The NADP-Dependent Methylene Tetrahydromethanopterin Dehydrogenase in Methylobacterium extorquens AM1

Abstract

An NADP-dependent methylene tetrahydromethanopterin (H₄MPT) dehydrogenase has recently been proposed to be involved in formaldehyde oxidation to CO₂ in Methylobacterium extorquens AM1. We report here on the purification of this novel enzyme to apparent homogeneity. Via the N-terminal amino acid sequence, it was identified to be the mtdA gene product. The purified enzyme catalyzed the dehydrogenation of methylene H₄MPT with NADP⁺ rather than with NAD⁺, with a specific activity of approximately 400 U/mg of protein. It also catalyzed the dehydrogenation of methylene tetrahydrofolate (methylene H₄F) with NADP⁺. With methylene H₄F as the substrate, however, the specific activity (26 U/mg) and the catalytic efficiency (V_max/K_m) were approximately 20-fold lower than with methylene H₄MPT. Whereas the dehydrogenation of methylene H₄MPT (E₀ = −390 mV) with NADP⁺ (E₀ = −320 mV) proceeded essentially irreversibly, the dehydrogenation of methylene H₄F (E₀ = −300 mV) was fully reversible. Comparison of the primary structure of the NADP-dependent dehydrogenase from M. extorquens AM1 with those of methylene H₄F dehydrogenases from other bacteria and eucarya and with those of methylene H₄MPT dehydrogenases from methanogenic archaea revealed only marginally significant similarity (<15%).

Methylobacterium extorquens AM1 is an aerobic methylotrophic bacterium which belongs to the α-subgroup of the proteobacteria (22, 30). In this organism methanol has been proposed to be metabolized to CO₂ via formaldehyde, N⁵,N¹⁰-methylene tetrahydrofolate (methylene H₄F), N⁵,N¹⁰-methenyl tetrahydrofolate (methenyl H₄F), N¹⁰-formyl tetrahydrofolate (formyl H₄F), and formate as intermediates in reactions involving pyrroloquinolinequinone-dependent methanol dehydrogenase, NADP-dependent methylene H₄F dehydrogenase, methenyl H₄F cyclohydrolase, formyl H₄F synthetase, and NAD-dependent formate dehydrogenase (7, 19, 21).

Recently, M. extorquens AM1 was shown to contain genes thought to be unique for methanogenic archaea, namely, genes predicted to encode methenyl tetrahydromethanopterin (H₄MPT) cyclohydrolase, formylmethanofuran:H₄MPT formyltransferase, and formylmethanofuran dehydrogenase (6). These enzymes catalyze the first three steps of methanogenesis during the growth of methanogens on CO₂ and H₂ and the last three steps of CO₂ formation during the growth of methanogens on methanol (15, 28). The genes in M. extorquens AM1 encoding the three methanogenic enzymes were shown by mutagenesis to be required for growth on C₁ compounds, suggesting an involvement of the corresponding enzymes in formaldehyde oxidation to CO₂ via N⁵,N¹⁰-methylene H₄MPT (methylene H₄MPT), N⁵,N¹⁰-methenyl H₄MPT (methenyl H₄MPT), N⁵-formyl H₄MPT (formyl H₄MPT), and formylmethanofuran as intermediates (6).

The genetic evidence for this novel metabolic pathway was substantiated by the finding that cell extracts of methanol-grown M. extorquens AM1 exhibit methenyl H₄MPT cyclohydrolase activity and formylmethanofuran:H₄MPT formyltransferase activity and contain dephospho-H₄MPT (6). H₄MPT is an H₄F analogue previously found only in methanogenic and sulfate-reducing archaea. Its properties are significantly different from those of H₄F (Fig. 1). Thus, the redox potential of the N⁵,N¹⁰-methenyl H₄MPT–N⁵,N¹⁰-methylene H₄MPT couple (−390 mV) is 90 mV more negative than that of the N⁵,N¹⁰-methenyl H₄F–N⁵,N¹⁰-methylene H₄F couple (−300 mV) (13, 29, 37).

