Glucosinolate biosynthetic genes in Brassica rapa

Gene 487 (2011) 135–142

Contents lists available at ScienceDirect

Gene

j ourna l homepage: www.e lsev ie r.com/ locate /gene

Glucosinolate biosynthetic genes in Brassica rapa

Hui Wang, Jian Wu, Silong Sun, Bo Liu, Feng Cheng, Rifei Sun, Xiaowu Wang ⁎Key Laboratory of Horticultural Crop Genetic Improvement, MOA, PR ChinaSino-Dutch Joint Lab of Horticultural Genomics Technology, PR ChinaInstitute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences, PR China

Abbreviations: GS, glucosinolate; CYP, cytochromeylmalate synthase; BCAT, branched-chain aminotransfe⁎ Corresponding author at: Institute of Vegetables and

Agricultural Sciences, Zhongguancun Nandajie No. 12, HPR China. Tel.: +86 10 82105971; fax: +86 10 621741

E-mail addresses: [email protected] (H. Wang), [email protected] (S. Sun), [email protected] (B. Liu), c(F. Cheng), [email protected] (R. Sun), wangxw@ma

0378-1119/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.gene.2011.07.021

a b s t r a c t

Article history:Accepted 15 July 2011Available online 30 July 2011

Received by Bronya Keats

Keywords:Secondary metabolismBiosynthesis pathwayComparative genomicsCruciferae

Glucosinolates (GS) are a group of amino acid-derived secondarymetabolites found throughout the Cruciferaefamily. Glucosinolates and their degradation products play important roles in pathogen and insectinteractions, as well as in human health. In order to elucidate the glucosinolate biosynthetic pathway inBrassica rapa, we conducted comparative genomic analyses of Arabidopsis thaliana and B. rapa on a genome-wide level. We identified 102 putative genes in B. rapa as the orthologs of 52 GS genes in A. thaliana. All butone gene was successfully mapped on 10 chromosomes. Most GS genes exist in more than one copy in B. rapa.A high co-linearity in the glucosinolate biosynthetic pathway between A. thaliana and B. rapa was alsoestablished. The homologous GS genes in B. rapa and A. thaliana share 59–91% nucleotide sequence identityand 93% of the GS genes exhibit synteny between B. rapa and A. thaliana. Moreover, the structure andarrangement of the B. rapa GS (BrGS) genes correspond with the known evolutionary divergence of B. rapa,and may help explain the profiles and accumulation of GS in B. rapa.

s P450; MAM, methylthioalk-rase.Flowers, Chinese Academy of

aidian district, Beijing, 100081,[email protected] (J. Wu),[email protected] (X. Wang).

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Glucosinolates (GS) are sulfur-rich plant secondary metabolitesderived from amino acids and sugars, whose hydrolysis productscontribute to the special flavors and tastes of Brassica vegetables(Padilla et al., 2007; Schonhof et al., 2004). When crucifer tissue isdamaged, GS are hydrolyzed bymyrosinase into degradation productssuch as isothiocyanate, nitrile, thiocyanates, cyano-epithioalkanes andoxazolidine-2-thiones, which have different bioactivities (Bones andRossiter, 2006; Rask et al., 2000; Wittstock et al., 2003). Degradationproducts of GS are thought to inhibit carcinogenesis by effecting cellcycle arrest and stimulating apoptosis (Hayes et al., 2008). Sulforaph-ane, the isothiocyanate derived from glucoraphanin (GRA), exhibitsstrong anticarcinogenic properties (Keck and Finley, 2004; Zhang etal., 1994). Indole-3-carbinol, a derivative of glucobrassicin, also hasanticarcinogenic properties (Choi et al., 2010). In addition, phenethylisothiocyanate can block the conversion of several carcinogens totheir carcinogenic forms (Hecht, 2000; Nakajima et al., 2001).

Glucosinolates can be classified into three major groups, namelyaliphatic, aromatic and indole GS, based on their precursor amino acids(Halkier and Gershenzon, 2006; Sønderby et al., 2010). Glucosinolate

biosynthesis generally involves three stages: (i) amino acid chainelongation; (ii) core structure formation; and (iii) secondary modifica-tion of initial GS (Sønderby et al., 2010). Methylthioalkylmalatesynthase (MAM), bile acid:sodium symporter family protein 5(BASS5), and branched-chain aminotransferase (BCAT) are involved inelongation of methionine (Gigolashvili et al., 2009; Kroymann et al.,2001; Sawada et al., 2009a; Textor et al., 2007). Core structure formationof GS is accomplished in five steps via oxidation by cytochromes P450 ofCYP79 and CYP83, followed by C-S lyase, S-glucosyltransferase andsulfortransferase (Brader et al., 2006; Grubb and Abel, 2006; Wittstockand Halkier, 2002). Subsequently, several loci such as GS-OX, GS-AOP,GS-OH, BZO1 and CYP81F2 participate in secondary modification(Sønderby et al., 2010). In addition, some nuclear-localized regulatorsand R2R3-Myb transcription factors take part in glucosinolate biosyn-thesis (Celenza et al., 2005; Gigolashvili et al., 2007a, 2007b;Gigolashviliet al., 2008; Levy et al., 2005; Skirycz et al., 2006; Sønderby et al., 2007).Several other genes also participate in co-substrate formation steps(Sønderby et al., 2010).

Chinese cabbage, broccoli, cabbage, cauliflower and other crucif-erous vegetables are rich in GS (Kushad et al., 1999; Padilla et al.,2007). Increasing daily intake of cruciferous vegetables is associatedwith reduced risk of certain cancers (Keck and Finley, 2004; Kim andPark, 2009; Tang et al., 2008; Wu et al., 2004). Chinese cabbage (B.rapa ssp. pekinensis) is one of the most important Brassica vegetables,especially in Asia. Brassica rapa comprises a variety of vegetables, ofwhich Chinese cabbage and pak choi are the most highly consumedvegetables in China and throughout eastern Asia.

Both Chinese cabbage and A. thaliana belong to the Cruciferaefamily. In recent decades, most of the A. thaliana genes involved in the

136 H. Wang et al. / Gene 487 (2011) 135–142

glucosinolate biosynthetic pathway have been identified and theirorthologs in B. rapa have been analyzed in comparative studies of A.thaliana and B. rapa based on sequencing of expressed sequence tags(ESTs) and bacterial artificial chromosome (BAC) libraries (Zang et al.,2009). However, because the genome sequence used in the earlierstudy covered only 40% of the euchromatic DNA of B. rapa, only 56genes were identified for the glucosinolate biosynthetic pathway in B.rapa. To overcome this limitation and to provide more comprehensiveinformation for the glucosinolate biosynthetic pathway in B. rapa, weperformed comparative genomic analysis between B. rapa and A.thaliana using the assembled genome sequence of B. rapa (Wang et al.,submitted for publication). We present several interesting observa-tions, including similarities of the glucosinolate biosynthetic path-ways between A. thaliana and B. rapa, multiple copies of BrGS genes,synteny of GS biosynthetic genes between A. thaliana and B. rapa, andexplanation of GS profiles and accumulation in B. rapa by BrGS genes.

