Elucidation of roles for vitamin B12 in regulation of folate, ubiquinone, and methionine metabolism

Abstract

Only a small fraction of vitamin B₁₂-requiring organisms are able to synthesize B₁₂ de novo, making it a common commodity in microbial communities. Initially recognized as an enzyme cofactor of a few enzymes, recent studies have revealed additional B₁₂-binding enzymes and regulatory roles for B₁₂ Here we report the development and use of a B₁₂-based chemical probe to identify B₁₂-binding proteins in a nonphototrophic B₁₂-producing bacterium. Two unexpected discoveries resulted from this study. First, we identified a light-sensing B₁₂-binding transcriptional regulator and demonstrated that it controls folate and ubiquinone biosynthesis. Second, our probe captured proteins involved in folate, methionine, and ubiquinone metabolism, suggesting that it may play a role as an allosteric effector of these processes. These metabolic processes produce precursors for synthesis of DNA, RNA, and protein. Thereby, B₁₂ likely modulates growth, and by limiting its availability to auxotrophs, B₁₂-producing organisms may facilitate coordination of community metabolism.

Keywords: chemical biology; cobalamin; metabolism; microbial regulation.

Fig. 1.

B₁₂-ABP synthesis. (A) Cyanocobalamin (CNB₁₂) was converted to a photoreactive probe by carbamate bond formation to a linker moiety containing a diazirine and click chemistry-compatible alkyne. (B) B₁₂-ABP was added directly to live Halomonas cells, incubated for 60 min, then the samples were UV-irradiated to induce covalent bond formation between the probe and B₁₂-binding proteins. Cells were lysed, azido-biotin was added to probe-labeled proteins by click chemistry, and labeled proteins were enriched on a streptavidin agarose resin. Enriched proteins were digested on-resin, followed by quantitative LC-MS proteomic analysis of probe-labeled proteins by the accurate mass and time tag method.

Fig. S1.

Assays of B₁₂-ABP specificity and live cell uptake. (A) Competition assays of B₁₂-ABP binding to transcobalamin, and (B) SDS/PAGE analysis of analysis of Halomonas HL-48 lysates after labeling with B₁₂-ABP. Fluorescence (Left) and Coomassie-stained (Right) images of the same gel show proteins bound by 0 (NP, no probe control), 10, 20, or 50 μM B₁₂-ABP probe.

Fig. S2.

Proteins (A) MetE, (B) the B₁₂-binding domain of MetH, (C) PhrR, and (D) FolD were recombinantly expressed and purified, then the proteins (2 μM) were labeled with B₁₂-ABP (2 μM). Protein labeling was also competed by concomitant addition of excess CNB₁₂ (10×, 25×, and 50× concentrations versus the B₁₂-ABP). Following labeling, azido-rhodamine was attached to B₁₂-ABP–labeled proteins via click chemistry, and the labeled protein solutions were separated by SDS/PAGE. The fluorescence gel images in the Left panels of A–D show probe labeling of the pure proteins and subsequent inhibition by increasing concentrations of CNB₁₂; panels on the Right side show Coomassie staining of protein abundance, displaying equivalent loading of gel lanes. Intensity of gel lanes was quantified using ImageJ, and shown in E as percent intensity compared with B₁₂-ABP labeling without competitive inhibition; n = 3 for the labeling and quantification experiments.

Fig. 2.

B₁₂-ABP captures 17 proteins in methionine, folate, and ubiquinone metabolism. Metabolites are shown in open boxes: 5,10-CH = THF, 5,10-methenyltetrahydrofolate; 5,10-CH2-THF, 5,10-methylene-THF; 5mTHF, 5-methyl-THF; 10f-THF, 10-formyl THF; DHF, dihydrofolate; H₂-MPt, dihydromonapterin; H₄-MPt, tetrahydromonapterin; H₂-NPt-P3, dihydroneopterin triphosphate; H₂-NPt, dihydroneopterin; HCY, homocysteine; Met, methionine; pABA, p-aminobenzoate; SAH, S-adenosylhomocysteine; SRH, S-ribosylhomocysteine; THF, tetrahydrofolate;. Enzyme abbreviations: CheR, chemotaxis signal relay system methyltransferase; FolD, bifunctional methylenetetrahydrofolate dehydrogenase (NADP+)/methenyltetrahydrofolate cyclohydrolase; FolE, GTP cyclohydrolase; FolK, 2-amino-4-hydroxy-6-hydroxymethyldihydropteridine diphosphokinase; FolM, alternative dihydrofolate reductase 1; GlyA, glycine hydroxymethyltransferase; LuxS, S-ribosylhomocysteine lyase; MsrB, methionine-R-sulfoxide reductase; PurN, phosphoribosylglycinamide formyltransferase 1; MetE, B₁₂-independent methionine synthase; MetF, 5,10-methylenetetrahydrofolate reductase; MetH, B₁₂-dependent methionine synthase; MetK, S-adenosylmethionine synthetase; Me-MCP, methyl accepting chemotaxis protein; MtnN, adenosylhomocysteine nucleosidase; MetX, homoserine O-acetyltransferase; MetZ, O-succinylhomoserine sulfhydrylase; PanB, 3-methyl-2-oxobutanoate hydroxymethyltransferase; PurU, formyltetrahydrofolate deformylase; ThyA, thymidylate synthase; UbiB, ubiquinone biosynthesis monooxygenase; UbiG, bifunctional 2-polyprenyl-6-hydroxyphenyl methylase/3-demethylubiquinone-9 3-methyltransferase; UbiE, 2-methoxy-6-polyprenyl-1,4-benzoquinol methylase; UbiI, 2-octaprenylphenol hydroxylase. ROS, reactive oxygen species.

