Plastid retrograde regulation of miRNA expression in response to light stress

Abstract

Background: MicroRNAs (miRNAs) are a class of endogenous noncoding RNAs that play a pivotal role in the regulation of plant development and responses to the surrounding environment. Despite the efforts made to elucidate their function in the adaptation of plants to many abiotic and biotic stresses, their role in high light (HL) stress is still vague. HL stress often arises upon plant exposure to full sunlight. Subsequent changes in nuclear gene expression are triggered by chloroplast-derived retrograde signals.

Results: In this study, we show that HL is involved in miRNA-dependent regulation in Arabidopsis thaliana rosettes. Microtranscriptomic screening revealed a limited number of miRNAs reacting to HL. To explain the miRNA regulation mechanisms at the different biogenesis stages, chemical and genetic approaches were applied. First, we tested the possible role of plastoquinone (PQ) redox changes using photosynthetic electron transport chain inhibitors. The results suggest that increased primary transcript abundance (pri-miRNAs) of HL-regulated miRNAs is dependent on signals upstream of PQ. This indicates that such signals may originate from photosystem II, which is the main singlet oxygen (¹O₂) source. Nevertheless, no changes in pri-miRNA expression upon a dark-light shift in the conditional fluorescent (flu) mutant producing ¹O₂ were observed when compared to wild-type plants. Thus, we explored the ¹O₂ signaling pathway, which is initiated independently in HL and is related to β-carotene oxidation and production of volatile derivatives, such as β-cyclocitral (β-CC). Pri-miRNA induction by β-CC, which is a component of this ¹O₂ pathway, as well as an altered response in the methylene blue sensitivity 1 (mbs1) mutant support the role of ¹O₂ signaling in miRNA regulation.

Conclusions: We show that light stress triggers changes in miRNA expression. This stress response may be regulated by reactive oxygen species (ROS)-related signaling. In conclusion, our results link ROS action to miRNA biogenesis, suggesting its contribution to inconsistent pri- and mature miRNA dynamics.

Keywords: Chloroplast; High light; Plastoquinone; Singlet oxygen; miRNAs.

Fig. 1

HL induces miRNA expression changes in Arabidopsis thaliana shoots. a F_v/F_m measured in 4-week-old Arabidopsis plants grown in LL under hydroponic conditions. LLc – control plants; HL—plants exposed to HL for 2 h (n = 18). b TT-qRT PCR for selected miRNA-targeted transcripts of nuclear encoded proteins localized in chloroplasts under LLc and HL. Transcript levels were normalized with respect to small nucleolar85 RNA (sno85) and small nucleolar101 RNA (sno101). c Results of a microtranscriptomic screening for miRNA expression changes in LLc, HL and LLr – plants exposed to HL for 2 h and a subsequent recovery for 4 h in LL. d TT-qRT PCR for miR163 and miR840 and e qRT-PCR for pri-miR163 and pri-miR840 in Col-0 LLc, HL and LLr plants. Transcript levels were normalized with respect to sno85 and sno101 (for miRNAs) or the PP2A and UPL7 genes (for pri-miRNAs), respectively. f Occupancy of PolII on MIR genes. Line charts present ChIP profile of total PolII on examined genes. Grey lines represent results for LLc plants and blue lines represent results for HL plants. Above each chart, gene structure is shown with black boxes representing miR, and grey boxes representing primary transcripts (pri-miRs).Orange lines show amplified regions (primer localization used for qRT-PCR analysis). Above each gene structure 100 bp scale is shown. g RNA stability assay performed on Arabidopsis seedlings in control plants (LLc) and plant exposed to HL (HL). Degradation curves after cordycepin treatment (plot) were used to calculate half-life of pri-miR840, pri-miR163, UBC (control) transcripts and short-lived mRNA transcribed from gene At3G45970 (table h). Presented values are averages from three biological replicates. For better clarity of chart only pri-miRNAs data were presented. Asterisks indicate significant differences according to Tukey’s HSD test (panel: a,b,d,e) or t-test (panel f) at the level of ** ≤ 0.01 and *** ≤ 0.001. Mean values ± SDs (n = 3) were provided

