Characterization of c-tocopherol methyltransferases from Capsicum annuum L and Arabidopsis thaliana Maria Koch 1,2 , Rainer Lemke 3 , Klaus-Peter Heise 2 and Hans-Peter Mock 1 1 Institute of Plant Genetics and Crop Plant Research, Gatersleben, Germany; 2 Albrecht-von-Haller-Institut fu ¨ r Pflanzenwissenschaften der Universita ¨ tGo ¨ ttingen, Germany; 3 Sungene GmbH & Co. KgaA, Gatersleben, Germany Tocopherols are essential micronutrients in human and animal nutrition due to their function as lipophilic anti- oxidants. They are exclusively synthesized by photosynthetic organisms including higher plants. Despite the attributed beneficial health effects and many industrial applications, research on the tocopherol biosynthetic pathway and its regulation in plants is still limited. In the work presented here we performed a detailed biochemical characterization of a c-tocopherol methyltransferase (c-TMT) from Arabidopsis thaliana and of a c-TMT purified from Capsicum annuum fruits, a tissue with high accumulation of tocopherols. The biochemical characteristics of both enzyme preparations were remarkably similar including substrate specificities. Both enzymes converted d-andc-intob-anda-tocopherol, respectively, but b-tocopherol was not accepted as a sub- strate, pointing to a specific methylation at the C(5)-position of the tocopherol aromatic head group. A kinetic analysis performedwiththeArabidopsis enzyme was consistent with an iso-ordered bi-bi type reaction mechanism. Our results emphasize the role of c-TMT in regulating the spectrum of accumulated tocopherols in plants. Keywords: Arabidopsis; Capsicum; c-tocopherol; methyl- transferase; vitamin E. a-Tocopherol belongs to a family of lipid-soluble hydrocar- bon compounds characterized by a chromanol ring with a phytyl side chain and summarized under the collective name Vitamin E. Putative biochemical functions of these com- pounds are the antioxidant properties as efficient scavengers of lipid peroxyl radicals and their action as membrane stabilizers [1]. Tocopherols have been found in all green tissues of photosynthetic organisms [2], but significant amounts are frequently observed in seeds. Plant tissues highly active in photosynthesis bear a great potential for the generation of reactive oxygen species and chloroplasts possess an elaborated protective system composed of enzy- mic and nonenzymic components [3]. It is assumed that the lipophilic tocopherols complement the antioxidative func- tion of the hydrophilic ascorbate in a concerted manner [4]. Besides their functions in plant metabolism, tocopherols are essential components of the human diet and serve as protectants in food and pharmaceutical technology [5]. Understanding the biochemical pathway of tocopherol biosynthesis therefore opens the perspective for improving the nutritional quality of crop plants [6]. Biosynthesis of tocopherols was demonstrated in plastid envelopes [7] from precursors originating from the plastidial isoprenoid path- way and from the shikimate pathway, providing the hydrophobic phytyl moiety and the polar head group homogentisic acid, respectively. Furthermore, plastidial tocopherol accumulation appears to depend on the up-regulation of genes encoding the enzymes being involved in the formation of these precursors, like 1-deoxyxylulose 5-phosphate synthase [8], geranylgeranyl reductase [9] and 4-hydroxyphenylpyruvate dioxygenase [10]. Based on earlier investigations [11] and on detailed work on the chemical synthesis of prenylquinones [12] the pathway for plastidial a-tocopherol biosynthesis has been elucidated [13,14]. The proposed pathway includes the phytylation of homogentisic acid to form 2-methyl-6-phytylquinol, the first ring methy- lation at position 3 to yield 2,3-dimethyl-5-phytylquinol, cyclization to yield c-tocopherol, and finally a second ring methylation at position 5 to yield a-tocopherol (Fig. 1). Detailed biochemical analysis of tocopherol synthesis and its regulation has largely been hampered by the lack of purified enzyme preparations catalysing individual steps of the pathway. Earlier reports have focused on the purifica- tion of c-TMT from bell pepper (Capsicum annuum)fruits [15], from spinach [16] and Euglena [17]. Consistent with previous tracer experiments these studies have shown that the c-TMT activities were membrane-associated and had to be solubilized prior to any additional purification step. A purified c-TMT enzyme preparation was reported for bell pepper indicating a molecular mass of 33 kDa for the active monomeric form [15]. Due to the instability of the solubilized enzyme, purification to homogeneity was not reported for the Euglena and spinach enzyme preparations. Recently genes encoding c-TMTs from Arabidopsis and Synechocystis have been identified [18]. Overexpression of the Arabidopsis enzyme with a seed-specific promoter resulted in a more than 80-fold increase of a-tocopherol at Correspondence to H P. Mock, Institute of Plant Genetics and Crop Plant Research, Corrensstrasse 3, D-06466 Gatersleben, Fax: + 49 39482 5139, Tel.: + 49 39482 5506, E mail: mock@ipk-gatersleben.de Abbreviations: AdoHcy, S-adenosyl- L -homocystein; AdoMet, S-adenosyl- L -methionine; c-TMT, c-tocopherol methyltransferase; toc, tocopherol. Enzymes: c-tocopherol methyltransferase (EC 2.1.1.95); accession number AF104220. (Received 23 July 2002, revised 3 October 2002, accepted 14 November 2002) Eur. J. Biochem. 