Rational gene stacking for seed-specific carbon partitioning to simultaneously enhance lipid and protein contents in Arabidopsis.

Increasing global population demands the development of oilseed crops such as canola-type Brassica napus L. and soybean varieties with high protein and oil content, despite the known negative correlation between them. We hypothesized that reallocating seed carbon from cellulose, a major compound of fiber, to these storage compounds via fine-tuned gene stacking could achieve this dual goal. We tested this hypothesis in Arabidopsis thaliana with a three-pronged gene stacking approach: (i) down-regulation of Arabidopsis CELLULOSE SYNTHASE 1 with RNAi (AtCESA1-RNAi) to reduce cellulose content; (ii) overexpression (OE) of native B. napus DIACYLGLYCEROL ACYLTRANSFERASE 1 (BnDGAT1) and a performance-enhanced variant BnDGAT1-L441P to increase or maintain oil content, respectively; and (iii) OE of protein biosynthesis-related genes, Arabidopsis AMINO ACID PERMEASE 1 (AtAAP1), ALANINE AMINOTRANSFERASE 1 (AtALAAT1), and ASPARAGINE SYNTHASE 1 (AtASN1), respectively, to increase seed protein content. The best line, AtCESA1-RNAi/BnDGAT1-L441P-OE/AtAAP1-OE, exhibited a relative increase of 19.5% in crude seed protein and 3.2% in total lipid content, and a 42.2% decrease in cellulose content compared with the empty vector control lines, alongside an 89% increase in seed yield. Collectively, the results demonstrate that fine-tuned gene stacking can mitigate the trade-offs between protein and oil accumulation for engineering high-value seed traits.