auxin biosynthesis: are the indole-3-acetic acid and ... · auxin biosynthesis: are the...

Auxin Biosynthesis: Are the Indole-3-AceticAcid and Phenylacetic Acid Biosynthesis PathwaysMirror Images?1[OPEN]

Sam D. Cook, David S. Nichols, Jason Smith, Prem S. Chourey, Erin L. McAdam, Laura Quittenden, andJohn J. Ross*

School of Biological Sciences (S.D.C., E.L.M., L.Q., J.J.R.), Central Science Laboratory (D.S.N.), School ofChemistry (J.S.), University of Tasmania, Sandy Bay, Tasmania, Australia, 7005; and U.S. Department ofAgriculture, Agricultural Research Service, Gainesville, Florida 32608 (P.S.C.)

ORCID IDs: 0000-0003-0950-6131 (S.D.C.); 0000-0003-3665-2062 (J.J.R.).

The biosynthesis of the main auxin in plants (indole-3-acetic acid [IAA]) has been elucidated recently and is thought to involvethe sequential conversion of Trp to indole-3-pyruvic acid to IAA. However, the pathway leading to a less well studied auxin,phenylacetic acid (PAA), remains unclear. Here, we present evidence from metabolism experiments that PAA is synthesizedfrom the amino acid Phe, via phenylpyruvate. In pea (Pisum sativum), the reverse reaction, phenylpyruvate to Phe, is alsodemonstrated. However, despite similarities between the pathways leading to IAA and PAA, evidence from mutants in peaand maize (Zea mays) indicate that IAA biosynthetic enzymes are not the main enzymes for PAA biosynthesis. Instead, weidentified a putative aromatic aminotransferase (PsArAT) from pea that may function in the PAA synthesis pathway.

Recently, the predominant pathway for auxin bio-synthesis in plants has been characterized, and it is nowwidely accepted that indole-3-acetic acid (IAA), themain auxin in plants, is primarily synthesized via a two-step pathway. This Trp-dependent synthesis of IAA isreported to utilize indole-3-pyruvic acid (IPyA) as the soleintermediate (Mashiguchi et al., 2011;Won et al., 2011;Daiet al., 2013).

The initial conversion of Trp to IPyA is catalyzed bytheTAA (TRPAMINOTRANSFERASE of ARABIDOPSIS)gene family, which includes multiple members inArabidopsis (Arabidopsis thaliana; Tao et al., 2008), maize(Zea mays; Chourey et al., 2010; Phillips et al., 2011), pea(Pisum sativum; Tivendale et al., 2012), and several otherspecies (Chourey et al., 2010; Liu et al., 2012). Enzymatic

assays on TAA family proteins have indicated that Trp isa preferred substrate for these enzymes (Tao et al., 2008).

The subsequent conversion of IPyA to IAA is cata-lyzed by a group of enzymes known as YUCCAs(Mashiguchi et al., 2011; Dai et al., 2013). The YUC genefamily is also quite extensive, with Arabidopsis pos-sessing 11 annotated YUC genes (Cheng et al., 2007).Unfortunately, IPyA is prone to substantial nonenzy-matic breakdown to IAA (Tam and Normanly, 1998),making quantification and analysis difficult. Neverthe-less, YUCCA activity has been reported in vitro, instudies using recombinant YUCprotein (Dai et al., 2013).

Such assays have been complemented by mutantstudies, although in Arabidopsis, high order yuc mu-tants (triple or quadruple) are required to elicit a dis-tinct phenotype (Cheng et al., 2007). In maize, however,single YUC knockouts can cause large reductions inIAA levels, resulting in severe alterations to vegetativeand reproductive development (Gallavotti et al., 2008;Phillips et al., 2011; Bernardi et al., 2012).

Trp-independent synthesis may also contribute to theIAA pool in a range of species (Michalczuk et al., 1992;Rapparini et al., 1999; Epstein et al., 2002; Sztein et al.,2002). Recent evidence suggests that this pathway maybe essential for embryo development (Wang et al.,2015), although this suggestion has been challenged(Nonhebel, 2015).

A second endogenous auxin, phenylacetic acid (PAA),has received substantially less attention than IAA. Thisauxin is present at physiologically similar levels to IAAand has been recorded at high levels in some tissues(Schneider and Wightman, 1986; Sugawara et al., 2015).Although the auxin activity of PAA in wheat (Triticum

1 This work was supported by the Australian Research Council(grant DP130103357 to J.J.R.).

* Address correspondence to [email protected] author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policy de-scribed in the Instructions for Authors (www.plantphysiol.org) is:John J. Ross ([email protected]).

S.D.C. conducted all experiments and interpreted the majority ofthe results, except for IAAld experiments; E.L.M. and L.Q. partici-pated in tar2-1 mutant experiments and conducted IAAld experi-ments; J.S. synthesized the D5phenylpyruvate; D.S.N. wasresponsible for the running and minor interpretation of UPLC-MS;P.S.C. provided material and contributed to maize-related results;S.D.C. and J.J.R. conceived the project and wrote the article withcontributions from other authors.

[OPEN] Articles can be viewed without a subscription.www.plantphysiol.org/cgi/doi/10.1104/pp.16.00454

1230 Plant Physiology�, June 2016, Vol. 171, pp. 1230–1241, www.plantphysiol.org � 2016 American Society of Plant Biologists. All Rights Reserved. www.plantphysiol.orgon September 16, 2018 - Published by Downloaded from

Copyright © 2016 American Society of Plant Biologists. All rights reserved.

http://orcid.org/0000-0003-0950-6131

http://orcid.org/0000-0003-3665-2062

mailto:[email protected]

http://www.plantphysiol.org

mailto:[email protected]

http://www.plantphysiol.org/cgi/doi/10.1104/pp.16.00454


aestivum) coleoptile and pea segment tests are not asstrong as that of IAA (Wightman and Lighty, 1982), PAAis reportedly capable of eliciting a stronger response ininitiation of root primordia (Wightman et al., 1980;Schneider et al., 1985).Like IAA, PAA has been found in numerous genera,

including bacteria, fungi, and some basal land plants(Abe et al., 1974; Kishore et al., 1976; Wightman andLighty, 1982; Schneider and Wightman, 1986; Hwanget al., 2001; Kim et al., 2004; Sugawara et al., 2015). Infungi and bacteria, PAA is produced from the aminoacid Phe, via the intermediate phenylpyruvate (Kishoreet al., 1976; Krings et al., 1996; Groot and de Bont, 1998;Somers et al., 2005). Bacterial phenylpyruvate is eitherconverted directly to PAA by a phenylpyruvate de-carboxylase or via an additional intermediate, phenyl-acetaldehyde (Somers et al., 2005; Spaepen et al., 2007).It has been suggested that PAA biosynthesis in plants

also occurs via phenylpyruvate (Taylor andWightman,1987). It has further been proposed that the enzymesresponsible for IAA biosynthesis are involved in theconversion of Phe to PAA (Sugawara et al., 2015). Invitro assays support a TAA1-mediated conversion ofPhe to phenylpyruvate (Tao et al., 2008). However, theTAA1 enzyme has a Km for L-Phe that is more than 30times that of L-Trp (Tao et al., 2008). There is also evi-dence for the conversion of phenylpyruvate to PAA byboth AtYUC6 and AtYUC2 protein in vitro (Dai et al.,2013; Sugawara et al., 2015). In fact, this reaction wasused to identify the biochemical mechanism of YUC(Dai et al., 2013).Phe is a precursor of many important compounds in