FIG. 1.

Structures of H₄MPT (25, 32, 33) and H₄F. Functionally, the most important difference between H₄MPT and H₄F is that H₄MPT has an electron-donating methylene group in conjugation to N¹⁰ via the aromatic ring, whereas H₄F has an electron-withdrawing carbonyl group in this position. One consequence is that the redox potential of the N⁵,N¹⁰-methenyl H₄MPT–N⁵,N¹⁰-methylene H₄MPT couple (−390 mV) is almost 100 mV more negative than that of the N⁵,N¹⁰-methenyl H₄F–N⁵,N¹⁰-methylene H₄F couple (−300 mV) (29). The H₄MPT derivative found in M. extorquens AM1 lacks the α-hydroxyglutaryl phosphate unit (6). The dephospho-H₄MPT is also present in methanogenic archaea (35).

Further support for the novel metabolic pathway came from the finding that cell extracts of M. extorquens AM1 contain an NADP-dependent methylene H₄MPT dehydrogenase activity catalyzing the formation of methenyl H₄MPT from methylene H₄MPT (6). The presence of an NADP-dependent methylene H₄MPT dehydrogenase in M. extorquens AM1 is of special interest because the methylene H₄MPT dehydrogenases in methanogenic and sulfate-reducing archaea are specific for coenzyme F₄₂₀, which is a 5′ deazaflavin derivative (8). The redox potential of the F₄₂₀–F₄₂₀H₂ couple (−360 mV) is 40 mV more negative than that of the NAD(P)⁺–NAD(P)H couple (−320 mV) (12, 34). Apparently, M. extorquens AM1 contains a novel enzyme combining features of both the F₄₂₀-dependent methylene H₄MPT dehydrogenase from methanogens and the NAD(P)-dependent methylene H₄F dehydrogenase found in bacteria and eucarya. We report here on the purification and the characterization of the novel enzyme.

MATERIALS AND METHODS

Coenzymes.

H₄MPT, methenyl H₄MPT, and coenzyme F₄₂₀ were purified from Methanobacterium thermoautotrophicum Marburg (DSM 2133) (5). H₄F was purchased from Sigma. Anoxic stock solutions of H₄MPT and H₄F were prepared in 50 mM Tricine-KOH (pH 7.0) containing 2 mM dithiothreitol. In giving concentrations of H₄F, it was considered that less than 50% of the commercially available H₄F was biologically active as determined enzymatically, due to the presence of the biologically inactive R isomer and dihydrofolate (36). Methenyl H₄F was generated from methylene H₄F by dehydrogenation via NADP-dependent methylene H₄MPT dehydrogenase from M. extorquens AM1 and was purified by high-performance liquid chromatography (HPLC). Methylene H₄F was generated from H₄F and formaldehyde by spontaneous reaction. Methylene H₄F was not purified prior to the enzymatic conversion to methenyl H₄F.

Growth of bacteria.

M. extorquens AM1 was grown on methanol (100 mM) at 30°C in minimal medium as described previously (11). The cultures were harvested in the late-exponential phase at a cell concentration of 3 g (wet mass)/liter. Cells were pelleted by centrifugation at 5,000 × g and stored at −20°C.

Preparation of cell extracts.

Frozen cells (9 g) were suspended in 18 ml of 50 mM morpholinepropanesulfonic acid (MOPS)-KOH (pH 7.0) at 4°C and passed three times through a French pressure cell at 1.2 × 10⁸ Pa. Cell debris and unbroken cells were removed by centrifugation at 27,000 × g for 30 min. The resulting supernatant is referred to as cell extract.

Protein concentration was determined by the Bradford assay (4) by using the Bio-Rad reagent with bovine serum albumin as the standard.

Determination of methylene H₄MPT dehydrogenase activity.

