Authors: Fernanda Raya Tonetti (1Department of Medicine, University of California San Diego, La Jolla, CA, USA), Alvaro Eguileor (1Department of Medicine, University of California San Diego, La Jolla, CA, USA), Marko Mrdjen (2Department of Molecular Medicine, Case Western Reserve University, Cleveland, OH; 3Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, OH.), Vai Pathak (4Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, Ohio, USA), Jared Travers (3Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, OH.; 5Department of Gastroenterology and Hepatology, University Hospital, Cleveland OH), Laura E Nagy (2Department of Molecular Medicine, Case Western Reserve University, Cleveland, OH; 3Department of Inflammation and Immunity, Cleveland Clinic, Cleveland, OH.; 6Department of Gastroenterology and Hepatology, Cleveland Clinic, Cleveland OH), Cristina Llorente (1Department of Medicine, University of California San Diego, La Jolla, CA, USA)
Categories: Article, Microbiome, therapeutic strategies, intestinal immune system, intestinal epithelial cells, enteric nervous system, single cell RNA sequencing, multiomics, hepatocytes, Kupffer cells, genome wide association studies (GWAS), inflammation, steatosis, fibrosis
Source: Hepatology (Baltimore, Md.)
Authors: Fernanda Raya Tonetti, Alvaro Eguileor, Marko Mrdjen, Vai Pathak, Jared Travers, Laura E Nagy, Cristina Llorente
The growing recognition of the role of the gut microbiome’s impact on alcohol-related diseases, especially in alcohol-associated liver disease, emphasizes the need to understand molecular mechanisms involved in governing organ-organ communication in order to identify novel avenues to combat alcohol-related diseases. The gut-liver axis refers to the bidirectional communication and interaction between gut and liver. Intestinal microbiota play a pivotal role in maintaining homeostasis within the gut-liver axis and this axis plays a significant role in alcohol-associated liver disease. The intricate communication between intestine and liver involves communication between multiple cellular components in each organ that enable them to carry out their physiological functions. In this review, we focus on novel approaches to understanding how chronic alcohol exposure impacts the microbiome, and individual cells within the liver and intestine, as well as the impact of ethanol on the molecular machinery required for intra- and inter-organ communication.
Chronic alcohol consumption can disrupt the balance of the gut microbiota leading to the breakdown of tight junctions in intestinal epithelial cells (IECs) (1) (2). Consequently, increased permeability allows for the passage of molecules and microorganisms from the intestinal lumen, which eventually translocate to the liver through the portal vein (2). This translocation of bacterial products triggers inflammatory responses in the liver, involving resident hepatic macrophages, Kupffer cells (KC), and other infiltrating immune cells (3). Bacterial toxins contribute to hepatocyte injury and death while persistent inflammation activates hepatic stellate cells (HSC), which, in turn, become significant contributors to fibrosis (4). Furthermore, due to collagen production by activated HSCs, liver sinusoidal endothelial cells (LSEC) undergo capillarization, including the loss of fenestrations and a shift toward a vascular phenotype (5). These changes collectively contribute to the pathogenesis of liver injury and fibrosis in alcohol-associated liver disease (ALD) (6). Unraveling the intricacies and molecular mechanisms behind the disruption of the gut-liver axis in ALD is pivotal for crafting innovative strategies to manage and treat this condition.
The human gastrointestinal tract hosts an extensive community of over 100 trillion microorganisms, encompassing bacteria, fungi, viruses, and archaea collectively known as the gut microbiota. These microorganisms play pivotal roles in metabolic processes, immune regulation, and even drug metabolism (1). The microbiome is susceptible to environmental factors, such as diet, toxins, pH levels, and oxygen availability (7). Indeed, alcohol consumption promotes intestinal bacterial overgrowth and alterations in the microbial composition, termed dysbiosis, in both preclinical and clinical studies in animal models and humans, with a more pronounced impact observed in the duodenum (8, 9). A minority of individuals who engage in heavy drinking develop severe ALD (10). There is a growing understanding that the microbiome plays a role in the progression to ALD (1). Table 1 provides a detailed overview of the microbiota changes observed in patients with ALD. In summary, a reduction in Bacteroidetes and an elevation in Proteobacteria is common in drinkers with or without ALD when compared to healthy controls (11). Alcohol-associated cirrhosis (AC) and alcohol-associated hepatitis (AH) are associated with an increase in Enterococcus, Streptococcaceae, Enterobacteriaceae, Bifidobacterium, and Lactobacillus, and a reduction in Lachonospiraceae, and Ruminococaceae (12–14). Just as the reduction of bacterial diversity is linked to the advancement of ALD in patients, the reduction of fungal diversity similarly correlates with ALD progression (15). Stool samples obtained from alcohol-consuming patients with all spectrums of ALD had increased Candida (15, 16) and reduced Epicoccum, Debaryomyces, and Galamyces (17). Finally, the effects of the viral microbiome (virome) in ALD patients remain poorly understood (18). Recent metagenomic sequencing characterized intestinal viromes, revealing heightened viral diversity in fecal samples from patients with ALD, particularly those with AH (Table 1).
Ethanol exposure impacts numerous molecular pathways, including those involved in the regulation of intestinal permeability, generation and disposition of microbial metabolites, bile acid (BA) metabolism, and recognition of microbial-derived antigens, metabolites, and neurotransmitters (19). Together, these multiple pathways contribute to the pathogenesis of ALD. This complexity reflects the potential for exploring new therapeutics targets for preventing and treating different stages of ALD.
Upon consumption, ethanol is primarily absorbed by the proximal small intestine prior to being metabolized by the liver via oxidative and non-oxidative pathways. In the oxidative pathway, alcohol is metabolized by hepatocytes into acetaldehyde by alcohol dehydrogenase (ADH1) before conversion of acetaldehyde to acetate by aldehyde dehydrogenase (ALDH) 1 and 2. In addition, in response to chronic, heavy alcohol consumption, cytochrome P450-2E1 (CYP2E1) expression is induced and contributes to ethanol metabolism. The production of these ethanol metabolites plays an important role in disease pathogenesis, promoting inflammation, endoplasmic reticulum (ER) stress, and even lead to epigenetic changes within the host (20). The negative impact of ethanol metabolism is due to the harmful effects of reactive oxygen species (ROS) on macromolecules and direct damage to hepatic mitochondria, leading to the release of damage-associated molecular patterns (DAMPs). DAMPs, in turn, further promote inflammation and liver injury (21). Acetaldehyde remains the most toxic ethanol metabolite due to its ability to both induce chromosomal damage and promote adduct formation leading to increased inflammation and lipid accumulation (22).
Similarly, acetaldehyde disrupts intestinal tight junction proteins like occludin, zonulin, and claudins, leading to structural and functional changes in the barrier and increased bacterial translocation. This disruption plays an important role in the pathogenesis of ALD (23). An in vitro study using Caco-2 cells demonstrated that acetaldehyde-induced permeability increase was driven by the inhibition of protein tyrosine phosphatases, particularly PTP1B, leading to hyperphosphorylation of tight junction proteins including occludin, Zonula occludens-1 (ZO-1), E-cadherin, and β-catenin (24). This effect is also observed in ALDH2-deficient mice at ileal epithelial junctions, along with a heightened redistribution of tight junction proteins in the colon, ultimately compromising gut barrier function and worsening ethanol-induced liver damage (25, 26). Interestingly, ethanol can paradoxically increase claudin-2, a pore-forming protein, further contributing to barrier permeability (27). Corticosterone promotes ethanol-induced tight junction redistribution in the colon (28). Moreover, alcohol and acetaldehyde elevate intracellular calcium (Ca2^+^) (29), and recent research identified transient receptor potential cation channel subfamily V Member 6 (TRPV6), a specific channel on the intestinal epithelium, as responsible for this increase. The TRPV6 activation was linked to alcohol-induced gut barrier dysfunction and systemic inflammation (30). Acetate from ethanol metabolism can directly influence gut microbiome composition and function, further exacerbating liver injury in ALD (31). Interestingly, probiotic treatment in ethanol-fed mice reversed tight junction disruption, upregulating ZO-1 and occludin mRNA and preventing the redistribution of those proteins from colonic epithelia (32, 33). Similarly, dietary glutamine protects against alcohol-induced intestinal barrier dysfunction in mice (34). Supplementation with glutamine preserved the organization of tight junction proteins occludin and ZO-1 in the distal colon, through an epidermal growth factor receptor (EGFR)-dependent mechanism (35). Several therapeutics have been developed targeting ADH and ALDH inhibition specifically for the treatment of alcohol use disorder (AUD) (36). While targeted inhibition of specific steps in ethanol metabolism could lead to accumulation of toxic metabolites like acetaldehyde, these therapies show promise in treating early-stage ALD.