plants, and its biosynthesis is well documented, al-though more than one biosynthetic pathway exists(Cho et al., 2007; Yamada et al., 2008; Tzin et al., 2009;Maeda et al., 2010; Yoo et al., 2013). The final product ofthe shikimate pathway, chorismate, is converted to pre-phenate, a branch point in Phe synthesis, by a chorismatemutase (Lee et al., 1995; Herrmann andWeaver, 1999). Insome species, prephenate is then primarily directed to-ward arogenate by a prephenate aminotransferase. This isfollowed by conversion to Phe by an arogenate dehy-dratase (Siehl et al., 1986; Cho et al., 2007; Maeda et al.,2010, 2011).The alternative pathway of Phe biosynthesis in plants

involves the conversion of prephenate to phenylpyruvateby a prephenate dehydratase (Maeda et al., 2010), al-though several arogenate dehydratases also possess thisability (Cho et al., 2007). Phenylpyruvate is subsequentlyconverted to Phe by a phenylpyruvate aminotransfer-ase (Tzin et al., 2009;Maeda et al., 2010; Yoo et al., 2013).This synthetic pathway appears incongruous with thepathway proposed for PAA biosynthesis where phe-nylpyruvate exists as a metabolite of Phe, rather than aprecursor (Sugawara et al., 2015). Although it should benoted that transamination is a reversible process (Jensenand Gu, 1996). Interestingly, the aminotransferases dis-cussed in relation to Phe biosynthesis in petunia (Petuniahybrida; Yoo et al., 2013) do not include TAA1 familyproteins nor their function on Trp.

Although Sugawara et al. (2015) favor the TAA1/YUC pathway for PAA synthesis, they do not identifyor quantify phenylpyruvate in their model system,Arabidopsis. In fact, the quantification of this com-pound is thought to be problematic. Maeda et al. (2010)reported that prephenate is subject to acid breakdownto phenylpyruvate during extraction procedures, cast-ing doubt on identifications of phenylpyruvate that donot account for this possibility. In fact, after accountingfor this conversion, Maeda et al. (2010) reported unde-tectable levels of phenylpyruvate in petals of theirmodel system, petunia. On the other hand,Manela et al.(2015) have recently reported detection of phenylpyruvatein grape (Vitis vinifera) using gas chromatography-massspectrometry; however, no chromatograms were pro-vided as evidence.

Results presented here support the theory that PAAis derived fromphenylpyruvate as suggestedbySugawaraet al. (2015). However, we propose that the enzymesinvolved are not those responsible for IAA biosynthe-sis. We show that while IAA synthetic enzymes arecapable of catalyzing reactions involved in PAA syn-thesis in vitro, it is unlikely that this reflects the in vivofunction of such enzymes. Furthermore, we present analternative candidate for the initial transamination re-action, identified here as PsArAT (KX236168). Phylo-genetic analysis of the translated amino acid sequenceof PsArAT suggests that it may be capable of utilizingPhe as a substrate. In addition to our studies on PAAbiosynthesis, we further clarify several aspects of theIAA and 4-Cl-IAA biosynthetic pathways.

RESULTS AND DISCUSSION

Phe Can Be Converted to PAA in Vivo

We sought to confirm that Phe is a precursor of PAAby investigating the metabolism of labeled (D5) Phe inexcised pea seeds (Fig. 1A). Over the course of 16 h, thesubstrate was metabolized, resulting in a clear dilutionof the phenylpyruvate pool with deuterium (35%). PAAwas also diluted by 18%, while the endogenous Phepool was diluted by 44%. The identity of D5phenyl-pyruvate and D5PAA was confirmed by obtainingproduct scan spectra (Supplemental Fig. S1). These re-sults are consistent with the proposition of a linear, two-step pathway (Sugawara et al., 2015). It is known thatdifferent tissue types may possess alternative auxinsynthetic pathways (Sugawara et al., 2009); therefore, toconfirm this chemical pathway, we investigated PAAsynthesis in vegetative tissue (Fig. 1B).

Germinating seedlings were hydroponically incu-bated in a solution containing D5Phe over a 7-d period.Following this, endogenous Phe was diluted in excessof 50% and PAA was labeled to a similar extent (49%;Fig. 1B). This indicates that Phe can be an in vivo pre-cursor to PAA in pea vegetative tissue. However, themajor proposed intermediate, phenylpyruvate, was onlydiluted 5% with deuterium in this experiment (Fig. 1B).

Plant Physiol. Vol. 171, 2016 1231

IAA and PAA: Similar Pathways, Different Enzymes

www.plantphysiol.orgon September 16, 2018 - Published by Downloaded from Copyright © 2016 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org/cgi/content/full/pp.16.00454/DC1


The low dilution may have been due to de novo syn-thesis of unlabeled phenylpyruvate from prephenate oralternatively the conversion of unlabeled prephenate tophenylpyruvate in the acidic conditions used duringpurification (Hermes et al., 1984).

Further analysis of possible PAA intermediatesidentified a 10% dilution of the phenylacetaldehyde(PAAld) pool (following stabilization to the thiazolidine-labeledderivative, PAAld-TAZ) in the homogenate of peaapical tissue (Supplemental Fig. S2). This is consistentwith previous studies on flavor volatiles in petuniawhere PAAld is an importantmetabolite of Phe (Boatrightet al., 2004). While PAAld is absent from the proposedPAA biosynthetic scheme in plants (Sugawara et al.,2015), whether this exclusion is justified remains unclear.

The biosynthetic scheme for IAApresentedbySugawaraet al. (2009) includes several compounds from alterna-tive IAA biosynthetic pathways as intermediates. Weused this scheme to identify analogous compounds todetermine if they are also intermediates in PAA bio-synthesis. Phenylethylamine is a direct precursor toPAAld in other systems (Boatright et al., 2004; Taylorand Wightman, 1987, Tieman et al., 2006). However,despite its presence in petunia and tomato (Solanumlycopersicum), we were unable to detect either the en-dogenous or labeled species of phenylethylamine in ourfeeding studies (Supplemental Fig. S2). In Tropaeolum,phenylacetonitrile has previously been proposed as apossible intermediate between glucosinolates and PAA(Ludwig-Müller and Cohen, 2002). Here, we sought todetermine if phenylacetonitrile and phenylacetamide,the remaining intermediate, were present in our sys-tem. However, we were unable to detect either of thesecompounds in either seeds or vegetative tissue of pea(Supplemental Fig. S2). Based on the analogous IAAsynthetic pathways (Sugawara et al., 2009), it is unlikely

that either of these compounds are major intermediatesinvolved in PAA biosynthesis in pea.

Our evidence from the labeling of Phe metabolitessupports the theory that PAA is mainly synthesizedthrough phenylpyruvate (Sugawara et al., 2015). How-ever, labeling patterns similar to those in vegetative tissue(i.e. a small degree of dilution of endogenous phenyl-pyruvate; Fig. 1B) have been previously used to excludetryptamine from the IAA biosynthetic scheme in to-mato (Cooney and Nonhebel, 1991). It is important tonote that phenylpyruvate may be compartmentalizedin the pea vegetative system, as reported for petuniaand Atropa, where phenylpyruvate is located in boththe cytosol and in plastids (Yoo et al., 2013; Bedewitzet al., 2014).

Phenylpyruvate Can Be Converted to Phe in Vivo

To further investigate the role of phenylpyruvate inour system, we incubated crude homogenates of peaapical tissue with synthesized D5phenylpyruvate (see“Materials and Methods”). In these experiments, thephenylpyruvate pool was diluted by an average of 32%with the labeled species following an overnight incu-bation (Fig. 2). We found that label was heavily incor-porated into PAA (Fig. 2; in excess of 200%), supportingthe suggestion that phenylpyruvate is a key interme-diate in PAA biosynthesis (Sugawara et al., 2015). Al-though the degree of labeling of Phe was only 5 to 10%,this nevertheless indicates strong conversion of phe-nylpyruvate to Phe, given the abundance of this aminoacid in the tissues concerned (up to 200 mg/g freshweight). PAAld was also labeled to a small extent(,1%; data not shown). The presence of label in Pheindicates that phenylpyruvate can also be a precursorfor Phe in our system.

