinflammation also disrupts BBB integrity through convergent molecular mechanisms [54]. Microglial activation, a hallmark of neuroinflammation, releases pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) and reactive oxygen species (ROS), which directly activate NF-κB signaling in brain endothelial cells, leading to downregulation and disorganization of tight junction proteins [55]. Simultaneously, activated microglia and infiltrating macrophages secrete matrix metalloproteinases (MMP-2 and MMP-9), which degrade basement membrane components (collagen-IV, laminin) and junctional complexes, further increasing paracellular permeability [54, 56]. Notably, elevated levels of LPS in the CNS directly activate microglial TLR4, perpetuating a self-amplifying cycle of cytokine release and oxidative stress that exacerbates BBB breakdown [57, 58]. Additionally, neuroinflammatory signals can trigger astrocytic end-foot swelling and release of vascular endothelial growth factor (VEGF), which increases transcytosis and fenestration of brain endothelial cells, further compromising barrier function [59–61].

In summary, LPS/TLR4/NF-κB signaling, cytokine/chemokine networks, MMP activation, and oxidative stress constitute a core molecular machinery that converges on the disruption of both tight junction integrity and basement membrane stability, ultimately leading to BBB leakage.

Effects of liver dysfunction on the nervous system

The liver, as the metabolic center of the body, when dysfunctional, not only disrupts systemic metabolic homeostasis but also interferes with the normal functions of the nervous system through multiple core pathways, including metabolic disorders, neurotoxin accumulation, hepatogenic substance regulation, inflammation mediation, and autonomic nerve dysregulation [62]. This explains why dysfunction of the CNS and PNS has become a common complication of CLD and liver failure [5].

On the one hand, the chemoreceptors in the porta hepatis can sense changes in metabolic substances, which are transmitted via the vagus nerve to the NTS and then to the hypothalamus, regulating feeding and energy metabolism [2]. On the other hand, the pro-inflammatory cytokines such as TNF-α and IL-6 released by the activation of hepatic KCs, can cross the BBB to activate microglia in the brain, inducing neuroinflammation and consequently impairing cognitive function and emotional regulation [4]. This suggests that the liver can regulate the nervous system through the the liver-nervous system axis, providing novel insights for the treatment of related complications (Table 2).

Table 2.

Regulatory effects of liver on nervous system function

Category	Key metabolites	Site/Conditions of production	Targets	Neurological Effects
Metabolic toxins	Ammonia (NH3) [1]	Liver failure → impaired urea cycle (hepatocytes, LSECs)	Glutamine synthetase in astrocytes	Inhibits LTP, disrupts BBB, causes cerebral edema and cognitive impairment
Heavy metals	Manganese (Mn) [63]	Impaired biliary excretion (hepatocytes, cholangiocytes)	Dopaminergic neurons in the basal ganglia	Deposition in globus pallidus → Parkinsonism-like symptoms, mood disorders
Bile acids	CDCA, DCA, UDCA [2]	Hepatic synthesis (hepatocytes) → 7α-dehydroxylation of intestinal bacteria (enterocytes)	FXR, TGR5	Inhibits Agrp neurons, activates POMC → reduces feeding, regulates GLP-1 secretion
Short-chain fatty acids (SCFAs)	Propionic acid, butyric acid [64]	Fermentation by intestinal microbiota (intestinal epithelial cells) → portal vein	Vagus nerve, MCT1	Promotes GLP-1, PYY secretion, inhibits mitochondria → neural energy crisis
Hepatogenic substances	FGF21, etc. [2]	Hepatic PPARα induction (hepatocytes)	FGFR1/β (Hypothalamic)	Inhibits HPA axis, reduces anxiety, ameliorates Aβ pathology
Aromatic amino acids (AAAs)	Phenylalanine, tyrosine [1]	Impaired liver function (hepatocytes, KCs)	LAT1 transporter in the BBB	Disrupts dopamine and 5-HT synthesis → psychiatric symptoms

Regulation of hepatic metabolites on nervous function

Studies have shown that liver-derived metabolites such as ammonia and bile acids can cross the BBB and exert damaging effects on the nervous system through multiple pathways, these include neuronal injury, activation of glial cells, alteration of neurotransmitter balance, and changes in synaptic plasticity [65, 66] (Fig. 2).

Fig. 2.

Regulation of hepatic metabolites on nervous function. Hepatic metabolites, such as ammonia, manganese, bilirubin, bile acids, regulate various aspects of nervous system function through direct and indirect pathways. Ammonia crosses the BBB, inducing glutamine accumulation and astrocytic edema, while disrupting neuronal excitation and neurotransmitter homeostasis. Manganese crosses the BBB, deposits in the basal ganglia, impairing dopaminergic neurotransmission, and additionally inducing oxidative stress, exerting neurotoxic effects. Bilirubin compromises BBB integrity and promotes neuroinflammation. It binds to 5-HT3 receptors and activates GABAergic neural pathways, ultimately altering pain thresholds. Bile acids, upon activation of TGR5 receptors, stimulate the release of spinal neuropeptides and GLP-1 secretion from intestinal L cells, and also act directly on hypothalamic neurons via crossing the BBB. Excessive SCFAs can ultimately lead to neuroinflammation by inhibiting mitochondrial function, promoting GLP-1/PYY secretion. AAAs competitively cross the BBB through the LAT1 transporter, producing pseudo-neurotransmitters that ultimately lead to psychiatric symptoms

Ammonia toxicity

Ammonia, as an important product of liver metabolism, elevates in blood levels when liver function is impaired. Excess ammonia crosses the BBB into the CNS, inducing nerve damage through dual mechanisms. One is directly interfering with the glutamine synthetase activity of astrocytes, causing glutamine accumulation and resulting in neurological dysfunction [14, 67]. The other is directly inhibiting the excitability of neurons, disrupting the balance of neurotransmitters, and triggering neurological symptoms such as cognitive impairment and confusion [68].

Importantly, the neurotoxic effects of ammonia are not uniformly distributed across the brain. The hippocampus is one of the most sensitive regions, where ammonia impairs astrocyte function, disrupts the glutamate cycle, and reduces brain-derived neurotrophic factor (BDNF), leading to deficits in synaptic plasticity that manifest as memory impairment and anxiety-like behaviors [69]. The cerebellum is another highly sensitive area, in which ammonia uniquely activates NMDA receptors and the nitric oxide-cyclic guanosine monophosphate (NO-cGMP) pathway, resulting in vasogenic edema and neuronal injury [70–72]. The cerebral cortex exhibits disturbed ion balance under ammonia toxicity, with astrocyte shrinkage impairing potassium (K⁺) clearance and causing neuronal hyperexcitability [73]. The basal ganglia, particularly the globus pallidus and caudate putamen, display vasogenic edema and exceptional metabolic inhibition, contributing to movement disorders and postural instability [74]. This region-selective vulnerability elucidates the diverse clinical manifestations of HE, including memory deficits, motor dysfunction, and cognitive impairment.

Manganese deposition

Excessive manganese deposition is also a crucial pathological mechanism of neurotoxicity induced by liver diseases [63]. In cases of severe liver insufficiency or portosystemic shunts, manganese cannot be effectively excreted through bile, leading to its accumulation in the blood [75]. Similar to ammonia, accumulated manganese can penetrate the BBB and selectively deposit in key regions of the brain, such as globus pallidus and substantia nigra, exerting neurotoxicity through multiple mechanisms. Firstly, it damages the presynaptic dopamine transporter protein and reduces the expression of postsynaptic dopamine D2 receptors (DRD2), directly interfering with the dopamine neurotransmission function [76]. Secondly, it accumulates in mitochondria, inducing oxidative stress and disrupting the balance of neurotransmitters such as glutamate, GABA, and NE [76].

Bilirubin accumulation

When the liver is damaged, the metabolic function of bilirubin is disturbed, resulting in abnormally elevated blood bilirubin levels, causing hyperbilirubinemia, which in turn damages the nervous system through dual mechanisms [77]. First, it disrupts the integrity of the BBB through multiple pathways. Unconjugated bilirubin (UCB) deposits in the phospholipid bilayer of cell membranes, leading to the breakdown of the BBB membrane [78, 79]. UCB downregulates tight junction proteins (such as ZO-1 and claudin-5), thereby increasing cellular permeability and resulting in the loss of barrier function [80, 81]. Additionally, UCB disrupts glutathione homeostasis, induces reactive nitrogen species production, and activates MMP-9 to degrade basement membrane components [80]. Collectively, these abnormalities impair BBB function. Accordingly, BBB disruption triggers neuroinflammation and neuronal apoptosis, and inhibits synaptic transmission in neurons [77, 82]. Second, it binds to 5-hydroxytryptamine 3 (5-HT3) receptors in the spinal glial region, activating GABAergic neural pathways and regulating the pain thresholds mediated by cholestasis [83].

Bile acids metabolic dysregulation

Bile acids synthesized by the liver act as signaling molecules and regulate neural functions through Takeda G-protein-coupled receptor (TGR) and farnesoid X receptor (FXR) [84, 85]. They can promote the secretion of intestinal GLP1 [86–88], regulate hypothalamic energy metabolism [89], and show dual effects of central analgesia and peripheral pain promotion in pain regulation [90, 91]. Their disorder is closely related to cholestasis-related neurological symptoms [89].

For instance, ursodeoxycholic acid (UDCA), as a hydrophilic bile acid, has shown therapeutic potential in the treatment of MASLD and metabolic dysfunction-associated steatohepatitis (MASH) [92–94]. Mechanistically, UDCA first activates the PPARα-CYP4A14 signaling axis, promoting fatty acid β-oxidation, reducing hepatic steatosis, and alleviating inflammation in high-fat diet-induced MASLD models [93]. Second, UDCA inhibits the production of pro-inflammatory cytokines by suppressing NF-κB activation and reducing the expression of TNF-α, IL-1β, and IL-6, thereby mitigating hepatic inflammation [95]. Third, UDCA modulates bile acid homeostasis by activating the FXR and TGR5, which regulate lipid metabolism and suppress inflammatory responses [96, 97]. Clinically, accumulating evidence supports the efficacy of UDCA in MASLD. A recent review indicated that UDCA significantly improves liver biochemical markers, including serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyl transpeptidase (GGT) and alkaline phosphatase (ALP), with longer treatment duration associated with greater ALT reduction [92]. In addition, Traussnigg et al. demonstrated a dose-dependent reduction in serum ALT levels with UDCA compared with placebo, and UDCA was safe and well tolerated [98]. These findings suggest that UDCA may serve as a potential therapeutic option for MASLD/MASH, although further large-scale randomized controlled trials are needed to establish its definitive role.

Other metabolites

Liver dysfunction can lead to the accumulation of various neurotoxins in the body, thereby exacerbating neurological damage. Phenolic substances can cause dopaminergic neurodegeneration [99]. Excessive SCFAs can induce neuro-energy crisis by inhibiting mitochondrial function and promoting GLP-1/PYY secretion [100]. AAAs competitively cross the BBB via the LAT1 transporter [101], where they disrupt dopamine and 5-HT and generate false neurotransmitters that replace NE and dopamine, ultimately inducing psychiatric symptoms such as confusion and behavioral abnormalities [102]. Notably, the abnormal levels of 24S-hydroxycholesterol, a cholesterol metabolite synthesized by the liver and a key regulator of cholesterol homeostasis in the brain, are also associated with the pathogenesis of Alzheimer's disease (AD) [103].

On the other hand, the liver also synthesizes a variety of metabolites with neuroprotective effects. Ketone bodies, such as β-hydroxybutyrate, can serve as alternative energy sources for the brain during starvation, and exert neuroprotective effects through inhibiting histone deacetylase [104]. Lactic acid, as a crucial energy substrate, plays a central role between astrocytes and neurons, maintaining energy homeostasis in the brain [105].

Thus, the liver has a dual role in maintaining nervous system health, and its functional status directly affects the survival and function of neurons. Collectively, these findings highlight how hepatic metabolites participate in both physiological and pathological processes of the nervous system through direct and indirect pathways, providing new perspectives for the study of metabolic-neural interactions.

Effects of hepatogenic substances on the nervous system

Bioactive substances secreted by the liver, including hormones, proteins, and exosomes, act as molecular messengers for the interaction between the liver and the nervous system. They can regulate nervous system function by crossing the BBB and targeting receptors on neural cells, and are closely associated with neurodegenerative diseases, cognitive impairment, and other disorders, representing potential therapeutic targets.

Hepatogenic hormones: core regulatory factor

Fibroblast growth factor 21 (FGF21) is a pleiotropic hormone primarily secreted by the liver [106], which can cross the BBB and directly act on the CNS [106]. Santoso et al. found that FGF21 can act on specific hypothalamic nuclei through the NUCB2/Nesf-1 axis to regulate appetite and energy expenditure [107]. Under fasting conditions, liver-derived FGF21 activates corticotropin-releasing hormone (CRH) neurons in the hypothalamus, on the one hand inhibiting lipolysis in white adipose tissue, and on the other hand enhancing thermogenic activity in brown adipose tissue [107]. In the feeding state, FGF21 can regulate energy intake through effector circuits in the CNS and inhibit the consumption of alcohol and carbohydrates [108]. Notably, the neuroprotective effect of FGF21 has attracted increasing attention. Its neuroprotective effects in AD and Parkinson's disease (PD) models suggest that it may exert therapeutic potential by inhibiting neuroinflammation and oxidative stress [109, 110].

Insulin-like growth factor 1 (IGF-1) is a multifunctional cell proliferation regulatory factor secreted by the liver, exerting trophic effects on neuronal regeneration in both the CNS and PNS [111–113]. Its secretion is regulated both by the classical endocrine pathway of pituitary growth hormone (GH) and by the vagal afferent pathway that conveys the hepatic metabolic status to the CNS [114]. Circulating IGF1 can be actively transported from the blood into the cerebrospinal fluid by lipoprotein receptor-related protein 2 (LRP-2), and can also cross the BBB by binding to the IGF1 receptor (IGF1R) on endothelial cells [115]. After IGF1 binds to IGF1R, it can activate the PI3K/mTOR/AKT1 and MAPK/ERK pathways, thereby inhibiting neuronal apoptosis [116, 117]. Additionally, IGF1 can also exert neuroprotective effects by directly regulating calcium-binding proteins [118]. More and more evidence indicates that IGF1 is closely associated with neurovascular dysfunction, aging, and age-related neurodegenerative diseases [119, 120].

