First Biochemical Characterization of a Methylcitric Acid Cycle from

Article pubs.acs.org/biochemistry

First Biochemical Characterization of a Methylcitric Acid Cycle from Bacillus subtilis Strain 168 Jason J. Reddick,* Sherona Sirkisoon, Rejwi Acharya Dahal, Grant Hardesty, Natalie E. Hage, William T. Booth, Amy L. Quattlebaum, Suzette N. Mills, Victoria G. Meadows, Sydney L. H. Adams, Jennifer S. Doyle, and Brittany E. Kiel Department of Chemistry and Biochemistry, University of North Carolina at Greensboro, Greensboro, North Carolina 27402, United States S Supporting Information *

ABSTRACT: The genome of Bacillus subtilis strain 168 contains the mother cell metabolic gene (mmg) operon that encodes homologues from the methylcitric acid cycle. We showed that the three genes, mmgDE and yqiQ(mmgF), provide three of the five steps of the methylcitric acid cycle. We also showed that the fourth step can be supplied by citB (aconitase), and we suggest that the fifth missing step, the propionyl-CoA synthetase, is probably skipped because the β-oxidation of methylbranched fatty acids by the enzymes encoded by mmgABC should produce propionyl-CoA. We also noted interesting enzymology for MmgD and MmgE. First, MmgD is a bifunctional citrate synthase/2-methylcitrate synthase with 2.3-fold higher activity as a 2-methylcitrate synthase. This enzyme catalyzes the formation of either (2S,3R)- or (2R,3S)-2-methylcitrate, but reports of 2-methylcitrate synthases from other species indicated that they produced the (2S,3S) isomer. However, we showed that MmgD and PrpC (from Escherichia coli) in fact produce the same stereoisomer. Second, the MmgE enzyme is not a stereospecific 2-methylcitrate dehydratase because it can dehydrate at least two of the four diastereomers of 2-methylcitrate to yield either (E)-2-methylaconitate or (Z)-2-methylaconitate. We also showed for the first time that the E. coli homologue PrpD exhibited the same lack of stereospecificity. However, the physiological pathways proceed via (Z)-2-methylaconitate, which served as the substrate for the citB enzyme in the synthesis of 2-methylisocitrate. We completed our characterization of this pathway by showing that the 2-methylisocitrate produced by CitB is converted to pyruvate and succinate by the enzyme YqiQ(MmgF).

T

carbon pool regains the molar equivalent of oxaloacetate in the form of succinate. Bacillus subtilis is the important model of Gram-positive bacterial species, but unlike E. coli and Salmonella enterica, it cannot grow with propionate as the sole carbon source;6 in addition, there have not been any reports of individual methylcitric acid cycle activities characterized from this species. Nevertheless, the B. subtilis mother cell metabolic gene (mmg) operon7 (see the top of Scheme 1) includes the genes mmgDE and yqiQ, which are homologues of the methylcitric acid cycle prpCD and prpB genes, respectively, from E. coli and Salmonella typhimurium.a The mmg operon also contains the three genes mmgABC that likely participate in fatty acid degradation, perhaps in handling the 2-methylbutyryl-CoA that would result from the degradation of branched-chain fatty acids or branched amino acids (also shown in Scheme 1).8,9 The β-oxidation of 2methylbutyryl-CoA by the mmgABC enzymes would produce propionyl-CoA, and the mmg operon is equipped to handle this metabolite via the methylcitric acid cycle enzymes encoded by

he methylcitric acid cycle (Scheme 1) feeds propionate into the carbon metabolite pool by the net oxidation of propionate to pyruvate.1 This pathway allows Escherichia coli,1 Salmonella species,2 and other bacteria to grow with propionate as the only carbon source. Methylcitric acid also accumulates in humans with various metabolic disorders that contribute to the buildup of propionate, such as propionic acidemia3 and others that arise from nutritional deficiencies such as those associated with vitamin B12.4 The methylcitric acid cycle begins with a propionyl-CoA synthetase (EC 6.2.1.17, PrpE from E. coli and Salmonella sp.), which combines propionate with CoA in an ATP-dependent reaction. This step is followed by 2methylcitrate synthase (EC 2.3.3.5, PrpC from E. coli and Salmonella sp.), which catalyzes the condensation of propionylCoA with oxaloacetate to yield 2-methylcitrate and CoA, by chemistry analogous to that of citrate synthase from the citric acid cycle. Next, the product is dehydrated by a 2-methylcitrate dehydratase (EC 4.2.1.79, PrpD from E. coli and Salmonella sp.) to form (Z)-2-methylaconitate, which is then hydrated again, usually by a canonical aconitase (EC 4.2.1.3) to give 2methylisocitrate.1 Finally, the 2-methylisocitrate is cleaved by a 2-methylisocitrate lyase (EC 4.1.3.30, PrpB from E. coli and Salmonella sp.) to produce pyruvate and succinate.5 The overall pathway produces pyruvate from propionate, and the metabolic © XXXX American Chemical Society

Received: August 11, 2017 Revised: September 26, 2017 Published: September 28, 2017 A

DOI: 10.1021/acs.biochem.7b00778 Biochemistry XXXX, XXX, XXX−XXX

Article

Biochemistry

Scheme 1. mmg Operon from B. subtilis (top)a and the Methylcitric Acid Cycle, Including Different Proposed Sources of Propionyl-CoA in B. subtilis (bottom)

a

The black genes are homologues that encode the methylcitric acid cycle, the focus of this report. Experiments with the light gray genes, mmgABC, are not presented in this report but encode homologues for the short-chain fatty acyl β-oxidation pathway, which because of the prevalence of anteiso methyl-branched fatty acids in B. subtilis could lead to the production of propionyl-CoA. The flanking promoter and transcription terminator sections are not drawn to scale.

E. coli K-12 (MG1655) was obtained from the American Type Culture Collection (ATCC 700926). E. coli BL21(DE3) and the pET-28a plasmid were obtained from Novagen, Inc. E. coli strains TOP10 and BL21-Star were obtained from Invitrogen (Life Technologies, Inc., Grand Island, NY). Custom DNA oligonucleotides were obtained from Integrated DNA Technologies, Inc. (Coralville, IA). B. subtilis mutant strain AWS19815 was kindly provided by A. Sonenshein at Tufts University (Medford, MA). This mutant strain contains a genome-encoded citB that expresses the protein as an Nterminal His6 fusion. This strain was grown; the protein was purified by cobalt-affinity chromatography, and the enzyme was activated and assayed for activity all as described by Serio et al.15 The pET-200 plasmid for other constructs originated from the pET-200/D-TOPO Champion pET Directional TOPO Expression Kit (Life Technologies, Inc.) and was used in topoisomerase-based cloning procedures performed according to the manufacturer’s instructions. All transformations and propagation of plasmids were performed using TOP10 chemically competent E. coli (also from Life Technologies) or E. coli DH5α, which was obtained from New England Biolabs. E. coli strains were grown in standard Lysogeny Broth (LB), and BS168 was grown in well-aerated LB medium. Genomic DNA from BS168 cultures was purified using a Wizard Genomic DNA Purification Kit from Promega, according to the manufacturer’s instructions. All plasmid DNA was purified using the Qiaprep Spin Miniprep Kit (Qiagen, Inc.). Standard methods were used for restriction endonuclease digestions, ligation, and chemical (CaCl2) bacterial transformations.16

mmgDE and yqiQ, as shown in Scheme 1, which we describe in this report. The mmg operon has a promoter that places the operon under the control of RNA polymerase sigma factor σE, which directs the transcription of a subset of genes in the mother cell only during an intermediate stage of sporulation.10 Immediately following the σE promoter is the operator named mmgO (which matches 13 of 14 nucleotides of the amyO operator) that makes the operon subject to the CcpA carbon catabolite repression system.7,11 The combination of the σE promoter and mmgO operator ensures that the transcription of the mmg operon proceeds only in the mother cell during an intermediate stage of sporulation, and under glucose-deficient conditions. Even if the mmg gene products provide a functional methylcitric acid cycle, the regulation by σE and CcpA together ensures that the operon is never available to support the metabolism of propionate during the main vegetative life cycle of B. subtilis, and thus, mutants lacking these genes were never developed to explore propionate metabolism. Although mutants showed that the mmg operon was unessential to growth in regular laboratory media and were still able to undergo sporulation,7,12 these mmg pathways may be important during sporulation under natural environmental conditions such as soil and in biofilms.13,14 In this report, we show the first in vitro biochemical evidence that this mmg operon may contribute to a functional methylcitric acid cycle in B. subtilis.

■

EXPERIMENTAL PROCEDURES Culture Growth and DNA Preparations. B. subtilis strain 168 (trpC2) (BS168) was obtained from the Bacillus Genetic Stock Center at The Ohio State University (BGSC code 1A1). B


