Genetics of aliphatic glucosinolates. I. Side chain elongation in Brassica napus and Arabidopsis thaliana

Computational prediction of plant metabolic pathways

Uncovering genes encoding enzymes responsible for the biosynthesis of diverse plant metabolites is essential for metabolic engineering and production of plant metabolite-derived medicine. With the availability of multi-omics data for an ever-increasing number of plant species and the development of computational approaches, the metabolic pathways of many important plant compounds can be predicted, complementing a more traditional genetic and/or biochemical approach. Here, we summarize recent progress in predicting plant metabolic pathways using genome, transcriptome, proteome, interactome, and/or metabolome data, and the utility of integrating these data with machine learning to further improve metabolic pathway predictions.

A Comparative Transcriptome and Metabolome Combined Analysis Reveals the Key Genes and Their Regulatory Model Responsible for Glucoraphasatin Accumulation in Radish Fleshy Taproots

Radish (Raphanus sativus L.) is rich in specific glucosinolates (GSLs), which benefit human health and special flavor formation. Although the basic GSLs metabolic pathway in Brassicaceae plants is clear, the regulating mechanism for specific glucosinolates content in radish fleshy taproots is not well understood. In this study, we discovered that there was a significant difference in the GSLs profiles and the content of various GSLs components. Glucoraphasatin (GRH) is the most predominant GSL in radish taproots of different genotypes as assessed by HPLC analysis. Further, we compared the taproot transcriptomes of three radish genotypes with high and low GSLs content by employing RNA-seq. Totally, we identified forty-one differentially expressed genes related to GSLs metabolism. Among them, thirteen genes (RsBCAT4, RsIPMDH1, RsMAM1a, RsMAM1b, RsCYP79F1, RsGSTF9, RsGGP1, RsSUR1, RsUGT74C1, RsST5b, RsAPK1, RsGSL-OH, and RsMYB28) were significantly higher co-expressed in the high content genotypes than in low content genotype. Notably, correlation analysis indicated that the expression level of RsMYB28, as an R2R3 transcription factor directly regulating aliphatic glucosinolate biosynthesis, was positively correlated with the GRH content. Co-expression network showed that RsMYB28 probably positively regulated the expression of the above genes, particularly RsSUR1, and consequently the synthesis of GRH. Moreover, the molecular mechanism of the accumulation of this 4-carbon (4C) GSL in radish taproots was explored. This study provides new perspectives on the GSLs accumulation mechanism and genetic improvements in radish taproots.

Effects of in vitro metabolism of a broccoli leachate, glucosinolates and S-methylcysteine sulphoxide on the human faecal microbiome

PURPOSE

Brassica are an important food source worldwide and are characterised by the presence of compounds called glucosinolates. Studies indicate that the glucosinolate derived bioactive metabolite sulphoraphane can elicit chemoprotective benefits on human cells. Glucosinolates can be metabolised in vivo by members of the human gut microbiome, although the prevalence of this activity is unclear. Brassica and Allium plants also contain S-methylcysteine sulphoxide (SMCSO), that may provide additional health benefits but its metabolism by gut bacteria is not fully understood.

METHODS

We examined the effects of a broccoli leachate (BL) on the composition and function of human faecal microbiomes of five different participants under in vitro conditions. Bacterial isolates from these communities were then tested for their ability to metabolise glucosinolates and SMCSO.

RESULTS

Microbial communities cultured in vitro in BL media were observed to have enhanced growth of lactic acid bacteria, such as lactobacilli, with a corresponding increase in the levels of lactate and short-chain fatty acids. Members of Escherichia isolated from these faecal communities were found to bioconvert glucosinolates and SMCSO to their reduced analogues.

CONCLUSION

This study uses a broccoli leachate to investigate the bacterial-mediated bioconversion of glucosinolates and SMCSO, which may lead to further products with additional health benefits to the host. We believe that this is the first study that shows the reduction of the dietary compound S-methylcysteine sulphoxide by bacteria isolated from human faeces.

Dissection of genetic architecture for glucosinolate accumulations in leaves and seeds of Brassica napus by genome-wide association study

Glucosinolates (GSLs), whose degradation products have been shown to be increasingly important for human health and plant defence, compose important secondary metabolites found in the order Brassicales. It is highly desired to enhance pest and disease resistance by increasing the leaf GSL content while keeping the content low in seeds of Brassica napus, one of the most important oil crops worldwide. Little is known about the regulation of GSL accumulation in the leaves. We quantified the levels of 9 different GSLs and 15 related traits in the leaves of 366 accessions and found that the seed and leaf GSL content were highly correlated (r = 0.79). A total of 78 loci were associated with GSL traits, and five common and eleven tissue-specific associated loci were related to total leaf and seed GSL content. Thirty-six candidate genes were inferred to be involved in GSL biosynthesis. The candidate gene BnaA03g40190D (BnaA3.MYB28) was validated by DNA polymorphisms and gene expression analysis. This gene was responsible for high leaf/low seed GSL content and could explain 30.62% of the total leaf GSL variation in the low seed GSL panel and was not fixed during double-low rapeseed breeding. Our results provide new insights into the genetic basis of GSL variation in leaves and seeds and may facilitate the metabolic engineering of GSLs and the breeding of high leaf/low seed GSL content in B. napus.

Chemical Aspects of Biological Activity of Isothiocyanates and Indoles, the Products of Glucosinolate Decomposition

There is growing evidence that cancer chemoprevention employing natural, bioactive compounds may halt or at least slow down the different stages of carcinogenesis. A particularly advantageous effect is attributed to derivatives of sulfur-organic phytochemicals, such as glucosinolates (GLs) synthesized mainly in Brassicaceae plant family. GLs are hydrolysed enzymatically to bioactive isothiocyanates (ITC) and indoles, which exhibit strong anti-inflammatory and anti-carcinogenic activity. Highly bioavailable electrophilic ITC are of particular interest, as they can react with nucleophilic groups of important biomolecules to form dithiocarbamates, thiocarbamates and thioureas. These modifications seem responsible for the chemopreventive activity, but also for genotoxicity and mutagenicity. It was documented that ITC can permanently bind to important biomolecules such as glutathione, cytoskeleton proteins, transcription factors NF-κB and Nrf2, thiol-disulfide oxidoreductases, proteasome proteins or heat shock proteins. Furthermore, ITC may also affect epigenetic regulation of gene expression, e.g. by inhibition of histone deacetylases. Some other derivatives of glucosinolates, especially indoles, are able to form covalent bonds with nucleobases in DNA, which may result in genotoxicity and mutagenicity. This article summarizes the current state of knowledge about glucosinolates and their degradation products in terms of possible interactions with reactive groups of cellular molecules.

