Characteristics of microbiome-derived metabolomics according to the progression of alcoholic liver disease

DOI: https://doi.org/10.21203/rs.3.rs-2408045/v1

Abstract

Background: Due to the global increase in alcoholic liver disease (ALD) rates, interest in gut-derived bacterial products is growing in targeted therapies. Identifying microbiome-derived metabolite signatures is challenging due to the complex patterns that have long-term effects on the development of ALD. We evaluated a feature of the gut-microbiota-derived metabolite signatures in patients with ALD.

Methods: A prospective cohort study was carried out between April 2017 and March 2022. Stool samples (healthy control [HC, n = 62], alcoholic fatty liver [AFL, n = 25], alcoholic hepatitis [AH, n = 80], and alcoholic cirrhosis [AC, n = 80]) were collected for the microbiota analysis by 16S rRNA sequencing and metabolite profiles by using GC-MS and LC-MS methods.

Results: Proteobacteria relative abundance increased in ALD, while Bacteroides decreased (p = 0.001). Fusobacteria levels were found to be higher in AH (p = 0.0001). A total of 103 metabolites were quantified and screened. 3-Indole propionic acid levels are significantly lower in AH and AC (p = 0.001). Surprisingly, AC increases indole-3-lactic acid (p = 0.04). AC had significantly lower levels of short-chain fatty acids (SCFAs) and bile acids (BAs). The levels of stercobilin, hexadecanedioic acid, and 3-methyladipic acid were significantly decreased in ALD. The pathways of linoleic acid metabolism, indole compounds, histidine metabolism, fatty acid degradation, and glutamate metabolism were closely related to ALD metabolism.

Conclusions: Short-chain fatty acids, bile acids, and indole metabolites were depleted according to the ALD progression. Microbial dysbiosis is associated with a shift in metabolite changes in ALD.

Clinicaltrials.gov, number NCT04339725.

Introduction

Alcoholic liver disease (ALD) is a fast-growing and highly prevalent threat to health, set to become a major cause of liver cancer and transplant. ALD is strongly linked to metabolic disorders and is caused by poor diet and lack of metabolic process. ALD comprises stages of hepatic changes and damage at the cellular level. It includes fatty changes in hepatic inflammation that result in lipid accumulation. These excruciating changes lead to the building of an extracellular matrix, resulting in fibrosis and then cirrhosis, which eventually progresses to hepatocellular carcinoma (HCC). This damage is instigated by alcohol overconsumption, which affects normal cellular and molecular functions [1]. Most heavy drinkers develop fatty liver. However, only a small proportion, about 10% of such a population, tends towards advanced ALD [2].

Accumulating evidence from the past decades, the gut microbiome has emerged as a contemporary yet consequential element of the gut-liver axis [3]. Compositional changes in the gut microbiome are directly associated with alcohol consumption despite liver disease development and are a risk factor for disease progression [3]. Bacterial translocation from the luminal space of the intestine to the liver via the portal vein, as well as elevated systemic levels of gut-derived microbial products, pose a greater threat to hepatocytes and other non-parenchymal cells through endotoxin exposure, resulting in inevitable liver abrasion [4]. Recent studies have shown that alcoholic consumers and ALD patients have higher bacterial endotoxins and circulating microbiomes in their peripheral blood than non-alcohol consumers [5]. Cirrhotic patients are very susceptible to bacterial infections and have a 4-5-fold increase in incident mortality and a 30–40% death rate with cirrhosis [6, 7]. From a comparative human cohort study, it was observed that ALD and alcoholic dependence were associated with intense shifts in the microbial community and that stabilizing the mucosal integrity or diminishing the cellular response to endotoxins might protect against experimental ALD [8, 9].

This investigates the clinical properties of metabolites in their networks that have been routinely applied as tools for clinical therapeutics [10, 11]. Metabolites can be discovered and used to develop key cancer therapies and reduce the cancer burden. The applications of metabolomics have continually been growing, which can lead to refinement of methods for measurement, analysis, and understanding of complex data sets [12] .

Metabolomics has a wide range of therapeutic applications in health and disease research [13], personal medicine [14], food and nutrients [15], microbiome research [16, 17] and more. The main analytical techniques of gas/liquid chromatography-mass spectrometry (GC/LC-MS) were used in a time-course experiment to better understand the molecular mechanisms of ALD and identify candidate metabolites involved in the liver under disease conditions [18, 19].

In this study, we investigated the role of microbiome and high-throughput untargeted metabolomics characterization of ALD using GC/LC-MS, which has shown that AFL, AH, and AC patient's metabolic systems. Our aim was to identify the metabolites mechanism that could propose new therapeutic approaches to ALD. We are well defined, effective, straightforward and powerful workflow to directly trace the metabolic regulations in various AFL, AH, and AC.

Materials And Methods

Study design

A total of 223 subjects, comprising HC, AFL, AH, and AC groups, were prospectively enrolled and analysed. The age range for eligibility was between 24 and 82 years. Baseline evaluation, BMI calculation, liver function tests, and screening for viral markers were done. Stool samples were collected prospectively from Hallym University Chuncheon Sacred Heart Hospital, Chuncheon, South Korea, from June 2018 to December 2019. These studies were approved.

