Alcohol-related liver disease (ALD) has become an important public health problem with high incidence and mortality (Llopis et al., 2016; EASL Clinical Practice Guidelines: European Association for the Study of the Liver, 2018). It has a variety of clinical presentations with a spectrum from alcoholic hepatitis to liver cirrhosis. ALD has become a worldwide public health concern (approximately 4.2% of the population) and caused approximately 5.9% of all deaths every year (2018). However, 2.3 billion people between the ages of 15 and 59 were still active alcohol drinkers (Borrelli et al., 2022). The economic burden of ALD contributes significantly to the healthcare cost, which has been estimated at 125 billion Euros in the European Union and 249 billion dollars in the United States in 2010 (Axley et al., 2019).

The gut microbiome thrives in our alimentary tract in a symbiotic manner and affects the function of the gastrointestinal tract or liver, and vice versa (Gupta et al., 2021). In the last decade, lots of articles illustrated the pathophysiology of ALD and gut microbiota (Mutlu et al., 2012; Leclercq et al., 2014b; Bajaj et al., 2017; Brandl et al., 2018; Ciocan et al., 2018; Chu et al., 2020; Lang et al., 2020; Bajaj et al., 2021). To be specific, the gut microbiota contributed to individual susceptibility to ALD through the metabolic pathway, immune signaling pathway, and gut-brain axis (Queipo-Ortuño et al., 2012; Llopis et al., 2016; Bajaj, 2019). Moreover, alcohol dependence and liver dysfunction have a negative impact on the gut microbiota in ALD patients (Dubinkina et al., 2017). This current review mainly focused on the pathophysiology of gut microbiota and clarifies the gut-brain cross talk in ALD, which may provide the opportunity of identifying target points for future therapeutic intervention in ALD. We believe that the review will enhance our understanding of the topic and provide clear information for the potential therapeutic intervention through the targeting therapy of gut microbiota.

The literature search was performed on the PubMed database for studies published between 1/2010 to 6/2022 by using the search strategy shown in the Supplementary Material S1. The current review included data from randomized controlled trials, observational cohort studies, and animal studies. Studies were excluded: 1) All reviews including meta-analysis or the sample size too small in the ALD arm (n < 10); 2) Patients or animals co-infected with other liver diseases. 3) Not related to our topic. Based on the keyword search on the aforementioned, a total of 1398 articles were identified, and 69 articles were finally enrolled (Supplementary Material S2). The literature search was performed by 2 authors (LZ and YW). The disagreement on the study selections was arbitrated by the discussions with the corresponding author (HX and CP) and resolved with the group consensus.

Alcohol has a direct toxic effect on endoplasmic reticulum structure and function in hepatocytes, also resulting in intestinal stem cell dysregulation and long-lasting intestinal damage (Howarth et al., 2012; Lu et al., 2017). Chronic alcohol intake also enhances Tumor Necrosis Factor (TNF)-α expression in the jejunum of mice, human intestinal monocytes, and macrophages (Chen et al., 2015a). Elevated systemic TNF-α contributed to the activation of TNF receptor I on intestinal epithelial cells and phosphorylation of myosin light chain kinase to redistribute tight junction proteins and increases intestinal permeability (Chen et al., 2015a). Additionally, alcohol-induced gut leakiness from binging alcohol has been linked to the increase of cytochrome P450-2E1(CYP2E1), apoptosis of enterocytes, and production of nitration followed by ubiquitin-dependent proteolytic degradation of the junctional complex proteins (Abdelmegeed et al., 2013). Alcohol suppressed antimicrobial-regenerating islet-derived (REG)-3B lectins and REG3G gene and protein expression. REG3B and REG3G, as secreted C-type lectins mainly expressed in the intestinal epithelial and Paneth cells, had bactericidal activity against Gram-positive and Gram-negative bacteria respectively (Wang et al., 2016). Moreover, alcohol feeding of mice also disrupted the gut mucus layer and diminished mucosal thickness (Grander et al., 2018). Several studies suggest that intestinal aerobes and facultative anaerobes, like Ruminococcus, Prevotella, Collinsella, Staphylococcus, Corynebacterium, Escherichia, Streptococcus, were responsible for alcohol dehydrogenase-mediated ethanol oxidation under aerobic and even microaerobic conditions (Suen et al., 2011; Tsuruya et al., 2016). The accumulation of acetaldehyde further inhibited the expression of tight junction protein and promotes mucosa-associated microbiota translocated through the intestinal barrier to mesenteric lymph nodes and liver (Wang et al., 2016).

