Phenylpropionic acid produced by gut microbiota alleviates acetaminophen-induced hepatotoxicity

Figures & data

Figure 1. Differential gut microbiota modulates susceptibility to APAP-induced hepatotoxicity.

(A-D) C57BL/6 mice from 6J or 6N were co-housed with mice from the other vendor for 4 weeks. After overnight fasting, mice were challenged i.p. with APAP (300 mg/kg) or PBS and sacrificed at 24 h post-treatment. (A and B) Fecal microbiota was analyzed by 16S rRNA gene amplicon sequencing of fecal samples collected at the end of co-housing. Shown are the nonmetric multidimensional scaling (NMDS) plot (a) and the Shannon index (b) based on the fecal microbiota analysis. Shown as indicators of APAP hepatotoxicity are serum ALT levels (c) and H/E staining of liver tissues (d).

(E-H) Pooled cecum contents from either 6J or 6N mice were orally gavaged to GF C57BL/6 mice. After 4 weeks, fecal samples were collected, and after overnight fasting, the mice were challenged i.p. with APAP (300 mg/kg) or PBS and sacrificed at 24 h post-treatment. Shown are NMDS plot (e) and Shannon index (f) based on 16S rRNA gene amplicon sequencing of the fecal samples. Serum ALT levels (g) and H/E staining of liver tissue (h). All data are shown as mean±S.D.: *, p < 0.05; ***, p < 0.001.

Figure 2. Identification of gut microbial metabolites potentially associated with differential susceptibility to APAP-induced hepatotoxicity using untargeted metabolomics.

(A and B) Respective liver and portal vein serum samples from 6J, 6N, GF, 6J_GF, and 6N_GF mice were subjected to LC-MS/MS-based untargeted metabolomic profiling. Principal component analysis and hierarchical clustering analysis of liver (a) and portal vein serum (b) samples.

(c) Schematics of filtering steps to compile a list of liver and portal vein serum metabolites associated with differential susceptibility to APAP hepatotoxicity.

(d) The signal ratios of metabolites that exhibited significant differences between 6J and 6N, as well as between 6J_GF and 6N_GF mice, are shown. Green and red colors denote metabolites that are more abundant in mice with 6J and 6N microbiota, respectively. ND, not detected.

Figure 3. The gut bacterial metabolite, phenylpropionic acid (PPA), alleviates APAP hepatotoxicity.

(a) A gut bacterial metabolic pathway for converting l-Phe to PPA and host metabolism of PPA in the liver. The metabolites identified by untargeted metabolomics associated with differential APAP hepatotoxicity are shown in green.

(b) PPA levels in cecal contents and cardiac serum of untreated mice were measured by LC-MS/MS. The dotted line denotes the limit of quantification (i.e., 0.2 μM).

(c) C57BL/6 mice from 6N were given access to PPA (0.4% in drinking water) or control water for 4 weeks. PPA levels in cecal contents or serum prepared from cardiac blood were measured using LC-MS/MS.

(d) C57BL/6 mice from 6N were given access to PPA (0.4% in drinking water) or control water for 4 weeks. After overnight fasting, the mice were challenged i.p. with APAP (300 mg/kg) or PBS and sacrificed at 24 h post-treatment. Serum ALT levels were measured as an indicator of APAP hepatotoxicity. All data are shown as mean±S.D. *, p < 0.05; **, p < 0.01; ***, p < 0.001

Figure 4. Phenylpropionic acid (PPA) alleviates APAP hepatotoxicity by lowering CYP2E1 levels.

(A-D) C57BL/6 mice from 6N were given access to PPA (0.4% in drinking water) or control water for 4 weeks. After overnight fasting, the mice were challenged i.p. with APAP (300 mg/kg) or PBS and sacrificed at 0, 0.5, 2, 6, 12, and 24 h post-treatment (n = 5–10/time point). Shown are serum ALT levels as an indicator of APAP hepatotoxicity (a), total hepatic GSH levels (b), APAP-CYS as an indicator of APAP-protein adducts (c), and hepatic APAP levels (d) at indicated time points.

(e and f) C57BL/6 mice from 6N were given access to PPA (0.4% in drinking water) or control water for 4 weeks. After overnight fasting, the mice were sacrificed, and hepatic microsomes were prepared to determine CYP2E1 and CYP1A2 protein levels and the extent of NAPQI formation. CYP2E1 and CYP1A2 protein levels in the microsomes were determined by western blot, with calnexin as a loading control (e). Hepatic microsomes were incubated with APAP, and the extent of NAPQI formation was determined by measuring the rate of APAP-GSH formation (f).

(G and H) C57BL/6 mice from either 6J or 6N were sacrificed after overnight fasting. CYP2E1 and CYP1A2 protein levels in the microsomes were determined by western blot, with calnexin as a loading control (g). Hepatic microsomes were incubated with APAP, and the extent of NAPQI formation was determined by measuring the rate of APAP-GSH formation (h). All data are shown as mean ± S.D. *, p < 0.05; ***, p < 0.001. NS, statistically not significant.

Figure 5. 6N cecal microbiota produces PPA as well as 6J cecal microbiota in vitro via a known l-Phe reductive pathway.

(a) Schematics showing factors that affect the PPA pool size in the gut. l-Phe is the primary substrate metabolized into PPA via an l-Phe reductive pathway involving Aat-FldH-FldBC-AcdA shown in (c). l-Phe amounts available for PPA production in the gut are governed by (1) l-Phe biosynthesized and secreted by gut bacteria and (2) l-Phe released from the proteolytic digestion of dietary proteins and gut microbial proteins by the gut microbiota. If the l-Phe level in the gut is not a limiting factor for PPA production, the presence and activity of gut bacteria (3) that mediate the conversion of L-Phe to PPA likely determine PPA levels in the gut.

(b) Cecal l-Phe levels in 6J and 6N mice were determined using LC-MS/MS.

(c) Two potential gut bacterial biosynthetic pathways for PPA production in the gut. The l-Phe reductive pathway involving Aat-FldH-FldBC-AcdA is previously known in four gut bacteria (see main text), and another biosynthetic pathway involving phenylalanine ammonia-lyase (PAL) has not been reported in mammalian gut bacteria.

(d) PPA production by the cecal contents of 6J mice in vitro.

(e) Comparison of PPA production by 6J or 6N cecal contents in vitro.

Data availability statement

16S rRNA amplicon sequencing data have been deposited under the NCBI BioProjects (https://www.ncbi.nlm.nih.gov/bioproject/) PRJNA604264 (6J/6N cohousing and gut microbiota transplantation), PRJNA940413 (6J/6N cecal microbiota), and PRJNA604973 (PPA supplementation). Any additional information required to reanalyze the data reported in this study will be available upon request.