Long noncoding RNAs contribute to DNA damage resistance in Arabidopsis thaliana

Abstract

Efficient repair of DNA lesions is essential for the faithful transmission of genetic information between somatic cells and for genome integrity across generations. Plants have multiple, partially redundant, and overlapping DNA repair pathways, probably due to the less constricted germline and the inevitable exposure to light including higher energy wavelengths. Many proteins involved in DNA repair and their mode of actions are well described. In contrast, a role for DNA damage-associated RNA components, evident from many other organisms, is less well understood. Here, we have challenged young Arabidopsis thaliana plants with two different types of genotoxic stress and performed de novo assembly and transcriptome analysis. We identified three long noncoding RNAs (lncRNAs) that are lowly or not expressed under regular conditions but up-regulated or induced by DNA damage. We generated CRISPR/Cas deletion mutants and found that the absence of the lncRNAs impairs the recovery capacity of the plants from genotoxic stress. The genetic loci are highly conserved among world-wide distributed Arabidopsis accessions and within related species in the Brassicaceae group. Together, these results suggest that the lncRNAs have a conserved function in connection with DNA damage and provide a basis for mechanistic analysis of their role.

Keywords: Arabidopsis; Brassicaceae; DNA damage; DNA repair; double strand break; long noncoding RNA; plant.

Fig. 1.

Identification of lncRNAs upon DNA damage induction. a) Processing of zeocin- or UV-C-treated plant material for RNA sequencing. Libraries were prepared from 5 (zeocin) or 3 (UV-C) biological replicates. b) Processing of sequence data for transcriptome assembly and lncRNA identification. c) Transcript assembly in libraries of zeocin- or UV-C-treated samples. mRNAs and known lncRNAs were counted if present in the annotation of the reference genome Araport 11. Novel lncRNAs refer to newly identified transcribed regions previously not annotated. d) Classification of lncRNAs according to their position in or between PC genes.

Fig. 2.

Comparison of differentially expressed genes. a) Differentially expressed genes after treatment with zeocin (upper panel, orange) or UV-C (lower panel, blue). Numbers of down- or up-regulated genes above the threshold (orange or blue) are indicated. b) Venn diagrams for overlap between mRNAs (upper) or lncRNAs (lower) differentially expressed between treated and mock-treated zeocin- or UV-C samples. c) RT-qPCR validation of differential expression of lncRNAs induced by zeocin- and UV-C-treatment Error bars indicate the standard deviation of 3 biological replicates (Welcher test **P-value <0.01, *P-value <0.05).

Fig. 3.

Characterization of DNA-damage-induced lncRNAs. a) Location of the genes encoding lncRNA B, C, and D on chromosome 3 and 4 of Arabidopsis thaliana. b) Scheme of the genes encoding lncRNA B, C, and D. Large boxed arrows (black) represent lncRNA transcripts confirmed by RACE-PCR; stripes represent the annotation in Araport11. Small arrows (green) indicate the position of the two primers used for 5′ or 3′ RACE-PCR. Arrow heads triangles (red) indicate the 5′ and the 3′ ends identified by RACE-PCR. c) RT-qPCR with specific primers for lncRNAs B, C, or D on chromatin samples immunoprecipitated with a PolII antibody recognizing the carboxy-terminal domain (CTD), from mock-treated or zeocin-treated samples. d) Expression of lncRNAs B, C, or D in WT or atm mutant in mock- or zeocin-treated samples, normalized to a constitutively expressed actin gene. Data were normalized to the values in WT mock samples. Error bars indicate standard deviation of 3 biological replicates (Welcher test **P-value <0.01, *P-value <0.05).

Fig. 4.

Characterization of deletion mutants. a) Northern blots with total RNA probed with the radioactively labeled amplicons for lncRNAs B, C, or D, in WT or plants in which the lncRNA gene had been deleted by CRISPR-Cas9 mutations. M: mock-treated; Z: zeocin-treated; Methylene Blue: loading control. b) True-leaf assay for plant sensitivity against DNA damage. Left: seedlings resistant to zeocin can grow and develop true leaves; sensitive seedlings are arrested after cotyledons have unfolded. Right: Resistance ratio in WT, ku70 as a known sensitive repair mutant, or the deletion mutants lacking lncRNA B, C, or D (P-values according to Mann–Whitney-test). c) Comet assay for plant sensitivity against DNA damage.