Alcohol can also affect gut bacteria profiles and features directly or change the composition of gut biofilms through bile acid synthesis indirectly. In the presence of alcohol, pathogenic bacteria like Acinetobacter baumannii enhanced its virulence significantly correlates with the acidification of bacterial cultures (Nwugo et al., 2012). Chronic alcohol abuse suppressed the bacterial genes involved in the biosynthesis of saturated long-chain fatty acids (LCFAs), inhibiting the proliferation of Lactobacilli which metabolizes saturated LCFA (Chen et al., 2015b). Primary bile acids such as cholic acid and chenodeoxycholic acid were synthesized through the Cholesterol 7-alpha-hydroxycholesterol (CYP7A1) classical pathway or the Cholesterol 27-hydroxycholesterol (CYP27A1) alternative pathway (Chiang, 2013). Alcohol upregulated the expression of CYP7A1 and CYP27A1 via the activation of hepatic cannabinoid receptor type 1 and suppressed fibroblast growth factor 15 gene expression to promote the synthesis of bile acids (Xie et al., 2013). Continued alcohol misuse also induced human fibroblast growth factor 19 gene expression in biliary epithelial cells and ductular cells, followed by total and conjugated bile acids increased significantly (Bajaj et al., 2017; Brandl et al., 2018). Bile acids exert a bacteriostatic effect, directly damaging DNA and destroying the bacterial membrane of bacterial. (Merritt and Donaldson, 2009; Kakiyama et al., 2013).

Alcoholics with dysbiosis had a reduced bacterial diversity, decreased abundance of Bacteroidetes, and increased Proteobacteria (Mutlu et al., 2012). The changes of gut microbiotas have been observed at both of the phylum and family levels in patients with alcohol dependence syndrome (ADS) (Leclercq et al., 2014b; Dubinkina et al., 2017). At the phylum level, Bacteroidetes and Firmicutes of alcoholics decreased significantly, whereas Proteobacteria, Fusobacteria, and Actinobacteria increased. At the family level, there was a decrease in Ruminococcaceae significantly but an increase in both Lachnospiraceae and Enterobacteriaceae (Leclercq et al., 2014b; Dubinkina et al., 2017). Also observed that Prevotellaceae was enriched in alcohol-related cirrhosis (ALC) compared with controls and other etiologies cirrhotics (Chen et al., 2011). Moreover, Bajaj has summarized articles about alcoholic cirrhosis and microbiota composition or function in humans from 2012 to 2018 (Bajaj, 2019). There were 2 updated novel-related studies from 2018 to now. It is shown that ALD patients who had a lower fungal diversity with an overgrowth of Candida and higher serum anti-Saccharomyces cerevisiae antibodies had increased mortality (Lang et al., 2020).

Upregulation of systemic inflammation including inflammation in the oral cavity, gut, and liver is postulated to be a motivating factor of ALD (Acharya et al., 2017). The oral mucosa serves as the first line of defense due to the function of salivary immunoglobulins, agglutinins, histatins, and lysozyme. Enrolled 102 cirrhotics (38% alcohol-related) to analyze salivary and stool microbiomes. They observed that the lower salivary microbiota ratio (calculated by Lachnospiraceae + Ruminococcaceae + Clostridiales Incertae Sedis XIV/Streptococcaceae), the lower cirrhosis dysbiosis ratio (calculated by Lachnospiraceae + Ruminococcaceae + Clostridiales Incertae Sedis XIV + Veillonelllaceae/Enterobacteriaceae + Bacteroidaceae), indicates oral dysbiosis and gut dysbiosis respectively (Bajaj et al., 2015). Furthermore, patients marked with salivary dysbiosis confront a higher 90-day liver-related hospitalization rate than those not. The possible cause may be related to oral microbiota invading the gut under the changes in gut pH or bile acid dysregulation of ALD. Patients with ADS had significantly increased oral-oriented microbes like Lactobacillus salivarius, Veillonella parvula, and Streptococcus salivarius (Dubinkina et al., 2017). When applying periodontal therapy in alcohol-related cirrhotics, the model for end-stage liver disease (MELD) scores were improved as the results of alleviating oral-gut dysbiosis, and decreased endotoxin, lipopolysaccharide-binding protein (Bajaj et al., 2018).

