SUMMARYThe ability to evolve novel metabolites has been instrumental for the defence of plants against antagonists. A few species in the Barbarea genus are the only crucifers known to produce saponins, some of which make plants resistant to specialist herbivores, like Plutella xylostella, the diamondback moth. Genetic mapping in Barbarea vulgaris revealed that genes for saponin biosynthesis are not clustered but are located in different linkage groups. Using co-location with quantitative trait loci (QTLs) for resistance, transcriptome and genome sequences, we identified two 2,3-oxidosqualene cyclases that form the major triterpenoid backbones. LUP2 mainly produces lupeol, and is preferentially expressed in insect-susceptible B. vulgaris plants, whereas LUP5 produces b-amyrin and a-amyrin, and is preferentially expressed in resistant plants; b-amyrin is the backbone for the resistance-conferring saponins in Barbarea. Two loci for cytochromes P450, predicted to add functional groups to the saponin backbone, were identified: CYP72As co-localized with insect resistance, whereas CYP716As did not. When B. vulgaris sapogenin biosynthesis genes were transiently expressed by CPMV-HT technology in Nicotiana benthamiana, high levels of hydroxylated and carboxylated triterpenoid structures accumulated, including oleanolic acid, which is a precursor of the major resistanceconferring saponins. When the B. vulgaris gene for sapogenin 3-O-glucosylation was co-expressed, the insect deterrent 3-O-oleanolic acid monoglucoside accumulated, as well as triterpene structures with up to six hexoses, demonstrating that N. benthamiana further decorates the monoglucosides. We argue that saponin biosynthesis in the Barbarea genus evolved by a neofunctionalized glucosyl transferase, whereas the difference between resistant and susceptible B. vulgaris chemotypes evolved by different expression of oxidosqualene cyclases (OSCs).

The genus Barbarea has emerged as a model for evolution and ecology of plant defense compounds, due to its unusual glucosinolate profile and production of saponins, unique to the Brassicaceae. One species, B. vulgaris, includes two ‘types’, G-type and P-type that differ in trichome density, and their glucosinolate and saponin profiles. A key difference is the stereochemistry of hydroxylation of their common phenethylglucosinolate backbone, leading to epimeric glucobarbarins. Here we report a draft genome sequence of the G-type, and re-sequencing of the P-type for comparison. This enables us to identify candidate genes underlying glucosinolate diversity, trichome density, and study the genetics of biochemical variation for glucosinolate and saponins. B. vulgaris is resistant to the diamondback moth, and may be exploited for “dead-end” trap cropping where glucosinolates stimulate oviposition and saponins deter larvae to the extent that they die. The B. vulgaris genome will promote the study of mechanisms in ecological biochemistry to benefit crop resistance breeding.

This study identifies six UGT73Cs all able to glucosylate sapogenins at positions 3 and/or 28 which demonstrates that B. vulgaris has a much richer arsenal of UGTs involved in saponin biosynthesis than initially anticipated. The wild cruciferous plant Barbarea vulgaris is resistant to some insects due to accumulation of two monodesmosidic triterpenoid saponins, oleanolic acid 3-O-β-cellobioside and hederagenin 3-O-β-cellobioside. Insect resistance depends on the structure of the sapogenin aglycone and the glycosylation pattern. The B. vulgaris saponin profile is complex with at least 49 saponin-like metabolites, derived from eight sapogenins and including up to five monosaccharide units. Two B. vulgaris UDP-glycosyltransferases, UGT73C11 and UGT73C13, O-glucosylate sapogenins at positions 3 and 28, forming mainly 3-O-β-D-glucosides. The aim of this study was to identify UGTs responsible for the diverse saponin oligoglycoside moieties observed in B. vulgaris. Twenty UGT genes from the insect resistant genotype were selected and heterologously expressed in Nicotiana benthamiana and/or Escherichia coli. The extracts were screened for their ability to glycosylate sapogenins (oleanolic acid, hederagenin), the hormone 24-epibrassinolide and sapogenin monoglucosides (hederagenin and oleanolic acid 3-O-β-D-glucosides). Six UGTs from the UGT73C subfamily were able to glucosylate both sapogenins and both monoglucosides at positions 3 and/or 28. Some UGTs formed bisdesmosidic saponins efficiently. At least four UGT73C genes were localized in a tandem array with UGT73C11 and possibly UGT73C13. This organization most likely reflects duplication events followed by sub- and neofunctionalization. Indeed, signs of positive selection on several amino acid sites were identified and modelled to be localized on the UGT protein surface. This tandem array is proposed to initiate higher order bisdesmosidic glycosylation of B. vulgaris saponins, leading to the recently discovered saponin structural diversity, however, not directly to known cellobiosidic saponins.

