Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 27.
Published in final edited form as: Biochemistry. 2015 Dec 1;54(49):7229–7236. doi: 10.1021/acs.biochem.5b00857

Mechanistic Investigation of cPMP Synthase in Molybdenum Cofactor Biosynthesis Using an Uncleavable Substrate Analogue

Bradley M Hover 1, Edward A Lilla 1, Kenichi Yokoyama 1,*
PMCID: PMC4847533  NIHMSID: NIHMS776048  PMID: 26575208

Abstract

Molybdenum cofactor (Moco) is essential for all kingdoms of life, plays central roles in various biological processes, and must be biosynthesized de novo. During its biosynthesis, the characteristic pyranopterin ring is constructed by a complex rearrangement of guanosine 5′-triphosphate (GTP) into cyclic pyranopterin monophosphate (cPMP) through the action of two enzymes, MoaA and MoaC. Recent studies revealed that MoaC catalyzes the majority of the transformation and produces cPMP from a unique cyclic nucleotide, 3′,8-cyclo-7,8-dihydro-GTP (3′,8-cH2GTP). However, the mechanism by which MoaC catalyzes this complex rearrangement is largely unexplored. Here, we report the mechanistic characterization of MoaC using an uncleavable substrate analogue, 3′,8-cH2GMP[CH2]PP, as a probe to investigate the timing of cyclic phosphate formation. Using partially active MoaC variants, 3′,8-cH2GMP[CH2]PP was found to be accepted by MoaC as a substrate and was converted to an analogue of the previously described MoaC reaction intermediate, suggesting that the early stage of catalysis proceeds without cyclic phosphate formation. In contrast, when it was incubated with wt-MoaC, 3′,8-cH2GMP[CH2]PP caused mechanism-based inhibition. Detailed characterization of the inhibited MoaC suggested that 3′,8-cH2GMP[CH2]PP is mainly converted to a molecule (compound Y) with an acid-labile triaminopyrimidinone base without an established pyranopterin structure. MS analysis of MoaC treated with 3′,8-cH2GMP[CH2]PP provided strong evidence that compound Y forms a tight complex with MoaC likely through a covalent linkage. These observations are consistent with a mechanism in which cyclic phosphate ring formation proceeds in concert with the pterin ring formation. This mechanism would provide a thermodynamic driving force to complete the formation of the unique tetracyclic structure of cPMP.

graphic file with name nihms-776048-f0001.jpg

Molybdenum cofactor (Moco) is a ubiquitous enzyme cofactor that mediates redox reactions in the active sites of a number of enzymes. Moco-dependent enzymes play central roles in purine and sulfur catabolism and drug metabolism in mammals (including humans), anaerobic respiration in pathogenic bacteria, and nitrate assimilation in plants.1,2 Importantly, Moco must be synthesized de novo in cells, as it is chemically unstable and cannot be taken up as a nutrient.1,2 Due to this essential role, Moco biosynthesis has been linked to a number of human health problems. For example, genetic mutations in human Moco biosynthetic genes cause a fatal metabolic disorder, Moco deficiency.1,3 Recently, Moco biosynthesis in pathogenic bacteria has been gaining significant interest because Moco is critical for the virulence of bacterial pathogens such as Mycobacterium tuberculosis and Pseudomonas aureus.4–6 In fact, the pharmacological inhibition of Moco biosynthesis was shown to be effective for the treatment of latent tuberculosis in a mouse model.7 Therefore, it is important to understand the molecular details of Moco biosynthesis in bacteria and humans.

During Moco biosynthesis in Escherichia coli, the characteristic pyranopterin ring is constructed by a complex rearrangement of GTP into cPMP.8 This unusual transformation involves the insertion of the guanine C-8 between C-2′ and C-3′ of ribose (Figure 1A)9 and requires two enzymes, MoaA and MoaC.9–11 Conventionally, MoaA, a member of the radical S-adenosyl-l-methionine (SAM) superfamily,12 was thought to be responsible for the formation of the pyranopterin backbone structure.11,13–17 However, recent characterization of the MoaA product by NMR and X-ray crystallography and functional characterization of MoaC revealed that MoaA catalyzes the transformation of GTP to 3′,8-cH2GTP (2; Figure 1A) and that the remaining conversion to cPMP is catalyzed by MoaC.18,19 Because of this unexpected finding regarding the catalytic function of MoaC, the mechanism of the MoaC-catalyzed transformation of 3′,8-cH2GTP to cPMP has been largely unexplored.

Figure 1.

Figure 1

Open in a new tab

Moco biosynthesis and the formation of cPMP. (A) Moco biosynthetic pathway. The symbols on GTP and cPMP indicate the origin of the carbon and nitrogen atoms in cPMP, as determined by isotope labeling studies.9 cPMP may also be in a hydrate form in solution.21 (B) Structure of the previously proposed MoaC substrate, pyranopterin triphosphate. (C) Current working hypothesis for the MoaC-catalyzed conversion of 3′,8-cH2GTP to cPMP.19 The proposed structural candidates for intermediate X are shown. (D) Uncleavable analogue of 3′,8-cH2GTP (3′,8-cH2GMP[CH2]PP).