Gut microbe-derived metabolites mainly included long-chain fatty acids (LCFAs), short-chain fatty acids (SCFAs), mucus, secondary bile acids, indole or phenol derivatives, and vitamin B (Table 1) (Fan and Pedersen, 2021). Methods like integrated analyses of the microbiome and linked metabolomes as well as microbiome-wide association studies can be used to profile the gut metabolome characteristic, including shotgun-based sequencing, various bioinformatic algorithms, strain-level profiling, and so on (Fan and Pedersen, 2021). Bioinformatics analyses such as the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis have been applied to identify potential mechanistic links between the gut microbiome and metabolome. KEGG is a database and helpful tool for understanding most of the known metabolic pathways based on sequence similarity to proteins with known functional characteristics. For ALD patients, seven KEGG pathways were significantly increased (Galactose metabolism Porphyrin, chlorophyll metabolism, ABC transporters, phosphotransferase system, fructose or mannose metabolism, glutathione metabolism, and biosynthesis of siderophore group non-ribosomal peptides) (Dubinkina et al., 2017). Besides, there were significantly lower fecal metabolites focused on bioenergetics (citrate, malate, or phosphate), amino acids (threonine, ornithine, or serine), and pyrimidine intermediates (ribosine, orotic acid, or hexonate) (Kakiyama et al., 2013; Bajaj et al., 2017; Brandl et al., 2018; Ciocan et al., 2018; Fan and Pedersen, 2021).

The appropriate immune response in the liver is dependent on the pattern recognition receptors expressed by liver sinusoidal endothelial cells, hepatocytes, Kupffer cells, dendritic cells, and liver-resident lymphocytes through the activation of receptors like Toll-like receptors (TLRs) (Jenne and Kubes, 2013). In ALD patients, gut microbiota can get across the intestinal barrier and circulate to the liver through the portal vein to interact with the liver (Bruellman and Llorente, 2021). As a result, a prolonged inflammatory immune response is activated, as shown in Figure 1.

The changes in gut microbiota also influence the function of the brain. The gut microbiota communicates with the brain via the neuroendocrine or neuroimmune mechanisms, known as the gut-brain axis. Gut microbiota can produce neurotransmitters, precursors like 5-hydroxytryptamine(5-HT), dopamine, noradrenaline, L-glutamic acid, γ-and aminobutyric acid (GABA), which can regulate the function of the central nervous system (Seo et al., 2020). In addition, other metabolites of gut microbiota, such as acetate, may affect the levels of 5-HT or other neurotransmitters and therefore influence human behavior (Holzer et al., 2012). Recently, Qamar et al., 2019 used the Ingenuity Pathway Analysis Software to identify the alteration in neurotransmitters associated with alcohol consumption.

Several bacteria such as Streptococcus, Acinetobacter, Escherichia coli, and Clostridia can produce 5-HT directly or generate it via a bacteria-mediated tryptophan metabolism pathway (Williams et al., 2014; Zhang et al., 2021). The 5-HT levels were always significantly increased in ALD patients when compared with those in non-ALD patients (Williams et al., 2014; Zhang et al., 2021). The activation of 5-HT receptors promoted dopamine release in the reward circuitry and increased the risk of alcohol addiction (Enoch et al., 2011). In addition, central noradrenergic neurons can convert dopamine to norepinephrine (Enoch et al., 2011). Norepinephrine can affect the growth of Prevotella and other anaerobic bacteria as it can increase virulence gene expression and enhance iron acquisition in Clostridium (Boyanova, 2017). Furthermore, several bacteria strains have been reported to be able to produce dopamine and norepinephrine. They included E. coli, Proteus vulgaris, Staphylococcus aureus, and Bacillus subtilis (Strandwitz, 2018). Noradrenergic neurons also can be regulated by transmitters like glutamate and GABA. The GABA was produced through the α-decarboxylation of glutamate by the enzyme glutamate decarboxylase (Sarasa et al., 2020). Chronic alcohol exposure increased the extracellular glutamate concentration and suppressed GABA activity, leading to decrease glutamatergic synaptic transmission and further alcohol consumption. However, acute alcohol exposure resulted in the opposite effects (Alasmari et al., 2018). As GABA-producing bacteria and neuroactive bacteria, L. reuteri and L. rhamnosus, act on intrinsic primary afferent neurons in the mouse model (Mao et al., 2013; Pokusaeva et al., 2017; Tsai et al., 2020; Zheng et al., 2020), which transmit microbial messages to the brain via the vagus nerve. As a result, alcohol-induced hepatitis was alleviated.

