Variations on a theme: synthesis and modification of plant benzoic acids Mary C Wildermuth

Plant benzoic acids (BAs) are critical regulators of a plant’s species of 30 different families and function as pollinator interaction with its environment. In addition, innumerable plant- attractants [3]. In addition, MeSA is involved in tri- derived pharmacological agents contain benzoyl moieties. trophic interactions; MeSA is emitted in response to Despite the prevalence and import of plant BAs, their herbivore damage (e.g. that caused by aphids or mites), biosynthetic pathways are not well-defined. Mounting attracting the predator of the herbivore [3,4]. As shown evidence suggests that BAs are synthesized both directly from in Figure 1, BAs are also incorporated into numerous plant shikimate/chorismate and from phenylalanine in plants; secondary metabolites that have established roles in however, few genes in these pathways have been identified. plant–herbivore or plant–pathogen interactions, such as Exciting progress has been made in elucidating genes that BA glucosinolate esters (e.g. in A. thaliana [5]), salicin (the modify BAs via methylation, glucosylation, or activation with major phenolic glycoside in willow [6]), xanthones (e.g. in Coenzyme A. As these modifications alter the stability, Hypericum androsaemum [7]), cocaine (in Erythroxylum coca solubility, and activity of the BAs, they impact the functional [8]), and taxol (in Taxus cuspidate [9]). Many of these BA roles of these molecules. The combination of multiple BA and benzoyl compounds are also important pharmacolo- biosynthetic routes with a variety of chemical modifications gical agents; for example, salicylate is the active ingre- probably facilitates precise temporal and spatial control over dient in aspirin, taxol is a potent anti-cancer drug, and active forms, as well as the channeling of intermediates to cocaine is a local analgesic. particular benzoate products. Despite the importance and prevalence of plant benzo- Addresses ates, the biosynthesis of BAs is not well defined. It is University of California, Department of Plant and Microbial Biology, 221 likely that several BA biosynthetic pathways exist in a Koshland Hall, Berkeley, California 94720-3102, USA given species, providing fine control over spatial and Corresponding author: Wildermuth, Mary C temporal synthesis and channeling intermediates to par- ([email protected]) ticular benzoate products. A now classic example is the synthesis of the active C5-unit isopentenyl diphosphate (IPP), the building block of all isoprenoids. A cytosolic Current Opinion in Plant Biology 2006, 9:288–296 mevalonate pathway synthesizes IPP, which is used in the This review comes from a themed issue on biosynthesis of sterols, sesquiterpenes, and triterpenoids. Physiology and metabolism By contrast, the plastidic, non-mevalonate 1-deoxy-D- Edited by Eran Pichersky and Krishna Niyogi xylulose-5-phosphate pathway (similar to that used by bacteria) is involved in the biosynthesis of plastidic iso- Available online 4th April 2006 prenoids, including carotenoids, phytol (the side chain of 1369-5266/$ – see front matter chorophylls), plastoquinone-9, and isoprene [10]. Modi- # 2006 Elsevier Ltd. All rights reserved. fications of BA (and SA) that influence their volatility, DOI 10.1016/j.pbi.2006.03.006 membrane permeability, solubility, and activity are also crucial to their transport and function. Thus, understand- ing the biosynthesis and modification of BAs in plants is a crucial first step to understanding the regulation and Introduction function of these important molecules. Plant benzoic acids (BAs) and their derivatives are com- mon and widespread mediators of plant responses to Biosynthesis of BAs in plants biotic and abiotic stress. For example, salicylic acid Plant BAs have been reported to originate either from (SA, 2-hydroxybenzoic acid) is a key signaling molecule phenylalanine (Phe) or directly from a shikimate-derived that mediates plant defense against a variety of pathogens product such as isochorismate, in which the carboxyl in a number of species, including tobacco and Arabidopsis carbon of shikimate is retained in the BA (Figure 2). thaliana. Its accumulation is required for the establish- By contrast, the carboxyl carbon of BAs that are synthe- ment of local and systemic required resistance (SAR) sized from Phe originates from the b-carbon of the Phe responses [1]. SA is also synthesized in response to side chain. Predominant early studies supported BA oxidative stressors, such as ozone, and is in part respon- synthesis from Phe, and thus most work has focused on sible for induced plant protective responses [2]. The Phe-derived pathways. As is readily apparent from volatiles methyl benzoate (MeBA) and methyl salicylate Figure 2, few of the plant genes that encode the enzymes (MeSA) are present in the floral scents of more than 100 of these pathways have been cloned; thus, much work is

Current Opinion in Plant Biology 2006, 9:288–296 www.sciencedirect.com Benzoic acids Wildermuth 289

Figure 1

Representative plant BAs and secondary compounds that incorporate or are biosynthesized from a benzoyl moiety (shown in blue).