Reconstructing Biosynthetic Pathway of the Plant-Derived Cancer Chemopreventive-Precursor Glucoraphanin in Escherichia coli

Epidemiological data confirmed a strong correlation between regular consumption of cruciferous vegetables and lower cancer risk. This cancer preventive property is mainly attributed to the glucosinolate products, such as glucoraphanin found in broccoli that is derived from methionine. Here we report the first successful reconstruction of the complete biosynthetic pathway of glucoraphanin from methionine in Escherichia coli via gene selection, pathway design, and protein engineering. We used branched-chain amino transferase 3 to catalyze two transamination steps to ensure the purity of precursor molecules and used cysteine as a sulfur donor to simplify the synthesis pathway. Two chimeric cytochrome P450 enzymes were engineered and expressed in E. coli functionally. The original plant C-S lyase was replaced by the Neurospora crassa hercynylcysteine sulfoxide lyase. Other pathway enzymes were successfully mined from Arabidopsis thaliana, Brassica rapa, and Brassica oleracea. Biosynthesis of glucoraphanin upon coexpression of the optimized enzymes in vivo was confirmed by liquid chromatography-tandem mass spectrometry analysis. No other glucosinolate analogues (except for glucoiberin) were identified that could facilitate the downstream purification processes. Production of glucoraphanin in this study laid the foundation for microbial production of such health-beneficial glucosinolates in a large-scale.

Consumption of a diet rich in Brassica vegetables is associated with a reduced abundance of sulphate-reducing bacteria: A randomised crossover study

SCOPE

We examined whether a Brassica-rich diet was associated with an increase in the relative abundance of intestinal lactobacilli and sulphate-reducing bacteria (SRB), or alteration to the composition of the gut microbiota, in healthy adults.

METHODS AND RESULTS

A randomised crossover study was performed with ten healthy adults who were fed a high- and a low-Brassica diet for 2-wk periods, with a 2-wk washout phase separating the diets. The high-Brassica diet consisted of six 84 g portions of broccoli, six 84 g portions of cauliflower and six 300 g portions of a broccoli and sweet potato soup. The low-Brassica diet consisted of one 84 g portion of broccoli and one 84 g portion of cauliflower. Faecal microbiota composition was measured in samples collected following 2-wk Brassica-free periods (consumption of all Brassica prohibited), and after each diet, whereby the only Brassica consumed was that supplied by the study team. No significant changes to the relative abundance of lactobacilli were observed (p = 0.8019). The increased consumption of Brassica was associated with a reduction in the relative abundance of SRB (p = 0.0215), and members of the Rikenellaceae, Ruminococcaceae, Mogibacteriaceae, Clostridium and unclassified Clostridiales (p < 0.01).

CONCLUSION

The increased consumption of Brassica vegetables was linked to a reduced relative abundance of SRB, and therefore may be potentially beneficial to gastrointestinal health.

Flowering Locus C (FLC) Is a Potential Major Regulator of Glucosinolate Content across Developmental Stages of Aethionema arabicum (Brassicaceae)

The biochemical defense of plants can change during their life-cycle and impact herbivore feeding and plant fitness. The annual species Aethionema arabicum is part of the sister clade to all other Brassicaceae. Hence, it holds a phylogenetically important position for studying crucifer trait evolution. Glucosinolates (GS) are essentially Brassicales-specific metabolites involved in plant defense. Using two Ae. arabicum accessions (TUR and CYP) we identify substantial differences in glucosinolate profiles and quantities between lines, tissues and developmental stages. We find tissue specific side-chain modifications in aliphatic GS: methylthioalkyl in leaves, methylsulfinylalkyl in fruits, and methylsulfonylalkyl in seeds. We also find large differences in absolute glucosinolate content between the two accessions (up to 10-fold in fruits) that suggest a regulatory factor is involved that is not part of the quintessential glucosinolate biosynthetic pathway. Consistent with this hypothesis, we identified a single major multi-trait quantitative trait locus controlling total GS concentration across tissues in a recombinant inbred line population derived from TUR and CYP. With fine-mapping, we narrowed the interval to a 58 kb region containing 15 genes, but lacking any known GS biosynthetic genes. The interval contains homologs of both the sulfate transporter SULTR2;1 and FLOWERING LOCUS C. Both loci have diverse functions controlling plant physiological and developmental processes and thus are potential candidates regulating glucosinolate variation across the life-cycle of Aethionema. Future work will investigate changes in gene expression of the candidates genes, the effects of GS variation on insect herbivores and the trade-offs between defense and reproduction.

Natural variation of root exudates in Arabidopsis thaliana-linking metabolomic and genomic data

Many metabolomics studies focus on aboveground parts of the plant, while metabolism within roots and the chemical composition of the rhizosphere, as influenced by exudation, are not deeply investigated. In this study, we analysed exudate metabolic patterns of Arabidopsis thaliana and their variation in genetically diverse accessions. For this project, we used the 19 parental accessions of the Arabidopsis MAGIC collection. Plants were grown in a hydroponic system, their exudates were harvested before bolting and subjected to UPLC/ESI-QTOF-MS analysis. Metabolite profiles were analysed together with the genome sequence information. Our study uncovered distinct metabolite profiles for root exudates of the 19 accessions. Hierarchical clustering revealed similarities in the exudate metabolite profiles, which were partly reflected by the genetic distances. An association of metabolite absence with nonsense mutations was detected for the biosynthetic pathways of an indolic glucosinolate hydrolysis product, a hydroxycinnamic acid amine and a flavonoid triglycoside. Consequently, a direct link between metabolic phenotype and genotype was detected without using segregating populations. Moreover, genomics can help to identify biosynthetic enzymes in metabolomics experiments. Our study elucidates the chemical composition of the rhizosphere and its natural variation in A. thaliana, which is important for the attraction and shaping of microbial communities.

