The inducibility of the glucosinolate resistance mechanism is an energy-saving strategy for plants, but whether induction would still be triggered by glucosinolate-tolerant Plutella xylostella (diamondback moth, DBM) after a plant had evolved a new resistance mechanism (e.g., saponins in Barbara vulgaris) was unknown. In B. vulgaris, aromatic glucosinolates derived from homo-phenylalanine are the dominant glucosinolates, but their biosynthesis pathway was unclear. In this study, we used G-type (pest-resistant) and P-type (pest-susceptible) B. vulgaris to compare glucosinolate levels and the expression profiles of their biosynthesis genes before and after infestation by DBM larvae. Two different stereoisomers of hydroxylated aromatic glucosinolates are dominant in G- and P-type B. vulgaris, respectively, and are induced by DBM. The transcripts of genes in the glucosinolate biosynthesis pathway and their corresponding transcription factors were identified from an Illumina dataset of G- and P-type B. vulgaris. Many genes involved or potentially involved in glucosinolate biosynthesis were induced in both plant types. The expression patterns of six DBM induced genes were validated by quantitative PCR (qPCR), while six long-fragment genes were validated by molecular cloning. The core structure biosynthetic genes showed high sequence similarities between the two genotypes. In contrast, the sequence identity of two apparent side chain modification genes, the SHO gene in the G-type and the RHO in P-type plants, showed only 77.50% identity in coding DNA sequences and 65.48% identity in deduced amino acid sequences. The homology to GS-OH in Arabidopsis, DBM induction of the transcript and a series of qPCR and glucosinolate analyses of G-type, P-type and F1 plants indicated that these genes control the production of S and R isomers of 2-hydroxy-2-phenylethyl glucosinolate. These glucosinolates were significantly induced by P. xylostella larvae in both the susceptiple P-type and the resistant G-type, even though saponins are the main DBM-resistance causing metabolites in G-type plants. Indol-3-ylmethylglucosinolate was induced in the G-type only. These data will aid our understanding of the biosynthesis and induction of aromatic glucosinolates at the molecular level and also increase our knowledge of the complex mechanisms underpinning defense induction in plants.

BackgroundThe diamondback moth (DBM, Plutella xylostella) is a crucifer-specific pest that causes significant crop losses worldwide. Barbarea vulgaris (Brassicaceae) can resist DBM and other herbivorous insects by producing feeding-deterrent triterpenoid saponins. Plant breeders have long aimed to transfer this insect resistance to other crops. However, a lack of knowledge on the biosynthetic pathways and regulatory networks of these insecticidal saponins has hindered their practical application. A pyrosequencing-based transcriptome analysis of B. vulgaris during DBM larval feeding was performed to identify genes and gene networks responsible for saponin biosynthesis and its regulation at the genome level.Principal FindingsApproximately 1.22, 1.19, 1.16, 1.23, 1.16, 1.20, and 2.39 giga base pairs of clean nucleotides were generated from B. vulgaris transcriptomes sampled 1, 4, 8, 12, 24, and 48 h after onset of P. xylostella feeding and from non-inoculated controls, respectively. De novo assembly using all data of the seven transcriptomes generated 39,531 unigenes. A total of 37,780 (95.57%) unigenes were annotated, 14,399 of which were assigned to one or more gene ontology terms and 19,620 of which were assigned to 126 known pathways. Expression profiles revealed 2,016–4,685 up-regulated and 557–5188 down-regulated transcripts. Secondary metabolic pathways, such as those of terpenoids, glucosinolates, and phenylpropanoids, and its related regulators were elevated. Candidate genes for the triterpene saponin pathway were found in the transcriptome. Orthological analysis of the transcriptome with four other crucifer transcriptomes identified 592 B. vulgaris-specific gene families with a P-value cutoff of 1e−5.ConclusionThis study presents the first comprehensive transcriptome analysis of B. vulgaris subjected to a series of DBM feedings. The biosynthetic and regulatory pathways of triterpenoid saponins and other DBM deterrent metabolites in this plant were classified. The results of this study will provide useful data for future investigations on pest-resistance phytochemistry and plant breeding.

Radish (Raphanus sativus) is an important cruciferous root crop with a close relationship to Chinese cabbage (Brassica rapa). RT-qPCR is used extensively to evaluate the expression levels of target genes, and accurate measurement of target gene expression with this method is determined by the valid reference genes used for data nomalization in different experimental conditions. Screening for appropriate reference genes with stable expression based on RT-qPCR data is important for gene expression and functional analysis research in radish and its relatives. However, many researches have thought that almost no single reference gene is widely suitable for all experimental conditions, and few researchers have paid attention to the validation of reference genes in radish gene expression analysis. In the present study, 12 candidate reference genes were selected for analysis. Their expression in 28 samples, including 20 radish samples from different organs and conditions, four Chinese cabbage organs and four organs of their distant hybrid, was assessed by RT-qPCR and then five software tools—ΔCt, geNorm, NormFinder, BestKeeper and RefFinder—were used to compare their expression stability. The results showed that the most suitable reference genes were different in different organs and conditions. GAPDH, DSS1, and UP2 were optimal reference genes for gene expression analysis in all organs and conditions in radish. UPR, GSNOR1, and ACTIN2/7 were the most stable reference genes in different radish organs. UP2 and GAPDH were suitable reference genes for radish pistil development studies. RPII, UBC9, and GAPDH had the most stable expression in radish under various stresses. DSS1, UP2, and TEF2 were the optimal reference genes for Chinese cabbage organs, whereas TUA was optimal for the distant hybrid. UP2, and TEF2 were appropriate reference genes for all of the samples together. The optimal reference genes we identified, UP2, GAPDH, UPR, and GSNOR1 were verified by normalizing the expression patterns of YAB3, RPL, and FUL. These results will provide important information for selecting target reference genes in different research contexts and improve the accuracy and precision of gene expression analysis for radish, Chinese cabbage and their distant hybrid.