2. Materials and methods

2.1. Database for glucosinolate biosynthetic genes identification in B.rapa

Sequences representing the complete set of glucosinolate biosyn-thetic genes in A. thaliana were acquired from the TAIR database(www.arabidopsis.org), and data for B. rapa ssp. pekinensis cv. Chiifugenome V1.0 and the set of annotated genes (http://Brassicadb.org)was searched for homologous genes. Estimation of the number ofBrGS genes in the genome of B. rapawas conducted by analysis of ESTdata for B. rapa downloaded from the NCBI EST database (accessed 10Jul. 2010) and mRNA sequencing data of B. rapa (unpublished data).Full length AOP protein sequences of 13 A. thaliana genes (includingAOP1, AOP2 and AOP3), one A. lyrata, two B. rapa, one B. oleracea, andfour B. juncea genes were retrieved from the TAIR database (www.arabidopsis.org) and the UniProt protein database (www.uniprot.org).

2.2. Homologous gene identification and analysis

We identified B. rapa homologous genes using BLASTN and BLASTPwith a cutoff E-valueb=1E−10 and coverageN=0.75, and we alsoestimated the number of BrGS genes expressed in the genome usingBLASTN with a cutoff E-valueb=1E−10 and coverageN=0.75. Weconducted synteny analysis using the CoGe comparative genomicssystem. Each synteny gene-pair was identified by not only theirhomozygosity with BLASTP (E-valueb=1E−20) but also the colin-earity of their flanking genes. A phylogenetic tree was constructedusing the neighbor-joining method with Mega 4.0 software (Tamuraet al., 2007), and bootstrap values with 500 replicates were calculated.Identity analyses were performed using the Needle program from theEMBOSS software package (http://emboss.sourceforge.net/).

3. Results and discussion

3.1. Glucosinolate biosynthetic gene identification from draft B. rapagenome

On the basis of the draft B. rapa genome v1.0 and 41,174 annotatedgenes (Wang et al., submitted for publication), 52 homologous A.thaliana glucosinolate (AtGS) biosynthetic genes were identifiedusing BLASTN and BLASTP. In total, 102 B. rapa glucosinolate (BrGS)biosynthetic candidate genes were identified as homologs of 44 of the52 known AtGS genes and a further eight AtGS genes showed no B.rapa orthologs (Table 1). The EST data and mRNA sequencing data forB. rapa supported the evidence that all BrGS genes were expressedexcept for two ST5b copies, Bra003726 and Bra027117 (Wang et al.,submitted for publication). Of the 102 BrGS genes, 101 were mapped

on the 10 chromosomes of B. rapa, with seven, 14, 21, 11, five, four, 10,nine, 16, four BrGS genes anchored on chromosomes A01–A10,respectively (Fig. 1). Another gene, Bra039818, which is an ortholog ofAPK1, was anchored on Scaffold000178, which has not yet beenmapped onto a chromosome. Based on the presence of thesehomologous genes, the glucosinolate biosynthetic pathway in B.rapa is clearly established.

Genes from different parts of the GS biosynthesis pathways sharedifferent sequence identities with their counterparts in A. thaliana. Ingeneral, BrGS biosynthetic genes shared 59–91% nucleotide sequenceidentity with their At orthologs (Table 1). Homologous gene pairs thatparticipate in regulation, amino acid chain elongation, and secondarymodification of initial glucosinolate pathway precursors sharesignificantly lower identity (79.9%) than those participating in corestructure formation (84.3%) at pb0.01. Genes involved in formation ofglucosinolate core structure are more highly conserved, sinceexcessive mutation of these genes would likely block biosynthesis ofGS. Therefore, strong sequence conservation is consistent with theirkey role in the glucosinolate biosynthesis pathway.

Zang et al. (2009) previously identified glucosinolate synthesisgenes in B. rapa. Although these authors carried out an exhaustivesearch, they did not identify all of the counterpart genes, but only87.5%, on the basis of cDNA and BAC libraries. In addition, 21.5% of thegenes found were partial CDS sequences, with many BrGS genesanchored only on the BAC, rather than on chromosomes. Here weidentified glucosinolate biosynthetic genes over the whole genome, indetail, by comparative genomic analysis between B. rapa and A.thaliana. In addition, we acquired data regarding a more intact set ofglucosinolate biosynthetic genes in B. rapa, which will pave the wayfor further improvement of agronomic traits via genetic engineeringof B. rapa.

3.2. Glucosinolate biosynthetic genes with no orthologs identified in B.rapa

No orthologs in B. rapa were observed for eight AtGS genescomprising a transcription factor (MYB76), two amino acid side chainelongation genes (IPMDH3 and IPMI SSU3), one core structureformation gene (CYP79F2) for long-chain aliphatic GS, and four sidechain modification genes (FMOGS-OX1, FMOGS-OX3, FMOGS-OX4 andAOP3). These genes may be specific to A. thaliana and their absencedoes not block the glucosinolate biosynthetic pathway in B. rapa.

MYB76 is a transcription factor involved in aliphatic glucosinolatebiosynthesis. It was reported that MYB76 over-expression linescontained increased levels of both short-chained and long-chainedaliphatic GS, and these increases correlated well with the MYB76expression level (Gigolashvili et al., 2008). However, a MYB76 loss-of-function mutant showed no significant change in glucosinolatecontent in A. thaliana (Gigolashvili et al., 2008). Therefore, MYB76 isnot indispensable for GS biosynthesis, so absence of MYB76 in B. rapais not fatal for GS biosynthesis.

