An organic O donor for biological hydroxylation reactions

Significance

Microorganisms colonize anoxic ecological niches and use water as an oxygen donor for hydroxylation reactions in such environments. Here, we identify a unique source of oxygen, namely the metabolite prephenate, which enables the anaerobic biosynthesis of ubiquinone, a lipid important for bacterial bioenergetics. We predict that prephenate is involved as an oxygen donor in several biological pathways, including RNA modification. This study opens prospects for deciphering anaerobic metabolism and designing antimicrobial agents.

Abstract

All biological hydroxylation reactions are thought to derive the oxygen atom from one of three inorganic oxygen donors, O₂, H₂O_2, or H₂O. Here, we have identified the organic compound prephenate as the oxygen donor for the three hydroxylation steps of the O₂-independent biosynthetic pathway of ubiquinone, a widely distributed lipid coenzyme. Prephenate is an intermediate in the aromatic amino acid pathway and genetic experiments showed that it is essential for ubiquinone biosynthesis in Escherichia coli under anaerobic conditions. Metabolic labeling experiments with ¹⁸O-shikimate, a precursor of prephenate, demonstrated the incorporation of ¹⁸O atoms into ubiquinone. The role of specific iron–sulfur enzymes belonging to the widespread U32 protein family is discussed. Prephenate-dependent hydroxylation reactions represent a unique biochemical strategy for adaptation to anaerobic environments.

Hydroxylation reactions are involved in the biosynthesis, degradation, and modulation of the biological activity of a wide variety of molecules, by introducing a single oxygen (O) atom into a carbon–hydrogen bond. The enzymes that catalyze hydroxylation reactions, collectively referred to as hydroxylases, typically derive the added O atom from O₂ or its reduction product H₂O₂ (1, 2), and therefore require aerobic conditions to function. Enzymes such as flavin monooxygenases or heme/nonheme monooxygenases form large families of O₂-dependent hydroxylases that have been studied for decades (3–7). In anaerobic environments, H₂O is currently the only known O-donor for biological hydroxylation reactions. H₂O-dependent hydroxylases contain a molybdenum cofactor and represent a diverse family with well-known members like xanthine oxidase, resorcinol hydroxylase, or ethylbenzene dehydrogenase (8, 9).

Ubiquinone (UQ) is a redox lipid that contributes to the function of energy-producing respiratory chains in both aerobiosis and anaerobiosis (10–12). The biosynthesis of UQ requires three hydroxylation steps, which are catalyzed by flavin-dependent hydroxylases (UbiI, UbiH, and UbiF in Escherichia coli) under aerobic conditions, using O₂ as the O-donor (13) (Fig. 1). Under anaerobiosis, the iron–sulfur (Fe–S) proteins UbiU and UbiV are essential for UQ biosynthesis, but the O-donor used remains unknown (Fig. 1) (14, 15). UbiU-V are widely distributed in Proteobacteria and belong to the U32 peptidase family. Most of the twelve subfamilies of U32 proteins remain uncharacterized, but two have been linked to O₂-independent hydroxylation reactions of RNA molecules (16–18), raising the possibility of a common hydroxylation function with an as yet unidentified O-donor.

Fig. 1.

Aerobic and anaerobic pathways of UQ₈ biosynthesis in E. coli. The Ubi-enzymes common to both pathways are in black, those specific to the O₂-dependent pathway are shown in red, and those specific to the O₂-independent pathway are shown in blue (? for the unknown O- donor). The O atoms derived from the three hydroxylation steps are highlighted in yellow. R, octaprenyl chain shown in green in the UQ₈ structure; 4-HB, 4-hydroxybenzoic acid; OPP, 3-octaprenylphenol; UQ₈, ubiquinone-8.

Here, we show, using genetics and isotopic labeling, that prephenate, a molecule derived from the shikimate pathway, is the O-donor for the hydroxylation reactions of the UQ biosynthetic pathway operating under anaerobic conditions. We propose a prephenate-dependent hydroxylation chemistry that is likely shared by U32 proteins.

Results

The aro Pathway is Required for the Biosynthesis of UQ in Anaerobic Conditions.

