Properties of Succinyl-Coenzyme A:L-Malate Coenzyme A Transferase and Its Role in the Autotrophic 3-Hydroxypropionate Cycle of Chloroflexus
http://www.100md.com
《细菌学杂志》
Mikrobiologie, Institut fur Biologie II, Albert-Ludwigs-Universitt Freiburg, Freiburg, Germany
ABSTRACT
The 3-hydroxypropionate cycle has been proposed to operate as the autotrophic CO2 fixation pathway in the phototrophic bacterium Chloroflexus aurantiacus. In this pathway, acetyl coenzyme A (acetyl-CoA) and two bicarbonate molecules are converted to malate. Acetyl-CoA is regenerated from malyl-CoA by L-malyl-CoA lyase. The enzyme forming malyl-CoA, succinyl-CoA:L-malate coenzyme A transferase, was purified. Based on the N-terminal amino acid sequence of its two subunits, the corresponding genes were identified on a gene cluster which also contains the gene for L-malyl-CoA lyase, the subsequent enzyme in the pathway. Both enzymes were severalfold up-regulated under autotrophic conditions, which is in line with their proposed function in CO2 fixation. The two CoA transferase genes were cloned and heterologously expressed in Escherichia coli, and the recombinant enzyme was purified and studied. Succinyl-CoA:L-malate CoA transferase forms a large ()n complex consisting of 46- and 44-kDa subunits and catalyzes the reversible reaction succinyl-CoA + L-malate succinate + L-malyl-CoA. It is specific for succinyl-CoA as the CoA donor but accepts L-citramalate instead of L-malate as the CoA acceptor; the corresponding D-stereoisomers are not accepted. The enzyme is a member of the class III of the CoA transferase family. The demonstration of the missing CoA transferase closes the last gap in the proposed 3-hydroxypropionate cycle.
INTRODUCTION
The phototrophic Chloroflexus aurantiacus, a thermophilic green nonsulfur bacterium, serves as a model organism for the study of bacterial photosynthesis (3). In its natural habitat, hot springs of neutral to slightly alkaline pH, it forms visible orange microbial mats and represents the dominant organism that grows photoautotrophically (35). Yet in the dark and in the presence of organic substrates in the light, the bacterium switches to heterotrophic growth.
This niche seems to favor a special type of central carbon metabolism that allows a flexible response to the offered carbon and electron sources. There are different lines of evidence that autotrophic CO2 fixation proceeds via the 3-hydroxypropionate cycle (Fig. 1). This new metabolic pathway allows the simultaneous assimilation of small organic fermentation products such as acetate, propionate, or succinate, which may be produced by fermenting bacteria underneath the microbial mat.
In brief, acetyl coenzyme A (acetyl-CoA) is carboxylated to succinyl-CoA by a series of steps driven by ATP as the energy source and NADPH as the reductant (1, 9, 13-19, 31-32). CO2 fixation proceeds via acetyl-CoA/propionyl-CoA carboxylase. Only two enzymes are required to transform acetyl-CoA to 3-hydroxypropionate, acetyl-CoA/propionyl-CoA carboxylase and malonyl-CoA reductase. The next enzyme, propionyl-CoA synthase, transforms 3-hydroxypropionate in a three-step process to propionyl-CoA. Carboxylation of propionyl-CoA to methylmalonyl-CoA by acetyl-CoA/propionyl-CoA carboxylase and rearrangement of methylmalonyl-CoA to succinyl-CoA is conventional, as is succinate oxidation to L-malate.
One intriguing problem of the proposed cyclic pathway consists in the regeneration of acetyl-CoA from succinyl-CoA (15) (Fig. 1). Extracts of autotrophically grown cells catalyzed the reaction succinyl-CoA + L-malate succinate + glyoxylate + acetyl-CoA. This reaction was ascribed to two enzymes: (i) a putative coenzyme A transferase that catalyzed the CoA transfer from succinyl-CoA to L-malate, forming succinate and L-malyl-CoA; (ii) L-malyl-CoA is then cleaved by L-malyl-CoA lyase to glyoxylate and acetyl-CoA, closing the cycle (13). The generation of glyoxylate from 2 bicarbonate consumes 3 NADPH and 4 energy-rich phosphate anhydride bonds of ATP and produces 1 reduced quinone in the course of succinate oxidation to fumarate. The assimilation of glyoxylate into cellular building blocks requires a second cyclic pathway whose outlines are just becoming visible (14).
The CoA transfer reaction has not been studied in detail; neither was the enzyme purified and studied nor was its gene(s) identified. Also, the transcriptional regulation of this characteristic new enzyme was of interest. In this investigation, we isolated the succinyl-CoA:L-malate CoA transferase from autotrophically grown cells of C. aurantiacus. Based on the N-terminal amino acid sequences, we identified two genes coding for the enzyme. Heterologous production yielded sufficient amounts of pure enzyme to allow molecular and kinetic characterization. This contribution closes the last gap in the proposed 3-hydroxypropionate cycle.
MATERIALS AND METHODS
Bacteria and growth conditions. C. aurantiacus strain OK-70-fl (DSMZ 636) was grown phototrophically in 2-, 5-, or 12-liter glass fermenters to an optical density at 578 nm (1-cm light path) of 3.5 to 4.0 at 55°C and a pH of around 8. The light exposure was 10,000 to 12,000 lx. Autotrophic growth occurred under anaerobic conditions on a minimal medium supplemented with trace elements and vitamins. The cultures were gassed with a mixture of H2 and CO2 (80%:20% [vol/vol]), as described elsewhere (31). Cells were also grown anaerobically under photoheterotrophic conditions on modified minimal medium D (5) supplemented with 0.25% (wt/vol) Casamino Acids, 0.1% (wt/vol) yeast extract, and trace elements. The medium was buffered with 0.05% glycylglycine-Na+ buffer. Cells were stored under liquid nitrogen until use. Escherichia coli strain BL21(DE3) (33) was grown at 37°C in Luria-Bertani (LB) medium (26). Ampicillin was added to E. coli cultures to a final concentration of 100 μg/ml. Growth was measured photometrically at 578 nm as optical density using cuvettes with a 1-cm light path.
Materials. Chemicals were obtained from Fluka (Neu-Ulm, Germany), Sigma-Aldrich (Deisenhofen, Germany), Merck (Darmstadt, Germany), or Roth (Karlsruhe, Germany). L-[1,4 (2,3)-14C]malate, a mixture of equal amounts of L-[1,4-14C]malate and L-[2,3-14C]malate, was obtained from Amersham Life Science (Braunschweig, Germany). Biochemicals were from Roche Diagnostics (Mannheim, Germany), Applichem (Darmstadt, Germany), or Gerbu (Craiberg, Germany). Materials for cloning and expression were purchased from MBI Fermentas (St. Leon-Rot, Germany), New England Biolabs (Frankfurt, Germany), Novagen (Schwalbach, Germany), Genaxxon Bioscience GmbH (Biberach, Germany), MWG Biotech AG (Ebersberg, Germany), or QIAGEN (Hilden, Germany). Materials and equipment for protein purification were obtained from Amersham Biosciences (Freiburg, Germany) or Millipore (Eschborn, Germany).
Syntheses. (i) Succinyl-CoA, acetyl-CoA, and propionyl-CoA. The CoA thioesters of succinate, acetate, and propionate were synthesized from their anhydrides (28-29) by a modified method described earlier (13), and dry powders were stored at –20°C.
(ii) Malonyl-CoA. Malonyl-CoA was chemically synthesized from monothiophenylmalonate as described previously (15), and dry powders were stored at –20°C.
(iii) Malyl-CoA. L-Malyl-CoA was chemically synthesized from L-malylcaprylcysteamine (S-[-hydroxysuccinyl]-N-caprylcysteamine) as described previously (8, 23), with a slight modification (13). L-Malyl-CoA was stored as freeze-dried powder at –20°C. It contained 80% CoA ester and 20% CoA, as determined by high-pressure liquid chromatography (HPLC) separation and detection at 260 nm.
Preparation of cell extract. C. aurantiacus and E. coli cells were suspended in a twofold volume of 50 mM morpholinepropanesulfonic acid (MOPS)/KOH (pH 7.0) containing 4 mM MgCl2 and 0.2 mg DNase I per ml of cell suspension and passed twice through a chilled French pressure cell at 137 kPa. The lysate was ultracentrifuged for 1 h at 100,000 x g at 4°C.
Heterologous expression and purification of recombinant L-malyl-CoA lyase. mcl from C. aurantiacus was heterologously expressed in E. coli, and the recombinant L-malyl-CoA lyase was purified in three purification steps by heat precipitation, DEAE-Sepharose fast flow chromatography, and size exclusion chromatography, as described elsewhere (13).
Enzyme assays. Succinyl-CoA:L-malate CoA transferase was tested at 55°C, routinely in the forward direction.
(i) HPLC analysis. The assay mixture (0.5 ml) contained 200 mM MOPS/KOH buffer (pH 6.5), 1 mM succinyl-CoA, 10 mM L-malate, and enriched or purified protein. L-Malate was omitted in a control experiment. The reaction was started by the addition of L-malate. Samples of 110 μl were taken after 1 min and 5 min of incubation at 55°C, and the reaction was stopped by the addition of 3 μl of 25% HCl. Precipitated protein was removed by centrifugation, and samples were analyzed for CoA thioesters by HPLC. A reversed-phase column (5 μm, 125 by 4 mm, endcapped, LiChrospher 100; Merck, Darmstadt, Germany) was used for separation of CoA-thioesters. A gradient from 1 to 8% acetonitrile in 50 mM potassium phosphate buffer, pH 6.7, with a flow rate of 1 ml min–1 over 30 min was used. CoA thioesters were detected at 260 nm. Retention times were 2 min (free organic acids), 8 min (L-malyl-CoA), 10 min (succinyl-CoA, free CoA), and 17 min (acetyl-CoA). For the separation of additional CoA thioesters, a gradient from 2 to 10% acetonitrile in 40 mM potassium phosphate buffer, 50 mM formic acid, pH 4.0, with a flow rate of 1 ml min–1 over 40 min was used. CoA thioesters were detected at 260 nm. Retention times were 2 min (free organic acids), 9 min (L-malyl-CoA), 10 min (free CoA, malonyl-CoA), 14 min (L-citramalyl-CoA), 17 min (succinyl-CoA), 18 min (acetyl-CoA), 20 min (itaconyl-CoA), and 26 min (propionyl-CoA).
(ii) Coupled spectrophotometric assay. The succinyl-CoA and L-malate-dependent formation of glyoxylate and acetyl-CoA in the presence of recombinant L-malyl-CoA lyase (13) was monitored photometrically at 324 nm with phenylhydrazine in a continuous assay (324 for glyoxylate-phenylhydrazone = 17,000 M–1cm–1). Succinyl-CoA:L-malate CoA transferase was rate-limiting. The assay mixture (0.5 ml) contained 200 mM MOPS/KOH buffer (pH 6.5), 5 mM MgCl2, 3.5 mM phenylhydrazinium chloride, 1 mM succinyl-CoA, 10 mM L-malate, succinyl-CoA:L-malate CoA transferase, and an excess of recombinant L-malyl-CoA lyase. Either substrate could be used to start the reaction. The apparent Km values were determined at the saturating concentration of the second substrate (10 mM L-malate; 5 mM succinyl-CoA) using 0.05 to 2.0 mM succinyl-CoA or 0.2 to 8.0 mM L-malate/L-citramalate as the other substrate. Buffers used to determine the pH optimum were 200 mM 2-(N-morpholino)ethanesulfonic acid/KOH buffer (pH 5.5 to 6.0) and 200 mM MOPS/KOH (pH 6.0 to 8.0).
The same assay was used to check the enzyme activity with D-malate as the substrate. D-Malate (Fluka) used was contaminated with L-malate (2.6%). We determined this impurity by using the coupled spectrophotometric assay with succinyl-CoA:L-malate CoA transferase, L-malyl-CoA lyase, and D-malate as the substrate. There was a formation of glyoxylate phenylhydrazone which cannot be explained by D-malate consumption because L-malyl-CoA lyase cannot cleave D-malyl-CoA.
Formation of L-[14C]malyl-CoA from L-[1,4 (2,3)-14C]malate and L-malyl-CoA. The assay mixture (0.4 ml) contained 200 mM MOPS/KOH buffer (pH 6.5), 10 mM L-malate, 0.1 mM L-[14C]malate (3 kBq), 1 mM L-malyl-CoA, and 6 μg of purified recombinant CoA transferase. The addition of L-malate and [14C]malate started the reaction. Samples of 110 μl were taken after 2 and 20 min of incubation, and addition of 3 μl of 25% HCl stopped the reaction. Precipitated protein was removed by centrifugation, and samples were analyzed by HPLC (see above). Simultaneous detection of standard compounds and 14C-labeled reaction products was possible by using two detectors (UV and radioactivity) in series.
Cloning and expression of putative succinyl-CoA:L-malate CoA transferase (smt) genes in E. coli BL21(DE3). Standard protocols were used for purification, preparation, cloning, transformation, and amplification of DNA (2, 26). Plasmid DNA was isolated with the QIAprep spin miniprep kit (QIAGEN).
(i) Heterologous expression of smtA from C. aurantiacus. Two oligonucleotides were designed upstream (5'-GACAAGGCATATGCCCCCCACAGGA-3'; 25-mer, NdeI restriction site is underlined) and downstream (5'-TCTGTGGTACCCAAGGTAGCAACT-3'; 24-mer, KpnI restriction site is underlined). PCR was performed with Pfunds polymerase (Genaxxon) for 25 cycles, an annealing temperature of 63°C, and extension at 72°C for 5 min. The 1,335-bp PCR product was purified and cloned into pUC19 vector (New England Biolabs), obtaining plasmid pAS5. The nucleotide sequence of the PCR product was confirmed to ensure that no errors had been introduced. The plasmid was restricted with NdeI and BamHI, and the fragment containing the smtA gene was ligated into pT7/7 (34), resulting in plasmid pAST5.
