* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Functional characterization of LePGT1, a membrane
Lipid signaling wikipedia , lookup
NADH:ubiquinone oxidoreductase (H+-translocating) wikipedia , lookup
Magnesium in biology wikipedia , lookup
Silencer (genetics) wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Western blot wikipedia , lookup
Enzyme inhibitor wikipedia , lookup
Point mutation wikipedia , lookup
Protein–protein interaction wikipedia , lookup
Expression vector wikipedia , lookup
Deoxyribozyme wikipedia , lookup
Genetic code wikipedia , lookup
Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup
Magnesium transporter wikipedia , lookup
Specialized pro-resolving mediators wikipedia , lookup
Evolution of metal ions in biological systems wikipedia , lookup
Protein structure prediction wikipedia , lookup
Two-hybrid screening wikipedia , lookup
Catalytic triad wikipedia , lookup
Proteolysis wikipedia , lookup
Biosynthesis of doxorubicin wikipedia , lookup
Biochemistry wikipedia , lookup
Metalloprotein wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Biochem. J. (2009) 421, 231–241 (Printed in Great Britain) 231 doi:10.1042/BJ20081968 Functional characterization of LePGT1, a membrane-bound prenyltransferase involved in the geranylation of p -hydroxybenzoic acid Kazuaki OHARA*1 , Ayumu MUROYA†, Nobuhiro FUKUSHIMA† and Kazufumi YAZAKI*2 *Laboratory of Plant Gene Expression, Research Institute for Sustainable Humanosphere, Kyoto University, Uji 611-0011, Japan, and †Science & Technology Systems, Inc., 1-20-1 Shibuya, Tokyo 150-0002, Japan The AS-PT (aromatic substrate prenyltransferase) family plays a critical role in the biosynthesis of important quinone compounds such as ubiquinone and plastoquinone, although biochemical characterizations of AS-PTs have rarely been carried out because most members are membrane-bound enzymes with multiple transmembrane α-helices. PPTs [PHB (p-hydroxybenzoic acid) prenyltransferases] are a large subfamily of AS-PTs involved in ubiquinone and naphthoquinone biosynthesis. LePGT1 [Lithospermum erythrorhizon PHB geranyltransferase] is the regulatory enzyme for the biosynthesis of shikonin, a naphthoquinone pigment, and was utilized in the present study as a representative of membrane-type AS-PTs to clarify the function of this enzyme family at the molecular level. Site-directed mutagenesis of LePGT1 with a yeast expression system indicated three out of six conserved aspartate residues to be critical to the enzymatic activity. A detailed kinetic analysis of mutant enzymes revealed the amino acid residues responsible for substrate binding were also identified. Contrary to ubiquinone biosynthetic PPTs, such as UBIA in Escherichia coli which accepts many prenyl substrates of different chain lengths, LePGT1 can utilize only geranyl diphosphate as its prenyl substrate. Thus the substrate specificity was analysed using chimeric enzymes derived from LePGT1 and UBIA. In vitro and in vivo analyses of the chimeras suggested that the determinant region for this specificity was within 130 amino acids of the N-terminal. A 3D (three-dimensional) molecular model of the substrate-binding site consistent with these biochemical findings was generated. INTRODUCTION structure of a representative enzyme, NphB (Orf2), was revealed [21,22], although there are no plant homologues of this soluble type AS-PT. In contrast, little detailed molecular characterization of membrane-bound AS-PTs has been reported regardless of origin. In particular, no characterization of plant AS-PTs has been reported. This is mainly because of the difficulty in handling these proteins. Representative of membrane-bound AS-PTs are the PPT [PHB (p-hydroxybenzoic acid) prenyltransferase] group. Genes coding for these enzymes have been identified in many organisms from prokaryotes to eukaryotes, and are mostly responsible for UQ biosynthesis [23–28]. The reaction catalysed by PPT is proposed to be a rate-limiting step in UQ biosynthesis [29], and all members characterized thus far show a broad specificity for prenyl diphosphates, in accepting substrates of different chain lengths. The only exception known to date is LePGT1 (Lithospermum erythrorhizon PHB geranyltransferase) [30], which only accepts GPP (geranyl diphosphate) and is involved in the biosynthesis of a red naphthoquinone, shikonin, a plant secondary metabolite. LePGT1 was reported to be the key regulatory enzyme of shikonin biosynthesis [31] and is not relevant to the formation of UQ [30]. The present study utilized LePGT1 to characterize the enzymatic function of membrane-bound AS-PTs by site-directed mutagenesis, as it has several advantages over other membranetype AS-PTs. For example, the expression of the recombinant protein in yeast was established, with which the enzymatic activity can be analysed quantitatively with a wide dynamic range using GPP as the native prenyl substrate, and it is an ER (endoplasmic The prenyltransferase family contains a large number of membrane-intrinsic proteins, which represent enzymes that accept various aromatic substrates. These AS-PTs (aromatic substrate prenyltransferases) catalyse the substitution of an aromatic proton with a prenyl group, leading to the formation of benzoand naphtho-quinones as well as prenylated polyphenols, which play various important biological roles in organisms ranging from bacteria to humans. For example, AS-PTs catalyse the key reaction in the biosynthesis of naturally occurring quinines, such as UQ (ubiquinone) [1], plastoquinone [2,3], and also vitamin E [4,5], in which their basic skeletons are formed via prenylated aromatic intermediates. In addition, prenylation reactions contribute to the diversification of plant aromatic natural products in terms of chemical structure and biological activity. For example, flavonoids [6–11], coumarins [12,13], and polyketides [14] are frequently prenylated, and cannabinoid [15] also represents a biologically active prenylated aromatic in higher plants. Some prenylated flavonoids are involved in plant defence mechanisms as phytoalexins [16,17], and some are beneficial to human health, exhibiting anti-cancer, anti-bacterial and antityrosinase activities etc., leading to their potential use as natural medicines [18–20]. Due to these attractive properties of prenylated compounds and the crucial role of prenyl moieties in their biological activities, intensive studies have been carried out. In fact, some soluble-type AS-PT genes involved in antibiotic biosynthesis have been cloned from Streptomyces and the crystal Key words: aromatic substrate prenyltransferase, coenzyme Q, mutagenesis, shikonin, substrate specificity, ubiquinone. Abbreviations used: 3D, three-dimensional; AS-PT, aromatic substrate prenyltransferase; FARM, first aspartate-rich motif; FPP, farnesyl diphosphate; GPP, geranyl diphosphate; LC, liquid chromatography; LePGT1, Lithospermum erythrorhizon PHB geranyltransferase; PHB, p -hydroxybenzoic acid; PPT, PHB prenyltransferase; SARM, second aspartate-rich motif; UQ, ubiquinone. 1 Present address: Central Laboratories for Frontier Technology, Kirin Holdings Co., Ltd., 1-13-5 Fukuura Kanazawa-ku, Yokohama-shi, Kanagawa 236-0004, Japan. 