Pauses in Transcription May Help Cells Fine-Tune Gene Expression
In eukaryotes, gene transcription is carried out by three related RNA polymerases. RNA polymerase (Pol) I and Pol III produce ribosomal and transfer RNAs, respectively, whereas Pol II transcribes protein-coding genes and produces mRNA, which serves as the template for protein synthesis.
Polymerase-associated factors enable the polymerases to recognize different promoters and to transcribe different classes of genes, to receive different regulatory signals, to direct the co-transcriptional processing of RNA transcripts, and to couple transcription to changes in chromatin. Therefore, RNA polymerases are not only the key enzymes of gene expression but also the central coordinators of nuclear events.
A recent investigation from the Stowers Institute for Medical Research shows that each time a gene is “turned on” or expressed, a molecule called RNA polymerase must position itself at a specific spot along the DNA and travel down its winding strands, transcribing the gene from start to finish. When one polymerase pauses on the track, it keeps other polymerases from entering the starting gate. Paradoxically, polymerases that pause for longer periods of time can mediate faster and more synchronized gene expression in response to the kinds of signals triggered by various stages of
development or dysregulated in cancer.“We discovered a traffic rule that appears to guide the process of transcription,” says Stowers Associate Investigator Julia Zeitlinger, Ph.D., who led the study. “Genes are transcribed through bursts of activity, like rush hour. Traffic is pretty dangerous. It makes sense to tightly control the number of cars on the road and minimize the number of accidents.”
Julia Zeitlinger and graduate student Wanqing Shao of the Stowers Institute for Medical Research in Kansas City, Missouri, stumbled upon the result by using a drug to freeze RNA polymerase II and transcription factors in place in Drosophila cells, then analyzing the positions of polymerases throughout the genome with a technique called ChIP-nexus. “We could clearly see minimal initiation in the presence of paused polymerase,” she says. The result was initially surprising, she adds, because it’s been shown that promoters with a strong propensity for downstream polymerase-pausing are associated with faster gene expression in response to a developmental signal.
This finding could be particularly important in the context of transcription bursts, a rapid succession of numerous transcribing polymerases interspersed by periods of inactivity that can last minutes or even hours. Since paused polymerases were observed to be so stable, the researchers think that they not only block other polymerases from immediately following them during bursts of transcription, but that they also sit there in between bursts of transcription.
“I think that the idea of having one rate-limiting step is sort of appealing, and a lot of biologists sort of intuitively think that way,” she says. “But it’s actually not a good way to design a system . . . because it makes it more stochastic, more random.” It makes sense that there would be a system in place to ensure transcription isn’t simultaneously paused and initiated on the same gene, she says, but the mechanics of how the paused polymerase wards off new initiation remain to be elucidated.
Craig Kaplan, a molecular biologist at Texas A&M University, says this and other recent studies also make it clear that pausing can occur for very different lengths of time depending on the promoter. This helps give cells a “buffet of choices in how expression may happen,” he notes. “The regulation doesn’t have to be thought of as on or off; it may be how frequently you make a transcript, or whether you’re making your transcript in bursts, or whether you’re making transcripts evenly.”
But Patrick Cramer of the Max Planck Institute for Biophysical Chemistry in Göttingen, Germany, wants to see the results confirmed by other studies. “Occupancy changes are certainly an indication for changes in kinetics, but are not a proof, because occupancy can change either because the number of factors bound to DNA changes or because their residency time on DNA changes, or both,” he says.
“Transcription is an old field, and I think it’s often seen as, ‘Oh, it’s so well studied,’” says Zeitlinger. “[But] there are a lot of things we don’t understand. . . . There are so many open questions in the field that are quite fundamental.”