Hepatocyte growth factor (HGF) is a chemokine and can also serve as a neurotrophic factor [121, 122]. Studies have found that HGF and its main receptor c-Met are widely expressed in the nervous system, jointly regulating the survival, migration and axon guidance. Furthermore, they significantly promote the maturation and functional maintenance of oligodendrocytes, thereby supporting normal neuronal development and network integration [123, 124].

Liver secreted proteins: secondary regulatory factor

Apolipoprotein E (ApoE) is a secreted molecule mainly synthesized by hepatocytes [125]. ApoE4 is mainly expressed by astrocytes in the brain and by hepatocytes in the periphery [126]. Liu et al. found that hepatogenic substances ApoE4, by damaging cerebrovascular function and synaptic plasticity, can reduce memory and cognitive abilities, increase the proliferation of vascular-related glial cells, and exacerbate cerebral amyloid pathology [127]. Zhang et al. found that liver-derived exosomes carrying ApoE4 are negatively correlated with serum thyroid hormone levels and cognitive function, and can be targeted to brain tissue, thereby leading to cognitive impairment [128]. Additionally, ApoE4 can affect lipid metabolism in astrocytes and specifically enhance neuronal energy metabolism [129]. Clinically, ApoE4 is also recognized as the most significant genetic risk factor for AD. It exacerbates the aggregation of amyloid-β (Aβ), reduces dendritic density, and impairs synaptic plasticity by decreasing synaptic glutamate receptors [130, 131].

Lysophosphatidyl choline (LPC) is an enzyme secreted by the liver [132]. When the liver secretes LPC, an enzyme called Autotaxin (ATX) in the blood rapidly converts it into lysophosphatidic acid (LPA) [133]. Previous studies have found that LPA plays a significant regulatory role in synaptic neurotransmission and plasticity [134, 135]. After fasting, the levels of LPA in the blood and cerebrospinal fluid of mice remain elevated, leading to increased neuronal activity in the cerebral cortex, and subsequently stimulating appetite [132].

Hemochromatosis type 2 protein (HFE2) is a protein secreted by the liver. After entering the bloodstream, it directly promotes the structural integrity of the blood-CNS barrier (BCB) [136]. Studies have shown that liver-specific knockout of the Hfe2 gene leads to the destruction of BCB structure, which in turn causes the accumulation of blood-derived toxic fibrinogen in the brain, a reduction in the number of cortical neurons, and accompanying behavioral deficits in mice [136].

The interactions between these hepatogenic substances and the nervous system not only expand the understanding of the liver-nervous system axis, but also provide theoretical support for the development of liver-derived therapeutic strategies for neurological diseases (Fig. 3).

Fig. 3.

Effects of hepatogenic substances on the nervous system. FGF21, inhibits neuroinflammation and oxidative stress. IGF1, inhibits neuronal apoptosis and regulates calcium-binding proteins. HGF, modulates neuronal survival and migration. ApoE4, impairs cerebrovascular function and affects astrocyte lipid metabolism. LPC, promotes synaptic neurotransmission and enhances neuronal activity in the cerebral cortex. HFE2, maintains the structural integrity of the blood-CNS barrier (BCB) and the number of cortical neurons

Liver-derived exosomes: novel regulatory vectors

Exosomes secreted by the liver, by encapsulating bioactive molecules such as miRNA and proteins, cross the BBB to act on the nervous system, exerting bidirectional regulatory effects in neurological diseases, and serving as a novel carrier for liver-neural interaction [137] (Fig. 4).

Fig. 4.

Effects of liver-derived exosomes on the nervous system. Overall mechanism: liver-derived exosomes act as carriers between the liver and the nervous system, crossing the BBB to deliver dozens of bioactive molecules. Their balanced or imbalanced functions critically regulate neuroinflammation, neuronal survival, and vascular health, impacting the progression of various neurological diseases

Notably, the specific detection and identification of hepatocyte-derived exosomes in mouse models and human samples rely on a combination of exosome isolation techniques and validation with hepatocyte-specific markers. In mouse models, hepatocyte-derived exosomes are isolated from serum, plasma or liver tissue homogenates by ultracentrifugation, density gradient centrifugation and size exclusion chromatography [138]. In human studies, exosomes are primarily isolated from peripheral blood (serum/plasma) or liver biopsy samples by commercial kits or ultracentrifugation [139]. Immunoaffinity capture utilizing antibodies against hepatocyte-specific surface markers, such as asialoglycoprotein receptor 1 (ASGR1) and CYP family 2E1 (CYP2E1), enables selective isolation of hepatocyte-derived exosomes from complex biological fluids including serum and plasma [140, 141]. Nanoscale flow cytometry (nFCM) with fluorescently labeled antibodies against these markers allows quantitative analysis of exosomes subsets from specific cellular origins [140]. These markers are summarized in Table 3.

Table 3.

Commonly used markers for detecting hepatocyte-derived exosomes in mice and humans

Category	Marker	Species	Detection Method	Key Application
Surface proteins	ASGR1	Human/Mouse	Immunoaffinity capture, nFCM	Selective isolation of hepatocyte-derived exosomes
	CYP2E1	Human/Mouse	Immunoaffinity capture, nFCM	Hepatocyte-exosomes identification
	ALB	Human/Mouse	Western blot, ELISA	Hepatocyte origin confirmation
Exosome markers	CD63, CD81, TSG101	Human/Mouse	Western blot, nFCM	Exosome characterization
miRNA cargo	miR-122	Human/Mouse	qPCR, sequencing	Hepatocyte-exosomes enrichment, liver injury biomarker
	miR-192	Human/Mouse	qPCR, sequencing	Hepatocyte-exosomesenrichment, MASLD severity
	miR-128-3p	Human	qPCR, sequencing	Hepatocyte-EV enrichment
	miR-155	Human/Mouse	qPCR, sequencing	Inflammation-associated liver diseases
	miR-146a	Human/Mouse	qPCR, sequencing	Inflammatory liver diseases
	miR-34a	Human/Mouse	qPCR, sequencing	Liver fibrosis, AD
DAMPs	HMGB1	Human/Mouse	Western blot, ELISA	Hepatocyte injury, inflammation

Hepatocyte-derived exosomes primarily cross the BBB via transcytosis rather than diffusion. Several interrelated mechanisms have been identified: receptor-mediated transcytosis, in which exosomes surface proteins (e.g., ApoE, LAMP2) bind to specific receptors (e.g., ApoER2) on brain endothelial cells [15]; adsorptive-mediated transcytosis, driven by electrostatic interactions between positively charged EV surface proteins and negatively charged heparan sulfate proteoglycans (HSPGs) on the endothelial glycocalyx [142]; clathrin- and caveolin-mediated endocytosis, involving energy-dependent vesicular internalization through clathrin-coated pits or caveolae [143]; macropinocytosis, a non-specific, actin-driven process particularly relevant for larger exosomes and under inflammatory conditions [144]; and direct membrane fusion, wherein exosomes release their contents directly into the endothelial cytoplasm [143]. Following internalization, exosomes are transported across the endothelial cell within endosomal compartments and released at the basolateral side via exocytosis, while maintaining the integrity of tight junctions [145]. Using in vitro transwell models, liver-derived exosomes have demonstrated efficient traversal of the BBB and delivery of hepatocyte-derived cargo into the brain parenchyma [15]. These mechanisms underscore the role of hepatic exosomes as key mediators of the liver-brain axis.

It is worth noting that,miR-122 is the most abundant miRNA in the liver, accounting for approximately 70% of the total hepatic miRNA pool, and is highly enriched in hepatocyte-derived exosomes [146]. Importantly, circulating exosomal miR-122 has been shown to be transferred from the liver to the brain via exosomes, where it can modulate neuroinflammatory responses [147]. Mechanistically, miR-122 exerts neuroprotective effects following intracerebral hemorrhage by targeting and inhibiting TP53 while upregulating SLC7A11 expression, thereby alleviating ferroptosis in neural cells [148]. Additionally, it targets and inhibits PDK4 expression to mitigate mechanical allodynia and thermal hyperalgesia resulting from chronic sciatic nerve injury, thus suppressing the development of neuropathic pain [149]. In addition to miR-122, other hepatocyte-derived exosomal miRNAs have been implicated in neuroinflammation. miR-192, which is released from hepatocytes under lipotoxic conditions, has been shown to reduce neuronal apoptosis by targeting apoptosis-related pathways, exerting a neuroprotective effect [150, 151]. miR-128-3p, another liver-enriched miRNA, has been associated with cognitive function in liver disease patients [152–155]. miR-155, a well-known pro-inflammatory miRNA, is upregulated in hepatocyte-derived exosomes during liver inflammation and can be delivered to the brain, where it targets SHIP1 and SOCS1 in microglia, promoting their activation and sustained neuroinflammation [156–159]. Conversely, miR-146a serves as an anti-inflammatory feedback regulator, targeting TRAF6 and IRAK1 to limit excessive inflammatory responses in the CNS [157, 160]. miR-34a, which is elevated in both liver fibrosis and Alzheimer’s disease, targets SIRT1 in neurons and contributes to neuronal apoptosis and synaptic dysfunction [150, 159, 161–163]. Beyond miRNAs, hepatocyte-derived exosomes also carry damage-associated molecular patterns (DAMPs) such as HMGB1, which activate TLR4 on microglia and brain-infiltrating macrophages, inducing the production of IL-1β and IL-6 and thereby amplifying neuroinflammatory cascades [164–167].

Hepatocyte-derived exosomes are characterized by significant disease dependence. In AD, these exosomes demonstrate a bidirectional regulatory effect. On the one hand, aberrantly expressed ApoE is delivered into the brain via exosomes, promoting the deposition and aggregation of Aβ and simultaneously activating microglia-mediated neuroinflammation, thereby accelerating neuronal damage [128, 168]. On the other hand, the exosomes carry miR-192, which targets and regulates apoptosis pathways, reduces neuronal death, and exerts a neuroprotective effect [169, 170]. In neuroinflammatory-related diseases, the function of hepatocyte-derived exosomes is regulated by their own state. Under normal physiological conditions, they primarily exhibit anti-inflammatory effects, whereas under pathological conditions such as lipotoxicity, they excessively transport pro-inflammatory factors like TNF-α, thereby exacerbating neuroinflammatory damage [168].

At the same time, hepatocyte-derived exosomes can transport CYP enzymes to play a role in nerve detoxification [171–173]. Specifically, multiple CYP isoforms have been identified in circulating exosomes, including CYP2E1, CYP1A1/2, CYP2A, and CYP4B [174]. Notably, CYP2E1, the major ethanol-inducible enzyme, is significantly elevated in exosomes from alcohol-exposed rodents and patients with alcoholism, and these CYP2E1-containing exosomes are functional and can be delivered to extrahepatic cells including brain cells, where they may participate in local alcohol metabolism and xenobiotic detoxification [171, 174]. Exosomes isolated from the serum of rats with hepatic ischemia–reperfusion injury (HIRI) have been shown to induce neuronal damage in the hippocampus and cortex of healthy rats [175]. These exosomes are delivered to the hippocampus and cortex, activating the NLRP3 inflammasome and caspase-1 pathway, which promotes oxidative stress and triggers neuronal pyroptosis [176]. When exosomes from hyperammonemic rats were injected into normal rats, these exosomes activated microglia through the TNFR1/NF-κB/GABA transporter 3 (GAT3) pathway, leading to the accumulation of pro-inflammatory factors such as IL-1β, TNF-α, and CD68, which induced neuroinflammation and impaired cognitive function in the rats [168, 177, 178].

These liver-derived exosomes, with the characteristics of crossing the BBB and the ability of precise molecular delivery, serve as carriersconnecting the liver and the nervous system. Their functional imbalance or proper regulation both influence the occurrence, development, and prognosis of neurological diseases, providing a novel entry point for the mechanistic research and therapeutic intervention of related diseases.

Synergistic effects of hepatic cytokines and autonomic dysregulation on the nervous system

In addition to the two major core pathways mentioned above, the inflammatory response mediated by liver-derived cytokines and the dysfunction of the hepatic autonomic nervous system, as auxiliary pathways, further amplify the influence of the liver on the nervous system and participate in the progression of neuropathological injury.

Role of cytokines in neural communication

During liver injury or metabolic abnormalities, KCs and HSCs are activated and release cytokines such as IL-6, TNF-α, and il-1β [179, 180]. These cytokines can cross the BBB or transmit through the vagal afferent pathways to target brain regions such as the hypothalamus and NTS, affecting neuroendocrine and autonomic nerve outputs, thereby regulating systemic glucose metabolism and energy allocation [181]. They also promote neuroinflammation and participates in the occurrence of HE and MASLD-related neurological dysfunction [182, 183].

Effects of hepatic autonomic dysfunction on the nervous system

Liver diseases can induce abnormal autonomic nerve signal transmission through pathways such as oxidative stress and imbalance of inflammatory mediators, specifically the overproduction of pro-inflammatory mediators (TNF-α, IL-1β, IL-6) and insufficient secretion of anti-inflammatory mediators (IL-10, TGF-β), further leading to nervous system dysfunction [5]. For example, in HE, vagus nerve-mediated neuroimmunomodulatory function is impaired [184–186]. In Wilson's disease (WD), copper deposition leads to extensive lesions in the CNS, manifesting as autonomic nervous dysfunction, such as sleep disorders and excessive daytime sleepiness, and non-motor symptoms (NMS) [187, 188]. These evidences suggest that hepatic autonomic dysfunction may directly or indirectly promote pathological changes in the nervous system through pathways such as neuroinflammation, metabolic disorders, and impaired toxin clearance.

Liver diseases affect the nervous system through multiple mechanisms, including metabolic disorders, accumulation of neurotoxins, and autonomic nerve dysfunction. These effects can lead to HE, cognitive dysfunction, neurodegenerative diseases, and other conditions. Therefore, for patients with liver diseases, in addition to treating liver lesions, it is also necessary to pay attention to the potential risks of the nervous system.

