Purification and properties of S-adenosyl-l -methionine:Nicotinic acid-N-methyltransferase from cell suspension cultures of Glycine max L.

ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 262, No.. 2, May 1, pp. 445-454,1988

Purification and Properties of S-Adenosyl-L-methionine:Nicotinic methyltransferase from Cell Suspension Cultures of G/y&e BARBARA

UPMEIER,

WILDTRUD GROSS, SUSANNE WOLFGANG BARZ2

Acid-Nmax L.’

KijSTER,

Lehrstuhl fiir Biochemie de-rPjlanzq Westfdische WilhelwwUnivemitit D-&o0 Miinster, Fe&al Republic of Gemnany

AND

Miinster,

Received August 21,1987, and in revised form December 22,1987

A soluble enzyme which catalyzes the transfer of the methyl group from S-adenosyl -L-methionine to the nitrogen atom of pyridine-3-carboxylic acid (nicotinic acid) could be detected in protein preparations from heterotrophic cell suspension cultures of soybean (Glycine max L.). Enzyme activity was enriched nearly loo-fold by ammonium sulfate precipitation, gel filtration, and ion-exchange chromatography to study kinetic properties. S-Adenosyl-L-methionine:nicotinic acid-N-methyltransferase (EC 2.1.1.7) showed a pH optimum at pH 8.0 and a temperature optimum between 35 and 40°C. The apparent KM values were determined to be ‘78 PM for nicotinic acid and 55 PM for the cosubstrate. S-Adenosyl-L-homocysteine was a competitive inhibitor of the methyltransferase with a KI value of 95 PM. The native enzyme had a molecular mass of about 90 kDa. The catalytic activity was inhibited by reagents blocking SH groups, whereas other divalent cations did not significantly influence the enzyme reaction. The purified methyltransferase revealed a remarkable specificity for nicotinic acid. No other pyridine derivative was a suitable methyl group acceptor. To study a potential methyltransferase activity with nicotinamide as substrate, an additional purification step was necessary to remove nicotinamide amidohydrolase activity from the enzyme preparation. This was achieved by affinity chromatography on S-adenosyl-L-homocysteineSepharose thus leading to a 580-fold purified enzyme which showed no methyltransferase activity toward nicotinamide as substrate. o Istxa Academic press, I,,~.

Nicotinic acid can be regarded as an important connecting link between primary and secondary metabolism in higher plants. Being a constituent of the pyridine nucleotide cycle, it is a precursor and a degradation product of the coenzymes NAD and NADP (l-3). Plant cells are able to completely degrade nicotinic acid (4) or to form conjugates (5-7). Furthermore, nicotinic acid may serve as a building moiety for pyridine alkaloids (1). Conjugation reactions play an impor1 This work was supported by the Deutsche Forschungsgemeinschaft and Fonds der Chemischen Industrie. ’ To whom correspondence should be addressed.

tant role in plant metabolism as they allow storage, detoxification, and metabolic regulation of primary and secondary constitutents in plant cells (8, 9). The methyl group is one of the most important conjugation moieties (10). Trigonelline, the N-methyl conjugate of nicotinic acid, is a well-known cell constituent of many different plant species (11, 12) and of numerous cell cultures (7). It is considered to be a storage form for nicotinic acid (13). After demethylation, a reaction which has been reported for several plant tissues (14), nicotinic acid can enter the pyridine nucleotide cycle and be reused for NAD synthesis (15). Trigonelline may have an additional physiological 0003-9861/88 $3.00 Copyright Q 1988 by Academic Press, Inc. All rights of reproduction in any form resewed.

446

UPMEIER ET AL.

significance because it is considered to be a plant hormone which has an important function in cell proliferation (16, 17). The enzymatic formation of trigonelline has already been studied using crude protein preparations from fenugreek (Trigonells foenum graecum L.) callus cultures (18). In this report we describe the characterization of S-adenosyl-L-methionine:nicotinic acid-N-methyltransferase with the intention to further elucidate the physiological role of triogonelline in plant cell metabolism. Soybean cell suspension cultures were used as a source of the enzyme because these cell cultures readily convert nicotinic acid to trigonelline. MATERIALS

