Arabidopsis latent virus 1, a comovirus widely spread in Arabidopsis thaliana collections

Summary Transcriptome studies of Illumina RNA‐Seq datasets of different Arabidopsis thaliana natural accessions and T‐DNA mutants revealed the presence of two virus‐like RNA sequences which showed the typical two‐segmented genome characteristics of a comovirus. This comovirus did not induce any visible symptoms in infected A. thaliana plants cultivated under standard laboratory conditions. Hence it was named Arabidopsis latent virus 1 (ArLV1). Virus infectivity in A. thaliana plants was confirmed by quantitative reverse transcription polymerase chain reaction, transmission electron microscopy and mechanical inoculation. Arabidopsis latent virus 1 can also mechanically infect Nicotiana benthamiana, causing distinct mosaic symptoms. A bioinformatics investigation of A. thaliana RNA‐Seq repositories, including nearly 6500 Sequence Read Archives (SRAs) in the NCBI SRA database, revealed the presence of ArLV1 in 25% of all archived natural A. thaliana accessions and in 8.5% of all analyzed SRAs. Arabidopsis latent virus 1 could also be detected in A. thaliana plants collected from the wild. Arabidopsis latent virus 1 is highly seed‐transmissible with up to 40% incidence on the progeny derived from infected A. thaliana plants. This has probably led to a worldwide distribution in the model plant A. thaliana with as yet unknown effects on plant performance in a substantial number of studies.


+31317480675
The following Supporting Information is available for this article:  Samples from newly-formed leaves after inoculation of healthy Col-0 plants with ArLV1_A (Wageningen isolate), ArLV1_B (Utrecht isolate) or mock inoculum (Mock). Sample numbers represent the individually sampled plants from two separate experiments (a and b) for which plant material was tested by qPCR for RNA1 and RNA2 of ArLV1 and dCT was calculated by subtraction from the average mock sample. dCT-values of RNA1 and RNA2 cluster into two main groups (positive and negative; k-means clustering). In total, 79 of the 90 inoculated plants tested positive for ArLV1, which represents an inoculation efficiency of 88%.       Study design, sampling, data cleaning and read mapping were performed as described earlier (Kloth et al., 2016).

Dataset Utrecht
Growth conditions and treatments are largely as described in Morales et al., (2022). A. thaliana Col-0 seeds were sown on a moist soil:perlite mix 1:2 (Primasta BV, Asten, The Netherlands) and and further cultivated under identical short-day conditions. Hoagland solution was applied at two, five and seven days after transplanting (10 mL, 20 mL and 10 mL per plant respectively). Plants were watered every two days unless subjected to drought or submergence treatment. Plants were randomized every 2-3 days up to 10-leaf stage. At 10-leaf stage, plants were subjected to different stress conditions: mild drought, high temperature, high temperature combined with mild drought, 5-day submergence in light, post submergence followed by recovery, and post submergence combined with mild drought.
Leaf material was harvested at 0, 5 and 10 days after the treatments started and all samples were collected at 2 hours after the photoperiod began. For each sample, five or six plants were selected and two young leaves (leaves 7, 8, 9 and 10 are defined as young leaves, counting from the oldest leaf at 10-leaf stage) from each plant were detached and pooled together as a biological replicate.
Leaf pools were immediately snap-frozen with liquid nitrogen. Frozen leaves were ground to a fine powder and processed to total RNA extraction using the RNeasy kit (Qiagen, Germany) following the manufacturers protocol. Samples were finally diluted to 25 ug/ul with a total volume of 60 uL using DEPC-treated water.
Quality control, library construction and Illumina sequencing were carried out by MACROGEN (Amsterdam, The Netherlands). Total RNA integrity and pureness were checked using an Agilent Technologies 2100 Bioanalyzer (Agilent, Santa Clara). Library preparations were performed based on the TruSeq stranded mRNA protocol (Illumina, San Diego). The constructed libraries were then sequenced by Illumina Novaseq6000 sequencer with 150 bp pair-end reading and finally provided with FASTQ files output.
Data cleaning, including adaptor trimming and reads filtering was conducted using CUTADAPT (Martin, 2011) under Linux operation system. The Truseq adaptor sequences "AGATCGGAAGAGCACACGTCTGAACTCCAGTCA" and "AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT" were trimmed with a maximum error rate of 0.07. Reads filtering was based on the criterion of the minimum cutoff length of 30 bp and quality score of 20. This brought at least 61 million clean pair-end reads per sample. The trimmed and filtered reads were then assembled and mapped onto A. thaliana transcriptome (Araport10) using KALLISTO (Bray et al., 2016).
For details of plant growth conditions and treatments used for the Utrecht transcriptome dataset, see Supplemental Information Methods S1. In short, 58 samples of young leaves from A. thaliana Col-0 at different timepoints and under different abiotic stress treatments (Morales et al., 2022) were harvested. Total RNA was extracted, a library was made and sequenced using the Truseq stranded mRNA protocol and an Illumina Novaseq6000. Before analysis, the transcripts per million Statistical analysis was done in R version 4.1.2 x64 using the log2-normalized TPM values in linear models ran for timepoint 5 for the control samples containing a high ArLV1 content when compared to the control samples with a low ArLV1 content at the same timepoint. The obtained significances were corrected using a Benjamini-Hochberg adjustment for multiple testing