needed to validate, define, and refine the biosynthetic [13] and 3-hydroxybenzoate (3HBA), which is incorpo- pathways for BAs. rated into amarogentin in Swertia chirata ([14]; Figure 1). In A. thaliana, the bulk of the SA that is produced in Biosynthesis of BAs directly from shikimate/chorismate response to pathogens is synthesized from isochorismate Bacteria provided the first evidence for BA biosynthesis and not from Phe, as confirmed by the lack of induced SA directly from shikimate/chorismate. For example, SA and accumulation in isochorismate synthase (ics1) mutants 2,3-dihydroxybenzoic acid (DHBA) are synthesized from [15]. This study provided the first genetic evidence for chorismate via isochorismate, and are precursors for side- a shikimate/chorismate direct pathway for the synthesis of rophores such as pyochelin in Pseudomonas aeruginosa [11] BAs in plants. Additional studies supporting, but not and enterobactin in Escherichia coli [12]. In plants, evi- providing direct evidence for, this pathway include: first, dence for the shikimate/chorismate direct pathway of BA the correlation of DHBA accumulation with ICS activity synthesis includes two 13C NMR studies in which the in elicted Catharanthus roseus cell cultures [16,17], and 13C-labeling of the carboxyl carbon is consistent with second, the incorporation of radiolabeled 3HBA (but not synthesis directly from shikimate/chorismate for gallic of Phe, trans-cinnamic acid [t-CA], or BA) into the ben- acid (3,4,5-trihydroxybenzoic acid) in the tree Rhus typina zoyl moiety of xanthone coupled with lack of induction of www.sciencedirect.com Current Opinion in Plant Biology 2006, 9:288–296 290 Physiology and metabolism

Figure 2

Biosynthesis of BAs in plants by the direct shikimate/chorismate pathway and via phenylalanine. The carboxyl carbon of shikimate is labeled (13C) as is the b-carbon of phenylalanine (*). The plant enzymes involved in BA biosynthesis for which genes have been cloned are indicated, as are the chorismate-utilizing enzymes anthranilate synthase (AS) and aminodeoxychorismate synthase (ADCS). Pathways from trans-cinnamic acid alone are shown for simplicity; similar pathways from precursors in which either hydroxyl or methoxyl functionalities decorate the benzene ring (e.g. p-coumaric acid) have been reported. In addition, possible glucosylated precursors are not shown. C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate:CoA ligase; CM, chorismate mutase; ICS, isochorismate synthase; PAL, phenylalanine ammonia lyase.

phenylalanine ammonia lyase (PAL) in elicited cells of confirmation of the non-oxidative route and a require- Centaurium erythraea [18]. The term shikimate/chorismate ment for CoA and the enzyme 4-coumarate:CoA ligase direct pathway is used to allow for the possibility that (4CL) in support of the CoA-dependent b-oxidation other intermediates (such as shikimate-3-phosphate or pathway. Support for the more recently proposed third anthranilate), in addition to isochorismate, might be route (Figure 2c) includes a requirement for CoA and utilized (Figure 2). Interconversion of the BA products 4CL coupled with detection of the benzaldehyde inter- (e.g. SA, DHBA, 3HBA, and BA) could then occur using mediate. There is an additional possibility that BA (or free or activated BAs (e.g. benzoyl-Coenzyme A [CoA] or BA-CoA) could be produced from Phe via phenylpyru- BA glucose conjugate). vate in a manner similar to that used by bacteria to catabolize Phe [19]. As limited conclusive evidence exists Biosynthesis of BAs from phenylalanine for the reported plant pathways from Phe to BA, alter- BA biosynthesis from Phe has been shown to proceed native possibilities such as this should be explored. from t-CA, which is produced from Phe via PAL. The conversion of t-CA to BA (or benzoyl-CoA) requires the The CoA-dependent b-oxidation pathway mirrors fatty cleavage of two carbons from the C3 side-chain of the acid b-oxidation ([20]; Figure 2a). In this pathway, t-CA is precursor. As shown in Figure 2, three possible routes activated by formation of the cinnamoyl-CoA ester, which have been reported for this production of BA from t-CA: is then hydrated to form 3-hydroxy-3-phenylpropionyl- CoA-dependent b-oxidation (Figure 2a), a CoA-indepen- CoA. This hydroxyl group is oxidized to a ketone, and the dent non-oxidative route (Figure 2b), and a CoA-depen- b-keto thioester is cleaved by a reverse Claisen conden- dent, non-oxidative pathway (Figure 2c). To distinguish sation to form benzoyl-CoA. 4CL catalyzes the activation between these pathways, researchers have used the iden- of t-CA but the subsequent enzymes have not been tification of the intermediate benzaldehyde as identified. In support of this pathway, radiolabeled Phe

Current Opinion in Plant Biology 2006, 9:288–296 www.sciencedirect.com Benzoic acids Wildermuth 291