AND METHODS

Plant material. Heterotrophic soybean (Glycine mux L.) cell suspension cultures were cultivated on Gamborg B6-medium (19) under conditions previously described (20). For enzyme purification cell cultures were harvested 4 days after inoculation. Chemicals. Nicotinic acid, nicotinamide, and nicotinic acid methyl ester were purchased from Merck (Darmstadt, FRG). Picolinic acid, iso-nicotinic acid, S-adenosyl+methionine, S-adenosyl-L-homocysteine, and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide were obtained from Sigma (Munich, FRG), 6-hydroxynicotinic acid was from Schuchardt (Munich, FRG). 3-Hydroxypyridine and quinolinic acid were supplied by EGA-Chemie (Steinheim, FRG); 2hydroxy- and 4-hydroxypyridine came from Aldrich (Beerse, Belgium). Pyridine-3-aldehyde, dithioerythritol, p-hydroxymercuribenzoate, and Dowex 1X 8 were purchased from Serva (Heidelberg, FRG); trigonelline was from Roth (Karlsruhe, Germany). Ultrogel AcA 34 and AcA 44 were obtained from LKB (Grlfelfing, FRG). Pharmacia (Uppsala, Sweden) supplied all other materials for protein purification (DEAE-Sephacel, AH-Sepharose 4B, Sephadex G-100, Sephadex G-25 (PD-10 columns)). The radiochemicals [7-‘4C]nicotinic acid (56 mCi/mmol), [7-‘4C]nicotinamide (55 mCi/mmol), and [methyl14C]Sadenosyl-L-methionine (60 mCi/mmol) were obtained from The Radiochemical Center Amersham (Amersham, UK). Bu&r systems. The buffers were buffer A, 200 mM potassium phosphate, 20 mM dithioerythritol, 2% glycerol (v/v), 0.02% sodium azide (w/v), pH 8.0; buffer B, 20 mM potassium phosphate, 4 mM dithioerythritol, 2% glycerol (v/v), 0.02% sodium azide (w/v), pH 8.0; buffer C, 100 mM potassium phosphate, pH 8.0. For the determination of the pH optimum the following buffer systems were used: pH range 4.0-5.0,20

mM Nas-citrate/HCl; pH 5.0-8.5, 20 mre potassium phosphate; pH range 8.0-10.0,20 mM Tris/HCl. Each buffer contained 2 mM dithioeythritol, 2% (v/v) glycerol, and 0.02% (w/v) sodium azide. Protein determination. Protein concentrations were determined according to Bradford (21) with bovine serum albumin as reference. Enzyme p&cation. The cell cultures were separated from the growth medium on a sintered glass funnel and washed with distilled water. Cells were homogenized in a chilled mortar with buffer A (1 ml/g fresh wt) and 10% (w/w) quartz sand. All procedures were carried out at 4°C. The homogenate was centrifuged at 25,OOOgfor 30 min. The supernatant was fractionated by dropwise addition of a saturated ammonium sulfate solution in two steps between O-35 and 35-65% saturation. The ammonium sulfate solution had been adjusted to pH 8.0 with NH40H and it was added to the protein extract at a rate of 2 ml/min. The protein solution was stirred for another 20 min before the precipitated protein was removed by centrifugation (3O,OOOg,30 min). The resulting sediment of the second precipitation was dissolved in 6-8 ml buffer B, immediately applied to an Ultrogel AcA 34 column (2.6 X 86 cm), and chromatographed using buffer B (flow rate 16 ml/h) and a fraction size of 4 ml. The fractions with pronounced methyltransferase activity were combined and subjected to ionexchange chromatography on a DEAE-Sephacel column (3.0 X 30 cm). After elution of unbound proteins with 200 ml of buffer B a linear gradient of NaCl in buffer B with a slope of O-O.5M NaCl in a total volume of 400 ml was applied. Protein was eluted with a flow rate of 40 ml/h and fractions of 4 ml each were collected. The fractions containing methyltransferase activity were pooled (32 ml) and concentrated by ultrafiltration (PM 10, Amicon, Osterhout, Netherlands) to a volume of 5 ml. This sample was desalted on Sephadex G-25 (PD-10) and immediately used for studies of enzyme properties. The separation of methyltransferase activity from nicotinamide amidohydrolase activity was achieved by affinity chromatography with AdoHcy3 bound to AH-Sepharose 4B. The purified enzyme preparation (1 ml) was applied to the AdoHcy-Sepharose column (0.7 X 5.0 cm) equilibrated with buffer B. The column was washed with 22.5 ml buffer B and subsequently eluted with 100 pM AdoMet in 15 ml buffer B and 80 mM and 200 ml NaCl, each dissolved in 22.5 ml buffer B (flow rate 20 ml/h). Fractions of 1.5 ml each were collected and assayed for nicotinic acid-N-

3 Abbreviations used: AdoMet, Sadenosyl-L-methionine; AdoHcy, S-adenosyl-L-homocysteine; EDTA, ethylenediaminetetraacetic acid; ECD, lethyl-3-(3-dimethylaminopropyl) carbodiimide; Tris, Tris (hydroxymethyl)aminomethane.