The assays were performed routinely at 30°C in 1-ml cuvettes (depth, 1 cm) in a total volume of 0.7 ml. The reactions were monitored photometrically by measuring the increase or decrease in absorbance at 340 nm. For the calculations, ɛ₃₄₀ values of 6.2 mM⁻¹ cm⁻¹ for NADPH, 20.8 mM⁻¹ cm⁻¹ for methenyl H₄MPT (10), and 21.7 mM⁻¹ cm⁻¹ for methenyl H₄F (18) were used. Units of enzyme activities are defined as 1 μmol/min at 30°C.

Methylene H₄MPT dehydrogenation with NADP⁺ was routinely measured in 120 mM potassium phosphate (pH 6.0) at a methylene H₄MPT concentration of 35 μM and an NADP⁺ concentration of 0.2 mM. The methylene H₄MPT was generated in the cuvette by a spontaneous reaction of H₄MPT (35 μM) with formaldehyde (3 mM). The excess of formaldehyde did not adversely affect the enzyme activity.

Methylene H₄F dehydrogenation with NADP⁺ was measured in 120 mM potassium phosphate (pH 6.0) at a methylene H₄F concentration of 70 μM and an NADP⁺ concentration of 0.2 mM. The methylene H₄F was generated in the cuvette by a spontaneous reaction of H₄F (70 μM) with formaldehyde (3 mM). The excess of formaldehyde did not adversely affect the enzyme activity.

Methenyl H₄MPT reduction with NADPH was measured in 120 mM potassium phosphate (pH 7.0) at a methenyl H₄MPT concentration of 30 μM and an NADPH concentration of 50 μM. The NADPH was regenerated with glucose-6-phosphate (2 mM) and glucose-6-phosphate dehydrogenase (10 U).

Methenyl H₄F reduction with NADPH was measured in 120 mM potassium phosphate (pH 7.0) at a methenyl H₄F concentration of 30 μM and an NADPH concentration of 50 μM.

Purification of methylene H₄MPT dehydrogenase.

All purification steps were performed at 4°C under aerobic conditions. To 23 ml of cell extract stirred on ice, 34.5 ml of saturated ammonium sulfate in 50 mM Tris-HCl (pH 7.0) was added to a final concentration of 60% saturation. After 20 min of stirring, the precipitated protein was removed by 30 min of centrifugation at 20,000 × g. The supernatant was applied to a phenyl Sepharose (High Performance 26/10; Pharmacia Biotech) column equilibrated with 2 M ammonium sulfate in 50 mM Tris-HCl (pH 7.0). With a linear gradient decreasing from 2 to 0 M (NH₄)₂SO₄ (600 ml), the dehydrogenase activity eluted at about 0.25 M (NH₄)₂SO₄. Combined active fractions were concentrated with Centricon 30 microconcentrators (Millipore), washed with 50 mM MOPS-KOH (pH 7.0) (1:5), and subjected to anion-exchange chromatography on a Q Sepharose column (High Performance 16/10; Pharmacia Biotech). The enzyme activity was recovered in the flowthrough of the column (50 mM MOPS-KOH [pH 7.0]). Active fractions were pooled, washed, and concentrated by using Centricon 30 microconcentrators and 50 mM morpholineethanesulfonic acid (MES)-NaOH (pH 5.5). The enzyme was further purified by cation-exchange chromatography on a Mono S column (5/5; Pharmacia Biotech) via a linear gradient from 0 to 0.2 M NaCl in 50 mM MES-NaOH (pH 5.5) (57 ml). Methylene H₄MPT dehydrogenase was recovered at 0.16 M NaCl, diluted (1:3) with 50 mM MES-NaOH (pH 5.5), and rechromatographed on a Mono S column by using the same gradient.

Determination of the N-terminal amino acid sequence.

Purified enzyme was electrophoresed in the presence of sodium dodecyl sulfate (SDS), and the 32-kDa band was electroblotted onto a poly(vinyl trifluoride) membrane (Applied Biosystems). Sequence determination was performed on a 477 protein/peptide sequencer from Applied Biosystems by D. Linder, Giessen, Germany.

