Identification of functional long non-coding RNAs in C. elegans

Abstract

Background: Functional characterisation of the compact genome of the model organism Caenorhabditis elegans remains incomplete despite its sequencing 20 years ago. The last decade of research has seen a tremendous increase in the number of non-coding RNAs identified in various organisms. While we have mechanistic understandings of small non-coding RNA pathways, long non-coding RNAs represent a diverse class of active transcripts whose function remains less well characterised.

Results: By analysing hundreds of published transcriptome datasets, we annotated 3392 potential lncRNAs including 143 multi-exonic loci that showed increased nucleotide conservation and GC content relative to other non-coding regions. Using CRISPR/Cas9 genome editing, we generated deletion mutants for ten long non-coding RNA loci. Using automated microscopy for in-depth phenotyping, we show that six of the long non-coding RNA loci are required for normal development and fertility. Using RNA interference-mediated gene knock-down, we provide evidence that for two of the long non-coding RNA loci, the observed phenotypes are dependent on the corresponding RNA transcripts.

Conclusions: Our results highlight that a large section of the non-coding regions of the C. elegans genome remains unexplored. Based on our in vivo analysis of a selection of high-confidence lncRNA loci, we expect that a significant proportion of these high-confidence regions is likely to have a biological function at either the genomic or the transcript level.

Keywords: C. elegans; CRISPR; Long non-coding RNA; Non-coding; lincRNA; lncRNA.

Fig. 1

LncRNAs sequence features in C. elegans. a Nucleotide compositions of exons and introns of lncRNAs and protein-coding genes classified according to their gene model. b GC content variation across metagenes. The x-axis represents non-overlapping windows each including 10% of the sequences across multi-exonic protein-coding and lncRNA loci; the solid lines represent respectively the 95% confidence interval, median and 5% confidence interval. The black band represents the GC content of flanking intergenic sequences. c Nucleotide conservation (PhyloP score) comparison among intergenic sequences, multi- or mono-exonic lncRNAs and protein-coding loci. d Enrichment of lncRNAs for chromatin annotations identified by Daugherty et al. [28]. Transcribed gene body: ensemble of ChromHMM states characterised by H3K79me2, H3K36me3, H3K4me1 and H4K20me1. Repressed enhancers: ensemble of ChromHMM states characterised by H3K4me1 and H3K27me3. Low signal: regions without histone modification signals

Fig. 2

a LncRNA expression properties. a Cumulative distribution of the proportion of multi-exonic, mono-exonic lncRNA and protein-coding loci identified as expressed across all libraries (FPKM > 1). b Expression (log₂ FPKM) across C. elegans development of 143 multi-exonic lncRNAs. Each column represents the average expression at one time point for whole individuals in standard conditions. GC composition (c) and nucleotide conservation (d) for lncRNA loci depending on the reproducibility of lncRNA model predictions across libraries. ≤ 20 of the libraries (low), 20 to 100 (others) and ≥ 100 libraries (high)

Fig. 3

Gene structure of the ten lncRNA deletions. lncRNAs (dark grey), neighbouring protein-coding genes (black) and deleted regions (light grey and named with the respective allele name, mj) are shown. Numbers are to an arbitrary start point and do not reflect chromosomal location. Chromosomes are indicated at the right hand side

Fig. 4

Phenotyping of ten lncRNA deletion mutants. a Viable brood sizes are presented with their standard deviations (blue area) and the 95% confidence interval of the mean (red area). Samples were compared to wild type animals using a pairwise two-sample t test with a multiple test (Bonferroni) correction. Samples are ordered by increasing p value and those found to be significant at (p ≤ 0.05) are shown to the left of the blue line (n = 15 animals/mutant). b Growth curves were compared to wild type animals, and those found not to be significantly different are shown by their mean across strains (black line) with the standard error of the mean (grey area). Those found to be significantly different from the control are shown individually as means only. Inset shows example images of the wild type (top) and linc-239 mutant (bottom) at 45 h post hatching with the computer-generated outlines, and computed area (black line = 500 μm)

Fig. 5

RNAi-mediated knock-down of lncRNAs. a Viable brood sizes are presented with their standard deviations (blue area) and the 95% confidence interval of the mean (red area). Samples were compared to “empty vector” control animals using a pairwise two-sample t test with a multiple test (Bonferroni) correction. Samples are ordered by increasing p value, and those found to be significant at (p ≤ 0.05) are shown to the left of the blue line (n = 18 animals/mutant). Empty vector (EV), GFP and linc-340 RNAi are negative controls. b Growth curves were compared to “empty vector” animals, and those found to be not significantly different are shown by their mean across strains (black line) with the standard error of the mean (grey area). Those found to be significantly different from the control are shown individually as means only. Inset shows example images of the “empty vector” (top) and linc-239 RNAi (bottom) at 45 h post hatching with the computer-generated outlines, and computed area (black line = 500 μm). c RT-PCR analysis of RNAi knock-down efficiency for linc-239 and linc-339. Actin is used as loading control

Fig. 6

Comparison of phenotypes arising from lncRNA genomic deletion mutants and phenotypes arising from RNAi-mediated knock-down of lncRNA transcript. To compare the effects of the disruption of a lncRNA genomic locus to the knock-down of the corresponding lncRNA transcript by RNAi, the mean brood size reduction compared to the control (a) and the ratio of the length at 50 h relative to the control (b) were plotted. These are shown as a scatter plot of the mean reduction (a, blue circle) or the mean ratio (b, blue circle) with the 95% confidence interval of the mean (orange lines). If the mutations or the RNAi yield an effect, data fall below the line y = 1 and to the left of x = 1. If mutants and RNAi yield similar effects, data fall along the red line; above the red line indicates that RNAi has a greater effect, while below the red line indicates that the genomic mutation has a greater effect. c Expression (log₂ FPKM) across C. elegans development for linc-239 and linc-339. d RT-PCR analysis of neighbouring genes in linc-239 (mj441) deletion mutant. Diagram shows the relative position of the lncRNA and the surrounding protein-coding genes. Col-73 expression (left panel) and F11G11.1 expression (right panel) are measured using end-point RT-PCR at three different cDNA concentrations. Actin is used as loading control. e qRT-PCR analysis of col-73 and F11G11.13 expression in linc-239 (mj441) mutants. Error bars = 95% CI, significance is tested using t test with multiple t test correction. f RT-PCR analysis of the T01C8.3 gene in linc-339 (mj601) deletion mutants (left panel) and in linc-339 RNAi knock-down (right panel) at three different cDNA concentrations. Actin is used as loading control. g qRT-PCR analysis of T01C8.3 expression in linc-339 (mj339) and linc-339 RNAi. Error bars = 95% CI, significance is tested using t test with multiple t test correction