Pathway analysis and statistical analysis

Using EzBioCloud, linear discriminant analysis (LDA) was performed to identify bacterial taxa that were differentially abundant in groups [21]. Using the LEfSe algorithm, taxonomic biomarker discovery of bacterial taxa that were differentially abundant in HC, AFL, AH, and AC groups was first acknowledged and verified using the Kruskal-Wallis H-test with adjustments for multiple comparisons (p < 0.05). Principal component analysis (PCA) was also performed for beta diversity. Microbial function was predicted by Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) based on the 16S rRNA gene [22].

A PCA score plot has been performed to suppress the outlier samples (data not shown). Using supervised techniques, particle least squares discriminant analysis (PLS-DA) and orthogonal PLS-DA (OPLS-DA) were classified. In the score plot, each point represents the patient's stool samples. R2 and Q2 values were assessed. Normalizations were estimated via the Pareto scaling algorithm. Pathway analysis was conducted with the Metabolomics Pathway Analysis (MetPa) using the global test algorithm for pathway enrichment and relative betweenness centrality to assess metabolite importance. MetaboAnalyst 5.0 software (https://www.metaboanalyst.ca/) was used extensively to handle metabolite values [23].

By using EzBioCloud Apps, LDA analysis was performed to identify bacterial taxa that were differentially abundant in groups. For all the data, the normalization was verified. The logarithmic transformation is applied before analyzing metabolic expression, metabolic networking, and data interpretation. P-values were calculated using a t-test (p < 0.05). If the Holm p < 0.05, FDR < 0.05, pathways impact > 0, and the number of metabolites hitting the pathway was greater than one, the pathway was considered significantly enriched. GraphPad Prism 8 software is used to obtain bar plots. The mean ± SD is used to represent metabolite values.

Microbiota and metabolites analysis: supplementary files

Results

Clinical characteristics of spontaneously ruptured ALD patients

A cohort study included 247 subjects, prospectively recruited and divided into 2 groups; metagenomics (n = 177) and metabolomics analysis (n = 70). Each analysis (metagenomics and metabolomics) is carried out according to disease conditions (healthy control [HC], n = 43 and 19; alcoholic fatty liver [AFL], n = 19 and 6; alcoholic hepatitis [AH], n = 53 and 27; and alcoholic cirrhosis [AC], n = 62 and 18) (Fig. 1A). The patients' median age (years), body mass index (BMI, kg/m2), aspartate transaminase (AST), alanine transaminase (ALT), gamma glutamyl transpeptidase (γ-GT), cholesterol, and creatinine (Cr) were recorded from liver function tests (Table 1). We estimated that p-values of AST, ALT and γ-GT were significantly high in the AH and AC groups (p < 0.001). Female patients in the AC group had a significant half-level γ-GT compared to male patients in the AC group, indicating that male patients are at a higher risk than female patients. The same phenomenon was concurrent with the previous study [24].

Table 1

Baseline characteristics of participants.
 

HC

AFL

AH

AC

p-value

Age (years)

61.6

± 7.6

58.7

± 6.8

52.1

± 12.7*

54.4

± 10.1*

< 0.001

BMI (kg/m2)

22.6

± 3.2

24.5

± 2.8

26.2

± 10.0*

23.6

± 3.3

0.028

AST (IU/L)

22.2

± 4.5

24.1

± 5.8

103.8

± 181.8*

100.3

± 129.1*

0.002

ALT (IU/L)

18.5

± 7.5

21.9

± 9.2

65.5

± 47.5*

37.9

± 43.0*

< 0.001

γ-GT (IU/L)

25.1

± 16.6

47.9

± 43.5

249.4

± 451.2*

418.4

± 830.1*

0.004

Chol (mg/dL)

172.1

± 42.5

183.9

± 42.4

165.2

± 47.5

129.7

± 50.4*

< 0.001

Cr(mg/dL)

0.9

± 0.2

0.94

± 0.5

0.9

± 0.2

0.8

± 0.3

NS
Note: * represents significant value, p < 0.05. Data are presented as mean ± standard error of the mean (SEM). n, number; HC, healthy control; AFL, alcoholic fatty liver; AH, alcoholic hepatitis; AC, alcoholic cirrhosis; BMI, body mass index; ALT, alanine aminotransferase; AST, aspartate aminotransferase; γ-GT, gamma glutamyl transferase; NS, not significant.

Characterisation of the bacterial component of the gut microbiota in stool and bacterial growth rate comparisons

Proteobacteria and Bacteroidetes have dominated the gut microbiota's composition and relative abundance at the class and genus levels. At the class level, Gammaproteobacteria (AH: 8.7%, AC: 21.1%, p = 0.0001), Negativicutes (AH: 14.8%, AC: 12.4%), Clostridia (AH: 17.0%, AC: 16.4%, p = 0.0001), and Bacteroidia (AH: 52.7%, AC: 35.2%, p = 0.0001) were the most abundant. The Fusobacteria_c (2.6%) and Betaproteobacteria (1.0%) were found in the AH group, whereas Bacilli (8.0%), Actinobacteria_c (1.2%), and Verrucomicrobiae (2.2%) were found in the AC group (Fig. 1B). At the genus level, the relative abundance of Escherichia, Prevotella, Alistipes, Bacteroides, Parabacteroides, Phascolarctobacterium, and Faecalibacterium was common in four groups. Eubacterium_g23, Oscillibacter, Dialister, and Paraprevotella were abundant only in the HC and AFL groups; Fusobacterium in the AH group; and Bifidobacterium, Haemophilus, Staphylococcus, Streptococcus, Akkermansia, and Lactobacillus in the AC group. Prevotella, Alistipes, Bacteroides, Parabacteroides, Phascolarctobacterium, and Faecalibacterium were markedly reduced in the AH and AC groups. Conversely, Escherichia abundance was much higher in the AH and AC groups (p = 0.001) (Fig. 1B).