The Genetics Underlying Natural Variation in the Biotic Interactions of Arabidopsis thaliana: The Challenges of Linking Evolutionary Genetics and Community Ecology

In the context of global change, predicting the responses of plant communities in an ever-changing biotic environment calls for a multipronged approach at the interface of evolutionary genetics and community ecology. However, our understanding of the genetic basis of natural variation involved in mediating biotic interactions, and associated adaptive dynamics of focal plants in their natural communities, is still in its infancy. Here, we review the genetic and molecular bases of natural variation in the response to biotic interactions (viruses, bacteria, fungi, oomycetes, herbivores, and plants) in the model plant Arabidopsis thaliana as well as the adaptive value of these bases. Among the 60 identified genes are a number that encode nucleotide-binding site leucine-rich repeat (NBS-LRR)-type proteins, consistent with early examples of plant defense genes. However, recent studies have revealed an extensive diversity in the molecular mechanisms of defense. Many types of genetic variants associate with phenotypic variation in biotic interactions, even among the genes of large effect that tend to be identified. In general, we found that (i) balancing selection rather than directional selection explains the observed patterns of genetic diversity within A. thaliana and (ii) the cost/benefit tradeoffs of adaptive alleles can be strongly dependent on both genomic and environmental contexts. Finally, because A. thaliana rarely interacts with only one biotic partner in nature, we highlight the benefit of exploring diffuse biotic interactions rather than tightly associated host-enemy pairs. This challenge would help to improve our understanding of coevolutionary quantitative genetics within the context of realistic community complexity.

BjuB.CYP79F1 Regulates Synthesis of Propyl Fraction of Aliphatic Glucosinolates in Oilseed Mustard Brassica juncea: Functional Validation through Genetic and Transgenic Approaches

Among the different types of methionine-derived aliphatic glucosinolates (GS), sinigrin (2-propenyl), the final product in 3C GS biosynthetic pathway is considered very important as it has many pharmacological and therapeutic properties. In Brassica species, the candidate gene regulating synthesis of 3C GS remains ambiguous. Earlier reports of GSL-PRO, an ortholog of Arabidopsis thaliana gene At1g18500 as a probable candidate gene responsible for 3C GS biosynthesis in B. napus and B. oleracea could not be validated in B. juncea through genetic analysis. In this communication, we report the isolation and characterization of the gene CYP79F1, an ortholog of A. thaliana gene At1g16410 that is involved in the first step of core GS biosynthesis. The gene CYP79F1 in B. juncea showed presence-absence polymorphism between lines Varuna that synthesizes sinigrin and Heera virtually free from sinigrin. Using this presence-absence polymorphism, CYP79F1 was mapped to the previously mapped 3C GS QTL region (J16Gsl4) in the LG B4 of B. juncea. In Heera, the gene was observed to be truncated due to an insertion of a ~4.7 kb TE like element leading to the loss of function of the gene. Functional validation of the gene was carried out through both genetic and transgenic approaches. An F2 population segregating only for the gene CYP79F1 and the sinigrin phenotype showed perfect co-segregation. Finally, genetic transformation of a B. juncea line (QTL-NIL J16Gsl4) having high seed GS but lacking sinigrin with the wild type CYP79F1 showed the synthesis of sinigrin validating the role of CYP79F1 in regulating the synthesis of 3C GS in B. juncea.

Comparative Transcriptome Analyses Reveal a Special Glucosinolate Metabolism Mechanism in Brassica alboglabra Sprouts

Brassica sprouts contain abundant phytochemicals, especially glucosinolates (GSs). Various methods have been used to enhance GS content in sprouts. However, the molecular basis of GS metabolism in sprouts remains an open question. Here we employed RNA-seq analysis to compare the transcriptomes of high-GS (JL-08) and low-GS (JL-09) Brassica alboglabra sprouts. Paired-end Illumina RNA-seq reads were generated and mapped to the Brassica oleracea reference genome. The differentially expressed genes were analyzed between JL-08 and JL-09. Among these, 1477 genes were up-regulated and 1239 down-regulated in JL-09 compared with JL-08. Enrichment analysis of these differentially expressed genes showed that the GS biosynthesis had the smallest enrichment factor and the highest Q-value of all metabolic pathways in Kyoto Encyclopedia of Genes and Genomes database, indicating the main metabolic difference between JL-08 and JL-09 is the GS biosynthetic pathway. Thirty-seven genes of the sequenced data were annotated as putatively involved in GS biosynthesis, degradation, and regulation, of which 11 were differentially expressed in JL-08 and JL-09. The expression level of GS degradation enzyme myrosinase in high-GS JL-08 was lower compared with low-GS JL-09. Surprisingly, in high-GS JL-08, the expression levels of GS biosynthesis genes were also lower than those in low-GS JL-09. As the GS contents in sprouts are determined by dynamic equilibrium of seed stored GS mobilization, de novo synthesis, degradation, and extra transport, the result of this study leads us to suggest that efforts to increase GS content should focus on either raising GS content in seeds or decreasing myrosinase activity, rather than improving the expression level of GS biosynthesis genes in sprouts.

Macroevolutionary patterns of glucosinolate defense and tests of defense-escalation and resource availability hypotheses

We explored macroevolutionary patterns of plant chemical defense in Streptanthus (Brassicaceae), tested for evolutionary escalation of defense, as predicted by Ehrlich and Raven's plant-herbivore coevolutionary arms-race hypothesis, and tested whether species inhabiting low-resource or harsh environments invest more in defense, as predicted by the resource availability hypothesis (RAH). We conducted phylogenetically explicit analyses using glucosinolate profiles, soil nutrient analyses, and microhabitat bareness estimates across 30 species of Streptanthus inhabiting varied environments and soils. We found weak to moderate phylogenetic signal in glucosinolate classes and no signal in total glucosinolate production; a trend toward evolutionary de-escalation in the numbers and diversity of glucosinolates, accompanied by an evolutionary increase in the proportion of aliphatic glucosinolates; some support for the RAH relative to soil macronutrients, but not relative to serpentine soil use; and that the number of glucosinolates increases with microhabitat bareness, which is associated with increased herbivory and drought. Weak phylogenetic signal in chemical defense has been observed in other plant systems. A more holistic approach incorporating other forms of defense might be necessary to confidently reject escalation of defense. That defense increases with microhabitat bareness supports the hypothesis that habitat bareness is an underappreciated selective force on plants in harsh environments.

