Eur. J. Biochem.

267, 37623769 (2000) q FEBS 2000

Characterization of a second methylene tetrahydromethanopterin dehydrogenase from Methylobacterium extorquens AM1

Christoph H. Hagemeier1, Ludmila Chistoserdova2, Mary E. Lidstrom2, Rudolf K. Thauer1 and Julia A. Vorholt1
1

Max-Planck-Institut fur terrestrische Mikrobiologie, Marburg, Germany; 2Department of Chemical Engineering, University of Washington, Seattle, WA, USA

Cell extracts of Methylobacterium extorquens AM1 were recently found to catalyze the dehydrogenation of methylene tetrahydromethanopterin (methylene H4MPT) with NAD1 and NADP1. The purification of a 32-kDa NADP-specific methylene H4MPT dehydrogenase (MtdA) was described already. Here we report on the characterization of a second methylene H4MPT dehydrogenase (MtdB) from this aerobic a-proteobacterium. Purified MtdB with an apparent molecular mass of 32 kDa was shown to catalyze the oxidation of methylene H4MPT to methenyl H4MPT with NAD1 and NADP1 via a ternary complex catalytic mechanism. The Km for methylene H4MPT was 50 mm with NAD1 (Vmax 1100 Umg21) and 100 mm with NADP1 (Vmax 950 Umg21). The Km value for NAD1 was 200 mm and for NADP1 20 mm. In contrast to MtdA, MtdB could not catalyze the dehydrogenation of methylene tetrahydrofolate. Via the N-terminal aminoacid sequence, the MtdB encoding gene was identified to be orfX located in a cluster of genes whose translated products show high sequence identities to enzymes previously found only in methanogenic and sulfate reducing archaea. Despite its location, MtdB did not show sequence similarity to archaeal enzymes. The highest similarity was to MtdA, whose encoding gene is located outside of the archaeal island. Mutants defective in MtdB were unable to grow on methanol and showed a pronounced sensitivity towards formaldehyde. On the basis of the mutant phenotype and of the kinetic properties, possible functions of MtdB and MtdA are discussed. We also report that both MtdB and MtdA can be heterologously overproduced in Escherichia coli making these two enzymes readily available for structural analysis. Keywords: C1-metabolism; methylotrophic bacteria; methanogenic archaea; tetrahydromethanopterin; tetrahydrofolate. The methylotrophic a-proteobacterium Methylobacterium extorquens AM1 possesses tetrahydrofolate (H4F) and dephospho tetrahydromethanopterin (H4MPT), two structural analogue C1 carriers [1,2]. In cell extracts of this aerobic a-proteobacterium H4MPT- and H4F-dependent enzymes were found. It was proposed that they participate in two separate C1 pathways, the one with the H4MPT-dependent enzymes being mainly involved in formaldehyde oxidation to CO2 [1,2]. Three H4MPT-dependent enzymes from M. extorquens AM1 have been purified until now and characterized: methenyl H4MPT cyclohydrolase (MchA) [3], formylmethanofuran: H4MPT formyltransferase (FfsA) (B. K. Pomper and J. A. Vorholt, unpublished results) and a NADP-specific methylene H4MPT dehydrogenase (MtdA) [2]. MchA and FfsA were found to have high sequence similarity to the respective enzymes from methanogenic and sulfate reducing archaea and their encoding genes were shown to be located within a cluster of genes whose deduced gene products show, with an exception of OrfX, high sequence similarity to proteins found until now
Correspondence to J. A. Vorholt, Max-Planck-Institut fur terrestrische Mikrobiologie, Karl-von-Frisch-Strasse, D-35043 Marburg, Germany. Fax: 1 49 6421178299, Tel.: 1 49 6421178331, E-mail: vorholt@mailer.uni-marburg.de Abbreviations: MtdA, NADP-dependent methylene tetrahydromethanopterin dehydrogenase; MtdB, NAD(P)-dependent methylene tetrahydromethanopterin dehydrogenase; H4MPT, tetrahydromethanopterin; H4F, tetrahydrofolate. Note: web page available at http://www.uni-marburg.de/mpi (Received 17 February 2000, accepted 17 April 2000)