Regulation of nervous system on hepatic function

The liver, as the metabolic center of the human, is precisely and efficiently regulated by the nervous system [189]. This process mainly involves the CNS and autonomic nervous system, which direct hepatic metabolic activities through electrical and chemical signals [190, 191]. Meanwhile, neurotransmitters, neuropeptides, and brain-derived exosomes also play significant roles in regulating hepatic function, collectively forming a multi-level network for the nervous system to control the liver.

The CNS acts as the "highest command center" for hepatic regulation

The regulation of the liver by the CNS is centered around the hypothalamus, which is the ultimate commander of metabolic regulation [192]. It does not directly connect to the liver but continuously receives signals from the blood, the liver, and other regions of the brain. It integrates these signals and adjusts the activity balance of the sympathetic and parasympathetic nerves through the brainstem, thereby indirectly regulating hepatic function [193].

Recent research has identified a novel hypothalamic–sympathetic–liver (HSL) axis, which operates independently of adrenal activity during acute stress responses to facilitate rapid glucose release and exerts multi-level regulatory effects on liver regeneration [194]. Specifically, corticotropin-releasing hormone (CRH) neurons in the paraventricular nucleus (PVN) of the hypothalamus (PVN-CRH neurons) activate the VMH-CRH2-RPA-IML pathway, promoting hepatic sympathetic nerve activity and inducing hepatic gluconeogenesis to achieve rapid blood glucose supply [194].

In addition to direct neural connections, the hypothalamus can also influence metabolic functions through neuroendocrine connections. The common ones can be categorized into the hypothalamic-pancreatic axis, the hypothalamic-adrenal axis and the hypothalamic-pituitary axis [195]. In the hypothalamic-pancreatic axis, the autonomic nerve releases glucagon and insulin, which directly act on the liver to affect metabolism [196]. In the hypothalamic-adrenal axis, the autonomic nerve prompts the adrenal medulla to release epinephrine, NE, and other catecholamines to regulate liver metabolism [196]. In the hypothalamic-pituitary axis, the release of glucocorticoids and thyroid hormones is stimulated by pituitary hormones, ultimately influencing hepatic metabolism through hormones such as cortisol [194, 197].

Furthermore, spinal cord injury (SCI) disrupts the CNS, leading to abnormal hepatic lipid metabolism andexacerbated inflammatory responses, confirming the indispensability of CNS regulation [198–200]. Specifically, SCI can induce the activation of HSCs in the rat liver [201], reduce the serum level of FGF21 and damage its signaling pathways in the liver and adipose tissue of mice [202] It causes metabolic disorders and steatohepatitis [203]. RNA-seq results further indicate that SCI severely affects the activity of the hepatic transcriptome and influences the occurrence of liver diseases [204]. In summary, SCI comprehensively impairs hepatic function by disrupting central command, from cellular activation and molecular signaling to the transcriptome level, fully confirming the core role of the CNS in maintaining hepatic homeostasis.

Direct regulatory effects of the dual branches of the autonomic nervous system on hepatic homeostasis

Under the regulation of the CNS, the autonomic nervous system, through the sympathetic and parasympathetic nerves (mainly the vagus nerve), achieves unconscious regulation of the liver. The two functions complement each other and work together to maintain liver homeostasis [205–208] (Fig. 1D).

Sympathetic nerve system promotes the progression of liver diseases

The SNS plays a complex and crucial role in liver physiological regulation and disease progression. It can not only directly innervate intrahepatic cells, but also deeply participate in multiple processes of liver inflammation, fibrosis, metabolic disorders and regeneration through indirect pathways such as immune regulation.

Firstly, the SNS is a significant driver of pro-inflammatory and pro-fibrotic responses in the liver [209]. During liver fibrosis progression, the nerve terminals behind the sympathetic ganglion release NE, which binds to the α-adrenergic receptors on the surface of HSCs, triggering downstream Ca²⁺ mobilization, ERK1/2 and PI3K/Akt phosphorylation cascades, further upregulating the expression of TGFβ, enhancing the transcriptional activity of pro-fibrotic genes, thereby stimulating collagen deposition and the expression of α-SMA, and accelerating liver fibrosis [16–19]. Moreover, activated HSCs can express adrenergic receptors and synthesize endogenous catecholamines, which upregulate TGF-β1 and promote collagen production through autocrine or paracrine mechanisms, forming a local positive feedback loop that accelerates fibrosis [17]. The sympathetic nerve can also indirectly induce inflammatory and fibrotic responses in HSCs by regulating immune cells such as KCs [19]. Under systemic stress conditions such as acute SCI, excessive activation of the sympathetic nerve can induce the liver to release pro-inflammatory factors, including IL-1β, CXCL10, CXCL1 and SAA-2, and further trigger systemic inflammatory responses [210].

Secondly, the SNS also plays a key role in hepatic glucose metabolism. Its activation can promote the breakdown of hepatic glycogen and generate coordinated calcium waves in the liver lobules through calcium signaling pathways, thereby enhancing hepatic glucose output capacity under physiological conditions such as exercise [194]. Notably, metabolic stress induced by a high-fat diet can trigger intrahepatic sympathetic neuropathy, characterized by reversible degeneration of nerve fibers caused by TNFα produced by CD11b⁺F4/80⁺ immune cells, exacerbating insulin resistance in hepatocytes [20]. Sarm1 is a crucial downstream mediator that regulates sympathetic nerve injury induced by metabolic stress, which can initiate programmed axonal degeneration and result in the loss of sympathetic nerve fibers. The deletion of the Sarm1 gene protects sympathetic nerves from degeneration, preserves NE-mediated immunosuppression, and consequently enhances metabolic stress resilience [20].

Moreover, the interaction between the SNS and immune cells plays a significant regulatory role in the progression of liver diseases. When a stroke occurs, NE secreted by the sympathetic nerve can activate natural killer T cells (NKT) in the liver of mice, generating cytokines and chemokines [211]. Conversely, blocking the innervation of the sympathetic nerve can inhibit NKT cell apoptosis, thereby alleviating hepatic inflammation [212, 213].

Notably, the SNS plays a dual role in liver regeneration and tumorigenesis. On the one hand, the SNS responds to benign stress through the β-adrenergic signaling pathway, triggering the rapid recruitment of group 1 innate lymphoid cells (ILC1) in the liver, which in turn promotes the secretion of IL-22 and ultimately enhances liver regeneration capacity [214]. It also promotes hepatic lymphangiogenesis by inducing VEGF-C expression, which can act as a compensatory protective mechanism to delay the progression of liver fibrosis [12]. On the other hand, during the progression of CLD and hepatocellular carcinoma (HCC), the sympathetic nerves maintains a pro-tumor inflammatory microenvironment by upregulating factors such as IL-6 and TGF-β, thereby promoting HCC development [19]. In advanced liver disease, it also contributes to the occurrence of extrahepatic complications such as cardiovascular dysfunction and disturbances in metabolic homeostasis, including glucose metabolism, lipid metabolism, and protein metabolism [215]. Excessive activation of the sympathetic nervous system disrupts hepatic regulation of glucose uptake and release [194], impairs lipid oxidation and storage [216], and alters protein synthesis and catabolism [217], leading to systemic metabolic disorders that further exacerbate the progression of liver disease and related complications.

These pieces of evidence collectively indicate that the SNS exerts a crucial and dual regulatory role in various processes, including liver inflammation, fibrosis, metabolic regulation, regeneration, and tumorigenesis, through multiple cellular targets and signaling pathways. It serves as a key entry point for understanding the pathophysiological mechanisms of the liver diseases and exploring novel therapeutic strategies.

Vagus nerve system exerts protective effects against liver diseases

Different from the dual role of the SNS, the vagus nerve, as the core of parasympathetic nervous system, mainly plays an anti-inflammatory and metabolic protective role in the liver. Although research on its regulatory mechanisms of hepatic lipid metabolism is still insufficient, recent studies have gradually revealed its key position in regulating liver immune and metabolic homeostasis.

Vagus nerve fibers are mainly distributed in the porta hepatis and intrahepatic bile duct system, regulating hepatocellular function by releasing neurotransmitters such as ACh [29, 31, 218]. Vagus nerve signal deficiency, such as vagotomy, can lead to enhanced hepatic glycolysis and fatty acid synthesis, as well as weakened β-oxidation, thereby promoting lipid accumulation in hepatocytes and exacerbating metabolic disorders [219].

In terms of inflammation regulation and immune responses, hepatic vagus nerve cholinergic signaling regulates α7nAChR on KCs during the pathogenesis of MASH. This inhibits inflammatory responses, significantly reduces hepatic lipid droplet area, suppresses NF-κB activation, inhibits inflammatory responses, and significantly reduces hepatic lipid droplet area, and thereby alleviating liver injury [31, 220]. In high-fat diet-induced obese and insulin-resistant mice, vagal nerve stimulation of the α7nAChR on KCs inhibits the secretion of IL-6 and the activator of signal transducer and activator of transcription 3 (STAT3), thereby reducing the expression of inflammatory factors and blocking the recruitment and activation of inflammatory cells in the liver [29]. Meanwhile, in acute-on-chronic liver failure (ACLF), the vagus nerve releases ACh to the liver, which can directly inhibit the production of pro-inflammatory chemokine C-X-C motif chemokine ligand 9 (CXCL9) by KCs, thereby blocking the recruitment and activation of inflammatory cells in the liver and preventing the development of ACLF [221]. Additionally, vagus nerve stimulation can alleviate HIRI [208, 222], whereas vagotomy significantly increases pro-inflammatory factor levels and exacerbates liver injury [223].

In the regulation of cell death, the vagus nerve reduces the generation of ROS and alleviates Fas-induced hepatocyte apoptosis by modulating α7nAChR [38]. Conversely, vagotomy (cutting the vagus nerve) leads to increased activity of caspase-3 (CASP3) and induces apoptosis, while the use of α7nAChR agonists can effectively alleviate liver injury [38].

Notably, accumulating evidence has demonstrated the critical role of the vagus nerve system in alcohol misuse and alcohol-related liver disease (ALD) [224, 225]. Chronic alcohol exposure not only directly damages hepatocytes and induces hepatic steatosis [226–228], but also impairs vagus nerve function through CNS-mediated neuroplasticity changes [229, 230]. Specifically, alcohol-induced CNS dysfunction suppresses vagal tone, leading to reduced ACh release in the liver and impaired α7nAChR-mediated anti-inflammatory signaling in KCs [231, 232]. This dysregulation exacerbates hepatic inflammation, promotes lipid accumulation, and accelerates ALD progression from simple steatosis to alcoholic steatohepatitis (ASH) and cirrhosis. Interestingly, vagotomy has been shown to completely abolish the alcohol deprivation effect, indicating that vagal afferent signals from the liver and intestinal mucosa to the brain play a critical role in alcohol craving and relapse [233]. These findings highlight the vagus nerve as a key mediator connecting alcohol consumption, hepatic steatosis, inflammation, and CNS effects, and suggest that α7nAChR agonists may hold therapeutic potential for ALD [232].

These studies indicate that the vagus nerve effectively alleviates liver inflammation and protects hepatocytes by regulating the release of the ACh and the activation of related receptors, providing important clinical translational prospects for the treatment of liver diseases based on vagus nerve modulation.

Existence and functional mechanism of hepatic sensory nerves

The hepatic sensory nerves are mainly responsible for transmitting real-time information such as the metabolic status, inflammatory level, hemodynamics, and nutritional signals of the liver to the CNS, thereby participating in the regulation of feeding behavior, energy metabolism, emotional responses, and immune responses [207]. Sensory nerves form a feedback pathway between the CNS and the liver, and act in coordination with the autonomic nervous system to complete the neural regulation of the liver.

In terms of distribution, hepatic sensory nerves mainly originate from the nodose ganglion of the vagus nerve, enter the liver via the hepatic branches of the vagus nerve, and are concentrated in the portal triad regions, including the hepatic artery, portal vein, and bile ducts. They are particularly abundant in the adventitia of portal vein branches and near the bile duct epithelium, and rarely extend into the hepatic parenchyma [207]. These sensory nerve fibers express various inflammatory mediator receptors and ion channels, such as transient receptor potential ankyrin 1 (TRPA1) and transient receptor potential vanilloid 1 (TRPV1). They can be activated when stimulated by inflammatory factors or pathogens, triggering action potentials and calcium ion influx, thereby promoting the release of SP and CGRP to exert regulatory effects [234, 235].

During the pathological process of the liver, SP can bind to its receptor, neurokinin 1 receptor (NK1R), activating HSC and promoting the progression of liver fibrosis [236–238]. Additionally, in an inflammatory environment induced by TNF-α, SP can also act on LSECs, enhancing cell viability and promoting the secretion of nitric oxide and HGF, thereby promoting hepatocyte repair and regeneration [239]. Meanwhile, the expression of CGRP significantly increased after partial hepatectomy [240]. Sensory nerves can activate the YAP/TAZ signaling pathway through the CGRP-RAMP1 axis, promoting liver regeneration in acute and chronic liver injury models [240].

Although the CNS cannot directly perceive liver fibrosis or stiffness at the conscious level due to the lack of specialized sensory endings in the liver parenchyma, emerging evidence indicates that the CNS can detect these pathological changes through specific afferent pathways and indirect mechanisms. Notably, hepatic sensory nerves may perceive physical signals associated with liver fibrosis, such as increased extracellular matrix stiffness [241]. Anatomically, the liver is innervated by sensory afferent fibers from the vagus and dorsal root ganglia that sense chemical, osmotic, inflammatory, and hemodynamic changes [207]. Mechanistically, ablation of hepatic vagal afferent neurons blocks the transmission of liver pathological signals to the brainstem [207]. Clinically, serum liver fibrosis biomarkers (APRI, FIB-4) correlate with globus pallidus MRI signal hyperintensities [242]. Furthermore, cirrhotic patients without overt HE exhibit widespread reductions in brain microstructural complexity (diffuse gray and white matter regions) [243]. Together, these findings demonstrate that liver fibrosis-related signals are sensed by the CNS via sensory nerve (vagal afferent) pathways as an important component of the liver-nervous system axis.