Deciphering allelic variations for seed glucosinolate traits in oilseed mustard (Brassica juncea) using two bi-parental mapping populations

QTL mapping by two DH mapping populations deciphered allelic variations for five different seed glucosinolate traits in B. juncea. Allelic variations for five different seed glucosinolate (GS) traits, namely % propyl, % butyl, % pentyl, aliphatics and total GS content were studied through QTL analysis using two doubled haploid (DH) mapping populations. While the high GS parents in two populations differed in their profiles of seed aliphatic GS, the low GS parents were similar. Phenotypic data of seed GS traits from three environments of the two populations were subjected to QTL analysis. The first population (referred to as DE population) detected a total of 60 QTL from three environments which upon intra-population meta-QTL analysis were merged to 17 S-QTL (Stable QTL) and 15 E-QTL (Environment QTL). The second population (referred to as VH population) detected 58 QTL from the three environments that were merged to 15S-QTL and 16E-QTL. In both the populations, majority of S-QTL were detected as major QTL. Inter-population meta-analysis identified three C-QTL (consensus QTL) formed by merging major QTL from the two populations. Candidate genes of GS pathway were co-localized to the QTL regions either through genetic mapping or through in silico comparative analysis. Parental allelic variants of QTL or of the co-mapped candidate gene(s) were determined on the basis of the significantly different R (2) values of the component QTL from the two populations which were merged to form C-QTL. The results of the study are significant for marker-assisted transfer of the low GS trait and also for developing lines with lower GS than are present in Brassica juncea.

Investigation of the multifunctional gene AOP3 expands the regulatory network fine-tuning glucosinolate production in Arabidopsis

Quantitative trait loci (QTL) mapping studies enable identification of loci that are part of regulatory networks controlling various phenotypes. Detailed investigations of genes within these loci are required to ultimately understand the function of individual genes and how they interact with other players in the network. In this study, we use transgenic plants in combination with natural variation to investigate the regulatory role of the AOP3 gene found in GS-AOP locus previously suggested to contribute to the regulation of glucosinolate defense compounds. Phenotypic analysis and QTL mapping in F2 populations with different AOP3 transgenes support that the enzymatic function and the AOP3 RNA both play a significant role in controlling glucosinolate accumulation. Furthermore, we find different loci interacting with either the enzymatic activity or the RNA of AOP3 and thereby extend the regulatory network controlling glucosinolate accumulation.

Cytoplasmic genetic variation and extensive cytonuclear interactions influence natural variation in the metabolome

Understanding genome to phenotype linkages has been greatly enabled by genomic sequencing. However, most genome analysis is typically confined to the nuclear genome. We conducted a metabolomic QTL analysis on a reciprocal RIL population structured to examine how variation in the organelle genomes affects phenotypic variation. This showed that the cytoplasmic variation had effects similar to, if not larger than, the largest individual nuclear locus. Inclusion of cytoplasmic variation into the genetic model greatly increased the explained phenotypic variation. Cytoplasmic genetic variation was a central hub in the epistatic network controlling the plant metabolome. This epistatic influence manifested such that the cytoplasmic background could alter or hide pairwise epistasis between nuclear loci. Thus, cytoplasmic genetic variation plays a central role in controlling natural variation in metabolomic networks. This suggests that cytoplasmic genomes must be included in any future analysis of natural variation. DOI: http://dx.doi.org/10.7554/eLife.00776.001.

Hierarchical nuclear and cytoplasmic genetic architectures for plant growth and defense within Arabidopsis

To understand how genetic architecture translates between phenotypic levels, we mapped the genetic architecture of growth and defense within the Arabidopsis thaliana Kas × Tsu recombinant inbred line population. We measured plant growth using traditional size measurements and size-corrected growth rates. This population contains genetic variation in both the nuclear and cytoplasmic genomes, allowing us to separate their contributions. The cytoplasmic genome regulated a significant variance in growth but not defense, which was due to cytonuclear epistasis. Furthermore, growth adhered to an infinitesimal model of genetic architecture, while defense metabolism was more of a moderate-effect model. We found a lack of concordance between quantitative trait loci (QTL) regulating defense and those regulating growth. Given the published evidence proving the link between glucosinolates and growth, this is likely a false negative result caused by the limited population size. This size limitation creates an inability to test the entire potential genetic landscape possible between these two parents. We uncovered a significant effect of glucosinolates on growth once we accounted for allelic differences in growth QTLs. Therefore, other growth QTLs can mask the effects of defense upon growth. Investigating direct links across phenotypic hierarchies is fraught with difficulty; we identify issues complicating this analysis.

Isothiocyanate concentrations and interconversion of sulforaphane to erucin in human subjects after consumption of commercial frozen broccoli compared to fresh broccoli

SCOPE

Sulforaphane (a potent anticarcinogenic isothiocyanate derived from glucoraphanin) is widely considered responsible for the protective effects of broccoli consumption. Broccoli is typically purchased fresh or frozen and cooked before consumption. We compared the bioavailability and metabolism of sulforaphane from portions of lightly cooked fresh or frozen broccoli, and investigated the bioconversion of sulforaphane to erucin.

METHODS AND RESULTS

Eighteen healthy volunteers consumed broccoli soups produced from fresh or frozen broccoli florets that had been lightly cooked and sulforaphane thio-conjugates quantified in plasma and urine. Sulforaphane bioavailability was about tenfold higher for the soups made from fresh compared to frozen broccoli, and the reduction was shown to be due to destruction of myrosinase activity by the commercial blanching-freezing process. Sulforaphane appeared in plasma and urine in its free form and as several thio-conjugates forms. Erucin N-acetyl-cysteine conjugate was a significant urinary metabolite, and it was shown that human gut microflora can produce sulforaphane, erucin, and their nitriles from glucoraphanin.

CONCLUSION

The short period of blanching used to produce commercial frozen broccoli destroys myrosinase and substantially reduces sulforaphane bioavailability. Sulforaphane was converted to erucin and excreted in urine, and it was shown that human colonic flora were capable of this conversion.

Conserved patterns of chromosome pairing and recombination in Brassica napus crosses

The patterns of chromosome pairing and recombination in two contrasting Brassica napus F1 hybrids were deduced. One hybrid was from a winter oilseed rape (WOSR) x spring oilseed rape cross, the other from a resynthesized B. napus x WOSR cross. Segregation at 211 equivalent loci assayed in the population derived from each hybrid produced two collinear genetic maps. Alignment of the maps indicated that B. napus chromosomes behaved reproducibly as 19 homologous pairs and that the 19 distinct chromosomes of B. napus each recombined with unique chromosomes from the interspecific hybrid between Brassica rapa and Brassica oleracea. This result indicated that the genomes of the diploid progenitors of amphidiploid B. napus have remained essentially unaltered since the formation of the species and that the progenitor genomes were similar to those of modern-day B. rapa and B. oleracea. The frequency and distribution of crossovers were almost indistinguishable in the two populations, suggesting that the recombination machinery of B. napus could cope easily with different degrees of genetic divergence between homologous chromosomes. Efficient recombination in wide crosses will facilitate the introgression of novel alleles into oilseed rape from B. rapa and B. oleracea (via resynthesized B. napus) and reduce linkage drag.