2 To whom correspondence should be addressed (email [email protected]). c The Authors Journal compilation c 2009 Biochemical Society 232 Table 1 K. Ohara and others The oligonucleotide sequences are listed below, where underlining indicates non-native restriction sites for subcloning Oligonucleotide Sequence PGT1_EcoRI_Fw PGT1-myc_XhoI_Rv UBIA_EcoRI_Fw UBIA-myc_XhoI_Rv LePGT1_PstI_Fw LePGT1_MfeI_Rv UBIA_XbaI_Fw UBIA_XbaI_Rv UBIA_SacI_Fw UBIA_SacI_Rv 5 -cgcgaattcAGAAATGGTTTCCAGCAAACAAAC-3 5 -cgctcgagCTACAGATCCTCTTCTGAGATGAGTTTTTGTTCAGGAAACAATCTTCCAAGTAAG-3 5 -cgcgaattcAGAAATGGAGTGGAGTCTGACGCAGAATA-3 5 -cgctcgagCTACAGATCCTCTTCTGAGATGAGTTTTTGTTCGAAATGCCAGTAACTCATTGCCAG-3 5 -cgctgcagATGGTTTCCAGCAAACAAACACAG-3 5 -cgcaattgGCTTGTCTAACCTTGCTAGATGAGC-3 5 - ggtctagaCCACTTCCCAGCGGCGCG -3 5 - ggtctagaCGCCGTGCGCTTAACATG-3 5 - gcgagctCCAATGGCTTTTGCCGCTG-3 5 -gcgagctccCCAGCCAAACGCCGCGCCC-3 reticulum)-localized protein without a signal peptide, which can cause trouble in yeast [30]. To clarify the molecular basis of the difference in prenyl-substrate specificity, several chimeric enzymes were also prepared using GPP-specific LePGT1 and a UQ biosynthetic PPT, UBIA with broad substrate specificity [23,32], and enzymatic function was characterized using an in vitro enzyme assay and in vivo complementation test with a yeast mutant. Moreover, a 3D (three-dimensional) molecular model of the substrate-binding domain of LePGT1 was generated, which is consistent with the biochemical data obtained. EXPERIMENTAL Construction of expression vectors for yeast A pDR196 vector [33] was used for the expression of LePGT1 in yeast, also for point mutants of LePGT1, UBIA, and chimeric enzymes. All recombinant enzymes were expressed as fusion proteins with a c-Myc tag at the C-terminus for immunodetection. Addition of the nucleotide sequence encoding the c-Myc epitope was carried out by PCR using the following primers, PGT1_EcoRI_Fw, PGT1-Myc_XhoI_Rv, UBIA_EcoRI_Fw and UBIA-Myc_XhoI_Rv, and the sequences are listed in Table 1. The PCR products were subcloned into pDR196 via the EcoRI and XhoI restriction sites. Site-directed mutagenesis was performed using the QuikChange site-directed mutagenesis kit (Stratagene) according to the manufacturer’s instructions. The PCR template for the mutagenesis was pDR196-LePGT1, and the introduction of the mutations was confirmed by nucleotide sequencing of the entire LePGT1 gene. The mutagenic oligonucleotides designed to produce the desired point mutations are listed in Supplementary Figure S1 (at http://www.BiochemJ.org/bj/421/bj4210231add. htm). Chimeric enzymes derived from LePGT1 and UBIA were created by combinations of restriction digestion and PCR. The parental LePGT1 and UBIA genes subcloned in pDR196 were digested with EcoRI/SacI, EcoRI/XbaI, XbaI/SacI or PstI/MfeI. DNA fragments of the correct sizes were purified from an agarose gel and ligated with the appropriate DNA fragments that were prepared by PCR to create chimeric cDNAs. To amplify the DNA fragments of LePGT1 or UBIA for ligation, the following primers were used, LePGT1_PstI_Fw, LePGT1_MfeI_Rv, UBIA_ XbaI_Fw, UBIA_XbaI_Rv, UBIA_SacI_Fw, UBIA_SacI_Rv, PGT1_EcoRI_Fw, PGT1-Myc_XhoI_Rv, UBIA_EcoRI_Fw and UBIA-Myc_XhoI_Rv (Table 1). All chimeric cDNAs were sequenced completely to confirm the absence of sequence errors. copy of the COQ2 gene (strain coq2 [30]) to eliminate the background AS-PT activity. The microsomal preparation from S. cerevisiae, and measurements of PPT activity using non-radioactive or radioisotope-labelled substrates were made according to Yazaki et al. [30] with some modifications. The standard reaction mixture used to determine levels of enzymatic activity was 50 mM Tris/HCl (pH 7.6) containing 400 μM GPP, 400 μM PHB and 10 mM MgCl2 in a total volume of 200 μl. After incubation at 30 ◦C for 60 min, the reaction was terminated by the addition of 5 μl of formic acid. The reaction product was extracted with 150 μl of ethyl acetate containing testosterone propionate as an internal standard. This organic phase was evaporated dry, dissolved in 50 μl of methanol and analysed by HPLC as described previously [30]. Lineweaver–Burk plots were employed to calculate the K m from the HPLC data. With a radioactive substrate of [14 C]PHB, the reaction products were analysed by silica gel TLC (Kiesel gel, Merck, 20 cm × 20 cm) with a solvent system of toluene/ethyl acetate (4:1). The TLC plates were exposed to an imaging plate (Fuji Film) at room temperature (22 ◦C) for 5 days and then analysed with a BAS1800 image analyser (Fuji Film). Detection of UQ6 from S. cerevisiae UQ6 produced in yeast cells was extracted according to the method of Uchida et al. [34] with slight modifications. LC (liquid chromatography)-MS was carried out on a Shimadzu model 2010A system liquid chromatograph and mass spectrometer equipped with a atmospheric-pressure chemical ionization source, in which an LC-10AD solvent delivery system was used as the LC unit under the following conditions: column LiChrosphere 100RP-18 (Merck) 4 × 250 mm; solvent system, ethanol/2propanol (1:1); flow rate, 0.2 ml/min. UQ6 was identified by direct comparison with a standard specimen. Immunoblotting A monoclonal antibody against the Myc epitope was purchased from Cell Signaling Technology (Danvers, MA, U.S.A.). SDS/PAGE and immunoblotting of microsomal membrane proteins was performed according to a method reported previously [35] with slight modifications. Aliquots (20 μg) of yeast microsomal fractions were used for SDS/PAGE. Protein concentrations were determined according to the method of Bradford [36], using BSA as the standard. Measurements of PPT activity 3D structural modelling of the catalytic centre of LePGT1 All point mutants and chimeric enzymes were constitutively expressed in Saccharomyces cerevisiae containing a disrupted A sequence alignment between FPP (farnesyl diphosphate) synthase and LePGT1 was prepared based on similarity around c The Authors Journal compilation c 2009 Biochemical Society Mutational analyses of membrane-bound prenyltransferase 233 Figure 1 Representative features of LePGT1 protein and a multiple alignment of regions conserved among PPT family members involved in UQ and shikonin biosynthesis (A) Putative transmembrane α-helices (TM) and positions of Regions I–III. TM domains are indicated in black and numbered from the N-terminus. (B) Amino acid residues corresponding to Regions I II, and III are shown. The multiple alignment was made using ClustalW (http://clustalw.ddbj.nig.ac.jp/top-j.html). Asterisks indicate the residues mutated in this study. Double asterisks are aspartate residues exclusively conserved in the Mg2+ -dependent prenyltransferase family. At, Arabidopsis thaliana ; Ce, Caenorhabditis elegans ; Hs, Homo sapiens ; Mm, Mus musculus ; Os, Oryza sativa ; Sp, Streptococcus pneumoniae . (C) Immunoblotting of microsomal proteins that were used in this study. To detect the recombinant protein, all point mutants as well as wild-type LePGT1 were expressed as fusion proteins with a c-Myc epitope tag at the C-terminus, and a Myc-tag monoclonal antibody (Cell Signaling Technology) was used for immunoblotting. the FARM (first aspartate-rich motif) and the