Authors: Dinakaran Vasudevan, Arulraj Ramakrishnan, Ganesan Velmurugan
Categories: Review Article, Blood microbiome, Fecal microbiota transplantation (FMT), Liver disease, Microbial components, Microbial translocation
Source: Heliyon
Liver diseases are a group of major metabolic and immune or inflammation related diseases caused due to various reasons including infection, abnormalities in immune system, genetic defects, and lifestyle habits. However, the cause-effect relationship is not completely understood in liver disease. The role of microbiome, particularly, the role of gut and oral microbiome in liver diseases has been extensively studied in recent years. More interestingly, the presence of blood microbiome and tissue microbiome has been identified in many liver diseases. The translocation of microbes from the gut into the portal circulation has been attributed to be the major reason for the presence of blood microbial components and its clinical implications in liver disorders. Besides microbial translocation, Pathogen associated Molecular Patterns (PAMPs) derived from gut microbiota might also translocate. The presence of blood microbiome in liver disease has been reviewed earlier. However, the role of blood microbiome as a biomarker and therapeutic target in liver diseases has not been analysed earlier. In this review, we confabulate the origin and physiology of blood microbiome and blood microbial components in relation to the progression and pathogenesis of liver disease. In conclusion, we discuss the translational perspectives targeting the blood microbial components in the diagnosis and therapy of liver disease.
Keywords: Blood microbiome, Liver disease, Microbial components, Microbial translocation, Fecal microbiota transplantation (FMT)
Liver disease is the fifth leading cause of mortality across the world and accounts for approximately two million deaths per year worldwide. Chronic Liver disease accounted for 2.2 % of deaths and 1.5 % of disability in 2016 [1]. In the United States, Liver diseases were the second leading cause of death amongst all digestive diseases [2].
Over the past years, viral hepatitis has been considered as the leading cause of mortality due to liver disease. However, this rate has decreased over the years due to increased prevention strategies [3]. On the other hand, obesity and alcohol consumption have become key risk factors of liver diseases. They are estimated to drive liver disease epidemiology and account for increased proportions of mortality due to liver disease in the future. The number of liver disease cases is estimated to be 1.5 billion worldwide. The most common causes of liver disease are non-alcoholic fatty liver disease (NAFLD), Hepatitis B virus (HBV), Hepatitis C virus (HCV), and alcoholic liver disease (ALD) [4]. Some of the common liver diseases are alcoholic liver disease (ALD), non-alcoholic fatty liver disease (NAFLD)/Non-alcoholic Steatohepatitis (NASH), Hepatitis B and C virus infections (HBV and HCV), Primary biliary cholangitis (PBC), Primary sclerosing cholangitis (PSC), liver fibrosis, liver cirrhosis and Hepatocellular carcinoma (HCC). The annual mortality rate due to liver cirrhosis has been represented in Fig. 1.
Fig. 1 Number of deaths due to cirrhosis every year globally, by etiology.Source: References [[2], [3], [4]]. ALD: Alcoholic liver disease, NAFLD: Non-alcoholic fatty liver disease
The blood microbiome has been recently identified in liver disease patients. The blood microbiome comprises of the microbial components including PAMPs and microbial metabolites released from the gut due to gut injury or gut leakage. The actual “meaning” of bacteria in blood is the presence of bacterial components and/or bacterial metabolites that are released into the circulation from internal tissues like gut. These bacterial components reach the liver through hepatic circulation.
Due to the extensive use of antibiotics, vaccines and improved sanitation, the occurrence of infectious diseases has decreased rapidly and this has resulted in the increase in the incidence of metabolic diseases like liver diseases. The blood microbiome identified in liver diseases over the past decade, could be a source of antigens responsible for the onset of liver diseases.