The dysregulation of microbial metabolites also contributes to changes in gut permeability, systemic inflammation, and liver damage in response to chronic, heavy alcohol consumption. Intestinal microorganisms anaerobically ferment dietary fibers into short-chain fatty acids (SCFAs), including acetic acid, propionic acid, butyric acid, and valeric acid. These regulate intestinal homeostasis, energy metabolism, and even blood-brain barrier and neuroimmunoendocrine functions (37). Chronic alcohol use induces dysbiosis, leading to changes in the relative concentrations of acetate, propionate, butyrate, isobutyrate, 2-methylbutyrate, isovalerate, valerate, and hexanoate in the stool; these changes are linked to gut barrier dysfunction (13). In addition, a study aimed at identifying novel metabolite mechanisms involved in the development of ALD found the lowest levels of acetic acid, butyric acid, propionic acid, iso-butyric acid, and iso-valeric acid in the stool of patients with cirrhosis (38). Notably, intestinal butyrate is also decreased in patients with AUD, further contributing to the weakening of intestinal tight junctions and increased permeability (39). The abundance of fecal SCFA-producing bacteria is decreased in patients with AH which was inversely correlated to presence of portal hypertension, endotoxemia, and systemic inflammation (40). Individuals with cirrhosis have lower plasma concentrations of the SCFA butyric acid, correlating with markers of inflammation and advanced liver disease (41). Beyond regulating intestinal integrity, SCFAs play a crucial role in regulating the expression of genes associated with hepatic lipid metabolism by modulating global histone acetylation and methylation in the liver (42). Butyrate acts as a histone deacetylase (HDAC), and supplementation of butyrate to mice reduces ethanol-mediated gut barrier disruption and hepatic steatosis (43, 44).
Metagenomic and metabolomic studies revealed that chronic alcohol consumption also impairs long-chain fatty acids (LCFAs) synthesis, and this adversely affects Lactobacillus species, leading to their reduction and subsequent disruption of tight junction barriers (27). Supplementation with saturated LCFAs reversed these effects and alleviated ethanol-induced liver disease in mouse models (27).
The microbial metabolism of choline generates methylamines like trimethylamine (TMA). While ethanol-related dysbiosis reduces choline availability (45), recent research indicates that ethanol-induced changes in the microbiota is associated with elevated concentrations of TMA. In liver, TMA is converted to TMA oxide (TMAO), with potential cardiometabolic consequences, by flavin-containing monooxygenase 3 (FMO3) enzyme. However, expression of FMO3 is reduced in patients with AH patients (45). This meta-organismal interaction of increased microbial generation of TMA, combined with decreased metabolism by FMO3 leads to increased plasma TMA concentrations in patients with AH (45). Use of small molecule drugs targeting specific microbial enzymes required to generate TMA is emerging as a potential therapeutic avenue (45).
In addition to the TMA/TMAO pathway, many other gut microbial metabolites are subject to intra-organismal metabolism by Phase II enzymes in the liver. Recent studies highlight the impact of ALD on expression of multiple enzymes involved in glucuronidation and sulfonation of microbial metabolites (46). Dysregulation of Phase II metabolism in patients with ALD has wide implications for both disposition of microbial metabolites as well as BA metabolism.
Neurotransmitters also emerge as key players in the gut-liver axis, influencing the dynamic communication between the gut microbiota, the central nervous system (CNS), and the liver during ALD. Increased levels of intestinal serotonin disrupt the integrity of the intestinal barrier, leading to increased permeability (47). Several bacteria, including Streptococcus, Acinetobacter, Escherichia (E.) coli, and Clostridia, can either directly produce or generate serotonin through a bacteria-mediated tryptophan metabolism pathway (48). Patients with ALD consistently exhibit elevated concentrations of serotonin compared to healthy individuals (49). Activation of serotonin receptors is linked to the promotion of dopamine release in the reward circuitry, heightening vulnerability to alcohol addiction (50). Moreover, central noradrenergic neurons can convert dopamine into norepinephrine which can influence the proliferation of anaerobic bacteria by upregulating virulence gene expression (51). Additionally, Escherichia coli (E. coli), Proteus vulgaris, Staphylococcus aureus, and Bacillus subtilis, produce both dopamine and norepinephrine (52). In a model of compulsive alcohol-seeking, changes in the gut microbiome were associated with altered striatal dopamine signaling (53). Furthermore, Lactobacillus and Bifidobacterium metabolize glutamate, an excitatory neurotransmitter in the brain, to produce γ-aminobutyric acid (GABA), an inhibitory neurotransmitter, in the gut. An inverse correlation between brain GABA levels and the severity of ALD was described (54).
BAs are steroid molecules derived from cytochrome p450-mediated catabolism of hepatic cholesterol. Following production and conjugation in the liver, BAs move through the bile canaliculi to the gallbladder, where they are stored. Upon consumption of dietary fats, BAs are released and absorbed into the small intestine to facilitate nutrient reabsorption in the terminal ileum (55).
The gut microbiota plays a pivotal role in influencing BAs metabolism. Several bacterial species can deconjugate BAs, including Clostridium, Lactobacillus, Bifidobacterium, Eubacterium, Escherichia, and Bacteroides (56). Conjugation with glycine or taurine reduces the pKa of BAs, improving water solubility and reducing lipophilicity. Therefore, after bacterial deconjugation, unconjugated BAs, being less polar—more hydrophobic and membrane permeable than conjugated forms—can be absorbed to a certain extent during transit through the intestine (57).
A minor portion of unabsorbed BAs enters the distal gastrointestinal tract, where bacterial metabolism leads to the formation of secondary BAs (58).
Along with their role in cholesterol metabolism and nutrient absorption, BAs are also important cell signaling molecules. Most notably, BAs can activate the nuclear receptor farnesoid X receptor (FXR) to exert their effects on the liver (59).
Hepatic BA synthesis is regulated by a negative feedback mechanism that includes activation of the FXR, which is expressed in intestinal and liver cells (60). Activation of FXR in the intestine, predominantly by primary BAs, leads to transcription of fibroblast growth factor 19 (FGF19), which reaches the liver through the portal vein and suppresses expression of CYP7A1 and thereby decreases BAs synthesis (60). In liver, FXR signaling activates expression of small heterodimer partner (SHP), which culminates in the repression of hepatic genes involved in BA synthesis such as CYP7A1 and CYP8B1(61). In addition, BAs can signal through G-protein-coupled receptor (GPCRs). For example, Takeda G protein-coupled receptor 5 (TGR5), expressed on multiple cell types including KCs, immune cells, and adipocytes, is involved in important metabolic processes, serving to maintain biliary and inflammatory homeostasis (62). Studies in various liver diseases have shown that TGR5 activation can reduce inflammation in KCs via modulation of lipopolysaccharide (LPS)-induced inflammatory signaling through cyclic adenosine monophosphate (cAMP) or antagonism of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB) signaling (63, 64). Additionally, FXR or TGR5 knockout mice both have increased susceptibility to alcohol-induced liver injury (65, 66).