was not incorporated into benzaldeyde in tobacco structurally related molecules as a substrate with varied mosaic virus (TMV)-inoculated tobacco (Nicotiana taba- affinity or catalytic efficiency. cum) leaf tissue [21], cucumber (Cucumis sativus) seed- lings, or smoke-treated coyote tobacco (Nicotiana Methylation attenuata) plants [22], whereas radiolabeled 3-hydroxy- Benzoid carboxyl methyltransferases catalyze the transfer 3-phenylpropanoic acid was detected as an intermediate of the methyl group of S-adenosyl-L-methionine (SAM) [22]. Similarly, in Lithospermum erythrorhizon cell free to the free carboxyl of BAs, forming volatile methyl esters extracts, p-hydroxybenzoic acid (4HBA) was synthesized such as MeBA and MeSA. These methyl esters are in a CoA-dependent manner requiring 4CL activity, and important constituents of many plant floral scents and p-hydroxybenzaldehyde was not detected as an inter- the first benzoid carboxyl methyltransferases were iden- mediate [23]. tified and isolated from floral tissue [3]. More recent work has also identified benzoid carboxyl methyltrans- The CoA-independent, non-b-oxidative pathway (or ferases that have a role in defense [29,30]. For example, a non-oxidative pathway [Figure 2b]) involves the hydra- reverse genomics approach was used to identify a putative tion of the free acid, side-chain degradation via a reverse Arabidopsis SA carboxyl methyltransferase (AtBSMT, aldol condensation, and NAD(P)-dependent oxidation of At3g11480) that has a role in defense from among the the intermediate aldehyde. Supporting evidence comes 24 predicted Arabidopsis SABATH family members. from studies examining 4HBA biosynthesis from p-cou- Expression of this gene correlated with induced MeSA marate in elicitor-treated carrot (Daucus carota) cell cul- emission in response to a variety of treatments, including tures and extracts [24], potato tuber cell free extracts [25], moth herbivory and treatment with alamethicin, a vol- and L. erythrorhizon cell extracts [26]; CoA was required tage-gated ion-channel-forming antibiotic from the fun- and p-hydroxybenzaldehyde was detected. However, gus Trichoderma viride [29]. This induction of MeSA there has been no subsequent work to provide detailed emission in Arabidopsis is similar to that observed in confirmation of these findings. By contrast, a later in vitro soybean, in which emitted MeSA has been shown to study with L. erythrorhizon supports the b-oxidation path- attract predators of the aphid herbivore [4]. In addition, way from p-coumaroyl-CoA [23] as the major 4HBA highly localized induction of AtBSMT around the site of biosynthetic pathway; this difference in results is largely feeding by piercing-sucking thrips suggests that AtBSMT attributable to modifications of in vitro assay conditions. might play additional roles in defense [29]. Similar to (In vitro assay conditions such as pH, redox status, and other SABATH family members, AtBSMT is able to substrate availability might differ from those in planta; utilize SA, BA, and structurally similar molecules, includ- thus care must be taken in interpreting all in vitro ing anthranilic acid and 3HBA [29]. Further experiments studies.) employing knockout and overexpresser Arabidopsis lines are therefore needed to confirm and dissect the in planta In Pseudomonas fluorescens, a feruloyl-CoA hydratase/lyase function(s) of AtBSMT. was isolated that catalyzes the metabolism of feruloyl- CoA to vanillin and acetyl-CoA via hydration and reverse In tobacco, a potential methyl salicylate esterase aldol cleavage [27,28]. This enzyme was active on a (NtSABP2) catalyzing the reverse reaction (i.e. the number of 4-hydroxycinnamoyl-CoA substrates, but not release of free SA from MeSA) has been identified by on cinnamoyl-CoA [27]. This led to the postulation of a structural and biochemical studies [31]. SABP2- similar CoA-dependent, non-oxidative pathway involved silenced tobacco plants failed to develop SAR after in BA biosynthesis in plants (Figure 2c), and to the inoculation with TMV [32]. As SAR development is subsequent discovery of a cinnamoyl-CoA hydratase/ dependent on SA, this suggests a possible role for MeSA lyase activity in protein extracts of elicited H. androsae- in defense-related SA-dependent signaling and/or trans- mum cells [7]. To date, this enzyme has not been further port [31,32]. Verification of this role requires measure- purified or characterized. ment of MeSA levels in the SABP2-silenced tobacco plants, particularly because neither the methylesterase Modification of BAs activity of NtSABP2 on structurally similar substrates Modifications of small molecules not only alter the critical (such as MeBA, methyl anthranilate, and methyl properties (i.e. volatility, stability, and activity) and thus 3HBA) nor possible hydroxynitrile lyase activity was the functional roles of these molecules, but also allow for reported. (SABP2 shares significant amino acid (45%) precise temporal and spatial control over active forms. In and structural similarity with hydroxynitrile lyases [31].) plants, these modifications can include methylation, gly- cosylation, amino acid conjugation, and activation with Glucosylation CoA (Figure 3). Over the past few years, exciting progress UDP-glucosyltransferases (UGTs) have been identified has been made in identifying the first genes and enzymes that catalyze the transfer of glucose to the carboxyl group that are involved in modifying the small molecule BA in of BA, forming a benzoyl glucose ester (BAE) in tobacco plants. Some of these enzymes can also use SA or other [33] and Arabidopsis [34] in vitro. These enzymes can also www.sciencedirect.com Current Opinion in Plant Biology 2006, 9:288–296 292 Physiology and metabolism