Conventionally, MoaC was thought to be responsible only for the formation of the cyclic phosphate moiety of cPMP on the basis of the proposal that the MoaC substrate is pyranopterin triphosphate (5; Figure 1B).14,16 However, in the revised function (Figure 1A), the MoaC substrate is now known to be 3′,8-cH2GTP and the catalysis involves not only cyclic phosphate formation but also a complex rearrangement of the ribose and base moieties of 3′,8-cH2GTP.18,19 This significant revision in the role of MoaC has raised a question about the timing of cyclic phosphate formation in the complex function of MoaC catalysis. The timing of cyclic phosphate formation is of mechanistic interest because it involves the cleavage of a high-energy phosphordiester bond and could provide a thermodynamic driving force for the reaction. In our most recent working hypothesis (Figure 1C),19 based mainly on the X-ray crystal structures of MoaC in complex with 3′,8-cH2GTP and cPMP, we proposed that cyclic phosphate formation is one of the last steps. In this model, MoaC catalysis is initiated by either aminal hydrolysis or C2′–C3′ bond cleavage to yield an as-yet structurally uncharacterized intermediate X followed by a rearrangement to yield N-hexosyl triaminopyrimidinone (6), which is then converted to cPMP by formation of the pterin and cyclic phosphate rings. While the mechanism is consistent with the X-ray crystal structures of MoaC, there is currently no enzymological evidence for the timing of the cyclic phosphate ring formation. As such, we set out to probe the mechanism of this transformation. Here, we report the results of our investigation using uncleavable substrate analogue 3′,8-cyclo-7,8-dihydro-guanosine [(α,β)-methyleno]triphosphate (3′,8-cH2GMP[CH2]PP, 7; Figure 1D). Our characterization revealed that the reaction yielded a species with an acid-labile triaminopyrimidinone base with no pterin structure that is either sufficiently electrophilic to form covalent modification of MoaC or spontaneously converted to a pterin molecule in the absence of MoaC. These characteristics of the observed reaction product are consistent with the uncleavable analogue of N-hexosyl triaminopyrimidinone (6). These observations also suggest that in MoaC catalysis the cyclic phosphate and pterin rings are formed in a concerted manner, as opposed to the previous notion, where cyclic phosphate formation takes place independent of pterin ring formation.14,16 These studies provide the first enzymological insights into the late stages of MoaC catalysis.

EXPERIMENTAL PROCEDURES

General Methods

Guanosine 5′-triphosphate (GTP), S-adenosyl-l-methionine (SAM), dithiothreitol (DTT), and sodium dithionite were purchased from Sigma-Aldrich. Guanosine 5′-[(α,β)-methyleno]triphosphate (GMP[CH2]PP) was purchased from Jena Biosciences. Nonlinear least-squares fitting of kinetic data was carried out using KaleidaGraph software (Synergy Software, Reading, PA). N-Terminally His6-tagged Staphylococcus aureus MoaA and E. coli MoaC were expressed and purified from E. coli as described previously.18,19 Anaerobic experiments were carried out in an UNIlab workstation glovebox (MBaun, Stratham, NH) maintained at 10 ± 2 °C with an O2 concentration < 0.1 ppm. 3′,8-cH2GTP was prepared as described previously.18 All MoaA and MoaC assays and HPLC sample preparations were performed at 25 °C under strict anaerobic conditions (O2 < 0.1 ppm). All HPLC experiments were performed on a Hitachi L-2130 pump equipped with an L-2455 diode array detector, an L-2485 fluorescence detector, an L-2200 autosampler, and an ODS Hypersil C18 column (Thermo Scientific) housed in an L-2300 column oven maintained at 40 °C.

Preparation of 3′,8-cH2GMP[CH2]PP

To prepare 3′,8-cH2GMP[CH2]PP, MoaA (65 μM) was incubated with SAM (1 mM), sodium dithionite (2 mM), and GMP[CH2]PP (0.3 mM) in assay buffer (50 mM Tris-HCl, pH 7.6, 2 mM MgCl2, 5 mM DTT, and 0.3 M NaCl) at 25 °C. After incubation for 120 min, MoaA was removed by filtration with a 10 kDa MWCO filter (Millipore). This reaction typically yields a solution with 80 μM 3′,8-cH2GMP[CH2]PP, which was used for subsequent MoaC assays without further purification. The amount of 3′,8-cH2GMP[CH2]PP was quantified based on its derivatization to dimethylpterin as described previously.18

Anaerobic HPLC Characterization of Products of the MoaC Assay with 3′,8-cH2GMP[CH2]PP as Substrate

MoaC assays were performed in the presence of 25 or 160 μM wt-MoaC or 160 μM MoaC variants and 52 μM 3′,8-cH2GMP[CH2]PP. In these assays, 3′,8-cH2GMP[CH2]PP was prepared as described above without further purification. The presence of residual SAM and GMP[CH2]PP did not affect the MoaC assay. After 120 min of incubation, the reactions were quenched by addition of a stoichiometric volume of anaerobic MeOH, and the protein precipitant was removed after pelleting by centrifugation at 13 000 rpm for 5 min. The supernatant was analyzed by ion-pairing HPLC under anaerobic conditions using anaerobically prepared solvents and buffers under a flow of Ar gas. An Alltech Apollo C18 column (4.6 × 250 mm, 2.7 μm, part no. 36511) was initially equilibrated in 0.1 M KH2PO4, pH 6.0, 8 mM tetrabutyl-ammonium hydrogen sulfate (Solvent A), to which 60 μL of the assay supernatant was applied. The elution was performed at a flow rate of 1.0 mL/min using solvents A and B (0.1 M KH2PO4, pH 6.0, 8 mM tetrabutylammonium hydrogen sulfate, 30% (v/v) ACN): 0% B for 6.5 min, 0–20% B for 6.5 min, 20–40% B for 13 min, and 40–100% B for 8 min. Chromatography was monitored by the L-2455 diode array detector.