Engineering glucosinolates in plants: current knowledge and potential uses

Glucosinolates (GSL) and their derivatives are well known for the characteristic roles they play in plant defense as signaling molecules and as bioactive compounds for human health. More than 130 GSLs have been reported so far, and most of them belong to the Brassicaceae family. Several enzymes and transcription factors involved in the GSL biosynthesis have been studied in the model plant, Arabidopsis, and in a few other Brassica crop species. Recent studies in GSL research have defined the regulation, distribution, and degradation of GSL biosynthetic pathways; however, the underlying mechanism behind transportation of GSLs in plants is still largely unknown. This review highlights the recent advances in the metabolic engineering of GSLs in plants and discusses their potential applications.

Combining genome-wide association mapping and transcriptional networks to identify novel genes controlling glucosinolates in Arabidopsis thaliana

BACKGROUND

Genome-wide association (GWA) is gaining popularity as a means to study the architecture of complex quantitative traits, partially due to the improvement of high-throughput low-cost genotyping and phenotyping technologies. Glucosinolate (GSL) secondary metabolites within Arabidopsis spp. can serve as a model system to understand the genomic architecture of adaptive quantitative traits. GSL are key anti-herbivory defenses that impart adaptive advantages within field trials. While little is known about how variation in the external or internal environment of an organism may influence the efficiency of GWA, GSL variation is known to be highly dependent upon the external stresses and developmental processes of the plant lending it to be an excellent model for studying conditional GWA.

METHODOLOGY/PRINCIPAL FINDINGS

To understand how development and environment can influence GWA, we conducted a study using 96 Arabidopsis thaliana accessions, >40 GSL phenotypes across three conditions (one developmental comparison and one environmental comparison) and ∼230,000 SNPs. Developmental stage had dramatic effects on the outcome of GWA, with each stage identifying different loci associated with GSL traits. Further, while the molecular bases of numerous quantitative trait loci (QTL) controlling GSL traits have been identified, there is currently no estimate of how many additional genes may control natural variation in these traits. We developed a novel co-expression network approach to prioritize the thousands of GWA candidates and successfully validated a large number of these genes as influencing GSL accumulation within A. thaliana using single gene isogenic lines.

CONCLUSIONS/SIGNIFICANCE

Together, these results suggest that complex traits imparting environmentally contingent adaptive advantages are likely influenced by up to thousands of loci that are sensitive to fluctuations in the environment or developmental state of the organism. Additionally, while GWA is highly conditional upon genetics, the use of additional genomic information can rapidly identify causal loci en masse.

Branched-Chain Amino Acid Metabolism in Arabidopsis thaliana

Valine, leucine and isoleucine form the small group of branched-chain amino acids (BCAAs) classified by their small branched hydrocarbon residues. Unlike animals, plants are able to de novo synthesize these amino acids from pyruvate, 2-oxobutanoate and acetyl-CoA. In plants, biosynthesis follows the typical reaction pathways established for the formation of these amino acids in microorganisms. Val and Ile are synthesized in two parallel pathways using a single set of enzymes. The pathway to Leu branches of from the final intermediate of Val biosynthesis. The formation of this amino acid requires a three-step pathway generating a 2-oxoacid elongated by a methylene group. In Arabidopsis thaliana and other Brassicaceae, a homologous three-step pathway is also involved in Met chain elongation required for the biosynthesis of aliphatic glucosinolates, an important class of specialized metabolites in Brassicaceae. This is a prime example for the evolutionary relationship of pathways from primary and specialized metabolism. Similar to animals, plants also have the ability to degrade BCAAs. The importance of BCAA turnover has long been unclear, but now it seems apparent that the breakdown process might by relevant under certain environmental conditions. In this review, I summarize the current knowledge about BCAA metabolism, its regulation and its particular features in Arabidopsis thaliana.

Glucosinolates in Brassica vegetables: the influence of the food supply chain on intake, bioavailability and human health

Glucosinolates (GLSs) are found in Brassica vegetables. Examples of these sources include cabbage, Brussels sprouts, broccoli, cauliflower and various root vegetables (e.g. radish and turnip). A number of epidemiological studies have identified an inverse association between consumption of these vegetables and the risk of colon and rectal cancer. Animal studies have shown changes in enzyme activities and DNA damage resulting from consumption of Brassica vegetables or isothiocyanates, the breakdown products (BDP) of GLSs in the body. Mechanistic studies have begun to identify the ways in which the compounds may exert their protective action but the relevance of these studies to protective effects in the human alimentary tract is as yet unproven. In vitro studies with a number of specific isothiocyanates have suggested mechanisms that might be the basis of their chemoprotective effects. The concentration and composition of the GLSs in different plants, but also within a plant (e.g. in the seeds, roots or leaves), can vary greatly and also changes during plant development. Furthermore, the effects of various factors in the supply chain of Brassica vegetables including breeding, cultivation, storage and processing on intake and bioavailability of GLSs are extensively discussed in this paper.

Glucosinolate polymorphism in wild cabbage (Brassica oleracea) influences the structure of herbivore communities

Natural plant populations often show substantial heritable variation in chemical structure of secondary metabolites. Despite a great deal of evidence from laboratory studies that these chemicals influence herbivore behaviour and life history, there exists little evidence for the structuring of natural herbivore communities according to plant chemical profiles. Brassica oleracea (Brassicaceae) produces aliphatic glucosinolates, which break down into toxins when leaf tissue is damaged. Structural diversity in these glucosinolates is heritable, and varies considerably at two ecological scales in the UK: both within and between populations. We surveyed herbivore attack on plants producing different glucosinolates, using 12 natural B. oleracea populations. In contrast to the results of previous studies in this system, which suffered low statistical power, we found significant differential responses of herbivore species to heritable glucosinolates, both within and between plant populations. We found significant correlations between herbivore infestation rates and the presence or absence of two heritable glucosinolates: sinigrin and progoitrin. There was variation between herbivore species in the direction of response, the ecological scale at which responses were identified, and the correlations for some herbivore species changed at different times of the year. We conclude that variation in plant secondary metabolites can structure the community of herbivores that attack them, and propose that herbivore-mediated differential selection deserves further investigation as a mechanism maintaining the observed diversity of glucosinolates in wild Brassica.