For centuries together, blood was strongly believed to be a germ-free environment [5]. Very rarely, it has been observed that bacteria could live inside red blood cells [6] and leukocytes [5]. Subsequently, L-forms and Corynebacteria like forms were visualized inside human red blood cells by electron microscope [7,8]. Until the mid- 20th century, microbes in the circulation were identified only in infectious disease cases and hence were considered pathogenic. It was in the beginning of 21st century, Nikkari et al. detected bacterial DNA for the first time in the blood of healthy individuals [9]. Later, bacteria were detected by amplification of the prokaryotic 16S rDNA gene in healthy human blood [10]. In the subsequent years, microbial components were identified in healthy circulation more too often by the availability of next generation sequencing techniques [11,12]. Metagenome shotgun sequencing of the blood has also shown the presence of viral and archaeal DNA in healthy blood [13]. In continuation with the earlier findings, intracellular L-forms or pleomorphic forms of bacteria were detected in healthy human blood [14]. More recently, there was a report on the detection of fungal microbiome in the blood of healthy individuals [15].
The healthy blood microbiota was found to vary from healthy gut microbiota. The healthy gut was found to be dominated by Firmicutes and Bacteroidetes. On the other hand, the healthy blood predominantly contains Proteobacteria [16,17]. The healthy blood microbiota is considered to be dormant because it does not induce clinical complications. However, the dormant state may enable microbes or their components to evade the host immune system. The microbial components remain dormant in blood for many years, either in the living or non-living state [18]. Earlier studies have reported that the dormant state of microbes plays a major role in their ability to sustain antibiotic treatment and causing disease [19]. Dormant bacteria in blood can shed cell membrane components like LPS and LBP during disease conditions to become pathogenic. The genetic material of non-viable bacterial cells may play a vital role in many disease conditions by acting on host cells.
The concept of microbial products including microbial metabolites from the gut entering the circulation has been known for many years. This was earlier denoted by many terms such as endotoxemia. Endotoxemia was identified in patients by the detection of lipopolysaccharides (LPS), LPS binding protein (LBP) and lipoteichoic acids (LTA) in the circulation. In recent years, this concept or phenomenon is known by the term called Microbial translocation. Microbial translocation from the gut or the oral tract to the circulation occurs by various mechanisms. Microbial translocation is usually assessed in blood samples by the quantification of PAMPs (LPS, LBP etc.) [20], bacterial DNA or RNA fragments [21], serum CD14 levels [22], endotoxin core antibodies and antiflagellin antibodies [23]. Over the past decades, 16S rDNA gene is detected and quantitated in blood to assess bacterial translocation [21].
The intestine functions as an effective immunological barrier against intraluminal bacteria from entering the circulation. The gut barrier is formed by a well functioning immune system, properly balanced microflora and an intact mucosa. The innate immune defense system of the intestine is formed by the Peyer's patches in the intestinal crypts, along with lymphocytes and macrophages [24]. A stable ecological balance is maintained in the intestine to prevent bacterial overgrowth and subsequent translocation. When one or more of these defensive mechanisms are impaired, viable microbes or their products such as PAMPs may pass through this barrier and reach the mesenteric lymph nodes or other organs via lymphatic drainage and circulation resulting in microbial translocation. These microbes or their components persist in circulation in diseased conditions. Also, practically, PAMPs such as LPS persist in the blood in healthy situation, albeit at a lower concentration.
Many immunological mechanisms are considered responsible for gut leakage. Tight junctions hold the enterocytes together. Bacteria initially adhere to the enterocytes by binding to the receptors on the cell membrane and slowly move to the basal membrane. The movement of bacterial products occurs either through the paracellular route through tight junctions or the transcellular route through enterocytes [25]. The movement of bacterial and viral components across the gut barrier involves an intricate immunological process. One proven process is by direct cellular uptake through the activation of NOD1 receptors in Microfold cells (M cells) of the Peyer's patches by damage to the gut epithelial cells [26]. Portal vein endotoxemia of gut origin occurs frequently in liver diseases leading to spontaneous bacterial peritonitis [27,21]. The intestinal bacteria are carried to mesenteric lymph nodes by lymphatic drainage from where they spread to other tissues via circulation [28]. Interestingly, the process of microbial translocation was also known to occur in the healthy human intestine [29] and appears to begin early in life [30] (Fig. 2).