IC50 Determination of MoaC Inhibition by 3′,8-cH2GMP[CH2]PP

wt-MoaC (0.2 μM) was incubated with 3′,8-cH2GTP (15 μM) and specified concentrations of 3′,8-cH2GMP[CH2]PP in assay buffer at 25 °C. The reaction was then quenched at 2, 4, 6, 8, and 10 min with 0.1 volume of 25% TCA. cPMP was quantified by HPLC after its conversion to compound Z.18

Characterization of 3′,8-cH2GMP[CH2]PP-Treated MoaC

wt-MoaC (10 μM) was incubated for 30 min at 25 °C in the presence or absence of 3′,8-cH2GMP[CH2]PP (19.3 μM). Subsequently, both samples were buffer-exchanged over a 10 kDa MWCO filter. This buffer exchange resulted in a 625-fold dilution of the small molecules in the protein fraction. MoaC (17 kDa) was found to be exclusively in the protein fraction based on SDS-PAGE. To test the activity, the resulting MoaC was diluted to 0.1 μM and incubated with 3′,8-cH2GTP (10.6 μM) in assay buffer at 25 °C. The reaction was quenched at 2, 4, 6, 8, and 10 min with 0.1 volume of 25% TCA, and cPMP was quantified by HPLC after its conversion to compound Z.18

To investigate for the presence of acid-labile triaminopyrimidinone, the retentate or filtrate after ultrafiltration of the solution from the MoaC assay with 3′,8-cH2GMP[CH2]PP was acidified with 0.1 volume of anaerobic 0.5 M HCl to quench the reaction. The resulting mixture was incubated at 95 °C for 5 min to facilitate hydrolysis, followed by an addition of 6 μL of 1 M NaOH to adjust the pH to 8.5. The resulting solution was combined with 25 μL of 0.66% (v/v) 2,3-butanedione solution in 0.9 M Tris-HCl, pH 8.5, and incubated at 95 °C for 45 min. After removal of precipitation by centrifugation, an aliquot (10 μL) of the supernatant was injected into the HPLC. The chromatography was performed by an isocratic elution with 92.5% 20 mM sodium acetate, pH 6.0, 7.5% MeOH at a flow rate of 1 mL/min and monitored by fluorescence (ex. 365 nm, em. 445 nm). Under these conditions, the void volume was 1.5 min and dimethylpterin was eluted at 5.6 min. For quantitation, a standard curve was generated using an authentic sample of trimethylpyrimidinone (Sigma) derivatized under the identical conditions.

To investigate for the presence of pterin compounds, the retentate or filtrate after the ultrafiltration of the solution from the MoaC assay with 3′,8-cH2GMP[CH2]PP was mixed with 0.1 volume of an anaerobic solution of I2 (1% w/v), KI (2% w/ v), and TCA (25% w/v) and incubated for 60 min at 22 °C. The resulting mixture was analyzed by ion-pairing HPLC using the conditions described above except that elution was performed by a linear gradient of 0–60% B over 20 min at 1.2 mL/min. Chromatography was monitored by an diode array detector and a fluorescence detector (ex. 365 nm, em. 445 nm). For LC-MS analysis, the HPLC peak at 4.7 min from the MoaC reaction with 3′,8-cH2GMP[CH2]PP was collected and batch purified with SP-Sepharose (ammonium form) to remove tetrabutylammonium ions. The resulting sample was lyophilized, resuspended in 20 mM ammonium formate, pH 8.0, and analyzed in negative ion mode using an Agilent 6224 LC-ESITOF-MS equipped with an Agilent Poroshell 120 C18 column (2.1 × 75 mm, 2.7 μm, part no. 697775-906). The elution was performed at a flow rate of 0.15 mL/min using 20 mM ammonium formate, pH 8.0, (solvent A) and MeOH (solvent B): 0–2% B for 10 min, 2–50% B for 7 min, and 50–100% B for 5 min. Data was processed using Agilent MassHunter software. The amount of 6-carboxypterin was quantified using an authentic standard (Sigma) and the reported molar extinction coefficient, ε344nm = 7410 M−1 cm−1 at pH 5.0.9

ESI-MS Analysis of MoaC

wt-MoaC (5 μM) was incubated for 120 min at 25 °C in the presence or absence of 3′,8-cH2GMPcPP (10 μM). Subsequently, samples were buffer-exchanged in 150 mM ammonium bicarbonate, pH 8.0, over a 10 kDa MWCO filter. This buffer exchange resulted in a 4096-fold dilution of the small molecules in the protein fraction. Then, the samples were diluted to 10 μM in 20% acetonitrile/30 mM ammonium bicarbonate, pH 8.0, and directly infused through a fused-silica capillary at 4 μL/min into a Waters Synapt G2 QTOF mass spectrometer operating in positive ionization mode through an electrospray interface. Source temperature was set to 80 °C, and the N2 nebulizing gas was set to 4 L/min. Approximately 30 full MS scans from m/z 50 to 2000 were collected for each condition. Spectra were summed and then deconvoluted with the Waters MaxEnt1 algorithm. Horse myoglobin was infused immediately prior to the test samples, and a static mass correction (−1.5 Da) was applied to all test samples based on the net difference from the expected myoglobin mass.