Figure 3

Reactions modifying benzoic acids in plants, shown for BA and SA. Specific enzymes are indicated where known. The letter ‘R’ represents an amino acid side chain. AtBSMT, Arabidopsis thaliana BA/SA carboxyl methyltransferase; CbBZL, Clarkia Breweri benzoate:CoA ligase; CoA, Coenzyme A; NtBAGT, Nicotiana tabacum benzoic acid glucosyl transferase; NtSABP2, N. tabacum methyl salicylate esterase; SAM, S-adenosyl-L-methionine; UDP, uridine diphosphate. glucosylate structurally related molecules such as SA, usage of BAs through the controlled interconversion of forming the salicyloyl glucose ester (SAE) or 2-O-b-D the acids and glucose esters/glucosides. Furthermore, glucoside (SAG) [33,34]. As there are numerous UGTs in glucosylated compounds might also serve as intermedi- plants (e.g. 107 in A. thaliana), evidence of in vitro ates in the biosynthesis of BAs. For example, the cinna- biochemical activity and the correlation of induction with moyl-glucose ester has been proposed as an intermediate the accumulation of the glucose ester or glucoside has in the formation of BAE (and BA) in tobacco [38], and been used to support specific functional roles. To date, no glucosylated o-coumarate has been proposed as a pre- single UGT knockouts deficient in the accumulation of cursor of SAE, SAG, and the salicyloyl-diglucose ester in a BAE, SAE, or SAG has been identified. number of plants, including willow [6]. The discovery of tobacco TOGT1, a UGT that shows high activity with Glucosylation alters the hydrophilicity, stability, subcel- t-CA and o-coumaric acid, supports this possibility [39]. lular localization and bioactivity of the acceptor molecule TOGT1 is induced during the tobacco response to TMV, [35]. For example, the bulk of pathogen-induced SA in parallel with the accumulation of BA, BAE, SA, and accumulates as SAG in the plant vacuole [36]. Because SA–glucose conjugates [38,39]. As the bulk of BA (and free SA is phytotoxic, the accumulation of the glucoside SA) is present as a glucose conjugate [38], identifying and allows the storage of SA in a stable, non-toxic, inactive characterizing the genes and proteins that are responsible form that can be hydrolyzed to active free SA [37]. for the formation, hydrolysis, and transport of these Glucosylation might play a role in the homeostasis and glucose esters and glucosides is essential.

Current Opinion in Plant Biology 2006, 9:288–296 www.sciencedirect.com Benzoic acids Wildermuth 293

Amino acid conjugation not membrane permeable, conversion of benzoyl-CoA to Conjugation of a plant hormone to amino acids can free BA would facilitate the inter- and intracellular trans- activate the hormone (e.g. jasmonic acid [40]) or inacti- port of BAs. Alternatively, benzoyl-CoA could be trans- vate it (e.g. auxin [indole-3-acetic acid [IAA]] [41,42]), ported across membranes by an ABC [54], or other, therefore playing a crucial role in phytohormone regula- transporter. Acyl-CoA thioesterases are an extremely tion and action. The enzymes that catalyze the formation diverse set of enzymes that comprise at least five distinct of these amide-linked phytohormone conjugates are families present in a variety of subcellular locations, GH3-like proteins belonging to the firefly luciferase including the cytosol, endoplasmic reticulum, mitochon- superfamily [43]. In addition, in Arabidopsis, a number drion, and peroxisome. In plants, acyl-CoA thioesterase of IAA–amino acid amidohydrolases that catalyze the activity is quite prevalent [55,56,57], but the first acyl- reverse reaction have been identified through genetic CoA thioesterase (A. thaliana ACH2) was cloned only screens [42]. Though no BA-specific amino acid synthe- recently [55]. Elucidating the function of plant acyl- tases or amidohydrolases have yet been identified, it is CoA thioesterases is an exciting area of research, and it possible that amino acid conjugates play a role in the will be interesting to determine whether a BA-specific temporal and spatial control of free benzoates. Of the 19 acyl-CoA thioesterase plays a role in modulating benzoate putative phytohormone amino acid synthetases in Arabi- biosynthesis, modification, and/or addition to benzoyl- dopsis [43,44], there are currently two candidate SA– containing secondary metabolites. amino acid synthetases: GH3.5 and PBS3. Recombinant GH3.5 (M4122.70, At4g27260) exhibited adenylase activ- Conclusions ity with SA (and IAA) [43]. PBS3 expression is highly Many readers might be surprised that we do not yet know correlated with expression of the pathogen-induced SA precisely how BA and other BAs are synthesized in plants. biosynthetic gene ICS1 (Spearman rank coefficient = 0.83 The observed complexity is probably due to the simpli- using AtGen Pathogen Treatment subset of NASCArray city of these molecules and their evolving chemical and Database [45]), and pbs3 mutants are more susceptible functional elaboration. Evidence is building to support to Pseudomonas syringae ([46]; R Innes, M Wildermuth, the idea that BAs could be synthesized both directly from unpublished). shikimate/chorismate and from Phe in most plants. There is a pressing need, however, to determine in what tissues Activation with CoA and under which conditions these distinct pathways are The incorporation of a benzoyl moiety into secondary utilized. In addition, it is likely that these pathways metabolites occurs by the transfer of an activated BA intersect and impact each other, complicating analysis intermediate, such as benzoyl-CoA, via an acyltransfer- and interpretation. Multiple routes for the synthesis of ase. Though benzoate:CoA ligases had been isolated and BAs from each of these two major pathways are possible, characterized from benzoate-degrading microorganisms perhaps facilitating fine control and channeling of inter- [47–50], the characterization of benzoate:CoA ligases mediates to particular benzoate products. Finally, the from plants is more recent. A 3HBA:CoA ligase was modification of BAs might not only control activity and isolated and characterized from the plant Centaruium function but also be essential to their synthesis. erythrae [51], and the purification and characterization of a BA:CoA ligase (BZL) from Clarkia breweri flowers To date, as seen in Figures 2 and 3, few of the genes that followed [52]. In addition to showing activity with BA, are involved in BA biosynthesis (and modification) in BZL also exhibited significant activity with other struc- plants have been identified. However, the use of turally related compounds, including anthranilic acid (50% genetic/genomic model plants such as A. thaliana has relative activity compared with BA) and 3HBA (25% and can facilitate the identification of these genes and relative activity) [52]. Though benzoyl-CoA could be provide the framework for careful and integrated dissec- formed by alternative pathways (Figure 2), the presence tion of the complexity associated with BA synthesis, of a benzoyl-CoA ligase could allow for the refined spatial modification, and function [15,29]). In particular, SA and temporal control of benzoyl-CoA (and BA) availability. and benzoyl glucosinolate biosynthesis can be readily This could facilitate the specific channeling of benzoyl- examined in Arabidopsis, and it appears that the shiki- CoA to the appropriate acyltransferase for addition of the mate/chorismate direct and Phe-derived pathways are benzoyl moiety to a given secondary metabolite. In fact, an operational in Arabidopsis [5,15,58,59]. Because the acyltransferase that is capable of synthesizing benzyl shikimate and phenylpropanoid pathways are closely benzoate (a volatile ester) from benzoyl-CoA and benzyl associated, it is possible that genetically altering or inhi- alcohol has been identified in C. breweri flowers [53], biting the activity of one pathway could impact another. though we do not yet know whether this enzyme and These impacts might be assessed readily in Arabidopsis by C. breweri BZL are co-expressed and co-localized. global expression profiling. Comparative genomic and biochemical approaches could then be used to extend An acyl-CoA thioesterase could catalyze the reverse reac- (and diversify) findings to other plants. In parallel, tion to release BA from benzoyl-CoA. As CoA esters are detailed biochemical studies to elucidate benzenoid www.sciencedirect.com Current Opinion in Plant Biology 2006, 9:288–296 294 Physiology and metabolism