NICOTINIC

ACID

N-METHYLTRANSFERASE

methyltransferase and nicotinamide amidohydrolase activity. Preparation of AdoHcy-Sepharose. A modified method according to Sharma and Brown (22) was used to prepare AdoHcy-Sepharose for affinity chromatography. AH-Sepharose 4B (1 g) was suspended in 25 ml 0.5 M NaCl solution and incubated for 30 min at room temperature. The gel was washed with 250 ml 0.5 M NaCl solution on a sintered glass funnel. NaCl was removed from the gel with 100 ml distilled water previously adjusted to pH 4.6. The moist gel was added to 4 ml of an aqueous ligand solution containing 15 mg S-adenosyl-L-homocysteine. AdoHcy was readily dissolved by decreasing the pH to pH 3.0. Before suspending the gel the pH of the ligand solution was adjusted to pH 4.6. ECD (115 mg) predissolved in 1 ml of distilled water was added to the gel suspension. The pH was maintained at pH 4.5-5.5 for 4 h, and the reaction was allowed to proceed another 24 h with the gel being gently agitated in an endover-end stirrer. The gel was finally washed with distilled water, pH 4.6, packed in a column (0.7 X 5.0 cm), and equilibrated with buffer B prior to chromatography. Fast-protein. liquid chromatography. Ion-exchange chromatography was performed on a Mono Q column (Pharmacia) equilibrated with buffer B. The proteins were eluted with a biphasic linear gradient of NaCl in buffer B, first with a slope of 25 mM/ml up to 100 mM NaCl and then with 8.5 m&ml up to 200 mM NaCl. The flow rate was 2 ml/min and fractions of 1 ml each were collected. Gel filtration was carried out with a Superose 6 column (Pharmacia) with buffer B at a flow of 0.5 ml/min (fraction size 0.5 ml). Enzyme assays. Standard assays consisted of 100 ~1 enzyme preparation, 50 nmol nicotinic acid containing 0.1 &i [7-r4C]nicotinic acid and 75 nmol AdoMet in a total volume of 150 ~1 buffer B. The enzyme reaction was started by adding the cosubstrate, incubated for 1 h at 35°C and stopped by transferring the assay tubes to a boiling water bath. Denaturated protein was removed by centrifugation and 100 ~1 of the supernatant were subjected to ion exchange chromatography on Dowex 1X8 (mesh 100-200, formate form). Dowex resin (2 ml) was equilibrated with distilled water and packed into a 5-ml syringe. After application of the enzyme assay the reaction product trigonelline could be removed by washing the resin with 5 ml distilled water. Radioactivity in this product fraction was measured by liquid scintillation counting to quantitate the enzyme reactions. Nicotinic acid could be eluted from the resin with 15 ml 8 M formic acid. The enzyme assay for nicotinamide amidohydrolase activity contained 100 ~1 enzyme preparation and 50-100 nmol nicotinamide with 0.1 pCi [‘I-“C]nicotinamide in a total volume of 150 pl buffer B. After 30-60 min of incubation at 35°C the reaction was

FROM

Glyoine

mom CELL

CULTURES

447

stopped by transferring the test tubes to a boiling water bath. Separation of product from substrate was achieved by ion-exchange chromatography as described above. The product of the enzyme reaction bound to the Dowex 1X8 resin and finally it was found in the formic acid fraction, while the remaining substrate could previously be separated by washing with 15 ml distilled water. Radioactivity of the product fraction was determined to quantitate the amidohydrolase activity. Product identi,fication The identity of the methylation product was verified using thin-layer chromatography on silica gel plates (20 X 20 cm, Si GFW, Merck, Darmstadt, FRG) using trigonelline as reference. Chromatographic solvents were acetone:water (4:l) n-butanol:acetic acid:water (4:1:1), and methylethylketone:methanol:formic acid:water (11:6:1:2). In addition, the product of the enzymatic reaction was subjected to high-performance liquid chromatography. The chromatogram was developed using a linear gradient of 100% acetonitrile to 50% acetonitrile in 1.5% (w/v) phosphoric acid within 30 min at a flow of 0.8 ml/min. Separation was performed on a Si 60 LiChrosorb column (250 X 4 mm, 5 pm, Merck) monitoring the uv absorption at 261 nm. The retention time of the methylated product was compared to that of trigonelline. Substrate specificity The reaction mixtures contained 200 pl enzyme preparation, 100 nmol of the substrate, and 150 nmol AdoMet with 0.2 WCi [methyl-‘4C]S-adenosyl-L-methionine in a total volume of 300 pl. After incubating the assays for 1 h at 35°C the reaction was stopped by addition of 506 ~1 methanol. The assays were brought to dryness by evaporation and the residues were redissolved in 100 ~150% (v/v) aqueous methanol. Enzyme assays with nicotinic acid methyl ester, pyridine-&aldehyde, 2-, 3-, and 4-hydroxypyridine were subjected to thinlayer electrophoresis on cellulose glass plates (20 X 20 cm, Fmr Merck) with buffer C to detect methylated products (loo-150 V, 40 mA, 150 min). Radioactivity on the thin-layer plates was located with a radioscanner LB 2723 (Berthold, Wildbad, FRG). To determine a potential methylation of B-hydroxynicotinic acid, quinolinic acid, iso-nicotinic, and picolinic acid, the methanolic solution obtained from the enzyme assay was analyzed using high-performance liquid chromatography with concomitant detection of radioactivity. Chromatographic analysis was performed under conditions described above (see Product identification). The retention times were compared to the radioactive signals monitored with a constant flow radiometer (Ramona-5, Raytest, Straubenhardt, FRG). Molecular moss. The molecular mass of the native enzyme was determined in the course of gel filtration using three different materials. Hexokinase, chymotrypsinogen (Serva), bovine serum albumin, and ovalbumin (Sigma) were used as marker proteins.

