Cloning and expression of a tomato cDNA encoding a methyl jasmonate cleaving esterase Christiane Stuhlfelder, Martin J. Mueller and Heribert Warzecha Lehrstuhl fu ¨ r Pharmazeutische Biologie, Julius-von-Sachs-Institut fu ¨ r Biowissenschaften, Universita ¨ tWu ¨ rzburg, Germany Jasmonic acid and its methyl ester are ubiquitous plant s ig- nalling compounds nec essary f or the regulation of growth and development, as well as for the response of plants to environmental stress f actors. To date, it is not clear w hether methyl jasmonate itself acts as a signal or if its conversion to jasmonic acid i s mandatory pri or t o the induction of a def- ense response. We have cloned a cDNA, e ncoding a m ethyl jasmonate-cleaving enzyme, from tomato cell suspension cultures. Sequence analysis revealed significant similarity to plant esterases and to (S)-hydroxynitrile lyases with an a/b-hydrolase fold structure. The coding sequence was heterologously expressed in Escherichia coli and purified in a c atalytically active form. Transcript levels, as well as en- zymatic a ctivity, were determined in different tomato tissues. High transcript levels and enzyme a ctivities w ere f ound in roots and flowers, while the m RNA level and activity w ere lowinstemsandleaves.Moreover,whentestedinmethyl jasmonate- and elicitor-treated cell s uspension cultures, transcript levels were found to decrease, i ndicating that this particular enzyme might b e a regulator of jasmonate sig- nalling. Keywords: a/b-hydrolase; cell suspension culture; Lyco- persicon e scule ntum; methyl jasmonate esterase; Solanaceae. Jasmonic acid (JA) is a ubiquitous plant compound, which plays a crucial role in t he response to w ounding or pathogen attack, as w ell as i n developmental proce sses, such a s fruit ripening, r oot growth, a nd fertility [ 1]. Most o f the en zymes involved in JA biosynthesis have been characterized at a biochemical and molecular level and the encoding genes have been cloned. Biosynthesis takes place mainly in chloroplasts and peroxisomes, initiating after the release of the precursor linolen ic acid from membrane stores by lipases. The enzymes lipoxygenase, allene oxide synthase, and allene oxide cyclase form the b iosynthesic i ntermediate 12-oxo-phytodienoic a cid (OPDA). Subsequent action of OPDA reductase and three cycles of b-ox idation lead to the formation of JA [ 2,3]. Thereafter, JA may b e esterified to its derivative methyl jasmonate (MeJA) [ 4], or c onjugated with an amino acid or glucose [5]. For most of t he jasmonates it h as been shown that t hey are capable of mediating a response by regulating gene expression [6,7]. Analysis of Arabidopsis thaliana mutants impaired in either JA biosynthe sis or signalling, gave a deeper insight i nto the function of single oxylipin s [8]. T he fad3–2fad7–2fad8 mutant, which forms almost no trienoic fatty acids [9], is male sterile and fertility could only be restored by application of linolenate or JA. Another mutant – opr3 – a rrests jasmonate biosynthesis at the OPDA level and is incapable of metabolizing exogenously applied OPDA to JA [10]. opr3 mutants d isplayed a normal defense response towards a variety of pathogens, indicating that OPDA alone is sufficient to initiate an effective defense response. However, mutant plants were male sterile and fertility c ould be restored by th e exoge nous application of JA. These experiments demonstrate that individual mem- bers of the jasmonate family are involved – at least in Arabidopsis – in different signalling pathways. An Arabidopsis(jar1)mutantwithadefectinthe jasmonate response has been described. The mutant is insensitive to MeJA and does not show root growth inhibition or vegetative storage protein (VSP) ind uction in response to MeJA [11]. Recent analysis of the jar1 locus revealed that its gene product modifies JA v ia adenylation, which is a pparently a p rerequisite for downstream signaling. The modification requires a free carboxyl group as the enzyme does not accept MeJA as a substrate [12]. T herefore, MeJA must be de methylated prior to becoming active. Thus, root growth inhibition and VSP expression are mediated by MeJA through JA, indicating that MeJA is not amediatoronitsowninthisparticularsystem. On the other hand, OPDA and JA can induce identical genes a s w ell as d istinct s ets o f t arget genes, suggesting that independent signalling pathways exist [13] and that the combined action of different inducers might be necessary for the full activation of responsive genes [14]. However, in Correspondence to H. Warzecha , Lehrstuhl fu ¨ r Pharmazeutische Biologie, Julius-von-Sachs-Institut, Julius-von-Sachs-Platz 2, 97082 Wu ¨ rzburg, Germany. Fax: + 4 9 9318886182, Tel.: + 49 9318886162, E-mail: warzecha@biozentrum.uni-wuerzburg.de Abbreviations: HNL, ( S)-hydroxynitrile lyase; JA, j asmonic acid; JMT, S-adenosyl- L -methionine jasmonic a cid carboxyl methyltrans- ferase; MeJA, methyl jasmonate; MJE, methyl jasmonate esterase; MeSA, methyl salicylate; OPD A , 12-oxo-phytodienoic acid; P I , proteinase inhibitors; PNAE, polyneuridine aldehyde esterase; RACE, rapid amplification of c D NA ends; VSP, vege t ative storage protein. Note: The sequence reported h erein was deposited under GenBank accession number AY455313. Note: A website is available at http://132.187.108.6/ (Received 2 8 March 20 04, revised 1 9 May 2004, accepted 25 May 2004) Eur. J. Biochem. 