only in methanogenic and sulfate reducing archaea [1]. The gene encoding MtdA is not located within this gene cluster and MtdA does not show significant sequence similarity at the protein level to archaea specific F420-dependent or H2-forming methylene H4MPT dehydrogenases [4]. MtdA shows only minor sequence identity (, 15%) to methylene H4F dehydrogenases from bacteria and eucarya [59], which is consistent with the fact that MtdA from M. extorquens AM1 exhibits activity with methylene H4F, although the catalytic efficiency of this reaction is 20-fold lower than the reaction with methylene H4MPT. Cell extracts of M. extorquens AM1 do not only catalyze the dehydrogenation of methylene H4MPT with NADP1 but also with NAD1. As purified MtdA is NADP-specific this finding indicates the presence of a second methylene H4MPT dehydrogenase. This enzyme, designated MtdB, was now purified, its encoding gene identified as orfX and its function deduced from the phenotype of orfX insertion mutants. NAD- and NADP-dependent methylene H4MPT dehydrogenase activities were also found in all methylotrophic and methanotrophic proteobacteria analyzed that assimilate formaldehyde by the serine or ribulose monophosphate pathway as well as autotrophic Xanthobacter strains [10]. Detailed analysis of these enzymes in M. extorquens AM1 might therefore help to understand C1 metabolism in other methylotrophic bacteria.

M AT E R I A L S A N D M E T H O D S
Tetrahydromethanopterin (H4MPT) and methenyl H4MPT were purified from Methanobacterium thermoautotrophicum strain

q FEBS 2000

Dehydrogenase from M. extorquens AM1 (Eur. J. Biochem. 267) 3763

Marburg (DSM 2133) [11]. Tetrahydrofolic acid (H4F) was from Sigma. Stock solutions of H4MPT and H4F were prepared in anoxic 120 mm potassium phosphate buffer pH 6.0 containing 2 mm dithiothreitol. Methylene H4MPT and methylene H4F were generated from H4MPT and H4F and the addition of excess of formaldehyde by spontaneous reaction. M. extorquens AM1 (pALS8) is the strain described in Chistoserdova et al. [1]. Growth of M. extorquens AM1 M. extorquens AM1 was grown on methanol (100 mm) at 30 8C in the minimal medium described previously [12]. The cells were cultivated in 10 L glass fermenters containing 8 L medium. The fermenters were inoculated with 800 mL of a culture grown in Erlenmeyer bottles. The glass fermenter was stirred with 500 r.p.m. and gassed with air (2 Lmin21). For the cultivation of M. extorquens AM1 (pALS8) [1] tetracycline (10 mgmL21 final concentration) was added to the medium. The cultures were harvested in the late-exponential phase at an optical density of D600 4.0 and harvested by centrifugation at 5000 g. Cell pellets were stored at 220 8C. Preparations of cell extracts and determination of enzyme activities Frozen cells of M. extorquens AM1 and Escherichia coli, respectively, were suspended in 50 mm Mops/KOH pH 7.0 at 4 8C and passed three times through a French pressure cell at 1.2 108 Pa. Centrifugation was performed at 150 000 g for 1 h to remove cell debris, whole cells and the membrane fraction which was shown not to contain NAD(P)-dependent methylene H4MPT dehydrogenase activity. Protein was determined by Bradford assay using the Bio-Rad reagent with bovine serum albumin as a standard. Enzyme assays were performed routinely at 30 8C in 1-mL cuvettes (d 1 cm) as described by Vorholt et al. [2]. Purification of the NAD(P)-dependent methylene H4MPT dehydrogenase from M. extorquens AM1 NAD(P)-dependent methylene H4MPT dehydrogenase was purified from M. extorquens AM1 or M. extorquens AM1 (pALS8) by five chromatographic steps. All chromatographic material was from Amersham Pharmacia Biotech. Ultracentrifugation supernatant (15 mL) was applied to a DEAESephacel column (2.6 10 cm) equilibrated with 50 mm Mops/KOH pH 7.0. Protein was eluted with an increasing NaCl gradient: 50 mL 160 mm NaCl and 250 mL 160 mm to 400 mm in 50 mm Mops/KOH pH 7.0. The NAD(P)-dependent methylene H4MPT dehydrogenase activity eluted at about 250 mm NaCl. Combined active fractions (43 mL) were diluted 1 : 3 in 50 mm Mops/KOH pH 7.0 and subjected to a Q Sepharose column (High Performance 1.6 10 cm) equilibrated with 50 mm Tricine/KOH pH 7.5. Protein was eluted with an increasing NaCl gradient from 160 mm to 400 mm in 200 mL. Active fractions were recovered at about 280 mm NaCl. To the combined active fractions (29 mL) ammonium sulfate was added (1 m final concentration) and stirred for one hour. Precipitated protein was removed by centrifugation at 27 000 g. The supernatant was loaded on a Source 15 Phe column (2.6 10 cm) equilibrated with 1 m ammonium sulfate in Tris/HCl pH 7.5. With a linear gradient decreasing from 1 m to 0 m ammonium sulfate (160 mL), the dehydrogenase was eluted at about 10 mm. Active fractions (16 mL)