amarogentin — a retrobiosynthetic 13C NMR study. Eur J Org metabolism in systems that have elaborated and analyz- Chem 2001:1459-1465. able benzoate products (e.g. floral scent from petunia See annotation for [13]. petals [60]) are both complementary and invaluable. 15. Wildermuth MC, Dewdney J, Wu G, Ausubel FM: Isochorismate synthase is required to synthesize salicylic acid for plant Acknowledgements defence. Nature 2001, 414:562-565. I thank UC Berkeley and the US National Science Foundation (NSF) 16. van Tegelen LJ, Moreno PR, Croes AF, Verpoorte R, Wullems GJ: MCB-0420267 for financial support, Marcus Strawn for assistance with Purification and cDNA cloning of isochorismate synthase from figures, and collaborator Roger Innes for allowing the inclusion of elicited cell cultures of Catharanthus roseus. Plant Physiol information prior to publication. 1999, 119:705-712. 17. Moreno PRH, van der Heijden R, Verpoorte R: Elicitor-mediated References and recommended reading induction of isochorismate synthase and accumulation of Papers of particular interest, published within the annual period of 2,3-dihydroxy benzoic acid in Catharanthus roseus cell review, have been highlighted as: suspension and shoot cultures. Plant Cell Rep 1994, 14:188-191. of special interest 18. Abd El-Mawla AM, Schmidt W, Beerhues L: Cinnamic acid is a  of outstanding interest  precursor of benzoic acids in cell cultures of Hypericum androsaemum L. but not in cell cultures of Centaurium 1. Dempsey DMA, Shah J, Klessig DF: Salicylic acid and disease erythraea RAFN. Planta 2001, 212:288-293. resistance in plants. Crit Rev Plant Sci 1999, 18:547-575. 19. Schneider S, Mohamed ME, Fuchs G: Anaerobic metabolism of 2. Sharma YK, Leon J, Raskin I, Davis KR: Ozone-induced L-phenylalanine via benzoyl-CoA in the denitrifying bacterium responses in Arabidopsis thaliana: the role of salicylic acid in Thauera aromatica. Arch Microbiol 1997, 168:310-320. the accumulation of defense-related transcripts and induced resistance. Proc Natl Acad Sci USA 1996, 93:5099-5104. 20. Hertweck C, Jarvis AP, Xiang L, Moore BS, Oldham NJ: A mechanism of benzoic acid biosynthesis in plants and 3. Effmert U, Saschenbrecker S, Ross J, Negre F, Fraser CM, bacteria that mirrors fatty acid beta-oxidation. Chem BioChem Noel JP, Dudareva N, Piechulla B: Floral benzenoid carboxyl 2001, 2:784-786.  methyltransferases: from in vitro to in planta function. Phytochemistry 2005, 66:1211-1230. 21. Alibert G, Ranjeva R: Research on the enzymes catalyzing the A recent review focused on benzenoid carboxyl methyltransferases that biosynthesis of phenolic acids in Quercus pedunculata 1. includes a very nice comparative analysis. Formation of the first members of the cinnamic series and benzoic series. FEBS Lett 1971, 19:11-14. 4. Zhu J, Park KC: Methyl salicylate, a soybean aphid-induced plant volatile attractive to the predator Coccinella 22. Jarvis AP, Schaaf O, Oldham NJ: 3-hydroxy-3-phenylpropanoic septempunctata. J Chem Ecol 2005, 31:1733-1746. acid is an intermediate in the biosynthesis of benzoic acid and salicylic acid but benzaldehyde is not. Planta 2000, 5. Graser G, Oldham NJ, Brown PD, Temp U, Gershenzon J: The 212:119-126. biosynthesis of benzoic acid glucosinolate esters in Arabidopsis thaliana. Phytochemistry 2001, 57:23-32. 23. Loscher R, Heide L: Biosynthesis of p-hydroxybenzoate from p-coumarate and p-coumaroyl-coenzyme A in cell-free 6. Ruuhola T, Julkunen-Titto R: Trade-off between synthesis of extracts of Lithospermum erythrorhizon cell cultures. salicylates and growth of micropropagated Salix pentandra. Plant Physiol 1994, 106:271-279. J Chem Ecol 2003, 29:1565-1588. 24. Schnitzler J-P, Madlung J, Rose A, Seitz HU: Biosynthesis of 7. Abd El-Mawla AM, Beerhues L: Benzoic acid biosynthesis in cell cultures of Hypericum androsaemum. Planta 2002, p-hydroxybenzoic acid in elicitor-treated carrot cell cultures. Planta 1992, 188:594-600. 214:727-733. 8. Bjorklund J, Leete E: Biosynthesis of the benzoyl moiety of 25. French CJ, Vance CP, Towers GHN: Conversion of p-coumaric cocaine from cinnamic acid via (R)-(+)-3-hydroxy-3- acid to p-hydroxybenzoic acid by cell free extracts of phenylpropanoic acid. Phytochemistry 1992, 31:357-360. potato tubers and Polyporus hispidus. Phytochemistry 1976, 15:564-566. 9. Walker K, Croteau R: Taxol biosynthesis: molecular cloning of a benzoyl-CoA:taxane 2a-O-benzoyltransferase cDNA from 26. Yazaki K, Heide L, Tabata M: Formation of p-hydroxybenzoic Taxus and functional expression in Escherichia coli. Proc Natl acid from p-coumaric acid by cell free extract of Acad Sci USA 2000, 97:13591-13596. Lithospermum erythrorhizon cell cultures. Phytochemistry 1991, 30:2233-2236. 10. Lichtenthaler HK: The 1-deoxy-D-xylulose-5-phosphate pathway of isoprenoid biosynthesis in plants. Annu Rev Plant 27. Mitra A, Kitamura Y, Gasson MJ, Narbad A, Parr AJ, Payne J, Physiol Plant Mol Biol 1999, 50:47-65. Rhodes MJ, Sewter C, Walton NJ: 4-hydroxycinnamoyl-CoA hydratase/lyase (HCHL) — an enzyme of phenylpropanoid 11. Serino L, Reimmann C, Baur H, Beyeler M, Visca P, Haas D: chain cleavage from Pseudomonas. Arch Biochem Biophys Structural genes for salicylate biosynthesis from chorismate 1999, 365:10-16. in Pseudomonas aeruginosa. Mol Gen Genet 1995, 249:217-228. 28. Gasson MJ, Kitamura Y, McLauchlan WR, Narbad A, Parr AJ, Parsons EL, Payne J, Rhodes MJ, Walton NJ: Metabolism of 12. Gehring AM, Bradley KA, Walsh CT: Enterobactin biosynthesis in ferulic acid to vanillin. A bacterial gene of the enoyl-SCoA Escherichia coli: isochorismate lyase (EntB) is a bifunctional hydratase/isomerase superfamily encodes an enzyme for the enzyme that is phosphopantetheinylated by EntD and then hydration and cleavage of a hydroxycinnamic acid SCoA acylated by EntE using ATP and 2,3-dihydroxybenzoate. thioester. J Biol Chem 1998, 273:4163-4170. Biochemistry 1997, 36:8495-8503. 29. Chen F, D’Auria JC, Tholl D, Ross JR, Gershenzon J, Noel JP, 13. Werner I, Bacher A, Eisenreich W: Retrobiosynthetic NMR Pichersky E: An Arabidopsis thaliana gene for methylsalicylate studies with 13C-labeled glucose. Formation of gallic acid in biosynthesis, identified by a biochemical genomics approach,  plants and fungi. J Biol Chem 1997, 272:25474-25482. has a role in defense. Plant J 2003, 36:577-588. [13C] NMR studies (particularly using labeled substrates such as glucose and shikimate) provide a fairly unbiased and physiologically relevant 30. Negre F, Kolosova N, Knoll J, Kish CM, Dudareva N: Novel approach for elucidating the biosynthesis of BAs under different condi- S-adenosyl-L-methionine:salicylic acid carboxyl tions. methyltransferase, an enzyme responsible for biosynthesis of methyl salicylate and methyl benzoate, is not involved in floral 14. Wang C-Z, Maier UH, Eisenreich W, Adam P, Obersteiner I, Keil M, scent production in snapdragon flowers. Arch Biochem Bacher A, Zenk MH: Unexpected biosynthetic precursors of Biophys 2002, 406:261-270. 