448

UPMEIER

The void volume and the pore volume of the columns (Sephadex G-100, Ultrogel AcA 34, and AcA 44, each 2.6 X 86 cm) were determined with Blue 2000 Dextran (Pharmacia, Freihurg, FRG) and dinitrophenylalanine (Serva), respectively. Sedimentation analysis was carried out according to Martin and Ames (23) using a linear isokinetic sucrose gradient (5-20% (w/v) in buffer B with a volume of 12 ml and protein samples of 100 ~1. A preparative ultracentrifuge (Damon/IEC B-60) equipped with a swinging bucket rotor 488 was employed. After 24 h of centrifugation (4’C, 40,000 rpm) the tubes were fractionated dropwise from the bottom and fractions of 400 ~1 each were collected. Alcohol dehydrogenase from horse liver (Boehringer, Mannheim, FRG) was used as internal reference protein. RESULTS

Enzyme Reaction Enzyme activity for the conversion of nicotinic acid to the N-methyl conjugate trigonelline (see Fig. 1) was detected in crude protein preparations from heterotrophic cell suspension cultures of G. max L. The addition of dithioerythritol to the extraction buffer led to a sevenfold increase in enzyme activity already indicating a strict dependence of the enzyme on sulfhydryl group containing compounds. The product of the enzymatic reaction was identified by comparing its chromatographic behavior to that of trigonelline during HPLC analysis and thin-layer chromatography in different solvent systems. Radioactive trigonelline could be detected when [methyZ-‘4C]AdoMet was added to the incubation assay instead of [7-14C]nicotinic acid, thus proving that AdoMet is the methyl donor of the transfer reaction.

f)COOH

““T

yy

0’00” Y

tn,

nicotinic acid

trigonelline

FIG. 1. Conjugation reaction catalyzed by S-adenosyl-L-methionine:nicotinic acid-N-methyltransferase.

ET AL.

Putification of the Nicotinic Acid-Nmethyltransfwase In order to successfully enrich the nicotinic acid-N-methyltransferase which was a labile enzyme and very susceptible to oxygen, a rapid and effective purification procedure was established. The results are summarized in Table I. After ammonium sulfate precipitation, the protein extract was immediately subjected to chromatography on Ultrogel AcA 34 (Fig. 2A). Gel filtration provided a 18-fold enriched enzyme preparation without any loss in total activity. The following ion-exchange chromatography (Fig. 2B), however, resulted in only 51% recovery of methyltransferase activity. This loss in activity could not be avoided replacing NaCl by potassium phosphate in the elution buffer. The resulting enzyme preparation was employed for studying the properties of the methyltransferase. In spite of an 10Zfold enrichment the methyltransferase was still contaminated with nicotinamide amidohydrolase activity which made it impossible to detect a potential catalytic activity of the enzyme against nicotinamide. Under standard assay conditions nearly 90% of the nicotinamide added was converted to nicotinic acid, which then served as a substrate. In order to further purify the methyltransferase the protein preparation obtained after DEAE-Sephacel chromatography was subjected to fast-protein liquid chromatography. Neither gel permeation on Superose 6 nor chromatography on a Mono Q column separated the two enzyme activities. However, by these steps the methyltransferase was enriched 130-fold and 205-fold, respectively. Separation ‘of the two enzyme activities was finally achieved by subjecting the protein preparation obtained from the DEAE-Sepharose column to affinity chromatography on AdoHcy-Sepharose 4B. When the partially purified enzyme preparation was applied to the AdoHcy-Sepharose column, a protein peak appeared by washing the gel material with buffer B. Apparently, a large amount of protein passed through the column containing none of the enzyme