Fig. 2

The role of PQ redox status in miRNA expression. a Scheme representing the action of DCMU and DBMIB in the photosynthetic electron transport chain. b Treatment with DCMU and DBMIB influences the maximum efficiency of PSII expressed as the Fv/Fm parameter. Plant material: dark—plants kept in darkness for 4 h; LLc—control plants in LL; LLtrt—plants treated with DCMU or DBMIB for 4 h and kept in LL. Asterisks indicate significant differences according to Tukey’s HSD test at the level of *** ≤ 0.001. Mean values ± SDs (n = 9) were provided. c qRT-PCR for pri-miR163 and pri-miR840 after using DCMU or DBMIB (upper panel). Transcript levels were normalized with respect to the PP2A and UPL7 genes. TT-qRT PCR for miR163 and miR840 after treatment with DCMU or DBMIB (bottom panel). Transcript levels were normalized with respect to sno85 and sno101. Asterisks indicate significant differences according to Tukey’s HSD test at the level of * ≤ 0.05, ** ≤ 0.01 and *** ≤ 0.001. Mean values ± SDs (n = 3) were provided

Fig. 3

Regulation of pri-miRNA expression is not dependent on EX1-dependent ¹O₂ signaling. a The F_v/F_m parameter measured in control (ctrl) plants (grown in constant light (CL)) and treated (trt) plants (plants grown for 2 weeks in CL, placed for 12 h in darkness, and re-exposed for 2 h to LL) of the Col-0, ex1, flu, and flu/ex1 genotypes (n = 9–10). b qRT-PCR of the DRP gene in the Col-0, ex1, flu, and flu/ex1 genotypes in ctrl and trt plants. c qRT-PCR for pri-miR163 and pri-miR840 in the Col-0, ex1, flu, and flu/ex1 genotypes in ctrl and trt plants. d qRT-PCR for pri-miR163 and pri-miR840 in Col-0 and ex1 plants LLc-control plants; HL-plants exposed to HL for 2 h; LLr-plants exposed to HL for 2 h and subsequent recovery in LL for 4 h. Transcript levels were normalized with respect to the PP2A and UPL7 genes. Asterisks indicate significant differences according to Tukey’s HSD test at the level of * ≤ 0.05, ** ≤ 0.01 and *** ≤ 0.001. In (b) and (c), * indicates significance within the same genotype, while ^ indicates comparison to Col-0 within the same conditions. Mean values ± SDs (n = 3) were provided

Fig. 4

Regulation of pri-miRNA by β-CC-dependent ¹O₂ signaling. a Scheme represents the β-CC-dependent ¹O₂ signaling pathway induced in HL conditions. β-CC is formed in PSII as a result of β-carotene oxidative breakdown under HL conditions. MBS1 is a downstream component that transduces stress information to the nucleus, where it influences the expression of singlet oxygen responsive genes (SORGs). b qRT-PCR for the DRP gene and c pri-miR163 and pri-miR840 in control plants (0.00) and after β-CC treatment (0.05 ml and 1.00 ml) for 4 h. d TT-qRT PCR for miR163 and miR840 level in plants treated by β-cc. ctrl- control plants, trt—plants treated with 1.0 ml of β-CC. Transcript levels were normalized with respect to sno85 and sno101. e qRT-PCR for pri-miR163 and pri-miR840 in Col-0 and mbs1 plants. LLc-control plants; HL-plants exposed to HL for 2 h; LLr-plants exposed to HL for 2 h and subsequent recovery in LL for 4 h. Transcript levels were normalized with respect to the PP2A and UPL7 genes. Asterisks indicate significant differences according to the Tukey HSD test at the level of * ≤ 0.05, ** ≤ 0.01, *** ≤ 0.001. Mean values ± SDs (n = 3), were provided