Chronic liver disease, cirrhosis, and its complications are epidemic worldwide. Most complications are mediated through a dysfunctional gut-liver axis. New techniques have made culture-independent analysis of the gut microbiome widespread. With insight into an unfavorable microbiome (dysbiosis) and how it affects liver disease, investigators have discovered new targets to potentially improve outcomes. Dysbiosis is associated with endotoxemia and propagates liver injury due to nonalcoholic steatohepatitis and alcohol. The composition and functionality of the microbiome changes with the development of cirrhosis, decompensation, and with treatments for these conditions. Gut microbiota can be used to predict clinically relevant outcomes in cirrhosis.
Key points
- •
Intestinal microbial dysbiosis has a large role in the progression of liver disease toward cirrhosis via endotoxemia, intestinal barrier dysfunction, and bile acid changes.
- •
Decompensation of cirrhosis results in significant changes to the gut microbiome that correlates with complications.
- •
Dysbiosis predicts decompensation and acute chronic liver failure; hence, every attempt has to be made to modulate this dysbiosis to prevent these outcomes.
- •
Bile acid changes are an important tool to study microbial function, and the microbiota-modulated bile acid profile plays an important role in progression of human liver disease.
Introduction
The fundamental understanding of liver disease, especially cirrhosis and its complications, has changed dramatically over the last decade with the introduction of the culture-independent microbiome analysis. Cirrhosis is estimated to affect 0.27% of the general population. Hepatitis C cirrhosis, alcoholic cirrhosis, and nonalcoholic steatohepatitis (NASH)-related cirrhosis are the most common causes, accounting for 53.5%. With the changing demography and increasing obesity, NASH-related cirrhosis is projected to be the most prevalent cause in the future.
There are multiple initial insults spanning from viral hepatitides, fatty liver (alcoholic and nonalcoholic), and biliary stasis to name a few. These initial insults result in inflammation, which is clinically detected by fatigue, malaise, and elevated liver functions. With repeated insult, the inflammation translates to fibrosis, and with continued insults, eventually cirrhosis. Clinically, the precirrhotic state is phenotypically different from the postcirrhotic state with portal hypertension being the major driver of clinical manifestations later. Cirrhosis is associated with complications of hepatic encephalopathy (HE), spontaneous bacterial peritonitis (SBP), variceal bleeding, ascites, and other manifestations of volume overload and also renal complications. However, all cirrhotics do not progress at the same pace, and decompensation is unpredictable. The rate of decompensation for alcoholic cirrhosis is estimated to vary between 4% and 25% for NASH cirrhosis, ∼2% over 5 years despite minimal histologic progression, and with hepatitis C cirrhosis, the cumulative probability for decompensation at 1 year is ∼5%. Continuing with the initial insult definitely leads to decompensation, but studies have noted progression of disease from fibrosis to cirrhosis despite cessation of the insulting factor. To understand why certain patients remain stable while others decompensate despite control of inciting insults, investigators have evaluated the gut microbiome and its associated changes in various stages and causes of liver disease to further explain this phenomenon.
The intestinal microbiome itself is a complex composition of microorganisms that is well known to be implicated in cirrhosis and its complications. The metabolic neural pathway also known as the gut-liver–brain axis is a key player in cirrhosis and particularly in HE, and this pathway is strongly regulated by the gut microbiome. Dysbiosis, or an unfavorable change in the composition of the microbiome with a reduction in autochthonous (Firmicutes) bacteria and growth of other taxa (Bacteroidetes, Actinobacteria), is well known to occur in advanced liver disease and other intestinal abnormality. Dysbiosis is thought to be central to the proposed pathophysiology of the microbiota and gastrointestinal abnormality for liver disease onset, progression, and development of complication. Typically in dysbiosis there occurs a change in the balance of native Firmicutes to Bacteroidetes species with the former decreasing and the latter increasing. These native bacteria are important for the harmony of the gastrointestinal flora, and as such, for the well-being of the entire body, which is why science now considers the human microbiome as an organ in itself. The autochthonous bacteria produce short-chain fatty acids (SCFAs) that nourish the colonic mucosal cells and reduce local colonic inflammation and also antibacterial peptides and hence help maintain the intestinal barrier. Hence, dysbiosis is associated with increased inflammation and endotoxemia in multiple gastrointestinal abnormality, and in particular, liver disease. The Cirrhosis Dysbiosis Ratio (CDR) is the ratio of autochthonous to nonautochthonous taxa in cirrhosis. The lower the CDR, the more the endotoxemia and more decompensated the cirrhosis.