NICOTINIC

ACID

N-METHYLTRANSFERASE

FROM

TABLE

Glycine

mace CELL

449

CULTURES

I

PURIFICATIONOFS-ADENOSYL-L-METHIONINE:NICOTINIC ACID-N-METHYLTRANSFERASE FROM HETEROTROPHICCELL SUSPENSIONCULTURESOF Glycine wzex L. Purification step Crude extract (NH&SO~ fractionation 35-65s saturation Ultrogel AcA 34 DEAE-Sephacel AdoHcy-Sepharose 4B AdoMet fraction 200 mM NaCl fraction

Protein (md

Total act (&at)

72.0

201.6

28.6 4.3 0.36

217.7 102.8

0.006 0.037

191.5

10.1 22.2

Sp act (&at/w)

Recovery (%)

Purification (fold)

2.8

100

1

6.7 50.4 285.6

95 108 51

2.4 18 102

1624.0 596.4

5

580

11

213

activities. Elution with 100 pM AdoMet in buffer B partially displaced the methyltransferase. Selective desorption of the amidohydrolase could be achieved with low concentrations of NaCl (80 IIIM in buffer B), while a second protein fraction with nicotinic acid-N-methyltransferase was eluted with 200 IIIM NaCl in the irrigant buffer (Fig. 3). Affinity chromatography resulted in two different protein fractions, which contained 580-fold and 213-fold purified nicotinic acid-N-methyltransferase, respectively. Both fractions were devoid of amidohydrolase activity. When the methyltransferase was chromatographed on a “control gel” to which no specific ligand had been linked, the enzyme activi.ty bound to the AH-Sepharose 4B matrix and could be eluted with 200 IIIM NaCl.

transferase could only be stored at +4”C for 24 h without significant loss in activity when kept under a nitrogen atmosphere.

Stability of the Enzyme

Molecular Mass of the Methyltransferase

Despite the presence of sulfhydryl group protecting agents in all buffers the purification procedure was severely hampered by the instability of the methyltransferase. Rigorous exclusion of oxygen by saturating the buffers with nitrogen allowed partial stabilization and thus an 580-fold enrichment of the enzyme was achieved. On freezing at -20°C or during storage in liquid nitrogen the catalytic activity of the purified enzyme was nearly totally lost. Attempts to stabilize the enzyme with varying amounts of glycerol (lo-40% (v/v)) or nicotinic acid failed. The methyl-

EJTectof Divalent Cations and Sufiydryl Group Blocking Agents The methyltransferase does not require divalent cations for enzymatic activity. A slight inhibition caused by Me ions was compensated by the addition of EDTA. Ca2’ and Mn2+ did not significantly influence the enzyme reaction. Hg2f and p-hydroxymercuribenzoate exercised a strong inhibitory effect on the methyltransferase reaction. Even in the presence of 2 InM dithioerythritol500 pM He ions totally inhibited the enzyme. Under these conditions iodoacetamide (lo-500 mM) caused only partial inhibition.

Chromatographic behavior on three different gel filtration columns pointed to a molecular mass in the range of 75-90 X 103.In order to more precisely estimate the molecular mass of the active enzyme, sedimentation analyses (ultracentrifugation) according to Martin and Ames (23) were performed. A molecular mass of (90 + 1.7) X lo3 was determined with alcohol dehydrogenase (80 X 103) as reference protein. Kinetic Properties The DEAE-Sephacel purification step yielded an about loo-fold enriched prepa-

UPMEIER ET AL.

:. ; ‘: : i :

i/’ :.;’.,‘.., ‘..., ./” /. .‘;

: : :

“..._

‘._. . ..(

.’ fj ,, .0 ,....’ ,, .Y’ , i’ .,.”..,. ,. .’ ,.c ,:’ ‘....__., ‘L, ..”_:’- ..‘.., .M./ .. ...‘.,.’ I r50

x)0

_.