Fine mapping of loci involved with glucosinolate biosynthesis in oilseed mustard (Brassica juncea) using genomic information from allied species

Fine mapping of six seed glucosinolate QTL (J2Gsl1, J3Gsl2, J9Gsl3, J16Gsl4, J17Gsl5 and J3Gsl6) (Ramchiary et al. in Theor Appl Genet 116:77-85, 2007a) was undertaken by the candidate gene approach. Based on the DNA sequences from Arabidopsis and Brassica oleracea for the different genes involved in the aliphatic glucosinolate biosynthesis, candidate genes were amplified and sequenced from high to low glucosinolate Brassica juncea lines Varuna and Heera, respectively. Of the 20 paralogues identified, 17 paralogues belonging to six gene families were mapped to 12 of the 18 linkage groups of B. juncea genome. Co-mapping of candidate genes with glucosinolate QTL revealed that the candidate gene BjuA.GSL-ELONG.a mapped to the QTL interval of J2Gsl1, BjuA.GSL-ELONG.c, BjuA.GSL-ELONG.d and BjuA.Myb28.a mapped to the QTL interval of J3Gsl2, BjuA.GSL-ALK.a mapped to the QTL interval of J3Gsl6 and BjuB.Myb28.a mapped to the QTL interval of J17Gsl5. The QTL J9Gsl3 and J16Gsl4 did not correspond to any of the mapped candidate genes. The functionality and contribution of different candidate genes/QTL was assessed by allelic variation study using phenotypic data of 785 BC(4)DH lines. It was observed that BjuA.Myb28.a and J9Gsl3 contributed significantly to the base level glucosinolate production while J16Gsl4, probably GSL-PRO, BjuA.GSL-ELONG.a and BjuA.GSL-ELONG.c contributed to the C3, C4 and C5 elongation pathways, respectively. Three A genome QTL: J2Gsl1harbouring BjuA.GSL-ELONG.a, J3Gsl2 harbouring both BjuA.GSL-ELONG.c and BjuA.Myb28.a and J9Gsl3, possibly the 'Bronowski genes', were identified as most important loci for breeding low glucosinolate B. juncea. We observed two-step genetic control of seed glucosinolate in B. juncea mainly effected by these three A genome QTL. This study, therefore, provides clues to the genetic mechanism of 'Bronowski genes' controlling the glucosinolate trait and also provides efficient markers for marker-assisted introgression of low glucosinolate trait in B. juncea.

A novel 2-oxoacid-dependent dioxygenase involved in the formation of the goiterogenic 2-hydroxybut-3-enyl glucosinolate and generalist insect resistance in Arabidopsis,

Glucosinolates are secondary metabolites found almost exclusively in the order Brassicales. They are synthesized from a variety of amino acids and can have numerous side chain modifications that control biological function. We investigated the biosynthesis of 2-hydroxybut-3-enyl glucosinolate, which has biological activities including toxicity to Caenorhabditis elegans, inhibition of seed germination, induction of goiter disease in mammals, and production of bitter flavors in Brassica vegetable crops. Arabidopsis (Arabidopsis thaliana) accessions contain three different patterns of 2-hydroxybut-3-enyl glucosinolate accumulation (present in leaves and seeds, seeds only, or absent) corresponding to three different alleles at a single locus, GSL-OH. Fine-scale mapping of the GSL-OH locus identified a 2-oxoacid-dependent dioxygenase encoded by At2g25450 required for the formation of both 2R- and 2S-2-hydroxybut-3-enyl glucosinolate from the precursor 3-butenyl glucosinolate precursor. Naturally occurring null mutations and T-DNA insertional mutations in At2g25450 exhibit a complete absence of 2-hydroxybut-3-enyl glucosinolate accumulation. Analysis of herbivory by the generalist lepidopteran Trichoplusia ni showed that production of 2-hydroxybut-3-enyl glucosinolate provides increased resistance. These results show that At2g25450 is necessary for the hydroxylation of but-3-enyl glucosinolate to 2-hydroxybut-3-enyl glucosinolate in planta and that this metabolite increases resistance to generalist herbivory.

Biochemical networks and epistasis shape the Arabidopsis thaliana metabolome

Genomic approaches have accelerated the study of the quantitative genetics that underlie phenotypic variation. These approaches associate genome-scale analyses such as transcript profiling with targeted phenotypes such as measurements of specific metabolites. Additionally, these approaches can help identify uncharacterized networks or pathways. However, little is known about the genomic architecture underlying data sets such as metabolomics or the potential of such data sets to reveal networks. To describe the genetic regulation of variation in the Arabidopsis thaliana metabolome and test our ability to integrate unknown metabolites into biochemical networks, we conducted a replicated metabolomic analysis on 210 lines of an Arabidopsis population that was previously used for targeted metabolite quantitative trait locus (QTL) and global expression QTL analysis. Metabolic traits were less heritable than the average transcript trait, suggesting that there are differences in the power to detect QTLs between transcript and metabolite traits. We used statistical analysis to identify a large number of metabolite QTLs with moderate phenotypic effects and found frequent epistatic interactions controlling a majority of the variation. The distribution of metabolite QTLs across the genome included 11 QTL clusters; 8 of these clusters were associated in an epistatic network that regulated plant central metabolism. We also generated two de novo biochemical network models from the available data, one of unknown function and the other associated with central plant metabolism.

Linking metabolic QTLs with network and cis-eQTLs controlling biosynthetic pathways

Phenotypic variation between individuals of a species is often under quantitative genetic control. Genomic analysis of gene expression polymorphisms between individuals is rapidly gaining popularity as a way to query the underlying mechanistic causes of variation between individuals. However, there is little direct evidence of a linkage between global gene expression polymorphisms and phenotypic consequences. In this report, we have mapped quantitative trait loci (QTLs)-controlling glucosinolate content in a population of 403 Arabidopsis Bay x Sha recombinant inbred lines, 211 of which were previously used to identify expression QTLs controlling the transcript levels of biosynthetic genes. In a comparative study, we have directly tested two plant biosynthetic pathways for association between polymorphisms controlling biosynthetic gene transcripts and the resulting metabolites within the Arabidopsis Bay x Sha recombinant inbred line population. In this analysis, all loci controlling expression variation also affected the accumulation of the resulting metabolites. In addition, epistasis was detected more frequently for metabolic traits compared to transcript traits, even when both traits showed similar distributions. An analysis of candidate genes for QTL-controlling networks of transcripts and metabolites suggested that the controlling factors are a mix of enzymes and regulatory factors. This analysis showed that regulatory connections can feedback from metabolism to transcripts. Surprisingly, the most likely major regulator of both transcript level for nearly the entire pathway and aliphatic glucosinolate accumulation is variation in the last enzyme in the biosynthetic pathway, AOP2. This suggests that natural variation in transcripts may significantly impact phenotypic variation, but that natural variation in metabolites or their enzymatic loci can feed back to affect the transcripts.