Article

Biochemistry Construction of Protein Expression Plasmids. PCR amplification of mmgD, mmgE, and yqiQ from BS168 genomic DNA17 and prpC and prpD from E. coli K-12 genomic DNA18 was accomplished using Phusion polymerase (New England Biolabs). The PCR primer pairs used to clone each gene are listed in Table S-1. Each PCR mixture was purified with the Qiagen PCR purification kit prior to the ligation step. In the case of mmgD and yqiQ, the PCR products were ligated into the NcoI and XhoI sites of pET-28a with T4 DNA ligase to yield either mmgD-pET-28a or yqiQ-pET-28a. In the case of mmgE, prpC, and prpD, the PCR product was immediately ligated into pET-200 with topoisomerase-based cloning, using the Champion Directional TOPO cloning kit from Invitrogen (now Life Technologies at Thermo Fisher Scientific), according to the manufacturer’s instructions, yielding mmgE-pET-200, prpC-pET-200, and prpD-pET-200. The pET-28a vectors were constructed such that the coding sequence of mmgD or yqiQ was placed in frame with a plasmid-encoded C-terminal His6 tag sequence. In the case of mmgE-pET-200, prpC-pET-200, and prpD-pET-200, each gene was placed in frame with a plasmid-encoded N-terminal His6 tag sequence. Assembly of the desired plasmids was verified by dye-terminated sequencing performed by SeqWright, Inc. (Houston, TX), or Eurofins Genomics (Louisville, KY). Expression and Purification of Proteins. BL21(DE3) transformed with mmgD-pET-28a or yqiQ-pET-28a was grown in LB containing kanamycin (30 μg/mL) and was shaken (220 rpm) at 37 °C. The mmgE-pET-200, prpC-pET-200, and prpDpET-200 plasmids were transformed and identically grown, except the expression strain used was BL21-Star from Invitrogen. All expression cultures were grown identically by the following procedure, except for yqiQ-pET-28a/BL21(DE3) and prpC-pET-200/BL21-Star where mentioned. When the culture reached an OD595 of 0.5−0.6, isopropyl β-D-1thiogalactopyranoside (IPTG) was added (final concentration of 1 mM), and the culture was shaken overnight at 37 °C. In the case of yqiQ expression, after an OD595 of 0.3−0.4 had been achieved, the culture was moved to an 18 °C incubator and shaking was continued. Once the OD595 reached 0.5−0.6, IPTG was added (final concentration of 1 mM) and the culture was shaken overnight at 220 rpm and 18 °C. In the case of prpC expression, the protein yield was poor with overnight expression; therefore, the culture was shaken at the same rate and 37 °C for 3 h after induction, and then the cells were harvested by centrifugation. In all cases, the cells were harvested by centrifugation (7500g for 30 min) and the pellet was immediately used or stored at −80 °C until cell lysis for protein purification was performed. All proteins were purified from their cell pellets using nickelnitrilotriacetic acid (Ni-NTA) affinity chromatography. The MmgE, YqiQ, PrpC, and PrpD proteins were purified at 4 °C using the following conditions. The bacterial pellet (from 500 mL of culture) was resuspended in 10 mL of binding buffer [5 mM imidazole, 500 mM NaCl, and 20 mM Tris-HCl (pH 7.9)] and sonicated for a total of 6 min (in 30 s increments) while on ice. The lysate was cleared by centrifugation (16000g for 30 min), passed through a 0.45 μm syringe filter, and applied to a Ni-NTA column (2 mL column bed) that was preequilibrated with binding buffer at 4 °C. After being loaded, the column was washed with 20 mL of binding buffer, then 12 mL of wash buffer [60 mM imidazole, 0.5 M NaCl, and 40 mM Tris-HCl (pH 7.9)], and then elution buffer with an increased imidazole concentration [200 mM imidazole, 0.25 M NaCl, and 10 mM

Tris-HCl (pH 7.9)]. Elution buffer fractions containing protein were combined and dialyzed at 4 °C, using Snakeskin dialysis tubing (from Thermo Fisher Scientific, 7000 molecular weight cutoff), in 4 L of 25 mM Tris-HCl (pH 8) buffer. The MmgD protein exhibited severe postcolumn insolubility under these conventional protein purification conditions. This precipitation was alleviated by handling all steps at room temperature and using binding, wash, and elution buffers identical to those described above except each also contained 10% glycerol. Fractions containing MmgD protein were combined and dialyzed overnight (at ambient room temperature) as described above but in 4 L of 25 mM Tris-HCl (pH 8) buffer containing 10% glycerol. All proteins were stored at −80 °C with 10% glycerol (which did not need to be added to the MmgD protein because the final dialysis buffer already contained the glycerol). The purity and size of all proteins were estimated by 12% Trisglycine sodium dodecyl sulfate−polyacrylamide gel electrophoresis (SDS−PAGE) according to the method of Laemmli.19 Citrate Synthase/2-Methylcitrate Synthase Activity Assay, Kinetics, and Analyses. To measure the citrate synthase activity20 of MmgD, reactions of acetyl-CoA (0.3 mM), 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB, 0.1 mM), in 200 mM Tris-HCl (pH 8.1), and MmgD protein (20 μg) were initiated by the addition of oxaloacetate (0.5 mM), and the absorbance at 412 nm was monitored using a Genesis 10 ultraviolet−visible (UV−vis) instrument. The 2-methylcitrate synthase activity of MmgD was measured identically, except propionyl-CoA was used instead of acetyl-CoA. Michaelis− Menten kinetics also utilized the UV assay, which was conducted by fixing either acyl-CoA or oxaloacetate at a saturating concentration while varying the concentration of the other substrate. To determine the kinetic constants for oxaloacetate, the concentration of acyl-CoA (either acetylCoA or propionyl-CoA) was held at 0.3 mM with varying concentrations of oxaloacetate (0.003−0.150 mM). In the other set of experiments, the concentration of oxaloacetate was held constant at 0.3 mM while the concentration of acyl-CoA was varied between 0.007 and 0.2 mM. The reactions catalyzed by MmgD were also characterized by analytical HPLC−UV and LC−MS using the methods described below in the analytical liquid chromatography sections. For HPLC or LC−MS analyses, the reaction components were the same as for the spectrometric assay described above, but without DTNB, and 1 mL samples were quenched with either 100 μL of 1 M sodium phosphate buffer (pH 2.9) or 0.1% formic acid. In the case of PrpC (E. coli) analysis, reactions of propionyl-CoA (10 mM) and oxaloacetate (20 mM) in 10 mM HEPES (pH 7.2) were initiated by the addition of PrpC protein (50 μg). The 1 mL mixture was incubated overnight at ambient temperature and then the reaction quenched with 100 μL of 1 M sodium phosphate buffer (pH 2.9). All quenched MmgD or PrpC samples were centrifuged at 15700g for 5 min before injection on the HPLC or LC−MS instrument. Peaks were correlated by retention time and m/z ratios with commercially available standards of citrate, 2-methylcitrate, and coenzyme A. Citrate Dehydratase and 2-Methylcitrate Dehydratase Activity Assay and Analyses. In the case of citrate dehydratase activity, a solution of citrate (1 mM) in 20 mM Tris-HCl (pH 7.5) was treated with MmgE enzyme, incubated at 37 °C overnight, and quenched and analyzed under the HPLC−UV conditions described below. In the case of the 2methylcitrate dehydratase activities of MmgE or PrpD, two sources of 2-methylcitrate were utilized. One source of 2C


Article

Biochemistry

3.6 units/mL lactic dehydrogenase in 20 mM HEPES buffer (pH 7.0). At different time intervals, 0.4 mL was removed from the enzyme reaction mixture and quenched by being heated at 90 °C for 5 min to denature the enzyme. Afterward, the sample was allowed to cool and was centrifuged to remove the precipitated protein. The supernatant was mixed with 500 μL of the LDH/NADH assay solution and incubated for 30 s, and the absorbance at 340 nm was measured. These absorbance values were correlated to pyruvate concentration using a linear calibration plot generated from standard solutions of pyruvate that were treated with the LDH/NADH assay in the same manner. The assay was always performed in triplicate and found to have a linear response between 0 and 0.1 mM pyruvate. All standards and all enzyme assays were measured in triplicate. The YqiQ activity was also analyzed via analytical chromatography by preparing 2-methylisocitrate in situ from oxaloacetate and propionyl-CoA using the combined activities of the MmgD, MmgE, and CitB enzymes, according to the following conditions. A complete 1 mL reaction mixture contained 0.4 mM propionyl-CoA, 0.2 mM oxaloacetate, 2 mM DTT, 2 mM MgCl2, 20 μg of MmgD, 20 μg of MmgE protein, and 20 mM Tris-HCl (pH 7.5). After 3 h at room temperature, 40 μg of activated CitB and 0.50 μg of YqiQ were added. The reaction mixture was incubated at room temperature overnight. Control reaction mixtures lacking individual components were prepared similarly and were also incubated at room temperature overnight. All reactions were quenched with 100 μL of 1 M sodium phosphate (pH 2.9) for HPLC analysis with UV detection or 100 μL of 0.1% formic acid for LC−MS analysis according to the method described below. Analytical Liquid Chromatography with High-Performance Liquid Chromatography and UV Detection (HPLC-UV). HPLC with UV detection was conducted on a Varian Prostar HPLC instrument with a 250 mm × 4.6 mm Synergi 4μ Hydro-RP 80A column (Phenomenex, Torrance, CA). The following chromatography solvents were used: solvent A, 20 mM sodium phosphate in water (pH 2.91); and solvent B, methanol. For reactions involving citrate (citrate synthase and citrate dehydratase/aconitase), chromatography was performed using 100% solvent A isocratically at a rate of 0.7 mL/min with detection at 220 nm. For the analysis of methylcitric acid cycle reactions, a gradient elution was employed as follows. After injection, the level of solvent B (methanol) was increased from 0 to 15% over 20 min. All of the products eluted in this time, but unreacted acyl-CoA substrates were retained on the column and eluted with a methanol flush involving a linear increase to 60% solvent B from 20 to 25 min, a plateau of 60% solvent B from 25 to 35 min, and then a linear decrease to 0% solvent B from 35 to 40 min, and then held for an additional 10 min at 100% solvent A until the next injection. Prior to injection of the samples onto the HPLC column, the reactions were quenched by the addition of 1 M sodium phosphate buffer (pH 2.9) (100 μL per 1 mL of sample). The addition of this phosphate buffer was critical to achieving a consistent separation and peak shape during HPLC elution. The identities of reaction mixture constituents were assigned using retention times of standards of CoA, acetyl-CoA, propionyl-CoA, oxaloacetate, citric acid, cisaconitate, trans-aconitate, succinate, pyruvate, and (2RS,3RS)-2methylcitric acid (a commercial mixture of diastereomers as described above). Unfortunately, standards of 2-methylaconitate or 2-methylisocitrate were not commercially available.

