The body's main molecular currency is most definitely proteins but less than 2 percent of the DNA in the human genome codes directly for them. So just what does that other 98 percent do? That’s where RNA comes in.
RNA is both the bridge between DNA and proteins and the toolbox that helps proteins do their everyday work. RNA exists in a variety of forms, each with a particular role and purpose, not all of which are fully understood. Penn scientists are at the forefront of exploring and pushing back the boundaries of the bewilderingly complex world of RNA. And, Jeremy Wilusz, PhD, a new faculty member in the department of Biochemistry & Biophysics, is adding to that knowledge with a recently published paper in Genes & Development on circular RNAs.
Until recently, only a handful of these curious RNAs were known to be found in nature in such things as virus particles called viroids and in hepatitis delta virus, which causes human liver disease. Genomic sequencing, however, has now revealed the presence of thousands of circular RNAs in organisms ranging from archaea to humans. For example, human fibroblasts alone have more than 25,000 circular RNAs, notes Wilusz in an earlier commentary in Science.
“Over the last few years, it has become increasingly clear that many protein-coding genes purposely don’t make protein-coding mRNAs (and eventually working proteins), but instead make circular non-coding RNAs,” says Wilusz. “This observation is completely against the central dogma of molecular biology. We were very intrigued by this and aimed to figure out the mechanism that determines why some genes make circles, while others do not.” The G&D paper describes how a cell "decides" to make these circular RNAs.
Genes contain introns – sequences that don’t get translated into proteins -- and splicing needs to occur so that introns are removed and exons – the sequences that do lead to proteins – are joined together. Circular RNAs are generated when the splicing machinery fails to join the exons in the expected order.
Wilusz explains: “Normally, exon 1 joins to exon 2, which joins to exon 3, and so forth, to make a final linear RNA that can produce a protein. However, the cell can purposely mis-splice and, for example, join the end of exon 2 to the beginning of exon 2, rather than to the beginning of exon 3. This loop creates a circular RNA out of exon 2. Considering that this circular RNA does not make a protein, why would the cell need the splicing machinery to purposely mis-splice?”
Wilusz and his team figured out that if an exon is flanked by introns that contain repeat sequences (the real so-called “junk DNA,” he says) in a particular order, these repeats are able to bind to each other and redirect the splicing machinery to make a circular RNA. This provides an interesting mechanistic explanation for why only certain genes form circular RNAs, he proposes.
Wilusz says that these findings also have interesting evolutionary implications because repeat sequences differ significantly across the genomes of different species and so it is possible that a gene that may only produce a linear RNA in mice may predominately make a circular RNA in humans.
What exactly these circular RNAs do is still rather murky, notes Wilusz. He and others have written about some circular RNA “sponges’ that mop up another class of non-coding RNAs called microRNAs. In these cases, circular RNAs have many binding sites for a specific microRNA, thereby allowing many copies of the microRNA to be bound to a single circle. Nevertheless, the jury is still out on what the other thousands of circular RNA do.
“We at least now have ways to make circular RNAs and we can test the effect of their over-expression on cells," Wilusz concludes. In the future, his lab will make designer circular RNAs to determine their functions. In addition, the lab will determine whether circular RNAs are misregulated in various diseases, such as cancer, and if they can serve as novel therapeutic targets.
Image from A Circuitous Route to Noncoding RNA, Jeremy E. Wilusz and Phillip A. Sharp, Science 26 April 2013: 440-441. [DOI:10.1126/science.1238522]