TY - JOUR
AU - Saito, Kazuki
AB - Abstract A novel acyltransferase committed to the final step of quinolizidine alkaloid biosynthesis, tigloyl-CoA:(–)-13α-hydroxymultiflorine/(+)-13α-hydroxylupanine O-tigloyltransferase, has been purified from Lupinus albus. The internal amino acid sequences were determined with protease-digested fragments of 25 and 30 kDa bands, allowing design of primers for amplification of cDNA fragments by polymerase chain reaction. Using an amplified fragment as the probe, a full-length cDNA clone was isolated. Sequence analysis revealed that the cDNA encodes a protein of 453 amino acids with a molecular mass of 51.2 kDa. Phylogenetic analysis of the deduced amino acid sequences indicated that this alkaloid acyltransferase belongs to a unique subfamily of a plant acyl-CoA-dependent acyltransferase gene family. The cDNA was expressed in bacterial cells as a recombinant protein fused to glutathione S-transferase. The fusion protein was affinity purified and cleaved to yield the recombinant enzyme for the study of catalytic properties. The recombinant enzyme catalyzed the acyltransfer reaction from tigloyl-CoA to (–)-13α-hydroxymultiflorine and (+)-13α-hydroxylupanine. Benzoyl-CoA could also serve efficiently as an acyl donor for these hydroxylated alkaloids. RNA blot analysis suggested that the gene was expressed in roots and hypocotyls but not in cotyledons and leaves. These results indicated that this specialized acyltransferase, isolated for the first time as tigloyltransferase from nature, is committed to control the quinolizidine alkaloid patterns in a tissue-specific manner. (Received October 12, 2004; Accepted November 10, 2004) Introduction The quinolizidine alkaloids are plant secondary products which are distributed mainly in the family Leguminosae, especially in the subfamily Papilionaceae (Ohmiya et al. 1995). More than 200 structurally related compounds belonging to this alkaloid group are found naturally. These alkaloids are assumed to play indispensable roles for the survival of plants producing these metabolites as defense compounds against pathogenic organisms or predators and allelopathic metabolites for competing with other plant species (Roberts and Wink 1998). Some alkaloids are beneficial for mankind as potential sources of medicines because of their pharmacological activities in animals. In fact, some plants containing the quinolizidine alkaloids have been used as traditional herbal medicines, and the quinolizidine alkaloids have been proven to be the principal components responsible for the pharmacological activities of these herbal medicines (Tang and Eisenbrand 1992). The quinolizidine alkaloids are biosynthesized from l-lysine via its decarboxylated metabolite cadaverine (Fig. 1). Three units of cadaverine are subjected to oxidative cyclization to form tetracyclic alkaloids such as (–)-multiflorine and (+)-lupanine. These de novo synthesized alkaloids are modified further by hydroxylation and by subsequent esterification to yield the ester-type alkaloids, e.g. 13α-tigloyloxymultiflorine. The ester-type quinolizidine alkaloids are widely distributed in the genera Lupinus, Cytisus, Perasonia, Calpurnia, Genista and Rothia, as the esters of acetic acid, tiglic acid, p-coumaric acid, ferulic acid and benzoic acid (Ohmiya et al. 1995). The tigloyl esters of quinolizidine alkaloids are the major forms found in Lupinus plants (Saito and Murakoshi 1995). These ester-type alkaloids are assumed to be the end-products of biosynthesis and thus the forms for transport and storage. A superfamily of acyl-CoA-dependent acyltransferases exhibiting a conserved sequence motif has been documented in plants (St-Pierre et al. 1998). This versatile gene family is assumed to evolve the catalytic ability of transfer of a variety of acyl groups to a number of different substrates having diverse functions in plants (St-Pierre and De Luca 2000). However, no gene and its protein that catalyzes tigloyl transfer have been characterized in this superfamily up to now. Also, no report is available on the presence of tigloyl esters and tigloyltransferase from other organisms, animals and microorganisms. Therefore, it is intriguing to isolate genes encoding tigloyltransferase responsible for the biosynthesis of ester-type quinolizidine alkaloids. The acyltransferases responsible for the conversion of hydroxylated alkaloids to their esters have been detected in several Lupinus species (Wink and Hartmann 1982a, Strack et al. 1991, Saito et al. 1993). From L. albus (=L. termis), we previously have purified and characterized tigloyl-CoA:(–)-13α-hydroxymultiflorine/(+)-13α-hydroxylupanine O-tigloyltransferase (HMT/HLT) that catalyzes the formation of (–)-13α-tigloyloxymultiflorine or (+)-13α-tigloyloxylupanine from (–)-13α-hydroxymultiflorine or (+)-13α-hydroxylupanine in the presence of tigloyl-CoA (Suzuki et al. 1994). In the continuing study on the enzymology of this biosynthetic pathway, we report here the isolation of a HMT/HLT cDNA from L. albus by improved purification of the enzyme proteins and determination of amino acid sequences of digested peptide fragments. The cloned HMT/HLT belongs to a large plant-specific acyltransferase superfamily. The recombinant protein produced in Escherichia coli efficiently catalyzes the acyltransfer reaction from several acyl-CoAs to the hydroxylated alkaloids 13α-hydroxymultiflorine and 13α-hydroxylupanine. HMT/HLT is the key enzyme that determines the alkaloid pattern in a plant, and it is also the first enzyme whose cDNA is isolated in the biosynthetic pathway of quinolizidine alkaloids. Results Improved purification of HMT/HLT and determination of internal peptide sequences An improved method requiring fewer chromatographic steps than previously described (Suzuki et al. 1994) was used to purify HMT/HLT from roots and hypocotyls of L. albus. The purification procedure included a novel butyl FF hydrophobic column step, together with RESOURCE S cation exchange chromatography and it omitted a few of the chromatographic separation steps described in Suzuki et al. (1994) to yield a high recovery of active pure HMT/HLT (Table 1). The specific activity of the 689-fold purified HMT/HLT was 7,305 pkat mg–1 protein and sufficient amounts of protein were produced to determine several internal HMT/HLT amino acid sequences. Analysis of the purified protein by SDS–PAGE and silver staining revealed three polypeptide bands with Mrs of 25, 30 and 60 kDa (Fig. 2). The N-terminal amino acid sequencing of trypsin-digested peptide fragments from these three polypeptides produced the six sequences shown in Fig. 2. In the digested peptides of the 25 kDa band, a motif sequence DFGWG that is highly conserved among the plant acyl-CoA-dependent acyltransferase family (St-Pierre et al. 1998, St-Pierre and De Luca 2000) was found, suggesting the authenticity of the 25 kDa polypeptide, at least as a part of functional HMT/HLT. The remaining four peptide sequences found in the 25 and 30 kDa bands showed no significant homology with the sequences registered in the protein data banks. The sequence found in the 60 kDa band exhibited a similarity to pectin esterase, implying other contaminating protein still in this fraction. Isolation of cDNA encoding HMT/HLT Since the 25 kDa polypeptide possessed a sequence motif common for plant acyltransferases, combinations of degenerated primers encoding the determined amino acid sequences from the 25 kDa band were designed for nested reverse transcription–polymerase chain reaction (RT–PCR) amplification. The overlapping 242 and 89 bp nucleotide fragments were amplified specifically with combinations of the forward primer P25K-3F-1 and the reverse primers P25K-1R or P25K-2R, respectively (Fig. 3). This 242 bp amplified