Current Opinion in Plant Biology 2006, 9:288–296 www.sciencedirect.com Benzoic acids Wildermuth 295

31. Forouhar F, Yang Y, Kumar D, Chen Y, Fridman E, Park SW, 45. Craigon DJ, James N, Okyere J, Higgins J, Jotham J, May S: Chiang Y, Acton TB, Montelione GT, Pichersky E et al.: NASCArrays: a repository for microarray data generated by  Structural and biochemical studies identify tobacco  NASC’s transcriptomics service. Nucleic Acids Res 2004, 32 SABP2 as a methyl salicylate esterase and implicate it (Database issue):D575-D577. in plant innate immunity. Proc Natl Acad Sci USA 2005, A great resource for those employing in silico reverse genetic approaches 102:1773-1778. in Arabidopsis. It is possible to examine gene expression patterns and Crystal structure and biochemical studies reveal that SABP2 is likely to determine correlated genes in response to diverse treatments (including function as an MeSA esterase. This work highlights the importance of abiotic and biotic stressors), in diverse tissues, etc. obtaining crystal structures for plant enzymes; structure supports a role for SABP2 as a MeSA esterase and not as a lipase. 46. Warren RF, Merritt PM, Holub E, Innes RW: Identification of three putative signal transduction genes involved in R gene- 32. Kumar D, Klessig DF: High-affinity salicylic acid-binding protein specified disease resistance in Arabidopsis. Genetics 1999, 2 is required for plant innate immunity and has salicylic 152:401-412. acid-stimulated lipase activity. Proc Natl Acad Sci USA 2003, 100:16101-16106. 47. Lopez Barragan MJ, Carmona M, Zamarro MT, Thiele B, Boll M, Fuchs G, Garcia JL, Diaz E: The bzd gene cluster, coding for 33. Lee HI, Raskin I: Purification, cloning, and expression of a anaerobic benzoate catabolism, in Azoarcus sp. strain CIB. pathogen inducible UDP-glucose:salicylic acid J Bacteriol 2004, 186:5762-5774. glucosyltransferase from tobacco. J Biol Chem 1999, 274:36637-36642. 48. Schuhle K, Gescher J, Feil U, Paul M, Jahn M, Schagger H, Fuchs G: Benzoate-coenzyme A ligase from Thauera 34. Lim EK, Doucet CJ, Li Y, Elias L, Worrall D, Spencer SP, Ross J, aromatica: an enzyme acting in anaerobic and aerobic Bowles DJ: The activity of Arabidopsis glycosyltransferases pathways. J Bacteriol 2003, 185:4920-4929. toward salicylic acid, 4-hydroxybenzoic acid, and other benzoates. J Biol Chem 2002, 277:586-592. 49. Egland PG, Gibson J, Harwood CS: Benzoate-coenzyme A ligase, encoded by badA, is one of three ligases able to 35. Bowles D, Isayenkova J, Lim EK, Poppenberger B: catalyze benzoyl-coenzyme A formation during anaerobic Glycosyltransferases: managers of small molecules. Curr Opin growth of Rhodopseudomonas palustris on benzoate. Plant Biol 2005, 8:254-263. J Bacteriol 1995, 177:6545-6551. 36. Dean JV, Mohammed LA, Fitzpatrick T: The formation, vacuolar 50. Altenschmidt U, Oswald B, Fuchs G: Purification and localization, and tonoplast transport of salicylic acid glucose characterization of benzoate-coenzyme A ligase and conjugates in tobacco cell suspension cultures. Planta 2005, 2-aminobenzoate-coenzyme A ligases from a denitrifying 221:287-296. Pseudomonas sp. J Bacteriol 1991, 173:5494-5501. 37. Hennig J, Malamy J, Grynkiewicz G, Indulski J, Klessig DF: 51. Barillas W, Beerhues L: 3-Hydroxybenzoate:coenzyme A ligase from cell cultures of Centaurium erythraea: isolation and Interconversion of the salicylic acid signal and its glucoside in  tobacco. Plant J 1993, 4:593-600. characterization. Biol Chem 2000, 381:155-160. The most detailed purification and biochemical characterization of a plant 38. Chong J, Pierrel MA, Atanassova R, Werck-Reichhart D, Fritig B, BA:CoA ligase to date. No protein sequence – or cDNA. Saindrenan P: Free and conjugated benzoic acid in tobacco plants and cell cultures. Induced accumulation upon 52. Beuerle T, Pichersky E: Purification and characterization of elicitation of defense responses and role as salicylic acid benzoate:coenzyme A ligase from Clarkia breweri. Arch precursors. Plant Physiol 2001, 125:318-328. Biochem Biophys 2002, 400:258-264. 