271, 2976–2983 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04227.x the case of MeJA and JA, the biological activities of the exogenously administered compounds are apparently iden- tical, possibly because rapid interconversion takes place in vivo. Another quality of JA-mediated plant defense is the systemic spread of defense responses after local induction. For instance, in tomato and potato plants the production of proteinase inhibitors (PI) as an inducib le defense r esponse against feeding C olorado potato beetle is not limited to the site of their attack, but also appears in distant leaves o f the plant [15]. The herbivore attack induces JA biosynthesis locally via systemin, an 18 amino acid s ignal p eptide. However, the induction of PI genes could be found also in distant parts of the plant, which requires a long distance signalling component. Grafting experiments with tomato mutants deficient i n either J A biosynthesis or J A perception proved that a jasmonate, rather than s ystemin, is the signal which is translocated t hrough the plant [ 16]. Rootstocks of the spr-2 mutant, which are impaired in JA biosynthesis, were not capable of generating a transmissible signal that could induce PI expression in wild-type scions, while wounded wild-type root stocks did induce PI e xpression in spr-2 scions. A strong candidate for a transmissible jasmonate signal is the JA-conjugate, MeJA. The volatile ester can diffuse through membranes and c an be found in the headspace above w ounded leaves [17], suggesting that M eJA m ight be an interplant c ommunication signal [18]. M oreover, it has been speculated that the physical role of MeJA is to mobilize JA [ 19]. The proof of this hypothesis might come from a more detailed understanding of how plants form MeJA from JA, and vice versa. A recent study identified an S-adenosyl- L -methionine:jasmonic acid carboxyl methyl- transferase (JMT) from Arabidopsis, w hich conver ts JA to MeJA [4]. Constitutive overexpression of JMT tripled the MeJA content of transgenic plants and also induced JA- responsive g enes. Therefore, t ransgenic plants with elevated JMT levels s howed enhanced resistance against the necro- trophic fungus Botrytis cinerea, suggesting a prominent role for the enzyme in jasmonate-mediated defense. It is not yet clear whether MeJA functio ns as a paracrine signal that is re leased from sites of pathogen attack to induce defense genes at distant sites. Moreover, i t r emains to be clarified whether MeJA is a mediator on its own that elicits JA responses without prior hydrolysis to JA. How- ever, if M eJA is considered to be a s ignal, there must be a way to regulate the signal by controlled formation and – perhaps more importantly – by its controlled inactivation. A candidate for performing the latter task is an est erase, previously characterized in tomato [20]. E nzyme activity has been found to be constitutively present in cell cultures of many taxonomically distant plant species. In tomato cell cultures, only one MeJA-hydro lysing enzyme could be identified by activity-guided protein purification. The tomato esterase has been purified and characterized from tomato cell suspension cultures. Owing to its MeJA-cleaving activity, we named the enzyme methyl jasmonate esterase (MJE). Yet, it remains to be established w hether MJE has a function in JA/MeJA signalling. As a first step to investigate this, we cloned a cDNA from tomato encoding MJE. Analysis of transcript incidence showed that MJE is differently expressed in different organs as well as after elicitation, indicating that this enzyme might be i nvolved i n jasmonate signalling. Experimental procedures Plant material Tomato plants (Lycopersicon e sculentum cv. M oneymaker) were grown in the greenhouse under con ditions of 16 h light and 8 h darkness. The growth temperature range was 16–22 °C with a relative humidity of 60–70%. Cell suspension cultures (L. esculentum )weregrownin 1 L Erlenmeyer flasks in Linsmaier & Skoog media [21] for 7 d ays under continuous light (600 lux) on orbital s hakers (100 r.p.m.) a t 24 ± 2 °C. Ce lls were harvested by suction filtration,shockfrozeninliquidnitrogen,andstoredat )80 °C until use. Nucleic acid isolation and blot analyses Plant material for the isolation of nucleic acid was either 4-day-old cell s uspension c ulture or 8-week-old plants. T otal RNA from suspension cultures w as isolated