Fig. 3.

Comparative genomics reconstruction of PhrR regulons in Gammaproteobactreria. (A) Predicted genes and candidate operons under regulation by PhrR in Halomonas HL-48. Genes, candidate PhrR-binding sites, and putative promoters are shown as rectangles, yellow circles, and small arrows, respectively. Sequence logo for PhrR-binding motif in the Halomonadaceae is shown in a box. Names and locus tags for PhrR-regulated genes are shown on top and bottom lines, respectively. The phrR (regulator) and phr (DNA photolyase) genes are in black and yellow, respectively. Genes in green and orange are involved in folate biosynthesis (fol) and cyclopropane fatty acid biosynthesis (cfa), respectively. Conserved members of the PhrR regulons in Gammaproteobacteria encoding functionally uncharacterized proteins (designated by Pfam/COG family numbers) are shown by gray rectangles. Experimentally tested PhrR sites are marked with a “T” in yellow circles. (B) Conserved core of PhrR regulons in 20 genomes of the Halomonadaceae. The table shows gene orthologs that are predicted to be regulated (light green squares) or not regulated (pink squares) by PhrR in each analyzed genome. The absence of a gene ortholog is shown by a blank space. (C) Genomic organization of PhrR-controlled loci in other Gammaproteobacteria.

Fig. S3.

(A) Palindromic DNA motifs are shown as sequence logos that were constructed by WebLogo using sequences of all predicted PhrR-binding sites for each taxonomic group. Candidate PhrR-binding motifs in different lineages of Gammaproteobacteria are characterized by similar 7-bp half-sites and an internal linker of variable length. (B) Common consensus of all PhrR-binding DNA motifs identified in Gammaproteobacteria. (C) Sequence logo of known binding sites for homologous light-inducible B₁₂-dependent regulator LitR from Bacillus megaterium and related Bacillus spp. (D) Consensus of known binding sites for homologous light-inducible B₁₂-dependent regulator LitR from Thermus thermophiles.

Fig. S4.

Phylogenetic footprinting of upstream regions of predicted PhrR regulated operons in Halomonas spp. Each gene is named according to Dataset S3; gene locus tags in Halomonas sp. HL-48 are given in parentheses. Candidate PhrR-binding sites are highlighted in yellow. Consensus sequences of the PhrR motif are shown in the top line in red. Nucleotides in the PhrR binding sites that correspond to the consensus motif are in red. PhrR binding site scores are given to the right of the first line of sequence for each entry. Strong binding sites have a score above 4.5; weak sites have scores between 4.0 and 4.5. Putative promoter elements (−35 and −10 boxes) are underlined. Coding regions of genes that are immediately downstream to PhrR binding sites are in blue.

Fig. S4.

Phylogenetic footprinting of upstream regions of predicted PhrR regulated operons in Halomonas spp. Each gene is named according to Dataset S3; gene locus tags in Halomonas sp. HL-48 are given in parentheses. Candidate PhrR-binding sites are highlighted in yellow. Consensus sequences of the PhrR motif are shown in the top line in red. Nucleotides in the PhrR binding sites that correspond to the consensus motif are in red. PhrR binding site scores are given to the right of the first line of sequence for each entry. Strong binding sites have a score above 4.5; weak sites have scores between 4.0 and 4.5. Putative promoter elements (−35 and −10 boxes) are underlined. Coding regions of genes that are immediately downstream to PhrR binding sites are in blue.

Fig. S4.