Fig. 5.

Conservation of genes for damage-associated lncRNAs B, C, or D. a) Relative expression level of lncRNAs B, C, and D in Col-0 and 5 nonreference accessions upon exposure to zeocin. The relative expression is the ratio between treated samples and mock controls for each accession. Error bars represent the standard deviation across 3 replicates. b) Multiple alignments of lncRNAs B, C, and D loci and their flanking 300 bp regions identified in the full genomes of 27 A. thaliana accessions. The red boxes mark the regions corresponding to the lncRNA transcript. The narrow black box indicates the Col-0 reference accession. The multiple alignments are sorted (descending) by length. c) Distribution of the number of SNPs per 1 kb for Araport11-annotated PC genes (blue line) and noncoding genes (red line). Dashed vertical lines show the exact number of SNPs per 1 kb for lncRNA B (dark green), lncRNA C (light aquamarine), and lncRNA D (dark aquamarine). The number of SNPs is calculated according to the SNP calling from 1,135 natural A. thaliana accessions (https://1001genomes.org/accessions.html). d) Percent of A. thaliana natural accessions that express (TPM > 0.5) lncRNAs B, C, or D. The ratios were calculated from RNA-seq data from seedlings, rosettes at the 9-leaf stage from 25 accessions, flowers, and pollen from 23 accessions (Kornienko et al. 2023), and leaves from mature prebolting rosettes from 461 accessions (Kawakatsu et al. 2016). e) Expression variability across 461 accessions ((Kawakatsu et al. 2016) for lowly expressed lncRNA B (left) and moderately expressed lncRNA C and D (right), compared to that for lowly expressed Araport11 PC and nonprotein-coding (NC) genes. The precise level of the expression variability of lncRNAs B, C, and D is indicated with horizontal dashed lines. Data source as in Fig. 5d. f) Expression levels of lncRNA C in accessions of different geographic origin defined by admixture groups. The red dashed horizontal line indicates the expression cutoff (TPM = 0.5). Data source as in Fig. 5d. The admixture group of each accession was determined based on genetic similarity (http://1001genomes.github.io/admixture-map/).

Fig. 6.

Conservation of genes for damage-associated lncRNAs B, C, or D among the Brassicaceae. Maximum Likelihood phylogenetic trees of homologous sequences to a) lncRNA B (At4g07235), b) lncRNA C (At4g09215), and c) lncRNA D (At3g00800), obtained from Brassicacean species. The support values at the branches have been obtained from 10,000 UFboot samples. The trees are rooted in the most basal species in the dataset, Aethionema arabicum. The sequence names denote the species as well as the chromosome (chr), supercontig (sc), or linkage group (LG) of the respective genome. If more than one sequences originate from the same chromosome, letters a, b, c, etc., were appended to the sequence name. The boxes denote Brassicaceae Lineages I and II as well as the basal lineages.

Fig. 6.

Conservation of genes for damage-associated lncRNAs B, C, or D among the Brassicaceae. Maximum Likelihood phylogenetic trees of homologous sequences to a) lncRNA B (At4g07235), b) lncRNA C (At4g09215), and c) lncRNA D (At3g00800), obtained from Brassicacean species. The support values at the branches have been obtained from 10,000 UFboot samples. The trees are rooted in the most basal species in the dataset, Aethionema arabicum. The sequence names denote the species as well as the chromosome (chr), supercontig (sc), or linkage group (LG) of the respective genome. If more than one sequences originate from the same chromosome, letters a, b, c, etc., were appended to the sequence name. The boxes denote Brassicaceae Lineages I and II as well as the basal lineages.

Fig. 6.

Conservation of genes for damage-associated lncRNAs B, C, or D among the Brassicaceae. Maximum Likelihood phylogenetic trees of homologous sequences to a) lncRNA B (At4g07235), b) lncRNA C (At4g09215), and c) lncRNA D (At3g00800), obtained from Brassicacean species. The support values at the branches have been obtained from 10,000 UFboot samples. The trees are rooted in the most basal species in the dataset, Aethionema arabicum. The sequence names denote the species as well as the chromosome (chr), supercontig (sc), or linkage group (LG) of the respective genome. If more than one sequences originate from the same chromosome, letters a, b, c, etc., were appended to the sequence name. The boxes denote Brassicaceae Lineages I and II as well as the basal lineages.