Persistent alcohol consumption leads to an augmented BA pool and diminished BA excretion in mice, indicating that alcohol intake influences the enterohepatic circulation (67). Chronic exposure of mice to ethanol dysregulates BA metabolism, manifested as altered gene expression in the liver, elevated levels of unconjugated BAs in plasma and feces, and a transition from primary to secondary BAs (68). In line with these findings, patients with AH have increased serum concentrations of both unconjugated and conjugated BAs, despite a decrease in de novo BA synthesis (60). Increased BAs are further evidenced by increased serum FGF19 and are linked to exacerbated gut barrier dysfunction (69), altered expression of tight junction proteins, and epithelial disruption (70). Furthermore, in patients with cirrhosis, dysbiosis increases secondary BAs, aggravating changes in intestinal permeability (71). These findings highlight the intricate relationship between ALD, BAs metabolism, microbiota, and gut barrier integrity in both preclinical models and clinical contexts. While there are no approved therapies targeting pathways related to BAs in ALD, clinical trials are currently underway testing the efficacy of FXR and TGR5 agonists in other metabolic liver diseases (72, 73).
Both the intestine and liver have a complex cellular architecture involving resident and infiltrating cells that interact to maintain homeostasis. Ethanol exposure impacts the function and activity of every cell investigated to date in intestine and liver. Therefore, to understand the pathophysiology of ALD, we must comprehend the impact of ethanol on individual cell types in gut and liver as well as their interactions at different disease states. Rapid advances in nucleotide sequencing capabilities now allow for the characterization of gene expression profiles of cells dissociated from tissues with single-cell resolution (single-cell RNA sequencing, scRNA-seq) and of intact tissues with preservation of spatial information (spatial transcriptomics). These technologies have dramatically increased our understanding of the mechanisms underlying liver zonation and the spatial and functional heterogeneity of liver and gut-resident cells both during homeostasis and in chronic disease states. The sensitivity of scRNA-seq and spatial transcriptomics allows investigators to have deep insights into the molecular and cellular interactions in intestine and liver from very small, precious tissue samples from patients with ALD.
The intestinal epithelium, composed of specialized IECs, including enterocytes, goblet cells, intestinal microfold (M) cells, enteroendocrine cells, Tuft cells, stem cells, and Paneth cells, plays a critical role in maintaining intestinal homeostasis and regulating the gut-liver axis in ALD (19). This section aims to explore the cellular components of the intestinal epithelium that influence liver function and review the molecular mechanisms involved in this interaction.
Enterocytes throughout the intestine play a crucial role in providing a protective function (74). They actively produce and react to cytokines influencing various immune cells (75). For instance, after bacterial recognition via TLRs, enterocyte-secreted proliferation-inducing ligand (APRIL) triggers the production of protease-resistant immunoglobulin A 2 (IgA2) to participate in the control of microbial growth (76). Enterocytes also aid adaptive immune responses by facilitating the transport of IgA with the assistance of polymeric Immunoglobulin Receptor (pIgR) across the epithelial barrier to the gut lumen (77). The secretion of IgA provides a vital defense mechanism against microbial invasion (78). A detailed discussion on the role of IgA in ALD will be reviewed in subsequent sections. Enterocytes also contribute significantly to controlling inflammatory immune reactions and bacterial translocation by forming tight junctions between epithelial cells, integrated by several proteins including occludin, zonula occludens 1, and claudins, and shedding proteins like Ly6/Plaur domain-containing 8 (Lypd8). Together, these proteins interact to create a spatial barrier at the epithelium (79). Ethanol and its metabolites disrupt these tight junctions (80), resulting in barrier dysfunction during ALD. Translocation of bacteria through this disrupted intestinal barrier triggers immune system disturbances, leading to increased production of inflammatory cytokines further exacerbating the intestinal barrier dysfunction (2). During ALD, heightened nitroxidative stress, enterocyte apoptosis, and altered tight junction proteins increased intestinal permeability, and bacterial translocation, exacerbating liver damage (81). Moreover, fucosylated carbohydrate structures in IECs, specifically α1–2-fucosylation, play a crucial role in regulating intestinal homeostasis (82). In individuals with AUD, decreased α1–2-fucosylation is observed in duodenal biopsies, and α1–2-fucosylation deficiency in mice worsens the ethanol-induced disease by promoting the expansion of intestinal Enterococcus (E.) faecalis (82).
Moreover, enterocytes secrete antimicrobial peptides (AMPs) such as ß-defensins and regenerating islet-derived protein 3 (REG3) lectins that target bacteria by creating specific pores in their membranes, and cathelicidins that disrupt bacterial and fungal cell membranes similar to REG3 (9, 83, 84). In patients with ALD, expression of multiple AMPs is diminished (15, 76). Alcohol leads to the reduction of REG3 protein expression in both humans and mice (9, 83, 84). Furthermore, chronic ethanol feeding decreases the expression of cathelicidin-related antimicrobial peptide (CRAMP) in mice and exacerbates ethanol-induced liver disease while its administration decreases ethanol-steatohepatitis (85, 86). Furthermore, human ß-defensin (hBD) is upregulated in inflammatory diseases including ALD (87, 88), and administration of hBD-2 ameliorates liver injury in mice (88).
Intestinal stem cells (ISCs) are vital for the gut’s renewal. This regeneration is crucial for a strong barrier against invading microbes. Ethanol-induced liver disease disrupts ISC function in the mouse small intestine by impacting ß-catenin signaling and other ISC markers, hindering epithelial regeneration and promoting increased bacterial translocation (89).
Paneth cells, located mainly in the small intestine crypts near ISCs, produce AMPs such as lysozyme C, α-defensins, phospholipases, cryptdins, and lectins that defend against microbial invasion. These AMPs, spread across the mucus layer, effectively kill bacteria, bolstering the intestinal barrier (90). Ethanol abuse alters Paneth cell biology, raising their numbers in the proximal small intestine (91) while reducing levels of AMPs like α-defensins (92). Patients with ALD and mice subjected to intragastric ethanol and 8 weeks of ethanol-containing Lieber-DeCarli liquid diet display decreased REG3 expression (9, 87, 92) linked to elevated mucosa-associated bacteria in the small intestine and increased bacterial infiltration into the liver (83). Dysfunction of Paneth cell α-defensins, mediated by zinc deficiency, contributes to the pathogenesis of AH, and promising avenues for therapeutic interventions have been preclinically tested, such as human α-defensin 5 (HD5) administration, to alleviate alcohol-induced liver damage (92). Notably, IL-22 treatment to alcohol-fed mice mitigated liver injury and reversed alcohol-reduced Reg3γ and α-defensins expression, correcting the antimicrobial response (93).
Goblet cells secrete mucin, forming the protective mucus layer (94), and produce AMPs like resistin-like molecule beta (RELM-ß) and trefoil factor (TFF2). RELM-ß boosts inflammation via increased interferon-gamma (IFN-γ) expression in CD4 T cells (95), while TFF2 aids in repairing damaged areas (96). Interestingly, elevated levels of REG3α and TFF3 in AH patients’ circulation, correlating with AH severity, microbial translocation, and inflammation, suggest their potential as biomarkers for assessing gut epithelial damage (97). The mucus layer extends throughout the digestive tract, acting as a barrier against pathogens. Mucin increases in production during ethanol-induced liver disease in mice, rats, and patients with AUD (87, 98) seemingly as a protective response against bacterial invasion. However, this thickened mucus layer could pose challenges for AMPs to control bacterial overgrowth due to increased physical barrier thickness (99). Mucin 2 (MUC2) is the predominant mucin in the intestine (99), when deficient results in a thinner mucus layer and milder ethanol-induced liver disease. Despite a more permeable gut barrier, these mice experience less microbial translocation attributed to reduced bacterial overgrowth, heightened immune surveillance, and increased expression of Reg3b and Reg3g (87).