(ii) Heterologous expression of smtB from C. aurantiacus. Two oligonucleotides were designed upstream (5'-CTAGGATCCTAGGAGTTAAGTCCATGGATGGAACGACCACAAC-3'; 43-mer, BamHI and NcoI restriction sites are underlined, Shine-Dalgarno sequence between the two restriction sites) and downstream (5'-AGGTGATAGAAGCTTCTATTCCTCGAGGACGTGTAGAGCGCAT-3'; 43-mer, HindIII and XhoI restriction sites are underlined) of the gene coding for C. aurantiacus smtB. PCR was performed as described for smtA. The 1,326-bp PCR product was purified and cloned into the pUC19 vector (New England Biolabs) using the BamHI and HindIII restriction sites, obtaining plasmid pAS7. The nucleotide sequence of the PCR product was determined to ensure that no errors had been introduced. The plasmid was restricted with NcoI and XhoI, and the fragment containing the smtB gene was ligated into pET16b (Novagen), resulting in plasmid pASE7.
(iii) Heterologous coexpression of smtA and smtB from C. aurantiacus. The fragment containing the smtB gene was cut out from pAS7 by restriction with BamHI and HindIII, and the fragment was cloned downstream of smtA by restriction of pAS5 with the same enzymes resulting in plasmid pAS12. The NdeI, HindIII fragment containing smtA and smtB (with its own introduced ribosome binding site) was cut out from pAS12 and cloned into pT7/7, resulting in plasmid pAST12. Competent E. coli BL21(DE3) cells (33) were transformed with pAST5, pASE7, or pAST12, grown at 28°C in Luria-Bertani medium containing 100 μg of ampicillin ml–1, and induced at an optical density of 0.8 with 0.5 mM isopropyl--D-thiogalactopyranoside. After additional growth for 14 h, the cells were harvested and stored in liquid nitrogen until use.
RT-PCR. Total RNA from C. aurantiacus cells grown anaerobically under autotrophic conditions was used for reverse transcription-PCR (RT-PCR). RNA was isolated with an RNeasy total RNA kit (QIAGEN) and was separated from contaminating DNA by treatment with fast protein liquid chromatography-purified DNase I (1 U per μg of total RNA; Amersham Pharmacia Biotech) for 30 min at 37°C. Complete removal of DNA from the RNA preparation was verified by amplifying the intergenic region between orf1 and smtA coding in different directions, with cDNA as the template (5'-ATCATCTTCCACCTGCTCAGTACG-3' and 5'-AATGCCACTCAACGGCAACTGCTC-3'), corresponding to nucleotides 20212 to 20764 of contig NZ_AAAH02000019 from C. aurantiacus. One microgram of purified total RNA was used to prepare cDNA by using a Moloney murine leukemia virus reverse transcriptase (RevertAid M-MuLV RT) and a mixture of completely random hexanucleotides for random priming (RevertAid first-strand cDNA synthesis kit; Fermentas). Gene expression was studied by amplification of intergenic regions between open reading frames (ORFs). The following primers were used to PCR amplify an intergenic region between smtA and smtB: 5'-ACGTATGAGGTGTTGCGCGAG-3' and 5'-AGCGTAACCGAACGCTTGTTG-3'. As a control, part of the gene coding for the -subunit of RNA polymerase was PCR amplified using primers 5'-TGAATTGCGTATCCTGACCACC-3' and 5'-TGAAGAGCAGTGACTCGATCAG-3'.
DNA sequencing and computer analysis. DNA sequence determination of purified plasmids was performed by G. L. Igloi (Institut Biologie II, Universitt Freiburg, Germany). DNA and amino acid sequences were analyzed with the BLAST network service at the National Center for Biotechnology Information (Bethesda, Md.), the local C. aurantiacus server (http://genome.jgi-psf.org/draft_microbes/chlau/chlau.home.html) at the Department of Energy (DOE) Joint Genome Institute (Walnut Creek, CA), and the program Clone Manager 7 (SciEd Software, Cary, NC). The phylogenetic tree of protein sequences was constructed using the multialignment program (http://prodes.toulouse.inra.fr/multalin/multalin.html).
Reconstitution experiments. Extracts of E. coli BL21(DE3) cells containing only heterologously expressed SmtA (plasmid pAST5) or SmtB (plasmid pASE7) were used in experiments to reconstitute the succinyl-CoA:L-malate CoA transferase activity measured by HPLC. Cell extracts of E. coli producing either SmtA or SmtB were mixed gently and used directly in an HPLC assay. Also, the two different cell extracts were preincubated for 30 min at 37°C and afterwards used in an HPLC assay. As a positive control, E. coli cell extracts containing recombinant SmtA and SmtB were used. The assay mixture (250 μl) contained 200 mM MOPS/KOH buffer (pH 6.5), 10 mM L-malate, 2 mM succinyl-CoA, and 0.5 mg of each cell extract. Samples of 110 μl were taken after 2 min and analyzed by HPLC.
Purification of succinyl-CoA:L-malate CoA transferase from C. aurantiacus. All purification steps were performed at 4°C. CoA transferase activity was measured using the coupled spectrophotometric assay with recombinant L-malyl-CoA lyase.
(i) Ammonium sulfate precipitation. Cell extract (100,000 x g supernatant) from 16 g of cell mass (wet weight) of autotrophically grown C. aurantiacus cells was precipitated with ammonium sulfate between 25 and 50% saturation at 4°C. After centrifugation (20,000 x g for 20 min), the precipitate was dissolved in 10 ml of 50 mM MOPS/KOH, pH 7.0 containing 4 mM MgCl2.
(ii) Gel filtration. The protein solution after ammonium sulfate precipitation was applied in two runs (5 ml each) to a Superdex 200 gel filtration column (bed volume, 320 ml; Amersham Biosciences), equilibrated with 20 mM Tris/HCl buffer, pH 7.0 (buffer A), containing 100 mM KCl with a flow rate of 2.5 ml min–1. Active protein eluted with a retention volume of 100 to 150 ml. Active fractions were immediately pooled, desalted, and concentrated to a final volume of 15 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(iii) MonoQ chromatography. The concentrated sample after size exclusion chromatography was applied onto a MonoQ 10/10 column (Amersham Biosciences) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. After washing the column with 5 bed volumes of buffer A and 5 bed volumes of buffer A plus 200 mM KCl, the enzyme eluted in a step between 200 and 350 mM KCl. Active fractions were immediately pooled, desalted, and concentrated to a final volume of 18 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(iv) Affinity chromatography. The concentrated sample obtained by MonoQ chromatography (18 ml) was applied onto an 8-ml Reactive Green 19 agarose column (Sigma) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. The column was washed with 5 bed volumes of buffer A and 10 bed volumes of buffer A containing 200 mM KCl and developed with an 80-ml (increasing) linear gradient of 200 mM to 400 mM KCl in buffer A. Active fractions (200 to 400 mM KCl) were pooled (100 ml) and concentrated to a final volume of 18 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(v) Resource phenyl chromatography. The active Reactive Green pool was adjusted to a final concentration of 1 M ammonium sulfate using saturated ammonium sulfate solution. The protein fraction was centrifuged (20,000 x g for 10 min), and the supernatant was directly applied to a 1-ml Resource phenyl column (Amersham Biosciences) at a flow rate of 1 ml min–1. The column had been equilibrated with buffer A containing 1 M ammonium sulfate. After washing the column with 5 bed volumes of this buffer and 10 bed volumes of buffer A plus 0.5 M ammonium sulfate, the column was developed with 15 bed volumes of a decreasing linear gradient of 0.5 to 0 M salt. The active protein eluted with 0.2 to 0 M ammonium sulfate, and active fractions were pooled (5 ml), concentrated, and stored at –20°C with 10% glycerol.
Purification of recombinant succinyl-CoA:L-malate CoA transferase from E. coli. Purification was carried out at 4°C. CoA transferase activity was measured using the coupled spectrophotometric assay with recombinant L-malyl-CoA lyase.
(i) Heat precipitation. Cell extract (100,000 x g supernatant) from 12 g of cells (wet mass) of E. coli with plasmid pAST12 was incubated at 65°C for 10 min to precipitate unwanted protein from E. coli cell extracts, followed by centrifugation (20,800 x g) at 4°C for 10 min. The supernatant was incubated at 70°C for 10 min, followed by centrifugation (20,800 x g) at 4°C for 10 min.
(ii) MonoQ chromatography. The supernatant, after heat precipitation (20 ml), was applied onto a MonoQ 10/10 chromatography column (Amersham Biosciences) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. The column was washed with 5 bed volumes of buffer A and 7 bed volumes of buffer A plus 150 mM KCl and developed with an 80-ml (increasing) linear gradient of 150 to 400 mM KCl in buffer A. Active fractions were immediately pooled (40 ml), desalted, and concentrated to a final volume of 16 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(iii) Affinity chromatography. The concentrated sample obtained by MonoQ chromatography (15 ml) was applied onto an 8-ml Reactive Green 19 agarose column (Sigma) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. The column was washed with 2 bed volumes of buffer A and 3 bed volumes of buffer A containing 150 mM KCl and developed with an 80-ml (increasing) linear gradient of 150 to 400 mM KCl in buffer A. Active fractions were pooled (40 ml), concentrated, and stored at –20°C with 10% glycerol.
Determination of molecular mass. The concentrated sample obtained by MonoQ chromatography was applied to a 120-ml Highload Superdex 200 16/60 column (Amersham Biosciences) equilibrated with buffer A containing 100 mM KCl. The column was developed at a flow rate of 1 ml min–1. The combined active fractions eluted with a retention volume of 53 to 61 ml were concentrated, and glycerol was added to a final concentration of 10% and stored at –20°C. The native molecular mass of the enzyme was estimated using the same gel filtration column. The column was calibrated with thyroglobulin (660 kDa), ferritin (450 kDa), catalase (240 kDa), alcohol dehydrogenase (150 kDa), bovine serum albumin (67 kDa), and ovalbumin (45 kDa) as molecular mass standards.
Inactivation experiments. Two types of experiments were performed using recombinant enzyme (22).
(i) Inactivation by sodium borohydride. Enriched succinyl-CoA:L-malate CoA transferase, after MonoQ chromatography (35 μg of the protein), was added to 460 μl of 200 mM MOPS/KOH buffer, pH 6.5, which either contained or lacked succinyl-CoA (2 mM). The enzyme was treated with 5 μl of 1 M NaBH4 in 1 M NaOH, and 5 μl of 1 M HCl was added immediately afterwards. The mixtures were incubated for 10 min at 55°C and tested for CoA transferase activity by the coupled spectrophotometric assay.
(ii) Inactivation by hydroxylamine. Enriched succinyl-CoA:L-malate CoA transferase, after MonoQ chromatography (35 μg of protein), was added to 420 μl of 200 mM MOPS/KOH buffer, pH 6.5, which either contained or lacked succinyl-CoA (2 mM). The enzyme was treated with 10 mM hydroxylamine for 10 min at 55°C, and subsequently, enzyme activity was tested with the spectrophotometric assay.
Other methods. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 12.5%) was performed as described by Laemmli (21). The following were used as molecular mass standards: rabbit phosphorylase b, 97 kDa; bovine serum albumin, 67 kDa; egg ovalbumin, 45 kDa; lactate dehydrogenase, 34 kDa; carbonic anhydrase, 29 kDa; lysozyme, 14 kDa. Proteins were visualized by Coomassie blue staining (36). Protein was determined by the method of Bradford (4) using bovine serum albumin as the standard. Determination of the N-terminal amino acid sequence of purified enzyme from C. aurantiacus after blotting on a polyvinylidene difluoride membrane was performed by TopLab (Martinsried, Germany) using an Applied Biosystems Procise 492 sequencer (Weiterstadt, Germany). The phenylthiohydantoin derivatives were identified with an online Applied Biosystems Analyzer 140 C. Determination of the purified recombinant proteins with peptide mass fingerprinting was performed by TopLab (Martinsried, Germany) using an Applied Biosystems Voyager-DE STR mass spectrometer.
RESULTS
Succinyl-CoA:L-malate coenzyme A transferase activity in cell extracts. Cell extracts of autotrophically grown C. aurantiacus possess succinyl-CoA:L-malate coenzyme A transferase activity (15). In previous studies, the CoA transferase activity and its regulation could not be studied accurately, since L-malyl-CoA lyase was not available. Here, succinyl-CoA:L-malate coenzyme A transferase activity was measured in a coupled spectrophotometric assay using purified L-malyl-CoA lyase in excess. This allowed us to determine the regulation of the CoA transferase independent of the regulation of malyl-CoA lyase. The lyase requires Mg2+ as a cofactor, which was included in the assay (13). The succinyl-CoA- and L-malate-dependent formation of the phenylhydrazone of glyoxylate was monitored at 55°C, the optimal growth temperature of the organism.
The specific succinyl-CoA:L-malate coenzyme A transferase activity in cell extracts of photoautotrophically grown C. aurantiacus was 20 to 33 nmol min–1 mg protein–1 with L-malate, depending on the batch of cells. The deviations in this kind of specific activity measurements were 10%. Extracts of photoheterotrophically grown cells catalyzed this reaction with a specific activity of 2 nmol min–1 mg protein–1 with L-malate. The 10- to 15-fold up-regulation of the succinyl-CoA:L-malate CoA transferase under autotrophic conditions suggests a role of the enzyme in the CO2 fixation pathway. Similar results were obtained before (15); however, the regulation could have been due to down-regulation of L-malyl-CoA lyase in heterotrophically grown cells.