either accord- ing to the protocol described previously [22], or with the TRIzol reagent ( Invitrogen), and used for RT-PCR or Northern blots, respectively. Isolation of RNA from p lant material was c arried out using the RNeasy Plant M ini Kit (Qiagen). For DNA purification the protocol described previously [23] was employed. Standard protocols were used for the transfer of RNA and DNA after electrophoretic sep aration [24]. Hybridiza- tion of RNA or DNA transferred to n ylon membranes was performed using a nonradioactive digoxigenin-probe label- ling system (Roche). RT-PCR and cloning of partial- and full-length cDNA Two degenerated primers were designed according to previously determined peptides. Sequences were YTTRTC RCANACNACRTANACNCKRTGNACNSWNCCRTA for MJErev4 and GAYATGGCNGCNWSNG GNATH AAYCC for MJEfor3.1. With total RNA from cell suspen- sion cultures and primer MJErev4 for first-strand synthesis, cDNA was produced using the RT-PCR s ystem fr om Qiagen . RT-PCR conditions were 30 min at 5 0 °C, 15 min at 9 5 °C; five cycles of 30 s at 94 °C, 1 m in at 40 °Cand1minat 72 °C, followed by 30 cycles of 30 s a t 94 °C, 1 min at 45 °C and 1 min at 72 °C. Gel electrophoresis, DNA elution a nd modification was carried out according to standard protocols [24]. A fter c loning o f th e P CR pr oduct into the v ector pG EM- T ( Promega) and s equencing (automated s equencer LI-COR 4200), two homologous primers were designed for the rapid amplification of cDNA ends (RACE) (RACEfor: GTGA CAGCTTTCATGCCTGG; and RACErev: ATCCTGT CCGTTGTTGTAAAC). 5¢-and3¢-RACE w as performed using the SMART II system (BD Bioscience). Full-length cDNA for expression was cloned by another RT-PCR with primers f ullMJEforMQ (GCA TGCAGGGTGATAAAAA TCACTTTGTA) and fullMJErev (AAGGATCCATAA TATTTTTGCGAA ATC), adding rest rictio n sites for SphI and BamHI, r espectively. Th e PCR pr oduct w as cloned i nto vector pDRIVE (Qiagen), sequenced and subcloned into Ó FEBS 2004 Methyl jasmonate esterase from tomato (Eur. J. Biochem. 271) 2977 expression vector pQE70 (Qiagen) via SphIandBglII restriction s ites. Overexpression and purification Escherichia c oli M15 cells harbouring the MJE e xpression plasmid w ere cultured a t 20 °C on a rotary shaker (200 r.p.m.). Twenty-four hours after induction with 1m M isopropyl thio-b- D -galactoside, cells of a 5 L suspen- sion were harvested and lysed by sonication. The crude extract was cleared by centrifugation (10 000 g) and separ- ated on Q-Sepharose fast-flow 26/20, gel-filtration on Sephacryl S-100 HR 26/60, and MonoQ HR 5/5 (all Amersham Biosciences), according t o a procedure described previously [20]. For metal affinity chromatography, Talon resin ( BD Bioscience) was utilized. A nalysis of proteins was performed with SDS/PAGE (12.5%) under denaturating conditions, a nd gels were silver stained as d escribed previously [25]. For Western blot analysis, proteins were blotted onto nitrocellulose filters and detected with anti-6·His primary antibody, alkaline phosphatase-labeled secondary antibody and chemoluminescent substrate (CDP-Star; Roche) . MJE activity was monitored according to a previously published protocol [20]. Results Isolation of MJE cDNA We previously described an MJE, which is the only or at least the predominant protein with MJE a ctivity i dentified in tomato cell c ulture s. On the basis of the p artia l amino a cid sequences obtained from the purified MJE [20], two degenerate primers were developed. T his method has been proven successful in several r eports [26,27] and s hould lead to the identification of the encoding gene rather than orthologous genes. Owing to the similarity of the peptide fragments with sequences of known proteins, the sequence PF18b (DMAASGINPK) was utilized to design a sense primer, and sequence PF23 (RVYVVCDKD) w as used for the generation o f an antisense prime r. Using tomato cDNA as a template, a 498 bp fragment was amplified by PCR. Sequencing revealed a DNA s tretch that e ncodes a peptide with sequence similarity to a/b-hydrolase fold proteins (data not shown), some of w hich have been p reviously aligned to the internal fragments of purified MJE [20]. To obtain the full-length cDNA by RACE, t wo seq u ence-specific prime rs were synthesized, generating overlapping fragments after 5¢-and3¢-RACE, respectively. Sequencing and annealing of the 5¢ and the 3¢ sequence revealed an ORF of 789 bp, encoding a 262 amino acid protein (Fig. 1). All four p eptides from the purified tomato protein could be identified in the deduced amino acid sequence, which substantiates that the cloned cDNA encodes the purified protein. The calculated molecular m ass o f the encoded protein is 29 524.93 Da and the pI ¼ 5.52. The calculated mass o f the encoded protein corresponds well with the m olecular mass of  28 000 determined by SDS/PAGE for the purified plant protein. Comparison of the N-termini showed that the