270, 84–92 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03364.x the expense of c-tocopherol without changing the total content. The recombinant enzyme expressed in E. coli accepted d-, but not b-tocopherol in addition to c-toco- pherol as a substrate. In the present paper we attempted a detailed characterization of c-TMT activities with respect to kinetic properties and substrate specificities. We investigated the properties of the recombinant enzyme from Arabidopsis and of a partially purified preparation from bell pepper fruit pericarp to compare the characteristics of c-TMT enzymes from different species and tissues. For purification of c-TMT we have chosen the fruit pericarp of Capsicum which is a tissue with a high enrichment of tocopherols. Materials and methods Plant material Mature Capsicum annuum L. fruits of the red variety were obtained from a local market. Chemicals The (+) c-and(+) d-tocopherols were purchased from Sigma (Deisenhofen, Germany). Residual (+/–)-b-tocopherols were obtained from Merck (Darmstadt, Germany) and additionally checked for purity by HPLC. [ 14 C]AdoMet (1.85 MBq) was from Pharmacia Biotech (Freiburg, Ger- many) and unlabelled AdoMet and AdoHcy were from Sigma. Chromatographic materials and columns were obtained from Bio-Rad (hydroxyapatite), Phenomenex (BioSep–Sec-S3000) and Pharmacia (all others). All other chemicals were of analytical grade and obtained from various suppliers. Preparation and purification of c-TMT from pepper fruits Chromoplast membranes were isolated from 12 kg of fruit pericarp as described by Arango and Heise [19–21] and precipitated with acetone according to d’Harlingue and Camara [15]. Solubilization of c-TMT was performed as described [19–21] using 0.1% (w/v) Tween 80 as a detergent. The resulting crude protein extract was either used for the characterization of c-TMT activities or further precipitated by sequential saturation (20–50%) with ammonium sul- phate and redissolved in 0.1 M potassium phosphate buffer of pH 8 containing 1 m M dithiothreitol and 1 m M EDTA. The crude protein extract was desalted through a Sephadex G25 column (200 mL bed volume; Pharmacia Biotech, Freiburg, Germany) against buffer A [50 m M Tris; 1 m M EDTA, 3 m M dithiothreitol, 3% (v/v) glycerol; pH 7.2] and purified by subsequent chromatography (FPLC system; Pharmacia Biotech, Freiburg, Germany) starting with a DEAE-Sepharose (fast flow material) column of 200 mL equilibrated in buffer A. After removal of nonbound proteins, elution of c-TMT was performed with a linear gradient from 0–1 M NaCl in buffer A. Fractions containing c-TMT activity were concentrated and applied to a second Fig. 1. Proposed pathway for the biosynthesis of tocopherols in plants. Ó FEBS 2003 c-tocopherol methyltransferases (Eur. J. Biochem. 270)85 DEAE-Sepharose column (20 mL bed volume) equilibrated in buffer B (buffer A adjusted to pH 7.8). After washing the column was developed with a linear gradient up to 1 M NaCl in buffer B. Active fractions were pooled and subjected to chromatography on a hydroxyapatite column equilibrated in buffer C [10 m M sodium phosphate, 1 m M EDTA, 3 m M dithiothreitol, 5% glycerol (v/v), pH 7.3]. Elution of protein was performed by increasing the sodium phosphate concentration to 400 m M . Fractions with c-TMT activity were further chromatographed on Blue Sepharose equilibrated in buffer D (buffer solution B with Tris reduced to 25 m M ). After washing, a linear gradient from 0–2 M NaCl in buffer D was applied to elute bound proteins. Between column separations the active fractions were desalted on Sephadex G25 or pooled and concentrated by dialysis against polyethyleneglycol 35 000 (Merck, Darms- tadt, Germany). Other methods for enzyme concentration such as ultrafiltration led to considerable losses of enzyme activity presumably due to the unspecific binding of the hydrophobic protein. Further purification of the concen- trated labile protein was attempted by precipitation with chloroform/methanol according to Wessel and Flu ¨ gge [22] and subsequent separation by HPLC under denaturing conditions on a BioSep–Sec-S3000 (300 · 7.8 mm) gel filtration column (Phenomenex, Aschaffenburg, Germany) using 20 m M potassium phosphate buffer containing 6 M guanidine hydrochloride. Molecular mass determination The native molecular mass was determined by gel filtra- tion on a Superdex 200 HR 30/10 column (1 · 30 cm) with a0.1 M potassium phosphate buffer of pH 7 containing 1m M EDTA and 3 m M dithiothreitol at a flow rate of 0.5 mLÆmin )1 . Fractions of 1.25 mL were collected. Col- umn calibration was with a protein standard containing aldolase (160 kDa), BSA (68 kDa), ovalbumin (45 kDa), carboanhydrase (30 kDa) and myoglobin (17.8 kDa). SDS polyacrylamide gel electrophoresis The samples were dissolved in a buffer medium containing 56 m M Na 2 CO 3 , 56 m M dithiothreitol, 2 m M EDTA, 2% (v/v) SDS, 12% (w/v) sucrose and 0.25% (w/v) bromophe- nol blue, incubated for 5 min at 95 °C and centrifuged in order to remove insoluble residues. Electrophoresis was according to Laemmli [23]. The gels were loaded with either 15 lgor0.5–3lg protein, electrophoresed and stained with Coomassie blue or silver according to Jungblut and Seifert [24]. Protein markers were from the LMW calibration kit of Pharmacia Biotech (Freiburg, Germany). Protein determination Protein was measured according to Bradford [25] using the reagent solution from Bio-Rad (Munich, Germany) and BSA as standard protein. Photolabelling of Capsicum c-TMT Radioactive assays with [ 14 C]AdoMet for c-TMT from the last purification step were performed with 20 lgprotein under UV-irradiation for 2 h according to Subbaramaiah and Simms [26]. The protein was precipitated and re-dissolved as described by Wessels and Flu ¨ gge [22] and separated by SDS/PAGE. Radioactively labelled proteins were visualized using a Phosphoimaging system (Storm system; Amersham Biotech, Freiburg, Germany). Purification of the recombinant c-TMT from Arabidopsis thaliana An E. coli strain for overexpressing Arabidopsis c-TMT [18] was a generous gift of SunGene GmbH & Co. KGaA company, Gatersleben, Germany. After harvesting the induced cells the recombinant protein was released by ultrasonication (6 · 15 s) of the cells in an ice-cold buffer medium (50 m M NaH 2 PO 4 , 300 m M NaCl, 10 m M imida- zol, 800 lg lysozyme; pH 8.0) and subsequent centrifuga- tion at 15 000 and 30 000 g, respectively. Purification was performed using an FPLC system on a Ni-agarose column (5–10 mg protein per ml Qiagen Ni-NTA Superflow; 1 · 10 cm; flow rate: 0.5 mLÆmin )1 ; 10 mL fractions) by stepwise elution with increasing imidazol concentrations in the buffer medium according to the manufacturer’s proto- col. The enzyme activity was preserved by additions of 10–20% glycerol or 3.8 M (NH 4 ) 2 SO 4 during storage of aliquots prior to subsequent enzyme assays. Assay conditions and analytical methods The assay for the Capsicum enzymeisbasedonthe methylation of exogenous c-intoa-tocopherol in the presence of [methyl- 14 C]AdoMet. The reactions were car- ried out for 2 h at 25 °Cin500lL medium containing 50 m M Tricine/NaOH (pH 7.5), 1 m M MgCl 2 , 50 l M c-tocopherol, 25 l M [ 14 C]AdoMet and 0.1–0.7 mg protein. c-Tocopherol or other tocopherols used as substrate were added from concentrated stock solutions in ethanol into the enzyme assays. The reaction products were extracted according to Arango and Heise [19,20] and separated on HPTLC-silicagel 60 plates (Merck, Darmstadt, FRG) with toluene as the solvent. The product formation was moni- tored using a Phosphoimager system (Storm system; Amersham Biotech, Freiburg, Germany). The recombinant enzyme was measured in a modified nonradioactive assay containing 50 m M Tris/HCl (pH 8.5), 25 l M AdoMet, 50 l M c-tocopherol, 5 m M dithiothreitol and 1–5 lg of the purified enzyme protein in a total volume of 500 lL. After termination the assay was processed as described above except that the residues of the organic phase were dissolved in methanol. The a-tocopherol content was quantified after HPLC (Waters 2690 Separation Module) separation on a Prontosil 200–3-C30-column (Bischoff Chromatography; NC; 230 · 4.6 mm, 3.0 lm) by fluorescence detection (Jasco FP-920 detector; k ex : 295 nm and k em : 332 nm). Elution of tocopherols was isocratically with 100% methanol at a flow rate of 1mLÆmin )1 . Enzyme kinetics The experiments were performed by varying the concentra- tion of substrates in the standard assay and by adding 86 M. Koch et al.(Eur. J. Biochem. 