Phylogenetic footprinting of upstream regions of predicted PhrR regulated operons in Halomonas spp. Each gene is named according to Dataset S3; gene locus tags in Halomonas sp. HL-48 are given in parentheses. Candidate PhrR-binding sites are highlighted in yellow. Consensus sequences of the PhrR motif are shown in the top line in red. Nucleotides in the PhrR binding sites that correspond to the consensus motif are in red. PhrR binding site scores are given to the right of the first line of sequence for each entry. Strong binding sites have a score above 4.5; weak sites have scores between 4.0 and 4.5. Putative promoter elements (−35 and −10 boxes) are underlined. Coding regions of genes that are immediately downstream to PhrR binding sites are in blue.

Fig. S4.

Phylogenetic footprinting of upstream regions of predicted PhrR regulated operons in Halomonas spp. Each gene is named according to Dataset S3; gene locus tags in Halomonas sp. HL-48 are given in parentheses. Candidate PhrR-binding sites are highlighted in yellow. Consensus sequences of the PhrR motif are shown in the top line in red. Nucleotides in the PhrR binding sites that correspond to the consensus motif are in red. PhrR binding site scores are given to the right of the first line of sequence for each entry. Strong binding sites have a score above 4.5; weak sites have scores between 4.0 and 4.5. Putative promoter elements (−35 and −10 boxes) are underlined. Coding regions of genes that are immediately downstream to PhrR binding sites are in blue.

Fig. S5.

Multiple alignment of Gammaproteobacterial PhrR regulators and homologous LitR and CarH regulators. The sequence alignment was constructed using ClustalX. Gene locus tags and species names are listed in Dataset S2. PhrR proteins are characterized by an N-terminal DNA-binding domain from the (A) MerR family and (B and C) two C-terminal B₁₂-binding domains. Secondary structure elements in the B₁₂-binding domain according to the known 3D structure of the T. thermophilus LitR (PDB ID code 3WHP) are shown by pink boxes (α-helices) and yellow arrows (β-strands). Residues involved in B₁₂ binding in 3WHP (as calculated in the PDB database) are shown by green circles.

Fig. S6.

Gel-filtration analysis for the purified recombinant PhrR protein from Halomonas sp. HL-48 after refolding. Retention volumes of 60 and 70 mL correspond by size to the dimer and monomer fractions of the protein, respectively.

Fig. S7.

Spectrometry of recombinant PhrR protein binding to B₁₂. UV spectrum of the monomer fraction of PhrR incubated with fourfold molar excess of vitamin B₁₂ (cyanocobalamin), a prospective ligand of PhrR, is shown by red line. As a control, UV spectrum of the recombinant PhrR protein in the absence of ligand is shown by black line.

Fig. 4.

Experimental validation of the PhrR regulon in Halomonas sp. HL-48 by fluorescence polarization (FP) binding assay. (A) Interactions between the recombinant PhrR protein and a DNA fragments containing consensus PhrR-binding site (6 nM) shows that DNA binding is specific and enhanced in the presence of CNB₁₂ or adenosylcobalamin (AdoB₁₂). (B) Titration of the effect of B₁₂-ABP and CNB₁₂ on the interaction between the recombinant PhrR protein (20 nM) and its predicted consensus DNA binding site (2.5 nM). (C) Light disrupts AdoB₁₂-dependent binding of PhrR to DNA. The PhrR protein (3 µM) was preincubated with AdoB₁₂ (66 µM) in the dark or irradiated with light for 5 min and then 0.7, 1.5, and 3 µM of the resulting PhrB:AdoB₁₂ complexes were checked for interaction with its consensus DNA motif (6 nM). (D and E) Effect of increasing concentrations of PhrR mixed with 33-bp DNA fragments (10 nM) containing candidate PhrR binding sites in the presence of AdoB₁₂ (4 µM). (F) Sequences of validated DNA fragments containing consensus PhrR-binding site and natural PhrR sites from Halomonas sp. HL-48 genome. Sequence logo represents the consensus PhrR-binding motif in the Halomonadaceae.

Fig. 5.

Effect of light vs. dark growth conditions on Halomonas HL-48 gene expression and intracellular metabolite production. (A) Relative gene-expression levels of PhrR regulated folate genes; rpoB is a control not regulated by PhrR. n = 3. (B) Cyclopropane fatty acid levels determined by metabolomic analysis in WT Halomonas HL-48 and a PhrR mutant. n = 3. (C) Levels of folate derivatives; 5mTHF, 5-methyl tetrahydrofolate; 5fTHF, 5-formyl tetrahydrofolate; DHF, dihydrofolate. n = 3. Statistically significant differences in measured gene or metabolite levels in A–C were evaluated by t test (n = 3): one asterisk (*): 0.05 < P < 0.1; two asterisks (**): 0.05 < P < 0.001.