A human gut Faecalibacterium prausnitzii fatty acid amide hydrolase

Abstract

Undernutrition in Bangladeshi children is associated with disruption of postnatal gut microbiota assembly; compared with standard therapy, a microbiota-directed complementary food (MDCF) substantially improved their ponderal and linear growth. Here, we characterize a fatty acid amide hydrolase (FAAH) from a growth-associated intestinal strain of Faecalibacterium prausnitzii cultured from these children. This enzyme, expressed and purified from Escherichia coli, hydrolyzes a variety of N-acylamides, including oleoylethanolamide (OEA), neurotransmitters, and quorum sensing N-acyl homoserine lactones; it also synthesizes a range of N-acylamides, notably N-acyl amino acids. Treating germ-free mice with N-oleoylarginine and N-oleolyhistidine, major products of FAAH OEA metabolism, markedly affected expression of intestinal immune function pathways. Administering MDCF to Bangladeshi children considerably reduced fecal OEA, a satiety factor whose levels were negatively correlated with abundance and expression of their F. prausnitzii FAAH. This enzyme, structurally and catalytically distinct from mammalian FAAH, is positioned to regulate levels of a variety of bioactive molecules.

Figure 1.. F. prausnitzii strain-specific degradation of N-acylethanolamines.

(A) Absolute abundances of organisms comprising the bacterial consortium in cecal contents collected at the time of euthanasia. (B) Oleoylethanolamide (OEA) and palmitoylethanolamide (PEA) levels in the cecal contents of animals from each of the two treatment groups. Each dot represents a sample from a mouse. Mean values ± SD are shown. ****P < 0.0001; ***P < 0.001 (t-test followed by Bonferroni correction). (C) Quantification of PEA-d₄ levels using targeted UHPLC-QqQ-MS of medium and cell pellets prepared from F. prausnitzii Bg7063 monocultures after incubation with PEA-d₄ for 1 h. Each dot represents a biologic replicate in the indicated treatment group. Mean values ± SD are shown. ****P < 0.0001; ***P < 0.001 (1-way ANOVA followed by post-hoc Tukey’s test). (D) Quantification of PEA-d₄ levels in monocultures of the indicated F. prausnitzii strains treated with PEA-d₄ for 1 h. NS, not significant (1-way ANOVA followed by post-hoc Tukey’s test).

Figure 2.. Characterization and identification of F. prausnitzii FAAH.

(A) Tandem mass spectrometry fragmentation profile of N-palmitoylarginine (C16:0-Arg) in F. prausnitzii Bg7063 monocultures incubated with (i) unlabeled PEA (indicated in green), C16:0-Arg produced from PEA labelled with ¹³C (blue asterisk) at the carbonyl carbon (highlighted in blue) and a chemically synthesized unlabeled C16:0-Arg standard (indicated in red). (B) Quantification of labeled PEA (PEA-d₄), palmitic acid-d₄ (C16:0-d₄) and N-palmitoyl-L-arginine-d₄ (C16:0-Arg-d₄) in reactions containing F. prausnitzii Bg7063 cell lysates. (C) Quantification of PEA-d₄ and C16:0-Arg-d₄ levels in reaction mixtures containing a cell lysate from E. coli transformed with a CGOBPECO_01956 expression vector, a lysate from E. coli containing the empty vector, purified protein (2 μg) or a no addition (water) control. (D) Reaction rates for purified F. prausnitzii FAAH catalyzed hydrolysis of PEA, OEA, C18:1-Arg, and C18:1-His at 100 μM (see Methods). (E) Reaction rates for purified F. prausnitzii FAAH catalyzed condensation of C16:0 or C18:1 (100 μM) with arginine (100 mM). Mean values ± SD are plotted in panels B-D. All assays involved 3 replicates per condition.

Figure 3.. F. prausnitzii FAAH synthetic and hydrolytic activity.

(A) Acylation activity determined in reactions containing the indicated fatty acid (100 μM), arginine (100 mM) and purified F. prausnitzii FAAH (20 μg/ml) (see Methods). The mass spectrometry ion intensity peak area of the most prominent acylation product, oleoylarginine (C18:1-Arg), was set as 100%. The relative utilization preference of the indicated fatty acid as a substrate for FAAH was calculated by dividing the peak area of the corresponding acylarginine by the peak area of C18:1-Arg. (B) Amine donor condensation activity measured by incubating each amine (10 mM) with the oleic acid (100 μM), in the presence of F. prausnitzii FAAH (20 μg/mL). The mass spectrometry ion intensity peak area of the most favored condensation product, C18:1-Arg, was set as 100%. The relative condensation activity of each amine was then calculated by dividing the peak area of the corresponding C18:1-amine by the peak area of C18:1-Arg. (C) FAAH hydrolytic activity for a variety of N-acyl amides with different acyl chains and amine donors. Results are expressed as a percentage of the hydrolytic rate of the indicated N-acyl amide divided by the hydrolytic rate of C18:1-Arg. (D) Reaction rates for F. prausnitzii FAAH catalyzed hydrolysis of quorum sensing N-acyl homoserine lactones. Mean values ± SD are plotted in panels A-D. All assays involved 3 replicates per condition.

Figure 4.. Host effects of F. prausnitzii FAAH synthetic products and correlation between F. prausnitzii FAAH abundance and fecal levels of N-acyl amides in undernourished children.

(A) Experimental design. Groups of 4–5-week-old male germ-free animals received the indicated metabolite by oral gavage on experimental days 1, 2, 3, and 4 (n=5 mice/treatment group). Cecal levels of OEA, C18:1-Arg, and C18:1-His in germ-free mice were measured using LC-QqQ-MS. Each dot represents a sample from a mouse in the indicated treatment group. Mean values ± SD are plotted. NS, not significant (1-way ANOVA followed by post-hoc Tukey’s test). (B) Heatmap of log2-fold change in levels of significantly differentially expressed transcripts (rows) that contribute to the 46 pathways (table S8C) whose expression was significantly enriched in both the ileum and jejunum upon treatment with C18:1-Arg and C18:1-His. (C) Fecal levels of OEA and PEA in children (n=117) prior to- and 3-months after initiation of RUSF or MDCF-2 treatment. Data are visualized using violin plots; median values for each analyte are shown. p-values shown were based on Wilcoxon matched-pairs signed rank test. NS, not significant. (D) Fecal levels of C18:1-Arg in samples obtained from the ‘low MAGBg0005 group’ and ‘high MAGBg0005 group’. The p-value was calculated using a Mann-Whitney test.