Purification of succinyl-CoA:L-malate coenzyme A transferase. The succinyl-CoA:L-malate CoA transferase activity was found in the soluble fraction after ultracentrifugation of cell extract for 1 h at 100,000 x g. Succinyl-CoA:L-malate CoA transferase was purified 56-fold from 16 g of autotrophically grown cells in five chromatography steps, with a yield of 8% (Table 1). The specific activity of the purified enzyme at 55°C was 1.2 μmol min–1 mg protein–1. SDS-PAGE after the last purification step revealed one strong band corresponding to 44 kDa and one weak additional band at approximately 46 kDa (Fig. 2, lane 6).
Sequencing of the 44-kDa band revealed the following N-terminal amino acid sequence: M (P/D) (P/G) T. The amounts of P/D and P/G at positions 2 and 3 were identical, suggesting the presence of two proteins with an N-terminal amino acid sequence of MPPT, MDGT, MPGT, or MDPT. These results suggest that succinyl-CoA:L-malate CoA transferase consists of two subunits with similar molecular masses. Therefore, we assume that the 46-kDa band was an impurity.
Identification of the genes coding for succinyl-CoA:L-malate CoA transferase. The protein sequence of succinyl-CoA:(R)-benzylsuccinate CoA transferase from Thauera aromatica (subunit BbsF, GenBank accession number AAF89841) was used to screen the incomplete C. aurantiacus genome sequence for possible succinyl-CoA:L-malate CoA transferase genes. Two open reading frames coding for proteins with 30% (each) sequence identity to BbsF were identified next to each other on contig NZ_AAAH02000019 (Fig. 3). The N-terminal amino acid sequences of the deduced protein sequences of those genes were MPPT and MDGT. These genes were named smtA and smtB. The predicted molecular mass of the protein encoded by smtB (44 kDa/405 amino acids) matches the molecular mass determined for the two subunits of succinyl-CoA:L-malate CoA transferase. Yet the predicted molecular mass of the protein encoded by smtA (46 kDa/428 amino acids) does not exactly match the observed apparent mass in SDS-PAGE of the subunit of native succinyl-CoA:L-malate CoA transferase.
Cloning and overexpression of succinyl-CoA:L-malate coenzyme A transferase genes (smtA and smtB) in E. coli, proof of function, and purification. The two genes, smtA and smtB, proposed to encode the subunits of succinyl-CoA:L-malate coenzyme A transferase (SmtAB) were expressed in E. coli using a T7 promoter/polymerase expression system and plasmid pAST12, providing optimized Shine-Dalgarno sequences in front of both genes. A cell extract of E. coli with vector pAST12 expressing both smtA and smtB was heat precipitated, and the soluble supernatant was measured using the coupled spectrophotometric assay. Succinyl-CoA:L-malate CoA transferase was easily detected, and the specific activity was 0.14 μmol min–1 mg protein–1; this activity was lacking in E. coli extracts of cells carrying plasmid pT7/7 but missing the DNA insert.
The recombinant succinyl-CoA:L-malate CoA transferase was further purified in two steps from 12 g of recombinant E. coli cells using heat precipitation, anion-exchange chromatography, and affinity chromatography. The yield was 12% (Table 2), and the specific activity at 55°C was 7.5 μmol min–1 mg protein–1. SDS-PAGE showed two distinct bands (46 kDa and 44 kDa) and three additional very faint protein bands (<5%) (Fig. 2). Sequencing these two strong bands by peptide mass fingerprinting confirmed that the two proteins represent SmtA and SmtB. Sequence coverage was 59% for SmtA and 79% for SmtB. This indicates that the genome of the strain studied here (OK-70-fl) does not differ significantly from that of the strain whose genome was sequenced (J-10-fl). As calculated from the specific activity of the purified recombinant enzyme, succinyl-CoA:L-malate CoA transferase represents approximately 0.3% of total soluble protein of autotrophically grown C. aurantiacus cells.
The two genes smtA and smtB were also separately expressed in E. coli using a T7 promoter/polymerase expression system. The corresponding proteins were produced (Fig. 4) and were present in the soluble fraction after ultracentrifugation. Both proteins were found in the soluble fraction after heat treatment of extracts at 55°C (10 min). However, after heat treatment at 65°C for 10 min followed by incubation at 75°C for 10 min, heterologously produced proteins were precipitated as judged by SDS-PAGE. No succinyl-CoA:L-malate CoA transferase activity could be measured either with cell extract containing SmtA or with cell extract containing SmtB after heat treatment at 55°C.
Reconstitution experiments were performed to check the ability of the subunits SmtA and SmtB to reconstitute active succinyl-CoA:L-malate CoA transferase after separate expression. Either cell extracts containing SmtA or SmtB were combined and tested for transferase activity at 55°C or the two subunits were preincubated together at 37°C for 30 min, followed by the measurement of enzyme activity. No CoA transferase activity could be reconstituted; however, cell extracts of heterologously expressed smtA and smtB were active in the appropriate control experiments. This indicates that the active enzyme consists of both subunits and that formation of active enzyme may need concerted biosynthesis of its subunits. Alternatively, the correct conditions for reconstitution have just not yet been found.
Molecular and catalytic properties of recombinant succinyl-CoA:L-malate CoA transferase. Purified recombinant succinyl-CoA:L-malate CoA transferase consisted of two polypeptides of 46 and 44 kDa. SDS-PAGE of the heterologously expressed purified protein yielded two distinct protein bands that stained equally. The mass of the native recombinant enzyme was determined as 550 ± 50 kDa by gel filtration chromatography on Superdex 200. The observed size fits best to a subunit composition of ()6, with a calculated mass of 546 kDa. The UV-visible spectrum ( = 200 to 600 nm) of the purified enzyme showed a single peak at 280 nm, with no indication for the presence of a chromophoric cofactor. The enzyme is stable for 1 week at 4°C in Tris buffer, pH 7, containing 100 mM KCl or kept frozen for several weeks with 10% glycerol (vol/vol) without significant loss of activity.
Purified recombinant enzyme was tested with succinyl-CoA and L- or D-malate as substrates; the products formed after 1 and 5 min of incubation were analyzed by HPLC. Only L-malate could be converted to its CoA thioester. This reaction did not require Mg2+ nor was it inhibited by 1 mM EDTA or stimulated by 1 mM dithioerythritol.
The kinetics and substrate specificities were determined using the coupled spectrophotometric assay including purified recombinant L-malyl-CoA lyase in excess. The reaction was linearly time dependent for minutes and protein dependent in a large range, as long as the lyase was not rate-limiting. The stoichiometry of the reaction was approximately 0.95 mol glyoxylate phenylhydrazone formed per 1 mol L-malate added, when succinyl-CoA was in excess. The enzyme exhibited optimal activity at 55°C at pH 6.5. Half-maximal activity was observed at pH 6.1 and 7.1, respectively. The turnover number per dimer ()1 was 11.4 s–1. As indicated above, the enzyme was specific for L-malate. It was inactive with D-malate, D-citramalate, mesaconate, acetoacetate, adipate, -hydroxy--methylglutarate, citrate, and methylmalonate, as tested by HPLC assay. The enzyme was even more active with L-citramalate and less active with itaconate (Table 3). Succinyl-CoA was the only CoA donor found so far which could not be substituted by propionyl-CoA, malonyl-CoA, or acetyl-CoA.
The apparent Km values were determined using the coupled spectrophotometric assay under saturating concentrations of the cosubstrate (5 mM succinyl-CoA and 10 mM L-malate) and were 0.5 mM for succinyl-CoA, 1.3 mM for L-malate, and 1.1 mM for L-citramalate (Table 3).
Comparison with other enzymes, exchange reactions, and inactivation assays of succinyl-CoA:L-malate CoA transferase. Protein sequence comparisons have been performed for the subunits SmtA and SmtB to three characterized enzymes in the CaiB-BaiF and CoA transferase family III: BbsF -subunit and BbsE -subunit of succinyl-CoA:(R)-benzylsuccinate CoA transferase of Thauera aromatica (GenBank accession numbers AAF89841 and AAF89840), Frc formyl-CoA:oxalate CoA transferase of Oxalobacter formigenes (GenBank accession number AAC45298), and CaiB crotonobetainyl-CoA:(R)-carnitine CoA transferase of E. coli (GenBank accession number CAA52112). Both subunits of succinyl-CoA:L-malate CoA transferase showed high amino acid sequence similarities to these enzymes with highest similarities to BbsF (amino acid identity and similarity, respectively: 30% and 46% to SmtA; 30% and 47% to SmtB). This indicates that the succinyl-CoA:L-malate CoA transferase also belongs to the family III transferases (Fig. 5).
Further experiments should give more evidences for this assignment. CoA transfer reactions imply that the high energy of the thioester bond is conserved in the catalytic cycle. This is only possible if the acyl or the CoA moiety is covalently linked to the enzyme via a reactive and energy-rich bond, e.g., as anhydride (family III) and thioester (family I, II), respectively. To test this hypothesis, we performed two types of experiments. We used L-malyl-CoA as the CoA donor and the substrate analogue L-citramalate as the potential CoA acceptor and tested by HPLC whether CoA transfer occurred. Indeed, L-citramalyl-CoA was formed. We also incubated L-malyl-CoA in the presence of L-[14C]malate and tested for the formation of L-[malyl-14C]CoA. This radioisotope-exchange reaction also occurred (data not shown), which is in line with the hypothesis, but could not definitely classify the succinyl-CoA:L-malate CoA transferase to a specific family of transferases.
Enzymes of family I use a ping-pong mechanism and are specifically inactivated by micromolar to millimolar concentrations of borohydride or hydroxylamine in the presence of the CoA donor. Enzymes of family II are subunits of complex lyases and contain a covalently bound (5'-phosphoribosyl)-3'-dephospho-CoA moiety. There is no indication for such a cofactor in succinyl-CoA:L-malate CoA transferase. Enzymes of family III seem to use a concerted mechanism via a ternary complex; this form is not very sensitive to borohydride and hydroxylamine. Therefore, we tested the effect of these compounds on succinyl-CoA:L-malate CoA transferase. Enzyme was incubated in the absence of CoA thioesters or in the presence of 2 mM succinyl-CoA for 10 min with 0.1 to 10 mM NaBH4 and with 10 mM hydroxylamine, respectively, and then assayed by the coupled spectrophotometric assay. No inactivation by 0.1 mM borohydride was recorded in the test with and without succinyl-CoA, and nearly 85% of the activity was retained when the enzyme was preincubated with 10 mM hydroxylamine for 10 min in the presence of 2 mM succinyl-CoA. This suggests that no reactive enzyme-CoA intermediate was present. Inactivation with 10 mM borohydride decreased activity to 28% in the presence of 2 mM succinyl-CoA.
Gene organization and cotranscription. The two genes encoding the subunits of succinyl-CoA:L-malate CoA transferase are located on a cluster of seven ORFs (Fig. 3). Cotranscription of genes smtA and smtB during autotrophic growth of C. aurantiacus was studied by performing RT-PCR experiments with mRNA isolated from autotrophically grown cells and comparing the results with results obtained with chromosomal DNA from C. aurantiacus (Fig. 6). The positive control, part of the gene for the -subunit of RNA polymerase, was amplified with both chromosomal DNA and cDNA. In contrast, the negative control, the intergenic region between adjacent orf1 and smtA (which are orientated in different directions), was negative with cDNA and positive with chromosomal DNA. The amplification of the intergenic region between smtA and smtB was positive in both instances. This indicates that smtA and smtB were cotranscribed and are part of the same operon.
DISCUSSION
Role of the enzyme. We have characterized a new CoA transferase whose function in autotrophic CO2 fixation of C. aurantiacus is to activate L-malate to its CoA thioester with succinyl-CoA as the CoA donor. The use of succinyl-CoA is an elegant means to couple the penultimate step of the cycle with a former one, making use of the high-energy thioester bond. In the next step, L-malyl-CoA is cleaved into glyoxylate and acetyl-CoA (13). Glyoxylate forms the primary net CO2 fixation product, and acetyl-CoA starts a new CO2 fixation cycle (15). The autotrophic 3-hydroxypropionate cycle is thus closed (Fig. 1). Mutant studies would be highly desirable in support of this proposed role but are difficult to perform due to the poorly developed genetic system and the difficult plating of this filamentous bacterium.
However, the following features corroborate the importance of succinyl-CoA:L-malate CoA transferase for autotrophic growth. (i) The specific enzyme activity in extracts of autotrophically grown cells amounted to 20 to 33 nmol min–1 mg protein–1. This in vitro rate is similar to the estimated minimal enzyme rate (12 nmol min–1 mg protein–1) that can explain the slow autotrophic growth (13). (ii) The 10- to 15-fold up-regulation under autotrophic conditions is in line with its function. (iii) The enzyme is highly specific for its CoA donor and acceptor, and the apparent Km values are in physiological range. (iv) The product of the reversible transferase reaction is specifically utilized by L-malyl-CoA/-methylmalyl-CoA lyase (13). This enzyme is also induced under autotrophic conditions, and the genes for these two proteins are located in the same gene cluster (Fig. 3). The fact that the enzyme also accepts L-citramalate may indicate that it may have two functions. The enzyme appears to be active with C4 dicarboxylic acids that have substitutions at C-2 provided the L conformation is given.
Succinyl-CoA:L-malate CoA transferase, a member of class III CoA transferases. CoA transferases catalyze the reversible transfer of CoA between organic acids and are known from many bacterial and eukaryotic species. Based on their reaction mechanism and amino acid sequences, the transferases can be divided into three distinct families (12). Family I enzymes consist of two dissimilar subunits in different aggregation states (22, 44). These enzymes operate by a ping-pong mechanism and can be specially inactivated by treatment with hydroxylamine or borohydride when they are preincubated with their CoA donor compound. Family II contains only a few members. The physiological substrate transferred by these enzymes is an enzyme-associated acyl carrier protein subunit.