IPMDH1 was identified based on strong co-expression withglucosinolate biosynthetic genes and prior knowledge of Leu biosyn-thesis, and an ipmdh1 knockout mutant showed a decrease inglucosinolate accumulation (He et al., 2009; Hirai et al., 2007; Sawada,et al., 2009b). IPMDH3 is a predicted enzyme that shares a similarfunction with IPMDH1 (Gigolashvili et al., 2009; Wentzell et al., 2007).Thus the absence of IPMDH3 in B. rapa, though predicted by genomicanalysis of A. thaliana, has no apparent effect on side chain elongation.IPMI is another locus indicated to be important for side chain elongationof methionine. IPMI LSU3 is a partially characterized enzyme, but IPMILSU1 and IPMI LSU2 may offset the loss of IPMI SSU3 (Sønderby et al.,2010). Therefore, the absence of IPMI LSU3 in B. rapa might not affectisomerization of the IPMI substrate 2-malate derivative.

The CYP79 family catalyzes the conversion of amino acids tooximes in GS biosynthesis. CYP79F1 metabolizes all chain-elongated

Table 1Gene inventory of the glucosinolate pathway and B. rapa orthologs.

Group name AGI BrIDa Blocksb Identity (%)c References

Transcription factorsDof1.1 At1g07640 Bra031588 A (A09) 76.8 (81.3) Skirycz et al. (2006)

Bra030696 A (A08) 79.9IQD1-1 At3g09710 Bra034081 F (A01) 76.1 (70.9) Levy et al. (2005)

Bra001299 F (A03) 70.8MYB28 At5g61420 Bra012961 X (A03) 79.1 (76.3–82.7) Gigolashvili et al. (2007b); Hirai et al. (2007)

Bra035929 X (A09) 81.5Bra029311 X (A02) 72.5

MYB29 At5g07690 Bra009245 R (A10) 92.1 (93.2) Gigolashvili et al. (2008); Hirai et al. (2007)Bra005949 R (A03) 75.7

MYB34 At5g60890 Bra013000 X (A03) 79.4 (79.4–83.9) Celenza et al. (2005)Bra035954 X (A09) 81.3Bra029350 X (A02) 79.7Bra029349 X (A02) 79.7

MYB51 At1g18570 Bra025666 A (A06) 78.5 (81.4–85.4) Gigolashvili et al. (2007a)Bra031035 A (A09) 77.5Bra016553 A (A08) 79.7

MYB76 At5g07700 Gigolashvili et al. (2008)MYB122 At1g74080 Bra015939 E (A07) 80.6 (84.9) Gigolashvili et al. (2007a)

Bra008131 E (A02) 82.3

Side-chain elongationBCAT-4 At3g19710 Bra022448 F (A05) 83.2 (93.0) Schuster et al. (2006)

Bra001761 F (A03) 83.0BAT5 At4g12030 Bra029434 P (A09) 85.7 (89.0) Gigolashvili et al. (2009); Sawada et al. (2009a)

Bra000760 P (A03) 88.5MAM1 At5g23010 Bra013007 Q (A03) 74.1 (72.4–75.6) Field et al. (2004); Kroymann et al. (2001)

Bra029355 Q (A02) 79.1Bra018524* O (A02) 81.3

MAM3 At5g23020 Bra013009 Q (A03) 82.8 (61.4–87.7) Field et al. (2004); Textor et al. (2007)Bra013011 Q (A03) 79.1Bra029356 Q (A02) 67.8Bra021947* J (A04) 74.3

IPMI LSU1 At4g13430 Bra032708 T (A04) 89.7 (92.3) He et al. (2010); Hirai et al. (2007);Knill et al. (2009); Sawada et al. (2009b);Textor et al. (2004)

Bra040341 T (A08) 89.5IPMI SSU2 At2g43100 Bra004744 J (A05) 82.8 Beekwilder et al. (2008); Gigolashvili et al. (2009);

Knill et al. (2009); Sawada et al. (2009b);Wentzell et al. (2007)

IPMI SSU3 At3g58990 Gigolashvili et al. (2009); Knill et al. (2009);Sawada et al. (2009b); Wentzell et al. (2007)

IPMDH1 At5g14200 Bra023450 R (A02) 89.5 He et al. (2009); Sawada et al. (2009b)IPMDH3 At1g31180 Gigolashvili et al. (2009); Wentzell et al. (2007)BCAT-3 At3g49680 Bra017964 M (A06) 85.3 (88.2) Knill et al. (2008)

Bra029966 M (A01) 86.1

Core structure formationCYP79F1 At1g16410 Bra026058 A (A06) 82.7 Chen et al. (2003); Hansen et al. (2001b);

Reintanz et al. (2001)CYP79F2 At1g16400 Chen et al. (2003); Hansen et al. (2001b);

Reintanz et al. (2001)CYP79A2 At5g05260 Bra009100 R (A10) 84.4 (87.4) Wittstock and Halkier (2000)

Bra028764 R (A02) 85.3CYP79B2 At4g39950 Bra011821 U (A01) 89.5 (91.1–92.4) Hull et al. (2000); Mikkelsen et al. (2000);

Zang et al. (2008a,b)Bra010644 U (A08) 90.8Bra017871* U (A03) 89.1

CYP79B3 At2g22330 Bra030246 I (A04) 89.7 Hull et al. (2000); Mikkelsen et al. (2000);Zang et al. (2008a,b)

CYP83A1 At4g13770 Bra032734 T (A04) 87.6 (85.6) Bak and Feyereisen (2001); Hemm et al. (2003);Naur et al. (2003); Zang et al. (2008a,b)

Bra016908* J (A04) 87.5CYP83B1 At4g31500 Bra034941* R (A08) 90.3 Bak and Tax (2001); Hansen et al. (2001a);

Naur et al. (2003); Zang et al. (2008a,b)GSTF9 At2g30860 Bra021673 IJ (A04) 90.0 (92.0) Wentzell et al. (2007)

Bra022815 IJ (A03) 91.5GSTF10 At2g30870 Bra022816 IJ (A03) 87.2 Wentzell et al. (2007)GSTF11 At3g03190 Bra032010 F (A05) 86.7 Hirai et al. (2005); Wentzell et al. (2007)GSTU20 At1g78370 Bra003645* E (A07) 87.2 Hirai et al. (2005); Wentzell et al. (2007)GGP1 At4g30530 Bra011201 U (A01) 86.2 (64.6–88.3) Geu-Flores et al. (2009)

Bra024068 U (A03) 85.8Bra010282 U (A08) 68.4Bra010283 U (A08) 86.2

(continued on next page)

137H. Wang et al. / Gene 487 (2011) 135–142

Table 1 (continued)

Group name AGI BrIDa Blocksb Identity (%)c References

SUR1 At2g20610 Bra036490 H (A07) 86.0 (90.7) Mikkelsen et al. (2004)Bra036703 H (A09) 87.3

