Comparative transcriptome and weighted correlation network analyses reveal candidate genes involved in chlorogenic acid biosynthesis in sweet potato

Abstract

Chlorogenic acids (CGAs) are important secondary metabolites produced in sweet potato. However, the mechanisms of their biosynthesis and regulation remain unclear. To identify potential genes involved in CGA biosynthesis, analysis of the dynamic changes in CGA components and RNA sequencing were performed on young leaves (YL), mature leaves (ML), young stems (YS), mature stems (MS) and storage roots (SR). Accordingly, we found that the accumulation of six CGA components varied among the different tissues and developmental stages, with YS and YL recording the highest levels, while SR exhibited low levels. Moreover, the transcriptome analysis yielded 59,287 unigenes, 3,767 of which were related to secondary-metabolite pathways. The differentially expressed genes (DEGs) were identified based on CGA content levels by comparing the different samples, including ML vs. YL, MS vs. YS, SR vs. YL and SR vs. YS. A total of 501 common DEGs were identified, and these were mainly implicated in the secondary metabolites biosynthesis. Additionally, eight co-expressed gene modules were identified following weighted gene co-expression network analysis, while genes in darkgrey module were highly associated with CGA accumulation. Darkgrey module analysis revealed that 12 unigenes encoding crucial enzymes (PAL, 4CL, C4H, C3H and HCT/HQT) and 42 unigenes encoding transcription factors (MYB, bHLH, WD40, WRKY, ERF, MADS, GARS, bZIP and zinc finger protein) had similar expression patterns with change trends of CGAs, suggesting their potential roles in CGA metabolism. Our findings provide new insights into the biosynthesis and regulatory mechanisms of CGA pathway, and will inform future efforts to build a genetically improve sweet potato through the breeding of high CGA content varieties.