Integrated metabolome and transcriptome revealed the flavonoid biosynthetic pathway in developing Vernonia amygdalina leaves.

Background Vernonia amygdalina as a tropical horticultural crop has been widely used for medicinal herb, feed, and vegetable. Recently, increasing studies revealed that this species possesses multiple pharmacological properties. Notably, V. amygdalina leaves possess an abundance of flavonoids, but the specific profiles of flavonoids and the mechanisms of fl avonoid bi osynthesis in developing leaves are largely unknown. Methods The total flavonoids of V. amygdalina leaves were detected using ultraviolet spectrophotometer. The temporal flavonoid profiles of V. amygdalina leaves were analyzed by LC-MS. The transcriptome analysis of V. amygdalina leaves was performed by Illumina sequencing. Functional annotation and differential expression analysis of V. amygdalina genes were performed by Blast2GO v2.3.5 and RSEM v1.2.31, respectively. qRT-PCR analysis was used to verify the gene expressions in developing V. amygdalina leaves. Results By LC-MS analysis, three substrates (p-coumaric acid, trans-cinnamic acid, and phenylalanine) for flavonoid biosynthesis were identified in V. amygdalina leaves. Additionally, 42 flavonoids were identified from V. amygdalina leaves, including six dihydroflavones, 14 flavones, eight isoflavones, nine flavonols, two xanthones, one chalcone, one cyanidin, and one dihydroflavonol. Glycosylation and methylation were common at the hydroxy group of C3, C7, and C4' positions. Moreover, dynamic patterns of different flavonoids showed diversity. By Illumina sequencing, the obtained over 200 million valid reads were assembled into 60,422 genes. Blast analysis indicated that 31,872 genes were annotated at least in one of public databases. Greatly increasing molecular resources makes up for the lack of gene information in V. amygdalina. By digital expression profiling and qRT-PCR, we specifically characterized some key enzymes, such as Va-PAL1, Va-PAL4, Va-C4H1, Va-4CL3, Va-ACC1, Va-CHS1, Va-CHI, Va-FNSII, and Va-IFS3, involved in flavonoid biosynthesis. Importantly, integrated metabolome and transcriptome data of V. amygdalina leaves, we systematically constructed a flavonoid biosynthetic pathway with regards to material supplying, flavonoid scaffold biosynthesis, and flavonoid modifications. Our findings contribute significantly to understand the underlying mechanisms of flavonoid biosynthesis in V. amygdalina leaves, and also provide valuable information for potential metabolic engineering.