In summary, sensory nerves real-time monitor hepatic status via multimodal receptors and transmit signals to the CNS. Meanwhile, they regulate hepatic inflammation, fibrosis, and regenerative repair by releasing neuropeptides. As such, they serve not only as a critical link between metabolic and psychiatric disorders but also provide novel targets and directions for neuromodulatory therapies of liver diseases.

Effects of neurotransmitters on hepatic function

As the core chemical messenger for signal transmission in the nervous system, neurotransmitters participate in regulating hepatic blood circulation, metabolic functions, immune-inflammatory responses, as well as regeneration and repair by binding to intrahepatic or extrahepatic receptors [244]. They serve as an important bridge connecting neural regulation and liver physiology and pathology (Table 4).

Table 4.

Effects of neurotransmitters on hepatic function

Neurotransmitters	Primary effects in the Liver
Dopamine [245]	It exerts complex dual regulatory effects in the liver. It can alleviate inflammation and fibrosis under different contexts, but may also promote the malignant transformation of fibrosis
Serotonin (5-HT) [246]	It not only promotes hepatocyte regeneration, but also exacerbates metabolic disorders, liver fibrosis and tumor metastasis by driving oxidative stress, inflammatory responses and immune microenvironment reprogramming
ACh [247, 248]	It creates an ACh-rich tumor microenvironment in the liver that drive cholangiocarcinoma metastasis, and simultaneously promotes hepatic regeneration by enhancing pro-regenerative function of KCs and suppressing overactivation and IFN-γ production of hepatic CD8 + T cells
NE [18]	It is the main neurotransmitter of the SNS. Mediated by different receptors, it can activate HSCs to drive fibrosis, inhibit the activation of pro-inflammatory macrophages to impede liver regeneration, and also improve insulin sensitivity by promoting lipolysis and reducing inflammation

Regulation of dopamine on hepatic inflammation and fibrosis

Dopamine plays a complex dual regulatory role in the process of hepatic inflammation and fibrosis [245]. Its regulatory mechanism involves both the indirect actions through neurometabolic circuits and direct effects on hepatocyte signal transduction.

In terms of protective effects, dopaminergic signaling significantly alters the hepatic physiological states by activating DRD2, suggesting an upstream regulatory role of the liver-nervous system axis in liver disease pathogenesis [249]. At the metabolic level, the dopaminergic system regulates hepatic glucose metabolism through glutamatergic neurons and GABA signaling inputs in the lateral hypothalamic area, indirectly influencing fibrotic progression associated with metabolic inflammation [250]. Notably, dopamine exhibits direct protective effects against MASH by inhibiting the activation of the p65 pathway, effectively alleviating inflammatory responses in both in vivo and in vitro models [251]. It can also upregulate the expression of interferon-stimulated gene 15 (ISG15) by activating the JAK/STAT signaling pathway, thereby inhibiting the expression of hepatitis B virus antigens, showing therapeutic potential in alleviating HBV-related hepatic inflammation and delaying fibrosis [252]. Notably, exercise serves as a key physiological stimulus for dopamine release [253], and the elevation of dopamine induced by exercise exerts significant regulatory effects on CLD and cirrhosis [250, 254]. During moderate-intensity exercise, the CNS releases dopamine from the substantia nigra and ventral tegmental area, which not only regulates exercise tolerance and motivation but also exerts hepatoprotective effects through both neural and humoral pathways [255]. In patients with cirrhosis, low-dose dopamine has been shown to improve systemic and splanchnic hemodynamics without adverse effects [256, 257]. Mechanistically, exercise-induced dopamine activates hepatic DRD2, inhibits hepatic inflammation by suppressing the NF-κB/p65 pathway [258], and reduces hepatic lipid accumulation by regulating glucose and lipid metabolism [259], thereby alleviating the progression of liver fibrosis in cirrhotic and CLD models.

In terms of pro-pathological effects, dopamine can induce the expression of sulfotransferase SULT1A3/4, promoting epithelial-mesenchymal transition (EMT) and tumor stemness in HCC. This is particularly critical in hepatocarcinogenesis against the background of chronic inflammation and fibrosis, suggesting that dopamine may exacerbate the malignant transformation of fibrosis in the microenvironment [260].

In summary, dopamine not only indirectly regulates the hepatic immune microenvironment through neuro-metabolic circuits, but also directly participates in hepatocyte signal transduction. Its multiple roles in inflammatory responses, fibrosis formation, and even HCC progression make it a target worthy of in-depth exploration in liver disease intervention.

The role of 5-hydroxytryptamine in hepatic metabolism and regeneration

5-hydroxytryptamine (5-HT) is not only a key neurotransmitter but also a neuroendocrine hormone synthesized in the CNS and enterochromaffin cells of the gastrointestinal tract. It plays a multifaceted and complex regulatory role in hepatic physiological and pathological processes [246].

In terms of physiological protection, 5-HT is a crucial signaling molecule that regulates systemic energy homeostasis, significantly affecting the balance of hepatic glucose and lipid metabolism [261]. In liver regeneration, 5-HT can promote hepatocyte proliferation by inducing the expression of autocrine factors such as TGF-α and IGF-I, demonstrating positive liver regeneration potential [262].

In terms of pathological promotion, abnormal activation of the 5-HT system is a key pathogenic mechanism of multiple liver injuries. In various liver injury models, abnormal activation of the 5-HT system has become a key pathogenic mechanism. For instance, in alcoholic liver disease, 5-HT exacerbates lipogenesis and endoplasmic reticulum stress, driving disease progression [263]. In CCl₄-induced acute liver injury or hyperglycemia-mediated diabetic liver disease, the expression of 5-HT2A receptor (5-HT2AR) and metabolic enzyme monoamine oxidase A (MAO-A) is upregulated, accelerating 5-HT degradation and inducing massive production of mitochondrial ROS. This further triggers oxidative stress, inflammatory responses, and HSCs activation, ultimately promoting fibrosis [264, 265]. In cholestasis and liver fibrosis models, cholangiocytes and HSCs highly express 5-HT2A/2B/2C receptors. Receptor activation exacerbates liver inflammation and fibrosis, while their antagonists can significantly alleviate these pathological changes [266]. Meanwhile, 5-HT also plays a significant role in liver immunity and tumor metastasis processes. By regulating macrophage foaming, vascular endothelial inflammation, and hepatic steatosis, 5-HT participates in the development of metabolic inflammatory liver diseases such as atherosclerosis [267]. 5-HT induced histone serotonylation (H3Q5ser) and orchestrated histone citrullination (H3cit) in neutrophils to trigger chromatin decondensation and facilitate the formation of neutrophil extracellular traps (NETs), thereby significantly promoting the liver metastasis of neuroendocrine cancers [268]. In addition, 5-HT is also involved in regulating the hepatic immune microenvironment. Using the selective 5-HT2AR antagonist fluoxetine or knocking out the Htr2a gene can enhance the cytotoxicity of CD8⁺ T cells, inhibit MAPK signaling, and significantly prolong the survival of mice with HCC, indicating that 5-HT2A is a potential immunomodulatory target [269].

Based on the above mechanisms, intervention strategies targeting the 5-HT system, such as 5-HT2AR antagonists, 5-HT synthesis inhibitors or MAO-A inhibitors, have shown promising therapeutic effects in various experimental models, suggesting that the 5-HT system may be a potential therapeutic target for the treatment of cancer, it is expected to provide new therapeutic strategies for a variety of liver diseases.

Effects of other neurotransmitters on the liver

ACh and NE, as the core neurotransmitters of the autonomic nervous system, also play complex roles in the physiological and pathological processes of the liver. They act synergistically with dopamine and 5-HT to form a delicate regulatory network of the liver-nervous system axis.

ACh exhibits dual regulatory effects. On the one hand, intrahepatic ACh can be synthesized by vagal nerve terminals, intrahepatic cholangiocarcinoma (ICC) cells, and even infiltrating macrophages, collectively forming an ACh-rich tumor microenvironment. It upregulates β-catenin through the CHRNA5-CAMKII-GSK3β axis, driving EMT and accelerating ICC transfer [247]. On the other hand, ACH can promote hepatic regeneration by acting on α7nAChR expressed on KCs and hepatic CD8 + T cells [248]. Specifically, ACh signaling via α7nAChR enhances the pro-regenerative function of KCs while restraining excessive activation of hepatic CD8 + T cells and inhibiting their production of pro-inflammatory interferon-γ (IFN-γ), thereby creating a favorable microenvironment for hepatocyte proliferation and effective liver regeneration [248].

The effects of NE are also multifaceted. During the process of liver fibrosis, NE released by the hepatic sympathetic nerve can activate HSCs through α-adrenergic receptors, promoting their proliferation and excessive secretion of extracellular matrix, thereby driving the progression of fibrosis [18]. During the process of liver regeneration, NE inhibits the activation of pro-inflammatory macrophages by acting on adrenergic receptor β2 (ADRB2) on hepatocytes, ultimately suppressing the regeneration process [22]. In terms of metabolic regulation, NE has a protective effect. It can significantly inhibit the progression of MASLD by promoting lipolysis [270], and also improve hepatocyte insulin sensitivity by reducing the inflammatory response mediated by immune cells, thereby alleviating metabolic liver injury [20].

In summary, the effects of neurotransmitters on liver function exhibit a significant double-edged sword characteristic. Under physiological conditions, they finely regulate regeneration, metabolism, and inflammation to maintain liver homeostasis. Whereas in pathological states, their signaling pathways can be hijacked by factors such as cancer or gut microbiota dysbiosis, thereby promoting tumor metastasis and drug resistance or triggering intense inflammatory cell death. Notably, these neurotransmitter systems do not operate independently. For instance, during the process of liver fibrosis, the cholinergic anti-inflammatory pathway forms a negative feedback regulatory network with the 5-HT-ergic system, while dopaminergic neurons regulate the release intensity of NE through presynaptic inhibition. These studies not only deepen our understanding of the role of the neuro-immune-metabolic network in liver diseases, but also provide new potential targets for therapeutic intervention.

Effects of neuropeptides on the liver

Neuropeptides, as a class of important signaling molecules, exert significant context-dependent regulatory effects on the liver. Their specific functions depends on the type of neuropeptides, pathological background and downstream signaling pathways, mainly influencing hepatocyte survival, metabolism, immune response, and the process of fibrosis.

During HIRI, neuropeptides can induce autophagy by activating the CREB-KLF4 signaling pathway, thereby alleviating hepatocyte damage and improving prognosis, highlighting their key role in maintaining hepatocyte homeostasis under stress [271].

However, in chronic pathological conditions, neuropeptides mostly exhibit pathological promoting effect. Studies have shown that neuropeptides can accelerate hepatic steatosis in mice fed a medium-fat/medium-fructose diet, suggesting that they may play an important role in the occurrence and development of MASLD [272]. Furthermore, the neuropeptide CGRP binds to its receptor component, the receptor activity-modifying protein 1 (RAMP1), and simultaneously activates two key pro-fibrotic signaling pathways, TGFβ1/Smad2 and Hippo/YAP. This significantly promotes HSCs activation and extracellular matrix deposition, ultimately driving the progression of liver fibrosis [235]. This functional diversity depends on the specific type of neuropeptide, the pathological context, and the downstream signaling pathways activated. An in-depth analysis of the specific mechanisms of neuropeptides in the liver will provide important scientific evidence for the development of novel targeted therapeutic strategies for hepatic steatosis, ischemia–reperfusion injury, and liver fibrosis.

An in-depth analysis of the specific mechanisms of neuropeptides in the liver will provide important scientific evidence for the development of novel targeted therapeutic strategies for hepatic steatosis, ischemia–reperfusion injury, and liver fibrosis.

Impact of brain-derived exosomes on the liver

Although liver-derived exosomes have been demonstrated to enter the brain under normal physiological conditions, the transport of brain-derived exosomes to the liver appears to be strictly restricted by the BBB, which can affect the liver under pathological conditions (Fig. 5).

Fig. 5.

Impact of brain-derived exosomes on the liver. Overall mechanism: brain injury alters brain-derived exosomes components, which cross the compromised BBB and circulate to the liver. These exosomes induce hepatic infiammation, oxidative stress, and senescence. Protective effects on liver function have not been observed

Existing studies indicate that brain-derived exosomes exert their effects mainly by inducing oxidative stress and inflammation in the liver. These exosomescross theBBB, activate inflammatory pathways, and promote oxidative damage in liver tissue.Hazelton I. et al. observed in a mouse model of traumatic brain injury (TBI) that neuronal injury activated microglia to release exosomes [273]. These exosomes are able to cross the damaged BBB and accumulate in the plasma. Exogenous administration of exosomes isolated from the plasma of TBI mice leads to secondary hepatic inflammation [273]. In addition, injection of exosomes from the brains of TBI mice into healthy mice results in liver inflammation and increased hepatocellular senescence [274]. Astrocyte-derived exosomes, which disseminate to the liver, induce hepatic inflammation by increasing the expression of IL-1β and TNF-α, as well as the infiltration and activation of neutrophils [275]. Lin H. et al. also observed that exosomes from the brains of septic rats cause damage and inflammation to the liver of healthy rats [276].

It is noteworthy that alcohol abuse and prolonged exposure to alcohol can trigger the release of brain-derived exosomes. Microglia are the primary producers of these exosomes [277], followed by astrocytes [278], while damaged neurons may also play a contributory role. Alcohol-induced oxidative stress and P2X7 receptor activation facilitate exosome biosynthesis and modify their composition [279]. These brain-derived exosomes are enriched with pro-inflammatory substances, including miR-155, heat shock proteins, and DAMPs [277, 279, 280]. Under conditions of alcohol-induced neuroinflammation and BBB damage, these brain-derived exosomes enter the bloodstream and are transported to peripheral organs, including the liver [279, 281]. Upon reaching the liver, they may be taken up by hepatocytes and KCs, where their pro-inflammatory substances could exacerbate the progression of ALD [281]. This inter-organ communication via exosomes represents a novel mechanism that links alcohol-induced neuroinflammation to peripheral organ pathologies.

In conclusion, brain injury alters the composition of brain-derived exosomes, enabling them to cross the BBB and transmit injury signals to the liver, thereby triggering hepatic inflammation. Currently, all relevant studies have not observed that brain-derived exosomes have a protective effect on the liver. These findings not only deepen our understanding of hepatic physiological and pathological processes but also provide new intervention targets for the treatment of liver diseases.

