GSH conjugation with intermediates is required for the biosynthesis of glucosinolate (GSL) by serving as a sulfur supply. Glutathione-S-transferases (GSTs) primarily work on GSH conjugation, suggesting their involvement in GSL metabolism. Although several GSTs, including GSTF11 and GSTU20, have been recently postulated to act in GSL biosynthesis, molecular evidence is lacking. Here, we demonstrated that GSTF11 and GSTU20 play nonredundant, although partially overlapping, roles in aliphatic GSL biosynthesis. In addition, GSTU20 plays a more important role than GSTF11, which is manifested by the greater loss of aliphatic GSLs associated with GSTU20 mutant and a greater number of differentially expressed genes in GSTU20 mutant compared to GSTF11 mutant. Moreover, a double mutation leads to a greater aggregate loss of aliphatic GSLs, suggesting that GSTU20 and GSTF11 may function in GSL biosynthesis in a dosage-dependent manner. Together, our results provide direct evidence that GSTU20 and GSTF11 are critically involved in aliphatic GSL biosynthesis, filling the knowledge gap that has been speculated in recent decades.

Total RNA was extracted from three-week-old plants with TRIzol (Invitrogen) and purified using a GeneJET plant RNA purification kit (Thermo Fisher Scientific). RNA integrity and concentration were assessed by gel electrophoresis and using the Qubit® 2.0 Fluorometer (Thermo Fisher Scientific). RNA (1.5 μg) was used for cDNA library preparation with the NEBNext® Ultra™ RNA Library Prep Kit for Illumina® (New England Biolabs, NEB) following the manufacturer’s protocol. Library quality was monitored using an Agilent Bioanalyzer 2100 (Agilent Technologies). The cDNA libraries were sequenced on an Illumina HiSeq 2500 platform, and 150-bp paired-end reads were generated.

Gene functional annotation was conducted by aligning reads to the Arabidopsis genome sequence (TAIR 10). Following alignment, the count of mapped reads from each sample was derived and normalized as RPKM (reads per kilobase of exon model per million mapped reads). Differentially expressed genes (DEGs) were identified using the DESeq R package (1.10.1). Genes with log2 fold change ≥ 1 and an FDR adjusted p value less than 0.05 were considered DEGs. GO term enrichment of DEGs was analyzed using tools in TAIR.