The alignment of the two genes, smtA and smtB, coding for the subunits of the new CoA transferase shows the highest similarity to known members of CoA transferase family III. This family is listed as the "CaiB/BaiF family" in the paralogous protein family database and forms a distinct third branch in the phylogenetic tree of the family of CoA transferases, as deduced from sequence comparison (Fig. 5). The function of most members of the gene family is unknown. The currently characterized enzymes are involved mainly in anaerobic bacterial metabolism. They share considerable amino acid sequence similarity and may use a similar concerted reaction mechanism via a ternary complex. Yet the quaternary structures vary considerably, with the known members being either homodimers (10-11, 20, 24-25, 27, 30), heterodimers (22), or more complex structures (7). High similarity of SmtAB exists to the subunits BbsE and BbsF of the heterodimeric succinyl-CoA(R)-benzylsuccinate CoA transferase of T. aromatica (amino acid identity and similarity, respectively, of BbsF of 30% and 46% to SmtA and 30% and 47% to SmtB). Sct represents succinyl-CoA:D-citramalate CoA transferase from Chloroflexus aurantiacus, an enzyme required to convert glyoxylate back to acetyl-CoA via citramalate (S. Friedmann, B. Alber, and G. Fuchs, unpublished results). A highly conserved aspartate residue (Asp 169 in the CaiB nomenclature), which is located in the active site and binds the organic acid in an anhydride bond (11, 20, 30), is conserved in SmtA and SmtB. Other residues that are important for binding are conserved as well, such as Arg 16, Gly 37, Ala 38, Val 40, Asp 90, Leu 184, Thr 190, and Gly 193 (30), with the exception of His 185 (Fig. 7).
Succinyl-CoA:L-malate CoA transferase preincubated with succinyl-CoA was almost insensitive against hydroxylamine treatment, which would inactivate CoA transferases of family I and II. Furthermore, the enzyme was not inactivated with 0.1 mM NaBH4 after preincubation with succinyl-CoA. The decrease in activity observed at 10 mM NaBH4 cannot be explained, but similar results were reported for the related enzymes succinyl-CoA:(R)-benzylsuccinate CoA transferase from Thauera aromatica (22) and (E)-cinnamoyl-CoA:(R)-phenyllactate CoA transferase from Clostridium sporogenes (7). A partial inhibition could possibly be due to partial reduction of the postulated Asp169-anhydride bond (25) during the CoA transfer reaction.
The native enzyme may form a heterohexamer ()6 of 550 kDa ± 50 kDa. The higher oligomeric state of this enzyme compared to other members of this enzyme family may provide a greater structural stability of the enzyme at thermophilic growth conditions (6). None of the individually produced two subunits was active alone nor could activity be reconstituted by combination of the two proteins. Rather, only coexpression of both genes resulted in active recombinant enzyme that formed a 550-kDa complex. This observation is plausible in view of the structure of different CoA transferases (10-11, 20, 24-25, 30) in which the two subunits form a tightly interconnected complex. The CoA binding site is formed at the interface of the two subunits. Such an interlocked dimer can possibly only be developed when both subunits are simultaneously formed but not from prefolded individual subunits. The folding pathway that might lead to such a structure is still at issue.
One observation cannot be explained so far. Purified native enzyme from C. aurantiacus exhibited on SDS-PAGE one broad band of approximately 44 kDa which contained both subunits in equal amounts. This follows from N-terminal amino acid sequencing of this protein band. However, when the two genes were cloned together and expressed in E. coli, the active enzyme showed two distinct bands in SDS-PAGE which corresponded to the expected sizes of 44 and 46 kDa. This may indicate that the 46-kDa subunit of the native enzyme in C. aurantiacus is somehow modified, which results in an altered migration behavior in electrophoresis. One possibility is a C-terminal processing of the larger subunit in the organism but not in the expression system. The additional band on SDS-PAGE after purification of the natural enzyme migrates somewhat higher than the recombinant 46-kDa subunit and is assumed to represent an impurity.
Genes adjacent to the succinyl-CoA:L-malate CoA transferase genes (smtA and smtB) on the chromosome of C. aurantiacus. The genes for succinyl-CoA:L-malate CoA transferase are located together in a gene cluster of seven ORFs which may play a role in autotrophic CO2 fixation (accession number NZ_AAAH02000019, 19,200 to 30,600 bps) (Fig. 3). RT-PCR studies with mRNA of autotrophically grown C. aurantiacus cells showed that smtA and smtB form a transcriptional unit, as one would predict. The proteins encoded by orf2 and orf3 have no similarities to other proteins in the database, whereas the protein encoded by ctr3 is similar to SmtA and SmtB and encodes a putative CoA transferase (COG1804). The protein encoded by mcl is L-malyl-CoA/-methylmalyl-CoA lyase, which has been characterized before (13) and which catalyzes the next two steps in the pathway, i.e., cleavage of L-malyl-CoA and formation of -methylmalyl-CoA from glyoxylate and propionyl-CoA (Fig. 1). -Methylmalyl-CoA is dehydrated to mesaconyl-CoA in the second cycle of the 3-hydroxypropionate pathway used for glyoxylate assimilation (14). The gene mcd downstream of mcl codes for a member of the enoyl-CoA hydratase family (COG2030) and could, therefore, encode -methylmalyl-CoA dehydratase. A related protein is postulated in the acetate assimilation pathway of Rhodobacter sphaeroides and is thought to catalyze the hydration of mesaconyl-CoA to -methylmalyl-CoA (B. Alber and G. Fuchs, unpublished results).
Similarity of SmtA and SmtB. The two subunits of succinyl-CoA:L-malate CoA transferase are very similar to each other (Fig. 7) (59% amino acid sequence identity and 74% similarity). The high sequence similarity and the fact that the two genes are located side by side in the gene cluster could result from an ancient gene duplication. This holds true also for the two subunits BbsE and BbsF of the heterodimeric succinyl-CoA:(R)-benzylsuccinate CoA transferase of T. aromatica (22).
ACKNOWLEDGMENTS
Thanks are due to Nasser Gad'on and Christa Ebenau-Jehle for expert technical assistance.
REFERENCES
Alber, B. E., and G. Fuchs. 2002. Propionyl-coenzyme A synthase from Chloroflexus aurantiacus, a key enzyme of the 3-hydroxypropionate cycle for autotrophic CO2 fixation. J. Biol. Chem. 277:12137-12143.
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1987. Current protocols in molecular biology. John Wiley and Sons, New York, N.Y.
Blankenship, R. E., M. T. Madigan, and C. E. Bauer. 1995. Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Bradford, M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.
Castenholz, R. W. 1969. Thermophilic blue-green algae and the thermal environment. Bacteriol. Rev. 33:476-504.
Dalhus, B., M. Saarinen, U. H. Sauer, P. Eklund, K. Johansson, A. Karlsson, S. Ramaswamy, A. Bjork, B. Synstad, K. Naterstad, R. Sirevag, and H. Eklund. 2002. Structural basis for thermophilic protein stability: structures of thermophilic and mesophilic malate dehydrogenases. J. Mol. Biol. 318:707-721.
Dickert, S., A. J. Pierik, D. Linder, and W. Buckel. 2000. The involvement of coenzyme A esters in the dehydration of (R)-phenyllactate to (E)-cinnamate by Clostridium sporogenes. Eur. J. Biochem. 267:3874-3884.
Eggerer, H., and C. H. Grunewlder. 1964. Zum Mechanismus der biologischen Umwandlung von Citronensure, IV. Synthese von Malyl-Coenzym A und seiner Diastereoisomeren. Liebigs Ann. Chem. 677:200-208.
Eisenreich, W., G. Strauss, U. Werz, G. Fuchs, and A. Bacher. 1993. Retrobiosynthetic analysis of carbon fixation in the phototrophic eubacterium Chloroflexus aurantiacus. Eur. J. Biochem. 215:619-632.
Gogos, A., J. Gorman, and L. Shapiro. 2004. Structure of Escherichia coli YfdW, a type III CoA transferase. Acta Crystallogr. Sect. D 60:507-511.
Gruez, A., V. Roig-Zamboni, C. Valencia, V. Campanacci, and C. Cambillau. 2003. The crystal structure of Escherichia coli YfdW gene product reveals a new fold of two interlaced rings identifying a wide family of CoA transferases. J. Biol. Chem. 278:34582-34586.
Heider, J. 2001. A new family of CoA transferases. FEBS Lett. 509:345-349.
Herter, S., A. Busch, and G. Fuchs. 2002. L-Malyl-coenzyme A lyase/-methylmalyl-coenzyme A lyase from Chloroflexus aurantiacus, a bifunctional enzyme involved in autotrophic CO2 fixation. J. Bacteriol. 184:5999-6006.
Herter, S., G. Fuchs, A. Bacher, and W. Eisenreich. 2002. A bicyclic autotrophic CO2 fixation pathway in Chloroflexus aurantiacus. J. Biol. Chem. 277:20277-20283.
Herter, S., J. Farfsing, N. Gad'on, C. Rieder, W. Eisenreich, A. Bacher, and G. Fuchs. 2001. Autotrophic CO2 fixation in Chloroflexus aurantiacus: study of glyoxylate formation and assimilation via the 3-hydroxypropionate cycle. J. Bacteriol. 183:4305-4316.
Holo, H. 1989. Chloroflexus aurantiacus secretes 3-hydroxypropionate, a possible intermediate in the assimilation of CO2 and acetate. Arch. Microbiol. 151:252-256.
Holo, H., and D. Grace. 1987. Polyglucose synthesis in Chloroflexus aurantiacus studied by 13C-NMR. Arch. Microbiol. 148:292-297.
Holo, H., and R. Sirevg. 1986. Autotrophic growth and CO2 fixation in Chloroflexus aurantiacus. Arch. Mirobiol. 145:173-180.
Hugler, M., C. Menendez, H. Schgger, and G. Fuchs. 2002. Malonyl-coenzyme A reductase from Chloroflexus aurantiacus, a key enzyme of the 3-hydroxypropionate cycle for autotrophic CO2 fixation. J. Bacteriol. 184:2404-2410.
Jonsson, S., S. Ricagno, Y. Lindqvist, and N. G. J. Richards. 2004. Kinetic and mechanistic characterization of the formyl-CoA transferase from Oxalobacter formigenes. J. Biol. Chem. 279:36003-36012.
Laemmli, U. K. 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227:680-685.
Leutwein, C., and J. Heider. 2001. Succinyl-CoA:(R)-benzylsuccinate CoA-transferase: an enzyme of the anaerobic toluene catabolic pathway in denitrifying bacteria. J. Bacteriol. 183:4288-4295.
Mohrhauer, H., K. Christiansen, M. Gan, M. Deubig, and R. T. Holman. 1968. Improved method for the preparation of malonyl coenzyme A. J. Lipid Res. 9:398-399.
Rangarajan, E. S., Y. Li, P. Iannuzzi, M. Cygler, and A. Matte. 2005. Crystal structure of Escherichia coli crotonobetainyl-CoA: carnitine CoA-transferase (CaiB) and its complexes with CoA and carnitinyl-CoA. Biochemistry 44:5728-5738.
Ricagno, S., S. Jonsson, N. Richards, and Y. Lindqvist. 2003. Formyl-CoA transferase encloses the CoA binding site at the interface of an interlcked dimer. EMBO J. 22:3210-3219.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Sidhu, H., S. D. Ogden, H.-Y. Lung, B. G. Luttge, A. L. Baetz, and A. B. Peck. 1997. DNA sequencing and expression of the formyl-coenzyme A transferase gene, frc, from Oxalobacter formigenes. J. Bacteriol. 179:3378-3381.
Simon, E. 1957. S-succinyl coenzyme A. Biochem. Prep. 5:30-32.
Stadtman, E. R. 1957. Preparation and assay of acyl coenzyme A and other thiol esters; use of hydroxylamine. Methods Enzymol. 3:931-941.
Stenmaek, P., D. Gurmu, and P. Nordlund. 2004. Crystal structure of CaiB, a type-III CoA transferase in carnitine metabolism. Biochemistry 43:13996-14003.
Strauss, G., and G. Fuchs. 1993. Enzymes of a novel autotrophic CO2 fixation pathway in the phototrophic bacterium Chloroflexus aurantiacus, the 3-hydroxypropionate cycle. Eur. J. Biochem. 215:633-643.
Strauss, G., W. Eisenreich, A. Bacher, and G. Fuchs. 1992. 13C-NMR study of autotrophic CO2 fixation pathways in the sulfur-reducing archaebacterium Thermoproteus neutrophilus and in the phototrophic eubacterium Chloroflexus aurantiacus. Eur. J. Biochem. 205:853-866.
Studier, F. W., and B. A. Moffatt. 1986. Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J. Mol. Biol. 189:113-130.
Tabor, S., and C. C. Richardson. 1985. A bacteriophage T7 RNA polymerase/promoter system for controlled exclusive expression of specific genes. Proc. Natl. Acad. Sci. USA 82:1074-1078.
van der Meer, M. T. J., S. Schouten, J. W. de Leeuw, and D. M. Ward. 2000. Autotrophy of green non-sulphur bacteria in hot spring microbial mats: biological explanations for isotopically heavy organic carbon in the geological record. Environ. Microbiol. 2:428-435.