270) Ó FEBS 2003 appropriate amounts of inhibitors (product inhibition experiments). Details are given in the individual experiments in the results section. Data were analyzed by linear regression using the statistic program of MS Office EXCEL (Microsoft, Deisenhofen, Germany). Statistics Substrate interaction kinetic experiments and product inhibition reactions were performed at least six times. All other experiments were conducted at least three times. Results Purification of c-TMT activity from Capsicum annuum fruits A crude protein extract was prepared by acetone precipi- tation from red bell pepper fruits characterized by the highest specific c-TMT activity when compared with other fruit varieties [27]. The crude protein extract could be stored at )20 °C without loss of c-TMT activity for 4 weeks, but approximately half of the activity was lost when the extract waskeptat4°C for 5 days. After solubilization and ammonium sulphate precipitation (saturation up to 50%) the crude c-TMT was further enriched by anion exchange (twice) followed by chromatography on hydroxyapatite and Blue Sepharose (Table 1). Additional gel filtration of the native enzyme on Superdex 200 showed no further purifi- cation effect. In total an approximately 45-fold purification with a 9% recovery of c-TMT activity was achieved. The enzyme purification during the subsequent steps was assessed by SDS gel electrophoreses as shown in Fig. 2 demonstrating the effectiveness of individual purification steps. All attempts to further enrich the native c-TMT by for example anion exchange chromatography or affinity chromatography on Adenosine-Sepharose in order to obtain an apparently pure fraction were hampered by the loss of enzyme activity. Addition of detergents into the buffer solutions did not stabilize enzyme activity (data not shown). Chromatography on reversed phase material led to severe loss of protein presumably by interactions with the gel matrix (data not shown). Further separation of proteins contained in the most purified active enzyme fraction was only achieved under denaturing conditions by HPLC on BioSep–Sec-S3000 (manufacturer) according to their mo- lecular size, but protein amounts were not sufficient for sequencing of candidate protein bands with a molecular mass predicted from gel filtration and photoaffinity labelling (data not shown). Addition of divalent cations, BSA and yolk lipids had no protective influence on the stability of the enzyme (data not shown). Molecular mass determination Gel filtration (Fig. 3) and photoaffinity labelling followed by SDS/PAGE were used to determine the molecular mass of c-TMT (Fig. 4). When applying the crude protein extract obtained after acetone precipitation to a Superdex 200 gel filtration column, c-TMT activity eluted in the high molecular mass fraction with a native molecular mass of more than 600 kDa. In contrast a mass of approximately 36 kDa was observed when the most purified fractions after affinity chromatography were analyzed (Fig. 3). It was tentatively concluded that this mass would represent the monomeric state of the enzyme. To further corroborate this assumption we used photoaffinity labelling and SDS/PAGE as an additional method for molecular mass determin- ation. For photoaffinity labelling of c-TMT a fraction after Blue Sepharose column purification was used. During the enzyme assay UV light (254 nm) was applied to enable the eventual covalent binding of radioactively labelled substrate to a fraction of the c-TMT as already demonstrated for other methyltransferases [26–28]. After termination of the Table 1. Purification protocol of c-TMT from red Capsicum fruits. Fraction Volume (mL) Total protein (mg) Total activity (fkat) Specific activity (fkatÆmg )1 Æprotein) Recovery (%) Purification (fold) Chromoplast membranes 500 2103 18164 9.5 100 1 Acetone precipitate 98 1161 20501 17.7 102 2 50% (NH 4 ) 2 SO 4 54 610 24086 37.0 113 4 DEAE (I) 110 187 15481 82.8 85 9 DEAE (II) 54 70 10834 154.1 60 16 Hydroxyapatite 48 19 4541 235.4 25 25 Blue Sepharose 92 4 1647 426.8 9 45 Fig. 2. SDS/PAGE analysis of fractions obtained during subsequent steps of c-TMT purification from Capsicum fruits. From each step, 1 lg of protein was loaded on the gel. Proteins were visualized by silver staining. The following abbreviations are used for labelling of the lanes: CM, chromoplast membrane; AE, acetone precipitate, AS, ammonium sulphate precipitate; IE I + II, active fractions from subsequent DEAE sepharose columns; HA, active fractions eluted from the hydroxyapatite column; AF, active fractions obtained after chromatography on Blue Sepharose; M, molecular mass marker. Ó FEBS 2003 c-tocopherol methyltransferases (Eur. J. Biochem. 