Integration of RNA-Seq profiling with genome-wide association study predicts candidate genes for oil accumulation in soybean

Abstract

The most important objectives of soybean (Glycine max (L.) Merr.) breeding are to increase oil content and to improve oil quality. Although the biochemical processes of oil accumulation in oilseeds are very clear, knowledge of their underlying genetic mechanisms and regulation is limited. We performed RNA-Seq of soybean seeds from six accessions with high, medium and low seed oil contents. Through comparative genome analysis, promoter-enrichment study, and protein–protein interaction (PPI) analysis, 80 lipid-metabolism-related genes and 31 transcription factors were detected. The pathways of fatty acid elongation, desaturation and export from plastid (P = 2.53E-4), and triacylglycerol biosynthesis (P = 2.31E-8), were significantly over-represented in accessions with high total oil content. Further, in an integration analysis of RNA-Seq and a genome-wide association study (GWAS) database, 62 candidate genes were found to be associated with seed oil content, 73 with oleic acid content, and 83 with linolenic acid content. Of these, 60 genes were found to be involved mainly in metabolism of lipids (25), carbohydrates (24), and amino acids (11). Thirty are known oil-synthesis-related genes; LOX1, CYP93D1 and GPT2 for oil content, SAD and FAD2 for oleic acid, and FAD2, CYP89A6 and GPT2 for linolenic acid were detected twice. There were 22 genes found to be associated with at least two oil-related traits, and of 154 pairs of PPIs, two genes for each pair of 95 PPIs (62%) were found to be associated with various oil-related traits, indicating the genetic foundations of oil-related traits. Three transcription factor genes were found to be associated with oil-related traits: HRE2 (Glyma.10G016500), ERF12 (Glyma.13G236600) and WRKY6 (Glyma.15G110300). This study provides an efficient strategy for further discovery of mechanisms of oil composition and accumulation.

Integrating omics approaches to discover and prioritize candidate genes involved in oil biosynthesis in soybean

“…Integration of the genes found by Yang et al and Niu et al (Niu et al, 2020; Yang et al, 2019). The heatmap shows the presence/absence of these genes in genomic regions for oil-related traits.…”

“…This remarkable representation support TFs as key regulators of transcriptional programs involved in oil accumulation during seed development. Originally, among the DEGs reported by Yang et al and Niu et al, 9.68% and 29.33% are TFs, respectively (Niu et al, 2020; Yang et al, 2019). This trend has prompted us to further investigate TFs related with oil content in soybean, which are discussed in the next section.…”

“…We also gathered other datasets to enrich our analyses (Supplementary figure S2): annotated transcription factors (TFs) (Moharana & Venancio, 2020; Niu et al, 2020; Yao et al, 2020); hub genes from co-expression modules for oil-related traits (Qi et al, 2018; Yang et al, 2019); differentially expressed genes (DEGs) between high and low-oil soybean accessions (Niu et al, 2020) and DEGs for a critical period of oil accumulation during soybean seed development (Yang et al, 2019). We leveraged these complementary datasets to better understand the roles of these genes and prioritize candidates for crop improvement strategies, as described below.…”

“…A recent study have identified 126 DEGs, during seed development of a single soybean line with ∼19% of oil content in a critical period of oil accumulation in seeds (Yang et al, 2019). Another study reported 359 DEGs, from comparisons among six contrasting accessions for oil content (11.9 to 12.5%; 17.2 to 17.8% and 20.9 to 22.3%) during seed development (Niu et al, 2020). By comparing the DEGs from both studies, we found only 16 genes in common.…”

“…In spite of the large volume of association and linkage mapping, genomic, and transcriptomic datasets, integrative approaches remain scarce (Liu et al, 2020; Niu et al, 2020; Ronne et al, 2019), resulting in an incomplete picture of metabolic and regulatory genes determining soybean oil content. Here, we integrate large-scale datasets from various sources to define stable genomic regions and identify the most promising genes involved in oil-related traits therein.…”