protein originally purified from tomato lac ked two amino acids: Met and Glu. It could not be concluded if this was a result of degradation of the protein during the purification process or whether t he protein was modified in vivo .Nopeptidesignal for subcellular targeting could b e i dentified. Sequence alignment Sequence analysis a nd alignment with known proteins f rom GenBank showed a high similarity of MJE to ethylene- induced esterase from Citrus sinensis (47% identity, 65% positivity) [28], the tobacco salicylic ac id binding protein 2 (SABP2) (47% identity, 65% positivity) [29], the polyneu- ridine aldehyde esterase (PNAE) from Rauvolfia serp entina (44% identity, 65% positivity) [27], and s everal lyases involved in the biosynthesis of cyanogenic compounds in different plant species [36% identity to (S)-hydroxynitrile lyase (HNL) from Hevea brasiliensis [30], 33% to (S)- Fig. 1. Tomato methyl jasmonate esterase cDNA and deduced protein sequence. Pe ptides determined by seq ue ncing of the p urified protein are boxed and the names of the peptides are indicated in italic letters above. Nucleic acids o f the ORF are shown in ca pital letters, while 5¢ and 3¢ untranslated regions are in lowercaseletters.Theputative amino acid r esidues of t he catalytical t riad of a/b-hydrolase fold pro- teins are marked with an asterisk. 2978 C. Stuhlfelder et al.(Eur. J. Biochem. 271) Ó FEBS 2004 acetone-cyanohydrin lyase from Manihot esculentum][31]. In addition, several putative proteins from the Arabidopsis genome exhibited high sequence similarity to MJE. For sequence alignment and analysis, only known or at least partially characterized proteins were included (Fig. 2). As all the aligned proteins belong to the extremely divergent family of a/b-hydrolase fold proteins, i t c ould b e assumed that MJE is a member of this protein family. As further support of this assumption, MJE shows the highly con- served amino acid residues forming the catalytic triad [32] – nucleophile, acid, and a his tidine – represented b y s erine at position 83, aspartic acid at position 2 11, and histidine at position 240 (Fig. 2). Moreover, it has been shown that MJE could be irreversibly in hibited by phenylmethanesulfo- nyl fluoride [20], a specific inhibitor o f s erine hydrolases [33]. Bacterial expression and purification of tomato MJE To obtain unequivocal evidence of the identity of the cloned sequence, MJE cDNA was subcloned i nto a bacterial vector for h eterologous expression. As amplification of the coding sequence with a forward primer homologous to the 5¢-end failed, the primer s equence had to be modified f or enhanced binding and amplification. Primer design was carried out using t he Vector NTI Software (Informax). Thereby, a modified N-terminus of the encoded protein was created in which Glu and Lys in positions 2 and 3, respectively, were replaced with a single Gln. At the C -terminus a 6 ·His extension was added to simplify subsequent purification of the protein. Crude extracts of E. coli M15 cells harbouring the pQE-MJE plasmid showed MJE-esterase activity (1.64 pkatÆmg )1 ) after isopropyl thio-b- D -galactoside induction, while wild- type M15 cells did not show any MJE activity. This v alue was comparable with the activity found in crude extracts from tomato cell suspension culture in previous experiments (1.77 pkatÆmg )1 ) [20], but was much less t han expected for an enzyme from heterologous bacterial expression of the cDNA. For visualization of proteins, bacterial extracts were subjected to SDS/PAGE. In a comparis on of MJE-produ- cing E. coli with wild-type cells, no p rominent protein w ith the approximate size of MJE (29 kDa) could be detected Fig. 2. Multiple sequence alignment. Alignment of methyl jasmonate esterase (MJE) with a/b-hydrolase fold proteins fr om different plant species. EIE, ethylene-induced este rase from Citr us sinensis (GenBank accession number AAK58599); SABP2, salicylic acid-binding protein from Nic- otiana tabac um (AY485932); PNAE, polyneuridine aldehyde esterase from Ra uv olfia serpentina (AAF22288); Pir7b, d efense-related ri ce ge ne from Oryza sativa (CAA84024); H NL, (S)-hydroxynitrile lyase f rom Hevea brasiliensis (P52704). Ó FEBS 2004 Methyl jasmonate esterase from tomato (Eur. J. Biochem. 