methylcitrate was that synthesized by the MmgD or PrpC enzyme in situ. These reaction mixtures were identical to the MmgD or PrpC mixtures described just above, except 20 μg of isolated MmgE (or 50 μg of the PrpD enzyme) was also included. The dehydratase reactions that included in situ 2methylcitrate were analyzed either within the same species enzyme pair, i.e., mixtures of MmgDE or PrpCD, or in crossspecies enzyme pairs, i.e., mixtures of MmgD with PrpD or PrpC with MmgE. A different source of 2-methylcitrate was the commercial (2RS,3RS) mixture of diastereomers of 2methylcitrate (obtained from Sigma-Aldrich). In these cases, reaction mixtures were comprised of 1 mM commercial (2RS,3RS)-2-methylcitrate and 20 μg of isolated MmgE protein (or 50 μg of the PrpD enzyme) in reaction buffer [20 mM Tris (pH 7.5)]. The change in absorbance at 240 nm due to the formation of the double bond in aconitate or 2-methylaconitate was measured on a Varian Cary spectrophotometer. The dehydratase reactions were also analyzed by HPLC−UV and LC−MS, according to the analytical chromatography sections below. For HPLC or LC−MS analyses, the reaction components were the same as for the spectrometric assay described above, and 1 mL samples were quenched with 100 μL of either 1 M sodium phosphate buffer (pH 2.9) or 0.1% formic acid. The samples were then centrifuged at 15700g for 5 min before injection on the HPLC instrument. Some of the peaks were correlated by retention time and m/z ratio with authentic standards of citrate, 2-methylcitrate, cis-aconitate, and trans-aconitate. Unfortunately, standards of (Z)- or (E)-2methylaconitate were not commercially available. 2-Methylaconitate Hydratase (CitB) Activity Analysis. The CitB protein was purified by Co2+-affinity chromatography from the AWS198 mutant strain as previously described.15 The Fe−S clusters of some aconitases, including CitB, are degraded by oxygen during purification, so a reducing activation buffer that also includes iron salts was used to reactivate the enzyme.21 Briefly, the enzyme was reactivated by first incubating two parts (by volume) of the purified CitB enzyme with three parts (by volume) of fresh reactivation buffer [8 mM dithiothreitol (DTT) and 0.8 mM Fe(NH4)2(SO4)2, all in 50 mM Tris-HCl (pH 8.0)] for 10 min at 25 °C before use. The success of this reactivation was confirmed by aconitase assays performed as previously described.15 To analyze the 2-methylaconitate hydratase activity, a 1 mL reaction mixture was prepared in 20 mM Tris-HCl (pH 7.5), 0.4 mM propionyl-CoA, 0.2 mM oxaloacetate, 20 μg of MmgD, and 50 μg of MmgE protein. After 3 h at room temperature, 40 μg of reactivated CitB protein was added and the mixture left at room temperature for 2 h. The reaction was quenched with 100 μL of 1 M sodium phosphate (pH 2.9) or 0.1% formic acid for HPLC analysis. Different control reactions lacking individual reaction components (either substrates or enzymes) were also performed. Standards of (Z)- or (E)-2-methylaconitate and 2-methylisocitrate were not commercially available. 2-Methylisocitrate Lyase (YqiQ) Activity Assay and Analyses. The 2-methylisocitrate lyase activity was analyzed by two methods. The first method involved titrating the amount of unreacted pyruvate in the reverse reaction between pyruvate and succinate, by using lactate dehydrogenase (LDH) and NADH.5 A 5 mL reaction mixture in 13 mM HEPES buffer (pH 7.0) containing 0.3 mM pyrvuate, 0.3 mM succinate, 0.2 mM magnesium chloride, and 0.1 mM dithiothreitol was initiated by the addition of 30 μg of YqiQ enzyme. A separate LDH/NADH assay solution consisted of 0.4 mM NADH and D


Article

Biochemistry

Figure 1. HPLC chromatograms showing the citric acid cycle activities of the MmgD and MmgE enzymes. All chromatograms utilized UV detection at 210 nm. For the sake of clarity, the absorbance y-axis was removed and only the elution time period between 6 and 17 min is shown. (A) Chromatogram of citrate. (B) Chromatogram of an MmgD reaction mixture lacking acetyl-CoA. (C) Chromatogram of a full MmgD reaction mixture containing acetyl-CoA and oxaloacetate. (D) Chromatogram of cis-aconitate. All commercial samples of cis-aconitate contained the transaconitate isomer at 14.0 min as an impurity, which was verified by analysis of a commercial sample of trans-aconitate (not shown). (E) Chromatogram of MmgE with citrate.

■

Analytical Liquid Chromatography with High-Performance Liquid Chromatography Coupled to Mass Spectrometry (LC−MS). Prior to LC−MS analysis, reactions of all 1 mL samples were quenched using 100 μL of 0.1% formic acid prior to injection. All liquid chromatography−mass spectrometry (LC−MS) analyses were performed on an Agilent 1100 Series high-performance liquid chromatograph coupled to a Thermo Fisher Scientific TSQ triple-quadrupole mass spectrometer with a heated electrospray (HESI) source. Each sample (5 μL) was injected onto the Phenomenex Synergi column described above for HPLC with UV detection. Analyses were conducted at a flow rate of 0.700 mL/min with the same gradient events as described in the preceding section, with the exception that solvent A was 0.1% formic acid. This substitution was made because the phosphate buffer used in the preceding section was incompatible with the MS detector. Analyses were conducted in both the positive and negative ion modes and with the stage 1 quadrupole, Q1. In positive ion mode, the TSQ instrument was operated over a scan range of m/z 50−1000 with the following settings: spray voltage, 3.80 kV; heated capillary offset, 35; heated capillary temperature, 350.0 °C; sheath gas flow rate, 45; auxiliary gas flow rate, 35; scan time, 0.50; peak width, 0.70. For negative ion mode analyses, the TSQ instrument was operated over a scan range of m/z 50− 1000 with the following settings: spray voltage, 3.00 kV; heated capillary offset, −35; heated capillary temperature, 380.0 °C; sheath gas flow rate, 45; auxiliary gas flow rate, 35; scan time, 0.50; peak width, 0.70.

RESULTS Genes mmgDE and yqiQ were cloned by PCR and ligated into plasmid vectors so that each protein could be separately expressed by E. coli with polyhistidine tags and purified by Ni2+affinity chromatography. The MmgD protein exhibited severe insolubility upon cell lysis after expression. This was improved by conducting all postlysis purification steps for MmgD at room temperature and by including 10% glycerol in all purification and dialysis buffers. We did not encounter any such problems with the remaining proteins described in this report, although the YqiQ protein was lost as inclusion bodies during overexpression at 37 °C and induction with 1 mM IPTG. Soluble YqiQ was obtained by inducing the culture at 18 °C followed by continued growth at that temperature overnight. This produced a yield that was lower than those of the other protein preparations (as evident in the difference in band intensities in Figure S-1), but nonetheless, the yield was adequate for this study. Finally, the CitB protein was purified from mutant B. subtilis strain AWS198 by Co2+-affinity chromatography and activated as previously described by Serio et al.15 SDS−PAGE indicated that each mmg protein was purified to acceptable homogeneity, as shown in Figure S-1. With purified proteins in hand, we proceeded with the analysis of their activities as described next. MmgD (citrate/2-methylcitrate synthase) Activity. B. subtilis was already known to have two citrate synthases, encoded by citZ and citA. CitZ is the “major” citrate synthase of B. subtilis,22 whereas CitA provided a minor amount of activity. CitZ and CitA are expressed at the end of exponential growth E


Article

Biochemistry

Figure 2. HPLC chromatograms from selected MmgD and MmgE reaction analyses. All chromatograms utilized UV detection at 210 nm. For the sake of clarity, the absorbance y-axis was removed and only the elution time period between 11 and 22 min is shown. (A) Chromatogram of commercial 2-methylcitrate, indicating the (2R,3S)/(2S,3R) set of diastereomers at 13.6 min and the (2R,3R)/(2S,3S) set at 16.2 min. (B) Chromatogram of the full reaction mixture of MmgD with propionyl-CoA and oxaloacetate showing one new product peak eluting at 13.6 min. (C) Chromatogram of an MmgE reaction mixture supplied with the commercial 2-methylcitrate mixture of diastereomers. Unreacted substrate peaks are seen at 13.6 and 16.2 min as well as two new product peaks at 16.5 and 19.6 min. (D) Chromatogram of a reaction mixture comprised of propionylCoA, oxaloacetate, and the two enzymes MmgD and MmgE. Unreacted 2-methylcitrate is visible at 13.6 min and one product peak at 16.5 min. Each of the reaction analyses was accompanied by various control reactions lacking individual components, and only the full reaction mixtures resulted in traces like those shown as chromatograms B−D.

steady-state kinetics with MmgD.20 The assay revealed that the enzyme accepted acetyl-CoA as a substrate, as expected given the sequence and mutant data cited above. The enzyme also showed activity when given propionyl-CoA instead of acetylCoA, thus confirming the expected dual role for this enzyme in both the citric acid and methylcitric acid cycles. The enzyme exhibited substrate inhibition by saturating oxaloacetate with both citrate and 2-methylcitrate synthase activities. Apparent Michaelis−Menten constants for the CoA−thioester substrates were therefore measured while holding oxaloacetate at a maximum uninhibiting concentration (300 μM). This likely translated to decreased kcat or Vmax values for all the substrates. Apparent specificity constants (kcat/KM) measured by this method showed that the enzyme favored propionyl-CoA (41 × 103 M−1 s−1) over acetyl-CoA (18 × 103 M−1 s−1) by a factor of 2.3. This factor is similar to relative activities reported for the 2methylcitrate synthase isolated from E. coli (PrpC). The Michaelis−Menten specificity constants for E. coli 2-methylcitrate synthase showed a 3-fold preference for 2-methylcitrate synthase over citrate synthase activity26 (and specific activities reported by others and not measured with Michaelis−Menten analyses showed an ∼2-fold preference27). The S. typhimurium LT2 PrpC enzyme was reported to favor propionyl-CoA by a factor of 30.28 In some cases (such as the S. typhimurium homologue), the PrpC/MmgD homologues are probably used strictly as 2-methylcitrate synthases, whereas MmgD is the only enzyme expressed in the sporulating B. subtilis mother cell with citrate synthase activity; its closer kinetic bifunctionality as a citrate and 2-methylcitrate synthase is likely important in the mother cell’s physiological milieu.