RESULTS AND DISCUSSION

3′,8-cH2GMP[CH2]PP has a methylene bridge in place of an oxygen between the α and β phosphate groups and therefore is expected to block cyclic phosphate formation. On the basis of our current working hypothesis (Figure 1C),19 the early steps of MoaC catalysis involve the rearrangement of the base and ribose moieties and one of the last steps is cyclic phosphate ring formation. Therefore, when it is incubated with MoaC, 3′,8-cH2GMP[CH2]PP is expected to be converted to an advanced reaction intermediate immediately prior to cyclic phosphate formation. Characterization of the trapped intermediate will provide insight into the timing of cyclic phosphate formation. To this end, we first prepared 3′,8-cH2GMP[CH2]PP from guanosine [(α,β)-methyleno]triphosphate (GMP[CH2]PP) using MoaA. Since the triphosphate moiety is not involved in MoaA catalysis, MoaA was expected to accept GMP[CH2]PP and convert it into 3′,8-cH2GMP[CH2]PP. Thus, GMP[CH2]-PP was incubated with MoaA under strict anaerobic conditions (O2 < 0.1 ppm), and the reaction products were analyzed anaerobically by ion-pairing HPLC. These analyses revealed a peak that migrated faster than 3′,8-cH2GTP (Figure 2A, compare traces 1 and 9) and exhibited a characteristic UV–vis absorption spectrum essentially identical to that of 3′,8-cH2GTP18 (Figure 2B). LC-MS analysis of this product was also consistent with the formation of 3′,8-cH2GMP[CH2]PP (Figure S1).

Figure 2.

Figure 2

Open in a new tab

Characterization of the products from assays of wt and variants of MoaC with 3′,8-cH2GMP[CH2]PP. (A) Anaerobic ion-pairing HPLC analysis. wt-MoaC (160 μM) or a MoaC variant (160 μM) was incubated with 3′,8-cH2GMP[CH2]PP (52 μM) and analyzed by HPLC with tetrabutylammonium hydrogen sulfate as an ion-pairing agent under a Ar atmosphere (traces 2–6). Chromatographs for 3′,8-cH2GMP[CH2]PP, cPMP, intermediate X, and 3′,8-cH2GTP under the same conditions are shown for comparison (traces 1 and 7–9, respectively). (B) UV–vis absorption spectra of 3′,8-cH2GMP[CH2]PP (solid black line) and [(α,β)-methyleno]-intermediate X (dotted black line) determined from diode array detection of HPLC traces 1 and 2 in (A), respectively. Also shown are the spectra for 3′,8-cH2GTP (solid gray line) and intermediate X (dashed gray line) for comparison.

The resulting 3′,8-cH2GMP[CH2]PP was then used to investigate its reaction with MoaC. Initially, the reaction of wt-MoaC with 3′,8-cH2GMP[CH2]PP was analyzed for the potential products by HPLC under anaerobic conditions. However, in such assays, no products were detectable (Figure 2A, trace 6), even though depletion of 3′,8-cH2GMP[CH2]PP was observed. The extent of 3′,8-cH2GMP[CH2]PP depletion was dependent on the ratio between the concentrations of 3′,8-cH2GMP[CH2]PP and wt-MoaC; complete depletion was observed when the amount of wt-MoaC (160 μM) was in excess of 3′,8-cH2GMP[CH2]PP (52 μM), whereas the depletion was minimal when wt-MoaC (25 μM) was substoichiometric to 3′,8-cH2GMP[CH2]PP (52 μM). These observations suggested that the product of the reaction between MoaC and 3′,8-cH2GMP[CH2]PP is tightly bound to MoaC and not readily released from the active site.

To obtain further insight into the reactivity of 3′,8-cH2GMP[CH2]PP with MoaC, we next tested whether 3′,8-cH2GMP[CH2]PP is converted to an analogue of a previously reported MoaC reaction intermediate. The conversion of 3′,8-cH2GTP to cPMP by wt-MoaC proceeds via intermediate X (Figure 1B). Because the formation of intermediate X precedes cyclic phosphate formation,19 we tested if 3′,8-cH2GMP[CH2]-PP is converted to an (α,β)-methyleno analogue of intermediate X. To this end, we used active-site variants of MoaC (K51A- and K131A-MoaC),19 which catalyze the transformation of 3′,8-cH2GTP to intermediate X but are incapable of converting intermediate X to the final product, cPMP. When 3′,8-cH2GMP[CH2]PP was incubated with K51A-MoaC or K131A-MoaC, 3′,8-cH2GMP[CH2]PP was quantitatively converted into a molecule that eluted earlier than 3′,8-cH2GMP[CH2]PP (Figure 2A, traces 2 and 5) and exhibited an absorption spectrum essentially identical to that of intermediate X (Figure 2B). Characterization of this compound by chemical derivatization was also consistent with the formation of [(α,β)-methyleno] intermediate X (Figure S2). No reaction was observed for the other variants, H77A-and D128A-MoaC (Figure 2A), which are incapable of forming intermediate X with the natural substrate, 3′,8-cH2GTP.19 These observations suggest that 3′,8-cH2GMP[CH2]PP serves as a substrate of MoaC and undergoes early steps of MoaC catalysis at least up until intermediate X.