Among the most important functions of goblet cells is the capacity to form goblet cell-associated antigen passages (GAPs). GAPs are dynamic structures involved in transporting luminal antigens and bacteria to lamina propria dendritic cells (LP-DCs), crucial for the education of the immune response (100). Whether GAPs have any implication in ALD remains to be elucidated.
Microfold (M) cells are distributed in Peyer’s patches, isolated lymphoid follicles, and colonic patches can modulate the microbiome by transporting antigens via endocytosis and transcytosis (101) to macrophages and LP-DCs, pivotal for immune response (101) by triggering immune responses in T and B cells (102). Long-term ethanol abuse in mice promoted structural changes in M cells (103) and was associated with fewer T and B cells in Peyer Patches (104). More research is needed on how ALD specifically alters M cell function.
Enteroendocrine cells wield significant influence in the gut microbiota. (105). Enteroendocrine cells house and release a range of peptide hormones, including cholecystokinin, somatostatin, motilin, neurotensin, vasoactive intestinal peptide (VIP), enteroglucagon, gastric inhibitory peptide (GIP), glucagon-like peptide (GLP) 1 and 2, cleaved peptide YY, oxyntomodulin, and histamine in response to nutrients and exposure to microbial byproducts. This release is triggered by the activation of G protein-coupled receptors (105). G protein receptors 41 and 43 (GPR41, GPR43) in enteroendocrine cells detect microbiota-derived SCFAs, prompting inflammatory responses in mice (106, 107). These activated cytokines can modify gut barriers by altering tight junctions (108). Furthermore, patients with AUD exhibit increased glucagon and GIP levels, suggesting an effect on specific enteroendocrine cells types (109). Reports also indicate impaired intestinal somatostatin production due to chronic ethanol exposure in rats (110). Other studies have noted a correlation between elevated ethanol intake, reduced defecation rates, and altered expression of enteroendocrine cells -produced hormones in the fly midgut, including diuretic hormone 31 (Dh31)(111).
Tuft cells, though few in number, act as communicators among intestinal immune cells. They produce interleukin-25 (IL-25), activating T helper type 2 (Th2) cells (112). Interestingly, the activation of epithelial tuft cells through the succinate receptor 1 (SUCNR1) leads to a type 2 immune response. This response, in turn, worsens the Paneth cell alterations, causing an imbalance in gut microbes and persistent inflammation (113). Tuft cells exhibited a significant reduction, particularly in left colon organoids, upon exposure to ethanol (114). However, prior research indicated that various dietary fats could influence the alterations induced by ethanol in the intestine (115). Interestingly, in fat-1 mice, which endogenously convert n6 to n3 PUFAs attenuating ethanol-mediated alterations in the gut-liver axis (116). In response to ethanol, these mice experience a downregulation of cell death, inflammation, and tuft cell markers (115). Further studies on the role of tuft cells in ALD are required.
In summary, the specialized cells of the intestinal epithelium play a pivotal role in maintaining intestinal homeostasis and regulating the gut-liver axis in ALD (Figure 1).
Intestinal equilibrium relies heavily on a harmonious synergy between physical, chemical, and immunological defenses against various pathogens (such as bacteria, viruses, and fungi), alongside immune tolerance toward dietary antigens and commensal bacteria (117). Intestinal homeostasis is orchestrated by the gut mucosal immune network. This network encompasses distinctive elements such as gut-associated lymphoid tissues (known as GALT or Peyer’s patches) mainly hosting intestinal B cells and secretory Ig A (SIgA) producing plasma cells, elements such as intraepithelial lymphocytes (IEL), LP innate immune cells including mononuclear phagocytes (MNPs), intestinal DCs, macrophages, eosinophils mast cells, mucosal-associated invariant T cells (MAIT cells), invariant natural killer T cells (iNKT cells) and innate lymphoid cells (ILCs), as well as adaptive immune cells such as CD4+ T lymphocytes (T helper type 1, 2 and 17 (Th1, Th2, Th17), T follicular helper (Tfh), and T regulatory (Treg cells)), and CD8+ T cells. Other elements include cytokines (e.g., IL-10), chemokines (e.g., CCR9), and cytokine-stimulated AMPs secreted by the innate and adaptive immune system (19).
Microbial antigens are transported to Peyer’s Patches through specialized M cells, where they are captured and presented by DCs. The following adaptive immune response stimulates the differentiation of B-cells to produce the IgA-secreting plasma cells (118). After migration to the LP, plasma cells actively produce IgA, which, upon binding to pIgR on enterocytes, is secreted into the mucus layer and serves as a critical defense mechanism (119). Mice exposed to chronic ethanol display diminished intestinal SIgA secretion, associated with reduced pIgR expression and alterations in immune cell functionality (120, 121). Impaired hepatic pIgR functionality exacerbates ALD by compromising IgA-mediated defense in the gut (122). Despite elevated systemic IgA levels and deposits in hepatic sinusoids in ALD patients, the focus on gut SIgA levels gains importance due to its pivotal role in regulating gut homeostasis. Studies consistently report reduced SIgA levels in the gut of individuals with ALD and metabolic dysfunction associated steatotic liver disease (MASLD)/ metabolic dysfunction-associated steatohepatitis (MASH) (123–125).
IEL, vital for host defense and wound repair, actively patrol the epithelial barrier, responding rapidly to infections or stressed cells and promoting cytolytic and Th1 cytokine responses (126). In a mouse model where cirrhosis was induced using carbon tetrachloride and 5% alcohol in drinking water for 12 weeks, or with alcohol alone, there was an increased number of IELs in both groups but compromised immune function was observed, with decreased proliferative response of IELs, reduced production of IFN-γ and increased bacterial translocation in both groups compared to the control group, but pronounces responses in the cirrhosis group (127). Consistently, histological changes in ALD patients involve increased IEL in the intestinal mucosa (128).
DCs recognize and eliminate pathogens while maintaining intestinal immune tolerance. There is increasing interest in understanding the functions of LP-DCs, particularly within the gut-liver axis. LP-DCs are categorized based on their expression, with major subgroups including CD103+ CD11b+, CD103+ CD11b−, CD103− CD11b+ (129).
The CD103+ CD11b− subset, known as the conventional type 1 DC (cDC1) subset, is predominant in the colonic LP, lymphoid tissues, and Peyer’s patches, and migrates to intestinal draining lymph nodes(130). cDC1 differentiation is driven by basic leucine zipper transcription factor ATF-4 like 3 (BATF3) and interferon regulatory factor 8 (IRF8) and plays a crucial role in the differentiation of CD8+ effector T cells under inflammatory conditions (131). Chronic ethanol feeding in mice reduces BATF3-dependent CD103+CD11b− cDC1s, decreasing ileal IL-12 levels. This proinflammatory cytokine induces IFN-γ-secreting Th1 and CD8 T cells, and IFN-γ-inducible genes involved in AMP secretion. Consequently, ethanol-induced cDC1 reduction resulted in decreased ileal Th1 and CD8+IFN-γ+ T cells. This led to the downregulation of downstream targets and AMPs (Defa5 and Reg3g), contributing to A. muciniphila reduction and tight junction disruption. The authors demonstrated that cDC1s, act as gatekeepers, preserving A. muciniphila, preventing tight junction disruption, and protecting against ethanol-induced liver disease (132, 133).