were diluted 1 : 3 with 10 mm potassium phosphate buffer pH 7.0 and subjected to chromatography on hydroxyapatite (1.6 10 cm) equilibrated in 10 mm potassium phosphate buffer pH 7.0. MtdB was eluted at 60 mm potassium phosphate within a linear increasing gradient from 10 mm to 125 mm potassium phosphate. Active fractions (11 mL) were pooled, washed, and concentrated by using Centricon 30 microconcentrators (Millipore) and 50 mm MES/NaOH pH 5.0. The enzyme was further purified by cation-exchange chromatography on a MonoS column (0.5 5 cm) via a linear increasing gradient from 0 to 240 mm NaCl in 40 mL. The dehydrogenase was recovered at 50 mm NaCl in 8 mL and concentrated by Centricon 30 microconcentrators. Determination of the N-terminal amino-acid sequence After electrophoresis of purified enzyme in SDS/PAGE, the 32-kDa band was electroblotted onto a poly(vinyl trifluoride) membrane (Applied Biosystems). The N-terminal sequence determination was performed on a 477 protein/peptide sequencer from Applied Biosystems by D. Linder, University of Giessen. Heterologous expression of the mtdB and the mtdA genes in E. coli For incorporation of mtdB and mtdA into the expression vector pET17b, mtdB and mtdA were amplified by PCR using chromosomal DNA from M. extorquens AM1 as a template [13]. For amplification of mtdB, the following two primers were used: 5 H -GAGAGGACGAAGCATATGGCCCGCTCGATC-3 H (sense) and 5 H -CGGAAGCGGGCTCGAGTTCGGACTCATCCGG-3 H (antisense). The sense primer was designed with a NdeI restriction site such that the amplified gene started with the ATG start codon and the antisense primer was designed with a XhoI restriction site located a few nucleotides downstream of the stop codon. For amplification, Expand-DNApolymerase (Roche Diagnostics) was used. The PCR product was digested with NdeI and XhoI and ligated into the pET17b expression vector previously digested with the same restriction enzymes. The construct thus obtained was designated pCH1 and amplified in E. coli TOP10 (Invitrogen). pCH1 was then used to transform E. coli BL21 (DE3) pLysS. Sequencing of the mtdB gene cloned into pET17b revealed no mutation. For expression of the mtdB gene in E. coli BL21 (DE3) pLysS (pCH1) the transformed cells were aerobically grown at 37 8C in 1 L minimal medium M9 [13] supplemented with ampicillin (150 mgmL21) and chloramphenicol (34 mgmL21) at 37 8C. When the D600 of the culture reached 0.5, cells were induced by IPTG (1 mm final concentration). After 4 h, the cells were harvested by centrifugation at 4200 g at 4 8C, yielding 3.5 g cells (wet mass). For amplification of mtdA, the following two primers were used: 5 H -GCCAGAGGACATATGTCCAAGAAGCTGCTC-3 H (sense) and 5 H -TTTGGGCCGGAATTCTCAGGCCATTTCCTTGGC-3 H (antisense). The sense primer was designed with a NdeI restriction site, the antisense primer with an EcoRI restriction site. The restriction sites were used to clone the amplified PCR product into pET17b. Cloning and expression of the mtdA gene was performed as described for the mtdB gene and E. coli BL21 (DE3) pLysS pCH2 was thus obtained. Purification of MtdB and MtdA from E. coli For purification of MtdB and MtdA heterologously overproduced in E. coli, the same purification protocol was used as

3764 C. H. Hagemeier et al. (Eur. J. Biochem. 267)

q FEBS 2000

described (for MtdB see above, for MtdA see Vorholt et al. [2]). For both enzymes the last chromatographic step, chromatography on MonoS, was not necessary. Generation of double mtdB-mxaF mutants Generation of insertion mutants in mtdB via marker-exchange technique was described earlier [1]. In this work, the same mutation was generated in the mxaF-background using a point mutant in mxaF, UV26 [14] as a parental strain. Mutants were selected on succinate plates protected from methanol vapors and then checked for growth on succinate in the presence of methanol. Introduction of mau genes in trans To obtain a derivative of mtdB mutant capable of methylamine oxidation in the presence of succinate, a cluster of genes encoding subunits of methylamine dehydrogenase and other genes involved in oxidation of methylamine (mau genes [15]), was cloned into mtdB mutant in trans, using pAYC269a (A.Y. Chistoserdov, unpublished results). Constitutive expression of mau genes in this construct is driven from the lac promoter of pRK310 [16].

