Mice with gut microbiota were shown to have an increased capability to harvest energy from the gut contents compared with germ-free rats[49]. Metagenomic analyses of the microbiota performed in obese mice and humans revealed an increased capacity for the degradation (fermentation) of carbohydrates[50,51]. This microbial fermentation increases the amount of short-chain fatty acids (SCFAs), such as acetate, propionate, butyrate, and L-lactate. These SCFAs have important roles in the reduction of intestinal pH, in the regulation of energy metabolism, immunity, and adipose tissue expansion and in modulating cancer cell development[52]. Butyrate is used as an energy substrate for colonocytes, acetate is potentially used as a cholesterol or fatty acid precursor and propionate is used as a gluconeogenic substrate in the liver[52-54]. In addition to these functions, SCFAs are also physiological ligands of G-protein coupled receptors GPR43 and 41 (also called free fatty acid receptor 2 and 3, respectively), which are expressed in several cell types (immune cells, endocrine cells, and adipocytes) of host tissues[55,56] (Table 1). Activation of GPR43 by the SCFAs contributes to the inhibition of lipolysis and to adipocyte differentiation, thereby increases the adipose tissue in high-fat-diet-fed mice[57,58]. Monosaccharides produced by microbial fermentation, absorbed and transferred to the liver via portal vein, activate the hepatic carbohydrate response element binding protein (ChREBP) that increases the transcription of several proteins involved in hepatic lipogenesis thus contributing to hepatic fat accumulation[55]. The effect of intestinal microbiota on obtaining energy from the gut contents had also been tested in human studies. Jumpertz et al[59] reported that the amount of stool energy in proportion of ingested calories was positively and negatively correlated with the abundance of phylum Bacteroidetes and phylum Firmicutes in the feces, respectively. Approximately a-150 kcal difference could be achieved with a change of 20% relative increases of Firmicutes and decreases of Bacteroidetes in the stool of lean individuals. Thus, excessive calories taken in the form of SCFAs from microbial fermentation of luminal contents may be a contributing factor in the obese state.

In addition to extract calories from otherwise indigestible dietary polysaccharides, the presence of the intestinal microbiota also stimulates changes in the expression of genes coding for peptides in host tissues, which control energy homeostasis and nutrient availability (Table 1). Bäckhed et al[60] firstly demonstrated that colonizing germ-free mice with gut microbiota had led to a decrease in the intestinal expression of angiopoietin-like factor IV [ANGPTL4, also called fasting-induced adipose factor (Fiaf)], thereby blunting the inhibition of lipoprotein lipase in the adipose tissue. In the study of Mandrad et al[61] inhibition of lipoprotein lipase blocked the dissociation of fatty acids from triglycerides for uptake into tissues and upregulated fatty acid oxidation and uncoupling proteins, and potentially reduced the amount of fat storage. These results may explain why conventionalized mice are more sensitive than germ-free mice to fat storage when fed with a high-fat diet and supports the role of the gut microbiota in the development of obesity[50,52,62].

Obesity, diabetes and insulin resistance are associated with a low grade systemic inflammation[63-65]. Several studies have been conducted on the effect of systemic inflammation on glucose and lipid metabolism, but little is known about its triggers.

Intestinal microorganisms have highly conserved microbial molecules, called “pathogen- associated molecular patterns” (PAMPs) and endogeneous products called “damage-associated molecular patterns” (DAMPs), which are recognized by pattern recognition receptors. These receptors include membranous toll-like receptors (TLRs) and intracellular NOD-like receptors (NLRs). TLRs recognise the potential pathogens in the intestinal lumen and induce the immune response. Microbial products such as LPSs, lipopeptides, DNA and RNA have potentially hepatotoxic effects as they are potent inducers of inflammation. Among 13 known TLRs, TLR2, TLR4 and TLR9 have been shown to play a role in the development of NAFLD. Stimulation of TLRs results in the activation of several different intracellular signaling cascades including stres-activated and mitogen activated protein (MAP) kinases, Jun N-terminal kinases, p38 MAP kinase, interferon regulatory factor 3 and nuclear factor kappa B (NF-κB) pathways[9,66]. NF-κB, is an important transcription factor in the cell, translocates to the nucleus and induces the transcription of a variety of inflammatory cytokines and chemokines such as TNF-α and IL-1β[67,68]. de la Serre et al[69] showed that rats prone to weight gain exhibited an increase in TLR4 activation associated with ileal inflammation, decreased intestinal alkaline phosphatase activity, a luminal enzyme that detoxifies the bacterial component (LPS), known to cause inflammation, and increased innate immune system activation in the luminal wall when compared to the rats resistant to obesity. In another study, Cani et al[46] demonstrated knockout rats of an immunoprotein (CD14), which was necessary to cause an inflammatory reaction to LPS, were resistant to weight gain. Vijay-Kumar et al[70] showed that TLR5 knockout mice exhibited hyperphagia and developed hallmark features of metabolic syndrome, including hyperlipidemia, hypertension, insulin resistance, and increased visceral fat deposition. These metabolic changes correlated with changes in the composition of the gut microbiota, and transfer of the gut microbiota from TLR5-deficient mice to wild-type germ-free mice conferred many features of metabolic syndrome to the recipients[70]. Lam et al[71] showed that mice fed with high-fat diet had reduced zona occludens-1 mRNA expression (40%) and increased permeability in proximal colon, and increased levels of TNF-α and IL-1 in mesenteric fat compared to mice fed with the control diet. In human studies, low grade endotoxemia (metabolic endotoxemia) was found associated with high-fat meal, obesity, NAFLD and diabetes[72-75]. All these data shows that the inflammatory mileu is the key component of the development of obesity and its complications.