Figure 1. Representative UPLC-MSchromatograms showing dilution ofPhe metabolic products after incu-bation with D5Phe in excised peaseeds (A) and seedlings (B). Peaksare relative intensities of deuteriumlabeled (green) and endogenous(black) species of Phe (approximateretention time: 1.11 min), phenyl-pyruvate (2.72 min), and phenyl-acetic acid (3.73 min). Retentiontimes in A vary due to preparationunder neutral pH conditions. Transi-tions indicated are for the endogenousspecies (all D5 transitions arem/z+5).

1232 Plant Physiol. Vol. 171, 2016

Cook et al.






Conversion of D5Trp to D5IPyA in Vivo

To confirm that labeled precursormetabolism studiesare effective in the investigation of auxin syntheticpathways, we incubated pea seedlings with deuteratedTrp to observe the fate of the deuterium label. In theseexperiments, deuterated IPyA was consistently detec-ted, following conversion to the stable thiazolidinederivative with cysteamine (Supplemental Fig. S3;Novák et al., 2012). To our knowledge, this is the firsttime that the conversion of labeled Trp to labeled IPyAhas been directly demonstrated in plants, despite thefact that the IPyA pathway is now recognized as themain pathway in many species (Tivendale et al., 2014;Sugawara et al., 2015).

ZmYUC1 Is Unlikely to Play a Role in PAA Biosynthesis

It has been suggested that, in vivo, PAA is synthe-sized by the same enzymes involved in the synthesis ofIAA (Sugawara et al., 2015). As there is currently nopublished YUC mutant in pea, we tested this sugges-tion by utilizing an auxin mutant from maize. In themaize defective endosperm mutant (de18; Bernardi et al.,2012), the ZmYUC1 protein is truncated resulting incomplete loss of enzyme function. This single mutationcauses up to a 100-fold drop in the levels of free IAA inmaize kernels throughout kernel maturation (Bernardiet al., 2012). Using mutant tissue, we sought to deter-mine if a nonfunctional ZmYUC1 affected the levels ofPAA. We found that, while IAA levels were reduced,consistent with previous results (Fig. 3C; Bernardi et al.,2012), there was no reduction in free PAA content,compared with the wild type, B37 (Fig. 3A).To eliminate the possibility that altered conjugation

of PAA was maintaining free auxin levels, sampleswere hydrolyzed to release both ester and amide de-rivatives. The total PAA levels show no significantdifference between the de18 mutant and its corre-sponding wild type, B37 (Fig. 3B). However, hydrolysisdid reveal a more pronounced reduction in total IAA(0.3% of the wild type; Fig. 3D) compared with that of

free IAA (1.8% of the wild type; Fig. 3C). These resultssuggest that ZmYUC1 does not play a role in PAA bio-synthesis. In Arabidopsis also, there is no evidencethat higher order YUC mutants (yuc1 yuc2 yuc6) affectPAA content, even though IAA levels were reduced inthe triple mutant (Sugawara et al., 2015). Interestingly,two of the three enzymes concerned in that study(AtYUC2 and AtYUC6) have been shown to convertphenylpyruvate to PAA in vitro (Dai et al., 2013;Sugawara et al., 2015).

It has been suggested that other members of theArabidopsis YUC family might contribute to PAA bio-synthesis in the yuc1 yuc2 yuc6 mutant (Sugawaraet al., 2015), and the samemay apply to the de18mutant.However, in maize endosperm, the ZmYUC1-likeYUCs, ZmYUC2 and ZmYUC3, are expressed at ex-ceptionally low levels (Bernardi et al., 2012), while an-other characterized YUC, ZmSpi1 (Gallavotti et al.,2008), is not expressed. Based on additional expressiondata collected by Sekhon et al. (2011), we sought toidentify any other YUC sequences that were expressedat any stage in the maize endosperm. These data showthat the only other gene with significant transcriptlevels (GRMZM2G109515) is expressed at a fraction ofZmYUC1 levels (;7%) throughout endosperm devel-opment (Supplemental Table S1). Given these expres-sion patterns, it is possible that the protein encoded atthe above locus may contribute to PAA biosynthesis;however, a substantial contribution appears unlikely.

Currently, the main in vivo evidence that YUCs areinvolved in PAA biosynthesis comes from over-expressing AtYUC1, AtYUC2, and AtYUC6 in Arabi-dopsis. This did appear to increase flux through thePAA biosynthesis pathway: In the overexpression lines,the level of the PAA conjugates PAA-Asp and PAA-Gluwere elevated between 14- to 41-fold and 1.6- to 3.8-fold, respectively (Sugawara et al., 2015). However, theextent of up-regulation of the YUC genes was quitehigh: Induction by b-estradiol resulted in over-expression of YUC2 by 142-fold and YUC6 by 29-fold(Sugawara et al., 2015). It is not unexpected that theelevated enzyme abundance can utilize endogenous

Figure 2. Representative UPLC-MS chromatograms from pea apical tissue incubated with labeled (D5) phenylpyruvate for 16 h.Peaks are relative intensities of deuterium labeled (green) and endogenous (black) species of Phe (approximate retention time:1.11 min), phenylpyruvate (2.78min), and phenylacetic acid (3.63min). Transitions indicated are for the endogenous species (allD5 transitions are m/z +5).







phenylpyruvate, as this capacity has been shown invitro (Dai et al., 2013; Sugawara et al., 2015). Further-more, since overexpressing an enzyme that normallyplays only a minor role in a pathway can dramaticallyaffect flux in that pathway (Yoo et al., 2013), suchoverexpression studies are not necessarily definitive inthemselves.

A Mutation in PsTAR2 Does Not Affect PAA LevelsDespite Drops in IAA, 4-Cl-IAA, and Their Derivatives

In maturing pea seeds, PsTAR2 encodes the mainenzyme for converting Trp and chlorinated Trp to IAAand chlorinated IAA, respectively (Tivendale et al.,2012). Here, we demonstrate that in addition to the re-duced 4-Cl-IAA content in tar2-1 seeds (Fig. 4A), thelevels of chlorinated IAA conjugates (including methylesters) were also reduced (Fig. 4B). We also report onstrong reductions in the content of both IAA (Fig. 4C)and IAA conjugates (Fig. 4D) in mutant tar2-1 seeds.These findings confirm that the vast majority of 4-Cl-IAAand IAA in maturing pea seeds is synthesized via theIPyA pathway. If this pathway made only a minorcontribution, it is difficult to envisage how the tar2-1mu-tation could dramatically reduce the content of bothfree 4-Cl-IAA and 4-Cl-IAA derivatives.

However, the data in Figure 4 show no reduction inPAA in tar2-1 seeds (Fig. 4E), and the levels of PAAconjugates (putative end products) were also unaf-fected in the mutant (Fig. 4F). The conjugate levels arein some ways more instructive than that of free PAA,since they provide a record of previous flux through thepathway. (However, in both pea and maize, little isknown about the specific nature of PAA conjugates.)Our results indicate that TAR2 is unlikely to play a rolein the synthesis of PAA in pea seeds. Interestingly, wedid not detect chlorinated PAA in pea seeds, indicatingsome substrate specificity for the enzymes responsiblefor chlorination in these organs (Tivendale et al., 2012).

Sugawara et al. (2015) reported a reduction in bothIAA and PAA levels in the TAA1 mutant (wei8-1) ofArabidopsis. However, the reduction in PAA was only20% compared with 80% for IAA, and no PAA conju-gate data were provided. Sugawara et al. (2015) alsosuggest that other TAA1-like enzymes may participatein the homeostatic regulation of PAA in the wei8-1 mu-tant, thus explaining the weak reduction in PAA.However, in our system, PsTAR2 is strongly expressedfrom themiddle to late stages of seed development, whilePsTAR1 and PsTAR3 are expressed at very low levelsduring this stage (Tivendale et al., 2012). In addition, ex-pression of these genes was not altered in tar2-1 seeds

Figure 3. Effects of the maize de18mutation on the levels of free andtotal auxin in developing endo-sperm (16 d after pollination) com-pared to the corresponding wildtype (B37). Data are means 6 SE

(ng g21 dry weight [DW], n = 3). Theasterisk indicates a significant dif-ference at the 0.05 level. A, FreePAA; B, total PAA; C, free IAA; D,total IAA.