and at the beginning of the stationary phase and are both repressed by glucose and glutamate.23,24 Mutants that lacked citA showed no substantial changes, but one that lacked citZ resulted in partial glutamate auxotrophy. Mutants that lacked both genes had complete glutamate auxotrophy and a diminished ability to enter sporulation. The gene sequences of citA and citZ (and encoded amino acids) differ from that of mmgD, and to the best of our knowledge, no reports have found that CitZ and CitA also had 2-methylcitrate synthase activity. It would be interesting to investigate that question, but so far, there have been no reports of in vitro measurements of any B. subtilis methylcitric acid cycle activities. The B. subtilis gene mmgD complemented various citrate synthase knockout mutants of E. coli as well as the mutant strain of B. subtilis lacking both citA and citZ,23 which suggested that mmgD encoded physiologically significant citrate synthase activity.7 The amino acid sequence of the mmgD protein is 35− 64% similar to those of citrate and 2-methylcitrate synthases from a variety of prokaryotes (comparisons made with some non-Bacillus species listed as MmgD homologues on UniProt25). The level of sequence similarity of mmgD, the gene’s ability to complement citrate synthase knockout mutants in both E. coli and B. subtilis, and finally the gene’s location alongside the apparent methylcitric acid cycle homologues mmgE (a homologue of 2-methylcitrate dehydratase) and yqiQ (a homologue of 2-methylisocitrate lyase) all suggested that MmgD possessed both citrate synthase and 2-methylcitrate synthase activities. A coupled spectrophotometric assay that measured the release of free coenzyme A was used for initial activity tests and F


Article

Biochemistry

Figure 3. LC−MS chromatograms demonstrating MmgD, MmgE, CitB, and YqiQ activities. (A) Chromatogram of a commercial mixture of diastereomers of 2-methylcitrate. (B) Chromatogram showing the full reaction mixture of MmgD with propionyl-CoA and oxaloacetate. (C) Chromatogram of a reaction mixture comprised of propionyl-CoA, oxaloacetate, and the two enzymes MmgD and MmgE. (D) Chromatogram of a reaction mixture generated from the commercial mixture of diastereomers of 2-methylcitrate and the MmgE enzyme. (E) Chromatogram of a reaction mixture comprised of propionyl-CoA, oxaloacetate, and the enzymes MmgD, MmgE, and CitB. (F) Chromatogram of a reaction mixture comprised of propionyl-CoA, oxaloacetate, and the enzymes MmgD, MmgE, CitB, and YqiQ.

at 16.2 and 13.6 min. Under the different mobile phase conditions compatible with MS detection, these two peaks showed similar relative intensities by integration and eluted from the same column at 17.26 and 14.71 min, respectively, as shown in Figure 3A. MS detection was optimal in negative mode, and both peaks displayed m/z ratios of 205, which is consistent with 2-methylcitrate, [M − H]−. Because diastereomers can be separated by achiral chromatography, the two components observed in the mixture were most likely the two possible sets of enantiomers of 2-methylcitrate. In keeping with the stereoselectivity reported for the Reformatsky synthesis of 2-methylcitrate, we assigned the major UV peak at 16.2 min (or 17.4 min in Figure 3A) as the homochiral (2R,3R)/(2S,3S) set of enantiomers and the minor peak at 13.6 min (or 14.7 min in Figure 3A) as the heterochiral (2R,3S)/(2S,3R) set of enantiomers. The MmgD reaction product of propionyl-CoA and oxaloacetate eluted at retention times identical to those for the minor heterochiral (2R,3S)/(2S,3R) components of the standard mixture for both LC methods (Figures 2B and 3B). In addition, the new reaction product of MmgD that appeared at 14.7 min in Figure 3B had a mass consistent with 2methylcitrate (m/z 205 for [M − H]−). The results described above indicated that MmgD produced citrate from acetyl-CoA and oxaloacetate and was also able to produce either (2R,3S)-2-methylcitrate or (2S,3R)-2-methylcitrate from propionyl-CoA and oxaloacetate. We note, however, that the 2-methylcitrate synthase homologue PrpC from E. coli was reported to produce the (2S,3S) enantiomer of 2methylcitrate.1 The PrpC amino acid sequence is 34% identical and 56% similar to that of MmgD, and we were surprised that

Both activities of MmgD were also verified by HPLC with UV detection and LC−MS, which confirmed the formation of free CoA from acetyl-CoA or propionyl-CoA (not shown). HPLC also showed that a reaction mixture containing MmgD, acetyl-CoA, and oxaloacetate yielded a new product peak (Figure 1C) that co-eluted with authentic citric acid (Figure 1A). The product of the MmgD enzyme reaction of propionylCoA and oxaloacetate was likewise compared to a commercially available sample of 2-methylcitrate. This synthetic preparation of 2-methylcitrate was available from Sigma-Aldrich as a mixture of diastereomers. This mixture was synthesized using a Reformatsky method [personal communication with technical support at Sigma-Aldrich (see the Supporting Information)]. The Reformatsky method for 2-methylcitrate has been reported in a couple of papers, including the most recent one by Krawczyk and Martyniuk.29 In their report, Krawczyk and Martyniuk claimed the Reformatsky method produced the “homochiral” (2R,3R) and (2S,3S) enantiomers [which they labeled as (2RS,3RS)]. Krawczyk and Martyniuk also made the opposite “heterochiral” set of (2R,3S) and (2S,3R) enantiomers [which they labeled as (2RS,3SR)] using a different synthetic method. To support their assignments, they provided extensive NMR data for the homochiral and heterochiral sets of enantiomers. We compared NMR spectra taken of the commercial 2-methylcitrate to the published data (see the Supporting Information), which indicated that the commercial mixture of diastereomers is a 9:1 homochiral/heterochiral mixture, according to the assignments made by Krawczyk and Martyniuk. HPLC chromatograms (Figure 2A) of the commercial mixture also showed two peaks with a 9:1 ratio G


Article

Biochemistry

min (or the one at 17.45 min in the LC−MS chromatogram) always exhibited significant tailing. This was reproducible and characteristic of this new product, but we do not know why the other later-eluting product peak did not have similar tailing. We also used MmgD as an in situ source for the biologically relevant 2-methylcitrate stereoisomer, which we have shown is either (2S,3R)- or (2R,3S)-2-methylcitrate. A reaction mixture containing both MmgD and MmgE was supplied with propionyl-CoA and oxaloacetate and allowed to incubate. As shown in Figure 2D, this combination of MmgD and MmgE activities gave only one of the MmgE reaction product peaks described above, the one that eluted at 16.5 min via UV detection. LC−MS analysis (Figure 3C) likewise showed one new product peak at 17.45 min, having a peak at m/z 187 consistent with 2-methylaconitate. In all experiments, this combination of MmgD and MmgE also led to a very minor peak in the LC−MS spectrum at 16.9 min with a peak at m/z 205. This peak was not visible in the UV chromatograms (Figure 2D) of the same mixture, and it was also absent in LC− MS analysis of MmgD reaction mixtures lacking MmgE (Figure 3B). This minor peak seemed to indicate that MmgE had some residual ability to interconvert different diastereomers of 2methylcitrate. We do not know how this is possible if the enzyme catalyzes only one type of elimination mechanism. Because the later-eluting MmgE product peak (at 19.6 min in Figure 2C and 23.19 min in Figure 3D) did not appear when MmgD was the source of the 2-methylcitrate, this second MmgE product originated from turnover of the (2S,3S) or (2R,3R) enantiomer, neither of which is produced by MmgD. Finally, this experiment showed that the product appearing at 16.5 min in Figure 2D and 17.45 min in Figure 3C is derived from the turnover of (2S,3R)- or (2R,3S)-2-methylcitrate from MmgD, indicating the compound eluting at 16.5 min (in Figure 2D) is the physiologically relevant product of MmgE. We were unable to purify the MmgE reaction products in quantities suitable for NMR analysis, nor were authentic standards of 2-methylaconitate commercially available. In lieu of those compounds, some considerations provide reasonable assignments of the identity of these two products. First, 2methylaconitate has two possible configurations around the double bond (E or Z), and the observation in Figures 2C and 3D of two product peaks, each at m/z 187, suggested that the reaction with the commercial 2-methylcitrate diastereomers produced 2-methylaconitate as a mixture of its two (Z) or (E) isomers. Second, cis-aconitate eluted before trans-aconitate using our reversed phase chromatography conditions (Figure 1D). The methyl group on 2-methylaconitate should not change the relative polarities of the cis (Z) and trans (E) isomers as compared to those of the nonmethylated aconitate analogues, so we expect that (Z)-2-methylaconitate would elute with a retention time earlier than that of (E)-2-methylaconitate. This comparison suggests that the earlier-eluting product peak at 16.5 min (Figure 2C) is likely (Z)-2-methylaconitate and the product peak observed at 19.6 min is the (E) isomer. This conclusion is consistent with every 2-methylcitrate dehydratase reported in the literature, which were so far all shown to produce the (Z) isomer from their physiological 2-methylcitrate precursor.1 Our experiments also showed that the mmgE enzyme is able to generate (E)-2-methylaconitate from either (2R,3R)- or (2S,3S)-2-methylcitrate, and this is the first such observation of a 2-methylcitrate dehydratase (but see our results with the E. coli homologue PrpD described below). Neither (2R,3R)- nor (2S,3S)-2-methylcitrate is generated by

such closely related homologues would perform the same reaction with different stereochemical outcomes. This greatly confused our initial efforts, and we will address the conflicting stereochemistry in the section describing our comparisons with the E. coli enzymes below, as well as in the Supporting Information. MmgE (2-methylcitrate dehydratase) Activity. The MmgE enzyme was expressed and purified with no difficulty. Sequence similarity indicated that this protein is a 2methylcitrate dehydratase, the step that follows the one catalyzed by the MmgD enzyme described above. Our initial approach to analyzing the 2-methylcitrate dehydratase activity of the MmgE enzyme involved using the commercially available mixture of 2-methylcitrate diastereomers described above as the substrate. UV measurements of these reactions indicated a substrate- and enzyme-dependent increase in absorbance at 240 nm, which was consistent with the formation of the new double bond within 2-methylaconitate.30 However, we did not use this method to attempt Michaelis−Menten kinetics (as others have done with homologues from other species),1 because the reaction product mixture was not as simple as we previously supposed, as revealed by HPLC (below). MmgE 2-methylcitrate dehydratase reactions were also analyzed by HPLC. Analyses of reaction mixtures supplied with the commercial 2-methylcitrate as the substrate were complicated because the compound was a mixture of diastereomers. As stated, these diastereomers eluted as two peaks with a 9:1 ratio, suggesting the four possible diastereomers were in a 45:45:5:5 mixture. We have shown that the reaction product of MmgD correlated with one of the minor components of the commercial mixture of 2-methylcitrate, indicating that the enzyme synthesized either the (2R,3S) or the (2S,3R) isomer of 2-methylcitrate. Because the natural substrate for MmgE was only ∼5% of this commercial mixture, we expected to observe just a slight amount of new product in the chromatograms taken from mixtures of MmgE with the commercial 2-methylcitrate. To our surprise, HPLC with UV detection indicated that the MmgE enzyme generated two intense product peaks from the commercial mixture of 2-methylcitrate, one minor product eluting at 16.5 min and the second major product at 19.6 min (Figure 2C). The relative integrations of these product peaks were approximately 1:9, a ratio identical to that of the originating substrate stereoisomer peaks shown in Figure 2A. These compounds probably had intense signals in the UV chromatograms because of their conjugated double bonds, but the mass analysis of these two new products from the mmgE reaction mixtures proved to be less sensitive; a representative of the clearest results is shown in Figure 3D. Although the 2methylcitrate diastereomers (m/z 205 for [M − H]−) could be readily detected by MS, the new peaks had far lower relative intensities compared to those of the UV-detected chromatograms shown in Figure 2C. The new peaks in Figure 3D could be viewed by LC−MS only when filtering ion chromatogram data for the mass of 2-methylaconitate (m/z 187), the expected product of 2-methylcitrate dehydratase activity. This experiment demonstrated that MmgE can produce more than one product, each with the mass of 2-methylaconitate. We reasoned that these two products were likely (Z)- and (E)-2methylaconitate and could have arisen from an ability to dehydrate more than one of the stereoisomers in the commercial 2-methylcitrate mixture, as discussed below. In both types of chromatograms, the new UV product peak at 16.5 H