fragment was used to isolate 14 cDNA clones by screening of the cDNA library generated from the roots of 10-day-old L. albus, and sequence analyses indicated that all clones were identical apart from their different lengths. The full-length HMT/HLT cDNA sequence was obtained by 5′-rapid amplification of cDNA end (RACE) to yield a 1,641 bp poly(A) tail-containing clone with a putative open reading frame (ORF) of 453 amino acids (Fig. 3). The calculated molecular mass of the encoded protein was 51.2 kDa, which was consistent with that of the purified protein determined by SDS–PAGE in our previous study (Suzuki et al. 1994). The ORF of the isolated cDNA contained the two peptide sequences obtained from the 30 kDa band in the purified HMT/HLT fraction (Fig. 3), although the cDNA was isolated solely with the probe encoding the partial 25 kDa band. It seemed that the 51.2 kDa protein might be cleaved to give 25 and 30 kDa bands. Then we determined the N-terminal amino acid sequences of these two 25 and 30 kDa bands isolated from the gel. The N-terminal sequence (APQTQ) of the 30 kDa band was identical to one of the peptides isolated upon trypsin digestion, and perfectly matched the sequence from the second to the sixth amino acid of the deduced sequence from the cDNA as shown in Fig. 3. The N-terminal sequence (FILQH) of the 25 kDa band was identical to the deduced sequence from the cDNA. From these results, we confirmed that the 25 and 30 kDa peptides were encoded by the single cDNA. Phylogenetic analyses of HMT/HLT Phylogenetic analysis of the deduced protein sequences revealed that HMT/HLT is separated from the subfamily of other alkaloid acyltransferases such as deacetylvindorine acetyltransferase (St-Pierre et al. 1998) and salutaridinol 7-O-acetyltransferase (Grothe et al. 2001) (Fig. 4). The most closely related sequence to HMT/HLT is benzoyl-CoA:benzylalcohol benzoyltransferase from Nicotiana tabacum (D’Auria et al. 2002) with 57% identity on an amino acid sequence level (Fig. 4). The sequence identities of HMT/HLT to other members of this subfamily are 55% to benzoyl-CoA:benzylalcohol benzoyltransferase from Clarkia brewei and 49% to acetyl-CoA:cis-3-hexen-1-ol acetyltransferase from Arabidopsis thaliana (D’Auria et al. 2002). Catalytic activity of recombinant HMT/HLT produced in Escherichia coli To confirm that the isolated cDNA encodes the catalytically active HMT/HLT, the functional analysis was carried out with the recombinant protein heterogeneously produced in E. coli. The recombinant HMT/HLT protein was purified by glutathione S-transferase (GST) tag affinity and subsequent cleavage of the tag (Fig. 5). By in vitro enzymatic assay using the protein extracts of E. coli and the purified recombinant protein, the single product in the reaction using 13α-hydroxymultiflorine and tigloyl-CoA as substrates was detected on high performance liquid chromatography (HPLC), while no products were formed in negative control reactions using E. coli extracts of the empty vector (Fig. 6a). This specific product was identified as 13α-tigloyloxymultiflorine by liquid chromatography-photodiode array detection-electrospray ionization-mass spectrometry (LC-PDA-ESI/MS) analysis (Fig. 6b). In a similar way, the activity for 13α-hydroxylupanine was also confirmed with tigloyl-CoA as the acyl donor. These results indicated that the isolated cDNA actually encoded the functional HMT/HLT. Substrate specificity and inhibition of recombinant HMT/HLT To investigate the substrate specificity of HMT/HLT, the activity of purified recombinant HMT/HLT protein was assayed with the five different acyl-CoA derivatives with 13α-hydroxymultiflorine (Fig. 7a) and 13α-hydroxylupanine (Fig. 7b). Interestingly, the activities with benzoyl-CoA as acyl donor were ∼1.8-fold higher than those with tigloyl-CoA with two alkaloid aglycons. The activities with acetyl-CoA, propionyl-CoA and 2-butenoyl-CoA were ∼2–20% of those with tigloyl-CoA. The Km values for 13α-hydroxymultiflorine, 13α-hydroxylupanine, tigloyl-CoA and benzoyl-CoA were determined (Table 2). Regarding alkaloid aglycon, 13α-hydroxymultiflorine is a better substrate than 13α-hydroxylupanine, judged from the Km values for each alkaloid and for each acyl donor with these alkaloids as co-substrate. However, no apparent difference in Km values was found between tigloyl-CoA and benzoyl-CoA. Phylogenetic analysis indicated that the sequence of HMT/HLT was close to that of benzoyl-CoA:benzylalcohol benzoyltransferase. Thus, to confirm whether HMT/HLT possesses the activity of benzoyl-CoA:benzylalcohol benzoyltransferase, we performed the assay for the reaction using benzoyl-CoA and benzyl alcohol as the substrates with the recombinant HMT/HLT. Determination of the possible reaction product, benzylbenzoate, was carried out with gas chromatography–mass spectrometry. However, no benzylbenzoate formation was detected with the recombinant HMT/HLT (data not shown), suggesting that HMT/HLT does not catalyze this reaction, despite its close sequence similarity to that of benzoyl-CoA:benzylalcohol benzoyltransferase. The HMT/HLT activity was inhibited by the pre-treatment of the enzyme protein with p-chloromercuribenzoic acid, a sulfhydryl blocker, and diethylpyrocarbonate (DEPC), a histidine blocker (Table 3). These results suggested that cysteine and histidine residues are essential for the catalytic function of HMT/HLT. mRNA expression of the HMT/HLT gene Southern blot analysis indicated that a small number (1–2 copies) of homologous genes are present in the genome of L. albus (data not shown). Northern blot analysis was also carried out as shown in Fig. 8. The mRNA expression of the HMT/HLT gene was the highest in roots followed by that in hypocotyls. In contrast, HMT/HLT gene expression in leaves and cotyledons was at a low level. No expression was detected in young developing leaves (data not shown). Discussion Although the quinolizidine alkaloids form a large family of plant secondary products, no genes encoding enzymes involved in their biosynthesis have been cloned, presumably because of the weak enzymatic activities found in plants. Only a limited number of reports are available for detection of the enzymatic activities in the cell-free protein extracts for quinolizidine alkaloid biosynthesis (Ohmiya et al. 1995). In the present study, we have cloned, for the first time, the cDNA encoding a biosynthetic enzyme for quinolizidine alkaloids. This study could lead to the molecular investigation of metabolism of this large family of plant secondary products. The present study is also the first example of the molecular cloning of a tigloyl transferase from any biological source including plants. Tigloyl esters are widely distributed in plants as minor components of secondary products, e.g. alkaloids (Lounasmaa 1988, Ohmiya et al. 1995, Saito and Murakoshi 1995). However, no tigloyl esters and tigloyl transferase have been found in microorganisms and animals. The tigloyl moiety is derived from isoleucine through several steps, as suggested for tigloyl esters of tropane alkaloids in Datura plants (Leete 1973, McGaw and Woolley 1977). Tigloyl-CoA formed by this pathway is assumed to be used for the esterification of a variety of aglycons including hydroxylated alkaloids by specific transferases such as HMT/HLT. The evolutionarily related plant acyl-CoA-dependent acyltransferases (EC 2.3.1.x) are referred to as the BAHD family based on the names of the first genes isolated (BEAT AHCT HCBT1 DAT) exhibiting characteristic sequence motifs (St-Pierre and De Luca 2000). This family consists of a large number of members; for example, there are approximately 60 BAHD gene family members in A. thaliana. However, only a limited numbers of genes have been characterized with their