Role and molecular mechanisms of the liver-nervous system axis in diseases

Liver dysfunction can affect the homeostasis of the CNS and PNS through multiple pathways, causing various neurological complications such as HE and hepatic myelopathy. Meanwhile, neurological abnormalities may also lead to liver lesions through neuroendocrine and immunomodulatory mechanisms. Although existing research has revealed some molecular mechanisms, there are still many unsolved mysteries regarding the complete pathway of the liver-nervous system crosstalk. The core pathogenic mechanisms mainly involve the accumulation of metabolic toxins (ammonia, manganese, bile acids, pseudo-neurotransmitters, etc.), activation of inflammatory responses, impairment of the BBB, and imbalance of neural regulation. Although different diseases present with distinct clinical manifestations, their development largely revolves around these common mechanisms. By sorting out the existing research results, we aim to provide new theoretical basis for the diagnosis and treatment of related diseases and look forward to future research directions.

Central nervous system diseases induced by hepatic disorders

The liver plays a crucial role in maintaining and eliminating toxic chemicals from the body. During the progression of liver diseases, when the liver fails to detoxify the blood, it leads to the accumulation of substances such as bilirubin, ammonia, manganese, and bile acids. These substances can damage the nervous system through multiple pathways, including neuronal injury, activation of glial cells, and alterations in neurotransmitter balance.

Hepatic encephalopathy

HE is a brain dysfunction disorder caused by cirrhosis. Its clinical features include acute or fluctuating brain dysfunction such as consciousness disorders, behavioral abnormalities, and asterixis [282]. The pathogenesis of HE centers on the synergistic effects of multiple toxic substances in the context of liver cirrhosis, which jointly damage the CNS (Fig. 6).

Fig. 6.

Pathogenesis and pathophysiology of hepatic encephalopathy. The pathological mechanism of HE caused by liver cirrhosis was systematically demonstrated from three aspects: systemic factors, BBB and brain changes. Cirrhosis leads to elevated blood ammonia, excessive manganese deposition, bile acid dysregulation, and pseudo-neurotransmitter accumulation, which disrupt the integrity of the BBB and allow neurotoxins to enter the brain. This subsequently induces astrocyte edema, neurotransmitter imbalance, manganese deposition, and neuroinflammation, and ultimately neuronal death, ultimately resulting in HE. (The BBB is mainly composed of endothelial cells, basement membrane, tight junctions, and astrocyte endfoot, etc. Neurotoxins damage BBB integrity by directly damaging membrane components, downregulating tight junction proteins, triggering neuroinflammation, and degrading basement membrane components.)

Ammonia metabolism disorder is the most crucial pathogenic mechanism. When cirrhosis leads to impaired liver urea synthesis, elevated blood ammonia crosses the BBB and enters the brain, disrupting glutamine synthetase activity in astrocytes. This leads to the accumulation of glutamine, inducing cellular hyperosmotic edema and the formation of characteristic Alzheimer type II astrocytes, while simultaneously disturbing cerebral energy metabolism and neurotransmission [65, 283]. In addition, hyperammonemia also affects the mitochondrial function of astrocytes. The transport of glutamine to mitochondria leads to partial consumption of α -ketoglutaric acid, weakening oxidative phosphorylation capacity and blocking the autophagy degradation process [21]. Furthermore, excessive activation of the SNS can promote HSCs activation through α-adrenergic receptors, exacerbating blood ammonia accumulation [20]. The synergistic effect of inflammatory factors and ammonia further amplifies neurotoxicity [21].

The accumulation of pseudo-neurotransmitters is another key pathogenic mechanism. In cirrhosis, phenylethylamine and tyramine produced by the decomposition of AAAs, such as phenylalanine and tyrosine, in the intestine cannot be cleared by the liver. After entering the brain tissue, they are converted into phenylethanolamine and octopamine by the action of β-hydroxylase. While structurally similar to the normal neurotransmitters NE and dopamine, their physiological effects are significantly weaker. The accumulation of these pseudo-neurotransmitters in the brain displaces normal neurotransmitters, leading to the buildup of inhibitory neurotransmitters such as GABA, thereby inducing cognitive dysfunction and altered consciousness [284].

Other toxic substances also exacerbate neurological dysfunction through various pathways. In the cirrhotic state, excessive deposition of manganese in the basal ganglia region can induce extrapyramidal symptoms [285]. Bile acids, such as allocholic acid, accumulate abnormally in cirrhotic patients, which can act as allosteric modulators of GABA receptors, enhancing inhibitory neurotransmission [286, 287]. Meanwhile, bile acids can also disrupt the integrity of the BBB to promote the entry of neurotoxins such as ammonia and manganese into the brain, and activate microglia to induce neuroinflammation [288].

It is worth noting that in liver disease animal models, neuroinflammation is the core pathological link connecting liver dysfunction and behavioral abnormalities in the CNS. In classic HE models, including bile duct ligation (BDL) or thioacetamide (TAA)-induced cirrhotic rats [289, 290], as well as the "double-hit" model combining LPS and ammonia treatment [291], the central TLR4/NF-κB signaling pathway is activated, which can directly induce neuroinflammation [292]. Furthermore, neuroinflammation directly drives behavioral abnormalities through microglia activation, inflammatory factor release, and cross-organ signaling [293, 294].

Based on the core pathology of liver cirrhosis, the aforementioned toxic substances accumulate synergistically and collectively drive the onset and progression of HE through multiple pathways. These pathways include disrupting the integrity of the BBB and contributing to neuronal death in key brain regions such as the hippocampus, basal ganglia, and cerebral cortex. This neuronal loss underlies the persistent cognitive impairment and motor dysfunction observed in patients with HE.

Acquired hepatocerebral degeneration

Acquired hepatocerebral degeneration (AHD) is a chronic and persistent neurological complication caused by liver cirrhosis, characterized primarily by organic damage to the brain and/or spinal cord [295, 296]. Its core pathogenic mechanism overlaps with HE, both involving the synergistic toxic effects of manganese accumulation and hyperammonemia.

The clinical manifestations of AHD are diverse and can occur alone or in combination, mainly including hepatic myelopathy, cirrhosis-related Parkinsonism, ataxia, as well as myoclonus, dysarthria, and cognitive impairment [297]. The pathogenesis of AHD is closely related to the deposition of manganese in basal ganglia [298]. In the state of liver failure, manganese cannot be effectively excreted through bile. Coupled with the formation of portosystemic shunt, it leads to an increase in blood manganese concentration, causing dysfunction of the dopaminergic system [299]. This process damages presynaptic dopamine transporters and postsynaptic DRD2, ultimately leading to ataxia and various movement disorders [299]. In addition, hyperammonemia also plays a synergistic role in the pathogenesis of AHD, which can exacerbate astrocyte edema and cerebral energy metabolism disorders, further promoting neurological damage [300].

Cirrhosis-related Parkinsonism

Cirrhosis-related Parkinsonism, the main manifestation of AHD, is a neurological complication caused by chronic liver failure. It is mainly characterized by clinical features such as bradykinesia, muscle rigidity, resting tremor, and gait instability [301]. The pathogenic mechanism is centered on the synergistic toxicity of manganese accumulation and hyperammonemia, sharing key pathological links with AHD.

The core molecular mechanism is dual in nature. Hyperammonemia can cause edema of astrocytes in the basal ganglia and disrupt neurotransmitter balance, particularly by enhancing inhibitory GABAergic signaling and depleting excitatory glutamate, thereby creating a braking state in the motor circuit. On the other hand, manganese accumulated due to impaired biliary excretion, not only triggers oxidative stress but also directly inhibits dopamine synthesis and impairs dopaminergic neurotransmission [302]. Furthermore, chronic manganese exposure promotes the misfolding of α-synuclein (α-Syn), thereby exacerbating neural damage [298, 303]. It is precisely the synergistic toxicity of ammonia and manganese, exacerbated by the systemic inflammation inherent in CLD, that collectively disrupts the delicate balance of basal ganglia circuits, ultimately leading to the development of parkinsonian symptoms [301].

This disease is significantly different from primary PD [300] (Table 5). It is characterized by symmetric symptoms, multiple HE, and poor response to levodopa therapy [306]. Liver transplantation can restore hepatic detoxification and eliminate systemic toxins, leading to significant improvement or even disappearance of symptoms in most patients, which confirms that the etiology is directly related to the impairment of liver function [301].

Table 5.

Comparison between Cirrhosis-related Parkinsonism and Primary Parkinson's disease

Diseases	Cirrhosis-related Parkinsonism [301]	Primary Parkinson's disease (PD) [304, 305]
Etiology	Secondary. It arises from manganese toxicity and hyperammonemia induced by liver failure/portosystemic shunting	Primary. It is associated with the degeneration and death of dopaminergic neurons in the substantia nigra. The etiology is unknown
Core Pathology	Manganese deposition in the basal ganglia, leading to impaired post-synaptic dopamine receptor signaling	Loss of dopaminergic neurons in the substantia nigra pars compacta, resulting in deficiency of the dopamine
Tremor Characteristics	Postural/action tremors are more common (similar to asterixis), resting tremors (pill-rolling tremor) are rare	Classic resting tremor (pill-rolling tremor) is a core feature
Symptom Symmetry	Symptoms are typically symmetric	Symptoms typically start on one side of the body and gradually spread to the contralateral side, showing asymmetry
Response to Levodopa	Poor or no efficacy	Favorable response. It serves as a diagnostic criterion and main therapeutic approach
Imaging Features	Bilateral symmetric hyperintensity in the globus pallidus on T1-weighted cranial MRI, which is a characteristic manifestation of manganese deposition	MRI is usually normal or shows non-specific cerebral atrophy. DAT-PET scans demonstrate decreased function of the nigrostriatal pathway
Associated Diseases	Always accompanied by a history of CLD and recurrent HE	Usually not associated with liver disease

Hepatic myelopathy

Hepatic myelopathy (HM) is a severe neurological complication in the advanced stages of cirrhosis, characterized by an insidious onset, progressive progressive, purely motor spastic diplegia (without sensory deficits or bladder/bowel dysfunction) [307–309]. Its core pathogenic mechanism is consistent with HE and AHD, both centered on ammonia toxicity, combined with factors such as inflammation and nutritional disorders.

The pathogenesis is based on cirrhosis, portal hypertension and portosystemic shunt, which allow ammonia absorbed from the intestine to bypass the liver and enter the systemic circulation [310]. Excessive ammonia is taken up by spinal cord astrocytes and converted into glutamine, inducing astrocyte edema. Meanwhile, ammonia can directly damage mitochondrial function, triggering oxidative stress responses and energy metabolism failure, resulting in dual damage to spinal cord function [311]. In addition, the inflammatory response caused by the accumulation of intestinal toxins disrupts the blood-spinal cord barrier [312, 313], vitamin B deficiency weakens the neural repair capacity [314, 315], and manganese deposition exerts additional neurotoxicity [316, 317], which forms a vicious cycle with ammonia toxicity, leading to progressive demyelination and axonal loss of the corticospinal tract [318].

Conventional ammonia-lowering therapies have limited efficacy in HM. However, motor evoked potential examination can be used for early diagnosis [319]. Currently, liver transplantation is the only fundamental treatment that may halt or reverse the progression of HM [320, 321].

Other central nervous system complications

Intracranial Hemorrhage: In patients with liver cirrhosis, due to thrombocytopenia and coagulation disorders (such as vitamin K deficiency, abnormal synthesis of coagulation factors, etc.), the risk of spontaneous intracranial hemorrhage increases [322]. Clinically, this complication often manifests as sudden headache, hemiparesis, altered consciousness, and other stroke-like symptoms [323]. The core cause is the coagulation metabolic disorder resulting from impaired liver function.

Wilson's disease (WD): Also known as hepatolenticular degeneration, Wilson's disease is an inherited disorder of copper metabolism that affects both the liver and the CNS [324]. The core pathogenic mechanism is the mutation of the ATP7B gene, which leads to impaired copper excretion in the liver and copper accumulation that damages hepatocytes [324]. Excessive copper enters the circulation and deposits in brain regions such as the basal ganglia, causing neuronal degeneration through oxidative stress and mitochondrial damage [325, 326]. Clinically, patients may present with manifestations including liver cirrhosis, dystonia, and tremor [326, 327]. Treatment mainly involves copper chelators, zinc agents, and dietary restrictions. In extreme cases, liver transplantation may be necessary [328–331].

CNS diseases associated with cholestatic liver diseases: Patients with chronic cholestatic liver disease (such as primary biliary cholangitis) may experience symptoms such as fatigue, cognitive impairment, and sleep disorders [332, 333]. The core mechanism involves the toxic effects of long-term accumulation of bile acids and bilirubin, combined with the inflammatory response triggered by cholestasis, which interferes with the function of nerve cells [334]. Additionally, manganese accumulation leads to morphological abnormalities in the CNS [335]. Bile acids can also indirectly regulate neural function, promoting depressive-like behaviors [336].

Bilirubin encephalopathy: An acute neurological injury directly caused by unconjugated bilirubin, mainly occurring in neonates and only in adults with severe acute liver failure [337]. Its core mechanism involves unconjugated bilirubin crossing the underdeveloped BBB and depositing in the basal ganglia, hippocampus and other brain regions [338], and it can also regulate GABAergic signaling by activating the 5-hydroxytryptamine 3 A (5-HT3A) receptors in the spinal cord, leading to pain desensitization [83]. Clinically, the condition is divided into acute and chronic phases. Permanent neurological deficits may persist in the chronic phase, highlighting the importance of early monitoring and intervention.

Increased Risk of CNS Infection: Patients with advanced liver disease exhibit impaired systemic immune homeostasis, which is accompanied by compromised intestinal barrier integrity and leaky gut syndrome. Collectively, these factors significantly elevate susceptibility to CNS infections, such as meningitis [339–341]. The primary trigger is systemic immune dysfunction resulting from hepatic insufficiency, while intestinal barrier damage (leaky gut) serves as a critical factor that amplifies vulnerability to CNS infections [342]. Specifically, cirrhosis leads to intestinal dysbiosis, a reduction in the expression of tight junction proteins (e.g., ZO-1 and claudin-5) in intestinal epithelial cells, and enhanced intestinal permeability, further exacerbating immune dysfunction [343, 344]. Coupled with the impaired clearance capacity of the damaged liver, circulating pathogens and inflammatory mediators can penetrate the compromised BBB, ultimately facilitating the onset and progression of central nervous system infections [345].

