Transcriptome Analysis Reveals Genes of Flooding-Tolerant and Flooding-Sensitive Rapeseeds Differentially Respond to Flooding at the Germination Stage

Abstract

Flooding results in significant crop yield losses due to exposure of plants to hypoxic stress. Various studies have reported the effect of flooding stress at seedling establishment or later stages. However, the molecular mechanism prevailing at the germination stage under flooding stress remains enigmatic. The present study highlights the comparative transcriptome analysis in two rapeseed lines, i.e., flooding-tolerant (Santana) and -sensitive (23651) lines under control and 6-h flooding treatments at the germination stage. A total of 1840 up-regulated and 1301 down-regulated genes were shared by both lines in response to flooding. There were 4410 differentially expressed genes (DEGs) with increased expression and 4271 DEGs with reduced expression shared in both control and flooding conditions. Gene ontology (GO) enrichment analysis revealed that "transcription regulation", "structural constituent of cell wall", "reactive oxygen species metabolic", "peroxidase", oxidoreductase", and "antioxidant activity" were the common processes in rapeseed flooding response. In addition, the processes such as "hormone-mediated signaling pathway", "response to organic substance response", "motor activity", and "microtubule-based process" are likely to confer rapeseed flooding resistance. Mclust analysis clustered DEGs into nine modules; genes in each module shared similar expression patterns and many of these genes overlapped with the top 20 DEGs in some groups. This work provides a comprehensive insight into gene responses and the regulatory network in rapeseed flooding stress and provides guidelines for probing the underlying molecular mechanisms in flooding resistance.

Keywords: DEG; flooding stress; germination; rapeseed; transcriptome analysis.

Figure 1

Phenotypic response of tolerant (Santana) and sensitive (23651) rapeseed lines to flooding stress. (A) Phenotypes of tolerant and sensitive lines under control treatment and after suffering flooding. Flooding-treated seeds were transplanted into moist vermiculite and cultured for 5 d at 24 °C. Then the growth of seedings was recorded and the pictures were taken. Shoot (B) and root length (C) of tolerant and sensitive lines under control treatment (CK) and after suffering flooding (F) (n = 4 for CK, n = 10 for F; medians: 3.10, 2.95, 3.45, and 1.40 for shoot length and 6.25, 84.75, 8.60, and 1.20 for root length). Flooding treatment was repeated three times with consistent results. Letters show significant differences checked by two-way ANOVA test.

Figure 2

Volcano plots showing the differentially expressed genes (DEGs) between different materials and treatments. (A,B) DEGs between materials responding to flooding; (C,D) DEGs between treatments of different materials. CK, control; T, tolerant; S, sensitive.

Figure 3

Venn diagrams showing the number of differentially expressed genes (DEGs) between materials and treatments. Down-regulated (A) and up-regulated (C) genes between materials; down-regulated (B) and up-regulated (D) genes between treatments. CK, control; T, tolerant; S, sensitive.

Figure 4

Validation of RNA-seq data by qRT-PCR. (A–D) The change of expression of 4 representative DEGs between control and flooding were compared based on RNA-seq and qRT-PCR data. Three biological replicates were used for the comparison. (E) The expression correlation between the RNA-seq and qRT-PCR. Actin7 was served as internal reference gene.

Figure 5

Gene ontology (GO) analyses of up- and down-regulated genes responding to flooding. (A) GO enrichment analysis of up-regulated DEGs. (B) GO enrichment analysis of down-regulated DEGs. In (A) and (B), GO enrichment was based on cellular component (C), molecular function (F) and biological process (P). The horizontal axis value shows enrichment factor (q < 0.05). All GO terms are displayed as –log₁₀ (Fisher’s exact test q value).

Figure 6

Mclust analysis of DEGs. Heatmap showing the expression pattern of DEGs specific to each group. Nine clusters were generated and named according to the highly significant GO terms found within them (based on q-value). Color bars represent enrichment (q values) of GO terms of respective DEGs. CK, control.