Effects of Abscisic Acid and Salicylic Acid on Gene Expression in the Antiviral RNA Silencing Pathway in Arabidopsis

Abstract

The RNA silencing pathways modulate responses to certain stresses, and can be partially tuned by several hormones such as salicylic acid (SA) and abscisic acid (ABA). Although SA and ABA are often antagonistic and often modulate different stress responses, they have similar effects on virus resistance, which are partially achieved through the antiviral RNA silencing pathway. Whether they play similar roles in regulating the RNA silencing pathway is unclear. By employing coexpression and promoter analyses, we found that some ABA- and SA-related transcription factors (TFs) are coexpressed with several AGO, DCL, and RDR genes, and have multiple binding sites for the identified TFs in the queried promoters. ABA and SA are antagonistic with respect to the expression of AGO1 and RDRs because ABA was able to induce these genes only in the SA mutant. Nevertheless, both hormones showed similarities in the regulation of other genes, for example, the induction of AGO2 by ABA was SA-dependent, indicating that ABA acts upstream of SA in this regulation. We inferred that the similar effects of ABA and SA on some genes resulted in the redundancy of their roles in resistance to bamboo mosaic virus, but that the two hormones are antagonistic with respect to other genes unrelated to their biosynthesis pathways.

Keywords: RNA silencing pathway; abscisic acid; bamboo mosaic virus; crosstalk; salicylic acid.

Figure 1

Transcription factor (TF) binding sites of ABA- and SA-related TFs in the promoters of RNA silencing genes: Binding sites of ABA- and SA-regulated TFs on the promoters of RNA silencing genes (AGO, DCL, and RDR gene families). Promoter regions 2000 bp upstream of the transcription start site (TSS) were subjected to promoter analyses using the Plant Promoter Analysis Navigator in PlantPan 2.0. The TFs and their binding sites (cis-regulatory elements) were visualized using PlantPan 2.0.

Figure 2

Effects of ABA and SA treatment on the relative expression (as determined by RT-qPCR) of (A) AGO1, (B) AGO2, (C) AGO3, (D) AGO4, (E) AGO6, (F) AGO7, (G) AGO10, (H) RDR1, (I) RDR2, (J) RDR6, (K) DCL1, (L) DCL2, (M) DCL3, and (N) DCL4 in seedlings of the Arabidopsis thaliana wild type (WT), the SA mutant sid2-1, the ABA mutant aao3, and their double mutant (D.M) sid2-1:aao3. Six-day-old seedlings were treated with SA (50 µM), ABA (10 µM), or mock (0.1% EtOH); expression was determined by RT-qPCR 2 days later. Values are means + SD of three biological replicates, each carried out with three technical replicates. Analyses of variance (ANOVAs) were conducted to determine significant differences (p < 0.05); when ANOVAs were significant, the means for each gene were compared with a Duncan’s post hoc test; means that share a lowercase letter are not statistically different, while different letters denote for statistical differences at least 95% confidence.

Figure 3

The susceptibility of wild type (WT) and ABA and SA mutant seedlings to BaMV. Six-day-old seedlings of the WT, sid2-1, aao3, and sid2-1:aao3 were infected with a BaMV infectious clone (pKB) using Agrobacterium-mediated transformation (AGROBEST method). Total proteins were extracted 4 days later, and BaMV coat protein was quantified by Western blot analysis as indicated in the figure. The experiment was carried out in 3 biological replicates with similar results (Figure S2). Band densities (anti-CP and RuBisCO from Ponceau-S staining) were measured with Image J software, and the CP level in each line was normalized to the corresponding Rubisco band. Values are means ± SD of three biological replicates. An asterisk indicates a significant increase relative to the WT (which was set at 1.0) according to a one-sided Student’s t-test (p < 0.05).