To give a brief overview of the pathophysiology, the intestine and its barrier, that is, epithelium, Peyers patches, and its lymphoid tissue, act as the first immune system to come into contact with bacteria endotoxins or lipopolysaccharides (LPS), also known as pathogen-associated molecular patterns (PAMPs), that are produced by human microbiota. Because of changes in the intestinal barrier, there is bacterial translocation (BT), which exposes the intestinal immune system to antigens. The intestinal cells have a system of receptors, namely the membranous Toll-like receptors (TLR) and intracellular nucleotide oligomerization domain– like receptors (NLR), that recognize bacterial LPS, bacterial DNA, and peptidoglycans. Recognition of the bacterial product by its receptors leads to upregulation of inflammatory mediators like tumor necrosis factor- (TNF-α). Another integral factor in this process is the portal vein that acts as the main conduit for transfer of LPS and other bacterial products from the intestines to the liver. The final step in this chain is that the metabolites interact with hepatocytes and Kupffer cells via the hepatic TLR and NLR, resulting in changes that promote a cirrhotic morphology.
In this article, the authors touch upon the proposed pathophysiology of how the microbiome is associated with different liver disease stages and microbiome, focusing mainly on human studies. In Tables 1 and 2 , the authors provide details about the main recent clinical studies that show dysbiosis in chronic liver disease (CLD). This review focuses primarily on human nonalcoholic fatty liver disease (NAFLD)/NASH-related liver disease, and alcohol-related liver disease (ALD), because these causes are the paradigms for microbiome-related endotoxemia, dysbiosis, and related changes in the liver. The onset of cirrhosis, regardless of cause, results in changes to the microbiome, which is furthered by decompensation.
| Authors and Year | Groups Compared | Sample Used for Analysis | Results |
|---|---|---|---|
| Human studies in NAFLD/NASH | |||
| Raman et al, 2013 | Healthy controls vs NASH | Stool | Increased Ruminococcaceae with reduced Lactobacillae and Lachnospiraceae |
| Zhu et al, 2013 | Adolescent health control vs obese vs NASH | Stool | Increased Bacteroidaceae and Bifidobacterium with reduced Ruminococcaceae and Lachnospiraceae |
| Mousaki et al, 2013 | Controls vs simple steatosis vs NASH | Stool | NASH had a lower (Bacteroidetes to total bacteria counts) compared the rest and higher fecal Clostridium coccoides |
| Bajaj et al, 2014 | Controls vs NASH cirrhosis | Stool | Increased Porphyromonadaceae, Bacterioidaceae, and reduced Veillonellaceae |
| Boursier et al, 2016 | Non cirrhotics—NASH vs no NASH | Stool | Increased Porphyromonadaceae, Ruminococcaceae, and Enterobacteriaceae with reduced Veillonellaceae and Ruminococcaceae |
| Human studies in ALD | |||
| Mutlu et al, 2012 | Healthy controls vs alcohol cirrhotics | Sigmoid mucosal biopsies | Increased Bacterioidaceae |
| Bajaj et al, 2014 | Controls vs alcohol cirrhotics | Stool | Increased Enterobacteriaceae and Halomonadaceae, reduced Lachnospiraceae, Ruminococcaceae, and Clostridiales XIV |
| Kakiyama et al, 2014 | Nonalcoholic cirrhotics vs alcoholic cirrhotics | Stool | Increased Bacterioidaceae and reduced Veillonellaceae |
| Authors and Year | Groups Compared | Sample Used for Analysis | Results |
|---|---|---|---|
| Bajaj et al, 2012 | Controls vs HE and no HE | Stool | Increased Veillonellaceae in OHE compared with no OHE cirrhotics, Porphyromonadaceae correlated with cognition |
| Bajaj et al, 2012 | Controls vs HE and no HE | Sigmoid mucosal biopsy | Increased Enterococcus , Megasphaera , and Burkholderia linked to poor cognition and higher inflammation; reduced Enterococcus , Megasphaera , and Burkholderia. Alcaligeneceae and Porphyromonadaceae were associated with poor cognition |
| Zhang et al, 2013 | Controls vs MHE vs without MHE | Stool | Veillonella parvula and S salivarius correlated negatively with cognitive function |
| Bajaj et al, 2014 | Cirrhotics of all causes studied prospectively | Stool | Increased pathogenic bacteria (Porphyromonadaceae, Bacterioidaceae) with no real change in autochthonous bacteria in OHE. ACLF was associated with increased Propionibacteriaceae and Halomonadaceae but reduced Lachnospiraceae and Veillonellaceae |