Determination of dephospho-H₄MPT and H₄F concentrations.

A cell extract (3 ml) of M. extorquens AM1 was ultrafiltrated with Centricon 3 microconcentrators. The filtrate, containing the low-molecular-mass compounds, was then supplemented with 15 mM formaldehyde, 2 mM NADP⁺, and 2 U of purified methylene H₄MPT dehydrogenase in order to convert dephospho-H₄MPT and H₄F into dephospho-methenyl H₄MPT and methenyl H₄F, respectively. Up to this point, all steps were performed under strictly anaerobic conditions. Subsequently, the filtrate was subjected to HPLC with a LiChrospher RP-18 column (14 mm by 125 mm) (Merck). Absorbed compounds were eluted (1 ml/min) with a linear gradient (30 ml) of 0 to 50% methanol in 25 mM sodium formate (pH 3.0). The eluate was continuously monitored for absorbance at 335, 356, and 274 nm, and UV-visible spectra were recorded by using a Hewlett-Packard 1050 diode array detector. The retention times were 20.0 min for dephospho-methenyl H₄MPT, 18.8 min for methenyl H₄MPT, and 17.9 min for methenyl H₄F. The methenyl H₄MPT had been added to the filtrate for calibration. The methenyl derivatives were analyzed by matrix-assisted laser desorption ionization time-of-flight (MALDI/TOF) mass spectrometry by J. Kahnt, Marburg, Germany, using Voyager-DE RP (PerSeptive Biosystems).

RESULTS

Cell extracts of M. extorquens AM1 grown on methanol catalyzed the reduction of NADP⁺ with methylene H₄MPT (2.6 U/mg), the reduction of NADP⁺ with methylene H₄F (0.2 U/mg), and the reduction of NAD⁺ with methylene H₄MPT (0.6 U/mg) rather than with methylene H₄F (<0.01 U/mg). The activities were associated with the soluble cell fraction. The two NADP-dependent activities were found to copurify and to be separated from the NAD-dependent activity upon ammonium sulfate precipitation. Purification was possible under aerobic conditions.

Cell extracts of M. extorquens AM1 did not catalyze the reduction of coenzyme F₄₂₀ with methylene H₄MPT or methylene H₄F. They also appeared not to contain coenzyme F₄₂₀, as determined spectrofluorometrically.

Purification and molecular properties.

The two NADP-dependent activities were purified by hydrophobic chromatography on a phenyl Sepharose column, by anion-exchange chromatography on a Q Sepharose column, and by cation-exchange chromatography on a Mono S column. In each step a complete copurification of NADP-dependent methylene H₄MPT dehydrogenase and NADP-dependent methylene H₄F dehydrogenase was observed. Purification of both activities was approximately 130-fold with 24% yields (Table 1).

TABLE 1.

Copurification of NADP-dependent methylene H₄F dehydrogenase and NADP-dependent methylene H₄MPT dehydrogenase activities from M. extorquens AM1 grown on methanol

Purification step	Protein (mg)	Activity (U)a with:		Fold purifi-cationb	Yield (%)b
		H4F	H4MPT
Cell extract	642	127	1,652	1	100
(NH4)2SO4 supernatant	129	107	1,408	4	85
Phenyl Sepharose	28	88	1,278	18	77
Q Sepharose	3.4	61	1,020	115	62
Mono S Ic	2.5	55	812	125	49
Mono S II	1.2	31	495	128	24

^aDetermined at 30°C under standard assay conditions. 

^bCalculated for activities determined with methylene H₄MPT. 

^cMono S I and Mono S II, first and second chromatographies, respectively, on a Mono S column. 