Preliminary evidence indicated that the aromatic amino acids (aro) pathway is involved in O₂-independent UQ biosynthesis in E. coli (14). When grown anaerobically in an LB medium containing aromatic amino acids, Δaro mutants showed a strong decrease in the level of UQ (Fig. 2 A and B), whereas the latter remained unaffected in cells grown aerobically (SI Appendix, Fig. S1A). Furthermore, the anaerobically grown Δaro mutants accumulated octaprenyl-phenol (OPP) (Fig. 2C), a UQ pathway intermediate downstream of 4-hydroxybenzoic acid (4-HB) (Fig. 1), suggesting that the UQ deficiency was independent of 4-HB. Accordingly, the addition of 4-HB to cultures of Δaro mutants did not increase their UQ levels (SI Appendix, Fig. S1B).

Fig. 2.

O₂-independent biosynthesis of ubiquinone depends on the aro pathway in E. coli. (A) High Perfomance Liquid Chromatography-Electrochemical Detection (HPLC-ECD) analysis of lipid extracts from E. coli Δaro strains grown anaerobically in LB medium, UQ₁₀ added as internal standard. UQ₈ (B) and OPP (C) content in the strains analyzed in A. Mean ± SD (n = 3). ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05 by unpaired Student’s t test comparing to WT. ns, not significant. (D) Metabolic pathway of aromatic amino acids in E. coli, modified from ref. 17. The chorismate mutase (CM) and prephenate dehydratase (PDT) domains of PheA convert chorismate to prephenate and prephenate to phenylpyruvate, respectively (19). The black and gray arrows indicate the mutants not tested or tested in this study. The red and green disks indicate the proteins dispensable or essential for UQ₈ biosynthesis under anaerobic conditions, respectively. The ΔpheA and ΔtyrA strains are not deficient for UQ₈ but the double-mutant ΔpheA ΔtyrA is. The UQ pathway is highlighted in gray, and (D)MK₈ corresponds to the (demethyl) menaquinone pathway.

The Δaro mutants were also deficient in demethylmenaquinone and menaquinone, but this was true under both anaerobic and aerobic conditions (SI Appendix, Fig. S1 C and D). This is likely due to a deficiency in the isochorismate precursor, which is derived directly from chorismate in the aro pathway (Fig. 2D). Overall, this genetic analysis shows that preventing the synthesis of chorismate affects the biosynthesis of UQ under anaerobic conditions at a step downstream of OPP (Fig. 1).

Prephenate Is Essential for the Biosynthesis of UQ under Anaerobic Conditions.

Single deletions of genes located downstream of chorismate in the aro pathway had no effect on UQ levels (Fig. 3A). In contrast, combined inactivation of pheA and tyrA resulted in a severe UQ deficiency under anaerobic conditions (Fig. 3A) associated with increased OPP levels (SI Appendix, Fig. S2A), while UQ levels were normal under aerobic conditions (SI Appendix, Fig. S2B). Expression of the chorismate mutase (CM) domain of PheA (PheA_CM) complemented the deficiency (Fig. 3B), supporting that prephenate, the product of PheA_CM (Fig. 2D), is required for the O₂-independent biosynthesis of UQ. Accordingly, the addition of prephenate to cultures of the ΔpheA ΔtyrA strain increased UQ levels (Fig. 3C). We observed a residual level of UQ in the ΔpheA ΔtyrA strain, which was abolished upon inactivation of the O₂-dependent hydroxylases, UbiI and UbiH (Fig. 3C). The residual level of UQ was therefore attributed to the activity of the O₂-dependent hydroxylases using trace amounts of O₂ in the anaerobic culture medium, a hypothesis supported by the complete absence of UQ in the ΔaroD ΔubiIH mutant (Fig. 3D). Importantly, prephenate increased UQ levels in both strains (Fig. 3D). Collectively, the data support that prephenate is essential for UQ biosynthesis under anaerobic conditions.

Fig. 3.