270)87 reaction, proteins were separated on SDS gel electrophor- esis. Imaging analysis revealed the presence of one single labelled protein band with a molecular mass of approxi- mately 36 kDa (Fig. 4A). Comparison of the properties of the partially purified c-TMT from Capsicum fruits with the recombinant c-TMT from Arabidopsis Recombinant Arabidopsis c-TMT containing a His-tag was purified from E. coli cell lysates by affinity chromatography. Analysis of the purified fraction by SDS/PAGE showed a single band with the expected molecular mass (Fig. 4B). Like the c-TMT from Capsicum fruits the Arabidopsis enzyme was slightly stimulated by dithiothreitol and was not dependent on divalent cations (data not shown). The pH dependence of both c-TMT sources was evaluated in the range of 5.5–10.0 with different buffer systems (Fig. 5) as described under materials and methods. The recombinant c-TMT from A. thaliana showed a more alkaline and sharper pH-optimum at pH 8.5 than the partially purified enzyme from pepper pericarp which showed a broader curve with a maximum at pH 7.5. Stability tests by preincubating the Arabidopsis enzyme at different pH values followed by assaying the activity at 8.5 indicated that the sharp decline of activity towards lower pH values was only partially due to enzyme inactivation (data not shown). For both enzyme preparations the methyltransferase reaction showed an identical temperature maximum of approximately 34 °C (data not shown). Substrate specificities To elucidate putative differences in the molecular properties of both enzymes, a detailed investigation of their substrate specificities was performed with the recombinant Arabidop- sis enzyme and a c-TMT fraction from Capsicum fruits obtained by solubilizing the acetone precipitate. Both c-TMT preparations were incubated with different Fig. 3. Elution profile of c-TMT on a Super- dex-200 gel filtration column. The insert shows the calibration curve obtained by using standard proteins (aldolase, 160 kDa; bovine serum albumin, 68 kDa; ovalbumin, 45 kDa; carboanhydrase, 30 kDa; myoglobin 17.8 kDa). (s), enzyme activity; (n), protein. Fig. 4. Photoaffinity labelling of Ca psicum c-TMT band from a Blue Sepharose column fraction (A) and SDS/PAGE analysis of puri- fied recombinant Arabidopsis c-TMT (B). (A) The protein extract (20 lg) was incubated with 14 l M [ 14 C]AdoMet under UV-irradi- ation. After termination of the assay the pro- tein fraction was separated by SDS/PAGE. Radioactively labelled proteins were visualized by phosphoimaging. (B) SDS/PAGE of recombinant c-TMT from A. thaliana purified by chromatography on Ni-agarose, loaded with 5 lgofpurifiedc-TMT. After electro- phoresis the gel was stained with Coomassie Brilliant blue. 88 M. Koch et al.(Eur. J. Biochem. 270) Ó FEBS 2003 methyl-substituted tocopherols. In the case of c-TMT from pepper the labelled reaction products were separated by HPTLC and visualized by phosphoimaging (Fig. 6). The unlabelled reaction products after incubation of the Ara- bidopsis enzyme were separated by HPLC and detected fluorimetrically. Both enzyme preparations showed the conversion of c-toa-tocopherol and of d-tob-tocopherol, respectively, whereas b-tocopherol was not accepted as substrate (Fig. 6; Table 2). The acceptance of c-tocopherol and of d-tocopherol by the crude TMT preparation was also kept through all the subsequent purification steps (data not shown). These results indicated that the final methylation step leading to the formation of a-andb-tocopherol, respectively, is exerted by one enzyme. Kinetic properties The properties of both enzyme preparations were further compared by a thorough analysis of kinetic parameters. The c-TMT activities from both sources showed regular Micha- elis–Menten behaviour for all substrates tested (data not shown). For all substrates investigated, the pepper c-TMT preparation showed very similar K m values and V max to K m ratios (Table 2). For the Arabidopsis enzyme the V max /K m - quotient was twofold higher for d- than for c-tocopherol. Kinetics of Arabidopsis TMT The following kinetic analysis was performed for the forward reaction in the presence and absence of inhibitors. Substrate interaction kinetic experiments were performed by varying one substrate at different fixed concentrations of the other substrate (Fig. 7). When either c-tocopherol or AdoMet were varied double reciprocal plots yielded lines converging to the left of the ordinate axis. Secondary plots of V )1 intercepts and of slopes were linear for either of the substrates. These results indicated that the c-TMT methy- lation reaction follows a sequential reaction mechanism and are accordingly not consistent with a ping-pong mechanism. Product inhibition studies In order to discriminate between the possible kinetic mechanisms suggested by the initial velocity studies, product inhibition experiments were performed with either one of the products of the reaction, a-tocopherol or AdoHcy (Fig. 8). Variation of both c-tocopherol or AdoMet as substrates in the presence of either a-tocopherol or AdoHcy as inhibitors always led to a noncompetitive inhibition (Fig. 8). This pattern of product inhibition is consistent by assuming that the methylation reaction follows an iso-ordered bi-bi mechanism. Fig. 6. Substrate specificity of partially