Mapping of QTL for resistance against the crucifer specialist herbivore Pieris brassicae in a new Arabidopsis inbred line population, Da(1)-12 x Ei-2

Background

In Arabidopsis thaliana and other crucifers, the glucosinolate-myrosinase system contributes to resistance against herbivory by generalist insects. As yet, it is unclear how crucifers defend themselves against crucifer-specialist insect herbivores.

Methodology/Principal Findings

We analyzed natural variation for resistance against two crucifer specialist lepidopteran herbivores, Pieris brassicae and Plutella xylostella, among Arabidopsis thaliana accessions and in a new Arabidopsis recombinant inbred line (RIL) population generated from the parental accessions Da(1)-12 and Ei-2. This RIL population consists of 201 individual F₈ lines genotyped with 84 PCR-based markers. We identified six QTL for resistance against Pieris herbivory, but found only one weak QTL for Plutella resistance. To elucidate potential factors causing these resistance QTL, we investigated leaf hair (trichome) density, glucosinolates and myrosinase activity, traits known to influence herbivory by generalist insects. We identified several previously unknown QTL for these traits, some of which display a complex pattern of epistatic interactions.

Conclusions/Significance

Although some trichome, glucosinolate or myrosinase QTL co-localize with Pieris QTL, none of these traits explained the resistance QTL convincingly, indicating that resistance against specialist insect herbivores is influenced by other traits than resistance against generalists.

MAM3 catalyzes the formation of all aliphatic glucosinolate chain lengths in Arabidopsis

Chain elongated, methionine (Met)-derived glucosinolates are a major class of secondary metabolites in Arabidopsis (Arabidopsis thaliana). The key enzymatic step in determining the length of the chain is the condensation of acetyl-coenzyme A with a series of omega-methylthio-2-oxoalkanoic acids, catalyzed by methylthioalkylmalate (MAM) synthases. The existence of two MAM synthases has been previously reported in the Arabidopsis ecotype Columbia: MAM1 and MAM3 (formerly known as MAM-L). Here, we describe the biochemical properties of the MAM3 enzyme, which is able to catalyze all six condensation reactions of Met chain elongation that occur in Arabidopsis. Underlining its broad substrate specificity, MAM3 also accepts a range of non-Met-derived 2-oxoacids, e.g. converting pyruvate to citramalate and 2-oxoisovalerate to isopropylmalate, a step in leucine biosynthesis. To investigate its role in vivo, we identified plant lines with mutations in MAM3 that resulted in a complete lack or greatly reduced levels of long-chain glucosinolates. This phenotype could be complemented by reintroduction of a MAM3 expression construct. Analysis of MAM3 mutants demonstrated that MAM3 catalyzes the formation of all glucosinolate chain lengths in vivo as well as in vitro, making this enzyme the major generator of glucosinolate chain length diversity in the plant. The localization of MAM3 in the chloroplast suggests that this organelle is the site of Met chain elongation.

Biology and biochemistry of glucosinolates

Glucosinolates are sulfur-rich, anionic natural products that upon hydrolysis by endogenous thioglucosidases called myrosinases produce several different products (e.g., isothiocyanates, thiocyanates, and nitriles). The hydrolysis products have many different biological activities, e.g., as defense compounds and attractants. For humans these compounds function as cancer-preventing agents, biopesticides, and flavor compounds. Since the completion of the Arabidopsis genome, glucosinolate research has made significant progress, resulting in near-complete elucidation of the core biosynthetic pathway, identification of the first regulators of the pathway, metabolic engineering of specific glucosinolate profiles to study function, as well as identification of evolutionary links to related pathways. Although much has been learned in recent years, much more awaits discovery before we fully understand how and why plants synthesize glucosinolates. This may enable us to more fully exploit the potential of these compounds in agriculture and medicine.

Genetic analysis of agronomic and quality traits in mustard (Brassica juncea)

To develop an efficient mustard (Brassica juncea) breeding programme, a better knowledge of the genetic control and relationships of the main selected characters is needed. Thus, doubled haploid (DH) lines were evaluated over 2 years in the field. Days to flowering, plant height, thousand-seed weight, fatty acid composition, seed oil content, sinigrin, gluconapin and total glucosinolate contents were determined in the DH population. The influence of seed coat colour was estimated. Results showed significant differences between yellow and brown seeds only for oil and fatty acid contents. Molecular analysis revealed that seed coat colour is associated with two Mendelian trait loci: Bjc1 [on linkage group (LG) 3] and Bjc2 (on LG6). The quantitative trait loci associated with characters-detected by composite interval mapping-were not co-localised and revealed a genetic independence. The results obtained in this study show that the most important agronomic and quality traits of brown mustard could be bred independently. Correlation between the studied traits is also discussed.

Molecular mapping of seed aliphatic glucosinolates in Brassica juncea

An RFLP genomic map with 316 loci was used to study the inheritance of aliphatic glucosinolates in Brassica juncea using doubled-haploid (DH) populations developed from a cross between RLM-514, an agronomically superior non-canola quality B. juncea (high erucic acid and high glucosinolates), and an agronomically poor canola quality B. juncea breeding line. Two QTLs (GSL-A2a and GSL-A2b) associated with 3-butenyl were consistent across years and locations, and explained 75% of the phenotypic variance in the population. Three QTLs (GSL-A2a, GSL-F, GSL-B3) affected 2-propenyl and explained 78% of the phenotypic variance in the population. For total aliphatic glucosinolates, five QTLs explained 30% to 45% of the total phenotypic variance in the population in different environments. Several QTLs (GSL-A7 and GSL-A3) were highly inconsistent in different environments. Major QTLs (GSL-A2a and GSL-A2b) associated with individual glucosinolates were non-significant for total aliphatic glucosinolates. A marker-assisted selection strategy based on QTLs associated with individual glucosinolates rather than total aliphatic glucosinolates is proposed for B. juncea.