The impact of alcohol on DCs beyond cDC1s is not well understood, and its implications for ALD remain unclear.
Derived from monocytes, intestinal macrophages are highly phagocytic and release anti-inflammatory cytokines like IL-10, promoting Treg differentiation (134). Patients with ALD, show an increase in CD68/TNFα positive monocytes and macrophages expressing elevated Tnfa mRNA in the duodenum (135). However, in ethanol-fed mice this effect is only observed in jejunum and is reversed with the use of antibiotics (135). It is interesting that, while chronic ethanol increases the numbers and types of inflammatory macrophages in the liver (136, 137), in the colon of rodents fed ethanol, macrophage numbers are reduced (138, 139). Further studies are indicated to understand the different roles of different intestinal macrophage populations in ALD.
MAITs play a crucial role in the defense against bacterial infections, and are found in the intestine, peripheral blood, and the liver (140). Limited data exists on MAITs in the intestine during ALD, but studies on binge-on-chronic alcohol exposure reveal significant reduction and hyperactivation of MAIT cells in the intestine, lung, and liver of mice. Fecal microbiota transplantation (FMT) from alcohol-fed mice replicated the effects on MAIT cells, even without direct alcohol exposure (141). In summary, these findings underscore the significance of MAIT cells in ALD.
Mast cells bolster innate immunity against pathogens, regulate angiogenesis, blood vessel integrity, gut barrier permeability, and can migrate to damaged tissues (142). In alcohol-induced carcinogenesis, using the APC^Δ468^ mice model, ethanol exposure increases polyp-associated mast cell numbers and promotes mast cell-mediated tumor migration in vitro, while depletion of mast cells was protective (142). It has been postulated that ALD involves the migration of mast cells and other immune cells to the damaged liver tissue (143), offering a key avenue for mitigating liver damage during ALD. Additionally, the interplay between IL-9 and mast cells regulates Candida pathogenicity at mucosal surfaces, and taking into account the role of Candida in the promotion of ALD, this fact, underscores the need to further explore the role of these cells in the pathogenesis of ALD (144). Notably, in the progression of cirrhosis, fibrosis, hepatitis, and diverse cholangiopathies, there is an observed increase in hepatic mast cell numbers (145).
iNKT cells, are innate-like T cells commonly found in the intestine, can migrate to the liver where they promote apoptosis and exacerbate ethanol-induced steatohepatitis (146).
ILCs within the LP, particularly ILC3s generate cytokines like IL-22 in response to stimuli (147). The microbiome produces metabolites such as butyrate and tryptophan decomposition products, enhancing gut integrity and ILC3 through activation of Aryl-hydrocarbon receptor-ligand (AhR) to produce IL-22 (147), which, in turn, induces the production REG3γ and REG3ß AMPs and epithelial fucosylation (148). Mice with ethanol-induced liver disease have impaired intestinal ILC3-secreting IL-22 production and decreased REG3 expression, thus fueling bacterial translocation translocation (149). AhR-ligand indole-3-acetic acid (IAA) administration and IL-22-producing bacteria administration reverted this effect and prevented ethanol-induced steatohepatitis (149). The overall survival rate in patients with AH depends on the presence of IL-22 cells from peripheral blood (150, 151). Clinical trials in AH patients using a recombinant fusion protein of human IL-22 showed a significant decrease in the model for end-stage liver disease (MELD) score, total bilirubin, ALT, and AST, a high rate of Lille score, and a reduction inflammatory marker in plasma (152, 153).
CD4+ T lymphocytes (Th1, Th2, Th17, Tfh, and Treg cells), and CD8+ T cells are also important components of the intestinal adaptive immune system. Research indicates that the consumption of ethanol may be associated with an immune shift towards Th2 cytokines, potentially contributing to harm to both the intestinal immune system and the liver (154, 155).
The elevation of IL-17 in the proximal small intestine during ethanol feeding is a result of the heightened presence of Paneth cells (91), and its abrogation with the use of aminosalicyclic acid (5-ASA) ameliorated ethanol-induced liver disease (156, 157). However, Th17 in the intestine plays a dual role by combating pathogens through inflammatory responses (158). However, promoting pro-inflammatory responses could be detrimental to the liver (157, 159). Additionally, extended exposure to an ethanol-containing Lieber-DeCarli diet in mice and AUD patients has been correlated with an elevated presence of Candida albicans (C. albicans) (15, 160). Indeed, intestinal C. albicans that can induce antigen-specific Th17 cells in humans (159) is expanded in the intestine of ALD patients and promotes migration of higher numbers of C. albicans-specific Th17 cells to the liver, promoting ethanol-induced liver disease (16, 161).
Interestingly, ethanol disrupts the balance of CD4+ T cell types by increasing inflammatory Th1 cells and decreasing Treg cells (162). Ethanol negatively impacts CD4+ T cell function and metabolism, promoting a pro-inflammatory state (162). On the other hand, ethanol exposure in mice reduces key markers and functions of Tfh cells, affecting their spatial organization and hindering the formation of Tfh -B cell connections in germinal centers (163).
Regarding CD8+ T cells, ALD patients have a decrease in the number of duodenal CD8+ T resident memory (TRM) cells. Even short-term abstinence does not reverse TRM cell death in ALD patients (164). However other studies have shown that chronic heavy drinking is associated with a reduction in T-cell numbers, loss of naïve T-cells, increased CD8+ T-cell activation and proliferation, or alterations in monocytes (164–166). Additionally, chronic binge alcohol administration results in marked loss of CD3+ T cells, accompanied by marked increases in intestinal CD4+ and CD8+ T cell proliferation and turnover, proliferation and turnover in rhesus macaques (167). Furthermore, alcohol use is associated with intestinal dysbiosis and dysfunctional CD8^+^ T-cell phenotypes in human immunodeficiency virus (HIV) persons (168). Given the complexities surrounding CD8+ T cells in ALD, it is imperative to conduct further investigations to elucidate the intricacies involved.
The gut microbiome significantly influences the development and function of the gut mucosal immune system, vital for intestinal homeostasis. For instance, germ free (GF) animals exhibit reduced IEL, IgA-secreting plasma cells, and Tregs (169, 170). Angiogenin-4 (Ang4), an antimicrobial crucial for microbe defense, shows decreased expression in GF mice (171). These immune deficits leave GF mice highly vulnerable to intestinal compromise making them susceptible to ethanol-induced liver disease mirroring binge drinking (172, 173).
In summary, the intricate interplay of various intestinal immune cells within the gut-liver axis highlights the complexity of ALD (Figure 1).
The enteric nervous system (ENS) is organized into two ganglionated the submucosal plexus and the myenteric plexus (174). These plexuses consist of enteric neurons including nitrergic, cholinergic, calretinin-expressing, neuropeptide-expressing, catecholaminergic, and inhibitory GABA neurons (174). This network forms a complete reflex circuit, encompassing intrinsic primary afferent neurons (IPANs), interneurons, and motor neurons within the gut wall (175). These glial cells, akin to astrocytes in the CNS, provide support to enteric neurons and possess neuroprotective properties (176). The ENS governs intestinal movement, blood flow, gastric acid secretion, nutrient absorption, and water intake, while also engaging with immune cells and intestinal epithelial barrier regulation (176).
The gut microbiota plays a crucial role in the proper development and upkeep of ENS functionality. Studies in GF mice indicate an underdeveloped ENS in the absence of gut microbes, which can be rectified through colonization in adult mice (177). The communication between the gut microbiota and the CNS, as well as the ENS, occurs through various pathways, encompassing microbial SCFA, microbial-derived neurotransmitters, hormones, the immune system, and PRRs on neurons and glial cells (178).