| Chen et al, 2015 | Controls vs ACLF cirrhotics all causes | Stool | ACLF was associated with increased Pasteurellaceae, Streptococcaceae, and Enterococcaceae with decreased Bacterioidaceae, Lachnospiraceae, and Ruminococcaceae |
| Ahluwalia et al, 2016 | Controls vs cirrhotics all causes | Stool | HE patients had higher relative abundance of autochthonous bacteria and higher abundance of Enterococcaaceae, Staphlyococcaceae, Porphyromonadaceae, and Lactobacillaceae. The autochthonous families correlated negatively with MRI brain and positively with pathogenic bacteria |
Introduction
The fundamental understanding of liver disease, especially cirrhosis and its complications, has changed dramatically over the last decade with the introduction of the culture-independent microbiome analysis. Cirrhosis is estimated to affect 0.27% of the general population. Hepatitis C cirrhosis, alcoholic cirrhosis, and nonalcoholic steatohepatitis (NASH)-related cirrhosis are the most common causes, accounting for 53.5%. With the changing demography and increasing obesity, NASH-related cirrhosis is projected to be the most prevalent cause in the future.
There are multiple initial insults spanning from viral hepatitides, fatty liver (alcoholic and nonalcoholic), and biliary stasis to name a few. These initial insults result in inflammation, which is clinically detected by fatigue, malaise, and elevated liver functions. With repeated insult, the inflammation translates to fibrosis, and with continued insults, eventually cirrhosis. Clinically, the precirrhotic state is phenotypically different from the postcirrhotic state with portal hypertension being the major driver of clinical manifestations later. Cirrhosis is associated with complications of hepatic encephalopathy (HE), spontaneous bacterial peritonitis (SBP), variceal bleeding, ascites, and other manifestations of volume overload and also renal complications. However, all cirrhotics do not progress at the same pace, and decompensation is unpredictable. The rate of decompensation for alcoholic cirrhosis is estimated to vary between 4% and 25% for NASH cirrhosis, ∼2% over 5 years despite minimal histologic progression, and with hepatitis C cirrhosis, the cumulative probability for decompensation at 1 year is ∼5%. Continuing with the initial insult definitely leads to decompensation, but studies have noted progression of disease from fibrosis to cirrhosis despite cessation of the insulting factor. To understand why certain patients remain stable while others decompensate despite control of inciting insults, investigators have evaluated the gut microbiome and its associated changes in various stages and causes of liver disease to further explain this phenomenon.
The intestinal microbiome itself is a complex composition of microorganisms that is well known to be implicated in cirrhosis and its complications. The metabolic neural pathway also known as the gut-liver–brain axis is a key player in cirrhosis and particularly in HE, and this pathway is strongly regulated by the gut microbiome. Dysbiosis, or an unfavorable change in the composition of the microbiome with a reduction in autochthonous (Firmicutes) bacteria and growth of other taxa (Bacteroidetes, Actinobacteria), is well known to occur in advanced liver disease and other intestinal abnormality. Dysbiosis is thought to be central to the proposed pathophysiology of the microbiota and gastrointestinal abnormality for liver disease onset, progression, and development of complication. Typically in dysbiosis there occurs a change in the balance of native Firmicutes to Bacteroidetes species with the former decreasing and the latter increasing. These native bacteria are important for the harmony of the gastrointestinal flora, and as such, for the well-being of the entire body, which is why science now considers the human microbiome as an organ in itself. The autochthonous bacteria produce short-chain fatty acids (SCFAs) that nourish the colonic mucosal cells and reduce local colonic inflammation and also antibacterial peptides and hence help maintain the intestinal barrier. Hence, dysbiosis is associated with increased inflammation and endotoxemia in multiple gastrointestinal abnormality, and in particular, liver disease. The Cirrhosis Dysbiosis Ratio (CDR) is the ratio of autochthonous to nonautochthonous taxa in cirrhosis. The lower the CDR, the more the endotoxemia and more decompensated the cirrhosis.