SARM (second aspartate-rich motif). The crystal structure of FPP synthase (PDB ID: 2F8Z) was used for molecular modelling. A tentative model of LePGT1 was built using the DS MODELLER module on a Discovery studio 1.7 (Accelrys). The docking simulation of magnesium ions in the catalytic centre of the LePGT1 model was performed with CDOCKER. The three magnesium ions with the top three scores are represented in Figure 8. The docked structure showed two magnesium ions that bound to the Oδ atoms of Asp84 , Asp87 and Asp91 , and one magnesium ion bound to the Oδ atoms of Asp208 and Asp212 . The model was optimized by 500 steps of potential energy minimization with CHARMm force field. PHB and GPP were docked into the model of LePGT1 including the magnesium ions. These models of ternary complex were optimized by 500 steps of potential energy minimization with CHARMm force field. Models were drawn using PyMOL. RESULTS Site-directed mutagenesis of LePGT1 (alanine-scan) Comparative topological analyses with five different prediction programs suggested that LePGT1 possesses nine transmembrane α-helices (see Supplementary Figure S2 at http://www.BiochemJ. org/bj/421/bj4210231add.htm). LePGT1 possesses three highly conserved amino acid sequences designated Regions I, II and III in the hydrophilic loops of the representative topology model predicted by the TMHMM program (http://www.cbs.dtu.dk/ services/TMHMM/), which are shown in Figure 1 [26,30,37]. c The Authors Journal compilation c 2009 Biochemical Society 234 K. Ohara and others Figure 2 PHB geranyltransferase activity of LePGT1 mutants in which conserved amino acids were substituted with alanine (alanine scan) The PHB geranyltransferase activities of fourteen mutant enzymes were measured using the microsomal fraction from S. cerevisiae transformants. The y -axis indicates specific activity (nmol/h per mg of protein), with relative activity (%) compared to the activity of wild-type LePGT1 (100 %) also shown. ud, undetectable. All these conserved hydrophilic loops face one side of the biological membrane, suggesting that the conserved domains interact to mediate enzyme function, such as substrate recognition. In order to investigate the role of each conserved amino acid, sitedirected mutations were introduced into LePGT1. Fifteen highly conserved and three non-conserved amino acid residues were individually changed to the aliphatic amino acid alanine: Region I, R76A, N83A, D84A, F86A, D87A, D91A and R96A; Region II, K152A and R153A; Region III, D201A, Y204A, H206A, Q207A, D208A, D211A, D212A, S219A and K229A (indicated with asterisks in Figure 1B). These mutant enzymes as well as the wild-type LePGT1 were heterologously expressed in S. cerevisiae strain coq2, in which the endogenous PPT gene was disrupted. The microsomal fraction was prepared first to check the expression level of the recombinant proteins by immunoblotting with an anti-Myc antibody to detect the c-Myc epitope tag attached to the C-terminus of LePGT1. The results showed that 13 mutant proteins had almost the same level of accumulation as the wild-type enzyme, whereas D91A accumulated less than the wild-type enzyme. Recombinant proteins were not detectable for four mutants (F86A, D201A, H206A and Y204A) (Figure 1C), which were not subjected to subsequent experiments. The enzymatic activity of 14 mutants was measured using GPP and PHB, the native substrates of LePGT1. The results are summarized in Figure 2. All mutant enzymes showed less activity than the wild-type protein, although the R153A mutant maintained a high level of activity (81.6 %). The molecular modelling of a bacterial orthologue, UBIA, revealed that this arginine residue (Arg137 in UBIA) is important for the recognition of phosphate groups via Mg2+ ions [37], but our biochemical data did not support this hypothesis. The mutant enzymes, in which one of four highly conserved aspartates was changed to alanine, did not show detectable activity (D87A, D91A and D208A), or exhibited extremely weak activity (D212A), i.e. 1000-fold lower than the control. In addition, seven mutant enzymes, in which a conserved amino acid was changed to alanine, showed only 0.15 % to 2.31 % of the level of activity of the wild-type, whereas S219A and K229A showed appreciable activity, i.e. 14.7 % and 12.9 %, respectively. Next, we determined apparent K m values to investigate the affinity of mutant enzymes for GPP and PHB (Figure 3). The K m of c The Authors Journal compilation c 2009 Biochemical Society Figure 3 Comparison of K m values of LePGT1 point mutants by alanine scanning (A) K m values for PHB of LePGT1 alanine mutants were determined from Lineweaver–Burk plots. (B) Those for GPP in LePGT1 alanine mutants are shown. Relative K m values compared with wild-type LePGT1 are shown as percentages. n.d., not determined. wild-type LePGT1 with the c-Myc tag was 66.4 μM for PHB and 29.5 μM for GPP. Among the mutant enzymes, N83A showed a strong increase in the K m value for PHB (1426 μM), i.e. 21.5-fold that of the wild-type, although its K m value for GPP (38.3 μM) was almost the same (1.3-fold) as that of the wild-type. Another mutant, R76A, also showed a strong decrease in affinity for PHB (K m 268 μM, 4-fold higher than the control), whereas this mutant revealed a 4-fold increase in affinity for GPP. In contrast, the K m values for GPP were increased in other mutants, such as S219A (147 μM, 5-fold) and K229A (88.6 μM, 3-fold); however, the influence on the K m value for GPP appeared to be weaker than that for PHB as compared with N83A (21.5-fold). The strong influence of N83A of the affinity for PHB may suggest the crucial involvement of Asn83 in the recognition of the PHB molecule. In fact, this asparagine residue is conserved in prenyltransferases accepting PHB as a substrate, but not in the homogentisate prenyltransferase family. The present study could not determine the K m value for PHB in R96A or the K m values for PHB and GPP in D212A because of the weak activities of these mutant enzymes. Substitution of four exclusively conserved aspartates with similar amino acids To investigate the role of the conserved aspartate residues (Asp87 , Asp91 , Asp208 and Asp212 ) in more detail, these residues were replaced with a similar amino acid: glutamate (namely D87E, D91E, D208E and D212E) or asparagine (namely, D87N, D91N, D208N and D212N). These eight mutant enzymes were expressed Mutational analyses of membrane-bound prenyltransferase 235 Figure 4 Substitution of four conserved aspartate residues with glutamate or asparagine Activities of mutated enzymes in which one of four exclusively conserved aspartate residues (Asp87 , Asp91 , Asp208 and Asp212 ) was replaced with glutamate (E) or asparagine (N) as similar amino acids. For comparison, the activity of the alanine (A) mutants is also shown. In addition to specific activity (nmol/h per mg of protein), relative activity compared with wild-type LePGT1 is also shown as a percentage. K m values of wild-type and mutated LePGT1 for PHB and GPP are indicated above each column. ud, undetectable; n.e., not expressed. in S. cerevisiae coq2, and their expression at the protein level was assessed by immunoblotting using the microsomal