biochemical functions. Phylogenetically, the deduced amino acid sequence of HMT/HLT is closely related to benzoyl-CoA:benzyl alcohol benzoyltransferase (BEBT) from C. breweri and N. tabacum, which is presumed to be involved in the formation of volatile esters in flowers (D’Auria et al. 2002). Indeed, HMT/HLT accepted benzoyl-CoA instead of tigloyl-CoA for acyltransfer to 13α-hydroxymultiflorine and 13α-hydroxylupanine. However, HMT/HLT could not catalyze the benzoyl-CoA:benzyl alcohol benzoyltransferase reaction, indicating a relatively strict specificity of HMT/HLT regarding hydroxylated aglycons. We have shown previously that the purified HMT/HLT only utilized the 13α(axial)-hydroxylated tetracyclic alkaloids as acyl acceptors (Suzuki et al. 1994). Although HMT/HLT, benzoyl-CoA:benzyl alcohol benzoyltransferase and acety-CoA:cis-3-hexene-1-ol acetyltransferase from A. thaliana (D’Auria et al. 2002) belong to the same subfamily of the BAHD acyltransferases, the substrate specificities of these enzymes are rather diverse. Similarly, the diverse substrate specificity in a single subfamily was reported for phylogenetically closely related acyltransferases from Taxus species (Walker and Croteau 2001). These results suggest that the substrate specificity of the BAHD acyltransferase could evolve relatively easily. Since Km values of HMT/HLT for tigloyl-CoA and benzoyl-CoA are nearly the same (Table 2) and the activity with benzoyl-CoA is higher than that with tigloyl-CoA (Fig. 7), this enzyme is more properly referred to as tigloyl/benzoyl-CoA:(–)-13α-hydroxymultiflorine/(+)-13α-hydroxylupanine O-tigloyl/benzoyltransferase. However, only a trace amount of benzoyloxylupanine was detected (Wink and Witte 1984) and no apparent accumulation of benzoyloxymultiflorine was reported in L. albus. This presumably is due to the limited supply of benzoyl-CoA compared with tigloyl-CoA or different compartmentation of each acyl-CoA and HMT/HLT in this plant. The accumulation of benzoyloxylupanine has been reported in other taxonomically related Lupinus species, e.g. L. polyphyllus (Wink et al. 1982b) and L. angustifolius (Strack et al. 1991, Hirai et al. 2000). HMT/HLT presumably is responsible for biosynthesis of benzoyloxylupanine in these species. Previously we purified HMT/HLT as a single protein of 50 kDa (Suzuki et al. 1994). However, we noticed that 30 and 25 kDa bands were associated with the activity and could not be separated. In the present study, we purified these two bands to obtain sufficient amounts of enzyme protein for peptide sequencing. The isolated cDNA encoded a single 50 kDa protein, which contained the partial peptide sequences from the 30 and 25 kDa bands. These results suggest that the cleavage of the 50 kDa protein into 30 and 25 kDa bands is an artifact of the purification procedure. Nevertheless, these fragments still retained the catalytic activity, presumably remaining physically associated. A similar situation has been reported for acetyl-CoA:deacetylvindoline 4-O-acetyltransferase from Catharanthus roseus (periwinkle) (Power et al. 1990, St-Pierre et al. 1998) and for hydroxycinnamoyl/benzoyl-CoA:anthranilate N-hydroxycinnamoyl/benzoyltransferase from Dianthus caryophyllus (Yang et al. 1997). This could be a common character of the proteins belonging to this family. As shown in Fig. 9, the members of the BAHD acyltransferase family have a few conserved sequence motifs. In addition to a highly conserved DFGWG motif as the signature of this family, cysteine and histidine residues indicated in Fig. 9 are also conserved in all sequences. The histidine residue in the HXXXD(G) motif has been postulated to function as a general base in catalysis of the acyltransfer reaction (Shaw and Leslie 1991, St-Pierre et al. 1998). Indeed, the activity of HMT/HLT was inhibited by DEPC, although the inhibition required a higher concentration of DEPC than previously reported for acyltransferases (for example, St-Pierre et al. 1998, St-Pierre and De Luca 2000, Grothe et al. 2001). The requirement for a higher concentration of DEPC could be due to the different protein folding pattern that prevents the access of DEPC to the catalytic histidine. The HMT/HLT activity was also inhibited by p-chloromercuribenzoic acid, indicating the involvement of a cysteine residue in the catalytic function. This could be the conserved cysteine residue indicated in Fig. 9, that may be a component of a catalytic triad as suggested previously (Grothe et al. 2001). The HMT/HLT gene was expressed specifically in roots and hypocotyls, and only a limited level of expression has been detected in cotyledons and leaves. This expression pattern is in good agreement with the enzymatic activity of HMT/HLT, representing ∼84% of the total activity found in roots and hypocotyls (Suzuki et al. 1994). These results indicate that roots and hypocotyls are the main organs for the biosynthesis of quinolizidine alkaloids in plant organs. Previously we also reported that the distribution of several different acyltransferases in different plant species is well correlated with the accumulation of acylated quinolizidine alkaloids (Suzuki et al. 1994), suggesting the importance of the terminal acyltransferase for alkaloid accumulation patterns in diverse plant species. Previously we detected HMT/HLT activities dominantly in the particle fractions associated with the mitochondrial marker enzymes (Suzuki et al. 1996). In fact, by the prediction of the iPSORT program (http://psort.nibb.ac.jp/), the N-terminal sequence of the deduced HMT/HLT showed the highest score for translocation to the mitochondria. However, the sequence of purified HMT/HLT started from the second amino acid of the deduced sequence derived from cDNA, indicating that no processing as an ordinary mitochondrial transit peptide takes place. Since there are some examples of a lack of a cleavable mitochondrial pre-sequence (Sjöling and Glaser 1998), the N-terminal sequence of HMT/HLT may act as a targeting signal to mitochondria without extensive processing. Otherwise HMT/HLT may localize in another particle fraction such as the peroxisomes, since there is a signal peptide-like sequence (SHI) for peroxisome translocation in the C-terminus of the HMT/HLT protein. This sequence shows a close similarity to the characteristic C-terminal peroxisome targeting signal type 1 (Reumann 2004). Thus, it still remains an open question in which subcellular compartment HMT/HLT localizes. In some Lupinus species, the alkaloid-rich ‘bitter’ form and the alkaloid-poor ‘sweet’ form are available. Recently, we have investigated the alkaloid accumulation patterns, gene expression profiles by cDNA-amplified fragment length polymorphism and the HMT/HLT activities of the ‘bitter’ and ‘sweet’ forms of L. angustifolius (Hirai et al. 2000). Although striking differences in alkaloid accumulation were observed between the two forms, no significant difference in the HMT/HLT activity was seen in the hypocotyl cell-free extracts of these forms. These results suggest that HMT/HLT is regulated independently from the genetic factor that determines the alkaloid accumulation in ‘bitter’ and ‘sweet’ forms. In conclusion, we have cloned and characterized the cDNA encoding HMT/HLT responsible for determination of alkaloid accumulation patterns in L. albus. The present findings can serve as the molecular information for a better understanding and a further metabolic engineering of quinolizidine alkaloid metabolism in plants. Materials and Methods Chemicals (–)-13α-Hydroxymultiflorine and (+)-13α-hydroxylupanine used in this study were from our laboratory stock. Tigloyl-, acetyl-, propionyl-, 2-butenoyl- and benzoyl-CoA were purchased from Sigma-Aldrich (St Louis, MO, U.S.A.). The molecular weight markers used for