The ability of 3′,8-cH2GMP[CH2]PP to serve as a MoaC substrate is also consistent with the X-ray crystal structure of MoaC.19 In the MoaC structure in complex with 3′,8-cH2GTP, the triphosphate moiety of 3′,8-cH2GTP is exposed to solvent and forms only one H-bond interaction between α-phosphate and H77. Therefore, it is reasonable for 3′,8-cH2GMP[CH2]PP to bind the MoaC active site in a manner identical to that of 3′,8-cH2GTP and undergo the early steps of the MoaC reaction that modify the base and sugar moieties.

On the basis of the above observation, the most reasonable interpretation of the results from the wt-MoaC reaction with 3′,8-cH2GMP[CH2]PP is that 3′,8-cH2GMP[CH2]PP is converted to an analogue of an advanced intermediate that binds to MoaC tightly without release. In this case, 3′,8-cH2GMP[CH2]PP should serve as a mechanism-based inhibitor of wt-MoaC. Thus, the inhibitory effects of 3′,8-cH2GMP[CH2]PP were investigated by performing MoaC assays using its natural substrate, 3′,8-cH2GTP, in the presence of varying concentrations of 3′,8-cH2GMP[CH2]PP. As a result, dose-dependent inhibition of MoaC activity was observed with an IC50 value of 8.6 ± 2.3 μM (Figure 3A). To test the reversibility of the inhibition, MoaC (10 μM) was preincubated with 3′,8-cH2GMP[CH2]PP (19.3 μM), followed by removal of small molecules by repeating the ultrafiltration. The resulting MoaC did not contain 3′,8-cH2GMP[CH2]PP above the detection limit (0.03 μM) and was still potently inhibited to an extent comparable to that prior to the removal of 3′,8-cH2GMP[CH2]PP (Figure 3B). These observations are consistent with the conclusion that 3′,8-cH2GMP[CH2]PP causes MoaC inhibition by a tight-binding or covalent mechanism.

Figure 3.

Figure 3

Open in a new tab

Characterization of wt-MoaC reaction with 3′,8-cH2GMP-[CH2]PP. (A) Dose-dependent inhibition of MoaC by 3′,8-cH2GMP-[CH2]PP. MoaC activity was determined by incubating MoaC (0.2 μM) with 3′,8-cH2GTP (15 μM) in the presence of varying concentrations of 3′,8-cH2GMP[CH2]PP. The solid line is a fit to the equation22 vi/vo = 1/((1 + 10log(I)–log(IC50)) × 1.6) with an IC50 value of 8.6 ± 2.3 μM. The activity of MoaC preincubated with GMP[CH2]PP, which should not have significant inhibitory effects on MoaC activity, is shown as a control. (B) Activity of 3′,8-cH2GMP[CH2]PP-treated MoaC after removal of unreacted 3′,8-cH2GMP[CH2]PP. MoaC (10 μM) was incubated with 3′,8-cH2GMP[CH2]PP (19.3 μM) or assay buffer, followed by the removal of unreacted 3′,8-cH2GMP[CH2]PP using a 10 kDa MWCO filter. The activity of the resulting MoaC was determined based on its ability to catalyze the conversion of 3′,8-cH2GTP to cPMP.

To obtain further insight into the chemical species causing the inhibition, the product of the MoaC reaction with 3′,8-cH2GMP[CH2]PP was characterized. Although we initially attempted to release the product from MoaC under various conditions, such as protein denaturing in acid or ethanol, no small molecule products were observed by anaerobic HPLC analysis. Attempts to detect the product using carbonyl-reactive reagents, such as O-(4-nitrobenzyl)hydroxylamine and O-(pentafluorobenzyl)hydroxylamine, did not result in detectable levels of small molecule derivatives. Therefore, we employed two chemical methods to detect characteristic structures expected in MoaC reaction intermediates. In the first method, pyranopterins, such as compound 5, are detected by oxidative conversion to pterins using KI/I2.2,8 The second method was to detect acid-labile triaminopyrimidinone base, which is expected to be present in precursors such as compound 6 (Figure 1C).18,20 In these analyses, 3′,8-cH2GMP[CH2]PP-treated wt-MoaC was first separated from small molecules using ultrafiltration (10 kDa MWCO). Then, the filtration retentate, the fraction containing macromolecules >10 kDa in size, was analyzed for the presence of pyranopterin or acid-labile triaminopyrimidinone by their conversion to fluorescence derivatives, pterin and dimethylpterin, respectively.18 The resulting derivatives were quantified by HPLC. In this analysis, dimethylpterin was observed in an amount stoichiometric to MoaC (1.2 ± 0.03 equiv per MoaC; Figure 4), whereas no significant amount of pterin compounds was detected. These results indicated that the inhibitory species contains an acid- labile triaminopyrimidinone but not a pyranopterin ring and that it bound MoaC tightly.

Figure 4.

Figure 4

Open in a new tab

Quantitation of triaminopyrimidinone- and pterin-containing species in the wt-MoaC assay with 3′,8-cH2GMP[CH2]PP. Chemical analysis of the wt-MoaC reaction with 3′,8-cH2GMP[CH2]-PP after separation of protein-bound and unbound molecules. MoaC assay with 3′,8-cH2GMP[CH2]PP was performed as in Figure 3B, and proteins >10 kDa were separated from small molecules by ultra-filtration. Each fraction was treated with TCA and butane-2,3-dione for detection of molecules with acid-labile triaminopyrimidinone (red bars) or with TCA and KI/I2 to detect pterin molecules (green bars).