Fig. 2 Schematic representation of an increase in blood microbiota concentration due to gut microbiota dysbiosis in liver disease patientsNAFL: Non-alcoholic Fatty Liver, NASH: Non-alcoholic Steatohepatitis, TLR: Toll-like receptor.
Transient bacteremia is known to occur in patients when exposed to chronic oral infections for a long period. Bacteria have been detected in blood after periodontal surgery, tooth extraction and even after normal tooth brushing [31]. Oral infections like dental caries led to destruction of periodontium, dentin, enamel and the root surfaces of teeth. Translocation of bacteria and their products into the circulation through the oral route has been reported in previous studies [32,33].
Microbial products, virulent components and metabolites are continuously shed into the circulation by disruption of oral biofilms on tooth surface. This disruption may occur because of oral dysbiosis during dental caries, periodontitis, endodontis etc., various dental procedures, daily oral hygiene efforts and even during gentle mastication. In particular, there are innumerable studies on P. gingivalis, which is well known to cause oral dysbiosis in many oral diseases [34]. The propinquity of bacteria in the root canal and periapical tissues to the bloodstream can cause bacteremia during clinical dental procedures like tooth extraction, periodontal and endodontic treatments.
The oral-gut-liver-immune system axis plays a significant role in the pathogenesis of liver disease. There are extensive studies on the dysbiosis of gut microbiota in many liver diseases. The healthy gut microbiota harbours a major proportion of Firmicutes and Bacteroidetes and a meagre proportion of Actinobacteria and Proteobacteria [35]. This healthy nature of the human gut is perturbed in liver diseases resulting in dysbiosis of the gut microbiota. Also, perturbed gut microbiota may lead to liver disease complications. It is like the “egg and the chick hypothesis”. The intestinal microbiota and the bacterial products may contribute to liver disease by mechanisms such as intestinal permeability, systemic inflammation, short chain fatty acid production and metabolic changes. Similarly, oral microbiota dysbiosis also plays a role in the pathogenesis of liver disease and augments the concentration of blood microbiome in liver disease patients. The blood microbiome induces the presence of liver tissue microbiome which in turn exacerbates the clinical complications of liver disease.
Blood microbiome in liver diseases originates predominantly due to microbial translocation through the gut-blood-liver axis. One of the clinical complications associated with liver disease and other metabolic diseases is called “the leaky gut phenomenon”, which is the chronic inflammation of intestinal mucosa leading to increased intestinal permeability. This might provoke bacterial translocation leading to a considerable amount and highly divergent bacterial traces in human blood samples. However, blood microbiome and its diagnostic potential has not been investigated in liver diseases until now. The approach employed to gather the comprehensive compilation of studies presented in Table 1 and this section involves utilizing Google Scholar and PubMed search methods. The search terms utilized include "Blood Microbiome" and "Liver," "Blood Microbiota" and "Liver," "Circulating Microbiome" and "Liver," "Circulating Microbiota" and "Liver," "Blood Virome" and "Liver," as well as "Circulating Virome" and "Liver."
Foot prints of bacteria or rather; bacterial DNA was detected for the first time in the circulation of cirrhosis patients in the year 2002 by PCR amplification and nucleotide sequencing of prokaryotic conserved 16S rRNA gene sequences in the serum of advanced liver cirrhosis patients. Presence of bacterial DNA was detected in the serum of 9 of 28 patients [36]. In the subsequent year, the same group reported the presence of 16S rRNA gene sequences in the serum of cirrhosis patients by quantitative PCR reaction. Seven out of 17 patients showed the presence of bacterial DNA in blood at the time of admission. Also, by 16S rRNA gene nucleotide sequencing, the authors found that bacteria identified during admission were identical to those found in subsequent detections after time [21]. In a similar study, the same group showed the presence of bacterial DNA fragments in the blood of liver cirrhosis patients to prove the concept of bacterial translocation in liver cirrhosis which they have mentioned in their earlier publications [37]. The same group later reported the association of bacterial DNA sequences with inflammatory markers of disease [38].
