Global identification of noncoding RNAs in Saccharomyces cerevisiae by modulating an essential RNA processing pathway

Abstract

Noncoding RNAs (ncRNAs) perform essential cellular tasks and play key regulatory roles in all organisms. Although several new ncRNAs in yeast were recently discovered by individual studies, to our knowledge no comprehensive empirical search has been conducted. We demonstrate a powerful and versatile method for global identification of previously undescribed ncRNAs by modulating an essential RNA processing pathway through the depletion of a key ribonucleoprotein enzyme component, and monitoring differential transcriptional activities with genome tiling arrays during the time course of the ribonucleoprotein depletion. The entire Saccharomyces cerevisiae genome was scanned during cell growth decay regulated by promoter-mediated depletion of Rpp1, an essential and functionally conserved protein component of the RNase P enzyme. In addition to most verified genes and ncRNAs, expression was detected in 98 antisense and intergenic regions, 74 that were further confirmed to contain previously undescribed RNAs. A class of ncRNAs, located antisense to coding regions of verified protein-coding genes, is discussed in this article. One member, HRA1, is likely involved in 18S rRNA maturation.

Fig. 1.

Experimental design and distribution of the expressed probes. (a) A maskless array synthesizer was used to synthesize 36-mer oligonucleotide probes covering both strands of the entire S. cerevisiae genome into eight arrays. Arrays were hybridized with RNA from Rpp1-depleted cells. (b) The pie chart displays the distribution of 72,633 expressed probes matching different genomic features: red, verified ORF; black, antisense to any annotated feature on the genome; green, uncharacterized ORF; blue, unannotated intergenic region; yellow, untranslated region, including 50-base upstream of ATG sequence and 200-base downstream of stop codon for all verified and uncharacterized ORFs; brown, other genomic features excluding ORF and RNA; gray, nonprotein coding RNA; and violet, dubious ORFs.

Fig. 2.

Transcription of a known and of a previously undescribed ncRNA gene. Hybridization data (y axis) are plotted along the genome (x axis) displaying signals for ncRNA genes of interest. Lines of different colors represent observations at eight different time points. Although the figure shows data only from the Watson strand, the locations of annotated genes on both strands at nearby regions are shown below each image (green rectangles, annotated genes). (a) SNR10 is a 245-bases-long essential H/ACA small RNA required for modification of rRNA precursor sequence. Based on its mature sequence, transcription at its 3′ end extends almost 400 bases beyond its annotation, suggesting that its precursor RNA is twice as long. The time dependence of the expression pattern for SNR10 is typical of other ncRNAs. The signal is strong at the first time point (blue), when RPP1 is overexpressed, but the signal falls immediately (green) to basal level (yellow = control). At later points, the signal increases gradually. (b) The previously undescribed RNA HRA1, described in this work, is located antisense to ORF DRS2. HRA1 is expressed only at later time points.

Fig. 3.

Promoter-based ncRNA that may function in RNA-mediated transcription regulation. (a) SRG1, located upstream of SER3, regulates the transcription of SER3 (9). (b) Confirmed putative novel RNA MAN1 and MAN2, respectively located upstream of GDH3 and YAL061C, are shown. (Inset) Lanes 1 and 2 show RT-PCR results for MAN1 and MAN2. MAN1 is located between the bases 31,200–31,500 on chromosome 1. MAN2 is located between the bases 33,000–33,350 on chromosome 1.