Cook et al.



compared with the wild type (Supplemental Fig. S4). Wetherefore conclude that the TAR genes, in general, do notappear to be involved in PAA biosynthesis in pea seeds.We also exploited the tar2-1 mutation to investigate

the origin of (chlorinated) indole-3-acetaldehyde. Theorigin of this compound, and of the analogous PAAld,is relevant to the overall question of biosynthetic routesleading to auxins. The presence of some deuterium inPAAld after feeds of deuterated phenylpyruvate isconsistent with phenylpyruvate being a precursor ofPAAld. We found that in tar2-1 seeds, the content of chlo-rinated indole-3-acetaldeyde was reduced (SupplementalFig. S5), indicating that in pea seeds, chlorinated acetalde-hyde originates mainly from the chlorinated IPyA path-way. This does notmean, however, that the acetaldehyde isfurther converted to 4-Cl-IAA in this system. Interest-ingly, in Arabidopsis, Mashiguchi et al. (2011) con-cluded that indole-3-acetaldehyde does not originatefrom the IPyA pathway, possibly indicating a differ-ence between the two species.

PsTAR1 Prefers Trp in Vitro

Another member of the pea TAR family, PsTAR1, isstrongly expressed during the early stages of seed de-velopment (Tivendale et al., 2012). Since no tar1mutant

is currently available, we instead expressed TAR1 as arecombinant protein to test its functional activity in anenzymatic assay. Based on previous results in Arabi-dopsis (Tao et al., 2008), we sought to confirm if peaTAR proteins were able to utilize a range of substratesin vitro.

As previously shown, PsTAR1 protein was able toconvert Trp to IPyA in vitro (Tivendale et al., 2012).Following 3 h of incubation, more than 99% of thesubstrate was consumed (Fig. 5A). Mass spectrometryidentified significant amounts of IPyA in the reactionsolution. As expected, large quantities of IAAwere alsopresent presumably due to the nonenzymatic degra-dation of IPyA to IAA (Tam and Normanly, 1998).

PsTAR1 also converted D5Phe to D5phenylpyruvate(Fig. 5B), and D5PAA was also detected in the reactionmixture. However, the ratio of product to substrate wasdramatically reduced for Phe, compared with Trp.Furthermore, less than half of the Phe substrate wasconsumed under identical conditions. These resultsshow that PsTAR1 strongly prefers Trp to Phe, pro-viding further evidence that TARs do not play a role inPAA biosynthesis in pea.

We also tested the ability of PsTAR1 to convert IPyAto Trp, given that transamination is reported to be areversible reaction (Jensen and Gu, 1996). IPyA haspreviously been identified as a preferred substrate for

Figure 4. Effects of the pea tar2-1 mutation on the levels of free auxins and auxin derivatives (including conjugates and methyl-esters) in maturing seeds compared to the corresponding wild type (TAR2). Data are means6 SE (ng g21 fresh weight [FW], n =3-4). The asterisk indicates a significant difference at the 0.05 level. A, Free 4-Cl-IAA; B, conjugated 4-Cl-IAA; C, free IAA; D,conjugated IAA; E, free PAA; F, conjugated PAA.








AtTAA1 in in silico docking experiments (Tao et al.,2008), although to our knowledge this has not beenconfirmed in a functional assay. Here, we report thatPsTAR1 can also catalyze the conversion of IPyA to Trpin vitro (Fig. 5C). Consistent with IPyA as a preferredsubstrate (Tao et al., 2008), utilization of D5Trp byPsTAR1 was reduced by 75% in the presence of IPyA(data not shown). These results suggest that given thesubstrate preference of TAR family proteins, it is un-likely that TARs play a major role in the PAA biosyn-thesis pathway in pea.

Alternative Aminotransferase Genes, Separate from theTAR Family, Exist in the Genome of Pea

Our results cast doubt on a role for TAA1 or YUC inPAA biosynthesis, although they do suggest that PAA

is synthesized from Phe through phenylpyruvate.Consequently, the question arises: Are there alternativecandidates for the enzymes that catalyze the conversionof Phe to PAA in pea? We therefore used a bio-informatics approach to identify enzymes that poten-tially play a role in PAA biosynthesis.

In petunia, the conversion of phenylpyruvate to Phecan be catalyzed by the phenylpyruvate aminotransfer-ase (PhPPY-AT; Yoo et al., 2013). This reaction is revers-ible, and PhPPY-AT can also use Phe as a substrate (Yooet al., 2013). In fact, this enzyme was initially identifiedusing sequence information from an aminotransferase inCucumis melo that was also shown to utilize Phe in vitro(Gonda et al., 2010). Using these sequences, we identifiedan aminotransferase (PsCam000495_1) in the pea geneatlas (Alves-Carvalho et al., 2015). We suggest that theencoded enzyme, a putative aromatic aminotransferase

Figure 5. UPLC-MS chromatograms of in vitroenzyme assay products following incubation ofrecombinant PsTAR1 with D5Trp (to D5 indole-3-pyruvate; A), D5Phe (to D5phenylpyruvate; B), andindole-3-pyruvate (to Trp; C). Peaks are relativeintensities accompanied by mass transitions. Ar-rows indicate the direction of reaction.


Cook et al.



(PsArAT), is a strong candidate for the enzyme thatconverts Phe to phenylpyruvate in pea. PsArAT shares astrong similarity to the PhPPY-AT and CmArAT aminoacid sequences (67 and 72% identity, respectively). In-terestingly, there are no other similar sequences in pea,suggesting that PsArAT may be solely responsible forthe transamination of Phe and phenylpyruvate in thisspecies.As transaminase activity for both CmArAT and

PhPPY-AT has been demonstrated (Gonda et al., 2010;Yoo et al., 2013), we constructed a multiple sequencealignment with these two sequences, and those ofPsArAT and TcTAT (Trypanosoma cruzi), an amino-transferase on which x-ray crystallography has beenconducted (Blankenfeldt et al., 1999; Supplemental Fig.S6.). The PsArAT amino acid sequence contains con-served residues responsible for anchoring of the PLPcofactor as well as residues responsible for substratebinding (Blankenfeldt et al., 1999; Gonda et al., 2010).These residues are absolutely conserved across multi-ple lineages, including mammals and protozoans

(Blankenfeldt et al., 1999), strongly supporting a po-tential role for PsArAT in the conversion of Phe tophenylpyruvate.

CONCLUSION

It is now widely accepted that for the biosynthesis ofIAA, the IPyA pathway predominates inmost, if not all,angiosperms (Zhao, 2012; Tivendale et al., 2014). Basedon the chemical similarities of IAA and PAA, it has beensuggested recently that the biosynthetic pathways forthese two auxins are analogous, beginning from Trpand Phe, respectively (Sugawara et al., 2015). Here,using labeled compounds, we show that Phe can in-deed be converted to phenylpyruvate, which can thenbe converted to PAA.However, while IAA biosynthesisappears to involve a simple two-step pathway, it ispossible that the intermediate phenylpyruvate is both aprecursor and a metabolite of Phe. Consistent with itspossible role as a Phe precursor, we show here thatphenylpyruvate can be converted to Phe in vivo. An

Figure 6. Comparison of the biosynthetic pathways for IAA and PAA (blue arrows) and Phe (red arrows). Catalytic enzymes areAtTAA1 and PsTAR1 (Trp aminotransferase related), YUC (flavin mono-oxygenase), PhPPY-AT and PsArAT (aromatic amino-transferase), ADT (arogenate dehydrogenase), PDT (prephenate dehydratase), and PhPPA-AT (prephenate aminotransferase).Arrows indicate direction of reaction, and the dashed line represents an as yet unknown catalytic mechanism.







important difference between the analogous auxinpathways in plants is that, to our knowledge, IPyA isformed only from Trp, whereas phenylpyruvate ap-pears to be synthesized from either Phe or prephenate(Fig. 6; Cho et al., 2007; Maeda et al., 2010).