Prephenate is required for the O₂-independent biosynthesis of ubiquinone. The indicated E. coli strains were grown anaerobically in either LB (A and B) or MOPS (C and D) media. UQ₈ content in the mutants downstream of chorismate (A), in the ΔpheA ΔtyrA double mutant containing either pTet (vec), pTet-pheA (PheA), or pTet-pheA-CM [PheA (CM)] (B), in the double mutant ΔpheA ΔtyrA and the quadruple mutant ΔpheA ΔtyrA ΔubiIH strains supplemented or not with 1mM of prephenate (Preph) (C), in the ΔaroD and the ΔaroD ΔubiIH strains supplemented with 1 mM of prephenate and/or 10 µM 4-HB and/or 100 µM of shikimate (Shik) (D). Mean ± SD (n = 3) except for D (n = 2). ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05 by unpaired Student’s t test comparing to WT for (A), to vec for (B) and to mutant strains without supplementation for (C and D). (E) Serial dilutions of the mutant strains ΔubiIH, or ΔaroD ΔubiIH containing either pBAD24 (vec), pES077 (UbiIH), pES154 (UbiUV), or pES185 (UbiU(C176A)V) on M9 succinate plates containing Phe, Tyr, Trp, 0.02% arabinose and supplemented or not with 200 µM shikimate. (F) Serial dilutions of the ΔpheAΔtyrA ΔubiH strain containing either pTet + pEB067 (vec), pES047 + pEB067 (PheA_CM), pES047 + pES232 (PheA_CM + UbiUV), pTet + pES154 (UbiUV), or pES077 + pEB067 (UbiIH on M9 succinate or glucose plates containing Phe, Tyr and 0.1% arabinose. Incubation at 37 °C was carried out for 48 h under aerobic conditions for succinate plates (+O₂ Succ) or under anaerobic conditions for glucose plates (−O₂ Glc, used as controls) (E and F). The results are representative of at least two independent experiments (E and F).

The Hydroxylation Steps Involving UbiU-V Are Dependent on Prephenate.

The expression of UbiU-V is tightly regulated by the transcription factor FNR (Fumarate and Nitrate Reductase); as a result, the two proteins are well expressed under anaerobiosis but very poorly expressed under aerobiosis (15). When expressed under aerobic conditions thanks to a constitutive promoter, UbiU-V can replace UbiI, UbiH, and UbiF in catalyzing the three hydroxylation reactions necessary for UQ biosynthesis, using an unidentified O-donor (15). In line with previous results, UbiU-V allowed robust aerobic growth of the ΔubiIH strain on succinate medium, which is strictly dependent on UQ biosynthesis (Fig. 3E). Consistent with the requirement of the 4Fe–4S cluster of UbiU for activity (15), the C176A mutation, which impairs the binding of the Fe–S cluster, abolished growth on succinate (Fig. 3E). Deletion of aroD (Fig. 3E) or pheA tyrA (Fig. 3F) abolished the growth conferred by UbiU-V on succinate, whereas anaerobic growth on glucose, which is independent of UQ, was maintained in all cases. The addition of shikimate rescued the growth of the ΔaroD ΔubiIH strain expressing UbiU-V (Fig. 3E), demonstrating that a metabolite downstream of shikimate in the aro pathway is required for UbiU-V function. Furthermore, the expression of PheA_CM in the ΔpheA ΔtyrA ΔubiH strain restored the function of UbiU-V (Fig. 3F). Taken together, these results demonstrate that prephenate is required for the UbiU-V-dependent hydroxylation steps and suggest that prephenate could be the O-donor for the O₂-independent hydroxylation reactions of UQ biosynthesis.

Biosynthesis of ¹⁸O₃-UQ from C₄[¹⁸O]-Shikimate under Anaerobic Conditions.

To test whether prephenate is the O-donor of the hydroxylation reactions of the anaerobic UQ pathway, we designed an in vivo labeling experiment using shikimate labeled with ¹⁸O at position 4 exclusively, which we termed C₄[¹⁸O]-shikimate. Shikimate was preferred over prephenate, which has a low bioavailability (20) and is unstable in solution, especially under acidic conditions (21–24). Moreover, the C4-hydroxyl group of shikimate remains intact during in vivo conversion to prephenate (25), and shikimate restored higher UQ levels when added to cultures of the ΔaroD strains (Fig. 3D and SI Appendix, Fig. S2C). The synthesis of C₄[¹⁸O]-shikimate was performed in five steps using commercial shikimate as the starting material (Fig. 4A and SI Appendix, Fig S4-S22). The general strategy was to protect all reactive functions of the molecule, except the hydroxyl at position 4 to selectively label it with ¹⁸O and then remove the protecting groups (Fig. 4A). LC–MS analysis of the final compound revealed that 64% of shikimate was labeled with ¹⁸O, while 36% remained unlabeled and contained ¹⁶O (SI Appendix, Fig. S3A).