Cognitive impairment: The liver secretes various hormones to act on key brain regions of the CNS, participating in the precise regulation of energy metabolism balance and insulin sensitivity. In patients with MASLD, hepatic steatosis leads to FGF21 resistance and impaired IGF-1 signaling, directly affecting synaptic plasticity and neuronal survival in the brain [346]. In addition, hepatic metabolic disorders can indirectly exacerbate cognitive impairment through multiple pathways. For example, abnormal lipid metabolism in the liver can activate the SNS, which in turn promotes central neuroinflammation, thereby disrupting brain microenvironment homeostasis, accelerating the abnormal deposition of β-amyloid protein, and increasing the risk of AD.

Peripheral nervous system issues caused by hepatic disorders

The impacts of liver diseases on the body extend far beyond these effects, as it also affects the PNS [3, 347]. The damage to peripheral nerves caused by liver diseases can be divided into two major pathways: direct and indirect [348–350]. Direct damage results from the accumulation of toxins such as ammonia during liver failure, which directly poisons the peripheral and autonomic nerves, causing symmetrical numbness in the hands and feet, orthostatic hypotension, and other manifestations [20, 351, 352]. Indirect damage is more common, it is associated with alcohol-induced neurotoxicity [353, 354], vitamin deficiency [355, 356], and immune vasculitis induced by viral infection [357–359].

Hepatic peripheral neuropathy

Hepatic peripheral neuropathy is the most common form of PNS involvement in liver disease. It typically presents as a mild, symmetrical distal sensory neuropathy and is particularly associated with alcoholic liver disease and cirrhosis [20]. The core pathogenic mechanisms are immunological abnormalities and nutritional metabolic disorders.

In terms of immunological abnormalities, peripheral neuropathy in patients with hepatitis B is mainly demyelinating damage, which is associated with immune dysfunction induced by the virus [360]. In contrast, hepatitis C induces mixed cryoglobulinemia, triggering immune-mediated vasculitis and resulting in ischemic nerve injury [361]. Liver diseases can affect the absorption, synthesis, and storage of various vitamins, particularly B vitamins, thereby interfering with the normal metabolism and nutritional support of the nervous system, ultimately leading to peripheral nerve dysfunction [362, 363].

Notably, HE often coexists with peripheral neuropathy clinically, which may be related to the combined effects of ammonia metabolism disorders and the accumulation of other neurotoxic substances, suggesting a certain overlap or synergistic effect in their pathogenesis.

Peripheral nervous system disorders associated with cholestatic liver disease

Chronic cholestatic liver disease can be accompanied by both CNS and PNS [364]. The core pathogenesis mechanism of its peripheral neuropathy is malabsorption of fat-soluble vitamins. Impaired bile excretion results in impaired absorption and metabolism of vitamin E, resulting in decreased serum vitamin E concentration. This causes axonal degeneration and demyelination of nerve fibers, ultimately leading to peripheral neuropathy [365]. This nutritional deficiency mechanism is consistent with that underlying CNS diseases associated with cholestatic liver diseases.

Neurological issues related to liver disease treatment

The neurological issues that arise during the treatment of liver diseases can be attributed to two core pathogenic mechanisms: direct neurotoxicity of drugs and treatment-induced metabolic disturbances. Different treatment approaches associated with distinct mechanisms of neural injury and varied clinical manifestations (Table 6), both of which are directly associated with physiological and metabolic alterations following liver-directed interventions.

Table 6.

Functional integration: physiological and pathological significance of neural regulation in the liver

Functional domain	Neuromodulatory mechanisms	Clinical/Experimental evidence
Glucose metabolism [2]	Sympathetic nerves promote glycogenolysis, parasympathetic nerves promote glycogenesis	Reduced hepatic glycogen synthesis after vagotomy
Lipid metabolism [366, 367]	Sympathetic nerves inhibit VLDL secretion and enhance fatty acid oxidation	Hepatic denervation leads to hyperlipidemia
Bile secretion [368]	Parasympathetic nerves enhance HCO3- secretion in cholangiocytes via ACh	Bile secretion decreases after parasympathetic denervation
Hepatic blood flow regulation [222, 368]	Sympathetic nerves constrict sinusoids, parasympathetic nerves dilate blood vessels	α-receptor blockers reverse reduced hepatic blood flow
Liver regeneration and immunity [369–371]	Sympathetic nerves regulate NKT cells and HSCs, influencing inflammation and tissue repair	β-receptor blockers improve liver fibrosis and regeneration
Behavior and cognition [372–374]	Hepatogenic inflammatory factors affect the CNS via the vagus nerve, inducing fatigue and anxiety	Abnormal functional connectivity in the basal ganglia observed in patients with CLD

Drug-related neurological side effects

Drug-related neurological side effects are the most common type of neurological problems caused by liver disease treatment. Different therapeutic drugs induce distinct manifestations and mechanisms of nerve injury. First are antiviral drugs used for treating viral hepatitis, such as interferon-α and nucleotide analogs. Interferon-α may damage peripheral nerves through immune-mediated inflammatory reactions, often leading to depression, anxiety, and in severe cases, mania or cognitive impairment, accompanied by symptoms such as limb weakness and diminished tendon reflexes [375]. Tenofovir may interfere with DNA synthesis in nerve cells, leading to axonal sensory neuropathy, which is manifested as numbness and tingling [376, 377]. Second, immunosuppressants used for autoimmune liver diseases or liver transplantation, such as tacrolimus, cyclosporine, and azathioprine, can also induce neurological effects. Tacrolimus and cyclosporine can inhibit the activity of neuronal calcineurin, affecting nerve signal transmission, and predisposing patients to tremors, headaches, and in severe cases, epilepsy [378, 379]. The metabolites of azathioprine exhibit cytotoxicity, directly damaging nerve fiber structures [378, 379].

Treatment-induced metabolic abnormality-associated neuropathy

Metabolic disorders arising during liver disease treatment, such as malabsorption or electrolyte imbalances, can also indirectly cause neurological damage.

In patients with cholestatic liver diseases, persistent impairment of bile excretion will affect the absorption of vitamin B12, folic acid and vitamin E, thereby increasing the risk of peripheral neuropathy [355, 380]. In cirrhotic patients with ascites, long-term use of diuretics may cause electrolyte imbalances such as hypokalemia and hypomagnesemia, which can trigger a series of neurological symptoms including confusion, epilepsy, psychiatric abnormalities, and even coma [372–374].

Neurological implications of hepatic denervation following liver transplantation

Liver transplantation is a critical therapeutic intervention for end-stage liver disease; however, a significant clinical challenge arises from the interruption of innervation between the donor liver and the recipient [381, 382]. This denervation has profound implications for graft function, metabolism, and potential long-term survival.

Firstly, the reinnervation of the liver following transplantation is severely limited. In human LT recipients, immunohistochemical staining for neural markers indicates that most small nerves disappear by day 5 post-transplantation, with recognizable nerve fibers reappearing in the portal area only 6 to 12 months later [383]. Even so, reinnervation is restricted to the portal area, with no regeneration of parenchymal nerve fibers observed [383, 384]. In rat models, exogenous reinnervation initiates approximately 3 days after transplantation; however, even at 4 months, the density of regenerating axons remains below normal levels [385]. Consequently, the transplanted liver remains largely denervated indefinitely.

Secondly, denervation disrupts the neural regulation of hepatic blood flow, bile secretion, and metabolic homeostasis. The absence of intrahepatic innervation limits the liver's capacity to respond rapidly to sympathetic nerve signals, resulting in impaired glycogenolysis, gluconeogenesis, reduced fatty acid oxidation, and decreased hepatocyte proliferation following injury [386, 387].

Furthermore, denervation is a potential factor influencing short-term graft survival. Clinical and experimental studies have demonstrated that denervated livers exhibit diminished regenerative capacity following ischemia–reperfusion injury (IRI) and are unable to coordinate immune and metabolic signals during acute stress [388]. The lack of neural input hampers the liver's ability to initiate anti-inflammatory and repair pathways, potentially exacerbating early transplant dysfunction and increasing the risk of acute rejection or poor transplant outcomes [389]. Recent studies suggest that engaging anti-inflammatory neuroimmune circuits through vagus nerve stimulation and other methods may represent a novel strategy to enhance organ vitality and graft survival [390].

In conclusion, denervation after liver transplantation is extensive and largely irreversible, with significant metabolic and immunological consequences. Although it may not be the primary cause of graft failure, denervation likely contributes to the multifactorial pathophysiology leading to reduced graft survival.

Hepatic disorders induced by neurological disorders

The precise regulation of the nervous system is crucial for maintaining liver functional homeostasis. Through signal transmission, it participates in modulating hepatocyte function, hepatic blood flow, and local immune-inflammatory responses [222]. The liver is directly and indirectly regulated by both the autonomic nervous system and CNS. When neurological function is disrupted, it can induce or exacerbate liver injury through various mechanisms, including neuroendocrine dysregulation, autonomic dysfunction, and impaired neuroimmune signaling. These disruptions may serve as potential triggers for the development and progression of various liver diseases, such as metabolic-associated fatty liver disease and autoimmune liver conditions [17].

Specifically, autonomic nervous system imbalance, manifested as sympathetic overactivation and parasympathetic inhibition, can directly promote HSCs activation via α1-AR and β2-AR, thereby accelerating hepatic fibrosis progression [391]. Secondly, dysfunction of the HPA axis leads to abnormal glucocorticoid rhythms, which subsequently disrupts lipid metabolism in hepatocytes and promote hepatic steatosis and insulin resistance [366]. Additionally, the CNS can regulate intestinal barrier function and microbiota composition through the vagus nerve, affecting the entry of enterogenous endotoxins into the liver and activating the inflammatory responses [77]. Clinical studies have shown that the incidence of MASLD is significantly higher in patients with mental illnesses such as depression and anxiety disorders, while autonomic neuropathy is independently associated with more severe liver fibrosis [26, 392]. These findings provide new theoretical foundations and potential therapeutic targets for the prevention and treatment of liver diseases from the perspective of neural regulation.

Imbalance in hepatic blood flow and metabolic regulation

Autonomic dysfunction caused by anxiety, depression, or chronic stress, through sympathetic hyperactivity, parasympathetic inhibition, and neurotransmitter imbalance, lead to insufficient hepatic blood perfusion and metabolic disorders, laying a pathological foundation for CLD. Hyperactivity of the SNS leads to excessive release of NE from sympathetic nerve terminals, which in turn activates HSCs, promotes collagen deposition, and accelerates liver fibrosis [17]. Conversely, inhibition of the parasympathetic nervous system weakens the protective effects on hepatocytes and increases the risk of autoimmune hepatitis [369]. Dopamine can regulate hepatic blood flow and bile secretion [368], while serotonin influences lipid metabolism and cell proliferation in the liver [367]. These neurotransmitters regulate hepatic blood flow, bile secretion, and lipid metabolism through a complex network. Dysregulation of this network can lead to persistent hepatic dysfunction, ultimately promoting the initiation and progression of liver injury.

Decline in hepatic repair and detoxification capacities

Peripheral nerves regulate hepatic blood supply, bile excretion, and nutrient transport by innervating hepatic blood vessels, bile ducts, and hepatocytes. Their damage can lead to a decline in liver repair and detoxification capabilities. The core mechanisms are the decreased neurotrophic factors and abnormal innervation of the bile ducts. Peripheral nerve damage results in a decrease of neurotrophic factors such as nerve growth factor (NGF), which reduces the proliferation and repair ability of hepatocytes [393]. Abnormal bile duct innervation leads to poor bile excretion, causing cholestasis and affecting hepatic detoxification [364, 394, 395].

Mechanistically, NGF binds to its high-affinity tyrosine kinase receptor TrkA on hepatocytes, activating two key downstream signaling cascades: the PI3K/Akt pathway and the ERK/MAPK pathway [396, 397]. The PI3K/Akt pathway represents the predominant mechanism, as PI3K inhibitors completely abolish NGF-induced proliferation, while MEK inhibitors only partially block this effect [396]. Additionally, NGF exerts hepatoprotective effects by activating Akt, which increases the Bcl-2/Bax ratio and suppresses oxidative stress-induced apoptosis, thereby indirectly supporting hepatocyte survival and regenerative capacity [398].

Hepatic lipid accumulation

Chronic stress activates the HPA axis, leading to elevated cortisol (CORT) levels [399]. High CORT levels suppress insulin signaling in hepatocytes, driving hepatic lipid accumulation and the development of MASLD [400, 401].

Beyond the HPA axis, functional abnormalities in central nervous nuclei can also directly regulate liver metabolism. For instance, dysfunction of the ventromedial hypothalamus (VMH) reduces its inhibitory effect on hepatic sympathetic nerve activity, thereby promoting intrahepatic lipid deposition [402].

Furthermore, a pathological vicious cycle and cross-organ pathological signaling exist between the CNS and the liver. In HE, disrupted glutamate metabolism not only induces neurotoxicity but also exacerbates hepatic inflammation and lipid deposition through peripheral–central signaling feedback, forming a self-perpetuating cycle of deterioration [403].

The role of neural regulation in liver diseases

The nervous system plays a crucial bidirectional regulatory role in liver diseases and regeneration, with the core mechanism centered around the imbalance between the sympathetic and parasympathetic nerves. In MASLD, sympathetic overactivation exacerbates hepatic lipid accumulation and fibrosis [18, 404, 405], whereas parasympathetic nerve activity helps alleviate lipid accumulation and inflammatory responses in the liver [220, 406]. α/β-adrenergic receptor blockers, such as carvedilol, can reverse CCl₄-induced liver fibrosis, highlighting the therapeutic potential of neuromodulatory strategies [407–409]. After partial hepatectomy, the sympathetic nerve can drive hepatocyte proliferation [410], while the vagus nerve is involved in liver regeneration, and nerve injury inhibits regeneration and exacerbates hepatocyte apoptosis [370, 371]. These findings not only reveal the central role of the nervous system in maintaining hepatic homeostasis, but also provide a theoretical basis for developing nerve-targeted intervention strategies for liver diseases. These findings further confirm the central role of neural regulation in maintaining hepatic homeostasis.