Article

Biochemistry

Figure 4. HPLC chromatograms with UV detection analyzing reactions involving 2-methylcitrate synthase and 2-methylcitrate dehydratase from E. coli or B. subtilis. For the sake of clarity, the y-axis is removed and the x-axis is shown between 10 and 20 min. (A) Chromatogram of the commercial mixture of diastereomers of 2-methylcitrate. (B) Chromatogram of the full reaction mixture of the E. coli prpC enzyme with propionyl-CoA and oxaloacetate. (C) Chromatogram of a reaction mixture comprised of propionyl-CoA, oxaloacetate, and the two enzymes PrpC and PrpD from E. coli. (D) Chromatogram of a reaction mixture generated from the commercial mixture of diastereomers of 2-methylcitrate from chromatogram A and the PrpD enzyme. (E) Chromatogram of a reaction mixture comprised of propionyl-CoA, oxaloacetate, and the PrpC enzyme from E. coli, and the MmgE enzyme from B. subtilis. (F) Chromatogram of a reaction mixture comprised of propionyl-CoA, oxaloacetate, and the MmgD enzyme from B. subtilis, and the PrpD enzyme from. E. coli. The retention times of the peaks shown in this figure are faster than the ones shown in Figure 2, probably because the column had aged, but what is important is that the general elution patterns are the same.

The dehydrations conducted by the MmgE protein (and its homologues elsewhere) are analogous to the first step performed by the conventional aconitases that also rehydrate the resulting aconitate to yield isocitrate. Although CitB is the only known aconitase in B. subtilis, we reasoned that the methyl group of 2-methylcitrate would fill a space within the MmgE active site that could just as easily be occupied by the C−H bond at the analogous position in citrate, though with fewer van der Waals contacts and less affinity. We therefore analyzed the MmgE enzyme for possible aconitase activity with citrate. As shown in Figure 1E, HPLC chromatograms of MmgE reaction mixtures with citrate clearly yield a new peak at 12.6 min, albeit with less apparent conversion compared to that of the reaction with 2-methylcitrate. This new product peak co-eluted with standard samples of cis-aconitate. Because aconitases ordinarily continue with a rehydration step to furnish isocitrate, we subjected MmgE to a variety of conditions in a search for the formation of isocitrate from citrate. All of these experiments failed to produce isocitrate. Extensive experiments under a variety of conditions also showed that the 2-methylaconitate products were not hydrated by MmgE to yield 2-methylisocitrate, which is consistent with experiments with other 2methylcitrate dehydratase homologues reported in the literature.1 Because 2-methylcitrate dehydratases like MmgE

the 2-methylcitrate synthase activity of MmgD, so we expect that the MmgE activity on these stereoisomers and the resulting (E)-2-methylaconitate are not physiologically relevant for B. subtilis. Our results here are the first to show a 2-methylcitrate dehydratase that does not have absolute selectivity with respect to the stereochemistry of the 2-methylcitrate substrate. The enzyme can produce either (Z)- or (E)-2-methylaconitate depending on the configuration of the substrate. On the other hand, earlier publications did not describe experiments that resolved stereoisomers in the manner we describe here, so it is possible that homologues of MmgE from other species may likewise lack absolute stereospecificity (as borne out below in a comparison with the E. coli homologue PrpD). However, reports describing the 2-methylcitrate dehydratase from E. coli (PrpD) (amino acid sequence 75% identical and 60% similar to that of MmgE) indicate that the (Z)-isomer is produced from (2S,3S)-2-methylcitrate, which would require an atypical syn elimination mechanism.1 Our observations with MmgD and MmgE thus suggest a disagreement in stereochemical assignments with other pairs of 2-methylcitrate synthases and 2methylcitrate dehydratases reported in the literature, which will together be addressed in the next section. I


Article

Biochemistry

shown in Figure 4, the 2-methylcitrate produced by PrpC matched the minor component of the commercial 2methylcitrate mixture (Figure 4A), which indicated that it was identical to the 2-methylcitrate produced by MmgD (compare Figures 4B and 2B). The PrpD 2-methylcitrate dehydratase turned over this PrpC reaction product to yield the same (Z)-2-methylaconitate as we have shown with the MmgDE pair (compare Figures 4C and 2D). Reactions between the PrpD enzyme and the commercial mixture of 2methylcitrate diastereomers exhibited the same mixture of (Z) and (E) products we observed with the MmgE enzyme (compare Figures 4D and 2C). This result showed that PrpD has the same lack of stereoselectvity for the substrate as MmgE, which is a new observation for the E. coli enzyme. We also performed cross-species enzyme analyses. Given the sequence similarity, we expected that PrpC and MmgD produced the same enantiomer of 2-methylcitrate, but because our HPLC methodology resolved diastereomers, but not enantiomers, we could not rule out the possibility that 2-methylcitrate synthases of B. subtilis and E. coli formed opposite enantiomers. If the two 2-methylcitrate synthases form the same enantiomer, then their reaction products would be compatible with the 2-methylcitrate dehydratase from both species. If PrpC and MmgD generated opposite enantiomers of 2-methylcitrate, then we thought it unlikely that each product would serve as a substrate for the 2methylcitrate dehydratase from the other species. We analyzed reaction mixtures involving mixtures of enzymes from the two species to help ascertain whether MmgD and PrpC generated the same enantiomer of 2-methylcitrate. PrpC/MmgE (Figure 4E) and MmgD/PrpD (Figure 4F) mixtures produced the same 2-methylaconitate product as MmgDE (Figure 2D) or PrpCD (Figure 4C), indicating that MmgD and PrpC probably generate the same 2-methylcitrate enantiomer and that MmgE and PrpD conduct their reactions with the same type of elimination mechanism leading to (Z)-2-methylaconitate. It is possible that the stereochemistry of the Reformatskysynthesized 2-methylcitrate we used here was incorrectly assigned by the commercial supplier (or by the authors who developed the synthesis in the cited work) or the stereochemical outcome of the E. coli 2-methylcitrate synthase PrpC was incorrectly assigned in reports by others in the literature. Reports in the literature either correlated the stereochemistry of 2-methylcitrate preparations with properties given in prior reports (as we are doing here) or provided stereochemical assignments without also providing the experimental data (see the Supporting Information).1 From our point of view, key experimental data that lead to the unambiguous stereochemistry assignments for enzyme-prepared 2-methylcitrate have not been published. We provide a more detailed analysis and discussion of the Reformatsky-synthesized 2-methylcitrate in the Supporting Information. CitB (aconitase/2-methylaconitate hydratase) Activity. The next step in the methylcitric acid cycle is the rehydration of 2-methylaconitate to produce 2-methylisocitrate. Because the only three obvious homologues of methylcitric acid cycle enzymes in the B. subtilis genome are clustered together in the mmg operon, we first addressed the possibility that any of these could have a second reactive capability to catalyze this rehydration step. As mentioned, we were unable to detect any 2-methylisocitrate formation in any MmgE reaction mixtures. We also sought a possible dual activity for YqiQ (2methylisocitrate lyase homologue), but MmgE assay mixtures

do not have a structurally homologous active site similar to that of canonical aconitases (as evidenced by the apparent lack of Fe−S clusters in 2-methylcitrate dehydratases),1,2 the active site probably lacks aconitase’s ability to “flip” or bind the 2methylaconitate in the two ways needed to remove the water from position 3 of 2-methylcitrate and place it at position 2 of 2-methylaconitate leading to 2-methylisocitrate. MmgE thus can handle only the first dehydration step of both the 2methylcitric and citric acid cycles. Any cis-aconitate formed physiologically by the MmgE enzyme is probably turned over by the well-established CitB aconitase enzyme (which can handle the complete isomerization pathway to produce isocitrate from citrate), and the 2-methylaconitate also requires hydration at position 2 if it is to proceed through the rest of the methylcitric acid cycle. Comparison of MmgDE to PrpCD from E. coli. The amino acid sequences of the MmgD and MmgE proteins are similar to those of the PrpC and PrpD proteins from E. coli, respectively. However, the reaction stereochemistries we observed with the B. subtilis MmgDE enzymes differed from reports of their E. coli homologues in two respects. First, the PrpC enzyme was reported to produce only the homochiral (2S,3S)-2-methylcitrate, whereas our observations suggested that MmgD produced one of the heterochiral stereoisomers, either (2S,3R)- or (2R,3S)-2-methylcitrate. We were surprised by this difference, but the level of sequence similarity (34% identical and 56% similar) between the two homologues was not so high that the difference was implausible. On the other hand, we report that the MmgE enzyme produced (Z)-2methylaconitate, which is the same product reported for the E. coli homologue PrpD. If the MmgE and PrpD enzymes both produce (Z)-2-methylaconitate but from different stereoisomers of 2-methylcitrate, then these homologues would need to perform their elimination chemistries by different mechanisms (anti or syn),31 which should require a different compliment of active site acids and bases. We were troubled by this difference because it seemed that the high degree of amino acid sequence similarity (60% identical and 75% similar) shared by MmgE and PrpD should indicate they use identical reaction mechanisms. Instead of comparing our B. subtilis results with published conclusions reached by others for homologues from E. coli or Salmonella sp. and using different methods, the goal of this section was to directly compare the stereochemical outcomes of the enzymes from two species using the methods we describe in this report. A published study of the PrpC enzyme from E. coli involved preparations of the wild-type protein, and not with the cloned and expressed enzyme.27 Studies of the E. coli PrpD gene involved the cloned and expressed product;1 in addition, studies of the closely related homologues of these two genes from S. enterica (the amino acid sequences of PrpC homologues from these two species are 99% similar to 96% identical, and PrpD homologues are likewise 98% similar to 94% identical) have both been performed with the heterologously expressed proteins.32 Although these proteins were already prepared by other workers using these varieties of methods, we performed our own cloning and expression of the prpC and prpD genes from E. coli K-12, using pET-200 expression vectors. The purification of the two encoded proteins using this expression system was straightforward and resulted in proteins that behaved the same as in the references cited above. The two E. coli homologues were analyzed in the same manner as the B. subtilis homologues as described above. As J


