Integrate Small RNA and Degradome Sequencing to Reveal Drought Memory Response in Wheat (Triticum aestivum L.)

Abstract

Drought has gradually become one of the most severe abiotic stresses on plants. Plants that experience stress training can exhibit enhanced stress tolerance. According to MicroRNA (miRNA) sequencing data, this study identified 195 candidate drought memory-related miRNAs in wheat, and targets of 64 (32.8%) candidate miRNAs were validated by degradome sequencing. Several drought memory-related miRNAs such as tae-miR9676-5p, tae-MIR9676-p3_1ss21GA, tae-miR171a, tae-miR531_L-2, tae-miR408_L-1, PC-3p-5049_3565, tae-miR396c-5p, tae-miR9778, tae-miR164a-5p, and tae-miR9662a-3p were validated as having a strong response to drought memory by regulating the expression of their target genes. In addition, overexpression of drought memory-related miRNA, tae-miR531_L-2, can remarkably improve the drought tolerance of transgenic Arabidopsisthaliana. Drought memory can regulate plant cellular signal transduction, plant biosynthetic processes, and other biological processes to cope with drought via transcriptional memory. In addition, drought memory-related miRNAs can promote starch and sucrose catabolism and soluble sugar accumulation and regulate proline homeostasis to improve plant drought resistance. Our results could contribute to an understanding of drought memory in wheat seedlings and may provide a new strategy for drought-resistant breeding.

Keywords: drought training; high-throughput sequencing; miR531; proline; transcriptional memory.

Figure 1

Experimental design and physiological index measurement. (A) Drought resistance detection of seedlings in three groups. Trileaf stage wheat seedlings in DD and DM groups were treated with half-strength Hoagland’s liquid medium containing 19.2% PEG for six days, and seedlings grown in normal conditions were used as control. (B) Trileaf stage seedlings were divided into three groups, including the control (CG), direct drought (DD), and drought memory (DM). The DM group was pre-treated with 10% (M/V) PEG. Seedlings of the DD and DM groups were then treated with 19.2% PEG, while CG was grown under normal conditions. Whole seedlings of three groups were collected at 0, 1, 6, and 12 h with three biological repetitions and immediately frozen in liquid nitrogen. (C) Proline content comparison amongst the three groups after treatment (19.2% (M/V) PEG) for 0, 24, 48, and 72 h. * p < 0.05; ** p < 0.01. (D) Comparison of the efficiency of photosystem II (Phi2) amongst the three groups after treatment (19.2% (M/V) PEG) for 0, 24, 48, and 72 h. Error bars represent the SD. Points on the polyline followed by different letters are statistically different according to the analysis of variance followed by Duncan’s Multiple Range Test (Comparison of different treatments at the same time).

Figure 2

Identification of drought memory-related miRNAs. (A–C) Venn diagrams of significantly different expressed miRNAs amongst CG, DD, and DM groups (D) Identification of drought memory-related miRNAs according to miRNA sequence data. (DM vs. CG) vs. (DD vs. CG) revealed 186 drought memory-specific miRNAs and 57 shared miRNAs. Fifty-seven common miRNAs were searched in the DM vs. DD miRNA set, and nine miRNAs were found to be significantly differentially expressed between DM and DD groups. Finally, we obtained 195 candidate drought memory-related miRNAs (inside the dotted red line).

Figure 3

Validation of 10 drought memory-related miRNAs and their target genes by RT-qPCR. (A–J) The relative expression level of ten candidate drought memory-related miRNAs and their target genes. The relative expression level of each miRNA and their target genes was compared between DD and DM groups at 0, 1, 6, and 12 h, with the CG as control. The expression levels of the miRNAs and target genes were normalized using U6 (GenBank: X63066.1) and Elongation Factor 1-Alpha genes, respectively. All reactions were performed in triplicate for each sample, and expression levels were calculated according to the 2^−ΔΔCT method. Error bars represent the SD, * p < 0.05; ** p < 0.01.

Figure 4

Overexpression of the tae-miR531_L-2 precursor in Arabidopsis thaliana. (A,B) The expression level of tae-miR531_L-2 in T₃ transgenic line (tae-miR531_L-2) and wild type (WT) seedlings under drought (20% PEG, 0 h, 1 h, 6 h and 12 h), cold (4 °C, 6 h), heat (37 °C, 6 h), and salt (300 mM NaCl, 6 h) stresses. Error bars represent the SD, * p < 0.05; ** p < 0.01. (C) Germination rate experiment of transgenic lines and WT. Seeds of WT and transgenic lines were germinated on half-strength Murashige and Skoog (MS) agar plates containing 0 mM and 350 mM D-mannitol for two weeks. (D) Drought resistance detection in the seedling stage. Two-week-old soil-grown seedlings were air-dried and then rehydrated to detect extreme drought resistance. The seven-day-old soil-grown seedlings were watered with 10% PEG solution for two weeks for drought resistance detection under prolonged mild drought.

Figure 5

The miRNA-gene-GO association analysis of drought memory-related miRNAs. Candidate drought memory-related miRNAs, their target transcripts, and annotated GO terms were used to investigate the relationship between them. The miRNAs are denoted in green, target transcripts are denoted in cyan, and the GO terms are red. The annotated functions of target transcripts are labeled in red.

Figure 6

A proposed regulatory mechanism involving the differentially expressed miRNAs and their target genes in wheat drought memory. Targets of eight miRNAs were enriched to fifteen pathways that might be crucial to drought memory.