Harding JL et al. (2014), Small RNA profiling of Xenopus embryos reveals ...

XB-ART-47876

Genome Res 2014 Jan 01;241:96-106. doi: 10.1101/gr.144469.112.

Show Gene links Show Anatomy links

Small RNA profiling of Xenopus embryos reveals novel miRNAs and a new class of small RNAs derived from intronic transposable elements.

Harding JL , Horswell S , Heliot C , Armisen J , Zimmerman LB , Luscombe NM , Miska EA , Hill CS .

???displayArticle.abstract???
Small RNA control of gene expression is critical for developmental processes in vertebrate embryos. To determine the dynamics of small RNA expression and to uncover novel small RNAs in the early vertebrate embryo, we performed high-throughput sequencing of all small RNAs in Xenopus tropicalis embryos at three developmental time points and in dissected halves of gastrula embryos. This analysis allowed us to identify novel microRNAs and we show that microRNA expression is highly dynamic and spatially localized in early embryos. In addition, we have developed a microRNA prediction pipeline and demonstrate that it has the power to predict new miRNAs that are experimentally detectable in frogs, mice, and humans. By combining the small RNA sequencing with mRNA profiling at the different developmental stages, we identify a new class of small noncoding RNAs that we name siteRNAs, which align in clusters to introns of protein-coding genes. We show that siteRNAs are derived from remnants of transposable elements present in the introns. We find that genes containing clusters of siteRNAs are transcriptionally repressed as compared with all genes. Furthermore, we show that this is true for individual genes containing siteRNA clusters, and that these genes are enriched in specific repressive histone modifications. Our data thus suggest a new mechanism of siteRNA-mediated gene silencing in vertebrates, and provide an example of how mobile elements can affect gene regulation.

???displayArticle.pubmedLink??? 24065776
???displayArticle.pmcLink??? PMC3875865
???displayArticle.link??? Genome Res
???displayArticle.grants??? [+]

Species referenced: Xenopus tropicalis Xenopus laevis

???displayArticle.gses??? GSE38605: Xenbase, NCBI

???attribute.lit??? ???displayArticles.show???

