NEW YORK (GenomeWeb News) – Circular RNAs are far more common than previously appreciated, according to new research from a Stanford University team.
As they reported online in PLoS ONE last night, the researchers did deep RNA-sequencing on several leukemia samples as well as human cell lines derived from normal and cancerous human tissues. In each of the tissues examined, they found examples of transcripts with exons that were out of sequence compared to their order in the genome.
These rearranged transcripts not only represent a sizeable proportion of isoforms present for hundreds of genes, they reported, but also tend to form circular rather than linear structures.
"This research shows that there are still important fundamental biological programs to be discovered by experimental exploration coupled with new statistical and computational approaches," Stanford biochemistry researcher Julia Salzman, co-first author on the study, said in a statement. "These circular RNAs may represent a yet-to-be discovered biological process."
Generally speaking, pre-messenger RNAs undergo splicing to produce linear messenger RNA molecules that are subsequently translated into proteins. The mRNAs characterized so far are typically comprised of a linear series of exons falling in the same order that they do in the genome, the study authors explained. And while there have been reports of exons shuffling that produces circular RNAs, these were generally believed to involve a very small fraction of transcripts in human cells.
"All examples of circular transcripts reported to date in humans have been found to be expressed at low levels compared to the dominant canonical linear isoform, requiring sensitive nested PCR experiments for detection," they wrote.
But Salzman and her colleagues suspected that circular RNAs might be more common than all that. The researchers started their search for transcripts representing non-canonical splicing events by doing RNA-sequencing in tumor samples from five children with acute lymphoblastic leukemia, or ALL, but soon found hints that exon rearrangement and circular RNA splicing occurs in normal human cells as well.
By comparing reads generated by paired-end sequencing of ALL samples with the Illumina GAII to a database containing data on known exon-exon junctions in the human transcriptome, the team tracked down more than 1,200 genes represented by transcripts with scrambled exons.
Moreover, many of these transcripts represented a relatively large proportion of the isoforms present for a given gene — results that the researchers verified using RT-PCR.
This exon shuffling was not limited to the leukemia cells, but also turned up in cells from the HeLa cell line and normal primary human blood cells tested by PCR. Moreover, the team's RNA-sequencing experiments using three white blood cell types pointed to more than 800 genes represented by rearranged isoforms that made up at least 10 percent of transcripts.
Another 2,748 isoforms with exons in a non-canonical sequence fell out of the researchers' analyses of publicly available transcriptome data for HeLa cells and human embryonic stem cells. And analyses of publicly available RNA-sequencing data generated on mouse brain samples suggest such isoforms are found in other animals as well.
From their subsequent statistical and biochemical analyses, meanwhile, researchers concluded that exon scrambling often produces circular rather than linear transcripts, with hundreds of these transcripts being highly expressed — in some cases at levels similar to canonical isoforms.
"Our results suggest that a non-canonical mode of RNA splicing, resulting in a circular RNA isoform, is a general feature of the gene expression program in human cells," the researchers wrote.
While it's too early to say what role circular transcripts might have in human cells, if any, those involved in the study argued that the extent to which unusual splicing turned up in their experiments points to possible functions for circular RNAs.
"We cannot yet rule out the possibility that they are incidental to regulated splicing of a conventional linear RNA and perhaps accumulate due to their relative resistance to degradation," the team noted.
"However," they added, "the high abundance of many circular RNA isoforms relative to their linear counterparts and lack of evidence for the predicted alternatively spliced linear RNA co-product suggest that circular RNA isoforms are not simply accidental byproducts of splicing."