Riboflavin Biosynthesis and Overproduction by a Derivative of the Human Gut Commensal Bifidobacterium longum subsp. infantis ATCC 15697

Abstract

Riboflavin or vitamin B₂ is the precursor of the essential coenzymes flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD). Despite increased interest in microbial synthesis of this water-soluble vitamin, the metabolic pathway for riboflavin biosynthesis has been characterized in just a handful of bacteria. Here, comparative genome analysis identified the genes involved in the de novo biosynthetic pathway of riboflavin in certain bifidobacterial species, including the human gut commensal Bifidobacterium longum subsp. infantis (B. infantis) ATCC 15697. Using comparative genomics and phylogenomic analysis, we investigated the evolutionary acquisition route of the riboflavin biosynthesis or rib gene cluster in Bifidobacterium and the distribution of riboflavin biosynthesis-associated genes across the genus. Using B. infantis ATCC 15697 as model organism for this pathway, we isolated spontaneous riboflavin overproducers, which had lost transcriptional regulation of the genes required for riboflavin biosynthesis. Among them, one mutant was shown to allow riboflavin release into the medium to a concentration of 60.8 ng mL^-1. This mutant increased vitamin B₂ concentration in a fecal fermentation system, thus providing promising data for application of this isolate as a functional food ingredient.

Keywords: gut commensal; health benefit; probiotic; vitamin B2; vitamin biosynthesis.

FIGURE 1

Flavin adenine dinucleotide (FAD) biosynthesis and corresponding gene loci predicted in the genome of Bifidobacterium longum. (A) General bacterial pathway of FAD biosynthesis. Stoichiometrically, the formation of riboflavin requires one equivalent of GTP and two equivalents of ribulose 5-phosphate. Reactions are catalyzed by enzymes: RibAB (1, 5), bifunctional GTP cyclohydrolase II [EC:3.5.4.25]/(3,4-dihydroxy-2-butanone 4-phosphate synthase [EC:4.1.99.12]; RibD (2, 3), bifunctional diaminohydroxyphosphoribosylaminopyrimidine deaminase [EC:3.5.4.26]/5-amino-6-(5-phosphoribosylamino)uracil reductase [EC:1.1.1.193]; unknown enzyme (4), 5-amino-6-(5-phospho-D-ribitylamino)uracil phosphatase [EC:3.1.3.104]; RibH (6), 6,7-dimethyl-8-ribityllumazine synthase [EC:2.5.1.78]; RibE (7) riboflavin synthase [EC:2.5.1.9]; RibCF (8, 9), bifunctional riboflavin kinase [EC:2.7.1.26]/FAD synthetase [EC:2.7.7.2]; Based on Bacher et al., 2000; (B) Gene loci encoding for the riboflavin synthesis enzymes in Bifidobacterium infantis ATCC 15697. Arrow marks the promoter, lollipop marks the FMN riboswitch upstream the rib cluster and ribU gene.

FIGURE 2

Comparative analysis of riboflavin biosynthesis in Bifidobacterium. (A) Comparative heatmap representing the distribution of riboflavin-associated genes from Bifidobacterium infantis ATCC 15697 across the Bifidobacterium genus. Color gradients indicate the percentage of identity in BLASTP alignments of open reading frames (ORFs) derived from bifidobacterial genomes grouped by origin of isolation; (B) Entanglement trees obtained from the concatenation of rpoB, groEL vs. ribU, ribCF genes across the genus Bifidobacterium. The two Neighbor Joining (NJ) trees were built using the MEGA package (statistical validation of 100 bootstrap replicates) and visualized using the “Dendextend” package in R v3.6.2. Connecting lines are color coded based on the presence (red) or absence (blue) of a rib cluster. The outgroup is indicated in green. (C) Sequence alignment of FMN region in B. infantis and 13 bifidobacterial strains containing a complete rib cluster. Sequence conservation is indicated with a logo for the consensus. Colored asterisks indicate mutations in FMN riboswitch region upstream of the rib cluster identified in ROS-resistant isolates. Prediction of the FMN folding was based on Rfam aligment (bases color-coded based on sequence conservation to the Rfam model RF00050) and was predicted by Rnafold. The obtained structures were obtained using the Infernal and Rnafold predictions in https://structrnafinder.integrativebioinformatics.me/ (Arias-Carrasco et al., 2018).

FIGURE 3

Roseoflavin-resistant isolates. Relative expression of ribD in wild-type (WT) and ROS25 cells grown to the mid-exponential growth phase in mSM7lac with 0 or 20 ng mL^–1 riboflavin measured by qRT-PCR. *indicate statistically significant differences between ROS25 and WT samples (p-value < 0.05).

FIGURE 4

Growth yield (OD_600nm) and riboflavin production in batch cultures of ROS25 and wild-type (WT) grown in mSM7 media containing varying carbohydrate sources (1%). (A) OD_600nm; (B) intracellular riboflavin ng mL^–1 of culture; (C) extracellular riboflavin ng mL^–1 of supernatant. Values are means ± SD, n = 3. * indicates statistically significant differences between ROS25 and WT samples (P < 0.05); lowercase letters indicate statistical significant differences for the same strains on the diverse carbon sources between (p-value < 0.05).

FIGURE 5

Controlled pH batch fermentation of ROS25 in mSM7lac. (A) Time course of biomass (gray) and riboflavin production (intracellular, fuchsia; extracellular, purple). (B) Time course of sugars (lactose, green; galactose, light blue; glucose, blue) and fermentation products (acetic acid, orange; lactic acid, green). The figure represents one representative experiment of the triplicate.

FIGURE 6

Permeability of wild-type (WT) and ROS25 cultures grown for 6 h in mSM7lac. Dosage of (A) extracellular riboflavin; (B) β-galactosidase, (C) proteins in the supernatant. Values are means ± SD, n = 3. * indicates statistically significant differences between ROS25 and WT samples (p-value < 0.05).

FIGURE 7

Extracellular riboflavin in fecal cultures. Riboflavin concentration was determined in the supernatant of fecal cultures at 0, 16, and 24 h of incubation in the absence of supplements (C) and in presence of riboflavin producing bifidobacteria [wild-type (WT) and ROS25]. (A) Fecal cultures with alive microbiota; (B) fecal cultures with pasteurized gut microbiota. Lower letters indicate statistically significant differences at the same time point among C, WT, and ROS25 cultures (p-value < 0.05); upper case letters indicate statistically significant differences for the same culture at various time points (p-value < 0.05).