39. Fraissinet-Tachet L, Baltz R, Chong J, Kauffmann S, Fritig B, 53. D’Auria JC, Chen F, Pichersky E: Characterization of an Saindrenan P: Two tobacco genes induced by infection, acyltransferase capable of synthesizing benzylbenzoate and other volatile esters in flowers and damaged leaves of Clarkia elicitor and salicylic acid encode glucosyltransferases acting on phenylpropanoids and benzoic acid breweri. Plant Physiol 2002, 130:466-476. derivatives, including salicylic acid. FEBS Lett 1998, 54. Verleur N, Hettema EH, van Roermund CW, Tabak HF, 437:319-323. Wanders RJ: Transport of activated fatty acids by the peroxisomal ATP-binding-cassette transporter Pxa2 in a 40. Staswick PE, Tiryaki I: The oxylipin signal jasmonic acid is semi-intact yeast cell system. Eur J Biochem 1997, activated by an enzyme that conjugates it to isoleucine in 249:657-661.  Arabidopsis. Plant Cell 2004, 16:2117-2127. The first report of plant hormone activation by conjugation to an amino 55. Tilton GB, Shockey JM, Browse J: Biochemical and acid. molecular characterization of ACH2, an acyl-CoA  thioesterase from Arabidopsis thaliana. J Biol Chem 2004, 41. Staswick PE, Serban B, Rowe M, Tiryaki I, Maldonado MT, 279:7487-7494. Maldonado MC, Suza W: Characterization of an Arabidopsis  The authors describe the first cloning and characterization of a plant acyl- enzyme family that conjugates amino acids to indole-3-acetic CoA thioesterase. The biochemical activity and expression patterns of acid. Plant Cell 2005, 17:616-627. this enzyme suggest that it is not involved in fatty-acid oxidation. The authors identify putative IAA–amino acid synthetases in Arabidopsis and describe an approach for dissecting the complexity (including partial 56. Obel N, Scheller HV: Enzymatic synthesis and purification of redundancy) of these enzymes’ function. Note that GH3.5 mutants exhibit caffeoyl-CoA, p-coumaroyl-CoA, and feruloyl-CoA. Anal weak auxin phenotypes; thus, a potential role for GH3.5 as an SA–amino Biochem 2000, 286:38-44. acid synthetase remains a possibility. 57. Tilton G, Shockey J, Browse J: Two families of acyl-CoA 42. Woodward AW, Bartel B: Auxin: regulation, action, and thioesterases in Arabidopsis. Biochem Soc Trans 2000, interaction. Ann Bot 2005, 95:707-735. 28:946-947. An excellent review on the well-studied phytohormone auxin. The authors provide a framework for thinking about the synthesis, modification, 58. Mauch-Mani B, Slusarenko AJ: Production of salicylic acid activity, and function of small molecules in plants. precursors is a major function of phenylalanine ammonia-  lyase in the resistance of Arabidopsis to Peronospora 43. Staswick PE, Tiryaki I, Rowe ML: Jasmonate response locus parasitica. Plant Cell 1996, 8:203-212. JAR1 and several related Arabidopsis genes encode enzymes Use of the PAL inhibitor AIP impacted the growth and reproduction of an of the firefly luciferase superfamily that show activity on avirulent isolate of Peronospora parasitica, but not of a virulent isolate. jasmonic, salicylic, and indole-3-acetic acids in an assay for Total SA accumulation was reduced by about two-thirds in the avirulent adenylation. Plant Cell 2002, 14:1405-1415. interaction when AIP was employed. AIP treatment abrogated induced PAL expression; thus, the specificity of its impact needs to be examined. 44. Shockey JM, Fulda MS, Browse J: Arabidopsis contains a large superfamily of acyl-activating enzymes. 59. Ferrari S, Plotnikova JM, De Lorenzo G, Ausubel FM: Arabidopsis Phylogenetic and biochemical analysis reveals a new class local resistance to Botrytis cinerea involves salicylic acid and of acyl-coenzyme A synthetases. Plant Physiol 2003,  camalexin and requires EDS4 and PAD2, but not SID2, EDS5 or 132:1065-1076. PAD4. Plant J 2003, 35:193-205. www.sciencedirect.com Current Opinion in Plant Biology 2006, 9:288–296 296 Physiology and metabolism