271) 2979 after Coomassie Blue staining (data not shown). This observation suggests that M JE is not abundantly produced or that it is not stable in E. coli. Because of the low abundance in E. coli, a four-step purification protocol was employed to purify the enzyme (Table 1). Catalytically active enzyme was obtained after anion exchange on Q-Sepharose, gel filtration with Sephacryl S-100, further anion-exchange chromatography on MonoQ, and finally separation on immobilized metal affinity chromatography (Talon resin). Starting from a 5 L bacterial s uspension culture, MJE(His) 6 could be enriched 203-fold, resulting in 52 lg of protein. Figure 3 A s hows the silver-s tained poly- acrylamide gel from the purified fraction. Although some impurities are visible, we assume that solely the MJE is responsible for MeJA-cleaving activity, as wild-type E. coli is not capable of c leaving M eJA. A n a ntibody s pecific f or hexa-histidine epitopes was used in an immunoblot experi- ment to con firm the p resence and the size of the recombin- ant protein. As shown in Fig. 3 B, extracts from E. coli harbouring pQE-MJE showed a b and o f  29 kDa repre- senting MJE(His) 6 , while wild-type bacteria did not. Southern blot analysis For Southern blot analysis, genomic DNA from cell suspension cultures or greenhouse-grown tomato plants (L. esculentum cv Moneymaker) was digested with BamHI , EcoRI, or HindIII restriction enzymes and probed with a full-length cDN A of MJE at high stringency. It should be noted that the MJE-coding sequence has a recognition site for EcoRI at position 204 and therefore should show at least two b ands in a Southern blot a nalysis. In addition to this , the probe hybridized with multiple bands (Fig. 4). In the case of HNL from Cassava – which shows high similarity to MJE – several gene copies were reported [ 34]. It could not be concluded from our data whether tomato contains several homologous genes, if some signals are a result of probe hybridization with pseudogenes, or if multiple bands occur o wing to the presence of recognition sites for the utilized restriction enzymes within introns (as assumed f or the HNL from Hevea) [30]. However, during the purification of MJE there was n o evidence for the expression of isoenzymes, although, if present, they might have different catalytic properties. Northern blot analysis and induction of MJE expression in cell cultures Northern blot analysis was u sed to determine MJE transcript levels in different plant organs. Therefore, total RNA from roots, leaves, stems, and flowers was probed with full-length MJE cDNA. Transcripts of  1kbwere present i n a ll plant tissues and significant v ariations in their amounts could be d etected (Fig. 5A). High levels of MJE mRNA could be found in roots a nd flowers, while low-to- moderate amounts w ere present in the leaves and stems of tomato plants. The RNA levels correspond well with the MJE activity found in different plant organs, showing that high enzymatic a ctivity c orrelates w ith a high transcript level (Fig. 5 B). Table 1. Purification of recombinant methyl jasmonate e sterase. Purification step Total protein (mg) Total activity (pkat) Specific activity (pkat) Purification (fold) Recovery (%) Crude extract 2044 3352 1.64 1 100 Q-Sepharose Fast Flow 712 1317 1.85 1.13 27.7 Sephacryl HR 26/60 67.5 652.7 9.67 5.89 19.4 Mono Q HR 5/5 1.9 100.6 54.39 33.16 3.0 Talon resin 0.052 24.1 464.3 283.11 0.8 Fig. 3. SDS/PAGE and Western blot a nalysis of me thyl jasmonate esterase (MJE) purified from rec ombinant bacteria. (A) Silver-stained proteins after separation b y SDS/PAGE (12.5% gel). M, marker proteins (sizes are indicated on t he left); lane 1, purified f raction after t he last pu rification step. MJE is indica- tedbyanarrow.(B)Westernblotofcrude bacterial extracts after transfer to n itrocellu- lose. An antibody specific for 6·His modified proteins was used. L ane M, pr otein standard; lane 1, Escherichia coli harbouring expre ssion plasmid p QE70-MJE; lane 2, E. coli without the expression plasmid. MJE is indicated with the arrow. 2980 C. Stuhlfelder et al.(Eur. J. Biochem. 271) Ó FEBS 2004 In order to evaluate whether transcript levels could be regulated b y external stimuli, cell s uspension c ultures were treated with MeJA, m ethyl salicylate (MeSA), or chitosan, and mRNA levels were determined in a time-dependent manner. While basal levels of m RNA w ere h igh in untreated cultures, t he levels decreased 1 h after MeJA treatment and returned to basal levels  8 h postinduction (Fig. 5C). A similar time c ourse could b e observed a fter treatment w ith the elicitor chitosan, although with a slightly delayed response. No changes in mRNA levels were observed after treatment with MeSA. Discussion Sequence alignment of the MJE revealed high similarity to a/b-hydrolase fold proteins of different origin a nd with diverse properties. Notably, all plant proteins of known function with high sequence s imilarity to MJE appear to be involved in the defense response and/or secondary metabo- lism. Among the related proteins is the ethylene-induced esterase from C. sinensis [28], a recently discovered SABP2 from tobacco [29], the Pir7b protein from Oryza sativa, hydroxynitrile lyases of different origins and the PNAE from the medicinal plant R. serpentina. Nevertheless, the substrate acceptance and enzymatic activity of those enzymes, if known, is highly diverse. With the increasing number o f s pecified plant enzymes of this g roup of the a/b- hydrolase f amily, i t might be assumed t hat