UGT74B1 At1g24100 Bra024634 B (A09) 83.8 Gachon et al. (2005); Grubb et al. (2004)UGT74C1 At2g31790 Bra005641 J (A05) 86.4 (88.8) Gachon et al. (2005)

Bra021743 J (A04) 85.0ST5a At1g74100 Bra015935 E (A07) 88.8 (93.4) Klein et al. (2006); Piotrowski et al. (2004)

Bra008132 E (A02) 86.5ST5b At1g74090 Bra015938 E (A07) 84.8 (76.2–88.4) Hirai et al. (2005); Klein et al. (2006);

Piotrowski et al. (2004)Bra015936 E (A07) 83.0Bra003817 E (A07) 73.6Bra003818 E (A07) 81.8Bra003726 E (A07) 68.1Bra027880* D (A09) 80.9Bra027117* D (A09) 77.5Bra027118* D (A09) 80.3Bra027623* D (A09) 79.7Bra031476* D (A01) 74.6

ST5c At1g18590 Bra025668 A (A06) 83.5 Klein et al. (2006); Piotrowski et al. (2004)

Secondary modificationFMOGS-OX1 At1g65860 Hansen et al. (2007)FMOGS-OX2 At1g62540 Bra027035* D (A09) 83.4 Li et al. (2008); Wentzell et al. (2007)FMOGS-OX3 At1g62560 Li et al. (2008); Wentzell et al. (2007)FMOGS-OX4 At1g62570 Li et al. (2008); Wentzell et al. (2007)FMOGS-OX5 At1g12140 Bra026988 A (A09) 83.0 (81.2) Li et al. (2008)

Bra016787 A (A08) 80.7AOP1 At4g03070 Bra034182 O (A09) 81.4 (81.8–90.1) Kliebenstein et al. (2001a)

Bra034181 O (A09) 81.1Bra000847 O (A03) 81.7

AOP2 At4g03060 Bra034180 O (A09) 69.6 (59.4–83.7) Kliebenstein et al. (2001a)Bra000848 O (A03) 58.5Bra018521 O (A02) 68.8

AOP3 At4g03050 Kliebenstein et al. (2001a)GSL-OH At2g25450 Bra021670* IJ (A04) 81.7 (78.6–97.3) Hansen et al. (2008); Wentzell et al. (2007)

Bra021671* IJ (A04) 81.0Bra022920* J (A03) 80.4

CYP81F2 At5g57220 Bra002747 W (A10) 87.6 (88.4–90.9) Bednarek et al. (2009); Clay et al. (2009);Pfalz et al. (2009)

Bra020459 W (A02) 86.5Bra006830 W (A03) 86.9

Co-substrate pathwaysBZO1p1 At1g65880 Bra004132 E (A07) 80.9 (74.6) Kliebenstein et al. (2007)

Bra039743 E (A02) 74.1APK1 At2g14750 Bra013120 GH (A03) 87.0 (88.6) Mugford et al. (2009)

Bra039818 GH (S178) 85.4APK2 At4g39940 Bra011822 U (A01) 86.5 (82.2–89.4) Mugford et al. (2009)

Bra010645 U (A08) 81.6Bra017872* U (A03) 85.6

GSH1/PAD2 At4g23100 Bra013675 U (A01) 87.6 (80.0–89.4) Bednarek et al. (2009); Schlaeppi et al. (2008)Bra019333 U (A03) 71.3Bra019332 U (A03) 88.9

CHY1 At5g65940 Bra031802 X (A02) 88.6 (77.6–88.2) Ibdah and Pichersky (2009)Bra039968* IJ (A04) 82.0Bra039975* IJ (A04) 79.9Bra018392* IJ (A05) 83.0

AAO4 At1g04580 Bra015330 A (A10) 89.1 Ibdah et al. (2009a)

a Asterisks indicate BrGS genes that do not exhibit synteny.b Characters referring to 24 conserved blocks (A–X), which represent the conserved segments identifiable among the ancestral karyotpye, A. thaliana and B. rapa (Schranz et al.,

2006). Two letters indicate that the gene is located at the boundary between two blocks; text in parentheses indicates the specific chromosome of B. rapa. Bra039818 was anchoredon Scaffold000178, which has not yet been mapped onto a chromosome.

c Values in parentheses refer to the identity range between different copies of the same BrGS gene obtained using the Needle program from the EMBOSS software package.

Core structure formation

138 H. Wang et al. / Gene 487 (2011) 135–142

methionine derivatives, and CYP79F2 exclusively converts the long-chained methionine derivatives. A CYP79F1 knockout mutant iscompletely lacking in short-chain aliphatic GS, but has an increasedlevel of long-chain aliphatic glucosinolates. In contrast, a CYP79F2knockout mutant has substantially reduced long-chain aliphatic GS(Chen et al., 2003). Therefore, absence of CYP79F2 is likely to affect theprofile and content of aliphatic GS in B. rapa.

The flavin monooxygenase FMOGS-OX1-5 catalyzes the conversionof both short and long-chained methylthioalkyl GS to the correspond-

ing methylsulfinylalkyl GS, albeit with different chain-length speci-ficity. FMOGS-OX1-4s has a broad substrate specificity and catalyzes theconversion from methylthioalkyl GS to the corresponding methylsul-finylalkyl GS independent of chain length. In contrast, FMOGS-OX5

shows substrate specificity toward the long-chain 8-methylthiooctylGS (Hansen et al., 2007; Li et al., 2008; Wentzell et al., 2007). The lossof FMOGS-OX1, FMOGS-OX3, and FMOGS-OX4 should not block thebiosynthetic pathway, because these paralogs share similar functions.AOP3 catalyzes formation of hydroxyalkyl GSL in A. thaliana, but no

Fig. 1. Maps of B. rapa chromosomes showing the distribution of glucosinolate biosynthesis related genes.

139H.W

anget

al./Gene

487(2011)

135–142

140 H. Wang et al. / Gene 487 (2011) 135–142

ortholog of AOP3was identified in B. rapa (Fig. 2). This couldmean thegene was lost in B. rapa or that it evolved as a gene specific to A.thaliana after the divergence of Brassica and A. thaliana.

3.3. BrGS genes corresponded with evolution of B. rapa

Both A. thaliana and B. rapa produce a variety of GS and sharesimilar pathways for glucosinolate biosynthesis, so the comparison ofglucosinolate biosynthetic genes between A. thaliana and B. rapa mayprovide insights into the evolution of Brassica and A. thaliana.