purified TMT from Capsicum. Enzyme assays were performed with [ 14 C]AdoMet in the presence of d-tocopherol (lane A) or c-tocopherol (lane B). Reaction products were separated by HPTLC and visualized by phosphoimaging. Prod- uct formation was verified by cochromatography with nonlabelled a-andb-tocopherol standards, which were detected by their fluores- cence under UV-light. In a control reaction, tocopherol was omitted as a substrate. Fig. 5. Influence of pH on c-TMT activity. Assays were performed at different pH values in the following buffers: Mes, pH 5.5–6.5; potas- sium phosphate, pH 6.5–8.0; Tris/HCl, pH 7.5–9.0; carbonate buffer, pH 9.2–10.0; (r), partially purified c-TMT from Capsicum;(d) Ara- bidopsis c-TMT. Table 2. Kinetic parameters of c-TMT partially purified from Capsicum fruits and of a recombinant c-TMT (A. thaliana ). Data sets were evaluated according to the method of Hanes–Wilkinson; n.d., not determined. Substrate K m [l M ] V max /K m [fkatÆmg )1 proteinÆl M )1 ] Capsicum Arabidopsis Capsicum Arabidopsis c-Tocopherol 3.1 ± 0.5 (n ¼ 6) 5.4 ± 0.6 (n ¼ 4) 15.8 2700 d-Tocopherol 2.9 ± 0.7 (n ¼ 5) 3.3 ± 0.5 (n ¼ 4) 12.8 6500 b-Tocopherol no product no product – – [ 14 C]-AdoMet 2.0 ± 0.4 (n ¼ 5) 5.2 ± 1.4 (n ¼ 4) n.d. n.d. Ó FEBS 2003 c-tocopherol methyltransferases (Eur. J. Biochem. 270)89 Discussion Our paper describes the first thorough characterization of the enzymic properties of c-TMTsfromhigherplants.The present purification protocol for c-TMT from pepper fruits was initially based on a previously published scheme [15]. Despite several modifications we were not able to purify c-TMT to complete homogeneity (Fig. 2) although the purification factor of 45 was similar to previously published results [15]. Analysis of the most enriched fraction by SDS/ PAGE and sensitive silver staining revealed the presence of a range of bands. A faint band at the expected molecular mass of 36 kDa was visible but obviously representing only a small portion of the total protein content of this fraction. To this end it remains unclear how c-TMT from pepper fruits could be purified to apparent homogeneity by a 69-fold enrichment starting from a crude membrane preparation as described in a previous publication [15]. The relatively high native molecular mass of more than 600 kDa, estimated for the crude pepper c-TMT after the first purification steps resembles the earlier findings of d’Harlingue and Camara [15] and may be, indeed, due to the tendency of membrane proteins to aggregate. This aggregation phenomenon may also explain the instability of the membrane enzyme in the diluted state and the high loss of enzyme activities during the subsequent purification procedure. In contrast, gel filtration of the purified protein at the end of the conventional purification procedure suggests a native molecular mass for the c-TMT from pepper pericarp of approximately 36 kDa (Fig. 3). This molecular mass for the monomer is supported by photoaf- finity labelling [28,29] of the pepper enzyme after SDS-gel electrophoresis of the Blue Sepharose column fraction (Fig. 4A) and agrees with the molecular mass of c-TMT from A. thaliana (Fig. 4B). The presence of only one labelled band is also indicative that the highly aggregated form of c-TMT observed during initial steps of purification contains only one protein involved in methylation of tocopherol. The low protein amount of the 36 kDa-band from pepper (Fig. 2) was not sufficient to obtain sequence information by EDMAN degradation. The 200-fold higher specific activity of the Arabidopsis c-TMT compared with the partially purified pepper enzyme also reflects the degree of purity as well as the loss of activity during lengthy conventional protein purification procedures. In spite of significant differences in the purification degree of the c-TMTs from Capsicum and Arabidopsis, both enzyme sources show remarkable conformities with respect to temperature maxima and pH-optima (Fig. 5), substrate specificities and kinetic parameters (Fig. 3; Table 2). Our data are consistent with the selected parameters from previously published initial studies on c-TMT from pepper and Euglena [15–18]. Both enzyme preparations accepted c-andd-tocopherol, but not b-tocopherol as a substrate. This observation points to the specific methylation by this enzyme at the C(5)-position (i.e. in ortho-position to the Fig. 7. Substrate interaction kinetics of Arabidopsis c-TMT. Left panels: Lineweaver–Burk-plots for the two-substrate reaction of c-TMT with (A) 1/v against 1/[AdoMet] with c-tocopherol at various fixed concentrations and (B) 1/v against 1/[c-tocopherol] with AdoMet at various fixed concentrations. Right panels: slope and intercept replots corresponding to A (upper two) or B (lower two) on the left panel. 90 M. Koch et al.(Eur. J. Biochem. 