The conservation landscape of the human ribosomal RNA gene repeats

Abstract

Ribosomal RNA gene repeats (rDNA) encode ribosomal RNA, a major component of ribosomes. Ribosome biogenesis is central to cellular metabolic regulation, and several diseases are associated with rDNA dysfunction, notably cancer, However, its highly repetitive nature has severely limited characterization of the elements responsible for rDNA function. Here we make use of phylogenetic footprinting to provide a comprehensive list of novel, potentially functional elements in the human rDNA. Complete rDNA sequences for six non-human primate species were constructed using de novo whole genome assemblies. These new sequences were used to determine the conservation profile of the human rDNA, revealing 49 conserved regions in the rDNA intergenic spacer (IGS). To provide insights into the potential roles of these conserved regions, the conservation profile was integrated with functional genomics datasets. We find two major zones that contain conserved elements characterised by enrichment of transcription-associated chromatin factors, and transcription. Conservation of some IGS transcripts in the apes underpins the potential functional significance of these transcripts and the elements controlling their expression. Our results characterize the conservation landscape of the human IGS and suggest that noncoding transcription and chromatin elements are conserved and important features of this unique genomic region.

Fig 1. Eukaryotic ribosomal DNA organization.

A) Head-to-tail tandem arrangement of rDNA repeat units. Typically, there are more units in an array than depicted. B) Each rDNA unit has an rRNA coding region (black) and an intergenic spacer (IGS; green). The coding region encodes the ~18S, 5.8S and ~28S rRNAs (black boxes) separated by two internal transcribed spacers (ITS-1 and 2) and flanked by two external transcribed spacers (5’- and 3’-ETS).

Fig 2. Primate rDNA repeat units.

A) Phylogenetic tree showing the relationships between primate species selected for rDNA phylogenetic footprinting [adapted from 91]. B) Human and primate rDNA unit structures are shown. The rRNA coding region (black line), including the 18S, 5.8S and 28S rRNA subunits (black boxes), and the IGS (grey line) are indicated along with the positions of repeat elements and a cdc27 pseudogene. Elements above the line are on the forward strand; those below on the reverse strand. The rRNA coding region/IGS coordinates and rDNA unit lengths are indicated.

Fig 3. Sequence similarity plot of the primate rDNA.

The horizontal axis represents the position in the human rDNA; the vertical axis the level of sequence similarity between 0 (no identity) and 1 (all bases the same). A 50 bp sliding window with 1 bp increment was used to generate the similarity plot. Conserved regions in the IGS (purple boxes) were identified using phastCons. The positions of Alu elements (green boxes), microsatellites (grey boxes), a cdc27 pseudogene (pink box), the rRNA promoter (blue lines), previously identified IGS noncoding transcripts (green wiggly lines), c-Myc binding sites (orange lines), p53 binding site (green line), and Sal boxes (terminator elements; red lines) are indicated. Conserved regions with a black circle or triangle below are conserved in common marmoset and mouse rDNA, respectively.

Fig 4. Two zones in the human IGS enriched for conserved regions and transcription associated factors.

The human IGS is shown at top, with the positions of Alu elements (green boxes), microsatellites (grey boxes), conserved regions (purple boxes), and previously identified IGS noncoding transcripts (black arrows) indicated. Below are chromatin and transcriptional features of seven human cell lines. The positions of the conserved regions are indicated by pale shading. For each cell line the presence of transcriptional start site (TSS), promoter (Prom), enhancer (Enh), and CTCF segmentation states, obtained by merging peaks for histone modification, Pol II and CTCF using Segway, are indicated. Below these, CAGE peaks are shown for the forward (black boxes) and reverse (red boxes) strands (CAGE stem cell data come from H9-hESC, not H1-hESC), followed by long poly(A+) and poly(A-) transcripts (green and blue arrows, respectively) with FPKM values >1; gray arrows indicate transcripts with FPKM < 1. Arrowheads indicate the direction of transcription. Peaks of small RNA are shown in pink. Zones 1 and 2 that are enriched for conserved regions and transcription-associated factors are boxed in red. Not all features have data available for all cell lines.

Fig 5. Origin replication complex (ORC) and double strand break (DSB) occurrence in the rDNA.

The black plot represents enrichment of ORC in Hela-S3 cells and grey boxes below represent the position of peaks. Scale on the left is the -fold enrichment, and the scale above shows the position in the rDNA. Purple boxes represent conserved regions. The predicted chromatin states: transcription start site (TSS; green boxes), promoter (pink boxes), and enhancer (orange boxes) are shown. CAGE peaks are shown as black boxes (positive strand). Long poly(A+) and poly(A-) transcripts with FPKM values > 1 are shown as green and blue boxes, respectively. Gray arrows show transcripts with FPKM < 1. Arrows indicate the direction of transcription. The purple plot at bottom represents the DSB sites in HEK293T cells.

Fig 6. Conservation of human IGS transcripts amongst primates.

The human IGS is indicated at top along with the conserved regions (purple boxes), Alu elements (green boxes) and cdc27 pseudogene (pink box). Below are poly(A+) IGS transcripts from the HUVEC cell line, followed by total RNA chimpanzee IGS transcripts (orange), and poly(A+) IGS transcripts from chimpanzee, orangutan, and rhesus macaque (green boxes). Only transcripts that are in common with human are shown. Transcript names and their start/end coordinates are indicated alongside, as are percent identities between each transcript and the human IGS (in parentheses). Arrowheads indicate direction of transcription.