Unlike A. thaliana, most of the glucosinolate biosynthesis geneswere present asmore than one copy in B. rapa. A comparison of B. rapaand A. thaliana genes involved in glucosinolate biosynthesis pathwaysis shown in Table 1. The origin of multi-copy genes might be becauseof genome triplication occurring after the divergence of B. rapa and A.thaliana (Wang et al., submitted for publication). Transpose genes andtandem genes might also have contributed to generation of multiplecopies of the BrGS genes.

In light of the presence of conserved blocks between B. rapa and A.thaliana, we analyzed the synteny of glucosinolate biosynthetic genesin the two species (Schranz et al., 2006). Among 44 AtGS genes, 40genes demonstrated synteny between B. rapa and A. thaliana(Table 1). We identified 83 homologous genes in B. rapa usingsynteny analysis. However, 96 homologous genes including the 83genes identified by synteny (from the above 40 genes), wereidentified in B. rapa by sequence similarity, indicating that 13.5% ofthese genes had no synteny, but maintained a high degree ofhomology with their orthologs in A. thaliana. The sequence identitiesfor syntenic genes (82.3%) and non-syntenic genes (82.1%) did notshow a statistical difference. These results suggest a triplication ortransposition event occurred not long after divergence of Brassica andA. thaliana. For the four genes without syntenic orthologs, we

Fig. 2. Phylogenetic relationship among AOP proteins. Note: At, Al, Bo, Bj and Br refer to ArabAOP proteins were identified in the present study; circles, triangles and squares refer to AO

identified six orthologs in total in B. rapa using sequence similarity.This could indicate that these genes were duplicated by a transpo-sition event accompanied by a large-scale chromosomal rearrange-ment after divergence.

Comparative physical mapping studies confirmed genome tripli-cation in B. rapa (Park et al., 2005). Based on the conserved syntenicblocks identified between the genomes of Brassica species and A.thaliana, the hypothesis of an ancestral karyotype was proposed(Schranz et al., 2006). Our data provide support for this hypothesis(Table 1). Multiple copies of the BrGS genes may have resulted fromgenome triplication, and retained synteny with their orthologs in A.thaliana, whereas most genes that appeared to comprise fewer thanthree copies might be caused by gene loss following triplication.However, ST5b consists of 10 copies in B. rapa, which probablyoccurred by duplication, transposition or tandem duplication aftertriplication, and this result is similar to the finding of Zang et al.(2009). We also found that two copies of ST5b are not expressed,which indicated that these two genes may be nonfunctional. Inaddition, 7% of genes do not retain co-linearity between A. thalianaand B. rapa, which perhaps resulted from transposition or/and large-scale chromosomal rearrangements.

3.4. BrGS genes could explain the profiles and accumulation of GS in B.rapa

More than 35 different GS have been documented, with 25 of thesecoming from differential elongation and side-chain modification ofmethionine in A. thaliana (Kliebenstein et al., 2001b). To date, 17different GS have been identified in B. rapa, of which 10 of arealiphatic GS, but the dominant GS were gluconapin (NAP), gluco-brassicanapin (GBN) and progoitrin (PRO) (Kim et al., 2010; Lou et al.,

idopsis thaliana, A. lyrata, Brassica oleracea, B. juncea and B. rapa, respectively; the otherP1, AOP2 and AOP3 respectively.

141H. Wang et al. / Gene 487 (2011) 135–142

2008; Padilla et al., 2007). Moreover, the profiles of these aliphatic GSindicate these are all short-chain GS in B. rapa.

In B. rapa, we observed all of the A. thaliana counterparts of genesthat participate in formation of the glucosinolate core structure exceptfor CYP79F2. A CYP79F2 knockout mutant showed substantiallyreduced long-chain aliphatic GS in A. thaliana (Chen et al., 2003).The absence of CYP79F2 corresponds well with the fact that all profilesof aliphatic GS in B. rapa are composed of short-chain GS.

The composition and amount of aliphatic GS is mainly controlledby the expression of AOP2 and AOP3, which, along with AOP1, arepresent in a tandem array in A. thaliana. AOP2 is directly responsiblefor biosynthesis of the alkenyl glucosinolates NAP or GBN (n=2–3),and the GS-OH locus can convert but-3-enyl glucosinolate (NAP) to 2-hydroxybut-3enyl glucosinolate (PRO) (Hansen et al., 2008). AOP3catalyzes methylsulfinylalkyl GS, and format hydroxyalkyl glucosino-lates (n=2) (Kliebenstein et al., 2001a). Phylogenetic analysisindicated the genome of B. rapa contains three orthologs of AOP loci,each containing two tandem duplicated genes that were revealed asorthologs of AOP1 and AOP2, while AOP3 does not exist (Fig. 2).Coincidentally, hydroxyalkyl glucosinolates are not detected in B. rapa(Kim et al., 2010; Lou et al., 2008; Padilla et al., 2007). In addition, anextra copy of AOP2 might catalyze formation of more alkenylglucosinolate and its downstream derivatives. This result correspondswell with the fact that the dominant GS were NAP, GBN and PRO, butonly a very low level of GRA is found in B. rapa (Kim et al., 2010;Padilla et al., 2007).

Duplicated genes, while providing redundancy to a system, mayalso enhance the potential resources for quantitative variation of aparticular trait (Li et al., 2008). This is illustrated by the fact that mostof the BrGS genes were present in multiple copies, yet there is onlyquantitative rather than qualitative variation in GS. While duplicatedgenes do show enhanced levels of genetic variation (Kliebenstein,2008), it remains to be seen whether duplicated gene families showany bias in controlling quantitative trait variation.

4. Conclusions

Glucosinolate biosynthetic genes were identified based on wholegenome analysis of B. rapa and comparative genomic analysesbetween A. thaliana and B. rapa. Glucosinolate biosynthetic pathwaysincluding 102 genes were established in B. rapa, and all genes but oneweremapped on 10 chromosomes in detail. Eight AtGS genes show nohomologs in B. rapa; however, this does not have a fatal effect onglucosinolate biosynthesis. Moreover, most of the BrGS genes exist inmore than one copy, and 93% of GS genes exhibit synteny between B.rapa and A. thaliana. In addition, copy number variation of BrGS genescorrelates with a triplication event in B. rapa, and BrGS gene contentcan explain the glucosinolate profiles and accumulation in B. rapa.

Acknowledgments

This work was mainly funded by the Chinese Ministry of Scienceand Technology, National Basic Research Program of China(2006CB101606, 2007CB108803), the National High TechnologyR&D Program of China (2006AA100108), and China InternationalScience and Technology Cooperation Project (2010DFA31730). Theresearch was conducted in Key Laboratory of Horticultural CropGenetic Improvement, Ministry of Agriculture and Sino-Dutch JointLab of Horticultural Genomics Technology.

