The Molecular and Mechanistic Insights Based on Gut–Liver Axis: Nutritional Target for Non-Alcoholic Fatty Liver Disease (NAFLD) Improvement

Bacteroidetes and Firmicutes account for the majority (above 90%) of human gut microbiota, while other phylums of bacteria exist in small amount, including Proteobacteria, Verrucomicrobia, Actinobacteria, Fusobacteria and Cyanobacteria [26]. Three major classes (Bacilli, Clostridia, and Mollicutes) encompassing over 250 genera are included in Firmicutes, almost all of which are gram-positive [27]. In contrast, all gut bacteria belonging to the phylum of Bacteroidetes are gram-negative, composed mainly of the genera of Bacteroides, Prevotella, Alistipes, and Parabacteroides [28]. Accumulating clinical and animal studies have indicated that NAFLD is intimately associated with disruption of the balance between Firmicutes and Bacteroidetes. Although available studies have pointed out the association between the increased abundance of gut Firmicutes and NAFLD, other findings confirm that overgrowth of Bacteroides plays a key role in the development of NAFLD [29,30,31,32]. In addition, the severity of NAFLD is associated with an increase in fecal abundance of Bacteroides, accompanied by a decrease in the level of Prevotella [33]. Other evidence suggests the fecal abundance of Anaerosporobacter and Faecalibacterium are lower, whereas Allisonella and Parabacteroides are higher in non-alcoholic steatohepatitis (NASH) patients [31]. Metagenome sequencing revealed Bacteroides vulgatus and Eubacterium rectale present the highest abundance in the feces from mild and moderate NAFLD, while Bacteroides vulgatus and Escherichia coli reach the most abundant in liver fibrosis [34]. To sum up, the above evidence supports the theory that the development and progression of NAFLD is closely linked to gut microbiota dysbiosis.

A disturbance of the intestinal microbiota in response to unbalanced diet (e.g., diet high in saturated fat and fructose) may elicit an increase in intestinal permeability, leading to chronic inflammatory condition [35]. Remarkably, a continuous low-grade inflammatory state may result in an acceleration in the progression from hepatic simple steatosis to NASH [36,37]. Serving as a key constituent of the outer membrane of cell wall in most gram-negative bacteria, lipopolysaccharides (LPS) (also known as endotoxins) has well been identified as a major factor in activation of innate immunity and are therefore suggested as a vital pathogenesis for NAFLD [38,39]. Indeed, patients with hepatic steatosis and inflammation present an elevation in the serum LPS [40]. This may be associated with a rise in the abundance of specific gram-negative bacterial genus, including Bacteroides, Enterobacteria, Escherichia, and Proteus, as observed in NAFLD patients [33,41]. Recently, study on germfree mice have uncovered that colonization of one of the three nonvirulent LPS-producing strains identified from obese human gut induces NAFLD in combination with high-fat diet feeding [42]. These results were further corroborated by either interference in the synthetic pathway of bacterial LPS or deletion of the LPS receptor toll-like receptor 4 (TLR4). As a main TLR4 ligand governing the release of pro-inflammatory cytokines, LPS can potentiate the susceptibility of NAFLD through the augmentation of inflammatory responses [43,44]. In addition to LPS, other pro-inflammatory bacterial components such as peptidoglycans, lipoteichoic acids, bacterial DNA, and extracellular vesicles have also been noted in recent years [45]. However, the underlying mechanism by which the role of these components produced by specific gut bacterial strains in NAFLD requires more extensive and explicit studies.

Considerable attention has been also paid to the role of gut microbiota-derived metabolites in the pathological process of NAFLD. A clinical research with 330 subjects by Barrea et al. reveals that the circulating levels of trimethylamine-N-oxide (TMAO) is a novel indicator of metabolic syndrome and NAFLD [46]. Although TMAO is synthesized in the liver, trimethylamine (TMA), the precursor of TMAO, is generated from gut bacteria. L-carnitine, choline, or betaine have been the major substrates for TMA synthesis by gut bacterial strains (e.g., Clostridium asparagiforme, Clostridium sporogenes, Clostridium hathewayi, Escherichia fergusonii, Anaerococcus hydrogenalis, and Proteus penneri) [47]. TMAO is formed by the oxidation of TMA following the catalysis of flavin-containing monooxygenase (FMO) enzymes in the liver [48]. The precise mechanisms involving the relationship between TMAO and the initiation and progression of NAFLD remains to be clarified. However, it has been verified that TMAO exacerbates hepatic steatosis by blocking the farnesoid X receptor (FXR) signaling activated by bile acid [49]. By contrast, the activation of FXR signaling have shown a protective effect against NAFLD including liver steatosis and inflammation [50,51]. It should be mentioned that N,N,N-trimethyl-5-aminovaleric acid (TMAVA), a newly identified gut microbiota metabolite, appears as a key molecule exacerbating high-fat diet-induced liver steatosis through the inhibition of carnitine synthesis accompanied by a reduction in mitochondrial fatty acid β-oxidation in hepatic tissue [52]. This metabolite is yielded by Enterococcus faecalis and Pseudomonas aeruginosa using trimethyllysine [52]. In future study, more metabolites are required to be identified and examined from particular gut strains to provide more extensive knowledge concerning the interaction of gut microbes with NAFLD.