‘......

iil i t Fractions d Lml

FIG. 2. (A) Ultrogel AcA 34 filtration of S-adenosyl-L-methionine:nicotinic acid N-methyltransferase after ammonium sulfate precipitation (35-s% saturation). (B) DEAE-Sephacel anion-exchange chromatography of 4adenosyl-L-methionine:nicotinic acid-iV-methyltransferase obtained from step (A).

ration of nicotinic acid-N-methyltransferase which was used for enzyme kinetic studies. The methylation reaction was lin-

ear only up to 80 min of incubation at 35°C. This effect is due to enzyme inactivation during the incubation period rather

NICOTINIC

ACID

5

10

N-METHYLTRANSFERASE

15

20

2s

FROM

30

35

Glycine

40

max CELL

4s Fmctiis

CULTURES

50 d 1,Sml

FIG. 3. Separation of S-adenosyl-L-methionine:nicotinic acid-N-methyltransferase from nicotinamide amidohydrolase activity by affinity chromatography on AdoHcy-Sepharose 4B. Between fractions 15 and 25 protein content could not be detected by optical density measurements (Am) because of interfering absorption of AdoMet in the elution buffer.

than to a decomposition of the cosubstrate AdoMet, as is mentioned by Dogbo and Camara (24). Enzyme assays carried out with AdoMet which had been preincubated for 1 h at 35°C and pH 8.0 showed no difference in nicotinate conversion compared to those obtained with a fresh AdoMet solution. By contrast, a l-h preincubation of the methyltransferase led to a significant loss (32%) of enzyme activity during the subsequent assay. The enzyme exhibited a pH optimum at pH 8.0. At pH 5.0 the methyltransferase still showed 50% of the maximum activity, whereas at pH 9.0 it was nearly inactive. The temperature optimum was determined to be between 35 and 4O”C, though at 45°C methyltransferase activity still could be detected in remarkable amounts. The apparent Kna values for nicotinic acid and AdoMet determined by a Lineweaver-Burk plot were found to be 78(flO) and 55(f7) PM, respectively, with a maximum velocity of 25.7 pkat/mg protein. Determination of the kinetic constants by two other methods of linear transformation (Eadie-Hofstee and Hanes (25)) differed by

10-15s. The reaction is competitively inhibited by S-adenosyl-L-homocysteine; the calculated K1 value is 95 PM. Substrate SpecQicity The purified methyltransferase exhibited a remarkable substrate specificity for nicotinic acid. A substantial number of other pyridine derivatives namely 6-hydroxynicotinic acid, iso-nicotinic acid, nicotinic acid methyl ester, pyridine-&aldehyde, 2-, 3-, and 4-hydroxypyridine, quinolinic acid, and picolinic acid were also tested. A methylation reaction could not be detected with any of these compounds. Furthermore nicotinamide was also not converted by the methyltransferase, as could be demonstrated with enzyme preparations devoid of amidohydrolase activity. DISCUSSION

In previous studies it could be demonstrated that nicotinic acid is converted to the iv-methyl conjugate trigonelline by

452

UPMEIER ET AL.

various Papillionaceae cell cultures (26). The methyl transfer reaction depicted in Fig. 1 has now been studied at the enzymatic level and the S-adenosyl-L-methionine:nicotinic acid-N-methyltransferase (EC 2.1.1.7) was purified and characterized from cell suspension cultures of G. 7r&uxL. The properties of this enzyme are very similar to those of other plant N-methyltransferases (for references see (11)). An alkaline pH is necessary for maximum enzyme activity. The enzyme requires sulfhydryl group protection and is very sensitive to He ions, p-hydroxymercuribenzoate, and iodoacetamide. This and other observations (27) indicate that essential sulfhydryl groups are necessary for either enzyme conformation or catalysis of the reaction. S-Adenosyl-L-homocysteine, a product of the methyl transfer reaction, is a competitive inhibitor of the nicotinateN-methyltransferase. With only a few exceptions (28, 29) the catalytic activity of other plant methyltransferases is affected by this cosubstrate analog in the same way (for references see (11)). Affinity chromatography on AdoHcylinked Sepharose makes use of these kinetic properties during purification of methyltransferases (30). This chromatographic method was successfully employed for the separation of nicotinic acid-N-methyltransferase from nicotinamide amidohydrolase activity. As was expected from other purification studies of methyltransferases (31,32) a release of nicotinic acid-N-methyltransferase activity occurred when the cosubstrate AdoMet was added to the elution buffer (Fig. 3). A second peak of methyltransferase activity was eluted with 200 mM sodium chloride. There are several examples in the literature for such specific desorption of methyltransferases from AdoHcy affinity columns by using changes in ionic strength or pH of the buffer (22,33). In order to elucidate whether the nicotinate-N-methyltransferase is nonspecifically adsorbed to the gel due to ionic binding or whether this chromatographic behavior is a question of different enzymes, the methyl-