References

Bak, S., Feyereisen, R., 2001. The involvement of two P450 enzymes, CYP83B1 andCYP83A1, in auxin homeostasis and glucosinolate biosynthesis. Plant Physiol. 127(1), 108–118.

Bak, S., Tax, F.E., 2001. CYP83B1, a cytochrome P450 at the metabolic branch point inauxin and indole glucosinolate biosynthesis in Arabidopsis. Plant Cell 13, 101–111.

Bednarek, P., et al., 2009. A glucosinolate metabolism pathway in living plant cellsmediates broad-spectrum antifungal defense. Science 323 (5910), 101–106.

Beekwilder, J., et al., 2008. The impact of the absence of aliphatic glucosinolates oninsect herbivory in Arabidopsis. PLoS One 3 (4), e2068.

Bones, A., Rossiter, J., 2006. The enzymic and chemically induced decomposition ofglucosinolates. Phytochemistry 67 (11), 1053–1067.

Brader, G., Mikkelsen, M., Halkier, B., Tapio Palva, E., 2006. Altering glucosinolateprofiles modulates disease resistance in plants. Plant J. 46 (5), 758–767.

Celenza, J., et al., 2005. The Arabidopsis ATR1 Myb transcription factor controls indolicglucosinolate homeostasis. Plant Physiol. 137 (1), 253–262.

Chen, S., Glawischnig, E., Jørgensen, K., 2003. CYP79F1 and CYP79F2 have distinctfunctions in the biosynthesis of aliphatic glucosinolates in Arabidopsis. Plant J. 33,923–937.

Choi, H., Cho, M., Lee, H., Yoon, D., 2010. Indole-3-carbinol induces apoptosis throughp53 and activation of caspase-8 pathway in lung cancer A549 cells. Food Chem.Toxicol. 48 (3), 883–890.

Clay, N.K., Adio, A.M., Denoux, C., Jander, G., Ausubel, F.M., 2009. Glucosinolatemetabolites required for an Arabidopsis innate immune response. Science 323(5910), 95–101.

Field, B., Cardon, G., Traka, M., Botterman, J., Vancanneyt, G., Mithen, R., 2004.Glucosinolate and amino acid biosynthesis in Arabidopsis. Plant Physiol. 135,828–839.

Gachon, C., Langlois-Meurinne, M., Henry, Y., Saindrenan, P., 2005. Transcriptional co-regulation of secondary metabolism enzymes in Arabidopsis: functional andevolutionary implications. Plant Mol. Biol. 58 (2), 229–245.

Geu-Flores, F., et al., 2009. Glucosinolate engineering identifies a gamma-glutamylpeptidase. Nat. Chem. Biol. 5 (8), 575–577.

Gigolashvili, T., Berger, B., Mock, H., Müller, C., Weisshaar, B., Flügge, U., 2007a. Thetranscription factor HIG1/MYB51 regulates indolic glucosinolate biosynthesis inArabidopsis thaliana. Plant J. 50 (5), 886–901.

Gigolashvili, T., Engqvist, M., Yatusevich, R., Müller, C., Flügge, U., 2008. HAG2/MYB76 andHAG3/MYB29 exert a specific and coordinated control on the regulation of aliphaticglucosinolate biosynthesis in Arabidopsis thaliana. New Phytol. 177 (3), 627–642.

Gigolashvili, T., Yatusevich, R., Berger, B., Müller, C., Flügge, U.I., 2007b. The R2R3-MYBtranscription factor HAG1/MYB28 is a regulator of methionine-derived glucosino-late biosynthesis in Arabidopsis thaliana. Plant J. 51 (2), 247–261.

Gigolashvili, T., Yatusevich, R., Rollwitz, I., Humphry, M., Gershenzon, J., Flügge, U.,2009. The plastidic bile acid transporter 5 is required for the biosynthesis ofmethionine-derived glucosinolates in Arabidopsis thaliana. Plant Cell 21 (6),1813–1829.

Grubb, C.D., Abel, S., 2006. Glucosinolate metabolism and its control. Trends Plant Sci.11 (2), 89–100.

Grubb, C.D., Zipp, B.J., Ludwig-Muller, J., Masuno, M.N., Molinski, T.F., Abel, S., 2004.Arabidopsis glucosyltransferase UGT74B1 functions in glucosinolate biosynthesisand auxin homeostasis. Plant J. 40 (6), 893–908.

Halkier, B.A., Gershenzon, J., 2006. Biology and biochemistry of glucosinolates. Annu.Rev. Plant Biol. 57, 303–338.

Hansen, B.G., Kliebenstein, D.J., Halkier, B.A., 2007. Identification of a flavin-monooxygenase as the S-oxygenating enzyme in aliphatic glucosinolate biosyn-thesis in Arabidopsis. Plant J. 50, 902–910.

Hansen, B.G., et al., 2008. A Novel 2-Oxoacid-Dependent dioxygenase involved in theformation of the goiterogenic 2-Hydroxybut-3-enyl glucosinolate and generalistinsect resistance in Arabidopsis. Plant Physiol. 148, 2096–2108.

Hansen, C.H., et al., 2001a. CYP83b1 is the oxime-metabolizing enzyme in theglucosinolate pathway in Arabidopsis. J. Biol. Chem. 276 (27), 24790–24796.

Hansen, C.H., Wittstock, U., Olsen, C.E., Hick, A.J., Pickett, J.A., Halkier, B.A., 2001b.Cytochrome P450 CYP79F1 from Arabidopsis catalyzes the conversion of dihomo-methionine and trihomomethionine to the corresponding aldoximes in thebiosynthesis of aliphatic glucosinolates. J. Biol. Chem. 276 (14), 11078–11085.

Hayes, J., Kelleher, M., Eggleston, I., 2008. The cancer chemopreventive actions ofphytochemicals derived from glucosinolates. Eur. J. Nutr. 47, 73–88.

He, Y., Chen, B., Pang, Q., Strul, J.M., Chen, S., 2010. Functional specification ofArabidopsis isopropylmalate isomerases in glucosinolate and leucine biosynthesis.Plant Cell Physiol. 51 (9), 1480–1487.

He, Y., et al., 2009. A redox-active isopropylmalate dehydrogenase functions in thebiosynthesis of glucosinolates and leucine in Arabidopsis. Plant J. 60 (4), 679–690.

Hecht, S., 2000. Inhibition of carcinogenesis by isothiocyanates 1. Drug Metab. Rev. 32(3–4), 395–411.