By coupling the use of a mutant in SA biosynthesis (via ICS) with a PAL A very nice integrated approach (involving modeling, isotope-labeling, inhibitor, the authors found that local resistance to Botrytis appears to be and metabolite analysis) for analyzing the complexity of benzenoid mediated by SA that is derived from t-CA. metabolism in petunia petal tissue. The authors predicted a benzyl benzoate intermediate, isolated a cDNA encoding a protein that is cap- 60. Boatright J, Negre F, Chen X, Kish CM, Wood B, Peel G, Orlova I, able of synthesizing benzyl benzoate, and characterized this enzyme Gang D, Rhodes D, Dudareva N: Understanding in vivo biochemically.  benzenoid metabolism in petunia petal tissue. Plant Physiol 2004, 135:1993-2011.

How to re-use Elsevier journal figures in multimedia presentations

It’s easy to incorporate figures published in Trends, Current Opinion or Drug Discovery Today journals into your multimedia presentations or other image-display programs.

1. Locate the article with the required figure on ScienceDirect and click on the ‘Full text + links’ hyperlink 2. Click on the thumbnail of the required figure to enlarge the image 3. Copy the image and paste it into an image-display program

Permission of the publisher is required to re-use any materials from Trends, Current Opinion or Drug Discovery Today journals or from any other works published by Elsevier. Elsevier authors can obtain permission by com- pleting the online form available through the Copyright Information section of Elsevier’s Author Gateway at http:// authors.elsevier.com. Alternatively, readers can access the request form through Elsevier’s main website at:

www.elsevier.com/locate/permissions

Current Opinion in Plant Biology 2006, 9:288–296 www.sciencedirect.com