Intestinal barrier performs an effective defensive action against the translocation of harmful substances including bacteria, bowel luminal antigens and inflammatory factors, and is usually assessed by intestinal permeability [53]. Gut microflora dysbiosis has been closely followed by increased intestinal permeability and subsequent bacterial translocation to the liver, leading to the release of inflammatory cytokines and free radicals from activated Kupffer cells [54]. Ample evidence reveals that NAFLD individuals develop an imbalanced gut microbiome and compromised intestinal permeability [55,56,57]. Inflammatory bowel disease (IBD) patients have a high incidence (up to 33.6%) of NAFLD even independent of metabolic risk [58], which has been closely associated with impaired intestinal barrier function. An impairment of the integrity of intestinal barrier initiated by disordered gut microbiome has been a prerequisite for NASH [59]. Consequently, interventions preserving the integrity of the intestinal barrier may contribute to attenuating or preventing the progression of NAFLD. It should be noted that zonulin serves as a key protein regulating intestinal permeability reversibly by adjusting the size of tight junctions between epithelial cells [60,61]. Serum circulation levels of zonulin present a significant positive correlation with pathological indicators in NAFLD, especially in NASH [62]. Of note, it has been shown that the endogenous ethanol content of NAFLD patients is substantially higher than that of healthy individuals [63]. The impaired intestinal integrity is partially associated with the endogenous production of ethanol. Intriguingly, recent evidence found that specific gut strains (Klebsiella pneumoniae) from NAFLD patients possess the ability to produce high level of endogenous ethanol, thus exacerbating the compromised intestinal barrier function and hepatic steatosis [64].

The circulation of bile acids between liver and intestine is quite active. Bile acids are known as amphipathic hydroxylated steroids synthesized from catabolism of cholesterol in the liver and are released into the small intestine in the form of bile salts from gallbladder [65]. They function not only to facilitate the emulsification, transport, and absorption of lipids and fat-soluble vitamins but also to regulate the balance of glucolipid metabolism and immune responses [66]. The synthesis of primary bile acids cholic acid (CA) from cholesterol is initiated by rate-limiting enzyme cholesterol 7a-hydroxylase (CYP7A1) in hepatocytes [67]. Alternatively, sterol-27-hydroxylase (CYP27A1) catalyzes the production of chenodeoxycholic acid (CDCA) or muricholic acids (MCAs, only in mice) from cholesterol [68]. The formation of secondary bile acids (deoxycholic acid (DCA), lithocholic acid (LCA), and ursodeoxycholic acid (UDCA)) is achieved via the modification of bile salts (primary bile acids conjugated with glycine or taurine) by gut microbiota [69]. In mice, other secondary bile acids including Omega-MCA (ωMCA), hyocholic acid (HCA), murideoxycholic acid (MDCA), and hyodeoxycholic acid (HDCA) are formed from MCAs by microbial epimerization and dehydroxylation [68]. About 95% of the bile acids are reabsorbed into the terminal ileum and are transferred to the liver via the portal vein [70]. Importantly, bile acids regulate bile acid receptors including FXR and Takeda G-protein receptor 5 (TGR5) which are known to modulate insulin sensitivity, glucolipid metabolism, energy homoeostasis, inflammatory responses, and intestinal barrier function [71,72,73,74]. FXR and TGR5, in turn, play a crucial role in the feedback regulation of bile acid homeostasis [75]. Imbalance of intestinal flora may lead to abnormal size and composition of bile acids and dysregulation of FXR and TGR5 signaling [76]. Specific bile acid conjugates esteem differential feedback of regulatory influences in FXR signaling. Since the FXR is a primary controlling actor for bile acid synthesis and bile flow, this would imply that patients with NAFLD manifest altered bile acid kinetics. For instance, a metabolomic study revealed that NASH patients presented higher plasma concentration of glycine or taurine-conjugated CA, and glycine-conjugated CDCA, compared with healthy subjects [77]. FXR and TGR5 have become important intervention targets for NAFLD [78]. Activation of FXR ameliorates NASH through multiple mechanisms, including the suppression of monocyte and neutrophil infiltration and the attenuation of NF-κB-mediated inflammation signaling [50]. TGR5 activation by agonist also has been proved to improve liver steatosis in a mouse model of NAFLD [79].