fraction (Figure 1C). The mutated proteins D208E, D208N and D212E were detected at the same level as the wild-type, whereas D87E had a lower level than the wild-type. However, no recombinant protein was detected for D87N, D91E, D91N or D212N. The activity of mutant proteins was confirmed as shown in Figure 4. No detectable geranyltransferase activity was found in D87E or D208N, although these recombinant proteins were clearly detected by immunoblotting. In contrast, D208E and D212E showed enzymatic activity, although only 0.41 % and 3.05 % respectively of that of the wild-type. These findings indicate the importance of the conserved aspartate residues to vary, i.e. the substitution with glutamate at Asp208 and Asp212 sustained the catalytic function, whereas the substitution at Asp87 abolished the activity. The K m values of D208E and D212E showed that the affinity for both substrates was the same as (D212E for GPP) or higher than that of the wild-type (D212E for PHB; D208E for both) (Figure 4). Substitution of other highly conserved residues with similar amino acids In addition to the four exclusively conserved aspartate residues, the alanine scan experiments indicated that other highly conserved amino acids could also be important for the catalytic mechanisms (Figure 2). To investigate the enzymatic function of other conserved amino acids, in the present study we tried to prepare six mutant enzymes, namely N83D, D84E, D84Q, R96K, K152R and K229R. The reason why these amino acids were selected was as follows: Asn83 had a strong influence on the K m value for PHB, Asp84 was used for comparison with the four conserved aspartate residues, Arg96 and Lys152 strongly affected the enzymatic activity as shown in Figure 2, and Lys229 is arginine in the mouse orthologue and the necessity of the lysine residue was to be confirmed. After the monitoring of protein levels using the c-Myc tag, N83D, D84E, D84Q and R96K were found to be successfully expressed in the host strain S. cerevisiae coq2, whereas the K152R and K229R mutants failed to accumulate in the recombinant yeast (Figure 1C). Subsequently, enzymatic activity was measured and the apparent K m values for GPP and PHB were determined with Figure 5 Substitution of conserved but non-essential residues with a similar amino acid Activities of mutated enzymes in which one of the conserved residues: (A), Asn83 ; (B), Asp84 ; (C), Arg96 was substituted with a similar amino acid. For comparison, the corresponding alanine mutants are also shown. The y -axis is specific activity (nmol/h per mg of protein). K m values of wild-type and mutated LePGT1 for PHB and GPP are indicated. these four mutant enzymes, which were compared with the alanine-substituted mutants (Figure 5). N83D showed a 7.4fold higher level of activity than N83A, whereas its affinity for PHB (K m = 70.0 μM) was the same as that of the wild-type, although N83A lost affinity for PHB, i.e. 21-fold lower than the D-substituted mutant (Figure 5A). The substitution of Asp84 with glutamate was much more preferable, namely 12-fold greater geranyltransferase activity than D84A; in contrast, D84Q showed a strong decrease in activity (Figure 5B). This finding suggests that an acidic amino acid at position 84 is crucial to maintain a high level of enzymatic activity. However, the K m values of D84E for the two substrates (PHB, 230 μM; GPP, 207 μM) were approx. 3.5- and 6.5-fold higher than those of the wild-type, and also apparently higher than those of D84A. The size of the amino acid would also be important for recognition of the substrate. The substitution of arginine at position 96 significantly reduced the enzymatic function. Even the change to a similar basic amino acid, lysine, brought about a substantial loss of activity, demonstrating that this arginine residue is important not only in providing a positive charge at this position but also in the size of the side chain. It would be interesting to compare the K m values for the substitutions at Arg96 , particularly with regard to the bulk of the side chain, but unfortunately reliable quantitative data were not obtained due to the detection limit of the reaction product. Analyses of substrate specificity with chimeric enzymes derived from LePGT1 and UBIA The main difference between LePGT1 and other PPTs involved in UQ biosynthesis is the availability of prenyl diphosphates c The Authors Journal compilation c 2009 Biochemical Society 236 Figure 6 K. Ohara and others PHB geranyltransferase activity and K m values of chimeric enzymes derived from LePGT1 and UBIA (A) Amino acid sequences of LePGT1 and UBIA. Arrowheads indicate the junctions for chimeras of LePGT1 and UBIA. Conserved Regions I, II, and III are underlined. (B) Schematic drawing of chimeric enzymes and PHB geranyltransferase (GT) activities. ud, undetectable. (C) K m values of chimeric enzymes for PHB and GPP. with various chain lengths, namely only LePGT1 is strictly GPPspecific, and all other characterized members of this family show broad specificity for prenyl substrates. However, the substrate preference was not influenced by a single amino acid substitution created in the above experiments (results not shown). Thus, to clarify the regions responsible for the difference in substrate specificity in the primary structure, several chimeric enzymes were prepared using LePGT1 and UBIA, the PPT of UQ biosynthesis in Escherichia coli [23] (Figures 6A and 6B). The native prenyl substrate of UBIA is octaprenyl diphosphate (C40), but UBIA is also able to utilize GPP as a prenyl substrate due to its broad specificity [32]. Utilizing measurements of PHB c The Authors Journal compilation c 2009 Biochemical Society geranyltransferase activity as a standard, the chimeric enzymes were expressed and their activity was detected. However, the activity could only be evaluated for Chimera 1 and Chimera 4 (Figure 6B), because the functional expression of these chimeras in yeast was very limited, i.e. some were not expressed at the protein level, and some showed no enzymatic activity. With chimeric enzymes 1 and 4, the K m values for both substrates were compared to those of the wild-type enzymes, UBIA and LePGT1. UBIA and Chimera 4 showed a considerably higher K m for GPP (3080 μM and 990 μM) than LePGT1 and Chimera 1 (29.5 μM and 33.5 μM), whereas these four enzymes had similar K m values for PHB (Figure 6C). Mutational analyses of membrane-bound prenyltransferase Figure 7 237 Specificity of chimeric enzymes for prenyl substrates and in vivo complementation testing of coq2 (A) PHB farnesyltransferase activities of LePGT1, UBIA and chimeric enzymes. PHB farnesyltransferase activity relative to PHB geranyltransferase activity is shown as a percentage. (B) Each chimeric enzyme, as well as wild-type LePGT1 and UBIA, was expressed in coq2, which was grown on an SD (Sabaraud dextrose) plate with either glucose (Glu) or glycerol (Gly) as the sole carbon source. pDR196, empty vector; W303-1A, wild-type S. cerevisiae ; coq2, COQ2 gene disruptant of S. cerevisiae . (C) UQ6 detection of UQ6 in coq2-expressing chimeric enzyme. The molecular mass of UQ6 (m /z 592) was scanned by LC-MS. Arrows indicate the retention time of UQ6. To