To give a brief overview of the pathophysiology, the intestine and its barrier, that is, epithelium, Peyers patches, and its lymphoid tissue, act as the first immune system to come into contact with bacteria endotoxins or lipopolysaccharides (LPS), also known as pathogen-associated molecular patterns (PAMPs), that are produced by human microbiota. Because of changes in the intestinal barrier, there is bacterial translocation (BT), which exposes the intestinal immune system to antigens. The intestinal cells have a system of receptors, namely the membranous Toll-like receptors (TLR) and intracellular nucleotide oligomerization domain– like receptors (NLR), that recognize bacterial LPS, bacterial DNA, and peptidoglycans. Recognition of the bacterial product by its receptors leads to upregulation of inflammatory mediators like tumor necrosis factor- (TNF-α). Another integral factor in this process is the portal vein that acts as the main conduit for transfer of LPS and other bacterial products from the intestines to the liver. The final step in this chain is that the metabolites interact with hepatocytes and Kupffer cells via the hepatic TLR and NLR, resulting in changes that promote a cirrhotic morphology.
In this article, the authors touch upon the proposed pathophysiology of how the microbiome is associated with different liver disease stages and microbiome, focusing mainly on human studies. In Tables 1 and 2 , the authors provide details about the main recent clinical studies that show dysbiosis in chronic liver disease (CLD). This review focuses primarily on human nonalcoholic fatty liver disease (NAFLD)/NASH-related liver disease, and alcohol-related liver disease (ALD), because these causes are the paradigms for microbiome-related endotoxemia, dysbiosis, and related changes in the liver. The onset of cirrhosis, regardless of cause, results in changes to the microbiome, which is furthered by decompensation.
| Authors and Year | Groups Compared | Sample Used for Analysis | Results |
|---|---|---|---|
| Human studies in NAFLD/NASH | |||
| Raman et al, 2013 | Healthy controls vs NASH | Stool | Increased Ruminococcaceae with reduced Lactobacillae and Lachnospiraceae |
| Zhu et al, 2013 | Adolescent health control vs obese vs NASH | Stool | Increased Bacteroidaceae and Bifidobacterium with reduced Ruminococcaceae and Lachnospiraceae |
| Mousaki et al, 2013 | Controls vs simple steatosis vs NASH | Stool | NASH had a lower (Bacteroidetes to total bacteria counts) compared the rest and higher fecal Clostridium coccoides |
| Bajaj et al, 2014 | Controls vs NASH cirrhosis | Stool | Increased Porphyromonadaceae, Bacterioidaceae, and reduced Veillonellaceae |
| Boursier et al, 2016 | Non cirrhotics—NASH vs no NASH | Stool | Increased Porphyromonadaceae, Ruminococcaceae, and Enterobacteriaceae with reduced Veillonellaceae and Ruminococcaceae |
| Human studies in ALD | |||
| Mutlu et al, 2012 | Healthy controls vs alcohol cirrhotics | Sigmoid mucosal biopsies | Increased Bacterioidaceae |
| Bajaj et al, 2014 | Controls vs alcohol cirrhotics | Stool | Increased Enterobacteriaceae and Halomonadaceae, reduced Lachnospiraceae, Ruminococcaceae, and Clostridiales XIV |
| Kakiyama et al, 2014 | Nonalcoholic cirrhotics vs alcoholic cirrhotics | Stool | Increased Bacterioidaceae and reduced Veillonellaceae |
| Authors and Year | Groups Compared | Sample Used for Analysis | Results |
|---|---|---|---|
| Bajaj et al, 2012 | Controls vs HE and no HE | Stool | Increased Veillonellaceae in OHE compared with no OHE cirrhotics, Porphyromonadaceae correlated with cognition |
| Bajaj et al, 2012 | Controls vs HE and no HE | Sigmoid mucosal biopsy | Increased Enterococcus , Megasphaera , and Burkholderia linked to poor cognition and higher inflammation; reduced Enterococcus , Megasphaera , and Burkholderia. Alcaligeneceae and Porphyromonadaceae were associated with poor cognition |
| Zhang et al, 2013 | Controls vs MHE vs without MHE | Stool | Veillonella parvula and S salivarius correlated negatively with cognitive function |
| Bajaj et al, 2014 | Cirrhotics of all causes studied prospectively | Stool | Increased pathogenic bacteria (Porphyromonadaceae, Bacterioidaceae) with no real change in autochthonous bacteria in OHE. ACLF was associated with increased Propionibacteriaceae and Halomonadaceae but reduced Lachnospiraceae and Veillonellaceae |