Zehr, B. D., T. J. Savin, and R. E. Hall. 1989. A one-step, low-background Coomassie staining procedure for polyacrylamide gels. Anal. Biochem. 182:157-159.(Silke Friedmann, Astrid S)
ABSTRACT
The 3-hydroxypropionate cycle has been proposed to operate as the autotrophic CO2 fixation pathway in the phototrophic bacterium Chloroflexus aurantiacus. In this pathway, acetyl coenzyme A (acetyl-CoA) and two bicarbonate molecules are converted to malate. Acetyl-CoA is regenerated from malyl-CoA by L-malyl-CoA lyase. The enzyme forming malyl-CoA, succinyl-CoA:L-malate coenzyme A transferase, was purified. Based on the N-terminal amino acid sequence of its two subunits, the corresponding genes were identified on a gene cluster which also contains the gene for L-malyl-CoA lyase, the subsequent enzyme in the pathway. Both enzymes were severalfold up-regulated under autotrophic conditions, which is in line with their proposed function in CO2 fixation. The two CoA transferase genes were cloned and heterologously expressed in Escherichia coli, and the recombinant enzyme was purified and studied. Succinyl-CoA:L-malate CoA transferase forms a large ()n complex consisting of 46- and 44-kDa subunits and catalyzes the reversible reaction succinyl-CoA + L-malate succinate + L-malyl-CoA. It is specific for succinyl-CoA as the CoA donor but accepts L-citramalate instead of L-malate as the CoA acceptor; the corresponding D-stereoisomers are not accepted. The enzyme is a member of the class III of the CoA transferase family. The demonstration of the missing CoA transferase closes the last gap in the proposed 3-hydroxypropionate cycle.
INTRODUCTION
The phototrophic Chloroflexus aurantiacus, a thermophilic green nonsulfur bacterium, serves as a model organism for the study of bacterial photosynthesis (3). In its natural habitat, hot springs of neutral to slightly alkaline pH, it forms visible orange microbial mats and represents the dominant organism that grows photoautotrophically (35). Yet in the dark and in the presence of organic substrates in the light, the bacterium switches to heterotrophic growth.
This niche seems to favor a special type of central carbon metabolism that allows a flexible response to the offered carbon and electron sources. There are different lines of evidence that autotrophic CO2 fixation proceeds via the 3-hydroxypropionate cycle (Fig. 1). This new metabolic pathway allows the simultaneous assimilation of small organic fermentation products such as acetate, propionate, or succinate, which may be produced by fermenting bacteria underneath the microbial mat.
In brief, acetyl coenzyme A (acetyl-CoA) is carboxylated to succinyl-CoA by a series of steps driven by ATP as the energy source and NADPH as the reductant (1, 9, 13-19, 31-32). CO2 fixation proceeds via acetyl-CoA/propionyl-CoA carboxylase. Only two enzymes are required to transform acetyl-CoA to 3-hydroxypropionate, acetyl-CoA/propionyl-CoA carboxylase and malonyl-CoA reductase. The next enzyme, propionyl-CoA synthase, transforms 3-hydroxypropionate in a three-step process to propionyl-CoA. Carboxylation of propionyl-CoA to methylmalonyl-CoA by acetyl-CoA/propionyl-CoA carboxylase and rearrangement of methylmalonyl-CoA to succinyl-CoA is conventional, as is succinate oxidation to L-malate.
One intriguing problem of the proposed cyclic pathway consists in the regeneration of acetyl-CoA from succinyl-CoA (15) (Fig. 1). Extracts of autotrophically grown cells catalyzed the reaction succinyl-CoA + L-malate succinate + glyoxylate + acetyl-CoA. This reaction was ascribed to two enzymes: (i) a putative coenzyme A transferase that catalyzed the CoA transfer from succinyl-CoA to L-malate, forming succinate and L-malyl-CoA; (ii) L-malyl-CoA is then cleaved by L-malyl-CoA lyase to glyoxylate and acetyl-CoA, closing the cycle (13). The generation of glyoxylate from 2 bicarbonate consumes 3 NADPH and 4 energy-rich phosphate anhydride bonds of ATP and produces 1 reduced quinone in the course of succinate oxidation to fumarate. The assimilation of glyoxylate into cellular building blocks requires a second cyclic pathway whose outlines are just becoming visible (14).
The CoA transfer reaction has not been studied in detail; neither was the enzyme purified and studied nor was its gene(s) identified. Also, the transcriptional regulation of this characteristic new enzyme was of interest. In this investigation, we isolated the succinyl-CoA:L-malate CoA transferase from autotrophically grown cells of C. aurantiacus. Based on the N-terminal amino acid sequences, we identified two genes coding for the enzyme. Heterologous production yielded sufficient amounts of pure enzyme to allow molecular and kinetic characterization. This contribution closes the last gap in the proposed 3-hydroxypropionate cycle.
MATERIALS AND METHODS
Bacteria and growth conditions. C. aurantiacus strain OK-70-fl (DSMZ 636) was grown phototrophically in 2-, 5-, or 12-liter glass fermenters to an optical density at 578 nm (1-cm light path) of 3.5 to 4.0 at 55°C and a pH of around 8. The light exposure was 10,000 to 12,000 lx. Autotrophic growth occurred under anaerobic conditions on a minimal medium supplemented with trace elements and vitamins. The cultures were gassed with a mixture of H2 and CO2 (80%:20% [vol/vol]), as described elsewhere (31). Cells were also grown anaerobically under photoheterotrophic conditions on modified minimal medium D (5) supplemented with 0.25% (wt/vol) Casamino Acids, 0.1% (wt/vol) yeast extract, and trace elements. The medium was buffered with 0.05% glycylglycine-Na+ buffer. Cells were stored under liquid nitrogen until use. Escherichia coli strain BL21(DE3) (33) was grown at 37°C in Luria-Bertani (LB) medium (26). Ampicillin was added to E. coli cultures to a final concentration of 100 μg/ml. Growth was measured photometrically at 578 nm as optical density using cuvettes with a 1-cm light path.
Materials. Chemicals were obtained from Fluka (Neu-Ulm, Germany), Sigma-Aldrich (Deisenhofen, Germany), Merck (Darmstadt, Germany), or Roth (Karlsruhe, Germany). L-[1,4 (2,3)-14C]malate, a mixture of equal amounts of L-[1,4-14C]malate and L-[2,3-14C]malate, was obtained from Amersham Life Science (Braunschweig, Germany). Biochemicals were from Roche Diagnostics (Mannheim, Germany), Applichem (Darmstadt, Germany), or Gerbu (Craiberg, Germany). Materials for cloning and expression were purchased from MBI Fermentas (St. Leon-Rot, Germany), New England Biolabs (Frankfurt, Germany), Novagen (Schwalbach, Germany), Genaxxon Bioscience GmbH (Biberach, Germany), MWG Biotech AG (Ebersberg, Germany), or QIAGEN (Hilden, Germany). Materials and equipment for protein purification were obtained from Amersham Biosciences (Freiburg, Germany) or Millipore (Eschborn, Germany).
Syntheses. (i) Succinyl-CoA, acetyl-CoA, and propionyl-CoA. The CoA thioesters of succinate, acetate, and propionate were synthesized from their anhydrides (28-29) by a modified method described earlier (13), and dry powders were stored at –20°C.
(ii) Malonyl-CoA. Malonyl-CoA was chemically synthesized from monothiophenylmalonate as described previously (15), and dry powders were stored at –20°C.
(iii) Malyl-CoA. L-Malyl-CoA was chemically synthesized from L-malylcaprylcysteamine (S-[-hydroxysuccinyl]-N-caprylcysteamine) as described previously (8, 23), with a slight modification (13). L-Malyl-CoA was stored as freeze-dried powder at –20°C. It contained 80% CoA ester and 20% CoA, as determined by high-pressure liquid chromatography (HPLC) separation and detection at 260 nm.
Preparation of cell extract. C. aurantiacus and E. coli cells were suspended in a twofold volume of 50 mM morpholinepropanesulfonic acid (MOPS)/KOH (pH 7.0) containing 4 mM MgCl2 and 0.2 mg DNase I per ml of cell suspension and passed twice through a chilled French pressure cell at 137 kPa. The lysate was ultracentrifuged for 1 h at 100,000 x g at 4°C.
Heterologous expression and purification of recombinant L-malyl-CoA lyase. mcl from C. aurantiacus was heterologously expressed in E. coli, and the recombinant L-malyl-CoA lyase was purified in three purification steps by heat precipitation, DEAE-Sepharose fast flow chromatography, and size exclusion chromatography, as described elsewhere (13).
Enzyme assays. Succinyl-CoA:L-malate CoA transferase was tested at 55°C, routinely in the forward direction.
(i) HPLC analysis. The assay mixture (0.5 ml) contained 200 mM MOPS/KOH buffer (pH 6.5), 1 mM succinyl-CoA, 10 mM L-malate, and enriched or purified protein. L-Malate was omitted in a control experiment. The reaction was started by the addition of L-malate. Samples of 110 μl were taken after 1 min and 5 min of incubation at 55°C, and the reaction was stopped by the addition of 3 μl of 25% HCl. Precipitated protein was removed by centrifugation, and samples were analyzed for CoA thioesters by HPLC. A reversed-phase column (5 μm, 125 by 4 mm, endcapped, LiChrospher 100; Merck, Darmstadt, Germany) was used for separation of CoA-thioesters. A gradient from 1 to 8% acetonitrile in 50 mM potassium phosphate buffer, pH 6.7, with a flow rate of 1 ml min–1 over 30 min was used. CoA thioesters were detected at 260 nm. Retention times were 2 min (free organic acids), 8 min (L-malyl-CoA), 10 min (succinyl-CoA, free CoA), and 17 min (acetyl-CoA). For the separation of additional CoA thioesters, a gradient from 2 to 10% acetonitrile in 40 mM potassium phosphate buffer, 50 mM formic acid, pH 4.0, with a flow rate of 1 ml min–1 over 40 min was used. CoA thioesters were detected at 260 nm. Retention times were 2 min (free organic acids), 9 min (L-malyl-CoA), 10 min (free CoA, malonyl-CoA), 14 min (L-citramalyl-CoA), 17 min (succinyl-CoA), 18 min (acetyl-CoA), 20 min (itaconyl-CoA), and 26 min (propionyl-CoA).
(ii) Coupled spectrophotometric assay. The succinyl-CoA and L-malate-dependent formation of glyoxylate and acetyl-CoA in the presence of recombinant L-malyl-CoA lyase (13) was monitored photometrically at 324 nm with phenylhydrazine in a continuous assay (324 for glyoxylate-phenylhydrazone = 17,000 M–1cm–1). Succinyl-CoA:L-malate CoA transferase was rate-limiting. The assay mixture (0.5 ml) contained 200 mM MOPS/KOH buffer (pH 6.5), 5 mM MgCl2, 3.5 mM phenylhydrazinium chloride, 1 mM succinyl-CoA, 10 mM L-malate, succinyl-CoA:L-malate CoA transferase, and an excess of recombinant L-malyl-CoA lyase. Either substrate could be used to start the reaction. The apparent Km values were determined at the saturating concentration of the second substrate (10 mM L-malate; 5 mM succinyl-CoA) using 0.05 to 2.0 mM succinyl-CoA or 0.2 to 8.0 mM L-malate/L-citramalate as the other substrate. Buffers used to determine the pH optimum were 200 mM 2-(N-morpholino)ethanesulfonic acid/KOH buffer (pH 5.5 to 6.0) and 200 mM MOPS/KOH (pH 6.0 to 8.0).
The same assay was used to check the enzyme activity with D-malate as the substrate. D-Malate (Fluka) used was contaminated with L-malate (2.6%). We determined this impurity by using the coupled spectrophotometric assay with succinyl-CoA:L-malate CoA transferase, L-malyl-CoA lyase, and D-malate as the substrate. There was a formation of glyoxylate phenylhydrazone which cannot be explained by D-malate consumption because L-malyl-CoA lyase cannot cleave D-malyl-CoA.
Formation of L-[14C]malyl-CoA from L-[1,4 (2,3)-14C]malate and L-malyl-CoA. The assay mixture (0.4 ml) contained 200 mM MOPS/KOH buffer (pH 6.5), 10 mM L-malate, 0.1 mM L-[14C]malate (3 kBq), 1 mM L-malyl-CoA, and 6 μg of purified recombinant CoA transferase. The addition of L-malate and [14C]malate started the reaction. Samples of 110 μl were taken after 2 and 20 min of incubation, and addition of 3 μl of 25% HCl stopped the reaction. Precipitated protein was removed by centrifugation, and samples were analyzed by HPLC (see above). Simultaneous detection of standard compounds and 14C-labeled reaction products was possible by using two detectors (UV and radioactivity) in series.
Cloning and expression of putative succinyl-CoA:L-malate CoA transferase (smt) genes in E. coli BL21(DE3). Standard protocols were used for purification, preparation, cloning, transformation, and amplification of DNA (2, 26). Plasmid DNA was isolated with the QIAprep spin miniprep kit (QIAGEN).
(i) Heterologous expression of smtA from C. aurantiacus. Two oligonucleotides were designed upstream (5'-GACAAGGCATATGCCCCCCACAGGA-3'; 25-mer, NdeI restriction site is underlined) and downstream (5'-TCTGTGGTACCCAAGGTAGCAACT-3'; 24-mer, KpnI restriction site is underlined). PCR was performed with Pfunds polymerase (Genaxxon) for 25 cycles, an annealing temperature of 63°C, and extension at 72°C for 5 min. The 1,335-bp PCR product was purified and cloned into pUC19 vector (New England Biolabs), obtaining plasmid pAS5. The nucleotide sequence of the PCR product was confirmed to ensure that no errors had been introduced. The plasmid was restricted with NdeI and BamHI, and the fragment containing the smtA gene was ligated into pT7/7 (34), resulting in plasmid pAST5.