R E S U LT S
Cell extracts of M. extorquens AM1 grown on methanol catalyzed the reduction of NADP1 with methylene H4MPT (2.6 Umg21) and the reduction of NAD1 with methylene H4MPT (0.6 Umg21). The NADP-specific methylene H4MPT dehydrogenase (MtdA) could be separated from the NADdependent methylene H4MPT dehydrogenase activity by chromatography on DEAE/Sephacel. The NAD-dependent enzyme was purified in principle via the procedure described in Table 1 and identified via its N-terminal amino-acid sequence as the orfX gene product. For routine purification of the NAD-dependent methylene H4MPT dehydrogenase, M. extorquens AM1 (pALS8) was used which carries a plasmid containing a genomic region including orfX [1]. Cell extract of M. extorquens AM1 (pALS8) showed a fivefold increase in the NAD-dependent methylene H4MPT dehydrogenase activity (3 Umg21). Purification and molecular properties of MtdB The NAD-dependent methylene H4MPT dehydrogenase activity was purified by five chromatographic steps. From 5 g of wet cells of M. extorquens AM1 (pALS8) the enzyme was enriched 144-fold with a yield of 8% to a specific activity of 460 Umg21 under standard assay conditions using NAD1 as the cofactor (Table 1). The enzyme could be purified and stored

Fig. 1. SDS/PAGE analysis of purified NAD(P)-dependent methylene H4MPT dehydrogenase (MtdB) from M. extorquens AM1. Proteins were separated on a 12% polyacrylamide gel, which was subsequently stained with Coomassie brilliant blue R250 [26]. Lane M, molecular mass standards (Amersham Pharmacia Biotech); lane B, 3 mg of purified MtdB.

at 220 8C under aerobic conditions. Half of the activity was lost after three weeks of storage. After the fifth chromatographic step, the preparation contained only one polypeptide, with an apparent molecular mass of 32 kDa, as revealed by SDS/PAGE (Fig. 1). The N-terminal amino-acid sequence was determined to be MARSILHMLTPLKHMS which is identical to that of the predicted orf X gene product. The gene orf X was therefore redesignated mtdB. The amino-acid sequence of MtdB is 30% identical to the NADP-specific methylene H4MPT dehydrogenase (MtdA) of M. extorquens AM1 [2]. An alignment of the two NAD(P)-dependent methylene H4MPT dehydrogenases from M. extorquens AM1 is shown in Fig. 2. Gel filtration (Superdex 200) of purified MtdB as well as native gel electrophoresis revealed an apparent molecular mass of the enzyme of about 180 kDa (data not shown) indicating a homohexameric structure of the native enzyme. The UV/visible spectrum of the enzyme was that of a protein lacking a chromophoric prosthetic group (data not shown). The :280 determined was within a range of 5% identical to that calculated from the tyrosine, phenylalanine, and tryptophan content of the enzyme.

Table 1. Purification of the NAD(P)-dependent methylene H4MPT dehydrogenase, MtdB, from M. extorquens AM1 (pALS8) grown on methanol. Activities were determined under standard-assay conditions as described in Materials and methods using NAD1 as cosubstrate. Purification step Cell extract DEAE-Sephacel Q-Sepharose Source 15 Phe Hydroxyapatite Mono S Protein (mg) 281 57 4.2 1.2 0.5 0.15 Activity (U) 893 790 435 253 205 69 Specific activity (Umg21) 3.2 13.9 104 211 410 460 Purification (-fold) 1 4 33 66 117 144 Yield (%) 100 89 49 28 23 8

q FEBS 2000

Dehydrogenase from M. extorquens AM1 (Eur. J. Biochem. 267) 3765

Catalytic properties of MtdB The pH optimum range for the dehydrogenation of methylene H4MPT was at about pH 6.0, the optimum temperature was at 3540 8C. In contrast to MtdA, MtdB did not catalyze the dehydrogenation of methylene H4F. Both NAD1 and NADP1 served as cosubstrates. The dependence of the rate of methylene H4MPT dehydrogenation with NAD1 and NADP1 on the concentrations of the substrates was examined (Fig. 3). Reciprocal plots of the initial rates vs. the concentration of one substrate at different concentrations of the second substrate yielded straight lines (not shown). For methylene H4MPT, Km values of 50 mm (reaction with NAD1) and 100 mm (reaction with NADP1) were determined. The Km value for NAD1 was 200 mm and for NADP1 20 mm. The Vmax value was 1100 Umg21 for the