assess the prenyl substrate specificity, PHB farnesyltransferase activity was employed. The results showed FPP was not a preferred substrate for Chimera 1 or for LePGT1 (2.20 % and 3.13 % of the activity of GPP); however, Chimera 4 as well as UBIA could clearly utilize FPP as a prenyl substrate (105 % and 49.6 % of the activity of GPP) (Figure 7A). A yeast complement test was also carried out. S. cerevisiae coq2 is defective in producing UQ due to a lack of PPT, the key prenyltransferase for the biosynthesis of the intermediate m-prenyl-p-hydroxyPHB, and this strain shows a clear growth defect on a minimal medium containing glycerol as the sole carbon source, because UQ is required to utilize this non-fermentable carbon source for growth [24]. The expression of UBIA and Chimera 4 successfully complemented the growth defect of the COQ2 disruptant on glycerol-containing medium in nearly the same manner as in the wild-type yeast, whereas no growth was observed for the coq2 strain harbouring either pDR-LePGT1 Chimera 1, or the empty vector pDR196 (Figure 7B). As this bioassay is much more sensitive than the HPLC assay, it was tested whether or not other chimeras (Chimeras 2, 3 and 5–7) with no detectable activity in vitro could compliment the growth defect on glycerol medium. However, they also failed to rescue the growth of the coq2 strain on the glycerol plate (Figure 7B). In addition to the growth complementation, the in vivo production of native UQ in yeast was confirmed, i.e. UQ6, in the extract of yeast transformants expressing Chimera 4 as well as UBIA used as a positive control (Figure 7C). The analysis showed that UBIA and Chimera 4 could indeed function as PHB hexaprenyltransferase in vivo leading to the accumulation of UQ6 together with the growth recovery of coq2 yeast on the glycerol plate. DISCUSSION Because LePGT1 is bound to the membrane via nine putative transmembrane α-helices, a structural analysis by X-ray c The Authors Journal compilation c 2009 Biochemical Society 238 Figure 8 K. Ohara and others 3D structure of LePGT1 determined by molecular modelling (A) Substrate-binding site of LePGT1. The polypeptide chain of LePGT1 is shown in yellow, the geranyl moiety of GPP is in magenta, diphosphate is shown in orange, PHB is drawn in green, and Mg2+ ions are shown as grey spheres. Amino acids affecting the catalytic function are also indicated. The distance between phosphate and the OH residue of PHB, and the carbon atom at the m -position of PHB and the C-1 atom of GPP are 2.76 Å and 7.11 Å respectively. The Figure was drawn using PyMOL. (B) Surface model of LePGT1. The geranyl chain of GPP is shown in magenta, and the N-terminal region (amino acids 41–170), which is involved in the substrate specificity for GPP, is drawn in cyan. crystallography was inapplicable. Thus a molecular modelling approach was attempted to visualize its 3D structure or at least that of the catalytic domain including bound substrates. Among various Mg-dependent prenyltransferases, FPP synthase provided a suitable model for generating a reasonable 3D structure of LePGT1 because this polypeptide consisted of more than 10 α-helices despite being a soluble protein. More importantly, FPP synthase shows extensive functional analogy with LePGT1, e.g. the preference for GPP as a substrate, which is recognized by an aspartate-rich motif, and the requirement for divalent cations with a preference for Mg2+ . By computational chemistry, the 3D structure of LePGT1 was generated, to which two substrates, GPP and PHB, were added by docking simulation (Figure 8A). In a similar manner to FPP synthase, the diphosphate moiety of GPP is recognized by two aspartate residues at positions 84 c The Authors Journal compilation c 2009 Biochemical Society and 87 via two Mg2+ atoms through chelation bonds. The PHB molecule is situated between the α-helix with Region I and that containing Region III, where the carboxyl residue of PHB seems to interact with another Mg2+ that is recognized by Asp208 and Asp212 in Region III, whereas the orientation of PHB is so that the m-position to be prenylated is facing the phosphate residue of GPP. It is noteworthy that the distance between the OH residue and the phosphate group is only 2.76 Å (1 Å = 0.1 nm). The distance between the C1 atom of GPP and the carbon atom at the m-position of PHB where the prenylation reaction takes place is calculated to be 7.11 Å, which is larger than that of a soluble type Streptomyces AS-PT, Orf2 (4 Å), whose crystal structure has been characterized [22], but the distance of LePGT1 is similar to that between the cysteine residue to be prenylated and the C1 atom of FPP in human protein farnesyltransferase (7.3 Å) [38]. This ‘bridge-like’ configuration of PHB between two αhelices containing either Region I and III, and the vicinity of two substrates with reasonable orientations strongly suggests that this model represents the structure of the substrate-binding site of LePGT1. Actually, this model supports the biochemical data using mutant enzymes demonstrated above with high conformity, i.e. both Region I and III influence the recognition of each substrate, and most amino acid residues (Asp84 , Asp87 and Asp91 ) involved in the catalytic functions in Region I as well as those in Region III (Asp208 , Asp211 and Asp212 ) face the inside of the binding pocket. Using this model, it was calculated that the surface structure of LePGT1-bound GPP, where the N-terminal region affecting the substrate specificity (positions 41–170) is shown in cyan (Figure 8B). This model demonstrates that the cavity where GPP is bound provides a narrow space and the N-terminal region takes the position close to the entrance of the cavity, suggesting that the tight binding pocket formed by the N-terminal region is the reason for the preference for GPP as the prenyl substrate. In prenyltransferase proteins catalysing trans-prenyl-chainelongation such as FPP synthase, the aromatic amino acid residue located at the fifth position upstream of the FARM (DDxxD) is essential for controlling the product chain length in trans prenyl chain elongation [39]. It was presumed that prenyl substrate specificity for the chain length could also be regulated in membrane-bound PPTs in a similar manner, but the substitution of the corresponding amino acids (Trp70 , Phe130 , Trp74 and His140 ) of LePGT1 did not alter the substrate specificity (results not shown). This suggests that the prenyltransferase activity for PHB has a different regulatory mechanism from soluble-type prenyl diphosphate synthases, and the molecular model of LePGT1 (Figure 8) indicates the specificity for the prenyl substrate to be determined by the spatial size of the hydrophobic pocket in the catalytic centre. This hypothesis is supported by the present study of chimeric enzymes showing that specificity for chain length is determined by the 130-amino acid segment of the N-terminus of LePGT1, which is involved in the GPP-binding cavity. The critical importance of the N-terminal region of PPT (e.g. Ser41 to Ala170 of LePGT1) for the availability of the prenyl substrates with different chain lengths was also demonstrated both in the analysis of K m values and in