The tight binding of the acid-labile triaminopyrimidinone-containing product (compound Y) was also demonstrated by MS analysis (Figure 5). In this experiment, MoaC was incubated with 3′,8-cH2GMP[CH2]PP as described above, thoroughly buffer exchanged using ultrafiltration into ammonium bicarbonate buffer (pH 8.0), and subjected to molecular weight determination by ESI-MS in 20% acetonitrile. The resulting MoaC retained 0.6 ± 0.1 equiv of compound Y based on the chemical derivatization assay. This sample was then subjected to ESI-QTOF MS analysis by direct infusion. The deconvoluted mass spectra revealed two clusters of mass signals (Figure 5). One corresponded to MoaC and its sodium and ammonium salts. The other corresponded to a species with a molecular weight 504 ± 1 Da greater than that of MoaC, which was absent in a control without 3′,8-cH2GMP[CH2]PP. The ratio between the intensities of the MS signals for unmodified vs modified MoaC (~1:2) was consistent with the stoichiometry of compound Y to MoaC (0.6 ± 0.1 equiv). Therefore, these observations suggest that the reaction of MoaC with 3′,8-cH2GMP[CH2]PP yields compound Y, which tightly binds MoaC, likely through a covalent bond.

Figure 5.

Figure 5

Open in a new tab

MS analysis of wt-MoaC with and without treatment with 3′,8-cH2GMP[CH2]PP. Deconvoluted MS spectra for wt-MoaC (A) and wt-MoaC treated with 3′,8-cH2GMP[CH2]PP (B) are shown. The 3′,8-cH2GMP[CH2]PP-treated sample was prepared as described in the caption to Figure 3 and buffer exchanged into 150 mM ammonium bicarbonate, pH 8.0, using ultrafiltration. The resulting sample was diluted into 30 mM ammonium bicarbonate, pH 8.0, 20% acetonitrile and directly injected to ESI-MS.

The ultrafiltration filtrate of the reaction, the fraction containing small molecules (<10 kDa), was also analyzed for the acid-labile triaminopyrimidinone and pterins. This analysis revealed the presence of only a small amount of acid-labile triaminopyrimidinone (<10% of MoaC; Figure 4), consistent with the tight binding of compound Y to MoaC. As for the pterin product, in contrast to the protein fraction, the small molecule fraction contained a detectable but low level of pterin species (Figure 4). This pterin species was absent in negative controls without MoaC or 3′,8-cH2GMP[CH2]PP (Figure 6A, traces 2 and 3), suggesting that it is derived from the product of the reaction of MoaC with 3′,8-cH2GMP[CH2]PP. Upon further UV–vis and ESI-MS analyses of this species and comparison with an authentic standard (Figures 6B and S3), the observed compound was identified as 6-carboxypterin (8; Figure 6C). However, quantitation of this species using an authentic standard revealed that the amount of 6-carboxypterin formed was small: 8.7 ± 3.2% of MoaC. This is significantly lower than the amount of compound Y bound to MoaC: 60–100% relative to MoaC. Therefore, together with the fact that 6-carboxypterin was the only pterin product detected, this observation suggests that in the reaction of wt-MoaC with 3′,8-cH2GMP[CH2]PP the pterin ring formation is significantly perturbed.

Figure 6.

Figure 6

Open in a new tab

Characterization of pterin species observed in the small molecule fraction (<10 kDa) of the wt-MoaC assay with 3′,8-cH2GMP[CH2]PP. (A) Ion-pairing HPLC analysis (ex. 365 nm, em. 445 nm) of the small molecule fraction of the MoaC assay with 3′,8-cH2GMP[CH2]PP after TCA/KI/I2 treatment. A control without 3′,8-cH2GMP[CH2]PP was performed using products of the MoaA reaction without GMP[CH2]PP so that it contains all other components (e.g., SAM and dithionite; see Experimental Procedures for detail). Compound Z, the oxidized derivative of cPMP, in traces 1 and 2 is derived from GTP co-purified with MoaA.18 (B) UV–vis spectra of compound Z and the 6-carboxypterin authentic sample observed in the MoaC assay (the peak at 7.2 min in trace 1 of (A)). (C) Structure of 6-carboxypterin.

On the basis of all of these observations, we propose the reaction model shown in Figure 7A, in which the product of the reaction of MoaC with 3′,8-cH2GMP[CH2]PP is [(α,β)-methyleno]N-hexosyl-triaminopyrimidinone (9; Figure 7A). The resulting 9 stalls in the active site, the majority of which results in the formation of a covalent complex with MoaC through a nucleophilic attack of an active site amino acid residue to the electrophilic α,β-unsaturated ketone in 9 (Figure 7B). This MoaC–9 conjugate is then dehydrated to give an adduct with a molecular weight of 503 Da, which is consistent with the observed mass increase of 504 ± 1 Da.

Figure 7.