Article

Biochemistry

comment on this peak further in the next section regarding the YqiQ enzyme. YqiQ (2-methylisocitrate lyase) Activity. The final step of the methylcitric acid cycle is the cleavage of 2methylisocitrate to yield pyruvate and succinate. Because 2methylisocitrate is not commercially available and we have had difficulty reproducing published syntheses,2,34 we were left with assessing this activity by two methods. We first analyzed the expected chemistry by determining whether the enzyme is active in the metabolic reverse sense: the synthesis of 2methylisocitrate from the commercially available pyruvate and succinate. This activity was analyzed by using a coupled enzyme activity assay in which the depletion of pyruvate was measured. By this assay, the enzyme was shown to deplete only 10% of the pyruvate given before changes in the mixture halted. This low conversion is consistent with data reported in the literature for the same reverse reaction of 2-methylisocitrate lyases identified in other organisms.5 The equilibrium achieved by this enzyme is evidently governed by thermodynamics that favor pyruvate and succinate over 2-methylisocitrate. The second method used to analyze the activity of the YqiQ enzyme was dependent on the enzymatic synthesis of 2methylisocitrate by the combined action of MmgDE and CitB, starting with propionyl-CoA and oxaloacetate. As shown in Figure 4F, LC−MS chromatograms show that the full set of enzymes from this pathway left no detectable 2-methylcitrate, 2-methylaconitate, or 2-methylisocitrate. Even though the LC− MS instrument’s range was m/z 50−1000, the method utilized here failed to show the peak at m/z 87 expected from authentic standards of pyruvate. Instead, all authentic pyruvate standards prepared in the same way as the enzyme reaction samples showed a peak at 4.06 min with an ion at m/z 191, in negative mode. This mass is consistent with the [M − H]− ion of the Schiff base of pyruvate with the Tris reaction buffer. The LC− MS chromatogram of the full reaction mixture with MmgDE, CitB, and YqiQ also led to a new peak at 4.05 min, with an ion at m/z 191, suggesting that the YqiQ enzyme produced pyruvate that likewise condensed with the Tris buffer in time for the MS analyses. Finally, the analysis also showed a new product at 11.63 min, a peak that co-eluted with authentic succinate and showed a mass consistent with the mass of succinate (m/z 118 for [M − H]−). The mass analyses thus demonstrated that the YqiQ enzyme produced pyruvate and succinate from the 2-methylisocitrate that was provided by the combined action of MmgDE and CitB. As described in the prior sections describing the activities of the MmgE and CitB enzymes, we noted (in Figure 4C,E) the appearance of a peak at 17.34 min, with an ion at m/z 205. This minor peak at 17.34 min co-eluted with the (2R,3S)/(2S,3R)-2methylcitrate we had shown is not produced by the MmgD enzyme. As shown in Figure 4F, these 2-methylcitrate and 2methylisocitrate peaks were eliminated during incubations that included the YqiQ enzyme. We suggest that the MmgE and CitB enzymes might equilibrate various stereoisomers of 2methylcitrate until the YqiQ enzyme has removed the physiological stereoisomer of 2-methylisocitrate generated by CitB. The results presented here indicate that the YqiQ enzyme provided the 2-methylisocitrate lyase step for the B. subtilis methylcitric acid cycle encoded within the mmg operon. When the mmg operon was reported by Bryan and co-workers in 1996, their sequence data covered the promoter/operator area (mmgO), the complete sequences of mmgABCD, and only part

that included YqiQ failed to deplete the MmgE reaction products and also did not produce succinate or pyruvate. Because there was no apparent 2-methylaconitate hydratase activity encoded within the mmg operon, and methylcitric acid cycles in other organisms like E. coli recruit aconitase for this purpose,1,2 we reasoned that a logical candidate for this activity in B. subtilis would be CitB. The CitB protein in B. subtilis is well-studied, and it is the only known aconitase enzyme in B. subtilis.15 The CitB protein is an iron−sulfur cluster enzyme that catalyzes the aconitase activity of the citric acid cycle, and the enzyme also possesses an interesting RNA binding activity that plays a key role in transcription regulation at an early stage of sporulation. Furthermore, expression of the citB enzyme is induced at an early stage of sporulation, much at the same time that the mmg operon can be induced, and like mmg, it is also subject to carbon catabolite repression.33 The citB enzyme is thus available at the time that the mmg operon becomes active and could therefore provide the 2-methylaconitate hydratase activity that is missing from the mmg methylcitric acid cycle. The CitB enzyme was isolated from the AWS198 mutant of B. subtilis using the method of Serio et al.15 The enzyme’s essential iron−sulfur cluster decomposes under affinity purification conditions, so the active iron−sulfur cluster was regenerated after purification using the methods of Serio et al.15 Success in obtaining an active enzyme was confirmed by an aconitase activity assay with this enzyme. We showed this activity via UV measurements as well as HPLC analysis demonstrating that cis-aconitate is converted to citrate by the enzyme (data not shown). With the active CitB enzyme in hand, we next assessed the enzyme’s possible supporting role for the mmg methylcitric acid cycle. MmgD/MmgE reaction mixtures were incubated to produce the (Z)-2-methylaconitate product. The freshly activated CitB was then added and allowed to incubate further. HPLC with UV detection showed the appearance of a new product in these mixtures (data not shown), but LC−MS provided clearer chromatograms along with mass data, as shown in Figure 4E. These chromatograms showed a new peak at 11.67 min, eluting before both sets of diastereomers of 2methylcitrate, but with an identical mass (m/z 205 for [M − H]−). We observed that isocitrate and citrate eluted from this HPLC column in the same pattern, with isocitrate eluting before citrate. Given the observed m/z, and the comparison to relative retention times of citrate/isocitrate, it is reasonable to conclude that this new CitB-dependent peak at 11.7 min is 2methylisocitrate. We did not attempt to resolve the stereochemical outcome of this conversion, but the E. coli methylcitric acid cycle was reported to proceed via (2R,3S)-2-methylisocitrate.1,5,30 Our results are the first to show that the CitB enzyme possesses 2-methylaconitate hydratase activity, thus adding to its two known functions in B. subtilis metabolism, namely, the citric acid cycle and transcription regulation. In CitB reaction mixtures, we also observed a minor peak at 17.33 min with m/z 205, as shown in Figure 4E. This peak eluted at a retention time very close to that of the 2methylcitrate diastereomers that were not produced by the MmgD enzyme (compare to parts A and B of Figure 4). This peak was also present in MmgDE reaction mixtures (as described above in the MmgE section, but the retention time in that case was 16.9 min at m/z 205); however, its relative intensity was larger with CitB present, and we did not perform any experiments to ascertain whether the CitB enzyme also contributed to the accumulation of this minor peak. We will K


Article

Biochemistry of the mmgE gene.7 The B. subtilis genome published in 199717 provided the rest of the mmgE gene sequence and revealed that these genes are followed immediately by yqiQ, and then a transcription terminator, which would include yqiQ in the same transcript as mmgABCDE. The convention in the literature has been to name methylcitric cycle genes as prp genes (for propionate metabolism), but mmgDE and yqiQ are expressed from the wider mmg operon with genes that encode a different pathway [mmgABC are involved in fatty acid β-oxidation (see below)]. We therefore propose to simply rename the yqiQ gene as mmgF, in keeping with the original mmg designation given by the operon’s discoverers.7

a propionyl-CoA synthetase. The acetyl-CoA synthetase (AscA or YtsI) or long-chain acyl-CoA synthetases (YhfL and LcfA) encoded by the B. subtilis genome17,35 could potentially provide the propionyl-CoA synthetase activity. However, homologues of these two synthetases are ubiquitous across biology (including E. coli), and it is striking that E. coli and other species still have dedicated propionyl-CoA synthetase homologues whereas B. subtilis does not encode one in the mmg operon or elsewhere in the genome. This is a problem because the mmg methylcitric acid cycle described here cannot function without the propionyl-CoA substrate. The lack of an obvious propionyl-CoA synthetase and a general inability to grow on propionate suggest that the function of the methylcitric acid cycle in B. subtilis is to cope specifically with propionyl-CoA, and not propionate, making this pathway slightly distinct from the propionate-metabolizing pathways of other species. What then is the source of the propionyl-CoA? We suggest that the answer to this question comes from the rest of the mmg operon not described here. Instead of a prpE propionyl-CoA synthetase homologue, this operon has the three genes mmgABC. These three genes encode enzymes having β-oxidation activity on n-butyryl-length acyl-CoAs.36−38 We propose that the mmgABC enzymes can participate in the final round of β-oxidation steps of the anteiso methyl-branched fatty acids, a major constituent of B. subtilis fatty acids,8,9 or isoleucine,39 whose α-keto-decarboxylase40,41 was reported to be constitutively expressed in B. subtilis and was involved in the uptake of isoleucine, perhaps for the synthesis of anteiso fatty acids.42 The final round of the β-oxidation cycle involving these metabolites would yield not two acetyl-CoAs, but rather one acetyl-CoA and one propionyl-CoA (Scheme 1). The acetyl-CoA would enter the citric acid cycle via the citrate synthase activity of MmgD, but that enzyme’s 2-methylcitrate synthase activity can likewise condense propionyl-CoA with oxaloacetate to begin the 2-methylcitric acid cycle whose remaining steps are encoded by mmgEF and citB. Our new understanding of the mmg operon suggests that it is involved in the scavenging of branched fatty acids and branched-chain amino acids under glucose-deficient conditions during sporulation, or perhaps it participates in the recovery of any lipids released during the early maturation of the endospore. The mmgABC enzymes are likely involved in the late stages of fatty acyl or isoleucine catabolism, and the mmgDEF (and citB) enzymes we describe in detail here are ready to pick up any propionyl-CoA released from that process. The mmg enzymes thus represent a system that is ready to utilize every carbon that is released from the fatty acid or isoleucine catabolism during sporulation. So far, this operon was shown to be unessential for vegetative and sporulating B. subtilis cells, but these mmg pathways nevertheless highlight the potential efficiency of a sporulating cell that does not waste available carbon atoms recovered during its nutrient-deficient state.