they have arisen from a common ancestral gene and the descendants partially occupied species-specific niches in secondary metabolism. Similar scenarios have been described for plant O-methyl transferases [35,36] or dioxygenases [37]. It should Fig. 4. Southern blot analysis of total genomic DNA from tomato plants. Fifty micrograms of DNA was digested overnight with restriction enzymes and separated on a 1% agarose gel. M, size standard; lane 1, Bam HI-digested DNA; lane 2, EcoRI digest; l ane 3, HindIII dig e st. Fig. 5. Levels of methyl jasmonate esterase (MJE) t ranscripts in dif- ferent tissues and after induction. (A)NorthernblotoftotalRNAfrom different tissues hybridized with the full-length cDNA probe. RNA was isolated from roots, leafs, stems and flowers. The lower panel shows 17S rRNA as a loading control (after methylene-blue staining); the results shown were consistent in three different experiments. (B) Specific activity of MJE in different plant organs. (C) Time course of MJE t ranscript a b undance after treatment with methyl jasmonate (MeJA), methyl salicylate (MeSA), an d chitosan. Lo ading contr ols show the ethidium b ro mide-stained gel p rior to transfer. Ó FEBS 2004 Methyl jasmonate esterase from tomato (Eur. J. Biochem. 271) 2981 be noted that in the Arabidopsis genome at least 20 genes with homologies to HNL and PNAE could b e found. It is unlikely that they have similar properties to the a/b- hydrolase fold proteins m entioned above, as they occur only in distinct plan t families o r even s pecies and t heir substrates are not present in Arabidopsis. O n the other hand, MJE activity could be found in several plant systems. From the 1 8 plant cell suspension cultures of taxonomically distant species tested to date, virtually all exhibited MJE activity [20]. O ne might s peculate that the M JEs of different species are encoded by orthologous genes. Whenever analysed, plant cells and tissues contain JA and its methyl ester, MeJA, side by side. In most reports, authors have made little effort to distinguish between the biological activity of the two compounds and usually only JA content is measured. If values of bo th JA and MeJA are published, the ratios depend on the plant species and the tissue a nalysed a nd vary from 3 : 2 (JA/MeJA) i n Arabid- opsis leaves [4] to a bout 10 : 1 in tomato flowers [38]. For tomato cell suspension cultures, a JA/MeJA ratio of almost 1 : 1 was found [39]. To d ate there is no way of distinguishing between the function of the two compound s on a physiological basis as they can be rapidly converted from one into the other. For the Arabidopsis jar1 mutant, i t has been shown that the role of JAR1 is to modify JA via adenylation of the carboxyl group [12]. As MeJA is not accepted as a substrate in this putative essential step involved in JA signalling, it is probable t hat MeJA has to be hydrolyzed by MJE in order to become metabolically activated. In this case, MeJA might represent a pool of inert JA conjugates or plays a role as signalling molecule between individual cells [19]. The inverse activity of JMT and MJE suggests that both enzymes should be spatially separated. MJE transcript levels were analyzed by Northern hybrid- ization a nd revealed a high, constitutive expression in roots and flowers and low RNA levels in leaves a nd stems, suggesting that enzyme activity could b e found in all tissues. These data were supported by activity assays of MJE in different plant organs, which signify that high enzymatic activity is contingent on high transcript levels. Interestingly, undifferentiated tomato cell suspension cultures accumulate both JA and MeJA at almost e qual levels while displaying high MJE activity, suggesting that individual cells may contain both JMT and MJE activity. However, it is unlikely that a cell forms MeJA under the expenditure of energy and degrades it immediately. Intra- cellular s eparation of MeJA synthesis and hydrolysis would be one way to avoid a treadmill situation, yet analysis of tomato MJE and Arabidopsis JMT reveals no evidence for subcellular targeting of the enzymes and, thus, both enzymes should reside in the cytosol. Alternatively, sub- strates may be presented for the enzymes in a highly regulated manner. To this end, it would be an a ppealing s cenario that a cell could distinguish between endogenously formed and exo- genous MeJA. Endogenously formed MeJA might be exported and not hydrolysed, while MeJA coming from outside the cell m ay be recognized as an alar m signal that – after hydrolysis to JA – functions as intracellular defense signal. In fact, a similar situation occurs in mammals. Stimulated neutrophils may synthesize (within minutes) leukotrienes, which are exported into the extracellular environment where they act as autocrine and paracrine signals. Ho wever, neu trophils are also capable of taking up leukotrienes for intracellular catabolism, thereby