(ii) Heterologous expression of smtB from C. aurantiacus. Two oligonucleotides were designed upstream (5'-CTAGGATCCTAGGAGTTAAGTCCATGGATGGAACGACCACAAC-3'; 43-mer, BamHI and NcoI restriction sites are underlined, Shine-Dalgarno sequence between the two restriction sites) and downstream (5'-AGGTGATAGAAGCTTCTATTCCTCGAGGACGTGTAGAGCGCAT-3'; 43-mer, HindIII and XhoI restriction sites are underlined) of the gene coding for C. aurantiacus smtB. PCR was performed as described for smtA. The 1,326-bp PCR product was purified and cloned into the pUC19 vector (New England Biolabs) using the BamHI and HindIII restriction sites, obtaining plasmid pAS7. The nucleotide sequence of the PCR product was determined to ensure that no errors had been introduced. The plasmid was restricted with NcoI and XhoI, and the fragment containing the smtB gene was ligated into pET16b (Novagen), resulting in plasmid pASE7.
(iii) Heterologous coexpression of smtA and smtB from C. aurantiacus. The fragment containing the smtB gene was cut out from pAS7 by restriction with BamHI and HindIII, and the fragment was cloned downstream of smtA by restriction of pAS5 with the same enzymes resulting in plasmid pAS12. The NdeI, HindIII fragment containing smtA and smtB (with its own introduced ribosome binding site) was cut out from pAS12 and cloned into pT7/7, resulting in plasmid pAST12. Competent E. coli BL21(DE3) cells (33) were transformed with pAST5, pASE7, or pAST12, grown at 28°C in Luria-Bertani medium containing 100 μg of ampicillin ml–1, and induced at an optical density of 0.8 with 0.5 mM isopropyl--D-thiogalactopyranoside. After additional growth for 14 h, the cells were harvested and stored in liquid nitrogen until use.
RT-PCR. Total RNA from C. aurantiacus cells grown anaerobically under autotrophic conditions was used for reverse transcription-PCR (RT-PCR). RNA was isolated with an RNeasy total RNA kit (QIAGEN) and was separated from contaminating DNA by treatment with fast protein liquid chromatography-purified DNase I (1 U per μg of total RNA; Amersham Pharmacia Biotech) for 30 min at 37°C. Complete removal of DNA from the RNA preparation was verified by amplifying the intergenic region between orf1 and smtA coding in different directions, with cDNA as the template (5'-ATCATCTTCCACCTGCTCAGTACG-3' and 5'-AATGCCACTCAACGGCAACTGCTC-3'), corresponding to nucleotides 20212 to 20764 of contig NZ_AAAH02000019 from C. aurantiacus. One microgram of purified total RNA was used to prepare cDNA by using a Moloney murine leukemia virus reverse transcriptase (RevertAid M-MuLV RT) and a mixture of completely random hexanucleotides for random priming (RevertAid first-strand cDNA synthesis kit; Fermentas). Gene expression was studied by amplification of intergenic regions between open reading frames (ORFs). The following primers were used to PCR amplify an intergenic region between smtA and smtB: 5'-ACGTATGAGGTGTTGCGCGAG-3' and 5'-AGCGTAACCGAACGCTTGTTG-3'. As a control, part of the gene coding for the -subunit of RNA polymerase was PCR amplified using primers 5'-TGAATTGCGTATCCTGACCACC-3' and 5'-TGAAGAGCAGTGACTCGATCAG-3'.
DNA sequencing and computer analysis. DNA sequence determination of purified plasmids was performed by G. L. Igloi (Institut Biologie II, Universitt Freiburg, Germany). DNA and amino acid sequences were analyzed with the BLAST network service at the National Center for Biotechnology Information (Bethesda, Md.), the local C. aurantiacus server (http://genome.jgi-psf.org/draft_microbes/chlau/chlau.home.html) at the Department of Energy (DOE) Joint Genome Institute (Walnut Creek, CA), and the program Clone Manager 7 (SciEd Software, Cary, NC). The phylogenetic tree of protein sequences was constructed using the multialignment program (http://prodes.toulouse.inra.fr/multalin/multalin.html).
Reconstitution experiments. Extracts of E. coli BL21(DE3) cells containing only heterologously expressed SmtA (plasmid pAST5) or SmtB (plasmid pASE7) were used in experiments to reconstitute the succinyl-CoA:L-malate CoA transferase activity measured by HPLC. Cell extracts of E. coli producing either SmtA or SmtB were mixed gently and used directly in an HPLC assay. Also, the two different cell extracts were preincubated for 30 min at 37°C and afterwards used in an HPLC assay. As a positive control, E. coli cell extracts containing recombinant SmtA and SmtB were used. The assay mixture (250 μl) contained 200 mM MOPS/KOH buffer (pH 6.5), 10 mM L-malate, 2 mM succinyl-CoA, and 0.5 mg of each cell extract. Samples of 110 μl were taken after 2 min and analyzed by HPLC.
Purification of succinyl-CoA:L-malate CoA transferase from C. aurantiacus. All purification steps were performed at 4°C. CoA transferase activity was measured using the coupled spectrophotometric assay with recombinant L-malyl-CoA lyase.
(i) Ammonium sulfate precipitation. Cell extract (100,000 x g supernatant) from 16 g of cell mass (wet weight) of autotrophically grown C. aurantiacus cells was precipitated with ammonium sulfate between 25 and 50% saturation at 4°C. After centrifugation (20,000 x g for 20 min), the precipitate was dissolved in 10 ml of 50 mM MOPS/KOH, pH 7.0 containing 4 mM MgCl2.
(ii) Gel filtration. The protein solution after ammonium sulfate precipitation was applied in two runs (5 ml each) to a Superdex 200 gel filtration column (bed volume, 320 ml; Amersham Biosciences), equilibrated with 20 mM Tris/HCl buffer, pH 7.0 (buffer A), containing 100 mM KCl with a flow rate of 2.5 ml min–1. Active protein eluted with a retention volume of 100 to 150 ml. Active fractions were immediately pooled, desalted, and concentrated to a final volume of 15 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(iii) MonoQ chromatography. The concentrated sample after size exclusion chromatography was applied onto a MonoQ 10/10 column (Amersham Biosciences) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. After washing the column with 5 bed volumes of buffer A and 5 bed volumes of buffer A plus 200 mM KCl, the enzyme eluted in a step between 200 and 350 mM KCl. Active fractions were immediately pooled, desalted, and concentrated to a final volume of 18 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(iv) Affinity chromatography. The concentrated sample obtained by MonoQ chromatography (18 ml) was applied onto an 8-ml Reactive Green 19 agarose column (Sigma) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. The column was washed with 5 bed volumes of buffer A and 10 bed volumes of buffer A containing 200 mM KCl and developed with an 80-ml (increasing) linear gradient of 200 mM to 400 mM KCl in buffer A. Active fractions (200 to 400 mM KCl) were pooled (100 ml) and concentrated to a final volume of 18 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(v) Resource phenyl chromatography. The active Reactive Green pool was adjusted to a final concentration of 1 M ammonium sulfate using saturated ammonium sulfate solution. The protein fraction was centrifuged (20,000 x g for 10 min), and the supernatant was directly applied to a 1-ml Resource phenyl column (Amersham Biosciences) at a flow rate of 1 ml min–1. The column had been equilibrated with buffer A containing 1 M ammonium sulfate. After washing the column with 5 bed volumes of this buffer and 10 bed volumes of buffer A plus 0.5 M ammonium sulfate, the column was developed with 15 bed volumes of a decreasing linear gradient of 0.5 to 0 M salt. The active protein eluted with 0.2 to 0 M ammonium sulfate, and active fractions were pooled (5 ml), concentrated, and stored at –20°C with 10% glycerol.
Purification of recombinant succinyl-CoA:L-malate CoA transferase from E. coli. Purification was carried out at 4°C. CoA transferase activity was measured using the coupled spectrophotometric assay with recombinant L-malyl-CoA lyase.
(i) Heat precipitation. Cell extract (100,000 x g supernatant) from 12 g of cells (wet mass) of E. coli with plasmid pAST12 was incubated at 65°C for 10 min to precipitate unwanted protein from E. coli cell extracts, followed by centrifugation (20,800 x g) at 4°C for 10 min. The supernatant was incubated at 70°C for 10 min, followed by centrifugation (20,800 x g) at 4°C for 10 min.
(ii) MonoQ chromatography. The supernatant, after heat precipitation (20 ml), was applied onto a MonoQ 10/10 chromatography column (Amersham Biosciences) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. The column was washed with 5 bed volumes of buffer A and 7 bed volumes of buffer A plus 150 mM KCl and developed with an 80-ml (increasing) linear gradient of 150 to 400 mM KCl in buffer A. Active fractions were immediately pooled (40 ml), desalted, and concentrated to a final volume of 16 ml by ultrafiltration (Amicon YM 30 membrane; Millipore).
(iii) Affinity chromatography. The concentrated sample obtained by MonoQ chromatography (15 ml) was applied onto an 8-ml Reactive Green 19 agarose column (Sigma) which had been equilibrated with buffer A with a flow rate of 1 ml min–1. The column was washed with 2 bed volumes of buffer A and 3 bed volumes of buffer A containing 150 mM KCl and developed with an 80-ml (increasing) linear gradient of 150 to 400 mM KCl in buffer A. Active fractions were pooled (40 ml), concentrated, and stored at –20°C with 10% glycerol.
Determination of molecular mass. The concentrated sample obtained by MonoQ chromatography was applied to a 120-ml Highload Superdex 200 16/60 column (Amersham Biosciences) equilibrated with buffer A containing 100 mM KCl. The column was developed at a flow rate of 1 ml min–1. The combined active fractions eluted with a retention volume of 53 to 61 ml were concentrated, and glycerol was added to a final concentration of 10% and stored at –20°C. The native molecular mass of the enzyme was estimated using the same gel filtration column. The column was calibrated with thyroglobulin (660 kDa), ferritin (450 kDa), catalase (240 kDa), alcohol dehydrogenase (150 kDa), bovine serum albumin (67 kDa), and ovalbumin (45 kDa) as molecular mass standards.
Inactivation experiments. Two types of experiments were performed using recombinant enzyme (22).
(i) Inactivation by sodium borohydride. Enriched succinyl-CoA:L-malate CoA transferase, after MonoQ chromatography (35 μg of the protein), was added to 460 μl of 200 mM MOPS/KOH buffer, pH 6.5, which either contained or lacked succinyl-CoA (2 mM). The enzyme was treated with 5 μl of 1 M NaBH4 in 1 M NaOH, and 5 μl of 1 M HCl was added immediately afterwards. The mixtures were incubated for 10 min at 55°C and tested for CoA transferase activity by the coupled spectrophotometric assay.
(ii) Inactivation by hydroxylamine. Enriched succinyl-CoA:L-malate CoA transferase, after MonoQ chromatography (35 μg of protein), was added to 420 μl of 200 mM MOPS/KOH buffer, pH 6.5, which either contained or lacked succinyl-CoA (2 mM). The enzyme was treated with 10 mM hydroxylamine for 10 min at 55°C, and subsequently, enzyme activity was tested with the spectrophotometric assay.
Other methods. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; 12.5%) was performed as described by Laemmli (21). The following were used as molecular mass standards: rabbit phosphorylase b, 97 kDa; bovine serum albumin, 67 kDa; egg ovalbumin, 45 kDa; lactate dehydrogenase, 34 kDa; carbonic anhydrase, 29 kDa; lysozyme, 14 kDa. Proteins were visualized by Coomassie blue staining (36). Protein was determined by the method of Bradford (4) using bovine serum albumin as the standard. Determination of the N-terminal amino acid sequence of purified enzyme from C. aurantiacus after blotting on a polyvinylidene difluoride membrane was performed by TopLab (Martinsried, Germany) using an Applied Biosystems Procise 492 sequencer (Weiterstadt, Germany). The phenylthiohydantoin derivatives were identified with an online Applied Biosystems Analyzer 140 C. Determination of the purified recombinant proteins with peptide mass fingerprinting was performed by TopLab (Martinsried, Germany) using an Applied Biosystems Voyager-DE STR mass spectrometer.
RESULTS
Succinyl-CoA:L-malate coenzyme A transferase activity in cell extracts. Cell extracts of autotrophically grown C. aurantiacus possess succinyl-CoA:L-malate coenzyme A transferase activity (15). In previous studies, the CoA transferase activity and its regulation could not be studied accurately, since L-malyl-CoA lyase was not available. Here, succinyl-CoA:L-malate coenzyme A transferase activity was measured in a coupled spectrophotometric assay using purified L-malyl-CoA lyase in excess. This allowed us to determine the regulation of the CoA transferase independent of the regulation of malyl-CoA lyase. The lyase requires Mg2+ as a cofactor, which was included in the assay (13). The succinyl-CoA- and L-malate-dependent formation of the phenylhydrazone of glyoxylate was monitored at 55°C, the optimal growth temperature of the organism.
The specific succinyl-CoA:L-malate coenzyme A transferase activity in cell extracts of photoautotrophically grown C. aurantiacus was 20 to 33 nmol min–1 mg protein–1 with L-malate, depending on the batch of cells. The deviations in this kind of specific activity measurements were 10%. Extracts of photoheterotrophically grown cells catalyzed this reaction with a specific activity of 2 nmol min–1 mg protein–1 with L-malate. The 10- to 15-fold up-regulation of the succinyl-CoA:L-malate CoA transferase under autotrophic conditions suggests a role of the enzyme in the CO2 fixation pathway. Similar results were obtained before (15); however, the regulation could have been due to down-regulation of L-malyl-CoA lyase in heterotrophically grown cells.