reaction with NAD1 as cosubstrate and 950 Umg21 for the reaction with NADP1 as cosubstrate. The plots of the initial rates (Fig. 3) favor a ternary complex catalytic mechanism which is supported by the finding that MtdB does not possess a redox active chromophoric group. For determination of the Km values the assays were started by the addition of NAD1 or NADP1, respectively. As expected, the initial rate of methylene H4MPT dehydrogenation with NAD1 was independent of whether the reaction was started by the addition of methylene H4MPT or NAD1 to the assay mixture. Surprisingly, the initial rate of methylene H4MPT dehydrogenation with NADP1 was dependent on the order of substrate addition. Reactions started with NADP1 were faster than reactions started with either methylene H4MPT or enzyme. An explanation for this finding at this time is difficult and may have to do with conformational changes of the enzyme.

Fig. 2. Alignment of the amino-acid sequences of the NAD(P)-dependent methylene H4MPT dehydrogenases, MtdA and MtdB, from M. extorquens AM1.

3766 C. H. Hagemeier et al. (Eur. J. Biochem. 267)

q FEBS 2000

The addition of NADPH to the enzymatic reaction catalyzed by MtdB indicated a noncompetitive inhibition. The kinetics of methylene H4MPT dehydrogenation with NAD1 and NADP1 in the presence of different concentrations of NADPH is shown in Fig. 4. NADH caused no inhibitory effect on MtdB activity. Purified MtdB did not catalyze the oxidation of methylene H4MPT with the reduction of flavin adenine dinucleotide, flavin mononucleotide, coenzyme F420, or dyes such as methylene blue or benzyl viologen as cosubstrates. It did not exhibit methenyl H4MPT cyclohydrolase activity. Analysis of mtdB insertion mutants Mutation of mtdB (formerly orf X) encoding the NAD(P)dependent methylene H4MPT dehydrogenase via insertion of a kanamycin-resistance cassette has been described previously [1]. These mutants lacking MtdB activity have lost their ability to grow on methanol, but were still able to grow on succinate. This finding indicates a specific involvement of MtdB in the oxidation of formaldehyde to CO2 upon growth on methanol.

As formaldehyde is a necessary but toxic intermediate during growth on C1 compounds, mtdB mutants were analyzed with respect to formaldehyde resistance in more detail. It was found that mutants in mtdB were highly sensitive to formaldehyde. Whereas wild-type M. extorquens AM1 could grow on succinate in the presence of 2.5 mm formaldehyde, mtdB mutants were not able to grow on succinate in the presence of more than 0.5 mm formaldehyde. Growth of mtdB mutants was also inhibited on plates containing succinate plus methanol. To test if methanol itself was toxic, or formaldehyde produced as a result of methanol oxidation by methanol dehydrogenase, double mutants were generated in mtdB and mxaF. Mutants in mxaF encoding the large subunit of methanol dehydrogenase were not able to oxidize methanol to formaldehyde. The double mtdBmxaF mutants could grow on succinate in the presence of methanol, indicating that not methanol itself but formaldehyde is the toxic compound. Another formaldehyde-producing substrate, methylamine did not affect growth of mtdB mutants on succinate. The methylamine oxidation system, however, is

Fig. 3. Kinetics of methylene H4MPT dehydrogenation with NAD1 and NADP1 catalyzed by the NAD(P)-dependent methylene H4MPT dehydrogenase (MtdB) from M. extorquens AM1. Assays were performed at 30 8C in 120 mm KPP pH 6.0, the NAD(P)1 and methylene H4MPT concentrations are indicated, 0.13 mg of purified MtdB was added. All assays were started by the addition of NAD1 and NADP1, respectively.

q FEBS 2000

Dehydrogenase from M. extorquens AM1 (Eur. J. Biochem. 267) 3767

Fig. 4. Kinetics of methylene H4MPT dehydrogenation with NAD1 and NADP1 in the presence of different concentrations of NADPH. For assay conditions see Fig. 3.