the complementation study with chimeric enzymes derived from LePGT1 and UBIA. The physiological prenyl substrate of UBIA is octaprenyl diphosphate (C40) in vivo, and UBIA showed a higher K m value for GPP than did LePGT1. The K m value of Chimera 1 for GPP was approximately equivalent to that of LePGT1, whereas Chimera 4 showed a much higher K m for GPP-like UBIA. It is worth noting that in the yeast complementation study both Chimera 4 and UBIA could utilize hexaprenyl diphosphate (C30) in vivo as a prenyl substrate, whereas LePGT1 and Chimera 1 could not. Mutational analyses of membrane-bound prenyltransferase Furthermore, the in vitro PHB farnesyltransferase assay showed that Chimera 1 had LePGT1-type strict specificity for GPP but Chimera 4 had relaxed specificity similar to UBIA. These results indicate that the prenyl substrate specificity of membrane-type PPTs is regulated by a different mechanism from that of soluble prenyltransferases. Plant prenyltransferases such as isoprene, monoterpene sesquiterpene and diterpene synthases, which are soluble proteins, require magnesium ions for their enzymatic activities, and their common aspartate-rich motifs are highly conserved [40–42]. The mechanism behind the requirement of Mg2+ for their catalytic function was clearly demonstrated by X-ray crystallographic studies. For instance, in FPP synthase, the catalytic centre contains the exclusively conserved aspartate residues in FARM and SARM (DDxxD), and the prenyl substrates were recognized directly by these aspartate residues through the magnesium ions in the catalytic centre of the enzyme [43]. The membrane-bound PPT family also needs magnesium ions for enzymatic activity. These enzymes have two highly conserved regions, i.e. Region I corresponding to FARM, where three aspartates are completely conserved, and Region III, which also contains three highly conserved aspartates but the amino acid sequence is specific only to this family accepting PHB as an aromatic substrate. Site-directed mutagenesis of LePGT1 showed that only one conserved aspartate residue either in Region I or in Region III acts concertedly in the catalytic centre of LePGT1 (Figure 2), perhaps in recognizing the diphosphate moiety of the prenyl substrate via magnesium ions. This hypothesis seems also to be valid for bacterial PPT, e.g. E. coli UBIA for UQ biosynthesis [37]. Bacterial soluble-type AS-PTs that do not require magnesium ions for their activity do not have these aspartate-rich motifs [21], but these orthologues have not been found in plants thus far. The substitution of aspartates in Region I tended to have severe negative effects on the expression of recombinant enzymes (failure to express: D87N, D91E or D91N) or enzymatic activity (no activity: D87E). However, substitution of the conserved aspartates in Region III with glutamate (D208E and D212E) resulted in detectable enzymatic activities with lower K m values for GPP than the native LePGT1 (Figure 4), suggesting that these glutamate residues partially complement the catalytic function in maintaining the substrate-binding affinity. From these results, the roles of conserved aspartates in the enzymatic function may differ in a similar manner as in soluble trans-prenyl transferases [44]. In the present 3D molecular model of LePGT1, Asp87 and Asp91 are involved in the binding of prenyl diphosphate via magnesium ions (Figure 8A). In addition, it is suggested in Figure 2 that Gln207 and Asp211 also have critical roles in the binding of substrates or in catalysing the reaction. Indeed, these amino acid residues, as well as Asp208 and Asp212 , are located on the inner face of the substratebinding pocket and seem to be involved in the recognition of both substrates (Figure 8A). Similarly, a contribution of Lys229 , located in Region III, to the enzymatic function, is predicted based on the present mutational study (Figure 3) in which the alanine mutant still displayed 12.9 % of the wild-type activity but the K m values for both substrates strongly increased (approx. 3-fold). This indicates that Lys229 is important for substrate recognition, and the residue also faces the inside of the substrate-binding pocket. Despite two limitations that the enzyme data is derived from the in vitro assay using the microsomal system of transgenic yeast and the structure is determined from molecular modelling based on a soluble protein, reasonable explanations for the biochemical data of LePGT1 can be drawn from the 3D structure shown in Figure 8. Previously, Regions I and III were predicted to be responsible for the binding of GPP and PHB respectively, due to sequence 239 similarity among various PPTs. However, the present mutational analysis has clearly demonstrated that Regions I and III are involved in the recognition of both substrates in a co-ordinated manner, e.g. N83A and R76A in Region I caused remarkable increases in the K m for PHB, whereas mutation of Ser219 and Lys229 in Region III strongly affected the K m for GPP (Figure 3). Although the structure is determined from molecular modelling with FPP synthase as the template, the 3D model provides a reasonable explanation of these biochemical findings, e.g. both GPP and PHB are held by Regions I and III, where the distance between the OH residue and phosphate group is 2.76 Å. It also newly suggests that a magnesium ion is necessary for the interaction between aspartate residues in Region III and the carboxyl group of PHB (Figure 8A). Recently, a 3D molecular model has been reported for the UBIA protein, in which 5-epi-aristolochene synthase was used as a template to generate the model [45]. A careful comparison was made with the UBIA model and the present LePGT1 model, especially at the active site of both enzymes, and some similar features were found, e.g. Asp87 (Asp71 of UBIA) and Asp91 (Asp75 of UBIA) are involved in the binding of the diphosphate residue via a magnesium ion. However, there are several discrepancies between the UBIA model and the LePGT1 model and biochemical data presented here. The most prominent difference involves Arg153 (Arg137 of UBIA), which was reportedly important for binding prenyl diphosphate [45]. In the present study, R153A retained strong enzymatic activity (81.6 % of the wild-type level), and thus Arg153 does not have a critical role in catalytic function, whereas the R153A mutant showed an influence on the K m value for PHB. The weak activity of the R137A mutant of UBIA observed in the previous study may not reflect the importance of this arginine residue for enzymatic function, but may be due to a failure of the expression of the recombinant protein in E. coli, which was not monitored in the study [45]. Other minor differences were also seen: Asp208 (Asp191 of UBIA) reportedly bound the OH group of PHB, but is involved in recognition of the carboxyl group of PHB in the present model; and Asp212 (Asp195 of UBIA) did not mediate binding to the phosphate group, but recognition of PHB was shown in the LePGT1 model. However, it should be emphasized that the GPP-binding pattern is very similar between the two models despite the usage of different templates. The configuration of PHB in the binding pocket of LePGT1 is slightly