Hemm, M.R., Ruegger, M.O., Chapple, C., 2003. The Arabidopsis ref2 mutant is defectivein the gene encoding CYP83A1 and shows both phenylpropanoid and glucosinolatephenotypes. Plant Cell 15 (1), 179–194.

Hirai, M., et al., 2005. Elucidation of gene-to-gene and metabolite-to-gene networks inArabidopsis by integration of metabolomics and transcriptomics. J. Biol. Chem. 280(27), 25590–25595.

Hirai, M.Y., et al., 2007. Omics-based identification of Arabidopsis Myb transcriptionfactors regulating aliphatic glucosinolate biosynthesis. PNAS 104 (15), 6478–6483.

Hull, A.K., Vij, R., Celenza, J.L., 2000. Arabidopsis cytochrome P450s that catalyze the firststep of tryptophan-dependent indole-3-acetic acid biosynthesis. PNAS 97 (5),2379–2384.

Ibdah, M., Chen, Y.T., Wilkerson, C.G., Pichersky, E., 2009. An aldehyde oxidase indeveloping seeds of Arabidopsis converts benzaldehyde to benzoic acid. PlantPhysiol. 150 (1), 416–423.

Ibdah, M., Pichersky, E., 2009. Arabidopsis Chy1 null mutants are deficient in benzoicacid-containing glucosinolates in the seeds. Plant Biol. (Stuttg.) 11 (4), 574–581.

Keck, A., Finley, J., 2004. Cruciferous vegetables: cancer protective mechanisms ofglucosinolate hydrolysis products and selenium. Integr. Cancer Ther. 3 (1), 5–12.

142 H. Wang et al. / Gene 487 (2011) 135–142

Kim, J., et al., 2010. Variation of glucosinolates in vegetable crops of Brassica rapa L. ssp.pekinensis. Food Chem. 119 (1), 423–428.

Kim, M., Park, J., 2009. Cruciferous vegetable intake and the risk of human cancer:epidemiological evidence. Proc. Nutr. Soc. 68 (01), 103–110.

Klein, M., Reichelt, M., Gershenzon, J., Papenbrock, J., 2006. The three desulfoglucosi-nolate sulfotransferase proteins in Arabidopsis have different substrate specificitiesand are differentially expressed. FEBS J. 273 (1), 122–136.

Kliebenstein, D.J., Lambrix, V., Reichelt, M., Gershenzon, J., Mitchell-Olds, T., 2001a.Gene duplication in the diversification of secondary metabolism: tandem 2-oxoglutarate-dependent dioxygenases control glucosinolate biosynthesis in Arabi-dopsis. Plant Cell 13 (3), 681–693.

Kliebenstein, D.J., et al., 2007. Characterization of seed specific benzoyloxyglucosinolatemutations in Arabidopsis thaliana. Plant J. 51, 1062–1076.

Kliebenstein, D.J., 2008. A role for gene duplication and natural variation of geneexpression in the evolution of metabolism. PLoS One 3 (3), 1–13.

Kliebenstein, D.J., et al., 2001b. Genetic control of natural variation in Arabidopsisglucosinolate accumulation. Plant Physiol. 126, 811–825.

Knill, T., Reichelt, M., Paetz, C., Gershenzon, J., Binder, S., 2009. Arabidopsis thalianaencodes a bacterial-type heterodimeric isopropylmalate isomerase involved inboth Leu biosynthesis and the Methionine chain elongation pathway ofglucosinolate formation. Plant Mol. Biol. 71 (3), 227–239.

Knill, T., Schuster, J., Reichelt, Michael, Gershenzon, J., Binder, S., 2008. Arabidopsisbranched-chain aminotransferase 3 functions in both amino acid and glucosinolatebiosynthesis. Plant Physiol. 146, 1028–1039.

Kroymann, J., et al., 2001. A gene controlling variation in Arabidopsis glucosinolatecomposition is part of the methionine chain elongation pathway. Plant Physiol. 127(3), 1077–1088.

Kushad, M., et al., 1999. Variation of glucosinolates in vegetable crops of Brassicaoleracea. J. Agric. Food Chem. 47 (4), 1541–1548.

Levy, M., Wang, Q., Kaspi, R., Parrella, M., Abel, S., 2005. Arabidopsis IQD1, a novelcalmodulin-binding nuclear protein, stimulates glucosinolate accumulation andplant defense. Plant J. 43 (1), 79–96.

Li, J., Hansen, B.G., Ober, J.A., Kliebenstein, D.J., Halkier, B.A., 2008. Subclade of flavin-monooxygenases involved in aliphatic glucosinolate biosynthesis. Plant Physiol.148, 1721–1733.

Lou, P., et al., 2008. Quantitative trait loci for glucosinolate accumulation in Brassicarapa leaves. New Phytol. 179, 1017–1032.

Mikkelsen, M., Hansen, C., Wittstock, U., Halkier, B.A., 2000. Cytochrome P450 CYP79B2from Arabidopsis catalyzes the conversion of tryptophan to indole-3-acetaldoxime,a precursor of indole glucosinolates and indole-3-acetic acid. J. Biol. Chem. 275(43), 33712–33717.

Mikkelsen, M.D., Naur, P., Halkier, B.A., 2004. Arabidopsis mutants in the C-S lyase ofglucosinolate biosynthesis establish a critical role for indole-3-acetaldoxime inauxin homeostasis. Plant J. 37 (5), 770–777.

Mugford, S.G., et al., 2009.Disruption of adenosine-5′-phosphosulfate kinase inArabidopsisreduces levels of sulfated secondary metabolites. Plant Cell 21 (3), 910–927.

Nakajima, M., Yoshida, R., Shimada, N., Yamazaki, H., Yokoi, T., 2001. Inhibition andinactivation of human cytochrome P450 isoforms by phenethyl isothiocyanate.Drug Metab. Dispos. 29 (8), 1110–1113.

Naur, P., et al., 2003. CYP83A1 and CYP83B1, two nonredundant cytochrome P450enzymes metabolizing oximes in the biosynthesis of glucosinolates in Arabidopsis.Plant Physiol. 133, 63–72.

Padilla, G., Cartea, M.a.E., Velasco, P., 2007. Variation of glucosinolates in vegetablecrops of Brassica rapa. Phytochemistry 68, 536–545.

Park, J.Y., et al., 2005. Physical mapping and microsynteny of Brassica rapa ssp. pekinensisgenome corresponding to a 222 kb gene-rich region of Arabidopsis chromosome 4and partially duplicated on chromosome 5. Mol. Gen. Genomics 274, 579–588.

