Identification and Characterization of Soybean Pigmentation Genes and Metabolic Engineering Seed Coat Color to Enable Visual Identification of Transgenic Grains

The inability to effectively detect and quantify the levels of adventitious presence of genetically modified (GM) plant grains in non-GM shipments represents a key issue that threatens the adoption of GM commodities in the worldwide market. A potentially safe and economical solution for effective detection is to produce distinct colors in GM grains of natural products by metabolic engineering. Successful attempts to engineer colored GM grains have been limited to the overexpression of plant pigment (anthocyanin) transcription factors in Arabidopsis, tobacco, and maize. The question of our research was whether soybean grain color could be manipulated by the suppression of late-stage pigment genes. The potential benefit of this approach would be to impart less unintended effects on the plant system than the induction of numerous genes by overexpression of a transcription factor. Towards testing our hypothesis, I first identified the anthocyanin branchpoint gene UGT78K1 encoding UDP-glycose:flavonoid-3-O-glycosyltransferase (UF3GT) from the black soybean seed coat and demonstrated its encoded activity in vitro, and by complementation of the Arabidopsis ugt78d2 knock-out mutation. Subsequently, I identified a second UF3GT gene (UGT78K2 ) and the gene OMT5 coding anthocyanin-3'-O-methyltransferase (AOMT) from the black seed coat by comparing metabolite and transcriptome data between isogenic black and brown soybean seed coats differing in alleles of the R locus. The proanthocyanidin (PA) branchpoint genes ANR1 and ANR2 coding anthocyanidin reductase (ANR) enzymes were then isolated from the brown soybean seed coat using RACE and the activities of their encoded enzymes were demonstrated in vitro . In the latter two studies we used liquid chromatography- tandem mass spectrometry (LC-MS/MS) and microarray and/or quantitative (q)RT-PCR for combined analyses of seed coat metabolite and gene expression data from brown, black, and red-brown soybean genotypes. From this data we hypothesized and confirmed that a red-brown soybean grain color could be engineered by the simultaneous suppression of ANR1 and ANR2 using RNA interference (RNAi) in brown soybean. The underlying mechanism was identified to involve positive feedback and feedforward mechanisms in the flavonoid pathway to redirect metabolic flux into anthocyanin biosynthesis. These results represent a novel approach to engineering pigmentation in plant tissues.