The liver and nervous system form a complex bidirectional regulatory network. Abnormal liver metabolism can affect neurological function through systemic inflammation, oxidative stress, metabolic disturbance, neurotoxin accumulation, and BBB disruption, while neural dysregulation can also feedback regulate hepatic immune and metabolic homeostasis via neuroendocrine mechanisms.

Alcohol misuse as a dual hit to the liver-nervous system axis

Alcohol misuse drives concurrent neuroinflammation and hepatic steatosis via TLR4/NF-κB signaling in both microglia and Kupffer cells [411]. This mechanism underpins the high comorbidity between alcohol use disorder (AUD) and ALD, highlighting the urgent need for dual-targeted therapeutic strategies [412, 413].

Glucagon-like peptide-1 receptor agonists (GLP-1RAs), initially developed for the treatment of diabetes and obesity, have now emerged as promising therapeutic agents for both AUD and ALD. A targeted trial simulation involving 16,080 individuals with harmful alcohol consumption demonstrated that GLP-1RAs treatment was associated with a 30% reduction in composite liver outcomes and decreased all-cause mortality [414]. A recent meta-analysis further validated that GLP-1RAs administration is strongly correlated with reduced alcohol-related adverse events and improved liver-related endpoints [415].

Mechanistically, GLP-1RAs act on the VTA and nucleus accumbens (NAc) to suppress alcohol-evoked dopamine release and alleviate alcohol craving. Meanwhile, they directly ameliorate hepatic steatosis, inflammation, and fibrosis through AMPK activation and SREBP-1c inhibition in the liver [416–419].

Therapeutic strategies, clinical significance, and future challenges of the liver-nervous system axis

The bidirectional regulatory network between the liver and the nervous system presents a novel concept for cross-organ collaborative intervention in treating liver-related and neurological diseases. Currently, liver-targeted interventions for neurological conditions have been preliminarily implemented and have yielded positive results. The therapeutic potential of neuroregulation strategies in liver diseases has been substantiated through numerous clinical trials. Furthermore, the development of drugs targeting the liver-nervous system axis is progressively advancing towards clinical translation. Therefore, it is crucial to focus on treatment strategies that demonstrate clear clinical feasibility, considering their clinical safety, efficacy advantages, and existing evidence from evidence-based medicine. A systematic analysis of the bottlenecks and challenges in current clinical translation, alongside an outlook on future development directions, holds significant theoretical and practical value for promoting clinical implementation in this field and enhancing the treatment of related diseases.

Application of hepatic interventions in nervous system diseases

The liver, as a central metabolic organ and immune regulatory hub, has seen improvements in its function or targeted interventions emerge as auxiliary treatment approaches for neurological disorders. By focusing on the regulation of liver function, repairing metabolic pathways, or eliminating pathological mediators, the pathological processes of the nervous system can be reversed and improved, demonstrating significant clinical feasibility.

Regulation of hepatic metabolic homeostasis reduces neuroinflammation risk

Neurological disorders are often accompanied by hepatic metabolic abnormalities such as hepatic steatosis and insulin resistance. Improving hepatic metabolic homeostasis has been proven to effectively reduce CNS damage, with mature intervention methods and clear safety profiles [1]. Clinically, dietary control (low-fat and low-sugar diets), regular exercise combined with drug intervention (such as metformin, an insulin sensitizer, and atorvastatin, a lipid-lowering agent) can significantly reduce intrahepatic lipid accumulation, inhibit the release of hepatic inflammatory factors, thereby attenuating central inflammatory responses and alleviating neurological dysfunction [420, 421]. Furthermore, hormones secreted by the liver, such as FGF21, play a critical role in neural repair. Clinical studies have confirmed that improving MASLD can upregulate the normal signaling of FGF21, thereby restoring synaptic plasticity in the brain, promoting neuronal survival, and ultimately alleviating cognitive impairment [422]. This intervention pathway has been preliminarily verified in patients with MASLD combined with cognitive impairment [288].

Hepatic immunomodulation repairs neurological function

The detoxification and immunomodulatory functions of the liver are key to blocking the spread of pathological substances and mitigating neuroinflammation, playing a vital role in the recovery of neurological disorders. For example, hepatoprotective drugs that repair hepatocellular damage and enhance detoxification capacity can effectively reduce sustained injury to the CNS [423]. Simultaneously, modulating the hepatic immune microenvironment to suppress the release of pro-inflammatory cytokines can interrupt the "liver inflammation–neural injury" signaling axis, thereby alleviating neurological damage [424, 425].

Intervention potential of neuromodulation in liver diseases

Neuromodulation has emerged as a pivotal strategy for liver disease intervention. Its core principle lies in regulating the autonomic nervous system and implementing multi-level interventions in neural pathways to restore hepatic microenvironmental homeostasis. Currently, many strategies have entered clinical application or phase III clinical trials. Compared with traditional liver protection and antiviral treatments, it has the advantages of minimally invasive, strong targeting, and the potential for synergistic enhancement. It provides diversified therapeutic options for MASLD, liver fibrosis, etc. (Table 7).

Table 7.

Neuromodulation therapy for different liver diseases

Liver disease types	Core regulatory targets	Main intervention methods	Clinical Status
MASLD [426, 427]	Sympathetic nerve GLP-1 receptor, β-adrenergic receptor	Pharmacological interventions: GLP-1 receptor agonists (semaglutide, liraglutide) [428]; β-adrenergic receptor blocker (metoprolol) [429]; Dietary intervention: sour taste stimulation (lemon, vinegar)	Semaglutide is received FDA approval for the treatment of MAFLD [428]; Liraglutide is also FDA-approved for weight loss, with a phase III MASH indication [430]; Acid stimulation is devoid of side effects and can serve as an auxiliary intervention [431]
Liver fibrosis/cirrhosis [432–434]	β-adrenergic receptor	Pharmacological interventions: β-adrenergic receptor blockers (propranolol [434], carvedilol [435])	Propranolol is a first-line medication for the secondary prevention of liver cirrhosis, as it can delay the progression of liver fibrosis [436]; Carvedilol is recognized as a novel non-selective β-blocker that is preferred for patients with liver cirrhosis [435]
HCC [437]	β-adrenergic receptor, 5-HT transporter	Pharmacological interventions: β-adrenergic receptor blocker (propranolol) [191]; Selective serotonin reuptake inhibitor (fluoxetine) [438, 439]	Propranolol is currently undergoing phase II clinical trials [440, 441]; Fluoxetine is in a phase I clinical trial for colorectal cancer [191]
Pruritus in liver diseases	Ileum bile acid transporter, κ-opioid receptor	Pharmacological interventions: Linerixibat [442–444], Nalfurafine [445–447]	Linexibatide is approved by the for the alleviation of cholestatic pruritus in primary biliary cholangitis [442–444]; Nefuraphine is FDA-approved for blocking itch signaling [445–447]

Inhibition of sympathetic overactivity improves hepatic blood flow perfusion and alleviates liver fibrosis

Sympathetic overexcitation is a key driver of liver disease progression. β-adrenergic receptor blockers (such as propranolol, carvedilol, and metoprolol), as core intervention drugs, can inhibit the activation of HSCs and delay the progression of liver fibrosis [434, 435]. At the same time, β-adrenergic receptor blockers can also inhibit the growth of HCC cells by inhibiting the ADRB2 signaling pathway, and enhance the anti-tumor sensitivity of sorafenib, a conventional targeted drug for HCC [448]. Among them, propranolol has been established as the first-line drug for secondary prevention of liver cirrhosis, and studies evaluating its use in HCC have entered Phase II clinical trials [436, 440, 441]. Carvedilol, as a new non-selective β-adrenergic receptor blocker, is considered the preferred intervention for patients with liver cirrhosis due to its antioxidant, anti-inflammatory properties and milder hemodynamic effects [435]. In addition to pharmacological interventions, minimally invasive sympathetic blockade techniques (such as stellate ganglion block and celiac plexus block) can effectively disrupt abnormal sympathetic signaling to the liver [437, 449, 450]. This disruption inhibits postoperative sympathetic storm, reduces inflammatory responses, enhances liver blood flow, and promotes liver regeneration [449]. Consequently, these techniques decrease the risk of liver function deterioration following liver resection and transplantation, thereby achieving perioperative liver protection [437, 449].

Furthermore, for MASLD, the GLP-1 receptor in the sympathetic regulatory pathway also represents an important intervention target.

Currently, the GLP-1 receptor agonist semaglutide has received FDA approval for clinical treatment of MASLD [428]. Additionally, another GLP-1 receptor agonist, liraglutide, has been approved for weight loss indications, and studies assessing its efficacy in treating MASH have progressed to Phase III clinical trials, thereby offering novel targeted strategies for the management of MASLD [430].

Activation of the vagus nerve improves hepatic inflammation and liver fibrosis

The vagus nerve, as a key component of the autonomic nervous system, plays a critical role in regulating hepatic inflammation and liver fibrosis through the cholinergic anti-inflammatory pathway, making it an important target for neuromodulatory interventions in liver diseases. Studies have confirmed that vagus nerve can inhibit the release of pro-inflammatory factors from KCs and downregulate the TGF-β pathway through the cholinergic anti-inflammatory pathway to alleviate liver fibrosis and inflammation [31, 221, 388, 432, 433, 451]. Furthermore, the vagus nerve modulates intrahepatic sympathetic tone, improving hepatic microcirculatory dysfunction and creating a favorable environment for hepatocyte repair [452]. Recent clinical trials have demonstrated that transcutaneous vagus nerve stimulation shows potential in reducing liver enzyme levels in patients with HIRI, though its long-term efficacy still requires validation through large-scale studies [453, 454].

Multi-target neuromodulation improves hepatic lipid metabolism and alleviates liver injury

The hypothalamus regulates hepatic lipid metabolism through neurohumoral pathways, targeting anorexia and appetite peptides, while integrating insulin and leptin signaling pathways, which effectively improve abnormal hepatic lipid accumulation [426, 427, 455–457].

Neurotrophic factors such as neuregulin 4 (NRG4) can regulate the function of immune cell populations in the tumor microenvironment, alleviate the severity of MASH, and serve as potential therapeutic targets for metabolic-related liver diseases [458–460].

The 5-HT transporter plays a crucial role in regulating 5-HT signaling and is abnormally expressed in liver fibrosis and HCC [268]. Fluoxetine, a well-known 5-HT inhibitor currently undergoing Phase I clinical trials for colorectal cancer, has demonstrated anti-HCC effects by inhibiting the ERK/NF-κB signaling pathway and inducing apoptosis in tumor cells [191].

Acid stimulation triggers neural signals by activating the taste receptors in the tongue, inhibiting the TAFA2-CCR2 pathway, which reduces macrophage activation and significantly decreases ALT and AST levels following hepatectomy [222]. Currently, this clinical trial has been registered in the Chinese Clinical Trial Registry and has been promoted in clinical practice across multiple hospitals, effectively improving the success rate of liver surgeries and enhancing patient recovery outcomes (Fig. 7).

Fig. 7.

Stimulation of sour taste neuron signals alleviates hepatic ischemia–reperfusion injury. Acidic taste inhibits the expression of neuron-derived TAFA2, reduces its interaction with the CCR2 receptor on hepatic macrophages, thereby attenuating macrophage activation and regulating the neuro-immune axis to alleviate HIRI

The primary regulatory targets for liver disease-related pruritus are the ileal bile acid transporter and the κ-opioid receptor, which are primarily addressed through pharmacological interventions. Commonly used drugs include linaclotide and nalfurafine [442–447]. Among these, linaclotide has received FDA approval and can reduce bile acid levels by inhibiting enterohepatic bile acid circulation, thereby indirectly alleviating pruritus associated with cholestasis in primary biliary cholangitis [442–444]. Nalfurafine has also been approved by the FDA and functions by activating κ-opioid receptors in the spinal cord and brain to block pruritus signals [445–447].

Pharmacological applications targeting the liver-nervous system axis

Repurposing of existing drugs

In recent years, some drugs have shown potential in the treatment of the liver-nervous system axis through repurposing [461] (Table 8).

Table 8.

New uses of old drugs targeting the liver-nervous system axis

Drugs	Original indication (Old Use)	Novel Liver–Brain Axis Mechanism	New Application Directions
Pioglitazone	Type 2 diabetes mellitus (T2DM)	Activates PPARγ [462], exerts anti-inflammatory and anti-steatotic effects in the liver [463, 464] and anti-inflammatory, neuron-protective effects in the brain [465–467]	Recommended as Grade 1A-level in the Chinese MASLD-RRB guidelines (2026 edition)
Fenofibrate	Hypertriglyceridemia	Activates PPARα [468–470], reduces hepatic lipid accumulation [471]	Not yet approved for use in Htg populations
Rifaximin	Intestinal infection, diarrhea	Modulates the gut–liver–brain axis, reduces blood ammonia and neurotoxins [472–474]	Approved as a first-line treatment for primary prophylaxis of HE in Europe and the United States [472, 474–476]
Lactulose	Chronic functional constipation	Acidifies the intestine and lowers blood ammonia levels [477]	Approved as a first-line treatment for HE [472]
Metformin	T2DM	Activates AMPK [478, 479], improves hepatic steatosis [478, 479], and exerts antioxidant and neurorepair-promoting effects in the brain [480]	Not yet approved for use in non-T2DM populations
Baicalin	Traditional Chinese Medicine	Reduces liver inflammation [481], inhibits neuronal apoptosis [482], and alleviates neuroinflammation [483]	Approved in China for adjuvant treatment of acute/chronic hepatitis and persistent hepatitis [484]
Curcumin	Traditional Chinese Medicine	Reduces hepatic lipid deposition [485, 486] and inhibits central neuroinflammation [487]	Approved in China as a hepatoprotective health food [488]

Both pioglitazone and metformin are well-established clinical drugs for the treatment of T2DM. Recent research on drug repurposing has revealed their significant clinical value in the management of liver and nervous system diseases. Pioglitazone functions by activating peroxisome proliferator-activated receptor γ (PPARγ) [462]. In the liver, it has been shown to significantly improve hepatic steatosis, reduce liver inflammation, and delay the progression of liver fibrosis [464, 489–491]. Its efficacy is classified as a grade 1A-level recommendation by the Chinese MASLD-RRB guidelines (2026 edition). In the context of the nervous system, pioglitazone effectively reduces neuroinflammation and slows cognitive decline, thereby offering a potential strategy for the intervention of neurodegenerative diseases [465–467]. Metformin primarily exerts its hepatic effects by activating the AMPK pathway, which improves insulin resistance, inhibits hepatic lipid synthesis, and reduces inflammatory responses, thus significantly alleviating hepatic steatosis and delaying liver fibrosis [478, 479]. In the nervous system, metformin can cross the BBB to provide antioxidant, anti-inflammatory, and neuroprotective effects, thereby opening new avenues for the preservation of neural function [480].