The use of probiotics has become an enticing and promising approach for the prevention and therapy of NAFLD. Probiotics refer to the live microorganism strains with sufficient quantity which provide the host with a health benefit or improved pathological conditions [80]. These strains induce a competitive exclusion of pathogenic bacteria to ensure a healthy and balanced intestinal microflora ecosystem favoring epithelial barrier and host immune function [81,82]. Plentiful studies have shown the extensive salubrious effects of probiotics (e.g., Lactobacillus and Bifidobacterium) on liver disease. As shown in high-fructose or high-fat diet-induced experimental NAFLD models, Lactobacillus rhamnosus GG ameliorated NAFLD by (1) control of gut microbiome; (2) repair of intestinal barrier; and (3) suppression of hepatic steatosis, inflammation, and lipid accumulation [83,84]. Bifidobacterium serves to protect against secretion of pro-inflammatory cytokines and dysfunction in intestinal barrier both in vitro and in vivo [85]. Results from a randomized clinical trial indicates that patients with NAFLD receiving a complex-probiotic prepared by 14 probiotic bacteria genera belonging to Lactobacillus, Lactococcus, Bifidobacterium, and Propionibacterium presented a reduction in fat accumulation in liver and aminotransferase activity, and pro-inflammatory factor levels including tumor necrosis factor-α (TNF-α) and interleukin 6 (IL-6) in serum [86]. However, probiotics therapy to NAFLD probably is not involved in the regulation of TMAO levels, according to the observations from high-fat diet feeding mice and patients with metabolic syndrome [87,88]. As for bile acids metabolism, a randomized double-blind crossover trial revealed that consumption of Bacillus subtilis R0179 and Bifidobacterium animalis subsp. lactis. B94 for 6 weeks showed a significant increase in serum deconjugated bile acids [89]. Evidence from mice indicated that administration of a probiotic mixture of VSL#3 facilitated hepatic synthesis and fecal excretion of bile acids via down-regulation of fibroblast growth factor 15 (Fgf15) [90], while Fgf15 is known to expedite hepatic fibrosis and even cancer in the background of NAFLD [91,92].

Dietary fiber is characterized as nondigestible carbohydrate polymers containing three or more monomeric units that are included in plant diets such as grains, legumes, fruit, and vegetables [93,94]. Comparative nutritional evaluation studies suggested that patients with NAFLD consumed less dietary fiber than healthy individuals [95,96]. In comparison, the consumption of dietary fiber has been a favorable factor resistant to the progression of NAFLD [97]. Fibers leaving the upper gastrointestinal tract are degraded into SCFAs including acetate, butyrate, and propionate through the fermentation of gut microbiota at the lower part of the gastrointestinal tract [98]. Such SCFAs participate in a broad variety of physiological processes related to metabolism, immunity, endocrine, and inflammation [99]. Furthermore, dietary fibers serve to adsorb bile acids and cholesterol in the gastrointestinal tract, thus changing the bioavailability of secondary metabolites and playing a role in management of hyperlipidemia [100,101]. An improvement in fatty livers index, hepatic steatosis index, and NAFLD hepatic fat score have been shown by subjects who consume higher insoluble fiber (≥7.5 g/day) [102]. Consumption of insoluble dietary fiber from soy hull induced an alteration in fecal microbiota composition, in particular the abundance of Lactobacillales and Bifidobacteriales known as probiotics [103]. Additionally, the increase in nutritional fibers improve hepatic steatosis and liver function affiliated with dramatically lowered serum level of zonulin protein, denoting an enhanced barrier function and thus a reduced intestinal permeability [104].