Our data from two in vivo systems indicate that themain enzymes for IAA biosynthesis are not the mainenzymes for PAA biosynthesis. We investigated therole of TAR and YUC in the synthesis of PAA by ana-lyzing hormone levels in hormone synthesis mutants.In both the pea tar2-1 and the maize de18 mutants,levels of IAA (and 4-Cl-IAA in pea) were dramaticallyreduced. Despite this, there were no consistent changesin the levels of PAA.We suggest that previous evidence(Sugawara et al., 2015) relies too heavily on geneoverexpression, which results in accumulation of totalPAA. However, overexpression of a gene that normallyfunctions in a minor way can conceivably result in in-creased product levels, particularly when transcriptsare elevated 100-fold. Based on our loss-of-functionmutant data, it is unlikely that the very same TAA1or YUC enzymes play a role in both IAA and PAAbiosynthesis. Since other TAR1-like (pea) and YUC-like (maize) genes are only weakly expressed in therelevant tissues, we suggest that members of thesefamilies are not significantly involved in PAA bio-synthesis.

In order to explain the lack of effect of TAA1 andYUCmutants on PAA content, Sugawara et al. (2015) raisedthe possibility that quite different pathways might beinvolved in IAA and PAA synthesis. However, sincesequence analysis of related aromatic aminotransferaseproteins indicates enzymes with different functions inplants, we suggest that, while the biosynthetic path-ways are analogous, the enzymes involved in thesepathways belong to different gene families. We suggestthat in vivo, there is a considerable degree of specificityfor the aromatic aminotransferases, such that amino-transferases that utilize Trp do not use Phe to any greatextent.

MATERIALS AND METHODS

Plant Material

The pea (Pisum sativum) tar2-1 (previously tar2; Tivendale et al., 2012) mu-tant line was previously obtained through targeting-induced local lesions ingenomes (TILLING; Dalmais et al., 2008; Tivendale et al., 2012) followed by atleast eight generations of backcrossing against a cv Cameor background.Nonmutant studies were conducted on cv Torsdag unless otherwise stated. Themaize (Zea mays) de18 mutant and its wild-type background, cv B37, wereobtained from Bernardi et al. (2012).

Chemicals

The following standards were obtained from Sigma-Aldrich: 13C2phenylaceticacid, D5Trp, D5Phe, phenylpyruvate, phenylethanol, phenethylamine, phenyl-acetaldehyde, and phenylacetonitrile.

13C6 indole-3-acetic acid was obtained from Cambridge Isotope Laboratories,phenylaceamide was sourced through Santa Cruz Biology, and D2tryptophol andD4 4-Cl-indole-3-acetic acid were synthesized as previously described (Quittendenet al., 2009; Tivendale et al., 2012).

Synthesis of D5Phenylpyruvate

D6-Azlactone

Ceric ammoniumnitrate (55 g, 100.3mmol)wasdissolved indilute nitric acid(3.5 M,150 mL) and added to a solution of D8-toluene (2.5 g, 2.5 mmol) in dilutenitric acid (3.5 M, 20 mL). The reaction mixture was heated at 80°C for 2.5 hbefore being cooled to room temperature and extracted with CH2Cl2 (3 3

75 mL). The combined organic extracts were washed with water (3 3 75 mL),dried over MgSO4, filtered, and evaporated to yield the crude D6-benzaldehyde,which was added to acetyl-Gly (2.92 g, 25.4 mmol), sodium acetate (1.04 g, 12.7mmol, 0.5 eq.), and acetic anhydride (5 mL, 43.1 mmol, 1.7 eq.). The mixture wasrefluxed for 1h before being removed from the heat and placed in a freezer over-night before collecting the D6-azlactone by filtration (;50%).

D6-Acetoaminocinnamic Acid

D6-azlactone (136 mg, 0.71 mmol) was refluxed in water (8 mL) and acetone(20 mL) for 4 h before being cooled to room temperature. The acetone was re-moved under reduced pressure and the aqueous residue was extracted withethyl acetate (3 3 20 mL). The combined organic extracts were washed withsaturated NaCl (20 mL), dried, filtered ,and removed in vacuo to give the crudeproduct (91 mg, 0.44 mmol).

D5Phenylpyruvate

D5acetoaminocinnamic acid (91 mg, 0.44 mmol) was refluxed in HCl (1 M,10mL) for 3 h before being cooled to room temperature and extractedwith ethylacetate (2 3 20 mL). The combined extracts were dried over MgSO4 andevaporated under reduced pressure to give crude D5phenylpyruvate as redbrown crystals (55 mg, 0.33 mmol, 75%) 1H NMR (400 MHz) 4.22 (s, 3H), 8.5(BS, 1H), e/z 169 (M+, 20%), 123 (60), 120 (30), 118 (30), 110 (30), 96 (100), 87 (95),and 82 (55).

Hormone Extraction

Whole tissue in excess of 200 mgwas placed in 15 to 20 mL of 80%methanolwith 250 mg/L of butylated hydroxytoluene and homogenized using a Phys-cotron (Microtec). Alternatively, vegetative tissue or seeds weighing less than200 mgwere placed in 2-mLmicrocentrifuge tubes containing tungsten-carbidebeads (Qiagen) and homogenized for 2 min at full speed with a Tissuelyser(Qiagen). Samples were then suspended in 1 mL of either 80%methanol or 65%isopropanol (with butylated hydroxytoluene) and homogenized for a furtherminute. All homogenate was then stored overnight at 4°C to extract.

Quantification of Hormones and Precursors

Samples were centrifuged to pellet debris, and 1-mL aliquots from largersamples or 250-mL aliquots from microextractions were taken and internalstandards added. Organic solvent was removed using a sample concentratorand hormoneswere resuspended in 500mL of 1% acetic acid in deionizedwater.Samples were partitioned twice against 0.6 volumes of diethyl ether, and theorganic fractions were transferred to fresh tubes and evaporated under N2 gas.Hormones were once again suspended in 1% acetic acid in deionized water andcentrifuged to clear extractant. The resulting supernatant was transferredto auto-sample vials (Waters) and analyzed by ultraperformance liquidchromatography-mass spectrometry (UPLC-MS) multiple reaction monitoring(MRM) as described below. All analyses were conducted in triplicate (n = 3)with replicates comprising individual plants unless otherwise stated.

Metabolism Experiments

Pea seeds were sterilized briefly with 1% bleach and rinsed thoroughly withsteriledistilledwater. Seedswereplanted in sterile gravel:vermiculitemix (50:50)toppedwithvermiculite. SeedswerewatereddailywithMilli-Qgradewater andgrown at 25°C with a 16-h photoperiod. Six-day-old seedlings were transferredto 50-mL falcon tubes containing full-strength Murashige and Skoog medium(Sigma-Aldrich) supplemented with D5Phe or D5Trp at concentrations of 0mM,10 mM, 1 mM, 100 mM, or 10 mM. The falcon tubes were covered in black plasticto simulate dark, and the taproot of each seedling was inserted into a small5-mm opening, up to the hypocotyl. Seedlings were grown for a further


Cook et al.



7 d under the same conditions with frequent wetting of the collar of the hypocotylto prevent drying. Seedlings were removed from the solution, and excess liquidwas removed. Seedling fresh weight was recorded and hormones were extractedas above.m/z values corresponding to deuterated (D5) phenylpyruvate or D5PAAwere selected and their products analyzed following fragmentation. Thespectra obtained were compared with those from D5phenylpyruvate andD5PAA standards.

