NEW YORK (GenomeWeb News) – New research is shedding light on the extensive transcription of the fission yeast genome — and revealing the extent to which the model organism’s transcriptome varies under different growth conditions and stages of the cell cycle.
In a paper published in the advance online edition of Nature yesterday, an international team of researchers used RNA-Seq and tiling microarrays to map the dynamics of the Schizosaccharomyces pombe transcriptome. The team reported that more than 90 percent of the genome is transcribed, including hundreds of new transcripts detected at low levels. They were also able to analyze untranslated regions in the genome and pinpoint splicing events that varied with time, location, and transcript levels.
In an effort to understand the S. pombe transcriptome as completely as possible, senior author Jürg Bähler, a functional genomics researcher at the Wellcome Trust Sanger Institute, and his team synthesized complementary DNA from poly(A)-enriched RNA isolated from S. pombe grown under different conditions. They then used an Illumina Genome Analyzer to sequence this cDNA and mapped the reads back to the spliced and un-spliced versions of the reference genome.
They also verified their results in rapidly growing cells, meiotic cells, cells grown under oxidative stress or heat shock, and RNA processing mutants by using high-density Affymetrix tiling arrays.
Based on more than 23 million reads from exponentially growing cells and more than 99 million reads from cells at five different stages of meiosis, they reported that almost the entire genome — more than 90 percent — was detectably transcribed.
“[W]e obtained sequence data from ~94 percent and >99 percent of the nuclear and mitochondrial genomes, respectively,” Bahler and his colleagues wrote, “suggesting that almost the entire genome is transcribed to some degree, consistent with the considerable overlap and complexity among different transcripts reported for other eukaryotes.”
And, like the budding yeast Saccharomyces cerevisiae, the fission yeast transcriptome is proving to be more complex than expected. The S. pombe transcriptome held surprises, both within coding and non-coding regions of the genome. Based on sequencing and tiling chip data, the team made 75 revisions to protein-coding regions in the fission yeast gene annotation and pinpointed some 20 new introns in previously identified genes.
They also identified at least 453 new transcripts within the fission yeast transcriptome. Just 26 of these were predicted to code for proteins, while three dozen or so were non-coding RNAs that overlapped with the antisense strand of known genes. When the researchers used RT-PCR to look more closely at 14 non-coding transcripts, they also found evidence for bi-directional transcription across the genome.
Still, the authors noted that, “[g]iven the ubiquitous transcription throughout the genome, the novel transcripts described here probably only hint at the true level of transcriptional complexity” in the S. pombe genome.
In general, the researchers found fewer reads for the newly transcribed regions and subsequent experiments suggest most of them are transcribed at low levels. Just 13 could be picked up in actively growing cells by tiling arrays and another 80 or so were only detected under specific conditions, particularly during meiosis or quiescence.
The researchers also reported new information about the nature of untranslated regions, splicing, and polyadenylation in fission yeast. For example, they found dozens of transcripts that had alternative start sites during meiosis or when S. pombe cells were grown under stressful conditions. About 187 transcripts had variable polyadenylation sites.
The team also found a surprisingly large pool of unprocessed messenger RNA, containing introns and exons. And more than 250 genes actually seemed to be spliced more efficiently during meiosis. Splicing efficiency also seemed to increase with higher gene expression, the researchers noted, suggesting that transcription and splicing are “mechanistically linked.”
“[O]ur results reveal a surprising genome-wide regulation of splicing, largely reflecting transcript levels during proliferation or differentiation,” the authors concluded. “These data point to a global and condition-specific coupling between splicing efficiency and transcription, which may help to optimize and streamline gene expression programs.”
Alterations from the study have been incorporated into the S. pombe database.