transferase fraction was chromatographed on a “control gel” bearing no biospecific ligand. The methyltransferase was retarded and completely eluted with 200 mM sodium chloride. These results clearly demonstrate that two different binding types are involved in the chromatography of nicotinic acid-N-methyltransferase on AdoHcy-Sepharose 4B. Partial separation can be related to the functional ability of the enzymatic binding site for the specific ligand AdoHcy attached to the gel. A second methyltransferase fraction is bound to the Sepharose matrix by unspecific adsorption. The various patterns of secondary plant products possessing one or more methyl groups can be explained by the high specificity of methylating enzyme systems involved in their biosynthesis (34, 35). The nicotinate-N-methyltransferase from soybean cell suspension cultures exhibits a very narrow substrate specificity, too. From various pyridine derivatives tested only nicotinic acid was accepted as a substrate. Nicotinamide was not converted to the N-methyl conjugate by the purified methyltransferase. Thus the soybean enzyme is significantly different from an animal N-methyltransferase which is specific for nicotinamide (36, 37). N-Methylnicotinamide plays an important role in the pyridine nucleotide metabolism of animal cells. The physiological significance of N-methylated nicotinic acid in plants is still under investigation. Methylation represents an important way to detoxify substances (38, 39) or to convert them into storage forms. In plants, trigonelline is mostly found in storage organs like seeds, roots, or old leaves (for references see (14)). There is evidence that after demethylation of trigonelline the nicotinic acid moiety is funneled into the pyridine nucleotide cycle to form NAD (14,40). Extensive studies on nicotinic acid metabolism in plant cell cultures have demonstrated that nicotinic acid is rapidly and alternatively converted to conjugates either by N-glucosylation or N-methylation (7). A significant turnover of the N-gluco-

NICOTINIC ACID iV-METHYLTRANSFERASE

FROM Glycine maa: CELL CULTURES

453

7. BARZ, W. (1985) in Primary and Secondary Mesyl conjugate in parsley cell cultures tabolism of Plant Cell Cultures (Neumann, proves its close linkage to the pyridine K. H., Barz, W., Reinhard, E., Eds.), pp. nucleotide cycle, whereas the N-demethyl186-195, Springer Verlag, Berlin/Heidelberg. ation of trigonelline in soybean cell cul8. BARZ,W., AND K~STER,J. (1981) in The Biochemtures seems to proceed too slowly to meet istry of Plants (Stumpf, P. K., and Conn, E. E., the nicotinate requirement of the cell for Eds.), Vol. 7, pp. 35-84, Academic Press, New pyridine nucleotides (B. Upmeier and W. York. Barz, unpublished). Therefore it is uncer9. BARZ, W., K~STER, J., WELTRING, K.-M., AND tain whether such a very slow demethylSTRACK, D. (1985) in Annual Proceedings of the Phytochemical Society of Europe (Van ation rate allows the methyl conjugate to Sumere, C. F., and Lea, P. J., Eds.), Vol. 25, pp. fulfil a role as a storage form for nicotinic 307-347, Clarendon Press, Oxford. acid. In cell cultures of G. ma2 L. the Sadenosyl-L-methionine:nicotinic acid-N- 10. POULTON,J. E. (1981) in The Biochemistry of Plants (Stumpf, P. K., and Conn, E. E., Eds.), methyltransferase synthesizes a methyl Vol. 7, pp. 667-723, Academic Press, New York. conjugate that exhibits the properties of a 11. MOTHES, K., AND SCH~TTE, H. R. (1969) Biotypical secondary plant product (12). synthese der Alkaloide, VEB Verlag, Berlin. The successful characterization of soy- 12. GROSS,D. (1970) in Fortschritte der Chemie Orbean nicotinic acid-N-methyltransferase gan&her Naturstoffe (Herz, W., Grisebach, in the combination with our previous reH., and Scott, A. I., Eds.), Vol. 28, pp. 109-161, Springer Verlag, Wien/New York. sults on nicotinic acid-N-glucosyltransferase (7) will allow more detailed investi- 13. BLAIM, K. (1960) Naturwissenschqfkn 14,332. gations at the enzymatic level on the 14. TAGUCHI, H., AND SHIMABAYASHI,Y. (1983) Biothem. Biophys. Res. Commun 113(2), 569-574. remarkable alternative formation of nico15. HEEGER, V., LEIENBACH, K.-W., AND BARZ, W. tinic acid conjugates in plant cell cultures (1976) Hoppe-Seyler’s 2. Physiol. Chem. 357, (26). Furthermore, future studies will re1081-1087. veal the metabolic linkage of nicotinate 16. EVANS, L. S., AND TRAMONTANO, W. A (1981) conjugates to the pyridine nucleotide cycle Amer. .I Bot. 68(9), 1282-1289. in plant cell cultures (3). 17. LYNN, D. G., LEWIS, D. H., TRAMONTANO,W. A., ACKNOWLEDGMENTS We thank Professor B. Surholt (Institute of Zoology, University of Miinster) for access to the fastprotein liquid chromatography system used in this study. The technical assistance of Mrs. J. Goldberg during chromatographical analyses is gratefully acknowledged. REFERENCES 1. WALLER, G. R., AND NOWACKI,E. K. (1978) Alkaloid Biology and Metabolism in Plants, Plenum, New York/London. 2. WAGNER, R., AND WAGNER, K. G. (1985) Planta 165,532-537. 3. WAGNER,R., FETH, F., AND WAGNER,K. G. (1986) Planta 167,226-232. 4. SCHWENEN, L., KOMOSSA, D., AND BARZ, W. (1986) 2. Naturforsch C 41,148-157. 5. MIZUSAKI, S., TANABE, Y., KISAKI, T., AND TAMAKI, E. (1970) Phytochemistry 9,549-554. 6. LEIENBACH, K.-W., HEEGER, V., AND BARZ, W. (1976) Hoppe-Seyler’s 2. Physiol. Chem. 357, 1089-1095.