Intestinal microbial system regulates entero-endocrine cells and promotes the release of several gut hormones. Peptide YY (PYY) is an enteroendocrine cell-derived hormone normally inhibits gut motility. Samuel et al[76] showed that PYY expression was lower in both Gpr41 -/- germ-free and conventionalized mice compared to Gpr41 +/+ mice. Reduced PYY expression resulted in increased intestinal transit time and reduced harvest of dietary energy[72-76]. In other studies, dietary fructo-oligo-saccharides increased the abundance of Bifidobacterium in the distal intestine, which led to increased colonic fermentation and glucagon-like peptide 1 (GLP-1) levels, decreased serum orexigenic peptide ghrelin and decreased food intake, fat mass, and hepatic steatosis in rats[77,78]. Based on these results, it can be concluded that dietary inulin-type fructans could play a role in the management of obesity and diabetes through their capacity to promote secretion of endogenous gastrointestinal peptides involved in appetite regulation[77,78].

Intestinal epithelial cells separate the intestinal microbial environment from the host immune system. Epithelial cells are linked to each other with tight junctions, which play a central role in maintaining intestinal barrier integrity[87]. In human studies NAFLD has been associated with increased LPS plasma levels, through mechanisms involving increased intestinal permeability, small intestinal bacterial overgrowth (SIBO), tight junction alteration and bacterial translocation[83,84,88-90].

SIBO is defined as an increase in the number and/or alteration in the composition of bacteria in the proximal gastrointestinal tract. Association of increased intestinal permeability and fatty liver was first demonstrated by Miele et al[84]. They reported that patients with fatty liver disease had increased gut permeability related to SIBO and disrupted tight junctions compared to healthy adults[84]. They also showed that gut permeability and SIBO were correlated with the severity of the liver steatosis. In the study of Gäbele et al[91], mice were fed with high-fat diet to stimulate NASH and then exposed to dextran sulphate sodium, an agent that causes intestinal epithelial injury. They showed that combined administration of high-fat diet and dextran sulphate sodium induced fibrosis in the liver[91]. All these data support the hypothesis that altered homeostasis between host and intestinal microbial system at the intestinal epithelial barrier level promotes bacterial translocation from the gut into the portal circulation and induces the liver damage[86,92,93].

Several studies demonstrated that the altered microbiota caused low-grade inflammation, which had a pivotal role for the development of obesity and its complications including NAFLD. Association of inflammation and fatty liver was firstly demonstrated by Cani et al[46]. They showed that continuous LPS infusion in mice increased the insulin resistance, liver triglyceride content and adipose tissue inflammation. After this study, other animal studies conducted in TLR4 knockout mice confirmed that TLR4 was essential for hepatic steatosis and NASH development[94-96]. Kupffer cells, which express the highest levels of TLR4 in the liver, are the primary cells in liver inflammation that respond to LPSs in order to produce inflammatory cytokines, chemokines and reactive oxygen species (ROS)[97,98]. In genetically obese mice, the administration of LPS induces changes in Kupffer cells function and increases liver parenchymal sensitivity to TNF-α[99]. TNF-α, which is the most important of LPS-TLR4 induced cytokines in these cells, is recognized as a mediator of hepatotoxicity, inflammation an NASH development in mice[100]. Besides, hepatic stellate cells might also have a substantial role in constituting the inflammatory cascade of the liver consequently associated with metabolic endotoxemia. Indeed, these cells are the major fibrogenic cell type in injured liver were shown to be the target through which TLR4 promoted fibrogenesis via enhancement of transforming growth factor-β (TGF-β) signalling[101-103].

The “second hit” mechanism of the NAFLD/NASH pathogenesis include enhanced lipid peroxidation and increased generation of ROS[104]. Inflammasomes, major contributors of inflammation, are cytoplasmic multiprotein complexes, which include nucleotide-binding domain (NLRPSs). NLRPSs are sensors of the bacterial PAMPs and DAMPs[105,106]. They manipulate the cleavage of pro-inflammatory cytokines such as pro-IL-1β and pro-IL-18. Most DAMPs induce the production of ROS, which is known to activate NLRP3 inflammasome[107,108]. Henao-Mejia et al[108] reported that inflammasome alterations or IL-18 deficiency cause intestinal microbial changes by enhancing portal influx of TLR4 and TLR9 ligands, which in turn increase hepatic TNF-α production in mice. Human studies also demonstrated that endotoxin levels were increased in both adult and pediatric obese patients with fatty liver disease[74,109,110]. Moreover, endotoxin levels were found to be correlated with the severity of the disease[75,111]. All these data support that chronic low-grade inflammation caused by obesogenic microbial ecosystem is a real “hepatotoxin” and has a key role in the pathogenesis of obesity related fatty liver disease.