different from that in the UBIA model, in which Arg72 (UBIA) is responsible for the interaction with the C-terminus of PHB [45], but this amino acid is not conserved in other PPT members, e.g. this position is glutamine in AtPPT1 and human COQ2, which is inconsistent with their hypothesis concerning the role of the arginine residue. In the present study, the amino acid residues of LePGT1 critical for enzymatic activity and the region responsible for the binding of substrates were elucidated biochemically. These findings should improve the basic understanding of the enzymatic mechanism of membrane-type AS-PTs. Other membrane-bound AS-PT families in plants include homogentisate prenyltransferases for vitamin E and plastoquinone biosynthesis and also flavonoid prenyltransferases, which are responsible for the production of many biologically active secondary metabolites in plants. These membrane proteins have a similar membrane topology to LePGT1 [46]. The molecular modelling method applied to LePGT1 will also be applicable for these members to understand their catalytic functions and will provide new molecular engineering designs of membrane-bound AS-PTs for the efficient production of prenylated aromatic compounds using heterologous organisms such as bacteria or higher plants. c The Authors Journal compilation c 2009 Biochemical Society 240 K. Ohara and others AUTHOR CONTRIBUTION Kazufumi Yazaki and Nobuhiro Fukushima designed research; Kazuaki Ohara and Ayumu Muroya performed research; Kazuaki Ohara, Ayumu Muroya, Nobuhiro Fukushima and Kazufumi Yazaki analysed data and Kazuaki Ohara, Ayumu Muroya and Kazufumi Yazaki wrote the paper. ACKNOWLEDGEMENTS The yeast shuttle vector pDR196 was a generous gift from Dr Wolf Frommer (Department of Plant Biology, Carnegie Institution for Science, Stanford, CA, U.S.A.). We also thank Dr Lutz Heide (Pharmaceutical Institute, University of Tübingen, Germany) for cDNA of the UBIA gene. We are grateful to Dr Toshiaki Umezawa of Kyoto University for LC-MS analyses. The analysis of DNA sequences was conducted by the Life Research Support Center of Akita Prefectural University. We thank Accelrys (San Diego, CA, U.S.A.) for approval to use the Accelrys Discovery Studio in the modelling. FUNDING This work was supported in part by a Grant-in-aid for Scientific Research [grant numbers 17310126 and 21310141 (to K. Y.)], a Grant from the Research for the Future Program: ‘Molecular mechanisms on regulation of morphogenesis and metabolism leading to increased plant productivity’ [grant number 00L01605 (to K. Y.)] of the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by a Research Fellowship from the Japan Society for the Promotion of Science for Young Scientists [grant number 17·2011 (to K. O.)]. REFERENCES 1 Kawamukai, M. (2002) Biosynthesis, bioproduction and novel roles of ubiquinone. J. Biosci. Bioeng. 94, 511–517 2 Swiezewska, E., Dallner, G., Andersson, B. and Ernster, L. (1993) Biosynthesis of ubiquinone and plastoquinone in the endoplasmic reticulum-Golgi membranes of spinach leaves. J. Biol. Chem. 268, 1494–1499 3 Sadre, R., Gruber, J. and Frentzen, M. (2006) Characterization of homogentisate prenyltransferases involved in plastoquinone-9 and tocochromanol biosynthesis. FEBS Lett. 580, 5357–5362 4 Collakova, E. and DellaPenna, D. (2001) Isolation and functional analysis of homogentisate phytyltransferase from Synechocystis sp. PCC 6803 and Arabidopsis . Plant Physiol. 127, 1113–1124 5 Schledz, M., Seidler, A., Beyer, P. and Neuhaus, G. (2001) A novel phytyltransferase from Synechocystis sp. PCC 6803 involved in tocopherol biosynthesis. FEBS Lett. 499, 15–20 6 Schroder, G., Zahringer, U., Heller, W., Ebel, J. and Grisebach, H. (1979) Biosynthesis of antifungal isoflavonoids in Lupinus albus . Enzymatic prenylation of genistein and 2 -hydroxygenistein. Arch. Biochem. Biophys. 194, 635–636 7 Yamamoto, H., Senda, M. and Inoue, K. (2000) Flavanone 8-dimethylallyltransferase in Sophora flavescens cell suspension cultures. Phytochemistry 54, 649–655 8 Zhao, P., Inoue, K., Kouno, I. and Yamamoto, H. (2003) Characterization of leachianone G 2”-dimethylallyltransferase, a novel prenyl side-chain elongation enzyme for the formation of the lavandulyl group of sophoraflavanone G in Sophora flavescens Ait. cell suspension cultures. Plant Physiol. 133, 1306–1313 9 Biggs, D. R., Welle, R., Visser, F. R. and Grisebach, H. (1987) Dimethylallylpyrophosphate:3,9-dihydroxypterocarpan 10-dimethylallyl transferase from Phaseolus vulgaris . Identification of the reaction product and properties of the enzyme. FEBS Lett. 220, 223–226 10 Welle, R. and Grisebach, H. (1991) Properties and solubilization of the prenyltransferase of isoflavonoid phytoalexin biosynthesis in soybean. Phytochemistry 30, 479–484 11 Laflamme, P., Khouri, H., Gulick, P. and Ibrahim, R. (1993) Enzymatic prenylation of isoflavones in white lupin. Phytochemistry 34, 147–151 12 Dhillon, D. S. and Brown, S. A. (1976) Localization, purification, and characterization of dimethylallylpyrophosphate:umbelliferone dimethylallyltransferase from Ruta graveolens . Arch. Biochem. Biophys. 177, 74–83 13 Hamerski, D., Schmitt, D. and Matern, U. (1990) Induction of two prenyltransferases for the accumulation of coumarin phytoalexins in elicitor-treated Ammi majus cell suspension cultures. Phytochemistry 29, 1131–1135 14 Zuurbier, K. W. M., Fung, S. Y., Scheffer, J. J. C. and Verpoorte, R. (1998) In-vitro prenylation of aromatic intermediates in the biosynthesis of bitter acids in Humulus lupulus . Phytochemistry 49, 2315–2322 15 Fellermeier, M. and Zenk, M. H. (1998) Prenylation of olivetolate by a hemp transferase yields cannabigerolic acid, the precursor of tetrahydrocannabinol. FEBS Lett. 427, 283–285 c The Authors Journal compilation c 2009 Biochemical Society 16 Tahara, S. and Ibrahim, R. K. (1995) Prenylated isoflavonoids-an update. Phytochemistry 38, 1073–1094 17 Morandi, D. (1996) Occurrence of phytoalexins and phenolic compounds in endomycorrhizal interactions, and their potential role in biological control. Plant Soil 185, 241–251 18 Wang, B. H., Ternai, B. and Polya, G. (1997) Specific inhibition of cyclic AMP-dependent protein kinase by warangalone and robustic acid. Phytochemistry 44, 787–796 19 Henderson, M. C., Miranda, C. L., Stevens, J. F., Deinzer, M. L. and Buhler, D. R. (2000) In vitro inhibition of human P450 enzymes by prenylated flavonoids from hops, Humulus lupulus . Xenobiotica 30, 235–251 20 Di Pietro, A., Conseil, G., Perez-Victoria, J. M., Dayan, G., Baubichon-Cortay, H., Trompier, D., Steinfels, E., Jault, J. M., de Wet, H., Maitrejean, M., et al. (2002) Modulation by flavonoids of cell multidrug resistance mediated by P-glycoprotein and related ABC transporters. Cell. Mol. Life Sci. 59, 307–322 21 Pojer, F., Wemakor, E., Kammerer, B., Chen, H., Walsh, C. T., Li, S. M. and Heide, L. (2003) CloQ, a prenyltransferase involved in clorobiocin biosynthesis. Proc. Natl. Acad. Sci. U.S.A. 100, 2316–2321 22 Kuzuyama, T., Noel, J. P. and Richard, S. B. (2005) Structural basis for the promiscuous biosynthetic prenylation of aromatic natural products. Nature 435, 983–987 23 Siebert, M., Bechthold, A., Melzer, M., May, U., Berger, U., Schroder, G., Schroder, J., Severin, K. and Heide, L. (1992) Ubiquinone biosynthesis. Cloning of the genes coding for chorismate pyruvate-lyase and 4-hydroxybenzoate octaprenyl transferase from Escherichia coli . FEBS Lett. 307, 347–350 24 Ashby, M. N., Kutsunai, S. Y., Ackerman, S., Tzagoloff, A. and Edwards, P. A. (1992) COQ2 is a candidate for the structural gene encoding para -hydroxybenzoate: polyprenyltransferase. J. Biol. Chem. 267, 4128–4136 25 Suzuki, K., Ueda, M., Yuasa, M., Nakagawa, T., Kawamukai, M. and Matsuda, H. (1994) Evidence that Escherichia coli ubiA product is a functional homolog of yeast COQ2, and the regulation of ubiA gene expression. Biosci. Biotechnol. Biochem. 