Fig. 4.

Addition of C₄[¹⁸O]-shikimate to the ΔaroD strain leads to the biosynthesis of ¹⁸O₃-UQ in anaerobic conditions. (A) Synthetic route of ¹⁸O-shikimate (3R,4S,5R)-3,4[¹⁸O],5-trihydroxy-4-cyclohex-1-ene-1-carboxylic acid. i) APTS, CH₃OH; ii) TBDMSOtf, 2,6-lutidine, CH₂Cl₂; iii) Dess-Martin Periodinane, CH₂Cl₂, rt; iv) H₂¹⁸O, CH3C[¹⁸O]₂H, THF; v) a) NaBH₄, THF/H₂O; b) APTS, THF/H₂O. Mass spectra of UQ₈ from ΔaroD ΔubiIH strain grown anaerobically in the presence of 10 µM unlabeled ¹⁶O-shikimate (B) or C₄[¹⁸O]-shikimate (C). The peaks corresponding to adducts of either unlabeled UQ₈ or UQ₈ labeled with two ¹⁸O (¹⁸O_2-UQ₈) or three ¹⁸O (¹⁸O₃-UQ₈) are represented in blue (B) and red (C), respectively. Mass spectra representative of three independent experiments. (D) Quantification of UQ₈ H⁺ adducts (m/z 727 to 735) in WT and ΔaroD ΔubiIH cells grown with ¹⁶O-shikimate or C₄[¹⁸O]-shikimate. Mean ± SD (n = 3), see also SI Appendix, Fig S3.

C₄[¹⁸O]-shikimate and unlabeled shikimate, used as control, were added to anaerobic cultures of the ΔaroD ΔubiIH strain and yielded similar levels of UQ (SI Appendix, Fig. S3B). The mass spectra of UQ differed significantly depending on whether the shikimate added to the cultures was labeled or not (Fig. 4 B and C). With C₄[¹⁸O]-shikimate, the most prominent ions corresponding to H⁺, NH₄⁺, and Na⁺ adducts of UQ₈ were predominantly shifted by +4 mass units, and lower peaks at +6 were also detected (Fig. 4 B and C), indicating the incorporation of two and three ¹⁸O atoms, respectively. Quantification of these signals by single-ion monitoring of the H⁺ adducts (SI Appendix, Fig. S3 C and D) yielded a relative abundance consistent with a quantitative incorporation of three oxygen atoms from a precursor labeled at 64% with ¹⁸O (compare Fig. 4D and SI Appendix, Fig. S3 E–G). Importantly, when 4-HB, which provides the hydroxyl group at position C4 of UQ (Fig. 1), was added together with C₄[¹⁸O]-shikimate, the labeling of UQ₈ was not affected (compare Fig. 4C and SI Appendix, Fig. S3H). This result established that the ¹⁸O labels incorporated from C₄[¹⁸O]-shikimate are present at the three positions C1, C5, and C6 of the UQ₈ molecule. Collectively, the data demonstrate that an organic molecule derived from shikimate is the O-donor for the three hydroxylation reactions required to convert OPP into UQ₈ under anaerobic conditions.

Discussion

We have established genetically that prephenate is essential for the biosynthesis of UQ under anaerobic conditions. In addition, isotopic labeling experiments demonstrated that the three O atoms of UQ are derived exclusively from the O atom at position C4 of shikimate, the unique precursor of prephenate. These data show conclusively that the hydroxylation reactions of the O₂-independent pathway leading to the biosynthesis of UQ use prephenate as the O-donor. This is the first example of an organic molecule serving as an O-donor in a biological reaction, with only O₂, H₂O₂, and H₂O previously known as O-donors (1, 2, 8).