270) Ó FEBS 2003 prenyl residue) of the tocopherol aromatic head group, recently described by Shintani and DellaPenna [18] and shown in Fig. 1. Calculation of the V max /K M ratios for c-andd-tocopherol showed similar values for the pepper enzyme. For Arabidopsis c-TMT a more than twofold higher value was deduced for d-tocopherol relative to c-tocopherol indicating a higher catalytic efficiency for this substrate. Initial velocity experiments in the absence of inhibitors with variable concentrations of c-tocopherol or AdoMet (Fig. 7) suggested that the methylation reaction follows a sequential and not a ping-pong type of reaction mechanism. In the product inhibition studies all substrate and inhibitor combinations investigated resulted in a noncompetitive inhibition pattern (Fig. 8) which is consistent with an iso- ordered bi-bi mechanism of the methylation reaction. The mechanism is a special case of the ordered bi-bi mechanism, which is a consequence of an isomerization of the enzyme in the central complex [30]. Kinetic analysis of methyltrans- ferases have revealed sequential as well as ping-pong mechanisms [31–33]. Two closely related methyltransferases involved in the biosynthesis of isoquinoline alkaloids displayed different types of reaction mechanisms [31]. Experimental techniques such as presteady state kinetic analysis, isotope–partitioning experiments and the use of mutants were applied to explore the kinetic and catalytic properties of methyltransferase reactions in more detail [33] and will help to further define the reaction mechanism of c-TMT. It has been recently shown that seed-specific overexpres- sion of a homogentisate phytyl transferase led to increased tocopherol levels in transgenic Arabidopsis lines [34] whereas overexpression of c-TMT resulted in a shift from c-to a-tocopherol [18]. As individual tocopherols have different properties, a detailed characterization of further enzymic steps in the tocopherol biosynthetic pathway such as shown here for c-TMT will be fundamental to support the rational design of engineered crop plants with modified profiles of tocopherols. Interplay between already known proteins and yet unknown factors will be elucidated by protein interac- tion studies using approaches such as the yeast two-hybrid system or pull-down assays. Analysis of transgenic lines and mutants with modified activities of individual components such as c-TMT will enable the study of the regulatory processes of the tocopherol biosynthetic pathway in planta. Acknowledgements This work was supported by grants of the SunGene GmbH & Co. KGaA company, Gatersleben, Germany to M.K., K P.H. and H P.M. References 1. Wang, X. & Quinn, P.J. (1999) Vitamin E and its function in membranes. Prog. Lipid Res. 38, 309–336. 2. Hess, J.L. (1993) Vitamin E, a-tocopherol. In Antioxidants in Higher Plants (Alscher, R. & Hess, J., eds), pp. 111–134. CRC Press, Boca Raton, USA. Fig. 8. Product inhibition kinetics of Ar abidops is c-TMT. Hanes–Wilkinson-plots are shown for the product inhibition of c-TMT by AdoHcy (upper panels) and c-tocopherol (lower panels). The set of data correspond to one of six independent experiments. All data points are derived from duplicate assays. Toc, tocopherol. (A1) [c-tocopherol]/v vs. [c-tocopherol] at various fixed concentrations of AdoHcy, (A2) [AdoMet]/v vs./ [AdoMet] at various fixed concentrations of AdoHcy. (B1) [c-tocopherol]/v vs. [c-tocopherol] at various fixed concentrations of a-tocopherol, (B2) [AdoMet]/v vs./[AdoMet] at various fixed concentrations of a-tocopherol. Ó FEBS 2003 c-tocopherol methyltransferases (Eur. J. Biochem. 270)91 3. Foyer, C.H., Decourvieres, P. & Kunert, K.J. (1994) Protection against oxygen radicals: an important defence mechanism studied in transgenic plants. Plant Cell Environ. 17, 507–523. 4. Fryer, M.J. (1992) The antioxidant effects of thylakoid vitamin E (a-tocopherol). Plant Cell Environ. 15, 381–392. 5. Traber, M.G. & Sies, H. (1996) Vitamin E in humans: demand and delivery. Annu. Rev. Nutr. 16, 321–347. 6. Grusak, M.A. & DellaPenna, D. (1999) Improving the nutrient composition of plants to enhance human nutrition and health. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 133–161. 7. Soll, J., Douce, R. & Schultz, G. (1980) Site of biosynthesis of a-tocopherol in spinach chloroplasts. FEBS Lett. 112, 243–246. 8. Bouvier, F., D’Harlingue, A., Suire, C., Backhaus, R.A. & Camara, B. (1998) Dedicated roles of plastid transketolases during the early onset of isoprenoid biogenesis in pepper fruits. Plant Physiol. 117, 1423–1431. 9. Keller, Y., Bouvier, F., D’Harlingue, A.D. & Camara, B. (1998) Metabolic compartmentation of plastid prenyllipid biosynthesis. Evidence for the involvement of a multifunctional geranylgeranyl reductase. Eur. J. Biochem. 251, 413–417. 