58, 1814–1819 26 Okada, K., Ohara, K., Yazaki, K., Nozaki, K., Uchida, N., Kawamukai, M., Nojiri, H. and Yamane, H. (2004) The AtPPT1 gene encoding 4-hydroxybenzoate polyprenyl diphosphate transferase in ubiquinone biosynthesis is required for embryo development in Arabidopsis thaliana . Plant Mol. Biol. 55, 567–577 27 Ohara, K., Yamamoto, K., Hamamoto, M., Sasaki, K. and Yazaki, K. (2006) Functional characterization of OsPPT1, which encodes p -hydroxybenzoate polyprenyltransferase involved in ubiquinone biosynthesis in Oryza sativa . Plant Cell Physiol. 47, 581–590 28 Forsgren, M., Attersand, A., Lake, S., Grunler, J., Swiezewska, E., Dallner, G. and Climent, I. (2004) Isolation and functional expression of human COQ2 , a gene encoding a polyprenyl transferase involved in the synthesis of CoQ. Biochem. J. 382, 519–526 29 Ohara, K., Kokado, Y., Yamamoto, H., Sato, F. and Yazaki, K. (2004) Engineering of ubiquinone biosynthesis using the yeast coq2 gene confers oxidative stress tolerance in transgenic tobacco. Plant J. 40, 734–743 30 Yazaki, K., Kunihisa, M., Fujisaki, T. and Sato, F. (2002) Geranyl diphosphate:4hydroxybenzoate geranyltransferase from Lithospermum erythrorhizon . Cloning and characterization of a key enzyme in shikonin biosynthesis. J. Biol. Chem. 277, 6240–6246 31 Heide, L. and Berger, U. (1989) Partial purification and properties of geranyl pyrophosphate synthase from Lithospermum erythrorhizon cell cultures. Arch. Biochem. Biophys. 273, 331–338 32 Melzer, M. and Heide, L. (1994) Characterization of polyprenyldiphosphate: 4-hydroxybenzoate polyprenyltransferase from Escherichia coli . Biochim. Biophys. Acta 1212, 93–102 33 Rentsch, D., Laloi, M., Rouhara, I., Schmelzer, E., Delrot, S. and Frommer, W. B. (1995) NTR1 encodes a high affinity oligopeptide transporter in Arabidopsis . FEBS Lett. 370, 264–268 34 Uchida, N., Suzuki, K., Saiki, R., Kainou, T., Tanaka, K., Matsuda, H. and Kawamukai, M. (2000) Phenotypes of fission yeast defective in ubiquinone production due to disruption of the gene for p -hydroxybenzoate polyprenyl diphosphate transferase. J. Bacteriol. 182, 6933–6939 35 Shitan, N., Bazin, I., Dan, K., Obata, K., Kigawa, K., Ueda, K., Sato, F., Forestier, C. and Yazaki, K. (2003) Involvement of CjMDR1, a plant multidrug-resistance-type ATP-binding cassette protein, in alkaloid transport in Coptis japonica . Proc. Natl. Acad. Sci. U.S.A. 100, 751–756 36 Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254 37 Brauer, L., Brandt, W. and Wessjohann, L. A. (2004) Modeling the E. coli 4-hydroxybenzoic acid oligoprenyltransferase (ubiA transferase) and characterization of potential active sites. J. Mol. Model. 10, 317–327 38 Long, S. B., Casey, P. J. and Beese, L. S. (2002) Reaction path of protein farnesyltransferase at atomic resolution. Nature 419, 645–650 Mutational analyses of membrane-bound prenyltransferase 39 Ohnuma, S., Narita, K., Nakazawa, T., Ishida, C., Takeuchi, Y., Ohto, C. and Nishino, T. (1996) A role of the amino acid residue located on the fifth position before the first aspartate-rich motif of farnesyl diphosphate synthase on determination of the final product. J. Biol. Chem. 271, 30748–30754 40 Miller, B., Oschinski, C. and Zimmer, W. (2001) First isolation of an isoprene synthase gene from poplar and successful expression of the gene in Escherichia coli . Planta 213, 483–487 41 Sasaki, K., Ohara, K. and Yazaki, K. (2005) Gene expression and characterization of isoprene synthase from Populus alba . FEBS Lett. 579, 2514–2518 42 Bohlmann, J., Meyer-Gauen, G. and Croteau, R. (1998) Plant terpenoid synthases: molecular biology and phylogenetic analysis. Proc. Natl. Acad. Sci. U.S.A. 95, 4126–4133 241 43 Marrero, P. F., Poulter, C. D. and Edwards, P. A. (1992) Effects of site-directed mutagenesis of the highly conserved aspartate residues in domain II of farnesyl diphosphate synthase activity. J. Biol. Chem. 267, 21873–21878 44 Song, L. and Poulter, C. D. (1994) Yeast farnesyl-diphosphate synthase: site-directed mutagenesis of residues in highly conserved prenyltransferase domains I and II. Proc. Natl. Acad. Sci. U.S.A. 91, 3044–3048 45 Brauer, L., Brandt, W., Schulze, D., Zakharova, S. and Wessjohann, L. (2008) A structural model of the membrane-bound aromatic prenyltransferase UbiA from. E. coli. Chembiochem 9, 982–992 46 Sasaki, K., Mito, K., Ohara, K., Yamamoto, H. and Yazaki, K. (2008) Cloning and characterization of naringenin 8-prenyltransferase, a flavonoid-specific prenyltransferase of Sophora flavescens . Plant Physiol. 146, 1075–1084 Received 30 September 2008/21 April 2009; accepted 24 April 2009 Published as BJ Immediate Publication 24 April 2009, doi:10.1042/BJ20081968 c The Authors Journal compilation c 2009 Biochemical Society Biochem. J. (2009) 421, 231–241 (Printed in Great Britain) doi:10.1042/BJ20081968 SUPPLEMENTARY ONLINE DATA Functional characterization of LePGT1, a membrane-bound prenyltransferase involved in the geranylation of p -hydroxybenzoic acid Kazuaki OHARA*1 , Ayumu MUROYA†, Nobuhiro FUKUSHIMA† and Kazufumi YAZAKI*2 *Laboratory of Plant Gene Expression, Research Institute for Sustainable Humanosphere, Kyoto University, Uji 611-0011, Japan, and †Science & Technology Systems, Inc., 1-20-1 Shibuya, Tokyo 150-0002, Japan Figure S1 Oligonucleotides used in the present study Mutagenic oligonucleotides designed to produce the desired point mutations are listed; underlining indicates changed codons for mutagenesis. 1 Present address: Central Laboratories for Frontier Technology, Kirin Holdings Co., Ltd., 1-13-5 Fukuura Kanazawa-ku, Yokohama-shi, Kanagawa 236-0004, Japan. 2 To whom correspondence should be addressed (email [email protected]). c The Authors Journal compilation c 2009 Biochemical Society K. Ohara and others Figure S2 Prediction of transmembrane domains of LePGT1 Predicted transmembrane domains (TM) are underlined, and the three conserved regions, Region I, II and III, are indicated with red lines. Predictions were carried out using the following programs; HMMTOP (http://www.enzim.hu/hmmtop/), TMHMM (http://www.cbs.dtu.dk/services/TMHMM/), TMPRED (http://www.ch.embnet.org/software/TMPRED_form.html), CONPRE (Conpred2; http://bioinfo.si.hirosaki-u.ac.jp/∼ConPred2/) and MEMSAT (http://saier144-37.ucsd.edu/memsat.html). H, α-helix; i, inside; o, outside. Received 30 September 2008/21 April 2009; accepted 24 April 2009 Published as BJ Immediate Publication 24 April 2009, doi:10.1042/BJ20081968 c The Authors Journal compilation c 2009 Biochemical Society