SDS–PAGE were purchased from Amersham Biosciences (Piscataway, NJ, U.S.A.). Bradford protein dye reagent was purchased from Bio-Rad (Hercules, CA, U.S.A.). Restriction enzymes were purchased from TaKaRa (Ohtsu, Japan) or Toyobo (Osaka, Japan). T4 DNA ligase was purchased from Promega (Madison, WI, U.S.A.) or Novagen EMD Biosciences (Madison, WI, U.S.A.). DNA amplification by PCR using Pyrobest or Ex Taq DNA polymerase (TaKaRa) was carried out in an iCycler thermocycler (Bio-Rad). Other chemicals and enzymes were of the highest grade available. Plant materials The seeds of L. albus Forsk were germinated in moistened vermiculite in daylight in a greenhouse at ∼25°C as described previously (Suzuki et al. 1994). Purification of HMT/HLT All procedures were performed at 4°C. Crude protein was extracted from roots, hypocotyls and epicotyls (2.47 kg, fresh weight) of 13-day-old L. albus seedlings as described (Suzuki et al. 1994). The precipitate appearing between 30 and 80% (NH4)2SO4 was collected and dissolved in buffer A (20 mM sodium phosphate, pH 7.0, containing 10 mM 2-mercaptoethanol and 1.5 M ammonium sulfate). The proteins were absorbed on butyl FF (Amersham) equilibrated with buffer A and were eluted at a flow rate of 0.8 ml min–1 with a linear gradient of 1.5–1.0 M ammonium sulfate. The fractions with the enzyme activity were collected and concentrated. The proteins were then separated on Superdex 200 pg (Amersham) equilibrated with buffer B (20 mM sodium phosphate, pH 6.3, containing 10 mM 2-mercaptoethanol) at a flow rate of 1.0 ml min–1. The fractions with the enzyme activity were collected and then applied to SP Sepharose XL (Amersham). The proteins were eluted at a flow rate of 2.0 ml min–1 with a linear gradient of 0–0.5 M NaCl, and the fractions with the enzyme activity were collected and concentrated. The portion (0.125 volume) of the concentrated enzyme solution was then separated on Superdex 200 HR (Amersham) equilibrated with buffer B at a flow rate of 0.3 ml min–1. The fractions with the enzyme activity were collected and then applied to RESOURCE S (Amersham) equilibrated with buffer B. Proteins were eluted at a flow rate of 4.0 ml min–1 with a linear gradient of 0–0.5 M NaCl. Determination of internal and N-terminal amino acid sequences After the final step of protein purification by RESOURCE S, the fraction with the highest enzymatic activity was subjected to 12.5% SDS–PAGE. Bands were detected by silver staining. The three observed bands of 25, 30 and 60 kDa proteins were subjected to determination of the internal amino acid sequences. The respective bands were digested in-gel by trypsin and then separated by reverse-phase HPLC. HPLC separation was performed on a TSKgel ODS-80Ts QA (2.0×250 mm) column (Tosoh, Tokyo, Japan) with a gradient of 0–90% acetonitrile in 0.1% trifluoroacetic acid (TFA) at a flow rate of 0.2 ml min–1. Separated peptides were monitored at 210 and 280 nm. Sequencing of the peptides was performed using a Procise 494 cLC (Applied Biosystems, Foster City, CA, U.S.A.) or Hewlett-Packard G1005A (Hewlett Packard, Palo Alto, CA, U.S.A.) protein sequencing system. To determine the N-terminal amino acid sequence of the 25 and 30 kDa bands, the peptides in the gel of SDS–PAGE were transferred to a PVDF membrane and then analyzed by protein sequencer, Procise 494 HT (Applied Biosystems). Molecular cloning of cDNA Isolation of partial fragments encoding the 25 kDa peptide by RT–PCR—Total RNA was extracted from roots of 10-day-old L. albus with the RNeasy Plant Mini Kit (QIAGEN, Hilden, Germany) or TRIzol reagents (Invitrogen, Carlsbad, CA, U.S.A.). mRNA was purified using an mRNA Purification Kit (Amersham). From total RNA extracted from roots, first-strand cDNA was synthesized using AMV reverse transcriptase XL (TaKaRa) with the Oligo dT-3sites Adaptor Primer (TaKaRa). The degenerate primers were designed from the amino acid sequence of the 25 kDa peptide. The first PCR was performed using the forward primers P25K-1F, GARYTNGAYGAYYTNTTYAA (corresponding to the amino acid sequence of ELDDLFK); P25K-2F, GAYGTNGAYTTYGGNTGGGGNAA (corresponding to the amino acid sequence of DVDFGWGK); or P25K-3F-1, CCNWSNTAYTTYTAYAAYGA (corresponding to the amino acid sequence of PSYFYND) and the reverse primer of the 3sites Adaptor Primer (TaKaRa) which hybridizes to a region synthesized by the Oligo dT-3sites Adaptor Primer. With the forward primer corresponding to these templates and the other reverse primers, which have complementary sequences to those of the forward primers, P25K-1R, P25K-2R and P25K-3R-1 (corresponding to P25K-1F, P25K-2F and P25K-3F-1, respectively), the nested PCR was performed with all possible combinations of the forward and reverse primers. The amplified fragments were cloned into pT7Blue T-vector (Novagen) by TA-cloning, and the nucleotide sequences of the inserts were determined. Isolation of the full-length clone by cDNA library screening and 5′-RACE—The cDNA library was constructed with the λZAPII vector (Stratagene, La Jolla, CA, U.S.A.) from mRNA of L. albus, and 2.0×105 plaques of primary cDNA library were screened by the PCR-amplified fragment. Plaques were blotted onto positively charged nylon membranes, Hybond N+ (Amersham). Duplicate filters were hybridized with the 32P-labeled probe prepared by the Random Primer DNA Labeling Kit Ver.2 (TaKaRa) using the 242 bp nucleotide fragment, which was amplified specifically by RT–PCR with the combination of P25K-3F-1 and P25K-1R as a template. The final post-hybridization wash was performed in 0.2×SSPE (Sambrook et al. 1989) and 0.1% SDS at 65°C. For extension of the 5′-terminus of the cDNA, 5′-RACE was performed using the 5′-Full RACE Core Set (TaKaRa). Expression and purification of recombinant HMT/HLT BamHI sites were created on both ends of the HMT/HLT coding region by PCR using the two primers: BamHI F, CGCGGATCCATGGCTCCCCAAACTCAATCTCTA; and BamHI R, CGCGGATCCTCAAATGTGTGACCTTAATATTTG. The engineered cDNA fragment was inserted into the BamHI site of pGEX-6P-2 (Amersham) with the correct orientation confirmed by sequence analysis as designated pGEX-HMT/HLT, which gives a recombinant gene product with an N-terminal GST protein tag. E. coli BL21 cells harboring pGEX-HMT/HLT and empty pGEX were grown at 20°C in 2×YT-ampicillin (100 µg ml–1) liquid medium to an OD600 = 0.7–1.0. The solution of 0.1 mM isopropyl-β-d-thiogalactoside was added immediately for inducible production of recombinant protein. The cells were suspended in 1×phosphate-buffered saline [140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4 and 1.8 mM KH2PO4 (pH 7.3)] and disrupted by sonication. After the addition of Trition X-100 to 1% concentration and following centrifugation, the supernatant was used as crude protein extracts. For purification of the recombinant HMT/HLT, a GSTrap column (Amersham), whose resin has glutathione as ligand resulting in affinity for GST, was used. GST tags were removed from GST fusion recombinant proteins by in-column digestion using PreScission Protease (Amersham). Assay of enzymatic activity The HMT/HLT activity was determined by detection of the ester alkaloid formed from alkaloid substrates and acyl-CoA derivatives by enzymatic reaction. The standard reaction mixture consisted of 100 mM potassium phosphate (pH 8.0), 0.5 mM EDTA, 1 mM dithiothreitol (DTT), 0.15 mM alkaloid substrate, 0.15 mM acyl-CoA derivative and protein. After incubation for 10–60 min at 30°C, the reaction was terminated by addition of a 0.67 volume of methanol. The reaction mixture was analyzed directly by HPLC. HPLC analysis was performed on a Mightysil RP-18 GP 150–4.6 (5 µm) column (Kanto Chemical, Tokyo, Japan) with a mobile phase of 15% acetonitrile and 20 mM sodium phosphate (pH 5.5) at a flow rate of 0.5–1.0 ml