Figure 7

Open in a new tab

Reaction of 3′,8-cH2GMP[CH2]PP with wt-MoaC. (A) Proposed mechanism of the reaction of 3′,8-cH2GMP[CH2]PP with wt-MoaC. 3′,8-cH2GMP[CH2]PP is converted to compound Y with acid-labile triaminopyrimidinone base, such as [(α,β)-methyleno]N-hexosyl triaminopyrimidinone (9). Accumulation of such a molecule in the active site could result in tight binding or covalent inhibition of MoaC (top path). The formation of a small amount of 6-carboxypterin is explained by the release of a minor amount of 9 to solution (bottom path; see Figure S4 for detail). (B) Possible mechanism of the covalent linking of 9 to MoaC. (C) Proposed mechanism of the MoaC-catalyzed formation of the tetracyclic structure of cPMP by a concerted cyclization of the pterin and cyclic phosphate rings.

The reaction yielded a small amount of pterin compound (6-carboxypterin; see Figure S4 for a possible mechanism of formation). However, considering that this amount was significantly lower than that of compound Y (9% vs >60%), the observation suggests that pterin ring formation is significantly perturbed. Therefore, these combined observations are most consistent with the concerted formation of the pterin and cyclic phosphate rings (Figure 7C). In this mechanism, pterin ring formation is coupled to the thermodynamically favored and likely irreversible cyclic phosphate formation. Therefore, this mechanism allows MoaC to use phosphodiester bond cleavage as a thermodynamic driving force to catalyze pterin ring formation and guide the complex rearrangement reaction to completion. Importantly, this mechanism is distinct from a previously accepted mechanism that was widely believed prior to the revision of MoaC's function, in which pterin ring formation takes place independently of and prior to cyclic phosphate formation.14−16 Therefore, the current observations also highlight the importance of the mechanistic coupling of the formation of the pterin and cyclic phosphate rings.

In conclusion, the data presented here provide the first insights into the late stages of the MoaC-catalyzed conversion of 3′,8-cH2GMP[CH2]PP to cPMP. Using 3′,8-cH2GMP-[CH2]PP as the mechanistic probe, our observations suggest that cyclic phosphate formation is the last step of catalysis coupled to pterin ring formation and that it likely provides the thermodynamic driving force for the formation of the characteristic tetracyclic structure of cPMP.

Supplementary Material

SI

ACKNOWLEDGMENTS

We would like to thank George R. Dubay (Duke University, Department of Chemistry) for assistance with the small molecule MS measurements. We thank Greg Waitt and Erik Soderblom in the Duke University School of Medicine Proteomics and Metabolomics Shared Resource for providing intact molecular weight determination of MoaC.

Funding

This work was supported by the Duke University Medical Center and the National Institute of General Medical Sciences (R01 GM112838 to K.Y.).

ABBREVIATIONS

Moco

molybdenum cofactor

GTP

guanosine 5′-triphosphate

3′,8-cH2GTP

3′,8-cyclo-7,8-dihydro-GTP

cPMP

cyclic pyranopterin monophosphate

GMP[CH2]PP

guanosine [(α,β)-methyleno]triphosphate

HPLC

high-pressure liquid chromatography

TCA

trichloroacetic acid

Footnotes

The authors declare no competing financial interest.

ASSOCIATED CONTENT

Supporting Information

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.5b00857.

Additional data and details of the characterization of 3′,8-cH2GMP[CH2]PP, [α,β-methyleno]intermediate X, and 6-carboxypterin(PDF)