The gut microbial signature suggests the severity of the disease is associated with the decline in phylogenetic diversity during disease progression in ALD. The α-diversity and β-diversity are measures of intra- and inter-variability in stool samples, respectively (Fig. 1C-D). The α-diversity was assessed using ACE, Shannon, and operational taxonomic unit (OTU) indices between 4 groups. The AH group's gut microbiome was significantly less diverse and decreased than the HC and AFL groups (p = 0.018). This result was more distinct and decreased in the AC group (p = 0.002), suggesting a decrease in the richness and evenness of the gut microbiome. Comparing the whole gut microbial composition in all the groups, PCA of generalized Unifrac distances was dispersed and lower in the AH group (p = 0.05), which became more distinct in the AC group (p = 0.001) than HC. The HC and HAC showed more uniform and close sample distances. This study revealed the phylum Proteobacteria showed an increased pattern in abundance and decreased in the phylum Bacteroidetes as the disease progressed, with the lowest in the AC group (p < 0.05) (Fig. 1D-E). The specific microbial taxa related with disease progression (the compositions of the gut microbiomes of AFL, AH, AC and HC) were compared using LEfSe method and Kruskal-Walli’s test. Figure 1F presents the heatmap of specific bacterial enrichment of the gut microbiome and the top 30 dominant bacterial genera in the ALD subjects. Prevotella, Faecalibacterium, Phascolarctobacterium, Oscillibacter, Alistipes, Megamonas, Parabacteroides, Lachnospira, and Roseburia were decreased in AC group while Escherichia, Veillonella, Enterobacteriaceae_g, Lactobacillus, Streptococcus, Staphylococcus, Klebsiella and Enterococcus was increased in AC group. One intriguing finding was noted that Fusobacterium which was markedly increased in only the AH group had shown to be significantly reduced in the AC group (p < 0.05). This could be further confirmed by correlation analysis which determines diversity shifts in each group. Microbiome associations remain significant for taxonomic assignment with metagenomics correlation and metabolic pathway regulation (see online supplementary Figure S1).

Effects of alcohol on gut microbial metabolites and metabolic transit profile

In Table S1, metabolites based on the volcano plot and fold change threshold (FC > 1) values, log2 (FC), and p-value from GC-MS spectra are listed. Initially, 103 metabolites were quantified and screened out in the metabolomics analysis of each stool sample. In AFL, chenodeoxycholic acid, valerate, and capsaicin are enlarged. An AH, stercobilin, azelaic acid, and tenofovir are significantly amplified. Sixteen metabolites such as IPA, 2-oxindole, methylimidazoleacetic acid, jasmonic acid, lithocholic acid, stercobilin, deoxycholic acid, azelaic acid, taurodeoxycholic acid, valerate, 3-methyladipic acid, adenosine, iso-valerate, capsaicin, guanine, and indole-3-acetic acid (IAA) were increased in AC. The FC > 1.0 and FC < 1.0 indicate a net increase and decrease in the metabolome concentration, respectively.

A univariate OPLS-DA score plot of AFL, 19.6%; AH, 23.2%; and AC, 27.7%, has discriminated in Fig. 2A-C. In Fig. 2D, the multivariate PLS-DA score plot of AFLD delivers 29.4% of metabolic variance. A PCA score revealed 32.3% (AFL), 29.2% (AH), and 38.2% (ALD). The OPLS-DA score revealed 25.7% (AFL), 26.5% (AH), and 33.2% (AC), which shows the metabolic discrimination in Figure S2A-C.

The untargeted relative quantification of ALD-containing metabolites was quantified. In Table S1, fold changes of metabolites based on valcano plot values are found to be increased in AFL, AH, and AC, separately. Similarly, AFL, AH, and AC are found to have decreased levels of 16, 16, and 41 metabolites, respectively. The FC values were calculated to indicate an abundance increase or decrease in the test sample group compared to the reference sample group. Table S2 lists metabolites with ANOVA significant abundance changes and post-hoc test results from fecal samples with a threshold of p < 0.05.

The multivariate metabolic characteristics (represented as dots of different colors) are shown in different groups. Compared to unsupervised PCA scores, the supervised PLS-DA score plot could better distinguish the different groups and screen out differential markers. A 95% confidence region and premutation test were done (data not shown). The R2 and Q2 values were confirmed, which provides the validity and suitability of this model. The GC-MS-based significant metabolites are summarized in Table S3.