The anxiety/depression-like behaviors in patients with chronic alcohol intake have been associated with the increase (Zhao et al., 2020) of pathogenic colonies like Actinobacteria and Adlercreutzia, which drive changes in inflammatory signaling and cytokine release, leading to GABA-1 receptor changes in the prefrontal cortex and the brain-derived neurotrophic factor/α1 subunit of γ-aminobutyric acid A (Kelly et al., 2016; Xu et al., 2019), and finally influenced people’s emotions. For example, at the withdrawal affect stage, microbiome-immune disruptions have been identified as potential mediators of front-limbic anomalies and derived emotional dysregulation in the alcohol addiction cycle (Gorky and Schwaber, 2016). On the one hand, the “leaky gut” allows inflammatory molecules to leak outside of the gut into the blood to generate peripheral inflammation, further activating microglia and causing a neuroinflammatory response (Skinner et al., 2009; Keita and Söderholm, 2010). On the other hand, such alterations may prime the vagus nerve to alter its neuroprotective afferent signals or promote vagal signals that are in some harmful way (Gorky and Schwaber, 2016; Carbia et al., 2021). With the brain primed by vagal afferent alterations, it may be more susceptible to influence from the peripheral inflammatory response that occurs in response to alcohol withdrawal (Figure 3 and Suplementary Material S3). Another common brain disorder among ALD patients is dementia which mostly occurs in women and may triple the risk of death (Bahorik et al., 2021). Gut microbiota dysbiosis such as the increase of Muribaculaceae and the decrease of Lactobacillus can lead to the peripheral accumulation of phenylalanine and isoleucine, which stimulates the proliferation of pro-inflammatory T helper 1 cells, contributing to dementia-associated neuroinflammation (Sahlman et al., 2016). Gut-resident Cyanobacteria-generated neurotoxins involving β-N-methylamino-l-alanine and saxitoxin may further aggravate neurodegenerative disease (Alkasir et al., 2017).

Hepatic encephalopathy (HE) is one of the end-stage manifestations of ALD, characterized by disturbances in mental status and neurological function. In cirrhosis, ammonia crossed the blood-brain barrier and led to astrocyte swelling and reactive oxygen species formation (Alsahhar and Rahimi, 2019). The increased Klebsiella, Proteus, and E. coli were related to the increased plasma levels of ammonia and endotoxin, contributing to the occurrence of hepatic encephalopathy in ALD (Wang et al., 2019). Moreover, the gut microbiota diversity was decreased during acute episodes of overt HE, and the relative abundances of Alistipes, Bacteroides, and Phascolarctobacterium were respectively associated with HE recurrence and overall survival during the subsequent 1-year follow-up (Sung et al., 2019).

The gut microbiota dysbiosis in ALD was initiated by oral-gut dysbiosis, followed by gut microbes’ colonization and circulating microbiota change. The factors that persistently aggravate disease progression in ALD patients have been explored in recent studies, which included Alcohol intake, gut microbiota change, gut metabolic dysregulation, activated inflammatory immune response, and neurological or behavioral deficits, giving a better understanding of the pathophysiology of gut dysbiosis in ALD and the pathway of the gut-brain cross talk. Further investigations into the pathogenic of dysbiosis and the mechanisms of certain protective bacteria in ALD may provide the therapeutic targeting points for the treatment of ALD and aid in selecting effective regimens with a favorable safety profile.