The concept of blood microbiota in liver disease patients originated for the first time when the group lead by Jacques Amar showed the association of blood microbiota with liver fibrosis in obese patients. The blood microbiota of obese patients was dominated by Proteobacteria with a meagre concentration of Actinobacteria, Firmicutes and Bacteroidetes [39]. Santiago et al. in the same year found difference in the serum microbiome of liver cirrhosis patients as compared to healthy individuals. A complex microbial community was detected in serum of cirrhosis patients as compared to healthy controls. Also, there was difference in the serum microbial diversity in cirrhosis patients with and without ascites. Cirrhotic patients with ascites had a higher concentration of Clostridiales as compared to patients without ascites [40].
In the subsequent year, an increased blood microbial diversity in conjunction with inflammatory response and systemic hemodynamic parameters was reported in decompensated cirrhotic patients as compared to healthy individuals. Intestinal Infections Microbial DNA qPCR Array was used to screen for 53 bacterial DNA from the gut in the blood of both the cirrhotic patients and healthy individuals [41]. Lebba et al. showed the presence of shared members of Proteobacteria phyla in peripheral and portal circulation of patients with liver cirrhosis [42]. They have identified impaired metabolism of short-chain fatty acids (SCFAs) and carbon/methane sources by faecal bacteria in Liver cirrhosis patients as compared to healthy individuals using feces and caecum samples. However, blood samples from healthy individuals were not analysed in this study. In another interesting study, the blood microbiome was compared in alcoholic hepatitis patients, heavy drinking controls and non-alcohol consuming controls. As expected, there was an increased bacterial concentration in alcoholic hepatitis patients as compared to controls. But, both the alcohol consuming groups (alcoholic hepatitis and heavy drinking controls) had an increased concentration of Fusobacteria and a decreased proportion of Bacteroidetes as compared to the non-alcohol consuming controls. This shows the influence of alcohol consumption in the gut and blood microbiome [43].
In a subsequent study, a group in Europe showed that bacteria in the different circulatory compartments were similar in Liver cirrhosis patients which is strong evidence for the concept of bacterial translocation. They characterised the blood microbiome in portal vein (first venous outflow in gut–liver axis), liver outflow, central venous blood and peripheral venous blood from seven patients with decompensated liver cirrhosis receiving transjugular intrahepatic portosystemic shunt (TIPS) for either variceal bleeding (n = 3) or refractory ascites (n = 4). The seven patients had a mean Model for End-stage Liver Disease (MELD) score of 8.4 (range 6–13), Child-Pugh-Score (CHILD A: n = 4, CHILD B: n = 3). They identified 65 genera belonging to four phyla, predominantly Proteobacteria, followed by Actinobacteria, Bactroidetes and Firmicutes [44]. In the same year, the fecal and blood microbiota was studied in obese and lean NAFLD patients. The fecal and blood microbiota profiles were similar among obese and lean NAFLD patients but differed between these two groups, which might serve as potential biomarkers to discriminate these two phenotypes of NAFLD. Obese NAFLD group showed a distinct bacterial community with a lower biodiversity and a far distant phylotype compared with the lean control group. In the blood microbiota alone, Succinivibrionaceae showed opposite correlations in the lean and obese NAFLD groups [45]. In a similar study in decompensated cirrhosis patients, Proteobacteria was identified as the predominant phyla in the blood and ascites samples. Proteobacteria was the most abundant phylum detected in ascites samples followed by Firmicutes, Actinobacteria, Bacteroidetes, and Gemmatimonadetes. The phylum composition in the blood samples was slightly different, with Proteobacteria accounting for nearly 90 %, while Actinobacteria, Firmicutes, and Bacteroidetes accounted for meager proportions [46].