After the second chromatography on Mono S, the preparation contained only one polypeptide, with an apparent molecular mass of 32 kDa, as revealed by SDS-polyacrylamide gel electrophoresis (PAGE) (Fig. 2). The N-terminal amino acid sequence (SKKLLFQFDTDATPSVFDVV) determined by Edman degradation and the apparent molecular mass conformed with those predicted for the mtdA product (7).

FIG. 2.

SDS-PAGE analysis of purified methylene H₄MPT dehydrogenase from M. extorquens AM1. Protein was separated on a 16% polyacrylamide gel and subsequently stained with Coomassie brilliant blue (17). Lane A, low-molecular-mass standards (Pharmacia Biotech); lane B, 5 μg of purified MtdA.

The purified enzyme eluted from Superdex 200, a gel filtration column, with an apparent molecular mass of 93 kDa (data not shown), suggesting that the dehydrogenase has a homotrimeric structure.

The UV-visible spectrum of the enzyme was that of a protein lacking a chromophoric prosthetic group (data not shown). The ɛ₂₈₀ determined and the ɛ₂₈₀ calculated from the tyrosine, phenylalanine, and tryptophan contents of the enzyme coincided within a 5% range.

The calculated pI for MtdA is 7.2, conforming with the property of the enzyme to bind to cation-exchange resins at pHs below 7.

The enzyme could be stored at −20°C under aerobic conditions for several weeks without loss of activity.

Catalytic properties.

The purified enzyme was found to be similarly active with the H₄MPT and the dephospho-H₄MPT derivatives, allowing for determination of the catalytic properties by using the more readily available H₄MPT. Apparently, the modification in the side chain (Fig. 1) has little effect on the enzyme activity, a phenomenon also observed for the various H₄MPT-dependent enzymes in methanogenic archaea (14, 23, 26, 27).

Under the standard assay conditions described in Materials and Methods, the purified enzyme catalyzed the dehydrogenation of methylene H₄MPT with NADP⁺ with a specific activity of 413 U/mg and the dehydrogenation of methylene H₄F with NADP⁺ with a specific activity of 26 U/mg. It also catalyzed the reverse reactions, the reduction of methenyl H₄MPT to methylene H₄MPT (9 U/mg) and of methenyl H₄F to methylene H₄F (15 U/mg) using NADPH. For thermodynamic reasons (see below), the reduction of methenyl H₄MPT to methylene H₄MPT could be monitored only when an NADPH-regenerating system was employed.

The dependence of the rate of methylene H₄MPT dehydrogenation with NADP⁺ on the concentrations of the substrates was determined. Reciprocal plots of the initial rates versus the concentration of one substrate at different fixed concentrations of the second substrate yielded straight lines (Fig. 3A and C). From reciprocal plots of 1/apparent V_max versus 1/[S], where [S] is the concentration of the substrate, a V_max of 600 U/mg, a K_m for methylene H₄MPT of approximately 20 μM, and a K_m for NADP⁺ of approximately 30 μM were obtained (Fig. 3B and D). With methylene H₄F as the substrate, the V_max was approximately 30 U/mg, the K_m for methylene H₄F was approximately 30 μM, and the K_m for NADP⁺ was approximately 10 μM (data not shown). Thus, the enzyme catalyzed the dehydrogenation of methylene H₄MPT with an approximately 20-fold-higher catalytic efficiency (V_max/K_m) than the dehydrogenation of methylene H₄F.

FIG. 3.

Kinetics of methylene H₄MPT dehydrogenation with NADP⁺ as catalyzed by NADP⁺-dependent methylene H₄MPT dehydrogenase from M. extorquens AM1. (A and C) Plots of 1/v versus 1/[S], where velocity (v) is given as micromoles of substrate converted per minute. (B and D) Replots of 1/apparent (app.) V_max versus 1/[S]. Assay conditions were as follows: temperature, 30°C; pH 6.0; NADP⁺ and methylene H₄MPT concentrations as indicated; 50 ng of purified methylene H₄MPT dehydrogenase.