Preparation of Total Hormone Levels

To determine total auxin levels in extracts, tissuewas homogenized as aboveand extracted overnight. Internal standards were added to aliquots of extractsand taken to dryness with a sample concentrator. The first aliquot was preparedas above to determine free hormone levels. The second aliquot was taken up in3 mL of 7 N NaOH and incubated at 100°C for 3 h to hydrolyze ester and amidelinkages. After 3 h, the samples were cooled to room temperature and the pHwas reduced with 10 N HCl (pH 2.7). Samples were partitioned as above with0.6 volumes of diethyl ether. All samples were taken to dryness and resus-pended in 1% acetic acid in deionized water. To obtain conjugate/derivativelevels, levels of free auxins were deducted from total levels.

Determination of Indole-3-Acetaldehyde and 4-Cl-Indole-3-Acetaldehyde

Seeds from tar2-1 and TAR2 lines were ground in liquid nitrogen and 600mLof isopropanol was added. Following an overnight extraction, 2 and 10 ng ofinternal standard (D2tryptophol) was added to tryptophol and indole-3-acetaldehyde (IAAld) samples, respectively. IAAld samples were furthertreatedwith sodiumborohydride in 100mL of 0.3 NNaOH and incubated for 1 hat 35°C to stabilize IAAld as tryptophol. The pH of samples was reduced with 2mL of 10 N HCl, and isopropanol was evaporated under N2 gas. Both IAAld andtryptophol samples were purified by SPE using SepPak cartridges (Waters)preconditioned with methanol and 1% acetic acid. Samples were eluted in 60%methanol and taken to dryness by rotary evaporation followed by resuspensionin 1% acetic acid for analysis by UPLC-MS. Levels of IAAld were calculated bysubtracting endogenous tryptophol levels from tryptophol-stabilized IAAldlevels. The 4-Cl-tryptophol levels were determined by comparison with addedD2-tryptophol internal standard.

Stabilization of Phenylacetaldehyde

Volatileand labile intermediateswere stabilizedusing cysteamine toproduceTAZ-derived species of PAAld (PAAld-TAZ) and IPyA (IPyA-TAZ), based onNovák et al. (2012). Samples were homogenized as above but were extracted in50 mM sodium phosphate buffer (pH 7). Aliquots (500 mL) were taken andadded to 3 mL of 0.25 M cysteamine solution. Samples were incubated at roomtemperature for 1 h with agitation and were reduced to pH 2.7 with 10 N HCl.Samples were purified by SPE using SepPak cartridges (Waters) preconditionedwith 1 mL of methanol, 1 mL of deionized water, and 500 mL of sodiumphosphate buffer (pH 2.7). The cartridges were washed with 5% methanol andeluted into round bottom flasks in 80% methanol. The eluate was taken todryness by rotary evaporation and resuspended in 1% acetic acid as above.

Expression and Purification of Functional Proteins

Previously isolated PsTAR1 sequence was amplified from pea apex cDNAusing primers containing a 59 CACC tag to facilitate directional cloning into apENTR TOPO vector (Invitrogen). PsTAR1-containing constructs were verifiedby sequencing (Macrogen), and the gene was transferred into a pET-53-DESTvector using an LR clonase II reaction (Novagen) to produce PsTAR1-DEST.The expression construct was transformed into NOVA F-cells for selection,followed by transformation into the Escherichia coli strain BL21 (DE3) for ex-pression. Culture and harvest of cells was conducted as described (Dai et al.,2013) followed by purification using Ni-NTA for metal chelation chromatog-raphy. Primer sequences are provided in Supplemental Table S2.

Quantitative Real-Time PCR

Mature green seeds of TAR2 and tar2-1were harvested, and RNA extractedas before (Tivendale et al., 2012). Four replicates were taken for each genotype,

each comprising one seed. cDNA was synthesized using random hexamers(Tetro cDNA synthesis kit; Bioline). First-strand cDNA was diluted 20-fold,with 1mL used per 10mL quantitative real-time PCR reaction using SYBRGreenchemistry (SensiFAST SYBR No-ROX kit; Bioline). Samples were set up with aCAS-1200N robotic liquid-handling system (Corbett Research) and run for 50 or55 cycles in a Rotor-Gene RG3000A dual-channel machine (Qiagen). Threetechnical replicates were performed for each biological replicate, and the con-centrations were calculated relative to a curve containing seven serial dilutions(1/10) of original pooled cDNA taken equally from each replicate. Reactionefficiencies and correlation coefficients were as before (Tivendale et al., 2012).Pea 18s rRNA levels were utilized as before (Ozga et al., 2003), in place of ahousekeeper gene. Primers used to amplify TAR1 (JN990988.1) and TAR3(JN990990.1) were qPsTAR-Mt5g90 F92, qPsTAR-Mt5g90 r93 (TAR1), andqPsTAR3 F85, qPsTAR3 r85 (TAR3), respectively (Supplemental Table S2).

Enzyme Assay

Assays were conducted in 100 mL of 50 mM sodium phosphate buffer (pH 7)containing 4 mM PsTAR1 enzyme, 50 mM substrate, and 1 mM PLP and 50 mM

sodium pyruvate as cofactors. Assays were incubated for 1 h at 37°C, and re-actions were terminated with the addition of 5% acetic acid in methanol.

UPLC-MS

Samples were analyzed as previously (Tivendale et al., 2012) with a solventcombination of 1% (v/v) acetic acid in water (solvent A) and acetonitrile (sol-vent B). The modified UPLC programwas 95% A, 5% B to 50% A, and 50% B at4.5 min, followed by immediate reequilibration to starting conditions for 3 min.The flow rate was 0.35 mL min21 with the column held at 35°C, and the samplecompartment was at 6°C.

The mass spectrometer was operated in positive and negative ion electro-spray mode with a needle voltage of 2.8 kV, and MRM was used to detect allanalytes (Supplemental Table S3). The ion source temperature was 130°C, the des-olvation gas was N2 at 950 L h21, the cone gas flow was 100 L h21, and the des-olvation temperature was 450°C. Data were processed using MassLynx software.

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL datalibraries under accession numbers PsArAT (KX236168), PsTAR1 (AFG31373),PsTAR2 (AFG31321), PsTAR3 (AFG31374), PhPPY-AT (AHA62827), CmArAT(NP_001284465), TcTAT (P33447), ZmYUC1 (NP_001105991), ZmYUC2(NP_001147961), ZmYUC3 (AFW60633), ZmSpi1 (DAA54585), GRMZM2G019515(AFW79552), GRMZM2G017193 (AFW83812), and GRMZM2G011622 (DAA47280).

Supplemental Data

The following supplemental materials are available.

Supplemental Figure S1. Product scan spectra of D5phenylpyruvate and PAA.

Supplemental Figure S2. Potential intermediates in PAA biosynthesis.

Supplemental Figure S3. Metabolism of D5Trp to D5IPyA.

Supplemental Figure S4. Expression levels of TARs in tar2-1mutant seeds.

Supplemental Figure S5. Effects of tar2-1 mutation on levels of (4Cl-)indole-3-acetaldehyde.

Supplemental Figure S6.Multiple sequence alignment of ArAT sequences.

Supplemental Table S1. Relative expression of YUC-like sequences inmaize endosperm.

Supplemental Table S2. Gateway and qPCR primer sequences.

Supplemental Table S3. MRM transition data for all observed analytes.

ACKNOWLEDGMENTS

We thank Noel Davies for his UPLC-MS expertise, Michelle Lang andTracey Winterbottom for their assistance with glasshouse operations, and theAustralian Research Council for financial support.

Received April 5, 2016; accepted April 21, 2016; published April 26, 2016.

