Fenofibrate, a classical peroxisome PPARα agonist originally indicated for hypertriglyceridemia, has recently seen expanded applications in the treatment of liver diseases. This drug significantly improves hepatic steatosis in patients with MASLD by activating the PPARα pathway, promoting hepatic fatty acid oxidation, reducing hepatic lipid deposition, and alleviating hepatic inflammation. Consequently, it offers a new option for the adjunctive treatment of MASLD [471, 492].

Rifaximin, an intestine-targeted antibiotic, plays a crucial role in preventing HE by inhibiting the proliferation of intestinal ammonia-producing bacteria and reducing the generation and absorption of blood ammonia [472–474]. It has been approved in Europe and the United States as a first-line treatment for the primary prevention of HE in patients with liver cirrhosis [472, 474–476]. Similarly, lactulose, which was originally indicated for the treatment of constipation, has been shown through subsequent clinical studies to reduce intestinal ammonia production and absorption by acidifying the intestinal lumen, thereby lowering blood ammonia levels [477]. Lactulose is currently recognized as a first-line treatment for HE [472].

Baicalin and curcumin, as active components of traditional Chinese medicine, demonstrate distinct advantages in protecting the liver–nervous system axis. Baicalein exerts a synergistic protective effect on the liver-brain axis by reducing liver inflammation [481], inhibiting neuronal apoptosis [482], and alleviating neuroinflammation [483]. It has been approved for the adjuvant treatment of both acute and chronic hepatitis, as well as protracted hepatitis, in China [484]. Curcumin has been shown to reduce hepatic lipid deposition [485], and inhibit central neuroinflammation [487], indicating its potential protective value for the liver–brain axis. Currently, curcumin is approved as a health food in China to aid in the protection against chemical liver injury [488].

The repurposing of these drugs not only reduces research and development costs but also provides a rapid pathway to clinical application, thereby offering new strategies for the treatment of diseases affecting the liver-nervous system axis.

Research progress of novel targeted drugs

The development of novel targeted drugs emphasizes neuroprotective hepatogenic substances and their receptor signaling pathways. Currently, several drugs have progressed to phase II clinical trials, where preliminary evaluations have confirmed their safety and efficacy. Compared to existing therapies, these novel agents exhibit enhanced targeting capabilities and dual effects on liver and neuroprotection, although their clinical applicability necessitates further validation [493, 494].

FGF21, as a liver-derived hormone, has been proven to cross the BBB and directly act on the CNS, improving metabolic disorders and neuroinflammation [346]. At the same time, it can also increase hepatic sympathetic nerve activity and reduce the level of hepatobiliary steroids [495, 496]. Presently, FGF21 analogs (Pegozafermin, Efruxifermin, Efimosfermin) have all entered phase II clinical trials [493, 494, 497], exhibiting significant efficacy and favorable safety profiles, with some progressing to phase III trials, marking a significant breakthrough for both MASLD and AD [110, 498].

In addition, the regulatory role of brain-derived neurotrophic factor (BDNF) in the liver–brain axis has garnered attention. Research has found that the liver can promote neuronal survival and synaptic plasticity by secreting exosome-carried BDNF precursors, providing new therapeutic targets for neurodegenerative diseases [499, 500]. These advancements have laid a theoretical foundation for developing drugs with both hepatoprotective and neuroprotective effects.

Optimization of drug delivery systems

To overcome the limitations of the BBB, the optimization of drug delivery systems has become a key strategy for enhancing drug bioavailability in the CNS. Nanocarriers, such as liposomes and polymer nanoparticles, can significantly enhance the BBB penetration efficiency of drugs through surface modification [501, 502]. For example, ionized lipid nanoparticles designed based on naturally active molecules can efficiently cross the BBB via the dopamine D3 receptor-mediated endocytosis pathway, and their therapeutic efficacy has been validated in several animal models of brain disorders including AD and glioma [503]. Fang et al. developed cellulose-mimetic chitosan microfibers loaded with quercetin selenium nanoparticles, which possess natural targeting ability and low immunogenicity, enabling simultaneous targeting of both liver and brain tissues [504]. By regulating inflammation and metabolism through the gut–liver–brain axis, this system alleviated cognitive impairment in AD [504]. In addition, Morrow et al. designed pH-responsive core-cross-linked micelle-encapsulated ferrostatin-1 (CCM Fer-1), which enables sustained release of the ferroptosis inhibitor Fer-1 and shows excellent anti-ferroptotic activity in various in vitro cell models [505]. This system can be transported across the BBB in vitro, and in view of the fact that the mechanism of ferroptosis is involved in various pathological processes such as neurodegeneration, inflammation, liver damage, and more, such delivery systems provide new ideas for the intervention of liver-nervous system axis diseases [505].

Current challenges and future directions

Although liver intervention has provided novel insights for treating neurological disorders, and neuromodulation demonstrates great potential in liver disease therapy, there are still numerous difficulties and challenges to overcome.

First, the specific molecular mechanisms underlying liver–nervous system interactions remain inadequately elucidated. The regulatory networks exhibit heterogeneity across various disease contexts, resulting in individual differences in the efficacy of existing intervention strategies, which complicates the achievement of precise targeted therapies. Therefore, it is essential to analyze the specific pathways under different disease backgrounds through large-sample clinical studies.

Second, the clinical applicability of existing interventions necessitates further optimization. The long-term efficacy of certain neuromodulation techniques, such as transcutaneous vagus nerve stimulation, remains ambiguous. Clinical trials for novel targeted drugs are constrained by small sample sizes, short follow-up durations, and an incomplete clinical evidence chain. Additionally, some agents are associated with hepatotoxicity or central side effects, for instance, certain β-blockers may induce bradycardia, underscoring the urgent need to develop more precise targeted preparations that balance efficacy and safety.

Moreover, significant differences in liver nerve distribution, nerve function, and metabolic status among individuals lead to poor uniformity in treatment responses to the same intervention strategies. Personalized treatment strategies have yet to be established. The absence of multi-center, large-sample, long-term follow-up clinical trials in this field hampers the systematic evaluation of the long-term efficacy and safety of various intervention strategies, thereby limiting their clinical promotion and application.

In the future, with the advancement of multi-omics technologies and cross-organ imaging techniques, further preclinical and clinical studies will elucidate the mechanisms of the liver–nervous system axis. These developments are expected to drive the creation of individualized targeted therapies, opening new treatment avenues for liver diseases and related neurological disorders.

Discussion and summary

Research on the liver-nervous system axis has revealed a multi-level regulatory network of cross-organ signal interactions, driving a paradigm shift in disease prevention and treatment from single-organ targeted therapies to multi-system coordinated precision regulation. Current clinical and basic evidence has preliminarily confirmed the translational value of this axis: ammonia-lowering interventions can effectively improve neurocognitive impairment in HE; β-blockers can delay the progression of liver fibrosis; vagus nerve stimulation can alleviate MASH-related liver inflammation, and liver transplantation can reverse the structural and functional damage to the CNS mediated by abnormal manganese deposition.

At the mechanistic level, recent breakthrough studies have further elucidated the specificity of inter-organ signaling circuits. For instance, Yan et al. identified a vagus nerve–liver–cortex circuit, wherein the DMV transmits efferent signals under stress, prompting the liver to produce Lipocalin-2 (LCN2), which subsequently suppresses prefrontal cortical activity and induces anxiety-like behaviors [506]. Meanwhile, Rose et al. reported that FGF21 acts directly on central glutamatergic neurons to promote hepatic triglyceride reduction and reverse fibrosis [346]. These findings not only provide novel targets for the collaborative prevention and treatment of metabolic liver diseases and neuropsychiatric comorbidities but also offer a referential research paradigm for studies on other cross-organ interaction systems, such as the gut-liver axis and brain-gut axis.

Despite rapid advancements in this field, existing research continues to present substantial contradictory evidence and critical scientific bottlenecks, failing to establish a unified theoretical framework. This has emerged as a core obstacle hindering clinical translation. First, there is significant divergence in anatomical evidence regarding hepatic cholinergic innervation. Multiple studies have failed to consistently detect stable cholinergic nerve fibers within the hepatic parenchyma; however, they have confirmed that ACh can significantly regulate hepatic inflammation, metabolism, and fibrosis. It remains unclear whether this effect originates from nerve terminal release or non-neurogenic paracrine signaling. This discrepancy between anatomical and functional findings may be closely related to the insufficient sensitivity of detection techniques, regional heterogeneity in nerve fiber distribution, and neural remodeling under pathological liver conditions. Second, there is a significant mismatch between the expression and functional effects of neurotransmitter receptors, and the same receptor subtype often shows completely opposite effects. The α1-AR can promote the activation and fibrosis of HSCs in most cases, but it can protect hepatocytes and inhibit apoptosis in the early stage of liver regeneration. The traditional view holds that the activation of β2-AR has an anti-fibrotic effect, but recent studies have found that its expression abnormally increases in liver cirrhosis and is associated with persistent inflammation. 5-HT usually aggravates fibrosis through the 5-HT2B receptor, but in cholestatic liver disease, it protects bile duct cells and reduces bile duct fibrosis. Such contradictory results are not accidental, and are mainly related to differences in animal model strains, stage specificity of pathology, concentration and dose dependence of neurotransmitters. Currently, there is still a lack of systematic comparative studies to clearly explain them. Third, the regulatory effects of sympathetic and vagal nerves are highly context-dependent. The same intervention measures often exhibit reversed effects across different disease models, such as liver regeneration, liver fibrosis and MASH. For instance, vagus nerve stimulation can demonstrate anti-inflammatory and anti-fibrotic effects in certain models, while exacerbating biliary epithelial proliferation and fibrosis progression in bile duct ligation models. Neuropeptides, including SP and neuropeptide Y, frequently display contradictory phenotypes, contributing to both liver protection and liver injury. Fourth, the causal link between neuroinflammation and behavioral changes associated with HE remains unclear in animal models. Although numerous studies have observed astrocyte swelling, microglia activation, elevated pro-inflammatory factors in animal models of HE, along with behavioral manifestations such as reduced motor activity, memory impairment, and anxiety-like behaviors, there is still a lack of direct causal evidence regarding whether neuroinflammation is the driving factor or merely a concomitant phenomenon of these behavioral abnormalities. Fifth, the clinical translation of neural regulation strategies encounters numerous unavoidable challenges. Although interventions like vagus nerve stimulation and sympathetic blockade have shown promising anti-inflammatory and anti-fibrotic potential in preclinical animal studies, they are hindered by critical limitations such as insufficient in vivo targeting, non-specific activation of neural signals, and potential cardiovascular and gastrointestinal side effects. Furthermore, most intervention studies remain in the preclinical phase, lacking large-sample randomized controlled clinical trials to validate their efficacy and safety, which complicates the direct advancement to clinical application.

Future research should abandon generalized descriptions and focus on key scientific questions and core translational bottlenecks. By utilizing single-cell multi-omics, spatial transcriptomics, and neural tracing techniques, researchers can quantitatively characterize the dynamic remodeling of liver-neural regulatory circuits across various stages of liver disease. This approach will clarify the concentration-dependent effects, cellular targets, and spatiotemporal specificity of controversial neurotransmitters. Furthermore, it is essential to develop novel organ- or cell-selective neuromodulatory tools, including optogenetics, chemogenetics, and targeted neurotransmitter prodrugs, to mitigate non-specific side effects and enhance intervention precision and safety. Additionally, prospective clinical cohort studies and early-phase clinical trials should be conducted, prioritizing the validation of intervention targets that exhibit manageable safety profiles and relatively clear mechanisms. This strategy will facilitate the steady translation of basic research findings into clinical protocols. Lastly, constructing multi-omics integrated models of cross-organ interactions is crucial. This involves systematically coupling the liver-nervous system axis with regulatory networks such as the gut-liver axis and brain-gut axis to dissect key nodes that underlie multi-organ comorbidity and disease progression.

In summary, the liver-nervous system axis serves as a pivotal hub linking metabolic disorders with neuropsychiatric diseases. A comprehensive investigation into this axis can elucidate the fundamental mechanisms underlying liver diseases and their neurological comorbidities, thereby fostering the development of innovative combined treatment strategies. Furthermore, it can provide both theoretical and technical support for precision medicine that is grounded in cross-organ interactions. To facilitate the transition of this field from basic research to clinical application, it is essential to critically integrate conflicting evidence, focus on core mechanisms, and address translational barriers in a targeted manner. This approach will enable the collaborative prevention and control of multi-organ diseases.

Authors’ contributions

Conceptualization, Z.-J., L.-W. and L.-S.H.; funding acquisition, Y.-Y.M. and Z.-J.; project administration, H.-D.L. and H.-Y.B.; writing—original draft, Z.-J.; writing—review & editing, Y.-Y.M.; All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (Grant number 82570746,82272421), Changzhou’s 14th five-year plan project to train high-level health professionals (Grant number 2022CZLJ027), Scientific Research Project of Jiangsu Commission of Health (K2024037), Medical Research Project of Jiangsu Health Commission (ZQ2025032), Open project of Jiangsu Provincial Key Laboratory of Key Laboratory of Laboratory Medicine (Grant number JSKLM-T-2025–01, JSKLM-Z-2025–01, JSKLM-Y-2024–005).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

As this manuscript pertains to a review article, so conventional requirements of ethical approval and participant consent are not applicable in this context.

Consent for publication

All authors grant their consent for the publication.

Competing interests

The authors declare no competing interests.