Acetylcholine, cholic acid, lithocholic acid, deoxycholic acid, taurochenodeoxycholic acid, taurocholic acid, glycocholic acid, glycochenodeoxycholic acid, and chenodeoxycholic acid were listed (Fig. 3A). Acetic acid, propionic acid, butyric acid, iso-butyric acid, and iso-valeric acid were summarized by their regulation in ALD (Fig. 3B). The metabolic charges of decreased metabolites such as IPA, butyrate, IAA, and acetate are shown in Fig. 2F and Fig. 3C. IPA has a beneficial effect on the epithelial homeostasis of the small intestine under HFD feeding conditions. IPA may affect metabolic processes through modulating the gut microbiota. IPA caused a significant induction of serum levels of IPA, suggesting a potential role for IPA in mediating cross-talk between the gut and extraintestinal organs (Fig. 3C) [25].

The palmitoylcarnitine in AF, AH, and AC is enlarged in Fig. 3C. Palmitoyl-CoA acts as a substrate for ACOX1 and mitochondrial oxidation in this case, stimulating peroxisomal action that may moderately inhibit palmitoylcarnitine accumulation. The decreased levels of stercobilin, 3-methyladipic acid, urocanic acid, hexadecanedioic acid, and 9-octadecadienoic acid are significantly altered. 2-hydroxycinnamic acid, hexadecanamide, palmitoylcarnitine, and linoleic acid were increased (Fig. 3C).

However, differences in their catabolism result in opposite effects on mitochondrial fatty acid oxidation and energy production, which partly modify their PPAR-dependent metabolic effects. Stercobilin is a tetrapyrrolic bile pigment and a one end-product of heme catabolism. [26].

In Fig. 3A-D, the relative concentration of acetylcholine is up-regulated. Secondary BAs of lithocholic acid (LCA) and deoxycholic acid (DCA) are reduced in various ALD conditions. BAs are synthesized in the liver via cholesterol catabolism in a multi-enzymatic pathway, with the rate-limiting step being the initial conversion by the cytochrome P450 enzyme CYP7A1[27]. Cholestatic liver disorders are associated with impaired cardiovascular function. Cholestatic disorders can eventually result in cirrhosis of the liver [28]. From this, we quantified the metabolites specifically in AFL, AH, and AC, as specifically shown in Fig. 3A-D.

Identification of significantly different metabolic pathways

Combined with all the differential metabolites identified by GC-MS, we can clearly see the results of potential biomarkers. All the potential metabolites mentioned above were imported into MetaboAnalyst 5.0 for the metabolic pathway analysis. The imbalanced metabolic pathways have been quantified with the metabolic pathway topology analysis (left) and enrichment ratio (right) (Fig. 4A-C). The results of metabolic pathway enrichment and topological analysis show that ALD severity is increasing. In Fig. 4A, pyrimidine metabolism, glycerophospholipid metabolism, the pentose phosphate pathway, the citrate cycle, glyoxylate and dicarboxylate metabolism, linoleic acid metabolism, and histidine metabolism are related to AFL metabolic changes. As per Fig. 4B, fatty acid degradation, vitamin B6 metabolism, glycerophospholipid metabolism, nicotinate, nicotinamide, linoleic acid metabolism, and histidine metabolism are metabolically regulated by the AH metabolism. Lysine degradation, histidine metabolism, fatty acid degradation, tryptophan metabolism, and aminoacyl-tRNA biosynthesis were all deregulated in AC samples, as shown in Fig. 4C. Many stool metabolites have been linked to validated linoleic acid. In particular, the enrichment ratio of histidine metabolism is steadily rising from AFL, AH, and AC. In addition, the enrichment ratio of linoleic acid metabolism is moderate in the early conditions of AFL and moves to a high enrichment ratio in AH. Reversibly, in AC samples, the enrichment ratio of linoleic acid metabolism is lower than in AFL and AH groups. According to KEGG enrichment analysis, the different expressed metabolites and metabolite to metabolite connections are established in LA metabolism and histidine metabolism. According to this, signaling pathways are significantly enriched in metabolisms (red and yellow captions are filled).

Discussion

We highlighted that the microbiome is associated with the progression of ALD in humans. Here, we demonstrated different structures and compositional changes in the gut microbiome at steatohepatitis and cirrhosis stages. While a lot of studies are documented to account for gut microbiome changes in ALD patients, differences during disease stage development have not been put together and studied in one place. This cohort study facilitates not only the identification of gut dysbiosis but also the differentiation and pull-out of potential taxa involved in pathogenesis at different stages in ALD.

According to previous research on humans and animals, alcohol consumption contributes to an overabundance of Proteobacteria, a major phylum of gram-negative bacteria, in the gut as the disease progresses [29]. However, proteobacteria evidently play an imperative role in the development of diseases linked to the gut microbiota. However, they are not alone[30]. Studies have shown some Firmicutes as well as pathogenic roles in disease development. Endotoxin (lipopolysaccharide, LPS) mediated proinflammatory responses through TLR-4, and lipoteichoic acid (LTA) mediated proinflammatory responses through TLR-2 [31]. Proteobacteria is the most studied gram-negative bacteria among the composition of relatively dominant phyla because of its long association with pathophysiology, and scientists have predicted it as a biomarker for gut dysbiosis [30]. Furthermore, they may have an additive effect on disease pathogenesis by mediating the immune response and thus activating the immune system in the intestinal mucosa [32]. At the phylum level, Enterobacteriaceae, Proteobacteria, Streptococcaceae, Staphylococcaceae, Enterococcaceae, Lactobacillaceae, and Firmicutes were more abundant at AH and became more significantly distinguishable in AC. These microbiome overgrowths can aggravate intestinal permeability, which is associated with deteriorating liver conditions and complications in cirrhosis [33, 34]. Proteobacteria enrichment indicates a potentially more inflammatory active gut microbiota within the AH and AC groups [35].