LITERATURE CITED

Abe H, Uchiyama M, Sato R (1974) Isolation of phenylacetic acid and itsp-hydroxy derivative as auxin-like substances from Undaria pinnatifida.AgricBiolChem 38:897–898

Alves-Carvalho S, Aubert G, Carrère S, Cruaud C, Brochot AL, Jacquin F,Klein A, Martin C, Boucherot K, Kreplak J, et al (2015) Full-length denovo assembly of RNA-seq data in pea (Pisum sativum L.) provides agene expression atlas and gives insights into root nodulation in thisspecies. Plant J 84:1–19

Bedewitz MA, Góngora-Castillo E, Uebler JB, Gonzales-Vigil E, Wiegert-Rininger KE, Childs KL, Hamilton JP, Vaillancourt B, Yeo Y-S,Chappell J, et al (2014) A root-expressed L-phenylalanine:4-hydroxyphenylpyruvate aminotransferase is required for tropane alka-loid biosynthesis in Atropa belladonna. Plant Cell 26:3745–3762

Bernardi J, Lanubile A, Li Q-B, Kumar D, Kladnik A, Cook SD, Ross JJ,Marocco A, Chourey PS (2012) Impaired auxin biosynthesis in the de-fective endosperm18 mutant is due to mutational loss of expression in theZmYuc1 gene encoding endosperm-specific YUCCA1 protein in maize.Plant Physiol 160:1318–1328

Blankenfeldt W, Nowicki C, Montemartini-Kalisz M, Kalisz HM, HechtHJ (1999) Crystal structure of Trypanosoma cruzi tyrosine aminotrans-ferase: substrate specificity is influenced by cofactor binding mode.Protein Sci 8:2406–2417

Boatright J, Negre F, Chen X, Kish CM, Wood B, Peel G, Orlova I, GangD, Rhodes D, Dudareva N (2004) Understanding in vivobenzenoidmetabolism in petunia petal tissue. Plant Physiol 135:1993–2011

Cheng Y, Dai X, Zhao Y (2007) Auxin synthesized by the YUCCA fla-vinmonooxygenases is essential for embryogenesis and leaf formation inArabidopsis. Plant Cell 19:2430–2439

Cho M-H, Corea OR, Yang H, Bedgar DL, Laskar DD, Anterola AM,Moog-Anterola FA, Hood RL, Kohalmi SE, Bernards MA, et al (2007)Phenylalanine biosynthesis in Arabidopsis thaliana. Identification andcharacterization of arogenate dehydratases. J BiolChem 282:30827–30835

Chourey PS, Li Q-B, Kumar D (2010) Sugar-hormone cross-talk in seeddevelopment: two redundant pathways of IAA biosynthesis are regu-lated differentially in the invertase-deficient miniature1 (mn1) seed mu-tant in maize. Mol Plant 3:1026–1036

Cooney TP, Nonhebel HM (1991) Biosynthesis of indole-3-acetic acid intomato shoots: Measurement, mass-spectral identification and incorpo-ration of22H from22H2O into indole-3-acetic acid, d-and l-tryptophan,indole-3-pyruvate and tryptamine. Planta 184:368–376

Dai X, Mashiguchi K, Chen Q, Kasahara H, Kamiya Y, Ojha S, DuBois J,Ballou D, Zhao Y (2013) The biochemical mechanism of auxin biosyn-thesis by an Arabidopsis YUCCA flavin-containing monooxygenase.J BiolChem 288:1448–1457

Dalmais M, Schmidt J, Le Signor C, Moussy F, Burstin J, Savois V,Aubert G, Brunaud V, de Oliveira Y, Guichard C, Thompson R,Bendahmane A (2008) UTILLdb, a Pisum sativum in silico forward andreverse genetics tool. Genome Biol 9:R43

Epstein E, Cohen JD, Slovin JP (2002) The biosynthetic pathway for indole-3-acetic acid changes during tomato fruit development. Plant GrowthRegul 38:15–20

Gallavotti A, Barazesh S, Malcomber S, Hall D, Jackson D, Schmidt RJ,McSteen P (2008) sparse inflorescence1 encodes a monocot-specificYUCCA-like gene required for vegetative and reproductive develop-ment in maize. ProcNatlAcadSci USA 105:15196–15201

Gonda I, Bar E, Portnoy V, Lev S, Burger J, Schaffer AA, Tadmor Y,Gepstein S, Giovannoni JJ, Katzir N, Lewinsohn E (2010) Branched-chain and aromatic amino acid catabolism into aroma volatiles inCucumismelo L. fruit. J Exp Bot 61:1111–1123

Groot MNN, de Bont JA, Nierop Groot MN, de Bont JAM (1998) Conver-sion of phenylalanine to benzaldehyde initiated by an aminotransferasein Lactobacillus plantarum. Appl Environ Microbiol 64:3009–3013

Hermes JD, Tipton PA, Fisher MA, O’Leary MH, Morrison JF, ClelandWW (1984) Mechanisms of enzymatic and acid-catalyzed decarboxyla-tions of prephenate. Biochemistry 23:6263–6275

Herrmann KM, Weaver LM (1999) Theshikimate pathway. Annu Rev PlantPhysiol Plant MolBiol 50:473–503

Hwang BK, Lim SW, Kim BS, Lee JY, Moon SS (2001) Isolation and in vivoand in vitro antifungal activity of phenylacetic acid and sodium phe-nylacetate from Streptomyces humidus. Appl Environ Microbiol 67:3739–3745

Jensen RA, Gu W (1996) Evolutionary recruitment of biochemically spe-cialized subdivisions of Family I within the protein superfamily ofaminotransferases. J Bacteriol 178:2161–2171

Kim Y, Cho J-Y, Kuk J-H, Moon J-H, Cho J-I, Kim Y-C, Park K-H (2004)Identification and antimicrobial activity of phenylacetic acid producedby Bacillus licheniformis isolated from fermented soybean, Chungkook-Jang. CurrMicrobiol 48:312–317

Kishore G, Sugumaran M, Vaidyanathan CS (1976) Metabolism ofDL-(+/-)-phenylalanine by Aspergillusniger. J Bacteriol 128:182–191

Krings U, Hinz M, Berger RG (1996) Degradation of [2H] phenylalanine bythe basidiomycete Ischnoderma benzoinum. J Biotechnol 51:123–129

Lee AY, Stewart JD, Clardy J, Ganem B (1995) New insight into the cat-alytic mechanism of chorismatemutases from structural studies.ChemBiol 2:195–203

Liu H, Ying Y-Y, Zhang L, Gao Q-H, Li J, Zhang Z, Fang J-G, Duan K(2012) Isolation and characterization of two YUCCAflavinmonooxygenasegenes from cultivated strawberry (Fragaria 3 ananassa Duch.). Plant Cell Rep31:1425–1435

Ludwig-Müller J, Cohen JD (2002) Identification and quantification ofthree active auxins in different tissues of Tropaeolum majus. Physiol Plant115:320–329

Maeda H, Shasany AK, Schnepp J, Orlova I, Taguchi G, Cooper BR,Rhodes D, Pichersky E, Dudareva N (2010) RNAi suppression of Ar-ogenate Dehydratase1 reveals that phenylalanine is synthesized predom-inantly via the arogenate pathway in petunia petals. Plant Cell 22:832–849

Maeda H, Yoo H, Dudareva N (2011) Prephenate aminotransferase directsplant phenylalanine biosynthesis via arogenate. Nat ChemBiol 7:19–21

Manela N, Oliva M, Ovadia R, Sikron-Persi N, Ayenew B, Fait A, GaliliG, Perl A, Weiss D, Oren-Shamir M (2015) Phenylalanine and tyrosinelevels are rate-limiting factors in production of health promoting me-tabolites in Vitis vinifera cv. Gamay Red cell suspension. Front Plant Sci6:538