| Chen et al, 2015 | Controls vs ACLF cirrhotics all causes | Stool | ACLF was associated with increased Pasteurellaceae, Streptococcaceae, and Enterococcaceae with decreased Bacterioidaceae, Lachnospiraceae, and Ruminococcaceae |
| Ahluwalia et al, 2016 | Controls vs cirrhotics all causes | Stool | HE patients had higher relative abundance of autochthonous bacteria and higher abundance of Enterococcaaceae, Staphlyococcaceae, Porphyromonadaceae, and Lactobacillaceae. The autochthonous families correlated negatively with MRI brain and positively with pathogenic bacteria |
Nonalcoholic fatty liver disease
NAFLD and its more sinister evolution, NASH, is slated to be the most common cause for CLD and liver transplantation in the near future. It is estimated that one-third of the general US population has a fatty liver. Up to 20% to 30% of NAFLD patients progress to NASH, and 30% of NASH cases progress to cirrhosis. NAFLD and NASH are commonly associated with the metabolic syndrome, and as noted in previous studies, is a proinflammatory state that is associated with higher levels of serum TNF-α, interleukin-6 (IL-6), and adipokines. The linkages of the gut microbiome and changes with the microbiome with regards to the causation in NAFLD/NASH have been studied for a while now, and the microbiome probably has a larger role in the causation than is known.
Precirrhosis Nonalcoholic Fatty Liver Disease Microbiota Changes
In order to understand why the microbiota is tied into NAFLD and NASH, one needs to first understand how obesity and the metabolic syndrome affect the human microbiome. The human microbiome has been noted to be altered in obesity, and dysbiosis has been well documented. A study by Mouzaki and colleagues looked at 50 patients (of which 22 had NASH) to understand the obesity and dysbiosis paradigm and showed an inverse relationship existed between percentage Bacteriodetes species and the presence of NASH; this confirmed a possible role of dybiosis in NAFLD. An aspect of dysbiosis in NAFLD that is often overlooked is the effect of the microbiome in extraction of energy from the gut where dysbiosis results in increased production of SCFAs in the intestine with increased monosaccharides, that on transport to the liver activates the proteins that promote hepatic lipogenesis and steatosis. Type 2 diabetes mellitus (DM), an integral a component of the metabolic syndrome, has no immediate effect on the microbiome per se, but dysbiosis seems to affect the development of DM. However, once DM sets in, it has a role in dysbiosis and pathogenesis of cirrhosis, as mentioned in a later section.
In humans with NAFLD/NASH, higher levels of LPS in the serum and increased expression of the TNF-α gene in hepatic tissues have been noted, confirming the proinflammatory state associated with these abnormalities. Disruption of the intestinal barrier, that is, disruption of intercellular tight junctions, has been seen in NAFLD and could directly contribute to LPS and other bacterial products reaching the liver. The mechanism underlying disruption of the barrier is thought to be from local colonic inflammation; however, the evidence for this is conflicting and indirect in adult humans, as seen in a study by Pendyala and colleagues, where weight loss in obese individuals resulted in reduced colonic inflammation. Other studies to study colonic inflammation via fecal calprotectin and leptin showed no difference between obese and lean adults. Regardless of the functionality of the intestinal barriers, it has been shown that bacterial LPS are transported out of the intestine to the liver via the portal vein along with chylomicrons that are formed during consumption of high-fat diets. This step is key to aid in the passage of proinflammatory bacterial products past the intestinal barrier, to the liver, where these products interact with their corresponding TLR and induce production of proinflammatory mediators like TNF-α and then subsequent procirrhotic changes. Hence, multiple mechanisms that underlie the development of microbial dysbiosis in NAFLD are seen.