■

DISCUSSION The biochemical data in this report demonstrate that there is a functional methylcitric acid cycle encoded within the genome of B. subtilis. We have shown that the enzymes encoded by mmgD, mmgE, citB, and mmgF(yqiQ) together turn over oxaloacetate and propionyl-CoA to yield pyruvate and succinate. Observed Stereochemistry. The apparent stereoselectivities of the MmgD and MmgE enzymes were interesting. First, our results showed that MmgD apparently produced either (2R,3S)- or (2S,3R)-2-methylcitrate, whereas homologues of this enzyme from other bacterial species, particularly E. coli and Salmonella sp., were all reported to produce (2S,3S)-2methylcitrate. Despite these conflicting reports, we showed that the PrpC enzyme from E. coli produces the same stereoisomer as MmgD from B. subtilis. Therefore, either the assignment we report here or the one reported in the literature must be incorrect. Second, the MmgE enzyme has relaxed substrate specificity that has not been reported for other 2methylcitrate dehydratases, being able to convert citrate to cisaconitate and dehydrate different stereoisomers of 2-methylcitrate leading to either (Z)- or (E)-2-methylaconitate. The MmgE and PrpD enzymes can accept substrates from either the heterochiral (2RS,3SR) or homochiral (2RS,3RS) sets of enantiomers of 2-methylcitrate, to produce (Z)- or (E)-2methylaconitate, respectively. We hypothesize that the active sites of MmgE and PrpD are permissive enough to bind 2methylcitrate in more than one orientation, as long as the outgoing α-proton and β-hydroxyl group can be oriented in an appropriate conformation complementary to any necessary active site residues.31 However, unlike conventional aconitases, the active site is evidently constrained enough so that a water molecule cannot be placed at position 2 forming 2methylisocitrate (or isocitrate). We predict that the MmgE and PrpD enzymes prefer a certain stereochemical configuration at just one of the two centers but do not distinguish the configuration on the other center. In all, the relaxed stereospecificity is an interesting property of the enzyme, and we will seek to understand the origins and potential synthetic utility of it in work that is outside the scope of this report. One Activity Is Missing from the B. subtilis Methylcitric Acid Cycle. The methylcitric acid cycle gene clusters in other organisms usually contain a propionyl-CoA synthetase, which is needed to start the pathway with propionate (Scheme 1). As one example, the prp operon from E. coli contains prpBCD (homologues of mmgF, mmgD, and mmgE, respectively) and also prpE, which encodes propionyl-CoA synthetase. The prp operon encodes a virtually complete system that allows E. coli to subsist on propionate. The mmg operon from B. subtilis explored in this report does not contain a homologue of

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.biochem.7b00778. PCR primers utilized for the cloning of mmg (B. subtilis) and prp (E. coli) genes used in this study, SDS−PAGE showing the purified MmgDEF(YqiQ) enzymes, and L


Article

Biochemistry

Involvement of methylcitrate dehydratase and aconitase. Eur. J. Biochem. 269, 6184−6194. (2) Horswill, A. R., and Escalante-Semerena, J. C. (2001) In vitro conversion of propionate to pyruvate by Salmonella enterica enzymes: 2-Methylcitrate dehydratase (PrpD) and aconitase enzymes catalyze the conversion of 2-methylcitrate to 2-methylisocitrate. Biochemistry 40, 4703−4713. (3) Ando, T., Rasmussen, K., Wright, J. M., and Nyhan, W. L. (1972) Isolation and Identification of Methylcitrate, a Major Metabolic Product of Propionate in Patients with Propionic Acidemia. J. Biol. Chem. 247, 2200−2204. (4) Allen, R. H., Stabler, S. P., Savage, D. G., and Lindenbaum, J. (1993) Elevation of 2-methylcitric acid I and II levels in serum, urine, and cerebrospinal fluid of patients with cobalamin deficiency. Metab., Clin. Exp. 42, 978−988. (5) Brock, M., Darley, D., Textor, S., and Buckel, W. (2001) 2Methylisocitrate lyases from the bacterium Escherichia coli and the filamentous fungus Aspergillus nidulans: Characterization and comparison of both enzymes. Eur. J. Biochem. 268, 3577−3586. (6) Sheu, C. W., Salomon, D., Simmons, J. L., Sreevalsan, T., and Freese, E. (1975) Inhibitory effects of lipophilic acids and related compounds on bacteria and mammalian cells. Antimicrob. Agents Chemother. 7, 349−363. (7) Bryan, E. M., Beall, B. W., and Moran, C. P., Jr. (1996) A σEdependent operon subject to catabolite repression during sporulation in Bacillus subtilis. J. Bacteriol. 178, 4778−4786. (8) Kaneda, T. (1967) Fatty Acids in the Genus Bacillus I. Iso- and Anteiso-Fatty Acids as Characteristic Constituents of Lipids in 10 Species. J. Bacteriol. 93, 894−903. (9) Klein, W., Weber, M. H. W., and Marahiel, M. A. (1999) Cold shock response of Bacillus subtilis: isoleucine-dependent switch in the fatty acid branching pattern for membrane adaptation to low temperatures. J. Bacteriol. 181, 5341−5349. (10) Steil, L., Serrano, M., Henriques, A. O., and Völker, U. (2005) Genome-wide analysis of temporally regulated and compartmentspecific gene expression in sporulating cells of Bacillus subtilis. Microbiology 151, 399−420. (11) Koburger, T., Weibezahn, J., Bernhardt, J., Homuth, G., and Hecker, M. (2005) Genome-wide mRNA profiling in glucose starved Bacillus subtilis cells. Mol. Genet. Genomics 274, 1−12. (12) Kobayashi, K., Ehrlich, S. D., Albertini, A., Amati, G., Andersen, K. K., Arnaud, M., Asai, K., Ashikaga, S., Aymerich, S., Bessieres, P., Boland, F., Brignell, S. C., Bron, S., Bunai, K., Chapuis, J., Christiansen, L. C., Danchin, A., Debarbouille, M., Dervyn, E., Deuerling, E., Devine, K., Devine, S. K., Dreesen, O., Errington, J., Fillinger, S., Foster, S. J., Fujita, Y., Galizzi, A., Gardan, R., Eschevins, C., Fukushima, T., Haga, K., Harwood, C. R., Hecker, M., Hosoya, D., Hullo, M. F., Kakeshita, H., Karamata, D., Kasahara, Y., Kawamura, F., Koga, K., Koski, P., Kuwana, R., Imamura, D., Ishimaru, M., Ishikawa, S., Ishio, I., Le Coq, D., Masson, A., Mauel, C., Meima, R., Mellado, R. P., Moir, A., Moriya, S., Nagakawa, E., Nanamiya, H., Nakai, S., Nygaard, P., Ogura, M., Ohanan, T., O’Reilly, M., O’Rourke, M., Pragai, Z., Pooley, H. M., Rapoport, G., Rawlins, J. P., Rivas, L. A., Rivolta, C., Sadaie, A., Sadaie, Y., Sarvas, M., Sato, T., Saxild, H. H., Scanlan, E., Schumann, W., Seegers, J. F. M. L., Sekiguchi, J., Sekowska, A., Seror, S. J., Simon, M., Stragier, P., Studer, R., Takamatsu, H., Tanaka, T., Takeuchi, M., Thomaides, H. B., Vagner, V., van Dijl, J. M., Watabe, K., Wipat, A., Yamamoto, H., Yamamoto, M., Yamamoto, Y., Yamane, K., Yata, K., Yoshida, K., Yoshikawa, H., Zuber, U., and Ogasawara, N. (2003) Essential Bacillus subtilis genes. Proc. Natl. Acad. Sci. U. S. A. 100, 4678−4683. (13) Lindsay, D., Brözel, V. S., and von Holy, A. (2006) Biofilmspore response in Bacillus cereus and Bacillus subtilis during nutrient limitation. J. Food Prot. 69, 1168−1172. (14) Earl, A. M., Losick, R., and Kolter, R. (2008) Ecology and genomics of Bacillus subtilis. Trends Microbiol. 16, 269−275. (15) Serio, A. W., Pechter, K. B., and Sonenshein, A. L. (2006) Bacillus subtilis aconitase is required for efficient late-sporulation gene expression. J. Bacteriol. 188, 6396−6405.

NMR analysis of the commercial 2-methylcitrate used in this study (PDF) Accession Codes

MmgD (citrate/2-methylcitrate synthase), UniProt entry P45858; MmgE (citrate/2-methylcitrate dehydratase), UniProt entry P45859; CitB (aconitase), UniProt entry P09339; YqiQ (MmgF) (2-methylisocitrate lyase), UniProt entry P54528; PrpC (citrate/2-methylcitrate synthase), UniProt entry P31660; PrpD (citrate/2-methylcitrate dehydratase), UniProt entry P77243. The experimental functions including new stereochemistry results will be submitted to UniProt after publication.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: 336-334-3941. ORCID

Jason J. Reddick: 0000-0001-8703-0795 Funding

The authors acknowledge generous support from the National Science Foundation (Award Number 0817793). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Professor Abraham Sonenshein (Tufts University) for the kind gift of the AWS198 mutant strain of B. subtilis. The authors also thank Dr. Brandie Ehrmann and Dr. Daniel Todd of the Triad Mass Spectrometry Facility for their aid with the various mass spectrometers and associated chromatography equipment and software. The authors thank Professor Sherri McFarland and Professor Kimberly Petersen for their assistance with the manuscript. Finally, the authors thank Professor Nicholas Oberlies and his colleague Dr. Jose Alberto Rivera-Chavez for their helpful comments regarding the stereochemistry of 2-methylcitrate.