min–1. The formation of 13α-hydroxymultiflorine and 13α-hydroxylupanine derivatives was monitored at 327 and 220 nm, respectively, as described (Saito et al. 1989). The identification of products was confirmed further by LC-ESI/MS. To elucidate the substrate specificity, the reaction with recombinant HMT/HLT protein was carried out with the various acyl-CoA derivatives and the aglycons. The products were identified with LC-ESI/MS LC-PDA-ESI/MS analysis The reaction mixtures were analyzed by an LC-PDA-ESI/MS system consisting of a Finnigan LCQ DECA mass spectrometer (Thermo Quest, San Jose, CA, U.S.A.) and an Agilent HPLC 1100 series (Agilent technologies, Palo Alto, CA, U.S.A.). LC separations were performed on a Mightysil RP-18 GP 150–4.6 (5 µm) column (Kanto Chemical) with a mobile phase of 15% acetonitrile and 0.1% TFA at a flow rate of 0.5 ml min–1 using UV photodiode array detection. Nitrogen was used as sheath gas for positive ion ESI/MS performed at a capillary temperature and voltage of 320°C and 5.0 kV, respectively. The tube lens offset was set at 10.0 V. Full scan mass spectra were acquired from 50 to 1,000 m/z at two scans s–1. Tandem MS analysis was carried out with helium as collision gas. The normalized collision energy was set to 40%. The specific products were identified from the theoretical unit mass for [M+H]+ in +ESI: 13-tigloyloxymultiflorine, 345; 13-acetyloxymultiflorine, 305; 13-propionyloxymultiflorine, 319; 13-(2-butenoyl)oxymultiflorine, 331; 13-benzoyloxymultiflorine, 367; 13-tigloyloxylupanine, 347; 13-acetyloxylupanine, 307; 13-propionyloxylupanine, 321; 13-(2-butenoyl)oxylupanine, 333; and 13-benzoyloxylupanine, 369. Fragmented ions derived from these quasi-molecular ions were also identified by the MS/MS technique. The relative activities regarding each acyl-CoA were calculated by relative integration of ion areas of quasi-molecular ions compared with the products of the tigloyl transferase reaction. Northern and Southern blot analyses For northern blot analysis, 40 µg of total RNA isolated from leaves, developing (folded) leaves, cotyledons, hypocotyls and roots was denatured and separated in a formaldehyde agarose (1.2%) gel. For Southern blot analysis, genomic DNA was extracted from L. albus using the DNeasy Plant Mini Kit (QIAGEN). Aliquots of 20 µg of genomic DNA were digested with BamHI or EcoRV and separated on a 0.8% agarose gel. After transfer to a Hybond N+ membrane, hybridization was carried out with the 32P-labeled probe prepared from the ORF of HMT/HLT cDNA using the Random Primer DNA Labeling Kit Ver.2 (TaKaRa). The final washes on post-hybridization were performed in 0.5× SSPE and 0.1% SDS at 65°C. Hybridization signals were detected with a STORM 860 image analyzer (Amersham). Miscellaneous techniques All recombinant DNA technology principally followed the methods of Sambrook et al. (1989). DNA sequencing was performed by the dideoxy chain termination method. Phylogenetic analysis was carried out by CLUSTAL W packaged in DNASpace Ver. 3.5 (Hitachi Software, Tokyo, Japan), and the tree was drawn by DNASpace. SDS–PAGE and subsequent staining of proteins were performed as described previously (Suzuki et al. 1994). Gas chromatography–mass spectrometry for detection of benzylbenzoate was carried out as described previously (Hirai et al. 2000, D’Auria et al. 2002). Acknowledgments This work was supported, in part, by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan, by CREST of Japan Science and Technology Agency (JST) and by Research for the Future Program (00L01605) ‘Molecular mechanisms on regulation of morphogenesis and metabolism leading to increased plant productivity’. We thank CREST-Plant Molecular Science Satellite Laboratory in the Life Science Research Support Center of Akita Prefectural University for DNA sequencing. View largeDownload slide Fig. 1 Biosynthetic pathway of the ester-type quinolizidine alkaloids, (–)-13α-tigloyloxymultiflorine and (+)-13α-tigloyloxylupanine, found in Lupinus plants. The tetracyclic skeleton of quinolizidine alkaloids is derived from l-lysine via cadaverine. HMT/HLT is committed in the final step of the biosynthesis of ester-type alkaloids, and thus this enzyme is responsible for determination of alkaloid accumulation patterns of the plants. View largeDownload slide Fig. 1 Biosynthetic pathway of the ester-type quinolizidine alkaloids, (–)-13α-tigloyloxymultiflorine and (+)-13α-tigloyloxylupanine, found in Lupinus plants. The tetracyclic skeleton of quinolizidine alkaloids is derived from l-lysine via cadaverine. HMT/HLT is committed in the final step of the biosynthesis of ester-type alkaloids, and thus this enzyme is responsible for determination of alkaloid accumulation patterns of the plants. View largeDownload slide Fig. 2 SDS–PAGE analysis and determined internal amino acid sequences of purified HMT/HLT from L. albus. The proteins from the final purification step were separated by 12.5% SDS–PAGE. Three bands of 25, 30 and 60 kDa polypeptides were detected by silver staining and subjected to determination of the internal amino acid sequences after in-gel trypsin digestion. The amino acid sequences determined are indicated in the boxes. The DFGWG in the 25 kDa band indicates the motif sequence highly conserved among the plant acyl-CoA-dependent acyltransferase family (St-Pierre and De Luca 2000). View largeDownload slide Fig. 2 SDS–PAGE analysis and determined internal amino acid sequences of purified HMT/HLT from L. albus. The proteins from the final purification step were separated by 12.5% SDS–PAGE. Three bands of 25, 30 and 60 kDa polypeptides were detected by silver staining and subjected to determination of the internal amino acid sequences after in-gel trypsin digestion. The amino acid sequences determined are indicated in the boxes. The DFGWG in the 25 kDa band indicates the motif sequence highly conserved among the plant acyl-CoA-dependent acyltransferase family (St-Pierre and De Luca 2000). View largeDownload slide Fig. 3 Nucleotide and deduced amino acid sequences of HMT/HLT cDNA. The amino acid sequences in blue and green indicate the partially determined amino acid sequences of the 25 and 30 kDa peptides, respectively. The blue and green arrowheads indicate the N-terminal sites of the 25 and 30 kDa polypeptides, respectively. The underlined amino acid sequence in red indicates a signature motif sequence highly conserved among the plant acyl-CoA-dependent acyltransferase family. The nucleotide sequence in blue indicates the 242 bp fragment specifically amplified by RT–PCR. Dotted arrows show the positions of primers designed for RT–PCR, P25K-3F-1, P25K-1R and P25K-2R. View largeDownload slide Fig. 3 Nucleotide and deduced amino acid sequences of HMT/HLT cDNA. The amino acid sequences in blue and green indicate the partially determined amino acid sequences of the 25 and 30 kDa peptides, respectively. The blue and green arrowheads indicate the N-terminal sites of the 25 and 30 kDa polypeptides, respectively. The underlined amino acid sequence in red indicates a signature motif sequence highly conserved among the plant acyl-CoA-dependent acyltransferase family. The nucleotide sequence in blue indicates the 242 bp fragment specifically amplified by RT–PCR. Dotted arrows show the positions of primers designed for RT–PCR, P25K-3F-1, P25K-1R and P25K-2R. View largeDownload slide Fig. 4 Phylogenetic tree of the plant acyl-CoA-dependent acyltransferases. The tree was constructed by neighbor-joining distance analysis. Line lengths indicate the relative distances between nodes. The deduced amino acid sequence of HMT/HLT (DDBJ/GenBank/EMBL accession number AB181292) was phylogenetically compared with benzoyl coenzyme A:benzyl alcohol benzoyl transferase (BEBT) from Nicotiana tabacum (AF500202) (D’Auria et al. 2002), BEBT from Clarkia breweri (AF500200) (D’Auria et al. 2002), acetyl-CoA:cis-3-hexen-1-ol acetyl transferase (CHAT) from Arabidopsis thaliana (AF500201) (D’Auria et al. 2002), taxadien-5α-ol O-acetyltransferase (TAT) from Taxus cuspidata (AF190130) (Walker et al. 2000c), taxoid C-13 O-phenylpropanoyltransferase (baccatin III:3-amino-3-phenylpropanoyltransferase, BAPT) from T. canadensis (AY082804) (Walker et al. 2002a), 3′-N-debenzoyl-2′-deoxytaxol N-benzoyltransferase (DBTNBT) from T. canadensis (AF466397) (Walker et al. 2002b), 2α-hydroxytaxane 2-O-benzoyltransferase (TBT) (2-debenzoyl-7,13-diacetylbaccatin III-2-O-benzoyl transferase, DBBT) from T. cuspidata (AF297618) (Walker and Croteau 2000a), 10-deacetylbaccatin III-10-O-acetyltransferase (DBAT) from T. cuspidata (AF193765) (Walker and Croteau 2000b), hydroxycinnamoyl-CoA quinate:hydroxycynnamoyl transferase (HQT) from Lycopersicon esculentum (AJ582652) (Niggeweg et al. 2004), hydroxycinnamoyl-CoA:hydroxyanthranilate N-hydroxycinnamoyltransferase 2 from Avena sativa (AsHHT2) (AB076981) (Yang et al. 2004), hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyltransferase (HCT) from N. tabacum (AJ507825) (Hoffmann et al. 2003), hydroxycinnamoyl/benzoyl-CoA:anthranilate N-hydroxycinnamoyl/benzoyltransferase 1 (HCBT1) from Dianthus caryophyllus (Z84383) (Yang et al. 1997), vinorine synthase (VS) from Rauvolfia serpentina (AJ556780) (Bayer et al. 2004), acetyl-CoA:deacetylvindoline-4-O-acetyltransferase (DAT) from Catharanthus roseus (AF053307) (St-Pierre et al. 1998), salutaridinol 7-O-acetyltransferase (SALAT) from Papaver somniferum (AF339913) (Grothe et al. 2001), malonyl-CoA:anthocyanin 5-glucoside 4′′′-O-malonyltransferase from Salvia splendens (Ss5MaT2) (AY383734) (Suzuki et al. 2004), agmatine coumaroyltransferase (ACT) from Hordeum vulgare (AY228552) (Burhenne et al. 2003), malonyl-CoA:anthocyanidin 3-O-glucoside-6′′-O-malonyltransferase from Dahlia variabilis (Dv3MaT) (AF489108) (Suzuki et al. 2002) and minovincinine 19-hydroxy-O-acetyltransferase (MAT) from C. roseus (AF253415) (Laflamme et al. 2001). View largeDownload slide Fig. 4 Phylogenetic tree of the plant acyl-CoA-dependent acyltransferases. The tree was constructed by neighbor-joining distance analysis. Line lengths indicate the relative distances between nodes. The deduced amino acid sequence of HMT/HLT (DDBJ/GenBank/EMBL accession number AB181292) was phylogenetically compared with benzoyl coenzyme A:benzyl alcohol benzoyl transferase (BEBT) from Nicotiana tabacum (AF500202) (D’Auria et al. 2002), BEBT from Clarkia breweri (AF500200) (D’Auria et al. 2002), acetyl-CoA:cis-3-hexen-1-ol acetyl transferase (CHAT) from Arabidopsis thaliana (AF500201) (D’Auria et al. 2002), taxadien-5α-ol O-acetyltransferase (TAT) from Taxus cuspidata (AF190130) (Walker et al. 2000c), taxoid C-13 O-phenylpropanoyltransferase (baccatin III:3-amino-3-phenylpropanoyltransferase, BAPT) from T. canadensis (AY082804) (Walker et al. 2002a), 3′-N-debenzoyl-2′-deoxytaxol N-benzoyltransferase (DBTNBT) from T. canadensis (AF466397) (Walker et al. 2002b), 2α-hydroxytaxane 2-O-benzoyltransferase (TBT) (2-debenzoyl-7,13-diacetylbaccatin III-2-O-benzoyl transferase, DBBT) from T. cuspidata (AF297618) (Walker and Croteau 2000a), 10-deacetylbaccatin III-10-O-acetyltransferase (DBAT) from T. cuspidata (AF193765) (Walker and Croteau 2000b), hydroxycinnamoyl-CoA quinate:hydroxycynnamoyl transferase (HQT) from Lycopersicon esculentum (AJ582652) (Niggeweg et al. 2004), hydroxycinnamoyl-CoA:hydroxyanthranilate N-hydroxycinnamoyltransferase 2 from Avena sativa (AsHHT2) (AB076981) (Yang et al. 2004), hydroxycinnamoyl-CoA:shikimate/quinate hydroxycinnamoyltransferase (HCT) from N. tabacum (AJ507825) (Hoffmann et al. 2003), hydroxycinnamoyl/benzoyl-CoA:anthranilate N-hydroxycinnamoyl/benzoyltransferase 1 (HCBT1) from Dianthus caryophyllus (Z84383) (Yang et al. 1997), vinorine synthase (VS) from Rauvolfia serpentina (AJ556780) (Bayer et al. 2004), acetyl-CoA:deacetylvindoline-4-O-acetyltransferase (DAT) from Catharanthus roseus (AF053307) (St-Pierre et al. 1998), salutaridinol 7-O-acetyltransferase (SALAT) from Papaver somniferum (AF339913) (Grothe et al. 2001), malonyl-CoA:anthocyanin 5-glucoside 4′′′-O-malonyltransferase from Salvia splendens (Ss5MaT2) (AY383734) (Suzuki et al. 2004), agmatine coumaroyltransferase (ACT) from Hordeum vulgare (AY228552) (Burhenne et al. 2003), malonyl-CoA:anthocyanidin 3-O-glucoside-6′′-O-malonyltransferase from Dahlia variabilis (Dv3MaT) (AF489108) (Suzuki et al. 2002) and minovincinine 19-hydroxy-O-acetyltransferase (MAT) from C. roseus (AF253415) (Laflamme et al. 2001). View largeDownload slide Fig. 5 Purification and SDS–PAGE analysis of recombinant HMT/HLT produced in E. coli. The crude extracts from E. coli harboring pGEX-HMT/HLT expressing the cDNA and the purified recombinant HMT/HLT were separated by 10% SDS–PAGE. Protein bands were detected by Coomassie brilliant blue staining. The arrowhead indicates the position of the purified HMT/HLT protein by GST tag and subsequent tag cleavage on a column. View largeDownload slide Fig. 5 Purification and SDS–PAGE analysis of recombinant HMT/HLT produced in E. coli. The crude extracts from E. coli harboring pGEX-HMT/HLT expressing the cDNA and the purified recombinant HMT/HLT were separated by 10% SDS–PAGE. Protein bands were detected by Coomassie brilliant blue staining. The arrowhead indicates the position of the purified HMT/HLT protein by GST tag and subsequent tag cleavage on a column. View largeDownload slide Fig. 6 Identification of the product of the reaction catalyzed by the recombinant HMT/HLT by LC-PDA-ESI/MS. (a) HPLC-PDA analysis of the reaction product of the recombinant HMT/HLT. The reaction with 13α-hydroxymultiflorine and tigloyl-CoA was carried out with the protein extract of E. coli harboring pGEX-HMT/HLT (upper chromatogram) or pGEX empty vector (lower chromatogram). HPLC analysis was performed on a Mightysil RP-18 GP 150–4.6 column (Kanto Chemical) with a mobile phase of 15% acetonitrile and 0.1% trifluoroacetic acid at a flow rate of 0.5 ml min–1. The substrate and the product, 13α-tigloyloxymultiflorine, were monitored at 318 nm, the maximum absorbance of these compounds. The arrow indicates 13α-tigloyloxymultiflorine specifically produced by the recombinant HMT/HLT. (b) LC-PDA-ESI/MS analysis of the specific product, 13α-tigloyloxymultiflorine, by the recombinant HMT/HLT. The product was identified by its quasi-molecular ion mass at m/z 345 ([M+H]+) (upper spectrum) and its fragmented ion at m/z 245 ([M-C5H7O2]+) by the MS/MS technique (lower spectrum). View largeDownload slide Fig. 6 Identification of the product of the reaction catalyzed by the recombinant HMT/HLT by LC-PDA-ESI/MS. (a) HPLC-PDA analysis of the reaction product of the recombinant HMT/HLT. The reaction with 13α-hydroxymultiflorine and tigloyl-CoA was carried out with the protein extract of E. coli harboring pGEX-HMT/HLT (upper chromatogram) or pGEX empty vector (lower chromatogram). HPLC analysis was performed on a Mightysil RP-18 GP 150–4.6 column (Kanto Chemical) with a mobile phase of 15% acetonitrile and 0.1% trifluoroacetic acid at a flow rate of 0.5 ml min–1. The substrate and the product, 13α-tigloyloxymultiflorine, were monitored at 318 nm, the maximum absorbance of these compounds. The arrow indicates 13α-tigloyloxymultiflorine specifically produced by the recombinant HMT/HLT. (b) LC-PDA-ESI/MS analysis of the specific product, 13α-tigloyloxymultiflorine, by the recombinant HMT/HLT. The product was identified by its quasi-molecular ion mass at m/z 345 ([M+H]+) (upper spectrum) and its fragmented ion at m/z 245 ([M-C5H7O2]+) by the MS/MS technique (lower spectrum). View largeDownload slide Fig. 