Intestinal Barrier Dysfunction and the Microbiome in Nonalcoholic Fatty Liver Disease
Another prominent factor that could promote endotoxemia in NAFLD and NASH is small intestinal bacterial overgrowth (SIBO). For unclear reasons, the prevalence of SIBO has been found higher in obesity and NAFLD. The estimated prevalence of SIBO in NALFD ranges from 50% to 70% from various studies. SIBO has been recognized as an independent risk factor for the severity of hepatic steatosis because of its role in the dysfunction of the intestinal barrier. SIBO in general results in a quantitative change in the microbiome that is associated with dysbiosis, which leads to increased intestinal permeability and increased gut bacterial product translocation, essential in the transition pathway from NAFLD to NASH to cirrhosis. Intestinal microbiota, apart from providing bacterial byproducts, increasing intestinal barrier permeability, also suppress small intestinal secretion of fasting-induced adipocyte factor, which results in increased triglyceride deposition in the hepatocytes. Studies that have looked at the microbiome change in NAFLD have not exactly been able to classify if the changes are SIBO related, and as such, there are no microbiome studies that document the exact changes in NAFLD/NASH from SIBO.
Role of Microbiota and Their Products as a “Second Hit” for Nonalcoholic Steatohepatitis Development
The human microbiome has a strong role in progression of NAFLD to NASH. The 2-hit hypothesis proposed by Day and James puts hepatic steatosis as the first hit. Multiple second hits are possible with this model. Prior papers have studied blood-ethanol concentrations in pediatric NASH subjects and obese non-NASH subjects and noted elevated levels in the former. They deduced that the production of endogenous alcohol by Escherichia coli in the intestinal microbiome played a role in NASH development. Another study of 15 adult female patients placed on regulated choline content diets noted that liver fat was inversely proportional to choline deficiency in the diets. Crespo and colleagues studied 52 adult obese patients and looked at the relationship between TNF-α p55 and p75 (TNF-α receptors) and noted overexpression of TNF-α mRNA in NASH patients with increased overexpression with increased severity of NASH. Last, TLR-mediated signaling in Kupffer cells is another proposed second hit. Whatever the second hit, given all the evidence, microbiome has an important role in NAFLD from the precirrhotic to the cirrhotic stages.
Another important mode of evaluating bacterial products is the bile acid (BA) profile. BAs are thought to regulate the microbiome by a potential detergent effect on the cell walls of the intestinal microbacteria and also by interacting with the Farnesoid X receptor (FXR) in the liver, which not only induces excretion of BAs from the liver but also induces antimicrobial peptides production.
The conjugated forms of primary acids BAs (cholic acid [CA] and chenodeoxycholic acid) have been found to be higher in the serum of NAFLD patients, and similarly, the conjugated form of primary and secondary BAs (lithocolic and deoxycholic acid) were noted to be 2.4 times higher in patients with NASH compared with controls. Lake and colleagues studied the composition of BAs in the human livers of NAFLD patients and interestingly found a reduced level of CA and glycodeoxycholic acid and an increase in taurocholic acid and taurodeoxycholic acid. A recent study looking at the fecal composition of NAFLD and NASH adults confirmed a higher fecal BA level with a predominance of primary BAs in the stool of NASH patients (NASH > NAFLD), further clarifying the mechanism by which BAs regulate dysbiosis. To further assess the interplay, Neuschwander-Tetri and colleagues in a multicenter randomized trial studied the effect of obeticholic acid (BA derivative that activates FXR) in NASH patients and noted improvement in the histologic features of NASH based on the NAFLD activity score on liver biopsies.
Related posts:
From Culture to High-Throughput Sequencing and Beyond
Impact of Antibiotics on Necrotizing Enterocolitis and Antibiotic-Associated Diarrhea
Fecal Microbiota Transplantation
Diet and the Microbiome
The Esophageal and Gastric Microbiome in Health and Disease
The Gut Microbiota in Inflammatory Bowel Disease
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