AND EVANS, L. S. (1984) Phytochemistry 23(6), 1225-1228. 18. ANTONY, A., GOPINATHAN,K. P., AND VAIDYANATHAN, C. S. (1975) Indian J. Exp. Bot. 13, 39-41. 19. GAMBORG, 0. L. (1966) Canad. J. B&hem. 44, 791-799. 20. HAHLBROCK, K., KUHLEN, E., AND LINDL, T. (1971) Planta (Berlin) 99,311-318. 21. BRADFORD, M. M. (1976) Anal. Biochem. 72, 248-254. 22. SHARMA,S. K., AND BROWN,S. A. (1978) J. Chrcmatogr. 157,427-431. 23. MARTIN, R. G., AND AMES, B. N. (1961) J. Biol. Chem 236,1372-1379. 24. DOGBO,O., AND CAMARA,B. (1986) C. R. Acad Sci Paris 303,93-96. 25. DIXON, M., AND WEBB,E. C. (1986) Enzymes, Academic Press, New York. 26. WILLEKE, U., HEEGER, V., MEISE, M., NEUHANN, H., SCHINDEL-MEISER,I., VORDEMFELDE,K., AND BARZ, W. (1979) Phytochemistry 18, 105-110. 27. WINK, M. (1984) Planta 161,339-344.

454

UPMEIER

28. KNOGGE, W., AND WEISSENB~CK, G. (1984) Eur. J. B&hem. 140,113-118. 29. JAY, M., DE LUCA, V., AND IBRAHIM, R. K. (1985) Eur. J. B&hem. X3,321-325. 30. MACK, J. P., AND SLAYTOR, M. B. (1978) J. Chre matogr. 157,153-159. 31. MEYER, E. (1982) Plant Cell Rep. 1,236-239. 32. HAUFFE, K. D., HAHLBROCK, K., AND SCHEEL, D. (1986) 2. Naturforch C 41,228~239. 33. WAT, C.-K., STEFFENS, P., AND ZENK, M. H. (1986) Z. Naturforch C 41,126-134. 34. PHILLIPSON, J. D., ROBERTS, M. F., AND ZENK, M. H. (1985) The Chemistry and Biology of Isoquinoline Alkaloids, Springer Verlag, Berlin.

ET AL. 35. IBRAHIM, R. K., DE LUCA, V., KHOURI, H., LATCHINIAN, L., BRISSON, L., AND CHAREST, P. M. (1987) Phytochemistry 26(5), 1237-1245. 36. CLARK, B. R., MURAI, J. T., POMERANZ, A., MILLS, P. A., HALPERN, R M., AND SMITH, R. A. (1975) Cancer Res. 35,1727-1733. 37. HOSHINO, J., K~HNE, U., AND KRUGER, H. (1982) B&him. Biophys. Acta 719,518-526. 38. DAMANI, L. A., SHAKER, M. S., CROOKS, P. A., GODIN, C. S., AND Nwosu, C. (1986) Xenobiotica 16(7), 645-650. 39. COBURN, S. P., AND MAHUREN, J. D. (1987) J. Biol

Chem 262(6), 2642-2644. 40. JOSHI, J. G., AND HANDLER,

Chem 237.3185-3188.

P. (1962) J. Biol.