Purification of succinyl-CoA:L-malate coenzyme A transferase. The succinyl-CoA:L-malate CoA transferase activity was found in the soluble fraction after ultracentrifugation of cell extract for 1 h at 100,000 x g. Succinyl-CoA:L-malate CoA transferase was purified 56-fold from 16 g of autotrophically grown cells in five chromatography steps, with a yield of 8% (Table 1). The specific activity of the purified enzyme at 55°C was 1.2 μmol min–1 mg protein–1. SDS-PAGE after the last purification step revealed one strong band corresponding to 44 kDa and one weak additional band at approximately 46 kDa (Fig. 2, lane 6).
Sequencing of the 44-kDa band revealed the following N-terminal amino acid sequence: M (P/D) (P/G) T. The amounts of P/D and P/G at positions 2 and 3 were identical, suggesting the presence of two proteins with an N-terminal amino acid sequence of MPPT, MDGT, MPGT, or MDPT. These results suggest that succinyl-CoA:L-malate CoA transferase consists of two subunits with similar molecular masses. Therefore, we assume that the 46-kDa band was an impurity.
Identification of the genes coding for succinyl-CoA:L-malate CoA transferase. The protein sequence of succinyl-CoA:(R)-benzylsuccinate CoA transferase from Thauera aromatica (subunit BbsF, GenBank accession number AAF89841) was used to screen the incomplete C. aurantiacus genome sequence for possible succinyl-CoA:L-malate CoA transferase genes. Two open reading frames coding for proteins with 30% (each) sequence identity to BbsF were identified next to each other on contig NZ_AAAH02000019 (Fig. 3). The N-terminal amino acid sequences of the deduced protein sequences of those genes were MPPT and MDGT. These genes were named smtA and smtB. The predicted molecular mass of the protein encoded by smtB (44 kDa/405 amino acids) matches the molecular mass determined for the two subunits of succinyl-CoA:L-malate CoA transferase. Yet the predicted molecular mass of the protein encoded by smtA (46 kDa/428 amino acids) does not exactly match the observed apparent mass in SDS-PAGE of the subunit of native succinyl-CoA:L-malate CoA transferase.
Cloning and overexpression of succinyl-CoA:L-malate coenzyme A transferase genes (smtA and smtB) in E. coli, proof of function, and purification. The two genes, smtA and smtB, proposed to encode the subunits of succinyl-CoA:L-malate coenzyme A transferase (SmtAB) were expressed in E. coli using a T7 promoter/polymerase expression system and plasmid pAST12, providing optimized Shine-Dalgarno sequences in front of both genes. A cell extract of E. coli with vector pAST12 expressing both smtA and smtB was heat precipitated, and the soluble supernatant was measured using the coupled spectrophotometric assay. Succinyl-CoA:L-malate CoA transferase was easily detected, and the specific activity was 0.14 μmol min–1 mg protein–1; this activity was lacking in E. coli extracts of cells carrying plasmid pT7/7 but missing the DNA insert.
The recombinant succinyl-CoA:L-malate CoA transferase was further purified in two steps from 12 g of recombinant E. coli cells using heat precipitation, anion-exchange chromatography, and affinity chromatography. The yield was 12% (Table 2), and the specific activity at 55°C was 7.5 μmol min–1 mg protein–1. SDS-PAGE showed two distinct bands (46 kDa and 44 kDa) and three additional very faint protein bands (<5%) (Fig. 2). Sequencing these two strong bands by peptide mass fingerprinting confirmed that the two proteins represent SmtA and SmtB. Sequence coverage was 59% for SmtA and 79% for SmtB. This indicates that the genome of the strain studied here (OK-70-fl) does not differ significantly from that of the strain whose genome was sequenced (J-10-fl). As calculated from the specific activity of the purified recombinant enzyme, succinyl-CoA:L-malate CoA transferase represents approximately 0.3% of total soluble protein of autotrophically grown C. aurantiacus cells.
The two genes smtA and smtB were also separately expressed in E. coli using a T7 promoter/polymerase expression system. The corresponding proteins were produced (Fig. 4) and were present in the soluble fraction after ultracentrifugation. Both proteins were found in the soluble fraction after heat treatment of extracts at 55°C (10 min). However, after heat treatment at 65°C for 10 min followed by incubation at 75°C for 10 min, heterologously produced proteins were precipitated as judged by SDS-PAGE. No succinyl-CoA:L-malate CoA transferase activity could be measured either with cell extract containing SmtA or with cell extract containing SmtB after heat treatment at 55°C.
Reconstitution experiments were performed to check the ability of the subunits SmtA and SmtB to reconstitute active succinyl-CoA:L-malate CoA transferase after separate expression. Either cell extracts containing SmtA or SmtB were combined and tested for transferase activity at 55°C or the two subunits were preincubated together at 37°C for 30 min, followed by the measurement of enzyme activity. No CoA transferase activity could be reconstituted; however, cell extracts of heterologously expressed smtA and smtB were active in the appropriate control experiments. This indicates that the active enzyme consists of both subunits and that formation of active enzyme may need concerted biosynthesis of its subunits. Alternatively, the correct conditions for reconstitution have just not yet been found.
Molecular and catalytic properties of recombinant succinyl-CoA:L-malate CoA transferase. Purified recombinant succinyl-CoA:L-malate CoA transferase consisted of two polypeptides of 46 and 44 kDa. SDS-PAGE of the heterologously expressed purified protein yielded two distinct protein bands that stained equally. The mass of the native recombinant enzyme was determined as 550 ± 50 kDa by gel filtration chromatography on Superdex 200. The observed size fits best to a subunit composition of ()6, with a calculated mass of 546 kDa. The UV-visible spectrum ( = 200 to 600 nm) of the purified enzyme showed a single peak at 280 nm, with no indication for the presence of a chromophoric cofactor. The enzyme is stable for 1 week at 4°C in Tris buffer, pH 7, containing 100 mM KCl or kept frozen for several weeks with 10% glycerol (vol/vol) without significant loss of activity.
Purified recombinant enzyme was tested with succinyl-CoA and L- or D-malate as substrates; the products formed after 1 and 5 min of incubation were analyzed by HPLC. Only L-malate could be converted to its CoA thioester. This reaction did not require Mg2+ nor was it inhibited by 1 mM EDTA or stimulated by 1 mM dithioerythritol.
The kinetics and substrate specificities were determined using the coupled spectrophotometric assay including purified recombinant L-malyl-CoA lyase in excess. The reaction was linearly time dependent for minutes and protein dependent in a large range, as long as the lyase was not rate-limiting. The stoichiometry of the reaction was approximately 0.95 mol glyoxylate phenylhydrazone formed per 1 mol L-malate added, when succinyl-CoA was in excess. The enzyme exhibited optimal activity at 55°C at pH 6.5. Half-maximal activity was observed at pH 6.1 and 7.1, respectively. The turnover number per dimer ()1 was 11.4 s–1. As indicated above, the enzyme was specific for L-malate. It was inactive with D-malate, D-citramalate, mesaconate, acetoacetate, adipate, -hydroxy--methylglutarate, citrate, and methylmalonate, as tested by HPLC assay. The enzyme was even more active with L-citramalate and less active with itaconate (Table 3). Succinyl-CoA was the only CoA donor found so far which could not be substituted by propionyl-CoA, malonyl-CoA, or acetyl-CoA.
The apparent Km values were determined using the coupled spectrophotometric assay under saturating concentrations of the cosubstrate (5 mM succinyl-CoA and 10 mM L-malate) and were 0.5 mM for succinyl-CoA, 1.3 mM for L-malate, and 1.1 mM for L-citramalate (Table 3).
Comparison with other enzymes, exchange reactions, and inactivation assays of succinyl-CoA:L-malate CoA transferase. Protein sequence comparisons have been performed for the subunits SmtA and SmtB to three characterized enzymes in the CaiB-BaiF and CoA transferase family III: BbsF -subunit and BbsE -subunit of succinyl-CoA:(R)-benzylsuccinate CoA transferase of Thauera aromatica (GenBank accession numbers AAF89841 and AAF89840), Frc formyl-CoA:oxalate CoA transferase of Oxalobacter formigenes (GenBank accession number AAC45298), and CaiB crotonobetainyl-CoA:(R)-carnitine CoA transferase of E. coli (GenBank accession number CAA52112). Both subunits of succinyl-CoA:L-malate CoA transferase showed high amino acid sequence similarities to these enzymes with highest similarities to BbsF (amino acid identity and similarity, respectively: 30% and 46% to SmtA; 30% and 47% to SmtB). This indicates that the succinyl-CoA:L-malate CoA transferase also belongs to the family III transferases (Fig. 5).
Further experiments should give more evidences for this assignment. CoA transfer reactions imply that the high energy of the thioester bond is conserved in the catalytic cycle. This is only possible if the acyl or the CoA moiety is covalently linked to the enzyme via a reactive and energy-rich bond, e.g., as anhydride (family III) and thioester (family I, II), respectively. To test this hypothesis, we performed two types of experiments. We used L-malyl-CoA as the CoA donor and the substrate analogue L-citramalate as the potential CoA acceptor and tested by HPLC whether CoA transfer occurred. Indeed, L-citramalyl-CoA was formed. We also incubated L-malyl-CoA in the presence of L-[14C]malate and tested for the formation of L-[malyl-14C]CoA. This radioisotope-exchange reaction also occurred (data not shown), which is in line with the hypothesis, but could not definitely classify the succinyl-CoA:L-malate CoA transferase to a specific family of transferases.
Enzymes of family I use a ping-pong mechanism and are specifically inactivated by micromolar to millimolar concentrations of borohydride or hydroxylamine in the presence of the CoA donor. Enzymes of family II are subunits of complex lyases and contain a covalently bound (5'-phosphoribosyl)-3'-dephospho-CoA moiety. There is no indication for such a cofactor in succinyl-CoA:L-malate CoA transferase. Enzymes of family III seem to use a concerted mechanism via a ternary complex; this form is not very sensitive to borohydride and hydroxylamine. Therefore, we tested the effect of these compounds on succinyl-CoA:L-malate CoA transferase. Enzyme was incubated in the absence of CoA thioesters or in the presence of 2 mM succinyl-CoA for 10 min with 0.1 to 10 mM NaBH4 and with 10 mM hydroxylamine, respectively, and then assayed by the coupled spectrophotometric assay. No inactivation by 0.1 mM borohydride was recorded in the test with and without succinyl-CoA, and nearly 85% of the activity was retained when the enzyme was preincubated with 10 mM hydroxylamine for 10 min in the presence of 2 mM succinyl-CoA. This suggests that no reactive enzyme-CoA intermediate was present. Inactivation with 10 mM borohydride decreased activity to 28% in the presence of 2 mM succinyl-CoA.
Gene organization and cotranscription. The two genes encoding the subunits of succinyl-CoA:L-malate CoA transferase are located on a cluster of seven ORFs (Fig. 3). Cotranscription of genes smtA and smtB during autotrophic growth of C. aurantiacus was studied by performing RT-PCR experiments with mRNA isolated from autotrophically grown cells and comparing the results with results obtained with chromosomal DNA from C. aurantiacus (Fig. 6). The positive control, part of the gene for the -subunit of RNA polymerase, was amplified with both chromosomal DNA and cDNA. In contrast, the negative control, the intergenic region between adjacent orf1 and smtA (which are orientated in different directions), was negative with cDNA and positive with chromosomal DNA. The amplification of the intergenic region between smtA and smtB was positive in both instances. This indicates that smtA and smtB were cotranscribed and are part of the same operon.
DISCUSSION
Role of the enzyme. We have characterized a new CoA transferase whose function in autotrophic CO2 fixation of C. aurantiacus is to activate L-malate to its CoA thioester with succinyl-CoA as the CoA donor. The use of succinyl-CoA is an elegant means to couple the penultimate step of the cycle with a former one, making use of the high-energy thioester bond. In the next step, L-malyl-CoA is cleaved into glyoxylate and acetyl-CoA (13). Glyoxylate forms the primary net CO2 fixation product, and acetyl-CoA starts a new CO2 fixation cycle (15). The autotrophic 3-hydroxypropionate cycle is thus closed (Fig. 1). Mutant studies would be highly desirable in support of this proposed role but are difficult to perform due to the poorly developed genetic system and the difficult plating of this filamentous bacterium.
However, the following features corroborate the importance of succinyl-CoA:L-malate CoA transferase for autotrophic growth. (i) The specific enzyme activity in extracts of autotrophically grown cells amounted to 20 to 33 nmol min–1 mg protein–1. This in vitro rate is similar to the estimated minimal enzyme rate (12 nmol min–1 mg protein–1) that can explain the slow autotrophic growth (13). (ii) The 10- to 15-fold up-regulation under autotrophic conditions is in line with its function. (iii) The enzyme is highly specific for its CoA donor and acceptor, and the apparent Km values are in physiological range. (iv) The product of the reversible transferase reaction is specifically utilized by L-malyl-CoA/-methylmalyl-CoA lyase (13). This enzyme is also induced under autotrophic conditions, and the genes for these two proteins are located in the same gene cluster (Fig. 3). The fact that the enzyme also accepts L-citramalate may indicate that it may have two functions. The enzyme appears to be active with C4 dicarboxylic acids that have substitutions at C-2 provided the L conformation is given.
Succinyl-CoA:L-malate CoA transferase, a member of class III CoA transferases. CoA transferases catalyze the reversible transfer of CoA between organic acids and are known from many bacterial and eukaryotic species. Based on their reaction mechanism and amino acid sequences, the transferases can be divided into three distinct families (12). Family I enzymes consist of two dissimilar subunits in different aggregation states (22, 44). These enzymes operate by a ping-pong mechanism and can be specially inactivated by treatment with hydroxylamine or borohydride when they are preincubated with their CoA donor compound. Family II contains only a few members. The physiological substrate transferred by these enzymes is an enzyme-associated acyl carrier protein subunit.