7 Substrate specificity of recombinant HMT/HLT. The acyltransfer reaction was carried out with recombinant HMT/HLT using 13α-hydroxymultiflorine or 13α-hydroxylupanine as acyl acceptors and various acyl-CoAs as acyl donors. The identification of the reaction products was made by LC-PDA-ESI/MS. The relative activities were determined by intensities of the [M+H]+ ions of the respective products. The data are expressed as the values relative to (a) 13α-tigloyloxymultiflorine or (b) 13α-tigloyloxylupanine. The bar indicates the data of duplicate assays. View largeDownload slide Fig. 7 Substrate specificity of recombinant HMT/HLT. The acyltransfer reaction was carried out with recombinant HMT/HLT using 13α-hydroxymultiflorine or 13α-hydroxylupanine as acyl acceptors and various acyl-CoAs as acyl donors. The identification of the reaction products was made by LC-PDA-ESI/MS. The relative activities were determined by intensities of the [M+H]+ ions of the respective products. The data are expressed as the values relative to (a) 13α-tigloyloxymultiflorine or (b) 13α-tigloyloxylupanine. The bar indicates the data of duplicate assays. View largeDownload slide Fig. 8 Organ-specific expression of the HMT/HLT gene by northern blot analysis. Total RNA (40 µg) isolated from roots, hypocotyls, cotyledons and leaves of 10-day-old L. albus was electrophoresed on an agarose gel (1.2%), transferred to a positively charged nylon membrane and then hybridized with a 32P-labeled probe of the ORF region of HMT/HLT cDNA. The final wash of the membrane was performed in 0.5×SSPE and 0.1% SDS at 65°C. rRNA bands are shown as the control of equal amounts of RNA loaded. View largeDownload slide Fig. 8 Organ-specific expression of the HMT/HLT gene by northern blot analysis. Total RNA (40 µg) isolated from roots, hypocotyls, cotyledons and leaves of 10-day-old L. albus was electrophoresed on an agarose gel (1.2%), transferred to a positively charged nylon membrane and then hybridized with a 32P-labeled probe of the ORF region of HMT/HLT cDNA. The final wash of the membrane was performed in 0.5×SSPE and 0.1% SDS at 65°C. rRNA bands are shown as the control of equal amounts of RNA loaded. View largeDownload slide Fig. 9 Multiple alignment of conserved motifs in the family of plant acyltransferases. Arrowheads indicate possible functional residues of cysteine and histidine that presumably are modified by p-chloromercuribenzoic acid and DEPC, respectively. The residues with a black and gray background indicate those of perfect conservation and >50% identity, respectively, in the 10 proteins. Abbreviations for the proteins are described in the legend to Fig. 4. View largeDownload slide Fig. 9 Multiple alignment of conserved motifs in the family of plant acyltransferases. Arrowheads indicate possible functional residues of cysteine and histidine that presumably are modified by p-chloromercuribenzoic acid and DEPC, respectively. The residues with a black and gray background indicate those of perfect conservation and >50% identity, respectively, in the 10 proteins. Abbreviations for the proteins are described in the legend to Fig. 4. Table 1 Summary of purification of HMT/HLT from L. albus Purification step  Activity (pkat)  Protein (mg)  Specific activity (pkat mg–1)  Yield (%)  Purification factor (-fold)  30–80% (NH4)2SO4  64,700  6,080  10.6  100  1  Butyl FF  23,780  103  231  36  22  Superdex 200 pg  12,870  19  677  20  64  SP Sepharose XL  10,740  3.2  3,356  16  317  Superdex 200 HR  5,201 (655) a  0.79 (0.1)  6,550  8.0  618  RESOURCE S  2,088 (263)  0.29 (0.036)  7,305  3.2  689  Purification step  Activity (pkat)  Protein (mg)  Specific activity (pkat mg–1)  Yield (%)  Purification factor (-fold)  30–80% (NH4)2SO4  64,700  6,080  10.6  100  1  Butyl FF  23,780  103  231  36  22  Superdex 200 pg  12,870  19  677  20  64  SP Sepharose XL  10,740  3.2  3,356  16  317  Superdex 200 HR  5,201 (655) a  0.79 (0.1)  6,550  8.0  618  RESOURCE S  2,088 (263)  0.29 (0.036)  7,305  3.2  689  a The numbers in parentheses indicates the data of the 0.125 aliquot of the purified protein solution from SP Sepharose XL. View Large Table 2 Summary of Km values of HMT/HLT Substrate for Km  Km  Co-substrate  13α-Hydroxymultiflorine  94 µM  Tigloyl-CoA (0.3 mM)  13α-Hydroxylupanine  112 µM  Tigloyl-CoA (0.3 mM)  Tigloyl-CoA   98 µM  13α-Hydroxymultiflorine (0.3 mM)    359 µM  13α-Hydroxylupanine (0.3 mM)  Benzoyl-CoA  93 µM  13α-Hydroxymultiflorine (0.3 mM)    405 µM  13α-Hydroxylupanine (0.3 mM)  Substrate for Km  Km  Co-substrate  13α-Hydroxymultiflorine  94 µM  Tigloyl-CoA (0.3 mM)  13α-Hydroxylupanine  112 µM  Tigloyl-CoA (0.3 mM)  Tigloyl-CoA   98 µM  13α-Hydroxymultiflorine (0.3 mM)    359 µM  13α-Hydroxylupanine (0.3 mM)  Benzoyl-CoA  93 µM  13α-Hydroxymultiflorine (0.3 mM)    405 µM  13α-Hydroxylupanine (0.3 mM)  View Large Table 3 Inhibition of HMT/HLT by sulfhydryl and histidine blockers Inhibitor  Concentration for 50% inhibition  Substrate  p-Chloromercuribenzoic acid  212 µM  13α-Hydroxymultiflorine and tigloyl-CoA    218 µM  13α-Hydroxylupanine and tigloyl-CoA  DEPC  23 mM  13α-Hydroxymultiflorine and tigloyl-CoA    37 mM  13α-Hydroxylupanine and tigloyl-CoA  Inhibitor  Concentration for 50% inhibition  Substrate  p-Chloromercuribenzoic acid  212 µM  13α-Hydroxymultiflorine and tigloyl-CoA    218 µM  13α-Hydroxylupanine and tigloyl-CoA  DEPC  23 mM  13α-Hydroxymultiflorine and tigloyl-CoA    37 mM  13α-Hydroxylupanine and tigloyl-CoA  The enzyme protein was pre-treated with the inhibitor for 15 min prior to the standard acyltransferase reaction as described in Materials and Methods. For the p-chloromercuribenzoic acid experiment, DTT was omitted from the standard reaction mixture. For the DEPC experiment, after pre-treatment by DEPC, imidazole (20 mM) was added to quench the reaction with DEPC. View Large Abbreviations DEPC diethylpyrocarbonate GST glutathione S-transferase HMT/HLT tigloyl-CoA:(–)-13α-hydroxymultiflorine/ (+)-13α-hydroxylupanine O-tigloyltransferase LC-PDA-ESI/MS liquid chromatography-photodiode array detection-electrospray ionization-mass spectrometry ORF open reading frame RACE rapid amplification of cDNA end RT–PCR reverse transcription–polymerase chain reaction. 3 These authors contributed equally to this work. 4 Present address: Kazusa DNA Research Institute, Kisarazu, Japan. 5 Corresponding author: E-mail, ksaito@faculty.chiba-u.jp; Fax, +81-43-290-2905. The nucleotide sequence reported in this paper has been submitted to GenBank, EMBL, DDBJ under the accession number AB181292 [tigloyl-CoA:(–)-13α-hydroxymultiflorine/ (+)-13α-hydroxylupanine O-tigloyltransferase from Lupinus albus]. References Bayer, A., Ma, X. and Stöckigt, J. ( 2004) Acetyltransfer in natural product biosynthesis—functional cloning and molecular analysis of vinorine synthase. Bioorg. Med. Chem.  12: 2787–2795. Google Scholar Burhenne, K., Kristensen, B.K. and Rasmussen, S.K. ( 2003) A new class of N-hydroxycinnamoyltransferases. Purification, cloning, and expression of a barley agmatine coumaroyltransferase (EC 2.3.1.64). J. Biol. Chem.  278: 13919–13927. Google Scholar D’Auria, J.C., Chen, F. and Pichersky, E. ( 2002) Characterization of an acyltransferase capable of synthesizing benzylbenzoate and other volatile esters in flowers and damaged leaves of Clarkia breweri. Plant Physiol.  