Due to the relatively low K_m values for both substrates, it was not possible to accurately measure the rate dependence on the substrate concentration at concentrations much lower than the K_m. Such measurements would be required to ascertain whether the lines in the double reciprocal plots shown in Fig. 3A and C really converge or are parallel, which would indicate a ternary complex (sequential) or a ping-pong catalytic mechanism, respectively. The absence of a chromophoric prosthetic group argues for a ternary complex (sequential) catalytic mechanism.

The optimum pH for methylene H₄MPT dehydrogenation was near 6.0, and the optimum temperature was near 45°C. The activity was slightly stimulated by salts, and maximal activity was reached at a potassium phosphate concentration of 120 mM (pH 6.0). NaCl concentrations higher than 100 mM were inhibitory.

The purified enzyme did not catalyze the reduction of NAD⁺, flavin adenine dinucleotide, flavin mononucleotide, coenzyme F₄₂₀, or dyes such as methylene blue or benzyl viologen using either methylene H₄MPT or methylene H₄F. It also did not exhibit methenyl H₄MPT cyclohydrolase or methenyl H₄F cyclohydrolase activity.

Thermodynamics.

Methylene H₄MPT dehydrogenation using NADP⁺ was found to proceed almost to completion. The reverse reaction, the reduction of methenyl H₄MPT using NADPH, ceased after less than 1% of the methylene H₄MPT had been reduced, as predicted from the E₀ of −390 mV of the N⁵,N¹⁰-methenyl H₄MPT–N⁵,N¹⁰-methylene H₄MPT couple and the E₀ of −320 mV of the NADP⁺-NADPH couple. In the case of methylene H₄F dehydrogenation, equilibrium was attained when about 30% of the methylene H₄F had been converted to methenyl H₄F, in agreement with the E₀ of −300 mV of the methylene H₄F-methenyl H₄F couple. From the concentrations at equilibrium, free energy changes (ΔG° values) of −13 and +3 kJ/mol were calculated for methylene H₄MPT dehydrogenation and methylene H₄F dehydrogenation, respectively.

Intracellular concentrations of dephospho-H₄MPT and H₄F.

Dephospho-H₄MPT and H₄F in cell extracts of M. extorquens AM1 were converted into more stable and readily detectable methenyl derivatives, which were subsequently subjected to reversed-phase HPLC and quantitated spectroscopically. Per milligram of cell extract protein, 1.4 nmol of dephospho-methenyl H₄MPT and 0.5 nmol of methenyl H₄F were detected, corresponding to intracellular concentrations of 0.4 and 0.15 mM, respectively, assuming that the intracellular volume is 3.3 μl/mg of protein, as in other procaryotes (3).

Dephospho-methenyl H₄MPT and methenyl H₄F were identified by their retention times on the HPLC column (internal standardization [data not shown]) and by the UV-visible spectra (Fig. 4), and their masses were determined by MALDI/TOF mass spectrometry (data not shown). The dephospho-methenyl H₄MPT isolated from M. extorquens AM1 was used as the substrate by the NADP-dependent methylene H₄MPT dehydrogenase with a specific activity similar to that for methenyl H₄MPT.

FIG. 4.

UV-visible spectra of dephospho-methenyl H₄MPT and methenyl H₄F isolated from M. extorquens AM1 grown on methanol. The compounds were separated by HPLC, and their spectra were recorded online by diode array detection.

The HPLC analysis did not yield evidence for the presence of tetrahydropterin derivatives other than dephospho-H₄MPT and H₄F.