Functional oligosaccharides have become an increasing focus of interest in NAFLD amelioration. Functional oligosaccharides (e.g., fructo-oligosaccharides, galacto-oligosaccharides and chitosan oligosaccharides) are prebiotics that have been shown (1) to facilitate the proliferation of beneficial flora in gut; (2) to suppress pathogenic bacteria; and (3) to participate in the control of glucolipid metabolism, immunity and oxidative injury [105,106]. Dietary fructo-oligosaccharides contribute to the recovery of normal intestinal microbiome and intestinal barrier function, thereby alleviating steatohepatitis in a methionine-choline-deficient mouse model [107]. Moreover, a fructo-oligosaccharides supply reversed high-fat diet feeding-induced reduction in level of Bifidobacteria and increase in plasma endotoxin content in mice [108]. Evidence from integrated lipidome and gut microbiome uncovered that chitosan oligosaccharide serves to ameliorate hepatic injury, steatosis, and inflammation as well as to reinforce the intestinal barrier [109]. Also, this oligosaccharide induced a reduction in the cecal abundance of Mucispirillum and an elevation in the content of Coprococcus [109]. Besides, the contribution of oligosaccharides to bile acid secretion and bile acid pool kinetics has been demonstrated in various animal models [110,111,112]. Overall, the findings from above animal studies hint that inclusion of oligosaccharides in food is a healthy way of providing beneficial effects in liver health. Future study is expected to evaluate the efficacy of oligosaccharides in NAFLD patients through clinical trials.

L-tryptophan is an amino acid essential for human and animals and exists in foods derived from protein such as meats, milk, nuts, and seeds [113]. The biochemical and physiological functions of tryptophan are not only restricted to its host metabolites (kynurenine, serotonin, and melatonin), but a wide variety of gut microbiota-sourced metabolites have also been identified [114,115]. Through the catalysis of enzymes expressed in gut microbiota, dietary tryptophan is converted into indole and its derivatives (indole lactic acid (ILA), indole acetic acid (IAA), indole aldehyde (IAld), indole propionic acid (IPA), and indoleacrylic acid (IA)). Currently, various evidence has associated indoles with the improvement of gut and liver health. For instance, IAld induced STAT3 (signal transducer and activator of transcription 3) phosphorylation by stimulating the secretion of interleukin-22 (IL-22) via aryl hydrocarbon receptor (AhR) to promote the proliferation of the intestinal epithelial cells, which in turn restores the barrier function [116]. Treatment with indole attenuated high-fat diet-induced severity of hepatic steatosis and inflammation in mice, which was connected with the regulation of PFKFB3 (6-Phosphofructo-2-Kinase/Fructose-2,6-Biphosphatase 3) and the normalization of gut microbiome [117]. Study on rats revealed that IPA facilitated the protection against steatohepatitis induced by high-fat diet, as demonstrated by the improvement of gut dysbiosis and intestinal epithelial barrier function and the inhibition of pro-inflammatory signaling [118]. In addition, it has been shown that IAA displays anti-oxidative and anti-inflammatory activity, which is beneficial in the resistance to NAFLD induced by high-fat feeding and pro-inflammatory response in macrophages [119,120]. Interestingly, Bifidobacterium strains collected from human infants have been turned out to produce ILA [121], the favorable effect of these strains may be associated with ILA, which deserves to be corroborated in future studies. L-glutamine is the most plentiful amino acid in the body dispersed predominantly in the skeletal muscle, small intestine, liver, and kidney [122]. This amino acid serves as a substrate for the synthesis of antioxidant glutathione [123]. Importantly, supplementation with parenteral glutamine improved intestinal barrier function and decreased the occurrence of infection [124]. These effects of glutamine are not only mediated by the redox status, but also depend on its regulation on immune response and signaling pathways (e.g., PI3K (phosphatidylinositol 3-kinase)/PKB (protein kinase B)/mTOR (mechanistic target of rapamycin)) [122,125]. Obese adults receiving l-glutamine for 14 consecutive day exhibited a reduction in fecal abundance of Dialister, Dorea, Pseudobutyrivibrio, and Veillonella, and an increase in Prevotella, compared with the overweight individuals daily supplemented with equivalent l-alanine [126], suggesting a modulation on the composition of gut microbiota by l-glutamine. Although more clinical and mechanistic studies in relation to the effect of l-glutamine on NAFLD remain to be performed, research on mice has shown a protective role of oral administration of L-glutamine against western-style diet-induced NASH [127].