repressed by succinate (L. Chistoserdova and M. E. Lidstrom, unpublished results). Therefore, in the presence of succinate no formaldehyde may be produced from methylamine. To test if this is the case, we introduced a cluster of mau genes from M. extorquens AM1 constitutively expressed from the lac promoter into mtdB. In such constructs, methylamine dehydrogenase activity was expressed during growth on succinate (data not shown), and these acquired methylamine sensitivity. These experiments show that MtdB is important also for heterotrophic growth when external formaldehyde is present or formaldehyde is formed from C1 substrates. Expression of mtdB and mtdA in E. coli and purification of the gene products To facilitate purification of MtdB and to enable structural/ functional analysis in the future, mtdB was expressed in E. coli BL21 (DE3) pLysS using pET17b. Cell extracts of IPTG-induced E. coli (pCH1) cells were found to contain high concentrations of a 32-kDa protein and to exhibit NAD(P)-dependent methylene H4MPT dehydrogenase activities of 89 Umg21 which corresponds to a 200-fold overexpression in comparison to M. extorquens AM1 wild type cells. MtdB was purified eightfold with a yield of 45% by the use of four chromatographic steps. The formation of active NAD(P)-dependent methylene H4MPT dehydrogenase indicates that the recombinant enzyme folded correctly in E. coli. The specific activity of the overproduced purified enzyme was 680 Umg21. The mtdA gene was also cloned into pET17b and actively expressed in E. coli. Cell extracts were found to exhibit NADP-dependent methylene H4MPT dehydrogenase activity of as much as 370 Umg21. The purification was performed by ammonium sulfate precipitation and two chromatographic steps. MtdA was purified with a yield of 44%. From about 2.5 g of wet cells 20 mg of purified MtdA could be obtained which had a specific activity of 1100 Umg21.

the NADP-dependent methylene H4MPT dehydrogenase, MtdA from the same organism (Table 2). Both enzymes catalyze the dehydrogenation of methylene H4MPT to methenyl H4MPT via a ternary complex catalytic mechanism and both are composed of only one type of subunits with a molecular mass of 32 kDa. The sequence identity is 30%.
Table 2. Comparison of properties of MtdA and MtdB. ND, not detectable. Parameter Apparent molecular mass (kDa) Native enzyme Monomer Vmax (Umg21) Methylene H4MPT as substrate NAD1 as cosubstrate NADP1 as cosubstrate Methylene H4F as substrate NAD1 as cosubstrate NADP1 as cosubstrate Km for methylene H4MPT (mm) NAD1 as cosubstrate NADP1 as cosubstrate Km for methylene H4F (mm) NADP1 as cosubstrate Km for NADP (mm) Methylene H4F as substrate Methylene H4MPT as substrate Km for NAD1 (mm) Methylene H4MPT as substrate Inhibition by NADPH NADH pH-optimum Temperature optimum (8C) Isoelectric point Chromophoric prosthetic group
1

MtdA

MtdB

90 32

180 32

ND 600 ND 30 ND 20 30 10 30 ND No No 6.0 45 7.2 No

1100 950 ND ND 50 100 ND ND 20 200 Yes No 6.0 3540 5.4 No

DISCUSSION
The NAD(P)-dependent methylene H4MPT dehydrogenase, MtdB, from M. extorquens AM1 shares many properties with