REFERENCES

  • 1.Mendel RR, Schwarz G. Molybdenum cofactor biosynthesis in plants and humans. Coord. Chem. Rev. 2011;255:1145–1158. [Google Scholar]
  • 2.Leimkuhler S, Wuebbens MM, Rajagopalan KV. The History of the Discovery of the Molybdenum Cofactor and Novel Aspects of its Biosynthesis in Bacteria. Coord. Chem. Rev. 2011;255:1129–1144. doi: 10.1016/j.ccr.2010.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Reiss J, Gross-Hardt S, Christensen E, Schmidt P, Mendel RR, Schwarz G. A mutation in the gene for the neurotransmitter receptor-clustering protein gephyrin causes a novel form of molybdenum cofactor deficiency. Am. J. Hum. Genet. 2001;68:208–213. doi: 10.1086/316941. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Fritz C, Maass S, Kreft A, Bange FC. Dependence of Mycobacterium bovis BCG on anaerobic nitrate reductase for persistence is tissue specific. Infect. Immun. 2002;70:286–291. doi: 10.1128/IAI.70.1.286-291.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Weber I, Fritz C, Ruttkowski S, Kreft A, Bange FC. Anaerobic nitrate reductase (narGHJI) activity of Mycobacterium bovis BCG in vitro and its contribution to virulence in immunodeficient mice. Mol. Microbiol. 2000;35:1017–1025. doi: 10.1046/j.1365-2958.2000.01794.x. [DOI] [PubMed] [Google Scholar]
  • 6.Filiatrault MJ, Tombline G, Wagner VE, Van Alst N, Rumbaugh K, Sokol P, Schwingel J, Iglewski BH. Pseudomonas aeruginosa PA1006, which plays a role in molybdenum homeostasis, is required for nitrate utilization, biofilm formation, and virulence. PLoS One. 2013;8:e55594. doi: 10.1371/journal.pone.0055594. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Wang F, Sambandan D, Halder R, Wang J, Batt SM, Weinrick B, Ahmad I, Yang P, Zhang Y, Kim J, Hassani M, Huszar S, Trefzer C, Ma Z, Kaneko T, Mdluli KE, Franzblau S, Chatterjee AK, Johnsson K, Mikusova K, Besra GS, Futterer K, Robbins SH, Barnes SW, Walker JR, Jacobs WR, Jr., Schultz PG. Identification of a small molecule with activity against drug-resistant and persistent tuberculosis. Proc. Natl. Acad. Sci. U. S. A. 2013;110:E2510–E2517. doi: 10.1073/pnas.1309171110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wuebbens MM, Rajagopalan KV. Structural characterization of a molybdopterin precursor. J. Biol. Chem. 1993;268:13493–13498. [PubMed] [Google Scholar]
  • 9.Wuebbens MM, Rajagopalan KV. Investigation of the early steps of molybdopterin biosynthesis in Escherichia coli through the use of in vivo labeling studies. J. Biol. Chem. 1995;270:1082–1087. doi: 10.1074/jbc.270.3.1082. [DOI] [PubMed] [Google Scholar]
  • 10.Reiss J, Cohen N, Dorche C, Mandel H, Mendel RR, Stallmeyer B, Zabot MT, Dierks T. Mutations in a polycistronic nuclear gene associated with molybdenum cofactor deficiency. Nat. Genet. 1998;20:51–53. doi: 10.1038/1706. [DOI] [PubMed] [Google Scholar]
  • 11.Hänzelmann P, Schindelin H. Crystal structure of the S-adenosylmethionine-dependent enzyme MoaA and its implications for molybdenum cofactor deficiency in humans. Proc. Natl. Acad. Sci. U. S. A. 2004;101:12870–12875. doi: 10.1073/pnas.0404624101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Frey PA, Hegeman AD, Ruzicka FJ. The Radical SAM Superfamily. Crit. Rev. Biochem. Mol. Biol. 2008;43:63–88. doi: 10.1080/10409230701829169. [DOI] [PubMed] [Google Scholar]
  • 13.Mehta AP, Abdelwahed SH, Xu H, Begley TP. Molybdopterin Biosynthesis: Trapping of Intermediates for the MoaA-Catalyzed Reaction Using 2 ′-DeoxyGTP and 2 ′-ChloroGTP as Substrate Analogues. J. Am. Chem. Soc. 2014;136:10609–10614. doi: 10.1021/ja502663k. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Mehta AP, Hanes JW, Abdelwahed SH, Hilmey DG, Hänzelmann P, Begley TP. Catalysis of a New Ribose Carbon-Insertion Reaction by the Molybdenum Cofactor Biosynthetic Enzyme MoaA. Biochemistry. 2013;52:1134–1136. doi: 10.1021/bi3016026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Iobbi-Nivol C, Leimkuhler S. Molybdenum enzymes, their maturation and molybdenum cofactor biosynthesis in Escherichia coli. Biochim. Biophys. Acta, Bioenerg. 2013;1827:1086–1101. doi: 10.1016/j.bbabio.2012.11.007. [DOI] [PubMed] [Google Scholar]
  • 16.Kanaujia SP, Jeyakanthan J, Nakagawa N, Balasubramaniam S, Shinkai A, Kuramitsu S, Yokoyama S, Sekar K. Structures of apo and GTP-bound molybdenum cofactor biosynthesis protein MoaC from Thermus thermophilus HB8. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010;66:821–833. doi: 10.1107/S0907444910019074. [DOI] [PubMed] [Google Scholar]
  • 17.Hänzelmann P, Schindelin H. Binding of 5′-GTP to the C-terminal FeS cluster of the radical S-adenosylmethionine enzyme MoaA provides insights into its mechanism. Proc. Natl. Acad. Sci. U. S. A. 2006;103:6829–6834. doi: 10.1073/pnas.0510711103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Hover BM, Loksztejn A, Ribeiro AA, Yokoyama K. Identification of a cyclic nucleotide as a cryptic intermediate in molybdenum cofactor biosynthesis. J. Am. Chem. Soc. 2013;135:7019–7032. doi: 10.1021/ja401781t. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Hover BM, Tonthat NK, Schumacher MA, Yokoyama K. Mechanism of pyranopterin ring formation in molybdenum cofactor biosynthesis. Proc. Natl. Acad. Sci. U. S. A. 2015;112:6347–6352. doi: 10.1073/pnas.1500697112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Bracher A, Fischer M, Eisenreich W, Ritz H, Schramek N, Boyle P, Gentili P, Huber R, Nar H, Auerbach G, Bacher A. Histidine 179 mutants of GTP cyclohydrolase I catalyze the formation of 2-amino-5-formylamino-6-ribofuranosylamino-4(3H)-pyrimidinone triphosphate. J. Biol. Chem. 1999;274:16727–16735. doi: 10.1074/jbc.274.24.16727. [DOI] [PubMed] [Google Scholar]
  • 21.Santamaria-Araujo JA, Fischer B, Otte T, Nimtz M, Mendel RR, Wray V, Schwarz G. The tetrahydropyranopterin structure of the sulfur-free and metal-free molybdenum cofactor precursor. J. Biol. Chem. 2004;279:15994–15999. doi: 10.1074/jbc.M311815200. [DOI] [PubMed] [Google Scholar]
  • 22.Copeland RA. Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for Medicinal Chemists and Pharmacologists. John Wiley & Sons, Inc.; Hoboken, NJ.: 2013. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

SI
NIHMS776048-supplement-SI.pdf (583.7KB, pdf)

RESOURCES