Alcohol exposure releases excitatory entero-neurotransmitters like ATP and GABA that activate P2X2/3 and GABA receptors in excitatory ascending neurons (179). Studies in mice have linked chronic alcohol-induced changes in neurotrophic factors brain-derived neurotrophic factor (BDNF) and α1 subunit of γ-aminobutyric acid A receptor (Gabra1) expression and gut microbial populations to neuropsychic behaviors (180). Furthermore, chronic alcohol intake alters the kynurenine pathway with a consequent reduction in serotonin synthesis that leads to cognitive alterations (181). New findings indicate that enteric neurons producing the neuropeptide VIP play a role in promoting α1,2-fucosylation of IECs, thereby regulating gut microbiota balance. The reduction of VIPergic neurons in the gut resulted in the expansion of E. faecalis and the reduction of Bifidobacterium. This imbalance ultimately increased susceptibility to ALD (182).
In summary, investigating the intricate interplay between the ENS, immune system, and gut microbiota in the context of ALD holds significant relevance. Further research in this domain is imperative for advancing our knowledge and improving clinical outcomes in ALD patients.
The liver is a highly vascularized organ that performs a diverse array of metabolic and immunologic functions because of its rich diversity of cell types and complex structural architecture. It is functionally divided into lobules with zonated metabolic compartmentalization. Herein, the pathophysiology of cell types within the liver both during homeostasis and their dysregulation with chronic alcohol consumption will be explored.
Hepatocytes, the predominant cell type within the liver, execute a myriad of functions, including detoxification, metabolism of carbohydrates and lipids, and synthesis of proteins and BAs (183), and are the principal site of alcohol metabolism. Chronic alcohol exposure modulates multiple enzymatic pathways to cause fat accumulation within hepatocytes (184). Hepatic steatosis is one of the earliest forms of disease and thus a contributing factor to the development of later stages of ALD. Triglycerides are the predominant lipid that accumulate in ALD, doing so through esterification of fatty acids (FAs), hepatic uptake of adipose-derived FAs, and through intestinal secretion of large triglyceride-rich lipoproteins known as chylomicrons (185). Alcohol has been shown to disrupt multiple facets of hepatic lipid metabolism, ranging from fatty acid transport to increased production of bioactive lipid mediators (186, 187). Circulating FAs, the predominant source of lipid accumulation in ALD, enter the liver through the action of fatty acid transporter proteins (FATPs) and fatty acid translocase (CD36), and are packaged by hepatocytes into very low-density lipoprotein (VLDL) to allow for their export and the prevention of triglyceride accumulation (188). However, in ALD both uptake and export of triglycerides is impaired as CD36 expression is increased in patients with ALD, and components of VLDL assembly such as peroxisome proliferator-activated receptor α (PPARα) and apolipoprotein B (apoB) are both dysregulated (188, 189). While this dysregulation is mostly attributed to ethanol metabolism, recent genome wide association studies (GWAS) revealed that some patients could be genetically predisposed to dysregulated lipid metabolism. Missense mutations in patatin-like phospholipase domain-containing protein 3 (PNPLA3), a gene involved in triglyceride hydrolysis, are significantly associated with high alcohol intake and increased risk of hepatocellular carcinoma (HCC) and cirrhosis (190). In addition, mutations in membrane-bound *O-*acyltransferase 7 (MBOAT7), a lysophosphatidylinositol acyltransferase gene involved in phospholipid remodeling, is linked to an increased risk of AC (191). Indeed, hepatocyte-specific deletion of Mboat7 dysregulates both autophagic flux, marked by accumulation of autophagosomes, and lysosomal biogenesis in mice exposed to chronic ethanol feeding (192).
scRNA-seq analysis of livers from alcohol-exposed mice revealed that alcohol dysregulates the expression of genes associated with oxidative stress, hypoxia, complement, and energy metabolism in hepatocytes (193). Hepatocytes also contribute to the progression of ALD through interactions with other liver-resident cell types. For example, apoptotic bodies released from hepatocytes activate KCs and HSCs to promote fibrosis (194), and hepatocyte-derived IL-6 polarizes liver macrophages to a pro-fibrogenic M2 phenotype (195).
KCs are the most populous immune cells within the liver under homeostatic conditions. Owing to their residence in sinusoids adjacent to endothelial cells (196), they promote liver homeostasis through scavenging of pathogens and damaged cells (183), secretion of immunoregulatory molecules (197), and iron and lipid metabolism (183). Three separate human liver scRNA-seq studies have identified Cluster of Differentiation 163 (CD163), macrophage Receptor With Collagenous Structure (MARCO), CD5 Molecule Like (CD5L), and T Cell Immunoglobulin And Mucin Domain Containing 4 (TIMD4) as markers of KCs (183, 198–200). Furthermore, proinflammatory and anti-inflammatory subpopulations of KCs exist in homeostasis (199). Zinc finger E-box binding homeobox 2 (ZEB2), a transcription factor widely recognized for mediating epithelial-to-mesenchymal transition, is critical to maintaining KC identity (201). KCs are also important in both acute and chronic responses to alcohol. Notably, almost a third of liver-resident KCs undergo apoptosis within the first days following alcohol exposure and are replaced with monocyte-derived neo-KCs (202). Furthermore, alcohol exposure increases portal delivery of intestine-derived DAMPs and pathogen-associated molecular patterns (PAMPs), including LPS, to induce proinflammatory cytokine and chemokine expression (195). Importantly, chronic alcohol exposure increases the sensitivity of KCs to stimulation by PAMPs and DAMPs, further exacerbating inflammation in the liver (136, 137). In addition to soluble mediator release, scRNA-seq performed on livers from mice with multiple alcohol exposure patterns revealed dramatic changes in expression of genes related to antigen presentation and energy metabolism in KCs (193).
In addition to resident hepatic macrophages, a rich diversity of both myeloid and lymphoid cell types are important for liver homeostasis and disease progression. In the setting of liver injury, circulating LY6C^hi^CCR2^+^ monocytes infiltrate the liver parenchyma and differentiate into monocyte-derived macrophages (MoMFs) that are distinct from KCs (183, 196). MoMFs are immunologically active as they produce proinflammatory cytokines upon sensing of DAMPS and PAMPs (197) and secrete the profibrogenic cytokine transforming growth factor beta (TGF-β) to activate and transdifferentiate HSCs (196). Notably, scRNA-seq analysis of human cirrhotic livers identified the etiology-independent presence of circulating monocyte-derived TREM2+CD9+ scar-associated macrophages specifically within fibrotic niches that activate PDGFRα-expressing scar-associated mesenchymal cells (183, 198). Alcohol impacts extrahepatic monocyte function as well. Single-cell transcriptional profiling of peripheral blood mononuclear cells (PBMCs) obtained from patients with AH identified three subsets of monocytes with dysregulated expression of gene profiles controlling cytokine production, differentiation, and antigen presentation (203). Additionally, an independent scRNA-seq analysis of PBMCs demonstrated diverse responses to LPS stimulation, and a subset of anti-inflammatory circulating monocytes exhibited a pro-inflammatory gene expression profile in patients with AH (204).
Single-cell analyses have uncovered two principal lineages of DCs within the liver plasmacytoid (pDCs) and cDC (183). In addition to engaging in antigen presentation, these DCs express immunoregulatory molecules to maintain homeostasis (197). Chronic alcohol consumption increases intrahepatic neutrophil levels and induces multiple phenotypic changes, including release of reactive oxygen and nitrogen species, pro-inflammatory cytokines, and DNA-containing extracellular traps (NETs), to propagate inflammation and injury (196, 197). Mast cell numbers in the liver increase during the progression of cirrhosis, fibrosis, hepatitis, and diverse cholangiopathies (145).