The alignment of the two genes, smtA and smtB, coding for the subunits of the new CoA transferase shows the highest similarity to known members of CoA transferase family III. This family is listed as the "CaiB/BaiF family" in the paralogous protein family database and forms a distinct third branch in the phylogenetic tree of the family of CoA transferases, as deduced from sequence comparison (Fig. 5). The function of most members of the gene family is unknown. The currently characterized enzymes are involved mainly in anaerobic bacterial metabolism. They share considerable amino acid sequence similarity and may use a similar concerted reaction mechanism via a ternary complex. Yet the quaternary structures vary considerably, with the known members being either homodimers (10-11, 20, 24-25, 27, 30), heterodimers (22), or more complex structures (7). High similarity of SmtAB exists to the subunits BbsE and BbsF of the heterodimeric succinyl-CoA(R)-benzylsuccinate CoA transferase of T. aromatica (amino acid identity and similarity, respectively, of BbsF of 30% and 46% to SmtA and 30% and 47% to SmtB). Sct represents succinyl-CoA:D-citramalate CoA transferase from Chloroflexus aurantiacus, an enzyme required to convert glyoxylate back to acetyl-CoA via citramalate (S. Friedmann, B. Alber, and G. Fuchs, unpublished results). A highly conserved aspartate residue (Asp 169 in the CaiB nomenclature), which is located in the active site and binds the organic acid in an anhydride bond (11, 20, 30), is conserved in SmtA and SmtB. Other residues that are important for binding are conserved as well, such as Arg 16, Gly 37, Ala 38, Val 40, Asp 90, Leu 184, Thr 190, and Gly 193 (30), with the exception of His 185 (Fig. 7).
Succinyl-CoA:L-malate CoA transferase preincubated with succinyl-CoA was almost insensitive against hydroxylamine treatment, which would inactivate CoA transferases of family I and II. Furthermore, the enzyme was not inactivated with 0.1 mM NaBH4 after preincubation with succinyl-CoA. The decrease in activity observed at 10 mM NaBH4 cannot be explained, but similar results were reported for the related enzymes succinyl-CoA:(R)-benzylsuccinate CoA transferase from Thauera aromatica (22) and (E)-cinnamoyl-CoA:(R)-phenyllactate CoA transferase from Clostridium sporogenes (7). A partial inhibition could possibly be due to partial reduction of the postulated Asp169-anhydride bond (25) during the CoA transfer reaction.
The native enzyme may form a heterohexamer ()6 of 550 kDa ± 50 kDa. The higher oligomeric state of this enzyme compared to other members of this enzyme family may provide a greater structural stability of the enzyme at thermophilic growth conditions (6). None of the individually produced two subunits was active alone nor could activity be reconstituted by combination of the two proteins. Rather, only coexpression of both genes resulted in active recombinant enzyme that formed a 550-kDa complex. This observation is plausible in view of the structure of different CoA transferases (10-11, 20, 24-25, 30) in which the two subunits form a tightly interconnected complex. The CoA binding site is formed at the interface of the two subunits. Such an interlocked dimer can possibly only be developed when both subunits are simultaneously formed but not from prefolded individual subunits. The folding pathway that might lead to such a structure is still at issue.
One observation cannot be explained so far. Purified native enzyme from C. aurantiacus exhibited on SDS-PAGE one broad band of approximately 44 kDa which contained both subunits in equal amounts. This follows from N-terminal amino acid sequencing of this protein band. However, when the two genes were cloned together and expressed in E. coli, the active enzyme showed two distinct bands in SDS-PAGE which corresponded to the expected sizes of 44 and 46 kDa. This may indicate that the 46-kDa subunit of the native enzyme in C. aurantiacus is somehow modified, which results in an altered migration behavior in electrophoresis. One possibility is a C-terminal processing of the larger subunit in the organism but not in the expression system. The additional band on SDS-PAGE after purification of the natural enzyme migrates somewhat higher than the recombinant 46-kDa subunit and is assumed to represent an impurity.
Genes adjacent to the succinyl-CoA:L-malate CoA transferase genes (smtA and smtB) on the chromosome of C. aurantiacus. The genes for succinyl-CoA:L-malate CoA transferase are located together in a gene cluster of seven ORFs which may play a role in autotrophic CO2 fixation (accession number NZ_AAAH02000019, 19,200 to 30,600 bps) (Fig. 3). RT-PCR studies with mRNA of autotrophically grown C. aurantiacus cells showed that smtA and smtB form a transcriptional unit, as one would predict. The proteins encoded by orf2 and orf3 have no similarities to other proteins in the database, whereas the protein encoded by ctr3 is similar to SmtA and SmtB and encodes a putative CoA transferase (COG1804). The protein encoded by mcl is L-malyl-CoA/-methylmalyl-CoA lyase, which has been characterized before (13) and which catalyzes the next two steps in the pathway, i.e., cleavage of L-malyl-CoA and formation of -methylmalyl-CoA from glyoxylate and propionyl-CoA (Fig. 1). -Methylmalyl-CoA is dehydrated to mesaconyl-CoA in the second cycle of the 3-hydroxypropionate pathway used for glyoxylate assimilation (14). The gene mcd downstream of mcl codes for a member of the enoyl-CoA hydratase family (COG2030) and could, therefore, encode -methylmalyl-CoA dehydratase. A related protein is postulated in the acetate assimilation pathway of Rhodobacter sphaeroides and is thought to catalyze the hydration of mesaconyl-CoA to -methylmalyl-CoA (B. Alber and G. Fuchs, unpublished results).
Similarity of SmtA and SmtB. The two subunits of succinyl-CoA:L-malate CoA transferase are very similar to each other (Fig. 7) (59% amino acid sequence identity and 74% similarity). The high sequence similarity and the fact that the two genes are located side by side in the gene cluster could result from an ancient gene duplication. This holds true also for the two subunits BbsE and BbsF of the heterodimeric succinyl-CoA:(R)-benzylsuccinate CoA transferase of T. aromatica (22).
ACKNOWLEDGMENTS
Thanks are due to Nasser Gad'on and Christa Ebenau-Jehle for expert technical assistance.
REFERENCES
Alber, B. E., and G. Fuchs. 2002. Propionyl-coenzyme A synthase from Chloroflexus aurantiacus, a key enzyme of the 3-hydroxypropionate cycle for autotrophic CO2 fixation. J. Biol. Chem. 277:12137-12143.
Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1987. Current protocols in molecular biology. John Wiley and Sons, New York, N.Y.
Blankenship, R. E., M. T. Madigan, and C. E. Bauer. 1995. Anoxygenic photosynthetic bacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands.
Bradford, M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254.
Castenholz, R. W. 1969. Thermophilic blue-green algae and the thermal environment. Bacteriol. Rev. 33:476-504.
Dalhus, B., M. Saarinen, U. H. Sauer, P. Eklund, K. Johansson, A. Karlsson, S. Ramaswamy, A. Bjork, B. Synstad, K. Naterstad, R. Sirevag, and H. Eklund. 2002. Structural basis for thermophilic protein stability: structures of thermophilic and mesophilic malate dehydrogenases. J. Mol. Biol. 318:707-721.
Dickert, S., A. J. Pierik, D. Linder, and W. Buckel. 2000. The involvement of coenzyme A esters in the dehydration of (R)-phenyllactate to (E)-cinnamate by Clostridium sporogenes. Eur. J. Biochem. 267:3874-3884.
Eggerer, H., and C. H. Grunewlder. 1964. Zum Mechanismus der biologischen Umwandlung von Citronensure, IV. Synthese von Malyl-Coenzym A und seiner Diastereoisomeren. Liebigs Ann. Chem. 677:200-208.
Eisenreich, W., G. Strauss, U. Werz, G. Fuchs, and A. Bacher. 1993. Retrobiosynthetic analysis of carbon fixation in the phototrophic eubacterium Chloroflexus aurantiacus. Eur. J. Biochem. 215:619-632.
Gogos, A., J. Gorman, and L. Shapiro. 2004. Structure of Escherichia coli YfdW, a type III CoA transferase. Acta Crystallogr. Sect. D 60:507-511.
Gruez, A., V. Roig-Zamboni, C. Valencia, V. Campanacci, and C. Cambillau. 2003. The crystal structure of Escherichia coli YfdW gene product reveals a new fold of two interlaced rings identifying a wide family of CoA transferases. J. Biol. Chem. 278:34582-34586.
Heider, J. 2001. A new family of CoA transferases. FEBS Lett. 509:345-349.
Herter, S., A. Busch, and G. Fuchs. 2002. L-Malyl-coenzyme A lyase/-methylmalyl-coenzyme A lyase from Chloroflexus aurantiacus, a bifunctional enzyme involved in autotrophic CO2 fixation. J. Bacteriol. 184:5999-6006.
Herter, S., G. Fuchs, A. Bacher, and W. Eisenreich. 2002. A bicyclic autotrophic CO2 fixation pathway in Chloroflexus aurantiacus. J. Biol. Chem. 277:20277-20283.
Herter, S., J. Farfsing, N. Gad'on, C. Rieder, W. Eisenreich, A. Bacher, and G. Fuchs. 2001. Autotrophic CO2 fixation in Chloroflexus aurantiacus: study of glyoxylate formation and assimilation via the 3-hydroxypropionate cycle. J. Bacteriol. 183:4305-4316.
Holo, H. 1989. Chloroflexus aurantiacus secretes 3-hydroxypropionate, a possible intermediate in the assimilation of CO2 and acetate. Arch. Microbiol. 151:252-256.
Holo, H., and D. Grace. 1987. Polyglucose synthesis in Chloroflexus aurantiacus studied by 13C-NMR. Arch. Microbiol. 148:292-297.
Holo, H., and R. Sirevg. 1986. Autotrophic growth and CO2 fixation in Chloroflexus aurantiacus. Arch. Mirobiol. 145:173-180.
Hugler, M., C. Menendez, H. Schgger, and G. Fuchs. 2002. Malonyl-coenzyme A reductase from Chloroflexus aurantiacus, a key enzyme of the 3-hydroxypropionate cycle for autotrophic CO2 fixation. J. Bacteriol. 184:2404-2410.
Jonsson, S., S. Ricagno, Y. Lindqvist, and N. G. J. Richards. 2004. Kinetic and mechanistic characterization of the formyl-CoA transferase from Oxalobacter formigenes. J. Biol. Chem. 279:36003-36012.
Laemmli, U. K. 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227:680-685.
Leutwein, C., and J. Heider. 2001. Succinyl-CoA:(R)-benzylsuccinate CoA-transferase: an enzyme of the anaerobic toluene catabolic pathway in denitrifying bacteria. J. Bacteriol. 183:4288-4295.
Mohrhauer, H., K. Christiansen, M. Gan, M. Deubig, and R. T. Holman. 1968. Improved method for the preparation of malonyl coenzyme A. J. Lipid Res. 9:398-399.
Rangarajan, E. S., Y. Li, P. Iannuzzi, M. Cygler, and A. Matte. 2005. Crystal structure of Escherichia coli crotonobetainyl-CoA: carnitine CoA-transferase (CaiB) and its complexes with CoA and carnitinyl-CoA. Biochemistry 44:5728-5738.
Ricagno, S., S. Jonsson, N. Richards, and Y. Lindqvist. 2003. Formyl-CoA transferase encloses the CoA binding site at the interface of an interlcked dimer. EMBO J. 22:3210-3219.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
Sidhu, H., S. D. Ogden, H.-Y. Lung, B. G. Luttge, A. L. Baetz, and A. B. Peck. 1997. DNA sequencing and expression of the formyl-coenzyme A transferase gene, frc, from Oxalobacter formigenes. J. Bacteriol. 179:3378-3381.
Simon, E. 1957. S-succinyl coenzyme A. Biochem. Prep. 5:30-32.
Stadtman, E. R. 1957. Preparation and assay of acyl coenzyme A and other thiol esters; use of hydroxylamine. Methods Enzymol. 3:931-941.
Stenmaek, P., D. Gurmu, and P. Nordlund. 2004. Crystal structure of CaiB, a type-III CoA transferase in carnitine metabolism. Biochemistry 43:13996-14003.
Strauss, G., and G. Fuchs. 1993. Enzymes of a novel autotrophic CO2 fixation pathway in the phototrophic bacterium Chloroflexus aurantiacus, the 3-hydroxypropionate cycle. Eur. J. Biochem. 215:633-643.
Strauss, G., W. Eisenreich, A. Bacher, and G. Fuchs. 1992. 13C-NMR study of autotrophic CO2 fixation pathways in the sulfur-reducing archaebacterium Thermoproteus neutrophilus and in the phototrophic eubacterium Chloroflexus aurantiacus. Eur. J. Biochem. 205:853-866.
Studier, F. W., and B. A. Moffatt. 1986. Use of bacteriophage T7 RNA polymerase to direct selective high-level expression of cloned genes. J. Mol. Biol. 189:113-130.
Tabor, S., and C. C. Richardson. 1985. A bacteriophage T7 RNA polymerase/promoter system for controlled exclusive expression of specific genes. Proc. Natl. Acad. Sci. USA 82:1074-1078.
van der Meer, M. T. J., S. Schouten, J. W. de Leeuw, and D. M. Ward. 2000. Autotrophy of green non-sulphur bacteria in hot spring microbial mats: biological explanations for isotopically heavy organic carbon in the geological record. Environ. Microbiol. 2:428-435.
Zehr, B. D., T. J. Savin, and R. E. Hall. 1989. A one-step, low-background Coomassie staining procedure for polyacrylamide gels. Anal. Biochem. 182:157-159.(Silke Friedmann, Astrid S)