NEW YORK (GenomeWeb News) – Coding and non-coding messenger RNAs can interact with — and regulate one another — through microRNAs, according to a paper appearing online today in Nature.
Researchers from Harvard University and the Memorial Sloan-Kettering Cancer Center looked at the interactions between the transcripts of the tumor suppressor gene PTEN and a related pseudogene PTENP1 in an effort to better understand the interplay between various RNAs. They found that the 3' untranslated region of the pseudogene could actually compete with the protein-coding transcript for miRNA binding, with increasing pseudogene levels leading to decreased repression of PTEN transcripts by these miRNAs.
"The miRNA becomes the letter of this code whereby the coding and non-coding RNA talk to each other," senior author Pier Paolo Pandolfi, director of research at Harvard's Beth Israel Deaconess Medical Center, told GenomeWeb Daily News. "Any messenger — coding or non-coding — can talk to another messenger simply by sequestering the miRNAs."
The presence of such rival RNAs, which the researchers dubbed "competitive endogenous RNAs" or ceRNAs, is offering clues about the function of pseudogenes and other non-coding RNAs. And, researchers say, unraveling these regulatory RNA interactions may help decipher a range of genomic patterns — including the consequences of mutations found in cancer genomes.
"[N]ot only have we discovered a new language for mRNA, but we have also translated the previously unknown language of up to 17,000 pseudogenes and at least 10,000 long non-coding RNAs," Pandolfi said in a statement. "Consequently, we now know the function of an estimated 30,000 new entities, offering a novel dimension by which cellular and tumor biology can be regulated, and effectively doubling the size of the functional genome."
MicroRNAs are small, non-coding RNAs that are thought to target and interfere with mRNA transcripts — usually leading to negative regulation of these transcripts, Pandolfi explained.
But he and his co-workers suspected there was more to RNA function than previously appreciated. "One day, I had this idea … whereby it's not the miRNA that regulates the messenger, but it's rather the messenger that regulates the miRNA by sequestering it," Pandolfi explained.
Because each miRNA interacts with many different genes, he added, it seemed feasible that miRNAs might act as intermediaries between competing mRNAs. If so, interactions with specific mRNAs could potentially sequester miRNAs, the team predicted, preventing them from inhibiting other mRNAs.
To test this, the researchers first looked at whether the expression of the PTEN tumor suppressor gene was related to the presence or absence of mRNA from a related pseudogene called PTENP1.
Although PTENP1 does not code for a functional protein, it contains several miRNA binding sites also found in PTEN, including those targeted by miRNAs in the miR-17, miR-21, miR-214, miR-19, and miR-26 families.
Indeed, the team found that when they over-expressed the 3' UTR from the pseudogene, the PTEN transcript was more highly expressed, consistent with a model in which the pseudogene and coding gene compete for inhibitory miRNAs.
Such interactions apparently occur in both normal and cancerous human tissues, the researchers noted, but seem to depend on the presence of mature miRNAs: when they did similar experiments in cells lacking an enzyme needed for miRNA maturation, the team did not see the same interaction between the pseudogene and coding gene transcripts.
The researchers found similar results when they investigated another oncogene, KRAS, and a related pseudogene, KRAS1P.
And Pandolfi says such interactions are not limited to the cancer-related genes. Rather, he and his colleagues argue that bioinformatics can be used to predict which RNAs are interacting with one another — and the consequences of these interactions.
"[T]he discovery of an miRNA decoy function for pseudogenes identifies these transcripts as biologically active units," the researchers wrote. "[T]he analysis of pseudogene expression level and genomic status in tumorigenesis needs to be undertaken systematically to further our understanding of disease progression."
The team is currently using the same approach to evaluate other sorts of long, non-coding RNA and is also studying how the newly identified RNA interactions might influence various cellular processes, including differentiation and cellular re-programming.
Down the road, Pandolfi noted, it may be possible to exploit such RNA interactions to come up with therapeutic strategies based on manipulating levels of one mRNA by altering those of another.
"Now, all of a sudden, we are able to functionalize all the transcribed genome," Pandolfi said. "This regulatory network is this huge RNA dimension, which now we start to decipher."