locally restricting and terminating the signal. Transcript levels were also monitored after stimulation of tomato cell cultures with exogenous MeJA or the elicitor chitosan, which induces intracellular synthesis of jasmonates in tomato [40]. Constitutively high MJE accumulation transiently declined within 2 and 3 h after the treatments, respectively. Decreasing transcript levels may not immediately affect enzyme activity and thus exogenous MeJA can still be hydrolysed to JA for some time. H owever, downregulation of MJE by exogenou s MeJA may limit MeJA hydrolysis and J A signalling when cells are exposed to elicitors/jasmonates over longer time- periods. Interestingly, basal transcript levels of JMT in Arabidopsis leaves are low, and stimulation by exogenous MeJA transiently increases JMT formation. Overexpres- sion of JMT cDNA has been shown to increase the synthesis o f MeJA [4], which, in turn, may leave producer cells and function as an intercellular signal. As MJE activity has been detected in all plant tissues examined so far, MJE-harbouring cells may trap the volatile and highly diffusible MeJA entering cells from the outside by hydrolysis to JA anions inside the cells. Inc reasing JA levels might t hen elicit specific responses. In the future, generation and careful analysis of transgenic plants that either constantly accumulate MJE or that are devoid of MJE will help to solve the question of whether or not MeJA is a paracrine or even a long-distance signal. Acknowledgements This work was supported by t he Sonderforschungsbereich (SFB) 567. The a u thors thank Susanne Michel for performin g DNA sequen cing. References 1. Creelman, R.A. & Mullet, J.E. (1997) Biosynthesis a nd action of jasmonates in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 355–381. 2. Mueller, M. (1997) Enzymes i nvolved in j asmonic acid b iosyn- thesis. Phys iol. Plant. 10 0, 653–663. 3. Schaller, F. (2001) Enzymes of the biosynthesis of octadecanoid- derived s ignalling molecules. J. Exp. Bot. 52 , 11–23. 4. Seo, H .S., S o ng, J .T. , C heon g, J.J., L ee , Y .H., Lee, Y.W., H wang, I., L ee, J.S. & Choi, Y .D. (2001) Jasmonic ac id carboxyl methyl- transferase: a key enzyme for jasmonate-regulated plant r esponses. Proc.NatlAcad.Sci.USA98, 4 788–4793. 5. Sembdner, G. & Parthier, B. (1993) The biochemistry and the physiological and molecular actions of jasmonates. Annu. Rev. Plant Physiol. Plant Mol. Biol. 44, 569–568. 6. Wasternack, C . & Parthier , B. ( 1997) Jasmonate-signalled p lant gene expression. Tre nds Pl ant Sci. 2, 302–307. 7. Kramell, R., Miersch, O., Atzorn, R., Parthier, B. & Wasternack, C. (2000) Octadecanoid-derived alteration of gene expression and the Ôoxylipin signatureÕ in stressed barley leaves. Implications for different signali ng pathways. Plant Physiol. 123 , 177–188. 8. Berger, S. (2002) Jasmonate-related mutants of Arabidopsis as tools f or studying s tress signaling. Planta 214, 497 –504. 9. McConn,M.&Browse,J.(1996)Thecriticalrequirementfor linolenic acid is pollen development, not photosynthesis, in an Arabidopsis mutant. Pl ant Cell 8, 403 –416. 2982 C. Stuhlfelder et al.(Eur. J. Biochem. 271) Ó FEBS 2004 10. Stintzi, A., Weber, H., Reymond, P., Browse, J. & Farmer, E.E. (2001) Plant defense in the absence of jasmonic acid: the role of cyclopentenones. Proc. Natl Acad. Sci. USA 98, 12837– 12842. 11. Staswick, P.E., Su, W. & Howell, S. H. (1992) Me thyl jas monate inhibition of root growth and induction of a leaf protein are decreased i n an Arabidopsis thaliana mutant. Proc. Natl A cad. Sc i. USA 89, 6837–6840. 12. Staswick, P.E., Tiryaki, I. & Rowe, M.L. (2002) Jasmonate response locus J AR1 a nd several related Arabidopsis genes enc ode enzymes of the firefly luciferase superfamily that show activity on jasmonic, salicylic, and indole-3-acetic acids in an a ssay for ade- nylation. Plant Cell 14, 1405–1415. 13. Reymond,P.,Weber,H.,Damond,M.&Farmer,E.E.(2000) Differential gene exp ression in respon se to mechanical wounding and i nsect f eeding in Arabidopsis. Plant Cell 12, 707–720. 14. Howe, G.A. (2001) Cyclopentenone signals for plant defense: remodeling the jasmonic acid response. Proc.NatlAcad.Sci.USA 98, 12317–12319. 15. Stratmann, J.W. (2003) Long distance run in the wound response – j asmonic acid is p ulling ahead. Trends Plant Sc i. 8, 247 –250. 16. Li, L., Li, C., Lee, G.I. & Howe, G.A. (2002) Distinct roles for jasmonate synthesis a nd ac tion in th e s ystemic w ound res ponse of tomato. Proc. Natl Acad. Sci. USA 99, 6416–6421. 17. Meyer, R., Rautenbach, G.F. & D ubery, I.A. (2003) Identification and q uantification of methyl jasm onate in leaf v olatiles o f Arabi- dopsis thaliana usin g s olid-phase m i croextraction i n combination with gas chromatography and mass spectrometry. Phytochem. Anal. 14, 155–159. 18. Farmer, E.E. & Ryan, C.A. (1990) Interplant communication: airborne methyl jasmonate induces synthesis of proteinase in- hibitors in plant l eaves. Proc . Natl Acad. Sci. USA 87, 7713–7716. 