As per Bajaj et al., they correlated chronic ALD with gut microbiome dysbiosis and demonstrated findings that Lachnospiraceae, Ruminococcaceae, and Rikenellaceae are negatively correlated, and Staphylococcae, Enterococcaceae, and Enterobacteriaceae are positively correlated with the severity of chronic ALD [36]. Decreased abundances of Lachnospiraceae, Ruminococcaceae, and Rikenellaceae are also found in AH and AC groups in relatively low abundances. Other studies have found conflicting results when considering the relative abundance of the Prevotellaceae family in ALD [37]. These studies portend a significantly negative trend with an increase in disease progression (AH to AC). This finding is consistent with a newly discovered cohort in which decompensated cirrhosis patients have a lower abundance of Prevotellaceae [38]. Likewise, the gut is enriched with a larger proportion of Prevotella species in the enterotype 2 classification of a healthy gut [39].

The Ruminococcaceae family includes genera Ruminococcus, Oscillibacter, Sporobacter, and Subdoligranulum, which are commensal microbes that benefit the host with their butyrate-producing functions and were found to be less enriched in cirrhotic patients [40]. Faecalibacterium enrichment is thought to protect against both gastrointestinal and extraintestinal environments [41]. In the present study, the group of ALD patients had a lower relative abundance of the genus Faecalibacterium and other Ruminococcaceae family members than the healthy control group did. This result was consistent with previous studies [8]. The Lachnospiraceae family includes such commensal genera as Agathobacter, Coprococcus, Dorea, Butyrivibrio_g1 and Roseburia, which benefit the host by producing short-chain fatty acids (SCFAs) and immunomodulatory proteins that exert beneficial metabolic and immune-modulating properties [42, 43]. Recently, Boram et al. proposed that Roseburia spp. supplementation for 10 days protects disrupted gut barrier functions and restores the gut microbial dysbiosis caused by alcohol consumption in an animal model [44]. Roseburia abundance was depleted in the human cohort, which was in accordance with the present study, which provided decreasing trends of Roseburia with the progression of ALD. Christensenellaceae and other butyrate-producing bacteria were found to be significantly lower in ALD patients. This beneficial bacteria was enriched in NAFLD caused by a high-fat diet when sodium butyrate was supplemented [45].

The presence of Fusobacteria and Phascolarctobacterium is an intriguing finding from this study. Puneet et al. discovered that Fusobacteria that are enriched in heavy drinking control and predict the severity of AH. As a result, they proposed a theory in which Fusobacteria mimic defense mechanisms that increase Fusobacteria in alcohol overconsumption, and failure to proliferate such bacteria may result in disease severity. The SCFA-producing bacteria of the genus Phascolarctobacterium followed a similar pattern to Fusobacterium. The gut microbiome can influence human physiology and pathology. Nevertheless, it is challenging to assess the potential impact of microbial abundance on the functionality of the microbiome. Therefore, the analytical tool for functional analysis of pathways related to 16S rRNA gene sequences and differences in the microbiomes indicated that some microbe-derived functions, including the biosynthesis of secondary metabolites, metabolism, and ABC transporters, might be augmented in ALD patients. Prokaryotes and eukaryotes possess ABC transporters to facilitate the movement of nutrients or other biomolecules in or out of the cell, and any functional alteration in transporter expression can result in damage to the hepatic cell under stress conditions [46]. It was noted in the present study that the ABC transporter system has an overexpression of 1.5-fold in the AC group, and 4% of the total KEGG orthologs are associated with the ABC transporter system. The impression that the ABC transporter system is involved in host cell infection and that inhibitors of this system could be used as therapeutic targets for drug delivery This notion was further supported by a study conducted by Meehan et al. that showed lateral gene transfer between microbial communities and infected hosts could impact the gut environment associated with the development of disease [47]. These results indicate the spatial disarrangement and altered composition of the gut microbiome. Those microbiomes may cultivate environments that promote either the initiation or progression of ALD through altered microbial function.

The present human cohort has exemplified and highlighted the characterization of the gut microbiota related to ALD development. The relative abundance of signature commensal taxa (e.g., Bacteroides, Prevotella, and Ruminococcus) and the low proportion of acknowledged pathogens (Proteobacteria and certain Firmicutes) may serve as key indicators of gut microbiome robustness and should be considered in analyses of gut microbiomes with respect to human health in ALD.

Recently, the molecular and analytical effects of chronic alcohol consumption have been reported. Excess alcohol consumption is one of the major factors in the development of a fatty liver, which is characterized by the hepatic accumulation of sugars, amino acids, lipids, and fatty acids [48]. Histidine regulates gene expression and the biological activity of proteins' metabolism.