Mashiguchi K, Tanaka K, Sakai T, Sugawara S, Kawaide H, Natsume M,Hanada A, Yaeno T, Shirasu K, Yao H, et al (2011) The main auxinbiosynthesis pathway in Arabidopsis. ProcNatlAcadSci USA 108:18512–18517

Michalczuk L, Ribnicky DM, Cooke TJ, Cohen JD (1992) Regulation ofindole-3-acetic Acid biosynthetic pathways in carrot cell cultures. PlantPhysiol 100:1346–1353

Nonhebel HM (2015) Tryptophan-independent indole-3-acetic acid syn-thesis: critical evaluation of the evidence. Plant Physiol 169:1001–1005

Novák O, Hényková E, Sairanen I, Kowalczyk M, Pospíšil T, Ljung K(2012) Tissue-specific profiling of the Arabidopsis thaliana auxin metab-olome. Plant J 72:523–536

Ozga JA, Yu J, Reinecke DM (2003) Pollination-, development-, and auxin-specific regulation of gibberellin 3b-hydroxylase gene expression in peafruit and seeds. Plant Physiol 131:1137–1146

Phillips KA, Skirpan AL, Liu X, Christensen A, Slewinski TL, Hudson C,Barazesh S, Cohen JD, Malcomber S, McSteen P (2011) vanishingtassel2 encodes a grass-specific tryptophan aminotransferase requiredfor vegetative and reproductive development in maize. Plant Cell 23:550–566

Quittenden LJ, Davies NW, Smith JA, Molesworth PP, Tivendale ND,Ross JJ (2009) Auxin biosynthesis in pea: characterization of the trypt-amine pathway. Plant Physiol 151:1130–1138

Rapparini F, Cohen JD, Slovin JP (1999) Indole-3-acetic acid biosynthesisin Lemna gibba studied using stable isotope labeled anthranilate andtryptophan. Plant Growth Regul 27:139–144

Schneider EA, Kazakoff CW, Wightman F (1985) Gas chromatography-mass spectrometry evidence for several endogenous auxins in peaseedling organs. Planta 165:232–241

Schneider EA, Wightman F (1986) Auxins of non-flowering plants. I. Oc-currence of 3-indoleacetic acid and phenylacetic acid in vegetative andfertile fronds of the ostrich fern (Matteucias truthiopteris). Physiol Plant68:396–402

Sekhon RS, Lin H, Childs KL, Hansey CN, Buell CR, de Leon N,Kaeppler SM (2011) Genome-wide atlas of transcription during maizedevelopment. Plant J 66:553–563

Siehl DL, Connelly JA, Conn EE (1986) Tyrosine biosynthesis in Sorghumbicolor: characteristics of prephenate aminotransferase. Z Naturforsch C41:79–86


Cook et al.



Somers E, Ptacek D, Gysegom P, Srinivasan M, Vanderleyden J (2005)Azospirillum brasilense produces the auxin-like phenylacetic acid by us-ing the key enzyme for indole-3-acetic acid biosynthesis. Appl EnvironMicrobiol 71:1803–1810

Spaepen S, Versées W, Gocke D, Pohl M, Steyaert J, Vanderleyden J(2007) Characterization of phenylpyruvate decarboxylase, involved inauxin production of Azospirillum brasilense. J Bacteriol 189:7626–7633

Sugawara S, Hishiyama S, Jikumaru Y, Hanada A, Nishimura T, KoshibaT, Zhao Y, Kamiya Y, Kasahara H (2009) Biochemical analyses ofindole-3-acetaldoxime-dependent auxin biosynthesis in Arabidopsis.ProcNatlAcadSci USA 106:5430–5435

Sugawara S, Mashiguchi K, Tanaka K, Hishiyama S, Sakai T, Hanada K,Kinoshita-Tsujimura K, Yu H, Dai X, Takebayashi Y, et al (2015)Distinct characteristics of indole-3-acetic acid and phenylacetic acid, twocommon auxins in plants. Plant Cell Physiol 56:1641–1654

Sztein AE, Ili�c N, Cohen JD, Cooke TJ (2002) Indole-3-acetic acid bio-synthesis in isolated axes from germinating bean seeds: The effect ofwounding on the biosynthetic pathway. Plant Growth Regul 36:201–207

Tam YY, Normanly J (1998) Determination of indole-3-pyruvic acid levelsin Arabidopsis thaliana by gas chromatography-selected ion monitoring-mass spectrometry. J Chromatogr A 800:101–108

Tao Y, Ferrer J-L, Ljung K, Pojer F, Hong F, Long JA, Li L, Moreno JE,Bowman ME, Ivans LJ, et al (2008) Rapid synthesis of auxin via a newtryptophan-dependent pathway is required for shade avoidance inplants. Cell 133:164–176

Taylor DC, Wightman F (1987) Metabolism of D,L-chloro-phenylalaninesby phenylalanine aminotransferase isozymes purified from bushbeanshoots. Phytochemistry 26:1279–1288

Tieman D, Taylor M, Schauer N, Fernie AR, Hanson AD, Klee HJ (2006)Tomato aromatic amino acid decarboxylases participate in synthesis ofthe flavor volatiles 2-phenylethanol and 2-phenylacetaldehyde. Proc-NatlAcadSci USA 103:8287–8292

Tivendale ND, Davidson SE, Davies NW, Smith JA, Dalmais M,Bendahmane AI, Quittenden LJ, Sutton L, Bala RK, Le Signor C, et al(2012) Biosynthesis of the halogenated auxin, 4-chloroindole-3-aceticacid. Plant Physiol 159:1055–1063

Tivendale ND, Ross JJ, Cohen JD (2014) The shifting paradigms of auxinbiosynthesis. Trends Plant Sci 19:44–51

Tzin V, Malitsky S, Aharoni A, Galili G (2009) Expression of a bacterial bi-functional chorismatemutase/prephenate dehydratase modulates pri-mary and secondary metabolism associated with aromatic amino acidsin Arabidopsis. Plant J 60:156–167

Wang B, Chu J, Yu T, Xu Q, Sun X, Yuan J, Xiong G, Wang G, Wang Y, Li J(2015) Tryptophan-independent auxin biosynthesis contributes to earlyembryogenesis in Arabidopsis. ProcNatlAcadSci USA 112:4821–4826

Wightman F, Lighty DL (1982) Identification of phenylacetic acid as anatural auxin in the shoots of higher plants. Physiol Plant 55:17–24

Wightman F, Schneider EA, Thimann KV (1980) Hormonal factors con-trolling the initiation and development of lateral roots. Physiol Plant 49:304–314

Won C, Shen X, Mashiguchi K, Zheng Z, Dai X, Cheng Y, Kasahara H,Kamiya Y, Chory J, Zhao Y (2011) Conversion of tryptophan to indole-3-acetic acid by TRYPTOPHAN AMINOTRANSFERASES OF ARABIDOPSISand YUCCAs in Arabidopsis. ProcNatlAcadSci USA 108:18518–18523

Yamada T, Matsuda F, Kasai K, Fukuoka S, Kitamura K, Tozawa Y,Miyagawa H, Wakasa K (2008) Mutation of a rice gene encoding aphenylalanine biosynthetic enzyme results in accumulation of phenyl-alanine and tryptophan. Plant Cell 20:1316–1329

Yoo H, Widhalm JR, Qian Y, Maeda H, Cooper BR, Jannasch AS, Gonda I,Lewinsohn E, Rhodes D, Dudareva N (2013) An alternative pathwaycontributes to phenylalanine biosynthesis in plants via a cytosolic ty-rosine:phenylpyruvate aminotransferase. Nat Commun 4:2833

Zhao Y (2012) Auxin biosynthesis: a simple two-step pathway convertstryptophan to indole-3-acetic acid in plants. Mol Plant 5:334–338





auxin biosynthesis: are the indole-3-acetic acid and ... · auxin biosynthesis: are the...

Documents