Similar results were obtained in another study performed in the same year in HCC and liver cirrhosis patients. Blood microbial diversity was significantly reduced in HCC patients, compared with cirrhosis patients and controls. Blood microbiomes in all the groups were dominated by Firmicutes and Proteobacteria, followed by Actinobacteria and Bacteroidetes in meager concentrations. 5 microbial gene markers with the potential to distinguish HCC from controls were identified [47]. Comparatively similar results were obtained in Hepatitis B-Acute on Chronic Liver Failure (HB-ACLF) patients as well. HB-ACLF patients showed a significant increase in bacterial DNA compared to that in the liver cirrhosis and controls. HB-ACLF patients showed a considerable decrease in blood microbial diversity. HB-ACLF patients showed an enrichment of Moraxellaceae, Sulfurovum, Comamonas and Burkholderiaceae. But there was a depletion of Actinobacteria, Deinococcus-Thermus, Alphaproteobacteria, Xanthomonadaceae and Enterobacteriaceae in HB-ACLF patients compared to controls [48]. Literally, the same results were obtained in Liver cirrhosis patients in another study performed the same year. Blood microbial diversity was higher in cirrhotics (183 genera) as compared to controls (123 genera). Enterobacteriaceae was found to be significantly higher in cirrhotics whereas Akkermansia, Rikenellaceae and Erysipelotrichales were significantly lower in cirrhotics compared to controls [49].
Another study was performed to study the link between microbiota and endogenous Hydrogen Sulphide (H2S) in the circulation of Liver cirrhosis patients with (HE) and without Hepatic Encephalopathy (NHE). Endogenous H2S production was found to be significantly associated with different abundances in three taxa between the HE and NHE groups [50]. Another study deals with the work on blood microbiome in HCV related cirrhosis patients. Corynebacteriales, Diplorickettsiaceae, Diplorickettsiales, Corynebacterium, Aquicella and Undibacterium parvum had higher relative abundances in patients who reached a decrease in clinically significant portal hypertension (measured by HVPG), while Halomonadaceae, Oceanospirillales, Rhodospirillales and Massilia had higher relative abundances in patients who did not show decrease in Hepatic venous pressure gradient (HVPG). Corynebacteriales and Massilia were significantly correlated with the plasma markers of inflammation and metabolites at baseline [51].
Another important study is the work on circulating microbiome in cirrhosis patients with portal hypertension. The circulating plasma microbiome profile in cirrhosis patients was different from those of the controls. Cirrhosis patients with portal hypertension (PH) had an enrichment of Escherichia, Shigella, Dialister, Cnuella, Prevotella, and Comamonas and depletion of Bradyrhizobium, Curvibacter, Pseudomonas, Diaphorobacter, and Pseudarcicella. Enrichment of the genera Escherichia, Prevotella, Shigella, and Bacteroides was associated with severe PH in both hepatic and peripheral vein compartments. Prevotella, Escherichia, and Shigella abundance was correlated with IL-8 levels in the hepatic vein [52]. In another study by Li et al., many microorganisms were detected in 74.4 % patients, including viruses, bacteria, fungi and chlamydia [53]. In subsequent plasma metabolomics studies, it was found that plasma urobilinogen was directly correlated with circulating bacterial peptides linked to bilirubin and increased levels of Salmonella enterica and Escherichia coli associated peptides were found in cirrhotic patients [54,55].
In another recent study, HCC and cirrhosis patients showed an increased abundance of Ruminococcaceae and Bacteroidaceae in blood as compared to NAFLD patients. The authors were able to prove the translocation of these bacteria from gut to blood in HCC and Cirrhosis patients [56]. In the most recent study by Israelsen et al. [57], ALD patients showed a significant but temporary increase of microbial DNA quantity in the hepatic and systemic venous blood because of alcohol intervention. But the intervention did not cause a significant change of microbial DNA quantity in the hepatic and systemic venous blood in the healthy controls and NAFLD patients. In a study on virome sequencing conducted by Zhang et al. [58], it was demonstrated that the occurrence of the expanded anellome from the Anello Virus Torque Teno Mini Virus to Torque Teno Midi Virus was significantly higher among individuals going through acute liver failure and receiving liver transplantation, as opposed to various other patient groups with liver disorders. Taken together, the blood microbiome studies in liver diseases show that the blood microbial diversity increases in liver disease patients as compared to controls. The blood microbiome in liver disease patients is mostly dominated by the phylum Proteobacteria with Actinobacteria, Bactroidetes and Firmicutes in meagre concentrations (Table 1). The methodology for analysis of blood microbiome in liver disease is pictorially represented in Fig. 3.