19. Weber, H . (2002) Fatty acid-derived signals in plants. Trends Plant Sci. 7, 217–224. 20. Stuhlfelder, C., Lottspeich, F. & Mueller, M .J. (2002) Purification and partial amino acid sequences of an esterase from tomato. Phytochemistry 60 , 233–240. 21. Linsmaier, E.M. & Skoog, F. (1965) Organic growth factor requirements of tobacco tissue cultures. Physiol. Plant. 18, 100– 127. 22. Ausubel, F.M., Brent, R ., King ston, R.E., Moore, D.D., S mith, J.A.,Seidman,J.G.&Struhl,K.E.(1994)Current Protocols in Molecul ar Biology . Greene a nd. Wiley, New York. 23. Rogers, S.O. & B endich, A.J. ( 1985) Extraction o f D NA from milligram amounts of fr esh, herbarium a nd mummified plant tis- sues. Plant Mol. Biol. 5, 69–76. 24. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning – a Laboratory Manual, 2nd ed n. Co ld Spring H arbor Laboratory Press, Cold Spring Harbor, New Y o rk. 25. Blum, H., Beier, H. & Gross, H.J. (1987) Improved silver staining of prote ins, RNA, DNA in polyacrylamide gels. Electrophoresis 8, 93–99. 26. Warzecha, H., Gerasimenko, I., Kutchan, T.M. & Stockigt, J. (2000) Molecular cloning and functional bacte rial expression of a plant glucosidase specifically involved in alkaloid biosynthesis. Phytochemistry 54 , 657–666. 27. Dogru,E.,Warzecha,H.,Seibel,F.,Haebel,S.,Lottspeich,F.& Stockigt, J. (2000) The gene encoding polyneuridine aldehyde esterase of monoterpenoid ind ole a lkaloid biosynthesis in p lan ts i s an orth olog of the alpha/beta h ydrolase super family [In Process Citation]. Eur. J. Biochem. 267, 1397–1406. 28. Zhong, G.Y., Goren, R., Riov, J., Sisler, E.C. & Holland, D. (2001) Characterization of an ethylene-induced este rase gene iso- lated from Citrus sinensis by competitive hybridization. Physiol. Plant. 113, 2 67–274. 29. Kumar, D. & Klessig, D.F. (2003) High-affinity salicylic acid- binding protein 2 i s req uired for p lan t innate imm unit y and has salicylic acid-stimulated lipase activity. Proc. Natl Acad. Sci. USA 100, 1 6101–16106. 30. Hasslacher, M., Schall, M., Hayn, M., Griengl, H., Kohlwein, S.D. & S chwab, H. (1996) Molecular cloning of the full-length cDNA o f (S)-hydroxynitrile lyase f rom Hevea brasi lien sis . Func- tional expre ssion in Escherichia c oli and Saccharomyces cerevisiae, and identification of an active site residue. J. Biol. Chem. 271, 5884–5891. 31. Hughes, J., Carv alho, F.J. & Hughes, M .A. (1994) Purification, characterization, and cloning of alpha-hydroxynitrile lyase f rom cassava (Manihot esculenta Crantz). Arch. Biochem. Biophys. 311, 496–502. 32. Heikinheimo, P., Goldman, A., Jeffries, C. & Ollis, D.L. (1999) Of barn owls and bankers: a lush variety of alpha/beta hydrolases. Structure F old Des. 7, R141–R146. 33. Gold, A.M. (1965) Sulfonyl fluor ides as inhibitors of esterases. Identification of serine as the site of sulfonation in phenyl- methanesulfonyl alpha-chymotryp sin. Biochemistry 4, 897–901. 34. Hughes, J ., Keresztessy, Z ., Brown, K., Su handono , S . & Hughes, M.A. (1998) Genomic organization and structure of alpha- hydroxynitrile lyase in cassava (Manihot e sculenta Crantz). Arch. Biochem. Biop hys. 356, 107–116. 35. Ibrahim, R.K., Bruneau, A. & Bantignies, B. (1998) Plant O- methyltransferases: m olecular analysis, common signature and classification. Plant Mol. Biol. 36, 1–10. 36. Pichersky, E. & Gang, D.R. (2000) Ge netics and b iochemistry o f secondary metabolites in plants: an evolutionary perspective. Trends Pla nt Sci. 5, 439–445. 37. Kliebenstein, D.J., Lambrix, V.M., Reichelt, M., Gershen zon, J. & Mitchell-Olds, T. (2001) Gene dupl ication in the diversification of secondary metabolism. Tandem 2-oxoglutarate-dependent dioxy- genases control glucosinolate biosynthesis in arabidopsis. Plant Cell 13 , 681–693. 38. Hause, B., Stenzel, I., Miersch, O., Maucher, H., Kramell, R., Ziegler, J. & Wasternack, C. (2000) Tissue-specific o xylipin sig- nature of tomato flowers: allen e oxide cyclase is highly expressed in dist inct flower o rgans a nd vascular b undles. Pla nt J . 24, 113– 126. 39. Gundlach, H., Mull er, M .J., Ku tcha n, T.M. & Z enk, M.H. (1992) Jasmonic acid is a signal transducer i n elicitor-induced plant cell cultures. Proc.NatlAcad.Sci.USA89, 2 389–2393. 40. Doares, S.H., Syrovets, T., Weiler, E.W. & Ryan, C.A. (19 95) Oligogalacturonides a nd chitosan activate plant defensive genes through the o ctadecanoid pathway. Proc. Natl Acad. Sc i. USA 92, 4095–4098. Ó FEBS 2004 Methyl jasmonate esterase from tomato (Eur. J. Biochem. 271) 2983 . Cloning and expression of a tomato cDNA encoding a methyl jasmonate cleaving esterase Christiane Stuhlfelder, Martin J. Mueller and Heribert Warzecha Lehrstuhl. was cloned by another RT-PCR with primers f ullMJEforMQ (GCA TGCAGGGTGATAAAAA TCACTTTGTA) and fullMJErev (AAGGATCCATAA TATTTTTGCGAA ATC), adding rest rictio