The source of ω-3 PUFAs include alpha-linolenic acid (ALA, 18:3) in plant oils and eicosapentaenoic acid (EPA, 20:5) and docosahexaenoic acid (DHA, 22:6) in fish oils. In contrast to the pro-inflammatory effect of ω-6 PUFAs (linoleic acid (18:2), γ-linolenic acid (18:3), arachidonic acid (20:4)), ω-3 PUFA (especially EPA and DHA) are well known to mitigate inflammation [128]. As a result, an increase in dietary intake ratio of ω-6/ω-3 PUFAs has been associated with various pathogenesis of diseases such as cardiovascular disease, arthritis, obesity, diabetes, and liver disease [129,130,131,132,133]. ω-3 PUFAs hold the potential to be a supportive option for the prevention and therapy of NAFLD as observed in various clinical and animal studies [134,135,136,137]. Previous studies have shown an improvement effect of ω-3 PUFAs on hepatic enzymes and triglycerides in liver and serum [138]. In recent years, more attention has been given to the role of ω-3 PUFAs in NAFLD by targeting the gut–liver axis. First, ω-3 PUFA direct exhibit an impact on the composition and diversity of gut microbiota [139]. For example, an endogenous generation of ω-3 PUFA by genetic approach resulted in lowered abundance of pro-inflammatory bacteria (e.g., Proteobacteria) and increased anti-inflammatory bacteria (e.g., Bifidobacterium, Akkermansia muciniphila) in the gut of mice, compared with wild-type littermates [140]. Secondly, dietary ω-3 PUFA supplementation elicits a decrease in plasma levels of endotoxin and TMA derived from bacteria. It has been reported that in contrast to the high concentration of serum endotoxin in response to saturated fat consumption, participants exposed to food with ω-3 PUFAs exhibited a reduction in endotoxin content [141]. Results from animal study have highlighted that DHA-enriched fish oil decreased the production of TMA and expedited TMAO metabolism through enhancing the activities of FMOs [142]. Thirdly, ω-3 PUFA contributes to the integrity of intestinal mucosa and to improvement of intestinal permeability as well as to the reduction of bacterial translocation, which has been reported in copious studies both in vivo and in vitro [143,144,145,146,147]. Lastly, ω-3 PUFAs produce a healthy bile acid reservoir and have been identified as FXR ligands, thereby conducing to the maintenance of liver health and the suppression of hepatic inflammation [50,148,149].

Food rich in carotenoids and polyphenols provides widespread benefits to humans and animals including gut and liver health. Carotenoids are a group of tetraterpenoids that mainly exist in plants, algae, fungi, and archaea [150]. One-month consumption of lycopene, the major carotenoid type in tomatoes, presented an increase in relative abundance of profitable strain Bifidobacterium adolescentis and Bifidobacterium longum in the gut of volunteers, along with an improvement of liver metabolism [151]. Serum level of carotenoids has been shown to be inversely correlated with the impaired intestinal barrier function [152]. Supplementation of carotenoids benefits to the gut barrier function, immune homeostasis, and dysbacteria through expediting immunoglobulin A (IgA) production in the gut [153,154]. Specific microbiota in the gut can be identified and enveloped by IgA, thereby making its translocation through the gut barrier hampered [155]. In addition to the protection against bacterial translocation, carotenoids have been known to provide anti-inflammatory and anti-oxidative effects. Results from animal models revealed the alleviative role of carotenoids on high-fat diet-induced NAFLD via attenuation of inflammatory response and oxidative stress [156,157,158]. The anti-inflammatory effect of carotenoids has also been ascertained in macrophages. For example, the inflammatory response in RAW264.7 macrophages or isolated primary macrophages from mice induced by LPS were shown to be mitigated by β-carotene, which was associated with the inactivation of nuclear factor-κB (NF-κB) and enhanced antioxidant capacity [159]. Although few reports support the regulation of carotenoids on bile acids, FXR, a palliative factor for hepatic steatosis, has been shown to be up-regulated in the liver of rats exposed to high-fat diet feeding with lycopene [160]. Polyphenols appertain to a highly diverse category of compounds that present naturally in plants [161]. Various polyphenols have been proved to display a protection effect on NAFLD by attenuation of insulin resistance, oxidative injury, and inflammation as well as by acceleration of fatty acid β-oxidation [162,163,164]. Clinical trial on volunteers presented ample evidence for reshaping gut microbiota by consumption of polyphenolic extract, leading to a reduction in the levels of serum TMAO [165]. Human study revealed that long-term strawberry polyphenols supplementation changed bile acid profiles and showed an impact on FXR/TGR5 signaling [166]. Polyphenols from berries and pomegranate fruit are metabolized by gut microbiota into urolithin A. This metabolite has been shown to possess anti-inflammatory and anti-oxidative properties and substantially boost intestinal barrier function [167], thus acting as a protector in chronic liver diseases including NAFLD. Additionally, polyphenols function to regulate intestinal barrier integrity. As detailed summarized by Yang et al. [168], a wide range of transcriptional factor, kinases, and enzymes correlate with the beneficial implications of polyphenols on intestinal barrier functions; therefore, these regulatory pathways may convey advantages to liver health.