Dysregulated meta-organismal metabolism of aromatic amino acids in alcohol-associated liver disease

Abstract

Background: Chronic alcohol consumption impairs gut barrier function and perturbs the gut microbiome. Although shifts in bacterial communities in patients with alcohol-associated liver disease (ALD) have been characterized, less is known about the interactions between host metabolism and circulating microbe-derived metabolites during the progression of ALD.

Methods: A large panel of gut microbiome-derived metabolites of aromatic amino acids was quantified by stable isotope dilution liquid chromatography with online tandem mass spectrometry in plasma from healthy controls (n = 29), heavy drinkers (n = 10), patients with moderate (n = 16) or severe alcohol-associated hepatitis (n = 40), and alcohol-associated cirrhosis (n = 10).

Results: The tryptophan metabolites, serotonin and indole-3-propionic acid, and tyrosine metabolites, p-cresol sulfate, and p-cresol glucuronide, were decreased in patients with ALD. Patients with severe alcohol-associated hepatitis and alcohol-associated cirrhosis had the largest decrease in concentrations of tryptophan and tyrosine-derived metabolites compared to healthy control. Western blot analysis and interrogation of bulk RNA sequencing data from patients with various liver pathologies revealed perturbations in hepatic expression of phase II metabolism enzymes involved in sulfonation and glucuronidation in patients with severe forms of ALD.

Conclusions: We identified several metabolites decreased in ALD and disruptions of hepatic phase II metabolism. These results indicate that patients with more advanced stages of ALD, including severe alcohol-associated hepatitis and alcohol-associated cirrhosis, had complex perturbations in metabolite concentrations that likely reflect both changes in the composition of the gut microbiome community and the ability of the host to enzymatically modify the gut-derived metabolites.

FIGURE 1

Gut-derived microbial metabolites are dysregulated in the plasma of patients in various stages of the alcohol-associated liver disease (ALD) spectrum. (A) Heatmap of metabolite concentrations in patients with ALD compared to healthy controls. Log fold change was measured for HD (n = 10), patients with mAH (n = 16), sAH (n = 40), and AC (n = 10) and compared to HC (n = 29). (B) Concentrations of indole-3-propionic acid, serotonin, p-cresol sulfate, and p-cresol glucuronide were measured in the plasma of patients with ALD using liquid chromatography with tandem mass spectrometry. (C) ROC curves for distinguishing patients with ALD from healthy controls. AUROC for each group is illustrated. Values represent means ± SEM. Statistical analysis was conducted in SAS; values with different superscripts are significantly different (p < 0.05). Abbreviations: 4-EPS, 4-ethylphenyl sulfate; AC, alcohol-associated cirrhosis; ALD, alcohol-associated liver disease; HC, healthy controls; HD, heavy drinkers, mAH, moderate alcohol-associated hepatitis; sAH, severe alcohol-associated hepatitis; TMAO, trimethylamine N-oxide.

FIGURE 2

Liver expression of enzymes involved in phase II sulfonation and glucuronidation pathways are perturbed in patients with severe alcohol-associated liver disease. Heatmaps were created by log transforming TPM and grouping samples by disease categories, such as HC (n = 10), EAH (n = 12), AHL (n = 18), explant tissue from patients with sAH with liver transplant (ExAH, n = 10), NAFLD (n = 8), HCV (n = 9), and HCV_Cirr (n = 9). (A) Expression of sulfotransferases and phosphosulfate synthase enzymes that comprise the liver sulfonation pathway. (B) Expression of glucuronosyltransferases and phosphoglucomutases involved in the liver glucuronidation pathway. (C) Boxplots of SULT1A1 and SULT2A1 gene expression in TPM of RNA sequencing data from patients with various liver diseases. (D) Boxplots of UGT1A6 and UGT2B4. Error bars indicate SD, and (*) indicates q < 0.05. Abbreviations: AH, alcohol-associated hepatitis; AHL, AH with liver failure; EAH, early alcohol-associated hepatitis; ExAH, explant tissue from patients with severe AH after liver transplant; HC, healthy control; HCV_Cirr, HCV with cirrhosis; sAH, severe alcohol-associated hepatitis; SULT, sulfotransferase; TPM, transcripts per million; UGT, UDP glucuronosyltransferase.

FIGURE 3

Meta-analysis reveals perturbations of phase II metabolism enzymes in 3 separate bulk RNA sequencing data sets. Violin plots were created by applying a random effects’ model on each of the individual fold changes to create the summarized diamonds, followed by a Bonferroni adjustment for p-values. (A) Log2 fold change, between healthy controls and sAH explant tissue, of sulfotransferases involved in the liver sulfonation pathway. (B) Log2 fold change of phosphoglucomutases involved in the liver glucuronidation pathway. Abbreviations: AH, alcohol-associated hepatitis; AHL, AH with liver failure; EAH, early alcohol-associated hepatitis; HC, healthy control; HCV_Cirr, HCV with cirrhosis; sAH, severe alcohol-associated hepatitis; SULT, sulfotransferase TPM, transcripts per million; UGT, UDP glucuronosyltransferase.

FIGURE 4

Hepatic gene expression of enzymes involved in tryptophan metabolism is impaired with severe ALD. Heatmaps were created by log transforming transcripts per million and grouping samples by disease categories, such as HC (n = 10), early AH (n = 12), AH with liver failure (n = 18), explant tissue from patients with sAH with emergency liver transplant (ExAH, n = 10), NAFLD (n = 8), HCV (n = 9), and HCV_Cirr (n = 9). (A) Expression of enzymes involved in tryptophan metabolism pathways. (B) Boxplots of TDO2 and MAOA gene expression in transcripts per million from RNA-seq data of patients with various liver pathologies. Error bars indicate SD, and (*) indicates q < 0.05. Abbreviations: AH, alcohol-associated hepatitis; ALD, alcohol-associated liver disease; ExAH, explant tissue from patients with alcohol-associated hepatitis; HC, healthy control; HCV_Cirr, hepatitis C virus with cirrhosis; sAH, severe alcohol-associated hepatitis.

FIGURE 5

Protein expression of enzymes involved in p-cresol metabolism is decreased in patients with sAH. (A) Liver SULT1A1 and UGT1A6 protein expression measured by western blot from healthy controls and patients with sAH. (B) Densitometric analysis of protein expression for SULT1A1 and UGT1A6 in healthy controls and patients with sAH. Abbreviations: sAH, severe alcohol-associated hepatitis; SULT, sulfotransferase; UGT, UDP glucuronosyltransferase.