DISCUSSION

The NADP-dependent methylene H₄MPT dehydrogenase from M. extorquens AM1 has several properties in common with the two kinds of methylene H₄MPT dehydrogenases found in methanogenic archaea, the F₄₂₀-dependent methylene H₄MPT dehydrogenase (16) and the H₂-forming methylene H₄MPT dehydrogenase (29). The three types of dehydrogenases are all composed of only one type of subunit with a molecular mass between 30 and 40 kDa, and they all lack a chromophoric prosthetic group and exhibit a ternary complex (sequential) catalytic mechanism. However, the three enzymes show only a very low degree of sequence similarity (<15%). Only minor sequence similarities are also found with NAD(P)-dependent methylene H₄F dehydrogenases from bacteria and eucarya, which, with a few exceptions (2, 20, 24, 31, 38), are generally more complex, multifunctional enzymes containing additionally methenyl H₄F cyclohydrolase in bacteria and both methenyl H₄F cyclohydrolase and N⁵-formyl H₄F synthetase activities (1, 9) in eucarya. Therefore, based on differences in amino acid sequence and coenzyme specificity, four families of methylene tetrahydropterin dehydrogenases can be defined; they catalyze the following reactions: methylene H₄MPT + NADP⁺ ⇌ methenyl H₄MPT⁺ + NADPH (ΔG° = −13 kJ/mol) methylene H₄MPT + F₄₂₀ + H⁺ ⇌ methenyl H₄MPT⁺ + F₄₂₀H₂ (ΔG° = −5.5 kJ/mol) methylene H₄MPT + H⁺ ⇌ methenyl H₄MPT⁺ + H₂  (ΔG° = +5.5 kJ/mol) methylene H₄F + NAD(P)⁺ ⇌ methenyl H₄F⁺ + NAD(P)H (ΔG° = +3.5 kJ/mol)

Within each of the four families, all enzymes show sequence similarity, even when they belong to organisms that are very distantly related phylogenetically.

In addition to the NADP-dependent methylene H₄MPT dehydrogenase, M. extorquens AM1 cell extracts contain an NAD-dependent methylene H₄MPT dehydrogenase activity. Genetic evidence indicates that this enzyme is encoded by orfX (6). The amino acid sequence of OrfX is 30% identical to that of the NADP-dependent methylene H₄MPT dehydrogenase. While the NADP-dependent dehydrogenase (MtdA) exhibits some activity with methylene H₄F, the NAD-dependent methylene H₄MPT dehydrogenase is specific for H₄MPT (unpublished results).

For understanding possible functions of the pyridine nucleotide-dependent methylene H₄MPT dehydrogenases found so far only in M. extorquens AM1, it is important that they preferentially catalyze the C₁ oxidation reaction, whereas the three other methylene tetrahydropterin dehydrogenases can, in principle, operate in both directions. It is also of importance that the NADP-dependent methylene H₄MPT dehydrogenase exhibits some NADP-dependent methylene H₄F dehydrogenase activity and that M. extorquens AM1 contains dephospho-H₄MPT and H₄F at similar concentrations. Under certain conditions the NADP-dependent methylene H₄MPT dehydrogenase could therefore also function in the reversible dehydrogenation of methylene H₄F. There is evidence that methanol-grown M. extorquens AM1 contains an H₄F-specific serine hydroxymethyltransferase and an H₄F-specific methenyl H₄F cyclohydrolase (unpublished results). There is also evidence for the presence of an H₄MPT-specific methenyl H₄MPT cyclohydrolase and an H₄MPT-specific formylmethanofuran:H₄MPT formyltransferase. It therefore appears that in M. extorquens AM1 two independent C₁ transfer pathways are simultaneously operative, one that involves dephospho-H₄MPT as the C₁ carrier and is most likely to operate in the oxidative direction and one that involves H₄F and might operate in either direction, depending upon cellular pools of intermediates (Fig. 5). Purification and characterization of the enzymes involved in these pathways are an important step in the elucidation of the roles of the two C₁ transfer pathways in the anabolism and catabolism of M. extorquens AM1 growing on methanol.

FIG. 5.

Possible role of methylene H₄MPT dehydrogenase (MtdA) and other enzymes in growth of M. extorquens AM1 on methanol. Note that the presence of methanofuran in M. extorquens AM1 has not been proven. It is not known whether N⁵-formyl H₄MPT can support purine biosynthesis.