The gut flora can also metabolize dietary tryptophan into indole and its derivatives, such as indole-3-acetic acid (IAA), indole-3-acrylic acid (IAA), indole-3-aldehyde (I3A), and indole-3-propionic acid (IPA). IPA indirectly inhibits hepatic NF-κB signaling. Inhibition of hepatic NF-κB signaling can significantly attenuate hepatic inflammation and liver injury and decrease the histological activity of NASH. Indoles and associated metabolites have been tested in an indomethacin-induced intestinal injury mode. IPA and IAA show important functions in maintaining intestinal epithelial homeostasis [25].

Acylcarnitines have been shown to activate proinflammatory signaling pathways in monocytes, resulting in the secretion of inflammatory cytokines and chemokines such as tumor necrosis factor (TNF-), interleukin-6 (IL-6), and monocyte chemoattractant protein-1 (MCP-1) [49]. Hepatic steatosis and dyslipidemia were prevented by inhibiting lipogenesis and increasing insulin sensitivity [50].

Tetrapyrrolic bile pigment and stercobilin levels were decreased in stool samples from AFL, AH, and AC. The intestinal bacterial metabolite stercobilin, a feces pigment, induced pro-inflammatory activities including TNF- and IL-1 induction. Bilirubin diglucuronide is deconjugated and converted by intestinal bacteria to form urobilinoids, which are a group of metabolites consisting of porphyrin derivatives. Stercobilinogen is oxidized to stercobilin, which is responsible for the pigmentation of feces. The microbial metabolites urobilinogen, urobilin, stercobilinogen, and stercobilin that are formed by intestinal bacteria play a main role in liver diseases [51].

The levels of 2-hydroxycinnamic acid and 3-methyladipic acid have significantly decreased. 2-hydroxycinnamic acid acts as a group of compounds and is extremely abundant in food, which may account for about one-third of the phenolic compounds in our diet. Hydroxycinnamic acids play the role of potent antioxidants. We investigated the effect of alcohol on human fecal metabolites using metagenomics and metabolomics approaches to better understand the metabolic disorders caused by four sample groups: AFL, AH, AC, and HC.

Our data suggest that metagenomics and metabolomics can reveal the therapeutic effects of metabolites against ALD. GC-TOF-MS was used to identify metabolites that have undergone significant alteration in human fecal samples collected with and without alcohol.

Our results show that the linolenic acid metabolism, SCFAs, BAs, and histone metabolism could be involved in the metabolic changes in all stages of liver disease. Stercobilin, hexadecanedioic acid, 2-hydroxycinnamic acid, 3-methyladipic acid, hexadecanamide, and palmitoylcarnitine could be potentially promising biomarkers for the diagnosis of ALD. Tryptophan metabolites of IPA and IAA expand gut dysbiosis and inhibit the synthesis of proinflammatory cytokines. The therapeutic potential of IPA and linolenic acid is highlighted by ALD analysis. Also, we found that alcohol consumption depletes IPA, IAA, linolenic acid, and some metabolites in human fecal samples and that such liver damage can be improved. Our study provides important therapeutic evidence that alcohol can enhance the metabolic burden of liver-gut microbial transportation.

Declarations

Supplementary Information

The online version contains supplementary material available. 

Author contributions

Conceptualization: R.G., D.J.K., K.T.S.; Data curation: all authors; Formal analysis: R.G., H.G., S.S.; Funding acquisition: K.T.S.; Investigation: K.T.S.; Methodology: R.G., H.G.; Software: R.G., H.G., S.J.Y.; Project administration: K.T.S.; Supervision: K.T.S.; Writing-original draft: R.G., H.G., D.J.K., K.T.S.; Writing – review & editing: R.G., H.G., Y.A.G., S.S., S.J.Y., D.J.K., K.T.S. All authors contributed to manuscript revision and read and approved the submitted version. 

Disclosure of potential conflicts of interest

There are no conflicts to declare. 

Ethical approval statement

This study was conducted in conformance with the ethical guidelines from the 1975 Helsinki Declaration as it is reflected by a priori approval of the institutional review board for human research. Informed consent on enrollment was obtained from each participant.

Consent for publication 

All subjects have written informed consent.

Funding

This research was supported by Hallym University Research Fund, the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF-2018M3A9F3020956, NRF-2019R1I1A3A01060447, and NRF-2020R1A6A1A03043026), and Korea Institute for Advancement of Technology (P0020622).

Availability of data and materials 

All data generated or analyzed during this study are included either in this article or in the supplementary information fles. 

Acknowledgements

Ki Tae Suk would like to thank the NRF of Korea, the Ministry of Education, Science, and Technology, and the Ministry of SMEs and Startups (MSS) for the funding support. R.G. specially thanks Ki Tae Suk for his support.