We have shown here that prephenate is required for the function of UbiU-V, two Fe–S proteins of the U32 family that are essential for the hydroxylation steps of the O₂-independent UQ biosynthetic pathway (14, 15). Interestingly, two other U32 members, RlhA and TrhP, are involved in O₂-independent hydroxylation reactions of RNA molecules (16–18), but a direct hydroxylation activity has not been demonstrated so far. Overall, an attractive hypothesis is that U32 proteins are bona fide hydroxylases that catalyze prephenate-dependent hydroxylation reactions, consistent with previous proposals (16, 17). While waiting for the development of in vitro activity assays, the availability of C₄[¹⁸O]-shikimate, through the chemical synthesis reported here, provides an opportunity to confirm in vivo the importance of prephenate for U32 protein-dependent pathways.

We propose the following mechanism for prephenate-dependent hydroxylation: i) a 2-electron oxidation of the substrate aromatic ring would lead to a cation intermediate, followed by ii) a concerted prephenate decarboxylation, releasing a nucleophilic hydroxide molecule that would quench the reactive cation to form the hydroxylated product and phenylpyruvate (Fig. 5). Interestingly, such a decarboxylation coupled to aromatization occurs in prephenate dehydratase (26) and in carboxy-S-adenosyl-L-methionine synthase (27). The unknown oxidant involved in step (i) probably has a high redox potential to oxidize the aromatic ring. The [4Fe–4S] clusters bound to UbiU and UbiV are functionally important (14, 15) and may serve as part of an electron transfer chain from the substrate to the yet unidentified electron acceptor. Remarkably, this role may be conserved in other U32 proteins as Fe–S clusters are likely to be a common feature of this family. Indeed, the function of TrhP and RlhA has been shown to depend either directly (18) or indirectly (16) on Fe–S clusters, and the four cysteine residues that coordinate the 4Fe–4S cluster in UbiU-V form a motif that is present in most U32 proteins (14).

Fig. 5.

Proposed molecular mechanism of prephenate-dependent hydroxylation: the reaction exemplifies the first hydroxylation step in the O₂-independent biosynthesis of UQ₈. R, octaprenyl chain.

The development of chemical analogs of prephenate as inhibitors of U32 proteins may be a promising antimicrobial strategy, considering that the aro pathway is found only in microorganisms and plants and that U32 proteins are not present in humans and are almost exclusively distributed in bacteria and to a lesser extent in archaea (17). UbiU-V has recently been implicated in the transition from chronic to acute infection in Pseudomonas aeruginosa (28), suggesting that inhibition of the anaerobic metabolism of this potent pathogen with prephenate analogs would affect its pathogenicity.

The existence of the O₂-independent UQ biosynthetic pathway was documented in 1978 (29), and it took 45 y to identify prephenate as the O-donor. Hydroxylation reactions occur in several other anaerobic pathways (30, 31), but the enzymes and O-donors involved remain largely uncharacterized. Of particular interest, one of the pathways for anaerobic microbial degradation of common aromatic pollutants, such as benzene and naphthalene, begins with hydroxylation of the aromatic ring with an unknown O-donor (32–34). Since about one-third of bacterial genomes show no evidence for the use of the molybdenum cofactor required for H₂O-dependent hydroxylation reactions (8, 35), these species must rely on other hydroxylation mechanisms under anaerobic conditions. Given the high conservation of the aro pathway among microbes (36), we propose that prephenate acts as an O-donor in some of these hydroxylation reactions.

Materials and Methods

Strains and Plasmids Constructions.

All the strains and plasmids used in this study are listed in SI Appendix, Tables S1 and S2, respectively. The knockout strains (SI Appendix, Table S1) were obtained by generalized Φ P1 transduction using donor strains from the Keio collection (37). For the generation of specific knockouts, PCR recombination with the λRed system was used (38). Alleles were combined in the same strain by generalized Φ P1 transduction. When necessary, the antibiotic resistance marker was removed using flippase (FLP recombinase expression from plasmid pCP20 as described previously (39). Cassette removal and plasmid loss were verified by antibiotic sensitivity and confirmed by PCR amplification.