The intrahepatic lymphoid compartment consists of both innate and adaptive immune cells. Distinct populations of intrahepatic NK, B, and T lymphocytes, including CD4+ memory, CD8+ effector, and γδ T cells, have been identified within the homeostatic liver by scRNA-seq (183, 200). Dysregulation of both innate and adaptive lymphoid cells have been observed in AC. NK cells clear activated stellate cells and release cytokines like IFN-γ to induce stellate cell apoptosis; alcohol exacerbates fibrogenesis by restraining these protective functionalities and depleting intrahepatic CD45+CD16− NK cells to exacerbate fibrogenesis (183). Additionally, scRNA-seq of liver immune cells from patients with AC identified expansion MAIT within fibrotic niches with evidence of intercellular communication with macrophages and NK cells (205).
HSCs are the principal reservoir of vitamin A in homeostasis. In liver fibrogenesis, they transdifferentiate into collagen-producing myofibroblasts (206) and are the predominant source of extracellular matrix production (183). ScRNA-seq of resting HSCs and activated myofibroblasts from murine CCl4-induced fibrotic livers found resting stellate cells to be a homogeneous population whereas myofibroblasts consist of multiple heterogenous populations (206). Topographically, stellate cells segregate to distinct regions within liver lobules (207), and central vein-associated HSCs, characterized by expression of LPAR1, are the predominant source of pathogenic extracellular collagen in CCl4-induced liver fibrosis (207). Furthermore, spatial transcriptomic analysis identified distinct stellate cell transcriptional states specifically within human fibrotic liver septa (208). Single-cell transcriptional profiling has enhanced our understanding of the complex interplay of factors that regulate stellate cell activation, including focal adhesion molecule signaling within matrix-associated stellate cells (209) and expression of the extracellular matrix protein microfibrillar-associated protein 4 (MFAP4) (210). Classically, stellate cell activation was widely-regarded as an end-stage phenomenon downstream of dysregulation of non-parenchymal cells; this paradigm is now called into question as stellate cells secrete multiple factors that continue regulate immune-related and profibrogenic pathways in other cells in the liver (211). In fact, scRNA-seq identified a large number of secreted factors of stellate cell origin, including cytokines and vasoactive peptide hormones, that act on endothelial and immune cells (212).
LSECs are marked by high expression of C-Type Lectin Domain Family 4 Member G (CLEC4G), CLEC4M, steroid receptor RNA activator (SRA); stabilin-2 (STAB2), and cluster of differentiation 14 (CD14) and have increased activity of the GATA Binding Protein 4 (GATA4) transcription factor (198). They execute an array of important functions during homeostatic conditions, including filtration/scavenging (213) and immunoregulation through cytokine release (197), antigen presentation (197), and priming the differentiation of MNP (211). LSECs contribute to the pathogenesis of ALD via several mechanisms, including endothelial dysfunction and capillarization (213). Endothelial dysfunction through loss of nitric oxide production from endothelial nitric oxide synthase (eNOS) impairs relaxation (213), resulting in increased vascular resistance and portal hypertension (214). Capillarization of LSECs, marked by loss of fenestrae and the generation of a basement membrane, leads to enhanced HSC activation (215) and deposition of extracellular matrix within the space of Disse (213). scRNA-seq of mouse liver with alcohol exposure identified expression changes in genes related to nitric oxide production, immunoregulation, and migration of epithelial and immune cells(193).
Spatial transcriptomics and scRNA-seq have uncovered mechanisms underlying the zonated functional specialization of hepatic cells, especially hepatocytes and LSECs. Hepatocytes perform distinct metabolic functions along the sinusoidal axis extending from central vein to portal vein, likely due at least in part to gradients of oxygen, nutrients, and other metabolic factors (183). Multiple scRNA-seq and spatial transcriptomics studies (197, 198, 200, 216, 217) have corroborated that hepatocytes in periportal (zone 1) are involved in gluconeogenesis and synthesis of cholesterol and urea whereas pericentral (zone 3) hepatocytes specialize in glycolysis, detoxification and synthesis of glutamine and BAs. Additionally, zone 2 is the predominant site of hepatocyte division and CYP-mediated metabolism of xenobiotics (183). Wnt signaling, hypoxia, and pituitary hormones are molecular drivers of hepatocyte zonation (216). Zonation is also well-described in LSECS with between one-third to two-thirds of LSEC genes exhibit zonated expression patterns (183, 218). Furthermore, capillarization has been found to be more severe in pericentral (zone 3) LSECs in murine cirrhotic livers (215).
In summary, spatial transcriptomics and scRNA-seq have greatly enhanced our understanding of liver architecture and subclasses of intrahepatic cell types, both in health and in multiple disease contexts. The application of these cutting-edge technologies to ALD is in its infancy and most likely will revolutionize both our knowledge of the pathophysiology and our currently limited therapeutic options.
The transfer of PAMPs and DAMPs caused by alcohol and its metabolites initiates the activation of Pattern Recognition Receptors (PRRs) like cell surface TLRs, intracellular nucleotide-binding oligomerization domain-like receptors (NOD) in liver immune cells (219) and C-type lectin receptors (CLRs) (15, 220). Upon recognition of bacterial moieties, TLRs on KCs and macrophages through both the myeloid differentiation primary response 88 (MyD88) and Toll/ILR-1 receptor domain-containing adapter-inducing interferon-β (TRIF) pathways induce transcription of proinflammatory cytokines, chemokines, co-stimulatory proteins and proteins that process antigens (4, 221). NOD1 and NOD2 are activated by components of bacterial cell walls and play a key role in the inflammatory processes associated with ALD (222). TLR4 activation by LPS from Gram-negative bacteria is associated with enhanced liver injury and proinflammatory cytokine expression in ALD (3). TLR2 and TLR9 recognize the surface of Gram-positive bacteria (223) and the CpG containing DNA fragments derived from virus and bacteria (223), and both contribute to ethanol-induced liver injury by promoting the expression of CXCL1 and the infiltration of neutrophils (224). For instance, in a mouse model with ethanol-induced liver disease, gavage of E. faecalis resulted in translocation to the liver and activated TLR2 on KCs to increase IL-1β secretion and promote liver cell damage (7). Treatment with the IL-1 receptor antagonist anakinra protected from E. faecalis -exacerbated ethanol-induced liver disease (7). Moreover, activation of TLR3 by mitochondrial double-stranded (ds) RNA released from exosomes stimulates neighboring KCs to produce IL-1β, contributing to liver inflammation in ALD (225). The activation of inflammasomes, notably the NOD-like receptor protein 3 (NLRP3) inflammasome, contributes to liver inflammation in response to PAMPs such as LPS (226). In the progression of ALD, the NLRP3 inflammasome pathway plays a role in the maturation of caspase-1, IL-1β, and IL-18, triggering a robust inflammatory response that contributes to fibrosis by activating HSCs with profibrogenic effects (227).
Furthermore, bacterial toxins can prompt cell death in the liver. In a recent investigation, it was discovered that cytolysin, a pore-forming toxin produced by E. faecalis, directly induces hepatocyte lysis and results in liver damage (228). The presence of cytolysin exhibited a strong correlation with unfavorable clinical outcomes and mortality in individuals with AH and in mice (228). Beyond bacteria, C. albicans was observed consistently across various stages of liver disease, with patients with nonprogressive ALD, AH, and AC showing similar levels of fungal dysbiosis in the intestine (15, 16). The 1,3-β-glucan derived from the fungal cell wall enters the liver via the portal vein and binds to the CLRs, Dectin-1/Clec7 on the surface of hepatic KCs, triggering the secretion of IL-1β resulting in steatosis and necrosis (229). Furthermore, yeast toxins such as candidalysin produced by C. albicans, were linked to the advancement of ALD in mice (160). In patients with AH, heightened fecal levels of candidalysin were associated with increased mortality (16). Endogenous CLR ligands also contribute to disease progression; β-glucosyl ceramides are increased in circulation of patients with AH and interact with another CLR, Mincle, to increase inflammation in murine models of ALD (220).