130: 466–476. Google Scholar Grothe, T., Lenz, R. and Kutchan, T.M. ( 2001) Molecular characterization of the salutaridinol 7-O-acetyltransferase involved in morphine biosynthesis in opium poppy Papaver somniferum. J. Biol. Chem.  276: 30717–30723. Google Scholar Hirai, M.Y., Suzuki, H., Yamazaki, M. and Saito, K. ( 2000) Biochemical and partial molecular characterization of bitter and sweet forms of Lupinus angustifolius, an experimental model for study of molecular regulation of quinolizidine alkaloid biosynthesis. Chem. Pharm. Bull.  48: 1458–1461. Google Scholar Hoffmann, L., Maury, S., Martz, F., Geoffroy, P. and Legrand, M. ( 2003) Purification, cloning, and properties of an acyltransferase controlling shikimate and quinate ester intermediates in phenylpropanoid metabolism. J. Biol. Chem.  278: 95–103. Google Scholar Laflamme, P., St-Pierre, B. and De Luca, V. ( 2001) Molecular and biochemical analysis of a Madagascar periwinkle root-specific minovincinine-19-hydroxy-O-acetyltransferase. Plant Physiol.  125: 189–198. Google Scholar Leete, E. ( 1973) Biosynthetic conversion of α-methylbutyric acid to tiglic acid in Datura meteloides. Phytochemistry  12: 2203–2205. Google Scholar Lounasmaa, M. ( 1988) The Alkaloids: The Tropane Alkaloids. Edited by Brossi, A. Vol. 33. Academic Press, Inc., CA. Google Scholar McGaw, B.A. and Woolley, J.G. ( 1977) Non-reversibility of the isoleucine-acetate pathway in Datura. Phytochemistry  16: 1711–1713. Google Scholar Niggeweg, R., Michael, A.J. and Martin, C. ( 2004) Engineering plants with increased levels of the antioxidant chlorogenic acid. Nature Biothechnol.  22: 746–754. Google Scholar Ohmiya, S., Saito, K. and Murakoshi, I. ( 1995) The Alkaloids: Lupine Alkaloids. Edited by Cordell, G.A. Vol. 47. Academic Press, Inc., CA. Google Scholar Power, R., Kurz W.G. and De Luca, V. ( 1990) Purification and characterization of acetylcoenzyme A:deacetylvindoline 4-O-acetyltransferase from Catharanthus roseus. Arch. Biochem. Biophys.  279: 370–376. Google Scholar Reumann, S. ( 2004) Specification of the peroxisome targeting signals type 1 and type 2 of plant peroxisomes by bioinformatics analyses. Plant Physiol.  135: 783–800. Google Scholar Roberts, M.F. and Wink, M. ( 1998) Alkaloids Biochemistry, Ecology, and Medicinal Applications. Plenum Press, New York. Google Scholar Saito, K., Kobayashi, K., Ohmiya, S., Otomasu, H. and Murakoshi, I. ( 1989) Analysis of lupine alkaloids in plants by high-performance liquid chromatography. J. Chromatogr. A . 462: 333–340. Google Scholar Saito, K. and Murakoshi, I. ( 1995) Studies in Natural Chemistry: Chemistry, Biochemistry and Chemotaxonomy of Lupine Alkaloids in Leguminosae. Edited by Rahman, A.-U. Vol. 15. Elsevier Science, Oxford. Google Scholar Saito, K., Suzuki, H., Takamatsu, S. and Murakoshi, I. ( 1993) Acyltransferases for lupin alkaloids in Lupinus hirsutus. Phytochemistry . 32: 87–91. Google Scholar Sambrook, J., Fritsch, E.F. and Maniatis, T. ( 1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Google Scholar Shaw, W.V. and Leslie, A.G. ( 1991) Chloramphenicol acetyltransferase. Annu. Rev. Biophys. Biophys. Chem.  20: 363–386. Google Scholar Sjöling, S. and Glaser, E. ( 1998) Mitochondrial targeting peptides in plants. Trends Plant Sci.  3: 136–140. Google Scholar St-Pierre, B. and De Luca, V. ( 2000) Recent Advances in Phytochemistry Evolution of Metabolic Pathways. Edited by John, R.I., Romeo, T., Varin, L. and De Luca, V. Vol. 34. Elsevier Science, Oxford. Google Scholar St-Pierre, B., Laflamme, P., Alarco, A.-M. and De Luca, V. ( 1998) The terminal O-acetyltransferase involved in vindoline biosynthesis defines a new class of proteins responsible for coenzyme A-dependent acyl transfer. Plant J.  14: 703–713. Google Scholar Strack, D., Becher, A., Brall, S. and Witte, L. ( 1991) Quinolizidine alkaloids and the enzymatic syntheses of their cinnamic and hydroxycinnamic acid esters in Lupinus angustifolius and L. luteus. Phytochemistry  30: 1493–1498. Google Scholar Suzuki, H., Koike, Y., Murakoshi, I. and Saito, K. ( 1996) Subcellular localization of acyltransferases for quinolizidine alkaloid biosynthesis in Lupinus. Phytochemistry  42: 1557–1562. Google Scholar Suzuki, H., Murakoshi I. and Saito, K. ( 1994) A novel O-tigloyltransferase for alkaloid biosynthesis in plants. Purification, characterization, and distribution in Lupinus plants. J. Biol. Chem.  269: 15853–15860. Google Scholar Suzuki, H., Nakayama, T., Yonekura-Sakakibara, K., Fukui, Y., Nakamura, N., Yamaguchi, M.A., Tanaka, Y., Kusumi, T. and Nishino, T. ( 2002) cDNA cloning, heterologous expressions, and functional characterization of malonyl-coenzyme A:anthocyanidin 3-O-glucoside-6′′-O-malonyltransferase from dahlia flowers. Plant Physiol.  130: 2142–2151. Google Scholar Suzuki, H., Sawada, S., Watanabe, K., Nagae, S., Yamaguchi, M.A., Nakayama, T. and Nishino, T. ( 2004) Identification and characterization of a novel anthocyanin malonyltransferase from scarlet sage (Salvia splendens) flowers: an enzyme that is phylogenetically separated from other anthocyanin acyltransferases. Plant J.  38: 994–1003. Google Scholar Tang, W. and Eisenbrand, G. ( 1992) Chinese Drugs of Plant Origin, Chemistry, Pharmacology, and Use in Traditional and Modern Medicine. Springer-Verlag, Berlin. Google Scholar Walker, K. and Croteau, R. ( 2001) Taxol biosynthetic genes. Phytochemistry  58: 1–7. Google Scholar Walker, K. and Croteau, R. ( 2000) Taxol biosynthesis: molecular cloning of a benzoyl-CoA:taxane 2α-O-benzoyltransferase cDNA from Taxus and functional expression in Escherichia coli. Proc. Natl. Acad. Sci. USA  97: 13591–13596. Google Scholar Walker, K. and Croteau, R. ( 2000) Molecular cloning of a 10-deacetylbaccatin III-10-O-acetyl transferase cDNA from Taxus and functional expression in Escherichia coli. Proc. Natl. Acad. Sci. USA  97: 583–587. Google Scholar Walker, K., Fujisaki, S., Long, R. and Croteau, R. ( 2002) Molecular cloning and heterologous expression of the C-13 phenylpropanoid side chain-CoA acyltransferase that functions in taxol biosynthesis. Proc. Natl. Acad. Sci. USA  99: 12715–12720. Google Scholar Walker, K., Long, R. and Croteau, R. ( 2002) The final acylation step in taxol biosynthesis: cloning of the taxoid C13-side-chain N-benzoyltransferase from Taxus. Proc. Natl. Acad. Sci. USA  99: 9166–9171. Google Scholar Walker, K., Schoendorf, A. and Croteau, R. ( 2000) Molecular cloning of a taxa-4(20), 11(12)-dien-5α-ol-O-acetyl transferase cDNA from Taxus and functional expression in Escherichia coli. Arch. Biochem. Biophys.  374: 371–380. Google Scholar Wink, M. and Hartmann, T. ( 1982) Enzymatic synthesis of quinolizidine alkaloid esters: a tigloyl-CoA:13-hydroxylupanine O-tigloyltransferase from Lupinus albus L. Planta  156: 560–565. Google Scholar Wink, M., Schiebel, H.M., Witte, L. and Hartmann, T. ( 1982) Quinolizidine alkaloids from plants and their cell suspension culture. Planta Med.  44: 15–20. Google Scholar Wink, M. and Witte, L. ( 1984) Turnover and transport of quinolizidine alkaloids: diurnal variation of lupanine in the phloem sap, leaves and fruits of Lupinus albus L. Planta  161: 519–524. Google Scholar Yang, Q., Reinhard, K., Schiltz, E. and Matern, U. ( 1997) Characterization and heterologous expression of hydroxycinnamoyl/benzoyl-CoA:anthranilate N-hydroxycinnamoyl/benzoyltransferase from elicited cell cultures of carnation, Dianthus caryophyllus L. Plant Mol. Biol.  35: 777–789. Google Scholar Yang, Q., Trinh, H.X., Imai, S., Ishihara, A., Zhang, L., Nakayashiki, H., Tosa, Y. and Mayama, S. ( 2004) Analysis of the involvement of hydroxyanthranilate hydroxycinnamoyltransferase and caffeoyl-CoA 3-O-methyltransferase in phytoalexin biosynthesis in oat. Mol. Plant Microbe Interact.  17: 81–89. Google Scholar
TI - Molecular Characterization of a Novel Quinolizidine Alkaloid O-Tigloyltransferase: cDNA Cloning, Catalytic Activity of Recombinant Protein and Expression Analysis in Lupinus Plants
JO - Plant and Cell Physiology
DO - 10.1093/pcp/pci021
DA - 2005-01-15
UR - https://www.deepdyve.com/lp/oxford-university-press/molecular-characterization-of-a-novel-quinolizidine-alkaloid-o-pEra4c8X0Y
SP - 233
EP - 244
VL - 46
IS - 1
DP - DeepDyve
ER -