3768 C. H. Hagemeier et al. (Eur. J. Biochem. 267)

q FEBS 2000

The dehydrogenation of methylene H4MPT to methenyl H4MPT coupled with the reduction of pyridine nucleotides in vivo probably is an irreversible reaction; the DG8 of the reaction is 213 kJmol21. The thermodynamics thus predict that MtdA and MtdB are involved in formaldehyde oxidation to CO2. An important question is why two different pyridinenucleotide dependent methylene H4MPT dehydrogenases are present in M. extorquens AM1. To obtain indications for the in vivo function of the two enzymes, kinetic analysis and the characterization of mutant phenotypes were performed. The two enzymes have different although overlapping substrate specificities. MtdA catalyzes the dehydrogenation of methylene H4MPT and methylene H4F with NADP1 as a cosubstrate. MtdB is specific for methylene H4MPT on one hand and uses NAD1 and NADP1 as cosubstrates on the other (Table 2). The Km-value of MtdB for NAD1 (200 mm) in comparison to NADP1 (20 mm) favors the possibility that MtdB is a NADP-dependent enzyme at first sight. The lower Km values of MtdA for NADP1 argues that MtdA is the main enzyme for the oxidation of methylene H4MPT to methenyl H4MPT. However, it seems to be important also to consider the concentration of the oxidized pyridine nucleotides in the cell. If MtdB would be a NADP-dependent enzyme in vivo the question arises, why, during a presumably catabolic reaction, NADPH is produced. For a direct re-oxidiation via NADH:Ubiquinone-oxidoreductase (complex I) [17,18] a production of NADH appears more likely. At this time we can not exclude the possibility that NADPH could also be oxidized by complex I or that a transhydrogenase could shuffle the electrons from NADPH to NAD1 before the transfer to complex I. Such a transhydrogenase was found for instance in Pseudomonas fluorescens [19] and also in other organisms [20]. For Saccharomyces cerevisiae a NAD1/NADP1 ratio of 10 : 1 was shown [21]. If it is also true that NAD1 concentration in M. extorquens AM1 is an order of magnitude higher than the concentration of NADP1, it might be that the reduction of NADP1 catalyzed by MtdA and MtdB consumes just a small part of the methylene H4MPT. According to this assumption, the main part of the reducing equivalents during this redox reaction might be obtained as NADH. It therefore seems also reasonable that the Km for NAD1 is higher than the Km for NADP1 so that NADP1 reduction to NADPH is not prevented, which is important for various biosynthetic pathways and polyhydroxybutyrate production [22]. MtdA seems to be an indipsensable enzyme in M. extorquens AM1 as no null-mutants could be generated in mtdA upon growth on succinate [23]. Single-crossover mutants in mtdA which have a lower level of MtdA-activity in comparison to wild type were not able to grow on methanol. Obviously, MtdA has an essential function in central metabolism of M. extorquens AM1. This might be due to the fact that MtdA has dual specificity with respect to the pterin cofactor and also catalyzes the reversible dehydrogenation of methylene H4F [2]. No H4F-specific methylene H4F dehydrogenase was found in M. extorquens AM1 so far. For this reason we can not exclude that MtdA is the only enzyme in M. extorquens AM1 that catalyzes the reversible dehydrogenation of methylene H4F to methenyl H4F, a reaction involved in basic cell metabolism, including purine biosynthesis [24]. Interestingly, mtdA is located 88 nucleotides upstream of fchA (formerly orf 4) [25] which encodes a methenyl H4F cyclohydrolase [3]. Null mutants in mtdB are readily obtained [1]. They are not able to grow on methanol and show a pronounced sensitivity towards formaldehyde (see above). This indicates that the

function of MtdB for growth on methanol could not be taken over by MtdA and that MtdB has a role also in formaldehyde detoxification. Possibly the reduction of NAD1 with methylene H4MPT is the main function of MtdB and the observed sensitivity towards formaldehyde might be a consequence of the missing NAD-dependent methylene H4MPT activity.

ACKNOWLEDGEMENTS
This work was supported by the Max-Planck-Gesellschaft and by an NIH grant (GM36296) to MEL. We would like to thank Johanna Moll, MPI Marburg, for the preparation of H4MPT.

REFERENCES
1. Chistoserdova, L., Vorholt, J.A., Thauer, R.K. & Lidstrom, M.E. (1998) C1 transfer enzymes and coenzymes linking methylotrophic bacteria and methanogenic archaea. Science 281, 99102. 2. Vorholt, J.A., Chistoserdova, L., Lidstrom, M.E. & Thauer, R.K. (1998) The NADP-dependent methylene tetrahydromethanopterin dehydrogenase in Methylobacterium extorquens AM1. J. Bacteriol. 180, 53515356. 3. Pomper, B.K., Vorholt, J.A., Chistoserdova, L., Lidstrom, M.E. & Thauer, R.K. (1999) A methenyl tetrahydromethanopterin cyclohydrolase and a methenyl tetrahydrofolate cyclohydrolase in Methylobacterium extorquens AM1. Eur. J. Biochem. 261, 475480. 4. Thauer, R.K. (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144, 23772406. 5. Uyeda, K. & Rabinowitz, J.C. (1967) Enzymes of clostridial purine fermentation. Methylenetetrahydrofolate dehydrogenase. J. Biol. Chem. 242, 43784385. 6. Moore, M.R., O'Brien, W.E. & Ljungdahl, L.G. (1974) Purification and characterization of nicotinamide adenine dinucleotide-dependent methylenetetrahydrofolate dehydrogenase from Clostridium formicoaceticum. J. Biol. Chem. 249, 52505253. 7. Ragsdale, S.W. & Ljungdahl, L.G. (1984) Purification and properties of NAD-dependent 5,10-methylenetetrahydrofolate dehydrogenase from Acetobacterium woodii. J. Biol. Chem. 259, 34993503. 8. Barlowe, C.K. & Appling, D.R. (1990) Isolation and characterization of a novel eukaryotic monofunctional NAD(1)-dependent 5,10-methylenetetrahydrofolate dehydrogenase. Biochemistry 29, 70897094. 9. Wohlfarth, G., Geerligs, G. & Diekert, G. (1991) Purification and characterization of NADP(1)-dependent 5,10-methylenetetrahydrofolate dehydrogenase from Peptostreptococcus productus Marburg. J. Bacteriol. 173, 14141419. 10. Vorholt, J.A., Chistoserdova, L., Stolyar, S.M., Thauer, R.K. & Lidstrom, M.E. (1999) Distribution of tetrahydromethanopterindependent enzymes in methylotrophic bacteria and phylogeny of methenyl tetrahydromethanopterin cyclohydrolases. J. Bacteriol. 181, 57505757. 11. Breitung, J., Borner, G., Scholz, S., Linder, D., Stetter, K.O. & Thauer, R.K. (1992) Salt dependence, kinetic properties and catalytic mechanism of N-formylmethanofuran: tetrahydromethanopterin formyltransferase from the extreme thermophile Methanopyrus kandleri. Eur. J. Biochem. 210, 971981. 12. Fulton, G.L., Nunn, D.N. & Lidstrom, M.E. (1984) Molecular cloning of a malyl coenzyme A lyase gene from Pseudomonas sp. strain AM1, a facultative methylotroph. J. Bacteriol. 160, 718723. 13. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA. 14. Nunn, D.N. & Lidstrom, M.E. (1986) Isolation and complementation analysis of 10 methanol oxidation mutant classes and identification of the methanol dehydrogenase structural gene of Methylobacterium sp. strain AM1. J. Bacteriol. 166, 581590. 15. Chistoserdov, A.Y., Chistoserdova, L.V., McIntire, W.S. & Lidstrom, M.E. (1994) Genetic organization of the mau gene cluster in