Expression plasmids for pheA and ubiIH were constructed using primers listed in SI Appendix, Table S3. DNA fragments obtained by PCR on genomic DNA from the MG1655 strain were cloned in EcoRI/XhoI restriction sites of the pTet vector (15). The compatible plasmid pACYC-pBAD-ubiUV was constructed by transferring the EcoRV/HindIII fragment from pES232 into the pACYC184 vector digested by the same enzymes.

Media and Growth Conditions.

Strains were grown in lysogeny broth (LB) medium (10 g/L of tryptone, 10 g/L of NaCl and 5 g/L of yeast extract) or MOPS (3- (N-morpholino)propanesulfonic acid, https://www.genome.wisc.edu/resources/protocols/mopsminimal.htm) supplemented with 0.4% glycerol as the carbon source and 19 amino acids (10 mM Ser, 0.8 mM Ala/Gly/Leu, 0.6 mM Gln/Glu/Val, 0.4 mM Arg/Asn/Ile/Lys/Phe/Pro/Thr, 0.2 mM His/Met/Tyr, and 0.1 mM Cys/Trp), 5 vitamins (0.02 mM thiamine hydrochloride, 0.02 mM calcium pantothenate, 0.02 mM 4-aminobenzoic acid, 0.02 mM 4-HB, 0.02 mM 2,3-dihydroxybenzoic acid) according to Sakai et al. (17). Ampicillin (100 µg/mL), kanamycin (50 µg/mL), chloramphenicol (35 µg/mL), 4-HB (10 or 100 µM), prephenate (1 mM) (Sigma-Aldrich) and shikimate (10, 100, or 300 µM) (Sigma-Aldrich) or synthesized labeled ¹⁸O-shikimate (10 µM) were added when needed from aqueous stock solutions sterilized by filtration through 0.2-µm filters. PheA and PheA-CM expression from pTet in ΔpheAΔtyrA strain was induced by promotor leak.

Aerobic cultures were performed in glass tubes (15 cm long and 2 cm in diameter) at 37 °C, with 180 rpm shaking. 5 mL of fresh medium was inoculated with 100 µL of an overnight culture and the culture was grown overnight.

Anaerobic cultures were performed in Hungate tubes as previously described (14), using LB or MOPS media. LB medium was supplemented with 2.5 mg/L resazurin as an anaerobic indicator, with 100 mg/L L-cysteine (adjusted to pH 6 with NaOH) in order to reduce residual dioxygen, and with anti-foam (Sigma Life Science, 0.5 mL/L). Then, 5 or 13 mL of this medium was distributed in Hungate tubes and deoxygenated by bubbling high-purity argon for 40 min. The Hungate tubes were sealed and autoclaved. The resazurin was initially purple, it turned to pink after deoxygenation and become colorless after autoclaving. The resazurin remained colorless during culture indicating anaerobic conditions. The sterile MOPS medium was supplemented with 100 mM sterile KNO₃ as the final electron acceptor and 5 mL of this medium was distributed to the Hungate tubes and deoxygenated under sterile conditions by bubbling high-purity argon for 40 min. The tubes were sealed and were then used for cultures. The precultures were performed overnight at 37 °C in Eppendorf tubes filled to the top with either LB or MOPS media. The Hungate tubes were then inoculated through the septum with disposable syringes and needles with 100 μL of precultures and incubated at 37 °C without agitation. For Hungate tubes containing LB medium, the shikimate, the prephenate, and the antibiotic were added after autoclave and at the same time as the preculture. In contrast, for the Hungate tubes containing MOPS medium, the supplements were added before deoxygenation.

For testing complementation of the ΔpheA ΔtyrA ΔubiH mutant by ubiUV, in the presence or absence of the PheA CM activity and for testing complementation of the ΔaroD or ΔaroD ΔubiIH or ΔubiIH mutants by ubiUV, in the presence or absence of shikimate, strains were transformed and selection was performed in −O₂. Precultures were done in −O₂ in LB at 37 °C. The cultures were diluted in series (10-fold dilution at each step) in an M9 medium and 5 μL drops were plated on a minimal medium with 50 mM succinate and incubated at 37 °C in +O₂ or with 0.1% glucose and incubated in the absence of O₂. Arabinose was added at the concentrations indicated in figure legends for inducing expression of ubiUV genes.