References

  1. Nath, B. and G. Szabo, Alcohol-induced modulation of signaling pathways in liver parenchymal and nonparenchymal cells: implications for immunity. Semin Liver Dis, 2009. 29(2): p. 166-77.
  2. Li, F., C.J. McClain, and W. Feng, Microbiome dysbiosis and alcoholic liver disease. Liver Research, 2019. 3(3): p. 218-226.
  3. Bajaj, J.S., Alcohol, liver disease and the gut microbiota. Nat Rev Gastroenterol Hepatol, 2019. 16(4): p. 235-246.
  4. Addolorato, G., et al., Gut microbiota compositional and functional fingerprint in patients with alcohol use disorder and alcohol-associated liver disease. Liver Int, 2020. 40(4): p. 878-888.
  5. Parlesak, A., et al., Increased intestinal permeability to macromolecules and endotoxemia in patients with chronic alcohol abuse in different stages of alcohol-induced liver disease. J Hepatol, 2000. 32(5): p. 742-7.
  6. Chan, C. and J. Levitsky, Infection and Alcoholic Liver Disease. Clin Liver Dis, 2016. 20(3): p. 595-606.
  7. Arvaniti, V., et al., Infections in patients with cirrhosis increase mortality four-fold and should be used in determining prognosis. Gastroenterology, 2010. 139(4): p. 1246-56, 1256 e1-5.
  8. Dubinkina, V.B., et al., Links of gut microbiota composition with alcohol dependence syndrome and alcoholic liver disease. Microbiome, 2017. 5(1): p. 141.
  9. Han, S.H., et al., Effects of probiotics (cultured Lactobacillus subtilis/Streptococcus faecium) in the treatment of alcoholic hepatitis: randomized-controlled multicenter study. Eur J Gastroenterol Hepatol, 2015. 27(11): p. 1300-6.
  10. Raja, G., et al., Advanced understanding of genetic risk and metabolite signatures in construction workers via cytogenetics and metabolomics analysis. Process Biochemistry, 2019. 86: p. 117-126.
  11. Coen, M., et al., NMR-based metabolic profiling and metabonomic approaches to problems in molecular toxicology. Chem Res Toxicol, 2008. 21(1): p. 9-27.
  12. Raja, G., et al., Recent Advances of Microbiome-Associated Metabolomics Profiling in Liver Disease: Principles, Mechanisms, and Applications. International Journal of Molecular Sciences, 2021. 22(3): p. 1160.
  13. Holmes, E., I.D. Wilson, and J.K. Nicholson, Metabolic phenotyping in health and disease. Cell, 2008. 134(5): p. 714-7.
  14. Li, B., et al., Novel Applications of Metabolomics in Personalized Medicine: A Mini-Review. Molecules, 2017. 22(7).
  15. Koulman, A. and D.A. Volmer, Perspectives for Metabolomics in Human Nutrition: An Overview. Nutr Bull, 2008. 33(4): p. 324-330.
  16. Li, M., et al., Symbiotic gut microbes modulate human metabolic phenotypes. Proc Natl Acad Sci U S A, 2008. 105(6): p. 2117-22.
  17. Noecker, C., et al., Defining and Evaluating Microbial Contributions to Metabolite Variation in Microbiome-Metabolome Association Studies. mSystems, 2019. 4(6).
  18. Ferrarini, A., et al., Metabolomic Analysis of Liver Tissues for Characterization of Hepatocellular Carcinoma. Journal of Proteome Research, 2019. 18(8): p. 3067-3076.
  19. Pan, H.Y., et al., LC/MS-Based Global Metabolomic Identification of Serum Biomarkers Differentiating Hepatocellular Carcinoma from Chronic Hepatitis B and Liver Cirrhosis. ACS Omega, 2021. 6(2): p. 1160-1170.
  20. Lu, Y., et al., Integrative transcriptomics and metabolomics explore the mechanism of kaempferol on improving nonalcoholic steatohepatitis. Food Funct, 2020. 11(11): p. 10058-10069.
  21. Segata, N., et al., Metagenomic biomarker discovery and explanation. Genome Biol, 2011. 12(6): p. R60.
  22. Douglas, G.M., R.G. Beiko, and M.G.I. Langille, Predicting the Functional Potential of the Microbiome from Marker Genes Using PICRUSt. Methods Mol Biol, 2018. 1849: p. 169-177.
  23. Pang, Z., et al., Using MetaboAnalyst 5.0 for LC–HRMS spectra processing, multi-omics integration and covariate adjustment of global metabolomics data. Nature Protocols, 2022. 17(8): p. 1735-1761.
  24. Guy, J. and M.G. Peters, Liver disease in women: the influence of gender on epidemiology, natural history, and patient outcomes. Gastroenterol Hepatol (N Y), 2013. 9(10): p. 633-9.
  25. Zhao, Z.H., et al., Indole-3-propionic acid inhibits gut dysbiosis and endotoxin leakage to attenuate steatohepatitis in rats. Exp Mol Med, 2019. 51(9): p. 1-14.
  26. Bjorndal, B., et al., Associations between fatty acid oxidation, hepatic mitochondrial function, and plasma acylcarnitine levels in mice. Nutr Metab (Lond), 2018. 15: p. 10.