Glucosinolates are found in plants of the order Brassicales and hydrolyzed to different breakdown products, particularly after tissue damage. In Barbarea vulgaris R.Br. (Brassicaceae), the dominant glucosinolate in the investigated "G-type" is glucobarbarin, (S)-2-hydroxy-2-phenylethylglucosinolate. Formation of the nitrile from glucobarbarin was observed in vitro, while a previously suggested thioamide (synonym thionamide) was not confirmed. Resedine (5-phenyl-1,3-oxazolidin-2-one) was detected after glucobarbarin hydrolysis in crushed B. vulgaris leaves and siliques, but not in intact parts. The abundance increased for several hours after completion of hydrolysis. The corresponding 1,3-oxazolidine-2-thione (OAT), with the common name barbarin, was also formed, and appeared to be the precursor of resedine. Addition of each of two non-endogenous OATs, (S)-5-ethyl-5-methylOAT and (R)-5-vinylOAT (R-goitrin), to a leaf homogenate resulted in formation of the corresponding 1,3-oxazolidin-2-ones (OAOs), confirming the metabolic connection of OAT to OAO. Formation of OAOs was inhibited by prior brief heating of the homogenate, suggesting enzyme involvement. We suggest the conversion of OATs to OAOs to be catalyzed by an enzyme ("oxazolidinethionase") responsible for turnover of OAT formed in intact plants. Resedine had been reported as an alkaloid from another species - Reseda luteola L. (Resedaceae) - naturally containing the glucosinolate glucobarbarin. However, resedine was not detected in intact R. luteola plants, but formed after tissue damage. The formation of resedine in two families suggests a broad distribution of putative OATases in the Brassicales; potentially involved in glucosinolate turnover that needs myrosinase activity as the committed step. In agreement with the proposed function of OATase, several candidate genes for myrosinases in glucosinolate turnover in intact plants were discovered in the B. vulgaris genome. We also suggest that biotechnological conversion of OATs to OAOs might improve the nutritional value of Brassicales protein. HPLC-MS/MS methods for detection of these glucobarbarin products are described.

In the wild cruciferous wintercress (Barbarea vulgaris), β-amyrin-derived saponins are involved in resistance against insect herbivores like the major agricultural pest diamondback moth (Plutella xylostella). Enzymes belonging to the 2,3-oxidosqualene cyclase family have been identified and characterized in B. vulgaris G-type and P-type plants that differ in their natural habitat, insect resistance and saponin content. Both G-type and P-type plants possess highly similar 2,3-oxidosqualene cyclase enzymes that mainly produce β-amyrin (Barbarea vulgaris Lupeol synthase 5 G-Type; BvLUP5-G) or α-amyrin (Barbarea vulgaris Lupeol synthase 5 P-Type; BvLUP5-P), respectively. Despite the difference in product formation, the two BvLUP5 enzymes are 98% identical at the amino acid level. This provides a unique opportunity to investigate determinants of product formation, using the B. vulgaris 2,3-oxidosqualene cyclase enzymes as a model for studying amino acid residues that determine differences in product formation. In this study, we identified two amino acid residues at position 121 and 735 that are responsible for the dominant changes in generated product ratios of β-amyrin and α-amyrin in both BvLUP5 enzymes. These amino acid residues have not previously been highlighted as directly involved in 2,3-oxidosqualene cyclase product specificity. Our results highlight the functional diversity and promiscuity of 2,3-oxidosqualene cyclase enzymes. These enzymes serve as important mediators of metabolic plasticity throughout plant evolution.