	Figure 1. Small RNA library preparation and quality control. (A) Schematic of library preparation. Small RNAs of 18–30 nt were purified from X. tropicalis embryos, and Illumina RNA adapters were ligated by an RNA ligation reaction. The adapter sequence is needed to attach the template DNA to the flow cell in the sequencing machine. The ligated RNAs were reverse-transcribed and the cDNA used as the template for a large-scale PCR. The PCR library was then subjected to single-end deep sequencing. (B) Length distributions of small RNA sequencing reads and unique sequence tags. Small RNA read and tag numbers are plotted against length for the five small RNA libraries. The sequencing reads are trimmed to remove sequencing adapters and mapped to the genome with zero mismatches. The arrow indicates the miRNA population that collapses when reads are converted to tags, and the arrowhead indicates the population of 28-nt RNAs.
	Figure 2. Heatmap analysis of miRNA expression in early Xenopus development. miRNA tags of 22–23 nt identical in sequence to a miRbase miRNA were clustered by expression similarity. miRNA expression is displayed using a color key where blue corresponds to low and red to high numbers of miRNA normalized reads. The miRNAs labeled are those mentioned in the text and/or the most abundant in the five libraries (Supplemental Table S3).
	Figure 3. miRNA expression is highly dynamic and localized. (A) Normalized read numbers for individual miRNAs at stages 10 and 18 are shown relative to the stage 8 values. The red line indicates the mean. (B) Normalized miRNA reads in the stage 10 animal and stage 10 vegetal libraries are plotted on a logarithmic scale with y = x superimposed to show the animal–vegetal spatial distribution of miRNAs at gastrulation.
	Figure 4. Identification of novel miRNAs. (A) Predicted pre-miRNA structures of X. tropicalis miR-A, C, and F. The black line indicates the mature miRNA sequence within the hairpin precursor. (B) qPCR detection of candidate novel miRNAs in X. tropicalis. Relative RNA levels to odc were calculated and error bars are standard deviations from triplicate repeats. (C) Northern blot for miR-F on RNA from X. tropicalis embryos of stages 8–40 (left panel) and X. laevis embryos of stages 10–20 (right panel). Both the mature miRNA (23 nt) and pre-miRNA (70 nt) are detected. Comparison to decade markers (Ambion) revealed the size of the small RNAs. (D) qPCR detection of miR-F in human and mouse cell lines. miR-F RNA levels relative to gapdh were calculated. Error bars are standard deviations of triplicate repeats.
	Figure 5. siteRNAs are conserved, cluster in introns, and align to TE remnants. (A) The percentage of unannotated small RNA tags that map to intergenic regions, exons, and introns (using RefSeq gene annotation) is shown for the stage 8, 10, 18, stage 10 animal, and stage 10 vegetal libraries. (B) Venn diagram showing the number and overlap of genes with small intronic noncoding RNAs in the stage 8, stage 10, and stage 18 libraries. (C) Small RNA alignments and mRNA-seq alignments (tracks I–III) to the sptlc1 gene at stage 10 shown in the Integrated Genome Viewer (IGV). Exons are indicated by dark blue boxes, introns by blue lines, and an arrow shows the direction of transcription. Small RNA tags are colored red for a sense alignment and blue for an antisense alignment to the genome, respectively. The gray histograms show peaks of read density with the scale indicated (e.g., 0–137 reads, track I). Track I shows small RNA-seq data before filtering by window-scanning analysis to reveal regions of high siteRNA density (track II). Track III shows mRNA reads, which align to exons. Reads that map over an exon–exon boundary are connected by a blue line. (D,E) Diagrams are shown for map7d2 and tdp1 to indicate the relationship between the siteRNA clusters and the TE remnants present in the introns. The representation of the genes is as in C. Small RNA reads are shown in the track labeled “small RNA-seq” in gray, or in green for those which passed the window scanning filter. The scale indicated is as in C. The green bars below indicate individual siteRNA clusters and the red arrow indicates the direction of transcription of the small RNA reads. In the track labeled “transposon,” the TE remnants are shown as colored boxes, with the names given below. The TE remnant from which the siteRNAs are derived is shown in red, and this entire TE is shown to scale above. The red box in the TE is the region of almost identical sequence to the intronic TE remnant. The purple line shows the protein-coding region and the arrow gives the direction of transcription of the entire transposon. Note that the small RNA reads are antisense relative to the direction of TE transcription.
	Figure 6. siteRNA clusters are associated with low expression of the gene to which they map. (A) mRNA expression levels obtained by RPKM normalization of mRNA-seq reads for all RefSeq genes (“transcriptome”) and genes containing siteRNA clusters (Supplemental File 2) are compared for stages 8–18. The box and whisker plot displays the median, first, and third quartiles and the whiskers indicate 1.5× the interquartile range. Pairwise Wilcoxon tests were performed to assess changes in the RPKM values between “transcriptome” and siteRNA cluster genes at each stage, and Bonferroni-corrected P-values are indicated. (B) mRNA expression levels for all RefSeq genes (“transcriptome”) and genes containing upstream RNA clusters are compared for stages 8–18 as in A. (C) siteRNA clusters can be amplified using specific qRT-PCR primers at stages 8, 10, and 18 in X. tropicalis embryos. The bars labeled “common” correspond to qRT-PCR using primers that would recognize RNA from all of the group 7 TE remnants. RNA levels relative to odc were calculated and error bars are standard deviations from duplicates. The groups refer to the TE sequence clusters shown in Supplemental Figure S3 and Table 1. (D) mRNA expression levels detected by qPCR at stages 8–18. mRNA levels relative to odc were calculated. Error bars are standard deviations from triplicates. mRNA levels of the developmentally regulated gene xbra are also plotted as a reference.
	Figure 7. siteRNA clusters are enriched for repressive H3K9me3 and H4K20me3 epigenetic marks. ChIP-qPCR profiles for tri-methylated H3K4 (H3K4me3), H3K9me3, and H4K20me3 histone modifications at the TSS and siteRNA cluster region. Chromatin was prepared from whole X. tropicalis embryos at stages 10 and 18. The quantity relative to the input after qPCR was calculated for each IP. The bars marked “beads” correspond to a control IP without antibody. Means and standard deviations of duplicates from a representative experiment are shown.

References [+] :

Akkers, A hierarchy of H3K4me3 and H3K27me3 acquisition in spatial gene regulation in Xenopus embryos. 2009, Pubmed, Xenbase