Choline is a water-soluble essential nutrient. It is an important phospholipid component of the cell membrane and is the precursor molecule for the neurotransmitter acetylcholine. Choline has important roles in fat metabolism in the liver and a very-low-lipoprotein assembly, and also it promotes lipid transport from the liver[86,112]. Exogenous sources of choline are meat, dairy products, fish, soybeans, nuts and whole grains. Endogenous sources of choline, in the form of phosphatidylcholine, are biliary lipids, exfoliated epithelial cells and intestinal bacteria[113]. Buchman et al[114] showed that choline-deficient nutrition stimulated the liver steatosis. Furthermore, they demonstrated that a 6-wk choline supplementation reversed this pathology in patients. Gut microbiota secrete enzymes that cleave the dietary choline to its toxic metabolites (dimethylamine and trimethylamine). Liver uptakes these toxic methylamines and converts them to trimethylamine-N-oxide which induce inflammation in the liver[115,116]. Spencer et al[117] showed that the compositions of the gastrointestinal microbial communities changed with dietary choline content and especially Gammaproteobacteria and Erysipelotrichi levels were directly associated with changes liver steatosis in each subject during choline depletion. The role of dietary choline in NAFLD can be explained by the bioavailibility of free choline to for lipoproteins in the liver (especially very- low-density-lipoprotein-VLDL), which allows the export of free fatty acids from this organ[113]. If the gut microbiota converts excessive amounts of dietary choline into trimethylamine, this leads to reduced choline bioavailibility and consequent fatty liver disease[118].

Bile acids modulate lipid absorption and cholesterol homeostasis. The nuclear bile acid receptor, called farnesoid X receptor (FXR), is strongly expressed at bile acid excretion (liver) and absorption (intestine) regions. Bile acids also act as signaling molecules and activate FXR and the G-protein coupled receptor TGR5. Through activation of downstream signaling pathways of these key receptors, bile acids regulate not only their own synthesis and enterohepatic circulation, but also impact on hepatic lipid, glucose, and energy homeostasis[119]. FXR plays a key role in the control of hepatic de novo lipogenesis, VLDL triglyceride export and plasma triglyceride turnover[120]. TGR5 binds secondary bile acids and promotes glucose homeostasis, by stimulating secretion of GLP-1[121]. Besides, bile acids have a bacteriostatic activity. Gut microbiota can modulate bile acid metabolism. Swann et al[82] showed that gut microbiota can indirectly promote hepatic steatosis and lipid peroxidation through FXR stimulation changes in bile acid secretion. On the other hand, high-fat diet changes the bile acid composition, which influences the conditions for gut microbial environment and causes dysbiosis[122,123].

Intestinal microbiota produces a number of potentially hepatotoxic compounds such as ethanol, phenols, ammonia and they are transported to liver by portal system. These toxins stimulate hepatic Kupffer cells for production of nitric oxide and cytokines such as TNF-α[9,124]. Acetaldehyde and acetate are two major metabolites of ethanol. Ethanol can increase acetate production via inhibition of the Krebs cycle. Acetate is a substrate for fatty acid synthesis. On the other hand, actaldehyde and its metabolites may lead to the formation of reactive oxygen species. ROS production could be involved in liver injury by contributing to the disruption of intestinal barrier function and to the two hit mechanisms of steatohepatitis[80,86,125]. Ethanol and LPS also stimulate the production of ROS by parenchymal and nonparenchymal liver cells. Gustot et al[124] showed that enteral ethanol exposure induced steatosis and increased liver weight, aminotransferase levels, and TLR1, 2, 4, 6, 7, 8, and 9 liver mRNA expressions in mice. They concluded that ethanol-fed mice exhibited an oxidative stress dependent on upregulation of multiple TLRs in the liver and were sensitive to liver inflammation induced by multiple bacterial products recognized by TLRs[124].

In a human study, Nair et al[126] demonstrated that obese women with NASH had higher breath ethanol concentrations than healthy controls detected by gas chromatography. Similarly, Zhu et al[127] showed in their pediatric age group study that Proteobacteria, Enterobacteriaceae, and Escherichia (is a well-known ethanol producer bacteria) were the only phylum, family and genus types exhibiting significant difference between the patients with and without NASH microbiomes. Similar blood-ethanol concentrations were observed between healthy subjects and obese non-NASH patients; however, NASH patients exhibited significantly elevated blood ethanol levels[127]. Ethanol contributes to iNOS-mediated intestinal hyperpermeability, and therefore enhances the passage of endotoxins from the intestinal lumen into the portal system[128].