Map-based cloning and CRISPR/Cas9-based editing uncover BoNA1 as the causal gene for the no-anthocyanin-accumulation phenotype in curly kale (Brassica oleracea var. sabellica)

Abstract

Abstract Brassica oleracea comprises several important vegetable and ornamental crops, including curly kale, ornamental kale, cabbage, broccoli, and others. The accumulation of anthocyanins, important secondary metabolites valuable to human health, in these plants varies widely and is responsible for their pink to dark purple colors. Some curly kale varieties lack anthocyanins, making these plants completely green. The genetic basis of this trait is still unknown. We crossed the curly kale inbred line BK2019 (without anthocyanins) with the cabbage inbred line YL1 (with anthocyanins) and the Chinese kale inbred line TO1000 (with anthocyanins) to generate segregating populations. The no-anthocyanin trait was genetically controlled by a recessive gene, bona1. We generated a linkage map and mapped bona1 to a 256-kb interval on C09. We identified one candidate gene, Bo9g058630, in the target genomic region; this gene is homologous to AT5G42800, which encodes a dihydroflavonol-4-reductase-like (DFR-like) protein in Arabidopsis. In BK2019, a 1-bp insertion was observed in the second exon of Bo9g058630 and directly produced a stop codon. To verify the candidate gene function, CRISPR/Cas9 gene editing technology was applied to knock out Bo9g058630. We generated three bona1 mutants, two of which were completely green with no anthocyanins, confirming that Bo9g058630 corresponds to BoNA1. Different insertion/deletion mutations in BoNA1 exons were found in all six of the other no-anthocyanin kale varieties examined, supporting that independent disruption of BoNA1 resulted in no-anthocyanin varieties of B. oleracea. This study improves the understanding of the regulation mechanism of anthocyanin accumulation in B. oleracea subspecies.