Lipid Extraction and Quinone Analysis.

First, 5 mL of each culture was cooled on ice for at least 30 min before centrifugation at 3,200 g at 4 °C for 10 min. Cell pellets were washed in 1 mL ice-cold phosphate-buffer saline and transferred to preweighted 1.5-mL Eppendorf tubes. After centrifugation at 12,000 g at 4 °C for 1 min, the supernatant was discarded, the cell wet weight was determined and pellets were stored at −20 °C until lipid extraction, if necessary. Quinone extraction from cell pellets was performed as previously described (40). The dried lipid extracts were resuspended in 100 µL ethanol, and a volume corresponding to 1 mg of cell wet weight was analyzed by HPLC electrochemical detection—mass spectrometry with a BetaBasic-18 column (Thermo Scientific) at a flow rate of 1 mL/min with a mobile phase composed of 50% methanol, 40% ethanol, and 10% of a mix [90% isopropanol, 10% ammonium acetate (1 M), and 0.1% formic acid]. When necessary, MS detection was performed on an MSQ spectrometer (Thermo Scientific) with electrospray ionization in positive mode (probe temperature, 400 °C; cone voltage, 80 V). Single-ion monitoring detected the following compounds: UQ₈ (M+H⁺), m/z 727 to 728, 6 to 10 min, scan time of 0.2 s; ¹⁸O-UQ₈ (M+H⁺), m/z 729 to 730, 6 to 10 min, scan time of 0.2 s; 2(¹⁸O)-UQ₈ (M+H⁺), m/z 731 to 732, 6 to 10 min, scan time of 0.2 s; 3(¹⁸O)-UQ₈ (M+H⁺), m/z 733 to 734, 6 to 10 min, scan time of 0.2 s; 4(¹⁸O)-UQ₈ (M+H⁺), m/z 735 to 736, 6 to 10 min, scan time of 0.2 s; UQ₈ (M+ NH₄⁺), m/z 744 to 745, 6 to 10 min, scan time of 0.2 s; UQ₁₀ (M+NH₄⁺), m/z 880 to 881, 10 to 17 min. MS spectra were recorded between m/z 600 and 900 with a scan time of 0.3 s. ECD and MS peak areas were corrected for sample loss during extraction on the basis of the recovery of the UQ₁₀ internal standard (40). The absolute quantification of UQ₈ based on the m/z 744 to 745 signal at 7.8 min was performed with a standard curve of UQ₈ ranging from 6.25 to 50 pmol UQ₈.

Synthesis of C₄[¹⁸O]-Shikimate.

The synthesis of C₄[¹⁸O]-Shikimate is described in SI Appendix, Supplementary text. To determine the percentage of labeled shikimate, 3 or 9 µL of an aqueous solution of the synthetized product (0.34 mg/mL) was analyzed by HPLC–MS with a BetaBasic-8 column (Agilent C8 (4.6 × 150 mm; 5 µm) at a flow rate of 0.7 mL/min with a mobile phase composed of 0.1% formic acid in water. MS detection was performed on an MSQ spectrometer (Thermo Scientific) with electrospray ionization in negative mode (probe temperature, 450 °C; cone voltage, 60 V). Single-ion monitoring detected the following compounds: ¹⁶O-shikimate (M-H), m/z 172.5 to 173.5, 0 to 6 min, scan time of 0.2 s; ¹⁸O-shikimate (M-H), m/z 174.5 to 175.5, 0 to 6 min, scan time of 0.2 s. MS spectra were recorded between m/z 150 and 300 with a scan time of 0.3 s. The percentage of C₄[¹⁸O]-Shikimate was calculated as a ratio (A175/(A175+A173))*100 in which A175 and A173 correspond to the peak area of the SIM signals at m/z 175 and m/z 173, respectively. The value obtained (64 ± 1.2%) corresponds to the average of the values measured for the two injections.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

All data are available in the main text or the SI Appendix.