Global Transcriptome and Co-Expression Network Analyses Revealed Hub Genes Controlling Seed Size/Weight and/or Oil Content in Peanut

Abstract

Cultivated peanut (Arachis hypogaea L.) is an important economic and oilseed crop worldwide, providing high-quality edible oil and high protein content. Seed size/weight and oil content are two important determinants of yield and quality in peanut breeding. To identify key regulators controlling these two traits, two peanut cultivars with contrasting phenotypes were compared to each other, one having a larger seed size and higher oil content (Zhonghua16, ZH16 for short), while the second cultivar had smaller-sized seeds and lower oil content (Zhonghua6, ZH6). Whole transcriptome analyses were performed on these two cultivars at four stages of seed development. The results showed that ~40% of the expressed genes were stage-specific in each cultivar during seed development, especially at the early stage of development. In addition, we identified a total of 5356 differentially expressed genes (DEGs) between ZH16 and ZH6 across four development stages. Weighted gene co-expression network analysis (WGCNA) based on DEGs revealed multiple hub genes with potential roles in seed size/weight and/or oil content. These hub genes were mainly involved in transcription factors (TFs), phytohormones, the ubiquitin-proteasome pathway, and fatty acid synthesis. Overall, the candidate genes and co-expression networks detected in this study could be a valuable resource for genetic breeding to improve seed yield and quality traits in peanut.

Keywords: co-expression network; hub genes; oil content; peanut (Arachis hypogaea L.); seed development; seed size/weight; transcriptome analysis.

Figure 1

Phenotype differences between ZH16 and ZH6 at four stages of seed development. (A) Graphical display of seeds at different stages of development (S1–S4) between ZH16 and ZH6. (B) Average 100-seed weights (g) between ZH16 and ZH6. (C) Average oil content between ZH16 and ZH6. ***, p < 0.001; ****, p < 0.0001 (Student’s t-tests).

Figure 2

Global gene expression profiles in ZH16 and ZH6. (A) PCA plot showing clustering of gene transcript levels at four stages of seed development in ZH16 and ZH6. (B) Proportion of genes with different expression levels (based on FPKM). (C,D) Venn diagrams of expressed genes amongfour stages of seed development in cultivars ZH16 (C) and ZH6 (D).

Figure 3

Genes with stage-specific expression during seed development in peanut cultivars ZH16 and ZH6. (A) Bar graph showing numbers of stage-specific expressed genes specifically or commonly present in ZH16 and/or ZH6 at each stage of seed development. (B) Heatmap showing the expression of common stage-specific expressed genes at different stages in ZH16 and ZH6. Color scale represents Z-score. (C,D) Enriched functional GO terms (biological process) of common stage-specific expressed genes in two cultivars at the S1 (C) and S2 (D) stages.

Figure 4

DEGs between ZH16 and ZH6 at different seed developmental stages. (A) Number of up-regulated and down-regulated genes. (B) Distribution of Log₂FC values of DEGs. (C,D) Venn diagrams showing numbers of DEGs concurrently or specifically expressed among four stages. (E) Enriched GO terms (biological process) of up- and down-regulated genes. The color scale represents significance (corrected p-value).

Figure 5

WGCNA of DEGs between ZH16 and ZH6 at each seed developmental stage. (A) Module–sample relationships. The number of genes within key modules is indicated next to the module name. Color bars represent negative (blue) and positive (red) correlations. (B–D) Expression patterns of DEGs in magenta (B), yellow (C), and red (D) modules.

Figure 6

Heatmap and co-expression networks of DEGs within three key modules. (A,C,E) Heatmaps of hub genes. Genes overlapping with reported QTLs are marked in red. (B,D,F) Co-expression networks. Purple, blue, orange, and red nodes represent hub genes involved in TFs, phytohormones, the ubiquitin–proteasome pathway, and fatty acid synthesis, respectively. Node size represents connectivity.