The galectins (Gal) are evolutionarily conserved β-galactoside-binding lectin proteins that are implicated in steatosis and fibrosis in multiple organs via activation of the NLRP3 pathway and through the TGF-β pathway (230). Systemic Gal-3 is increased in ALD and negatively correlates with liver function (231). Patients with AC have elevated Gal-3 levels and correlated with Child-Pugh score (232).
Moreover, systems related to hepatic pathobiont clearance are also altered in ALD. The complement receptor of the immunoglobulin superfamily (CRIg) on liver macrophages that facilitates the phagocytosis of complement component C3b or Gram-positive bacteria is reduced in liver tissues of ALD patients (233). CRIg-deficient mice exhibit worsened ALD, and less efficiency in clearing translocated Gram-positive bacteria like E. faecalis which facilitates the progression of liver disease emphasizing the intricate immunological factors that contribute to the progression of liver disease (233).
Complement, part of the innate immune system that also bridges to the adaptive immune response, is implicated in ALD. Accumulating evidence from murine models indicates that complement activation and release of anaphylatoxins contributes to liver inflammation and drives progression of ethanol-induced liver injury (136, 234), while complement Factor D (CFD), a component of the alternative pathway, protects mice from chronic ethanol-induced injury (235). Importantly, studies in patients with AH indicate that the quantity and activation of circulating and hepatic complement may be important diagnostic and prognostic biomarkers in patients with severe AH (sAH) (236).
Studying the gut-liver axis and microbial interactions in the context of ALD requires multifaceted approaches, including computational methodologies. Leveraging patterns from extensive datasets, machine learning (ML) excels at identifying predictions and associations challenging for traditional methods. A recent study, aimed to predict short-term mortality in AH, used different ML algorithms and considered various types of data, including serum metabolites, serum lipids, fecal bacteria, and fecal fungi and viruses (261). Gradient boosting outperformed the current MELD score, achieving the highest area under the curve (AUC) in predicting 30-day mortality using the intestinal microbiota and metabolic pathways dataset and 90-day mortality with the fungi dataset (237).
Transcriptomic analysis across multiple tissues is another example of the use of advanced computational tools that are furthering our understanding on gut-liver interactions in ALD. For example, a transcriptomic analysis on liver and ileum of mice subjected to chronic plus binge ethanol feeding identified a negative correlation between the downregulation of solute carrier organic anion transporter family, member 1a1 (Slco1a1) mRNA in the liver and serum cholic acid levels (238). Furthermore, prokineticin 2 (Prok2) mRNA, was significantly downregulated in ethanol-fed mice and negatively correlated with the abundance of Allobaculum, Coprococcus, Lachnospiraceae, Lactococcus, and Cobriobacteriaceae while positively correlated with Bacteroides (238). Nevertheless, future computational studies should validate these findings in human patients and address their pathobiological significance.
GWAS and epigenetic studies have played pivotal roles in the identification of potential genetic risk factors and the comprehension of epigenetic modulations in the progression of ALD (239). Because GWAS requires an extensive number of samples for accurate statistical power, effect size detection, multiple testing correction, and rare variant detection, the United Kingdom Biobank (UKB) has emerged as a widely utilized resource, harboring more than 500,000 patients, although the sample size for patients specific to AUD and ALD is significantly smaller and underrepresented (240). Several GWAS studies have juxtaposed heavy drinkers with patients suffering from later stages of progressive ALD, including AC and HCC. Through this comparative approach, genes of interest PNPLA3, MBOAT7, transmembrane 6 superfamily member 2 (TM6SF2), and hydroxysteroid 17-beta dehydrogenase 13 (HSD17B13) correlated with a higher genetic risk and increased liver fibrosis have been identified, presenting themselves as potential therapeutic targets (241). For instance, a study by Linden et. al. demonstrated utilizing in vivo murine models that downregulation of Pnpla3 in MASLD can ameliorate fibrosis in the liver (242). Similarly, another study by Varadharajan et. al. showed that a loss of function in Mboat7 resulted in further progression of ALD in mice (192). It should be noted that diseases like MASLD can have similar pathologies and some mechanisms of action as ALD. For example, genetic polymorphisms in MBOAT7 are associated with both ALD and MAFLD (243). It is noteworthy that the development of liver disease is not universally observed among heavy drinkers, and numerous genetic alterations identified in association with heavy drinking and AUD may not manifest in patients with ALD (244); however, this may be a consequence of the inclusion of heavy drinkers as controls in most GWAS studies of ALD. These distinctions contribute to the inherent complexity of studying ALD. Finally, it is important to recognize that the number of patients enrolled in GWAS studies for ALD is very limited; future studies with larger sample sizes targeting genetic susceptibility for different types of ALD (AH vs AC) are needed, as well as pharmacogenetic studies to potentially identify AH patients resistant to glucocorticoid therapies. Interestingly, to date, none of the genes identified to be associated with increased risk for AC have been implicated in the gut-liver axis of disease progression. This is a promising area for future investigations.
In addition to GWAS, epigenetics offers an alternative avenue for understanding how post-translational modifications impact progression of ALD, employing assays such as DNA methylation and chromatin immunoprecipitation through DNA Sequencing (ChIp-seq). Dysregulation of HDACs is observed in cases of excessive alcohol consumption, with one study noting increased histone acetylation of the promoter region of Pnpla3 in mouse models (245)(246). This accumulated knowledge of disease progression mechanisms has given rise to a class of medications termed “epidrugs” designed to potentially target aberrant epigenetic modifications stemming from alcohol damage (245). However, further research is imperative to validate findings and acquire a more systematic understanding of ALD pathology. Additionally, computational approaches that combine meta-analyses and multiomics could provide an accurate and holistic comprehension of the mechanism of action alcohol presents on liver damage, though further exploration and availability of samples are needed.
Emerging therapies directed at the gut-liver axis show promise in mitigating ALD (Table 2). The diverse outcomes and historical success of antibiotic treatment, FMT, and probiotic therapies in ALD are outlined in Table 2. While preclinical studies on prebiotics, postbiotics, BAs regulation, and bacteriophage therapies are also summarized in Table 2, further research is essential to translate these findings into clinical studies.
FMT has gained significant relevance in the field showing promise in clinical trials for ALD (247, 248). However, FMT’s exact mechanisms remain unclear and challenges persist due to variations in administration routes and frequencies among clinical trials, along with concerns about transferring multidrug-resistant bacteria (249). Recent studies emphasize the importance of characterizing and standardizing donor stool for FMT (249). To tackle these issues, ongoing clinical trials are examining the safety, tolerability, and effectiveness of capsules that incorporate bacteria obtained from healthy individuals for the treatment of alcohol craving and consumption (NCT05548452 and NCT04014413). Other studies seek to characterize the intestinal microbiome in individuals with severe AH and examine the safety and trends in enhancing microbiome diversity after administering lyophilized capsules containing microbiota suspension from carefully screened healthy donors (NCT05006430). Additionally, there is an investigation into assessing the survival benefit at 3 months for patients with severe AH who are ineligible for steroid treatment (NCT05285592).
Use of advanced technologies and computational methods, coupled with the availability of animal models of ALD and biological samples from patients with ALD, has led to a growing understanding of the molecular and cellular effects of alcohol on the gut-liver axis. Importantly, this knowledge has contributed to a robust interest in the scientific community to test therapeutics that have potential to maintain and/or restore homeostatic interactions to mitigate the development and progression of ALD.