Genetic analysis, expression and molecular characterization of BoGSL-ELONG, a major gene involved in the aliphatic glucosinolate pathway of Brassica species

We cloned a major aliphatic glucosinolate (GSL) gene, BoGSL-ELONG in Brassica oleracea, using the Arabidopsis sequence database. We based our work on an Arabidopsis candidate gene forming part of a gene family coding for isopropyl malate synthetase-like enzymes (IPMS). This gene is presumably responsible for synthesis of GSL possessing side chains consisting of four carbons (4C). The similarity of the Brassica homolog IPMS-Bo from broccoli to its Arabidopsis counterpart IPMS-At was on the order of 78%, both sharing the same number of exons. A nonfunctional allele of the BoGSL-ELONG gene from white cauliflower, based on the absence of 4C GSL in this crop, displayed a 30-bp deletion, which allowed us to develop a codominant marker for 4C-GSL. Gene expression analysis based on RT-PCR revealed a splicing site mutation in the white cauliflower allele. This resulted in a longer transcript containing intron 3, which failed to excise. Perfect cosegregation was observed for broccoli and cauliflower alleles at the IPMS-Bo gene and 4C-GSL content, strongly indicating that this gene indeed corresponds to BoGSL-ELONG. Cloning of two other major genes, BoGSL-ALK and BoGSL-PRO, is underway. The availability of these genes and BoGSL-ELONG is essential for the manipulation of the aliphatic GSL profile of B. oleracea.

Glucosinolate research in the Arabidopsis era

The wide range of biological activities of products derived from the glucosinolate-myrosinase system is biologically and economically important. On the one hand, the degradation products of glucosinolates play an important role in the defence of plants against herbivores. On the other hand, these compounds have toxic (e.g. goitrogenic) as well as protective (e.g. cancer-preventing) effects in higher animals and humans. There is a strong interest in the ability to regulate and optimize the levels of individual glucosinolates tissue-specifically to improve the nutritional value and pest resistance of crops. Recent advances in our understanding of glucosinolate biosynthesis have brought us closer to this goal.

A gene controlling variation in Arabidopsis glucosinolate composition is part of the methionine chain elongation pathway

Arabidopsis and other Brassicaceae produce an enormous diversity of aliphatic glucosinolates, a group of methionine (Met)-derived plant secondary compounds containing a beta-thio-glucose moiety, a sulfonated oxime, and a variable side chain. We fine-scale mapped GSL-ELONG, a locus controlling variation in the side-chain length of aliphatic glucosinolates. Within this locus, a polymorphic gene was identified that determines whether Met is extended predominantly by either one or by two methylene groups to produce aliphatic glucosinolates with either three- or four-carbon side chains. Two allelic mutants deficient in four-carbon side-chain glucosinolates were shown to contain independent missense mutations within this gene. In cell-free enzyme assays, a heterologously expressed cDNA from this locus was capable of condensing 2-oxo-4-methylthiobutanoic acid with acetyl-coenzyme A, the initial reaction in Met chain elongation. The gene methylthioalkylmalate synthase1 (MAM1) is a member of a gene family sharing approximately 60% amino acid sequence similarity with 2-isopropylmalate synthase, an enzyme of leucine biosynthesis that condenses 2-oxo-3-methylbutanoate with acetyl-coenzyme A.

Genetic control of natural variation in Arabidopsis glucosinolate accumulation

Glucosinolates are biologically active secondary metabolites of the Brassicaceae and related plant families that influence plant/insect interactions. Specific glucosinolates can act as feeding deterrents or stimulants, depending upon the insect species. Hence, natural selection might favor the presence of diverse glucosinolate profiles within a given species. We determined quantitative and qualitative variation in glucosinolates in the leaves and seeds of 39 Arabidopsis ecotypes. We identified 34 different glucosinolates, of which the majority are chain-elongated compounds derived from methionine. Polymorphism at only five loci was sufficient to generate 14 qualitatitvely different leaf glucosinolate profiles. Thus, there appears to be a modular genetic system regulating glucosinolate profiles in Arabidopsis. This system allows the rapid generation of new glucosinolate combinations in response to changing herbivory or other selective pressures. In addition to the qualitative variation in glucosinolate profiles, we found a nearly 20-fold difference in the quantity of total aliphatic glucosinolates and were able to identify a single locus that controls nearly three-quarters of this variation.

The TASTY locus on chromosome 1 of Arabidopsis affects feeding of the insect herbivore Trichoplusia ni

The generalist insect herbivore Trichoplusia ni (cabbage looper) readily consumes Arabidopsis and can complete its entire life cycle on this plant. Natural isolates (ecotypes) of Arabidopsis are not equally susceptible to T. ni feeding. While some are hardly touched by T. ni, others are eaten completely to the ground. Comparison of two commonly studied Arabidopsis ecotypes in choice experiments showed that Columbia is considerably more resistant than Landsberg erecta. In no-choice experiments, where larvae were confined on one or the other ecotype, weight gain was more rapid on Landsberg erecta than on Columbia. Genetic mapping of this difference in insect susceptibility using recombinant inbred lines resulted in the discovery of the TASTY locus near 85 cM on chromosome 1 of Arabidopsis. The resistant allele of this locus is in the Columbia ecotype, and an F(1) hybrid has a sensitive phenotype that is similar to that of Landsberg erecta. The TASTY locus is distinct from known genetic differences between Columbia and Landsberg erecta that affect glucosinolate content, trichome density, disease resistance, and flowering time.

The biosynthesis of benzoic acid glucosinolate esters in Arabidopsis thaliana

The siliques and seeds of Arabidopsis thaliana accumulate a series of glucosinolates containing an alkyl side chain of varying length with a terminal benzoate ester function. The biosynthesis of these unusual nitrogen- and sulfur-containing natural products was investigated by feeding isotopically-labeled precursors to detached flowering stems. Glucosinolates were extracted, purified and analyzed by tandem mass spectrometry. Phenylalanine and benzoic acid were incorporated into the benzoate ester function, and methionine and acetate were incorporated into the aliphatic portion of the side chain in a position-specific manner. The labeling patterns observed were consistent with the chain extension of methionine by a three-step elongation cycle which begins with the condensation of acetyl-CoA with a 2-oxo acid derived from methionine and ends with an oxidative decarboxylation forming a new 2-oxo acid with an additional methylene group. Incorporation of desulfo-4-methylthiobutyl glucosinolate into 4-benzoyloxybutyl olucosinolate suggested chain-extended methionine derivatives are first converted to their corresponding methylthioalkyl glucosinolates with further side chain modification occurring later. Transformation of the methylthiol function to a hydroxyl group is followed by esterification with benzoic acid. The siliques appear to possess the complete machinery for carrying out all of the reactions in the biosyntheis of these complex glucosinolates.