q FEBS 2000

Dehydrogenase from M. extorquens AM1 (Eur. J. Biochem. 267) 3769

nucleotide transhydrogenase in Saccharomyces cerevisiae and its effect on product formation during anaerobic glucose fermentation. Appl. Environ. Microbiol. 65, 23332340. Korotkova, N.A., Doronina, N.V. & Trotsenko, Y.A. (1999) Biosynthesis of the copolymer of 3-hydroxybutyrate and 3hydroxyvalerate in Methylobacterium extorquens. Metabolism of propanol., propionate, pentanol, and valerate. Mikrobiologiya (Engl. Transl.) 68, 296303. Chistoserdova, L.V. & Lidstrom, M.E. (1994) Genetics of the serine cycle in Methylobacterium extorquens AM1: identification of sgaA and mtdA and sequences of sgaA, hprA, and mtdA. J. Bacteriol. 176, 19571968. MacKenzie, R.E. (1984) Biogenesis and interconversion of substituted tetrahydrofolates. In Folates and Pterins. Chemistry and Biochemistry of Folates (Blakley, R.L. & Benkovic, S.J., eds), pp. 256306. John Wiley & Sons, New York, USA. Chistoserdova, L.V. & Lidstrom, M.E. (1994) Genetics of the serine cycle in Methylobacterium extorquens AM1: Identification, sequence, and mutation of three new genes involved in C1 assimilation, orf4, mtkA, and mtkB. J. Bacteriol. 176, 73987404. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685.

16.

17. 18. 19.

20. 21.

Methylobacterium extorquens AM1: complete nucleotide sequence and generation and characteristics of mau mutants. J. Bacteriol. 176, 40524065. Ditta, G., Schmidhauser, T., Yakobson, E., Lu, P., Liang, X.W., Finlay, D.R., Guiney, D. & Helinski, D.R. (1985) Plasmids related to the broad host range vector, pRK290, useful for gene cloning and monitoring gene expression. Plasmid 13, 149153. Saraste, M. (1999) Oxidative phosphorylation at the fin de siecle. Science 283, 14881493. Yagi, T., Yano, T., Di Bernardo, S. & Matsuno-Yagi, A. (1998) Procaryotic complex I (NDH-1), an overview. Biochim. Biophys. Acta. 1364, 125133. French, C.E., Boonstra, B., Bufton, K.A.J. & Bruce, N.C. (1997) Cloning, sequence, and properties of the soluble pyridine nucleotide transhydrogenase of Pseudomonas fluorescens. J. Bacteriol. 179, 27612765. Voordouw, G., van der Vies, S.M. & Themmen, A.P.N. (1983) Why are two different types of pyridine nucleotide transhydrogenase found in living organisms? Eur. J. Biochem. 131, 527533. Anderlund, M., Nissen, T.L., Nielsen, J., Villadsen, J., Rydstrom, J., Hahn-Hagerdal, B. & Kielland-Brandt, M.C. (1999) Expression of the Escherichia coli pntA and pntB genes, encoding nicotinamide

22.

23.

24.

25.

26.