Pfalz, M., Vogel, H., Kroymann, J., 2009. The gene controlling the indole glucosinolatemodifier1 quantitative trait locus alters indole glucosinolate structures and aphidresistance in Arabidopsis. Plant Cell 21, 985–999.

Piotrowski, M., et al., 2004. Desulfoglucosinolate sulfotransferases from Arabidopsisthaliana catalyze the final step in the biosynthesis of the glucosinolate corestructure. J. Biol. Chem. 279 (49), 50717–50725.

Rask, L., Andreasson, E., Ekbom, B., Eriksson, S., Pontoppidan, B., Meijer, J., 2000.Myrosinase: gene family evolution and herbivore defense in Brassicaceae. PlantMol. Biol. 42 (1), 93–114.

Reintanz, B., et al., 2001. bus, a bushy Arabidopsis CYP79F1 knockout mutant withAbolished synthesis of short-chain aliphatic glucosinolates. Plant Cell 13, 351–367.

Sønderby, I., Hansen, B., Bjarnholt, N., Ticconi, C., Halkier, B.A., Kliebenstein, D.J., 2007. Asystems biology approach identifies a R2R3 MYB gene subfamily with distinct andoverlapping functions in regulation of aliphatic glucosinolates. PLoS One 2 (12),e1322.

Sønderby, I.E., Geu-Flores, F., Halkier, B.A., 2010. Biosynthesis of glucosinolate genediscovery and beyond. Trends Plant Sci. 15 (5), 283–296.

Sawada, Y., et al., 2009a. Arabidopsis bile acid: sodium symporter family protein 5 isinvolved in methionine-derived glucosinolate biosynthesis. Plant Cell Physiol. 50(9), 1579–1586.

Sawada, Y., et al., 2009b. Omics-based approaches to methionine side-chain elongationin Arabidopsis: characterization of the genes encoding methylthioalkylmalateisomerase and methylthioalkylmalate dehydrogenase. Plant Cell Physiol. 50,1181–1190.

Schlaeppi, K., Bodenhausen, N., Buchala, A., Mauch, F., Reymond, P., 2008. The glutathione-deficient mutant pad2-1 accumulates lower amounts of glucosinolates and is moresusceptible to the insect herbivore Spodoptera littoralis. Plant J. 55 (5), 774–786.

Schonhof, I., Krumbein, A., Brückner, B., 2004. Genotypic effects on glucosinolates andsensory properties of broccoli and cauliflower. Nahrung/Food 48 (1), 25–33.

Schranz, M.E., Lysak, M.A., Mitchell-Olds, T., 2006. The ABC's of comparative genomicsin the Brassicaceae: building blocks of crucifer genomes. Trends Plant Sci. 11 (11),535–542.

Schuster, J., Knill, T., Reichelt, M., Gershenzon, J., Binder, S., 2006. Branched-chainaminotransferase4 is part of the chain elongation pathway in the biosynthesis ofmethionine-derived glucosinolates in Arabidopsis. Plant Cell 18, 2664–2679.

Skirycz, A., et al., 2006. DOF transcription factor AtDof1. 1 (OBP2) is part of a regulatorynetwork controlling glucosinolate biosynthesis in Arabidopsis. Plant J. 47 (1),10–24.

Tamura, K., Dudley, J., Nei, M., Kumar, S., 2007. MEGA4: molecular evolutionary geneticsanalysis (MEGA) software version 4.0. Mol. Biol. Evol. 24 (8), 1596–1599.

Tang, L., et al., 2008. Consumption of raw cruciferous vegetables is inversely associatedwith bladder cancer risk. Cancer Epidemiol. Biomark. Prev. 17 (4), 938–944.

Textor, S., et al., 2004. Biosynthesis of methionine-derived glucosinolates in Arabidopsisthaliana: recombinant expression and characterization of methylthioalkylmalatesynthase, the condensing enzyme of the chain-elongation cycle. Planta 218 (6),1026–1035.

Textor, S., de Kraker, J.W., Hause, B., Gershenzon, J., Tokuhisa, J.G., 2007. MAM3catalyzes the formation of all aliphatic glucosinolate chain lengths in Arabidopsis.Plant Physiol. 144, 60–71.

Wang, et al., submitted for publication. The genome of the mesohexaploid crop speciesBrassica rapa. Nat. Genet. (Acceptd, 3rd Aug, 2011)

Wentzell, A.M., Rowe, H.C., Hansen, B.G., Ticconi, C., Halkier, B.A., Kliebenstein, D.J.,2007. Linking metabolic QTLs with network and cis-eQTLs controlling biosyntheticpathways. PLoS Genet. 3 (9), 1687–1701.

Wittstock, U., Halkier, B., 2000. Cytochrome P450 CYP79A2 from Arabidopsis thaliana L.catalyzes the conversion of L-phenylalanine to phenylacetaldoxime in thebiosynthesis of benzylglucosinolate. J. Biol. Chem. 275 (19), 14659–14666.

Wittstock, U., Halkier, B., 2002. Glucosinolate research in the Arabidopsis era. TrendsPlant Sci. 7 (6), 263–270.

Wittstock, U., Kliebenstein, D., Lambrix, V., Reichelt, M., Gershenzon, J., 2003. Chapterfive glucosinolate hydrolysis and its impact on generalist and specialist insectherbivores. Recent Adv. Phytochem. 37, 101–125.

Wu, L., et al., 2004. Dietary approach to attenuate oxidative stress, hypertension, andinflammation in the cardiovascular system. PNAS 101 (18), 7094–7099.

Zang, Y.X., et al., 2009. Genome-wide identification of glucosinolate synthesis genes inBrassica rapa. FEBS J. 276, 3559–3574.

Zang, Y.X., Kim, D.H., Lim, M., Park, B.S., Hong, S.B., 2008a. Metabolic engineering ofindole glucosinolates in Chinese cabbage plants by expression of ArabidopsisCYP79B2, CYP79B3, and CYP83B1. Mol. Cells 25 (2), 231–241.

Zang, Y.X., Kim, J.H., Park, Y.D., Kim, D.H., Hong, S.B., 2008b. Metabolic engineering ofaliphatic glucosinolates in Chinese cabbage plants expressing Arabidopsis MAM1,CYP79F1, and CYP83A1. BMB Rep. 41 (6), 472–478.

Zhang, Y., Kensler, T., Cho, C., Posner, G., Talalay, P., 1994. Anticarcinogenic activities ofsulforaphane and structurally related synthetic norbornyl isothiocyanates. PNAS 91(8), 3147–3150.