Fig. 3 Methodology for analysis of blood microbiome in Liver disease patients.
At present, many advanced techniques including next-generation sequencing and metabolomics are being employed to detect the presence of blood microbiota and their metabolites. Biosensors and nanosensors can be developed for the detection of specific bacteria or their metabolites both in the gut and blood samples of liver disease patients. The sequencing techniques, metabolomics and biosensors should be incorporated into routine clinical analysis. The concentration of blood microbiota and their metabolites released from the gut into circulation can be decreased by employing different strategies. Antibiotics and drugs have been used to target microbes and their components in systemic and portal circulation earlier [59,60]. Small synthetic chemical molecules are being developed as drugs targeting the microbiota, their metabolites, and some of these molecules are in clinical trials [61,62]. Prebiotics [63], synbiotics and probiotics [64] have been used to restore the perturbed blood microbiota and darn the gut barrier integrity [61,65]. They can also reduce metabolic endotoxemia (LPS, LBP & LTA in blood), subsequent Toll-like receptor (TLR) activation and insulin resistance. Prebiotics and probiotics restore the normal gut microbiota and thereby reduce the other clinical complications of liver diseases such as diabetes, portal hypertension, cardiomopathy etc.
In addition, fecal microbiota transplantation (FMT) from healthy individuals has been used to restore the gut microbiota in liver disease patients. This will also reduce the release of microbiota derived metabolites like short chain fatty acids (SCFAs), aromatic amino acid (AAA) derived metabolites, trimethyl amines (TMA) and cholines into portal circulation. Fecal microbiota transplantation has the potential to mitigate intestinal permeability and it has been performed earlier in many liver diseases [[66], [67], [68]]. Further, this will reduce bile acid imbalance or dysregulation, lipogenesis and steatosis.
It is now recognized that blood microbial components derived from gut microbiome are present in the circulation of liver disease patients. In particular, the presence of microbiome and their metabolites in portal circulation has been clearly studied in liver disease. However, it is also quite important to note that the detection of microbial nucleic acids in the circulation may denote the presence of microbial signatures only and not the presence of live bacteria in the blood. Profound alterations in the blood microbiome and blood metabolome during liver disease indicate that these microbial components may play a role in the etiology and progression of the disease. Further research will lead to understanding of the molecular mechanisms of microbial translocation and their physiology in the development and progression of liver disease. This is also essential to develop the blood microbiome and their metabolites as therapeutic targets and biomarkers in liver disease.
The authors declare that the data associated with our study has not been deposited into a publicly available repository and no data was used for the research described in the article.
No additional information is available for this paper.
Dinakaran Vasudevan: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Arulraj Ramakrishnan: Supervision, Software, Resources, Project administration, Conceptualization. Ganesan Velmurugan: Supervision, Software, Resources, Project administration, Conceptualization.
The authors declare that there is no conflict of interest. All the authors have read and accepted the Manuscript.
DV thanks the Science and Research Engineering Board (SERB), Government of India for support through the National Postdoctoral Fellowship (File No. PDF/2019/001910). All authors thank the Founding trustees, KMCH Research Foundation for generous research support and research facilities. A special word of thanks to Dr. Krishnan Swaminathan for his constant support and encouragement.
Dinakaran Vasudevan, Email: dinakaran.svgev@gmail.com.
Ganesan Velmurugan, Email: vel@kmchrf.org.
The authors declare that the data associated with our study has not been deposited into a publicly available repository and no data was used for the research described in the article.