10. Norris, S.R., Shen, X. & DellaPenna, D. (1998) Com- plementation of the Arabidopsis pds1 mutation with the gene encoding p-hydroxyphenylpyruvate dioxygenase. Plant Physiol. 117, 1317–1323. 11. Whistance, G.R. & Threlfall, D.R. (1970) Biosynthesis of iso- prenoid quinones and chromanols. In Aspects of Terpenoid Chemistry and Biochemistry (Goodwin, T.W., ed.), pp. 357–404. Academic Press, London, UK. 12. Mayer, H. & Isler, O. (1971) Synthesis of vitamins E. Methods Enzymol. 18C, 241–348. 13. Soll, J. & Schultz, G. (1980) 2-Methyl-6-phytylquinol and 2,3- dimethyl-5-phytylquinol as precursors of tocopherol synthesis in spinach chloroplasts. Phytochemistry 19, 215–218. 14. Soll, J., Schultz, G., Joyard, J., Douce, R. & Block, M.A. (1985) Localization and synthesis of prenylquinones in isolated outer and inner envelope membranes from spinach chloroplasts. Arch. Bio- chem. Biophys. 238, 290–299. 15. D’Harlingue, A. & Camara, B. (1985) Plastid enzymes of ter- penoid biosynthesis. Purification and characterization of c-tocopherol methyltransferase from Capsicum chromoplasts. J. Biol. Chem. 260, 15200–15203. 16. Ishiko, H., Shigeoka, S., Nakano, Y. & Mitsunaga, T. (1992) Some properties of c-tocopherol methyltransferase solubilized from spinach chloroplasts. Phytochemistry 31, 1499–1500. 17. Shigeoka, S., Ishiko, H., Nakano, Y. & Mitsunaga, T. (1992) Isolation and properties of c-tocopherol methyltransferase in Euglena gracilis. Biochim. Biophys. Acta 1128, 220–226. 18. Shintani, D. & DellaPenna, D. (1998) Elevating the vitamin E content of plants through metabolic engineering. Science 282, 2098–2100. 19. Arango, Y. & Heise, K P. (1997) a-Tocopherol synthesis by Capsicum fruit chromoplasts. J. Plant Physiol. 150, 509–513. 20. Arango, Y. & Heise, K P. (1998) Localization of a-tocopherol synthesis in chromoplast envelope membranes of Capsicum ann- uum L. fruits. J. Exp. Bot. 49, 1259–1262. 21. Arango, Y. & Heise, K P. (1998) Tocopherol synthesis from homogentisate in Capsicum annuum L. (yellow pepper) chromo- plast membranes: evidence for tocopherol cyclase. Biochem. J. 336, 531–533. 22. Wessels, D. & Flu ¨ gge, U.I. (1984) A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem. 138, 141–143. 23. Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. 24. Jungblut, P.R. & Seifert, R. (1990) Analysis by high-resolution two-dimensional electrophoresis of differentiation-dependent alteration in cytosolic protein pattern of HL-60 leukemic cells. J. Biochem. Biophys. Methods 21, 47–58. 25. Bradford, M.M. (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dry binding. Anal. Biochem. 72, 248–254. 26. Subbaramaiah, K. & Simms, S.A. (1992) Photolabeling of CheR methyltransferase with S-adenosyl- L -methionine (AdoMet). Stu- dies on the AdoMet binding site. J. Biol. Chem. 267, 8636–8642. 27. Koch, M., Arango, Y., Mock, H P. & Heise, K.P. (2002) Factors influencing a-tocopherol synthesis in pepper fruits. J. Plant Physiol. 159, 1015–1019. 28. Teyssier, E., Block, M.A., Douce, R. & Joyard, J. (1996) Is E37, a major polypeptide of the inner membrane from plastid envelope, an S-adenosyl methionine-dependent methyltransferase? Plant J. 10, 903–912. 29. Herbik, A., Koch, G., Mock, H P., Dushkov, D., Czihal, A., Thielmann, J., Stephan, U.W. & Ba ¨ umlein, H. (1999) Isolation, characterization and cDNA cloning of nicotianamine synthase from barley. A key enzyme for iron homeostasis in plants. Eur. J. Biochem. 265, 231–239. 30. Cleland, W.W. (1963) The kinetics of enzyme-catalyzed reactions with two or more substrates. Biochim. Biophys. Acta 67, 104–137. 31. Morishige, T., Tsujita, T., Yamada, Y. & Sato, F. (2000) Mole- cular characterization of the S-Adenosyl- L -methionine: 3¢-iso- quinoline alkaloid biosynthesis in Coptis japonica. J. Biol. Chem. 275, 23398–23405. 32. Jay, M., De Luca, V. & Ibrahim, R.K. (1985) Purification, properties and kinetic mechanism of flavonol 8-O-methyl- transferase from Lotus corniculatus L. Eur. J. Biochem. 153, 321–325. 33. Vilkaitis, G., Merkiene, E., Serva, S., Weinhold, E. & Klima- sauskas, S. (2001) The mechanism of DNA cytosine-5 methyla- tion. J. Biol. Chem. 276, 20924–20934. 34. Savidge, B., Weiss, J.D., Wong, Y.H.H., Lassner, M.W., Mitsky, T.A., Shewmaker, C.K., Post-Beittenmiller, D. & Valentin, H.E. (2002) Isolation and characterization of homogentisate phytyl- transferase genes from Synechocystis sp. PCC 6803 and Arabi- dopsis. Plant Physiol. 129, 321–332. 92 M. Koch et al.(Eur. J. Biochem. 270) Ó FEBS 2003 . Characterization of c-tocopherol methyltransferases from Capsicum annuum L and Arabidopsis thaliana Maria Koch 1,2 , Rainer Lemke 3 , Klaus-Peter. single labelled protein band with a molecular mass of approxi- mately 36 kDa (Fig. 4A). Comparison of the properties of the partially purified c-TMT from Capsicum fruits