  27. Vasavan, T., et al., Heart and bile acids - Clinical consequences of altered bile acid metabolism. Biochim Biophys Acta Mol Basis Dis, 2018. 1864(4 Pt B): p. 1345-1355.
  28. Ghonem, N.S., D.N. Assis, and J.L. Boyer, Fibrates and cholestasis. Hepatology, 2015. 62(2): p. 635-43.
  29. Bull-Otterson, L., et al., Metagenomic analyses of alcohol induced pathogenic alterations in the intestinal microbiome and the effect of Lactobacillus rhamnosus GG treatment. PLoS One, 2013. 8(1): p. e53028.
  30. Shin, N.R., T.W. Whon, and J.W. Bae, Proteobacteria: microbial signature of dysbiosis in gut microbiota. Trends Biotechnol, 2015. 33(9): p. 496-503.
  31. Warshakoon, H.J., M.R. Burns, and S.A. David, Structure-activity relationships of antimicrobial and lipoteichoic acid-sequestering properties in polyamine sulfonamides. Antimicrob Agents Chemother, 2009. 53(1): p. 57-62.
  32. Litvak, Y., et al., Dysbiotic Proteobacteria expansion: a microbial signature of epithelial dysfunction. Curr Opin Microbiol, 2017. 39: p. 1-6.
  33. Quigley, E.M., C. Stanton, and E.F. Murphy, The gut microbiota and the liver. Pathophysiological and clinical implications. J Hepatol, 2013. 58(5): p. 1020-7.
  34. Bajaj, J.S., Alcohol, liver disease and the gut microbiota. Nature Reviews Gastroenterology & Hepatology, 2019. 16(4): p. 235-246.
  35. Bajaj, J.S., et al., Specific Gut and Salivary Microbiota Patterns Are Linked With Different Cognitive Testing Strategies in Minimal Hepatic Encephalopathy. Am J Gastroenterol, 2019. 114(7): p. 1080-1090.
  36. Bajaj, J.S., et al., Altered profile of human gut microbiome is associated with cirrhosis and its complications. Journal of Hepatology, 2014. 60(5): p. 940-947.
  37. Chen, Y., et al., Dysbiosis of small intestinal microbiota in liver cirrhosis and its association with etiology. Sci Rep, 2016. 6: p. 34055.
  38. Bajaj, J.S., et al., Cognition and hospitalizations are linked with salivary and faecal microbiota in cirrhosis cohorts from the USA and Mexico. Liver Int, 2020. 40(6): p. 1395-1407.
  39. Hollister, E.B., C. Gao, and J. Versalovic, Compositional and Functional Features of the Gastrointestinal Microbiome and Their Effects on Human Health. Gastroenterology, 2014. 146(6): p. 1449-1458.
  40. Bajaj, J.S., et al., Colonic mucosal microbiome differs from stool microbiome in cirrhosis and hepatic encephalopathy and is linked to cognition and inflammation. Am J Physiol Gastrointest Liver Physiol, 2012. 303(6): p. G675-85.
  41. Eppinga, H., et al., Similar Depletion of Protective Faecalibacterium prausnitzii in Psoriasis and Inflammatory Bowel Disease, but not in Hidradenitis Suppurativa. J Crohns Colitis, 2016. 10(9): p. 1067-75.
  42. Chen, Y., et al., Characterization of fecal microbial communities in patients with liver cirrhosis. Hepatology, 2011. 54(2): p. 562-72.
  43. Ahn, S.B., et al., Randomized, Double-blind, Placebo-controlled Study of a Multispecies Probiotic Mixture in Nonalcoholic Fatty Liver Disease. Scientific Reports, 2019. 9(1): p. 5688.
  44. Seo, B., et al., Roseburia spp. Abundance Associates with Alcohol Consumption in Humans and Its Administration Ameliorates Alcoholic Fatty Liver in Mice. Cell Host Microbe, 2020. 27(1): p. 25-40 e6.
  45. Wang, L. and Y.-J.Y. Wan, The role of gut microbiota in liver disease development and treatment. Liver Research, 2019. 3(1): p. 3-18.
  46. Davidson, A.L., et al., Structure, Function, and Evolution of Bacterial ATP-Binding Cassette Systems. Microbiology and Molecular Biology Reviews, 2008. 72(2): p. 317-364.
  47. Meehan, C.J. and R.G. Beiko, Lateral gene transfer of an ABC transporter complex between major constituents of the human gut microbiome. BMC Microbiology, 2012. 12(1): p. 248.
  48. Liu, Y., et al., Gut Microbiome and Metabolome Response of Pu-erh Tea on Metabolism Disorder Induced by Chronic Alcohol Consumption. J Agric Food Chem, 2020. 68(24): p. 6615-6627.
  49. Enooku, K., et al., Altered serum acylcarnitine profile is associated with the status of nonalcoholic fatty liver disease (NAFLD) and NAFLD-related hepatocellular carcinoma. Sci Rep, 2019. 9(1): p. 10663.
  50. Jeyapal, S., et al., Substitution of linoleic acid with alpha-linolenic acid or long chain n-3 polyunsaturated fatty acid prevents Western diet induced nonalcoholic steatohepatitis. Sci Rep, 2018. 8(1): p. 10953.
  51. Sanada, S., et al., Intestinal microbial metabolite stercobilin involvement in the chronic inflammation of ob/ob mice. Sci Rep, 2020. 10(1): p. 6479.