■

ABBREVIATIONS CoA, coenzyme A; PCR, polymerase chain reaction; LB, Lysogeny Broth; HPLC, high-performance liquid chromatography; LC−MS, liquid chromatography−mass spectrometry; mmg, mother cell metabolic gene; HEPES, 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid; Tris, tris(hydroxymethyl)aminomethane.

■

ADDITIONAL NOTE In this report, we followed the accepted (but often misunderstood) formatting convention using names derived from bacterial gene codes. As an example, the italicized form mmgD (first letter lowercase last letter uppercase) indicates a gene (gene D from the mmg operon), whereas the unitalicized form MmgD (first letter uppercase) represents the protein encoded by the mmgD gene. It is often efficient to describe multiple, successive genes or proteins in the manner of mmgDEF (meaning the three genes mmgD, mmgE, and mmgF) or MmgDEF (meaning the proteins encoded by the three genes mmgD, mmgE, and mmgF). a

■

REFERENCES

(1) Brock, M., Maerker, C., Schütz, A., Völker, U., and Buckel, W. (2002) Oxidation of propionate to pyruvate in Escherichia coli: M


Article

Biochemistry (16) Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, Plainview, NY. (17) Kunst, F., Ogasawara, N., Moszer, I., Albertini, A. M., Alloni, G., Azevedo, V., Bertero, M. G., Bessieres, P., Bolotin, A., Borchert, S., Borriss, R., Boursier, L., Brans, A., Braun, M., Brignell, S. C., Bron, S., Brouillet, S., Bruschi, C. V., Caldwell, B., Capuano, V., Carter, N. M., Choi, S. K., Codani, J. J., Connerton, I. F., Cummings, N. J., Daniel, R. A., Denizot, F., Devine, K. M., Dusterhoft, A., Ehrlich, S. D., Emmerson, P. T., Entian, K. D., Errington, J., Fabret, C., Ferrari, E., Foulger, D., Fritz, C., Fujita, M., Fujita, Y., Fuma, S., Galizzi, A., Galleron, N., Ghim, S. Y., Glaser, P., Goffeau, A., Golightly, E. J., Grandi, G., Guiseppi, G., Guy, B. J., Haga, K., Haiech, J., Harwood, C. R., Henaut, A., Hilbert, H., Holsappel, S., Hosono, S., Hullo, M. F., Itaya, M., Jones, L., Joris, B., Karamata, D., Kasahara, Y., KlaerrBlanchard, M., Klein, C., Kobayashi, Y., Koetter, P., Koningstein, G., Krogh, S., Kumano, M., Kurita, K., Lapidus, A., Lardinois, S., Lauber, J., Lazarevic, V., Lee, S. M., Levine, A., Liu, H., Masuda, S., Mauel, C., Medigue, C., Medina, N., Mellado, R. P., Mizuno, M., Moestl, D., Nakai, S., Noback, M., Noone, D., O’Reilly, M., Ogawa, K., Ogiwara, A., Oudega, B., Park, S. H., Parro, V., Pohl, T. M., Portetelle, D., Porwollik, S., Prescott, A. M., Presecan, E., Pujic, P., Purnelle, B., Rapoport, G., Rey, M., Reynolds, S., Rieger, M., Rivolta, C., Rocha, E., Roche, B., Rose, M., Sadaie, Y., Sato, T., Scanlan, E., Schleich, S., Schroeter, R., Scoffone, F., Sekiguchi, J., Sekowska, A., Seror, S. J., Serror, P., Shin, B. S., Soldo, B., Sorokin, A., Tacconi, E., Takagi, T., Takahashi, H., Takemaru, K., Takeuchi, M., Tamakoshi, A., Tanaka, T., Terpstra, P., Tognoni, A., Tosato, V., Uchiyama, S., Vandenbol, M., Vannier, F., Vassarotti, A., Viari, A., Wambutt, R., Wedler, H., Weitzenegger, T., Winters, P., Wipat, A., Yamamoto, H., Yamane, K., Yasumoto, K., Yata, K., Yoshida, K., Yoshikawa, H. F., Zumstein, E., Yoshikawa, H., Danchin, A., and Wedler, E. (1997) The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature 390, 249−256. (18) Blattner, F. R., Plunkett, G., III, Bloch, C. A., Perna, N. T., Burland, V., Riley, M., Collado-Vides, J., Glasner, J. D., Rode, C. K., Mayhew, G. F., Gregor, J., Davis, N. W., Kirkpatrick, H. A., Goeden, M. A., Rose, D. J., Mau, B., and Shao, Y. (1997) Complete genome sequence of Escherichia coli K-12. Science 277, 1453−1462. (19) Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680−685. (20) Brock, M., Fischer, R., Linder, D., and Buckel, W. (2000) Methylcitrate synthase from Aspergillus nidulans: implications for propionate as an antifungal agent. Mol. Microbiol. 35, 961−973. (21) Dingman, D. W., and Sonenshein, A. L. (1987) Purification of aconitase from Bacillus subtilis and correlation of its N-terminal amino acid sequence with the sequence of the citB gene. J. Bacteriol. 169, 3062−3067. (22) Jin, S., and Sonenshein, A. L. (1996) Characterization of the major citrate synthase of Bacillus subtilis. J. Bacteriol. 178, 3658−3660. (23) Jin, S., and Sonenshein, A. L. (1994) Identification of two distinct Bacillus subtilis citrate synthase genes. J. Bacteriol. 176, 4669− 4679. (24) Jin, S., and Sonenshein, A. L. (1994) Transcriptional regulation of Bacillus subtilis citrate synthase genes. J. Bacteriol. 176, 4680−4690. (25) The UniProt Consortium (2017) UniProt: the universal protein knowledgebase. Nucleic Acids Res. 45, D158−D169. (26) Gerike, U., Hough, D. W., Russell, N. J., Dyall-Smith, M. L., and Danson, M. J. (1998) Citrate synthase and 2-methylcitrate synthase: structural, functional and evolutionary relationships. Microbiology 144, 929−935. (27) Textor, S., Wendisch, V. F., De Graaf, A. A., Müller, U., Linder, M. I., Linder, D., and Buckel, W. (1997) Propionate oxidation in Escherichia coli: evidence for operation of a methylcitrate cycle in bacteria. Arch. Microbiol. 168, 428−436. (28) Horswill, A. R., and Escalante-Semerena, J. C. (1999) Salmonella typhimurium LT2 catabolizes propionate via the 2-methylcitric acid cycle. J. Bacteriol. 181, 5615−5623.

(29) Krawczyk, H., and Martyniuk, T. (2007) Characterisation of the H and 13C NMR spectra of methylcitric acid. Spectrochim. Acta, Part A 67, 298−305. (30) Blank, L., Green, J., and Guest, J. R. (2002) AcnC of Escherichia coli is a 2-methylcitrate dehydratase (PrpD) that can use citrate and isocitrate as substrates. Microbiology 148, 133−146. (31) Babbitt, P. C., Hasson, M. S., Wedekind, J. E., Palmer, D. R. J., Barrett, W. C., Reed, G. H., Rayment, I., Ringe, D., Kenyon, G. L., and Gerlt, J. A. (1996) The enolase superfamily: a general strategy for enzyme-catalyzed abstraction of the α-protons of carboxylic acids. Biochemistry 35, 16489−16501. (32) Horswill, A. R., and Escalante-Semerena, J. C. (2001) In vitro conversion of propionate to pyruvate by Salmonella enterica enzymes: 2-methylcitrate dehydratase (PrpD) and aconitase enzymes catalyze the conversion of 2-methylcitrate to 2-methylisocitrate. Biochemistry 40, 4703−4713. (33) Jourlin-Castelli, C., Mani, N., Nakano, M. M., and Sonenshein, A. L. (2000) CcpC, a novel regulator of the LysR family required for glucose repression of the citB gene in Bacillus subtilis. J. Mol. Biol. 295, 865−878. (34) Muñoz-Elias, E. J., Upton, A. M., Cherian, J., and McKinney, J. D. (2006) Role of the methylcitrate cycle in Mycobacterium tuberculosis metabolism, intracellular growth, and virulence. Mol. Microbiol. 60, 1109−1122. (35) Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D. J. (1997) Gapped BLAST and PSIBLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389−3402. (36) Reddick, J. J., and Williams, J. K. (2008) The mmgA gene from Bacillus subtilis encodes a degradative acetoacetyl-CoA thiolase. Biotechnol. Lett. 30, 1045−1050. (37) Russell, S. A. (2008) Overexpression, purification, and characterization of MmgB and MmgC from Bacillus subtilis strain 168. M.S. Thesis, The University of North Carolina at Greensboro, Greensboro, NC. (38) Vegunta, Y. (2011) Biochemical characterization of MmgB, a gene encoding a 3-hydroxybutyryl-CoA dehydrogenase from Bacillus subtilis 168 and genetic evidence for existence of the methylcitric acid cycle/propionate metabolism in Bacillus subtilis 168. M.S. Thesis, The University of North Carolina at Greensboro, Greensboro, NC. (39) Massey, L. K., Sokatch, J. R., and Conrad, R. S. (1976) Branched-chain amino acid catabolism in bacteria. Bacteriol. Rev. 40, 42−54. (40) Wang, G.-F., Kuriki, T., Roy, K. L., and Kaneda, T. (1993) The primary structure of branched chain α-oxo acid dehydrogenase from Bacillus subtilis and its similarity to other α-oxo acid dehydrogenases. Eur. J. Biochem. 213, 1091−1099. (41) Debarbouille, M., Gardan, R., Arnaud, M., and Rapoport, G. (1999) Role of BkdR, a transcriptional activator of the SigL-dependent isoleucine and valine degradation pathway in Bacillus subtilis. J. Bacteriol. 181, 2059−2066. (42) Namba, Y., Yoshizawa, K., Ejima, A., Hayashi, T., and Kaneda, T. (1969) Coenzyme A- and nicotinamide adenine dinucleotide-dependent branched chain α-keto acid dehydrogenase. J. Biol. Chem. 244, 4437−4447. 1

N


First Biochemical Characterization of a Methylcitric Acid Cycle from

Recommend Documents