Noncoding RNA Shows Context-Dependent Function

Sense and antisense transcription on track. Single-molecule imaging of sense and antisense transcription reveals that the same DNA strand is transcribed in two fundamentally different ways. Antisense is synthesized with random intervals as single RNAs, whereas sense RNA is produced in bursts of transcription. Antisense transcription inhibits sense transcription when the sense gene is uninduced. However, it has no effect on the actively transcribed gene, even when both are transcribed at the same time, as depicted in the image.

In addition to well-studied protein coding sequences, it is known that the genomes of higher organisms produce numerous noncoding RNAs (ncRNAs). Important roles for some ncRNAs in cell function have been demonstrated, though usually on a case-by-case basis, leading some scientists to argue that the majority of ncRNA production is just “noise” that results from the imperfect transcription machinery. The fact that many ncRNAs overlap with coding genes has hampered studies of their activities. Thus, a general understanding of whether ncRNA production is functional or not is lacking. To address this issue, Daniel Larson, Ph.D., of CCR’s Laboratory of Receptor Biology and Gene Expression, and his colleagues developed a new approach using single-molecule imaging in living cells. The researchers specifically labeled coding and ncRNAs from the GAL locus in yeast, which regulates the galactose response. Glucose is the preferred source of carbon for yeast, but when it is scarce, genes within the GALlocus, including GAL10 and GAL1, are activated to allow the metabolism of galactose.

The researchers labeled GAL10 RNA, which overlaps with the GAL10 ncRNA but on the opposite DNA strand, with red fluorescent protein and GAL10 ncRNA with green fluorescent protein by incorporating targeting sequences into their genes. These sequences did not affect the transcription level of either gene. Through observations of single cells, the investigators found thatGAL10 ncRNA was produced when yeast were grown in raffinose, another sugar that fails to induce the GALlocus, and in galactose prior to switching on GAL10 transcription. The GAL10 ncRNAs remained at the transcription site only long enough to be synthesized, and the distribution of their transcription events suggested initiation events were not correlated and showed no bursts of transcription. When cells received galactose, the scientists observed GAL10 transcription bursts, the first directly visualized in yeast. They also showed that GAL10 RNAs only remained at the site of transcription during synthesis. These studies demonstrated that the same DNA template can be transcribed in fundamentally different ways.

To investigate the switch from GAL10 ncRNA to GAL10 RNA, the researchers tracked cells over time after galactose exposure. They found that GAL10 ncRNA transcription occurred before or with GAL10 transcription, but mostly disappeared after the switch to GAL10. Previous conflicting reports suggested GAL10 ncRNA acts positively or negatively on GAL10 transcription. However, comparing cells exposed to galactose with and without GAL10 ncRNA transcription revealed no difference in GAL10 transcription, showing that GAL10 ncRNA does not affect GAL10 activation and, in these conditions, that GAL10 ncRNA is not functional.

The investigators next wanted to assess the role of GAL10 ncRNA in other conditions. They applied CRISPR interference, which can strand-specifically repress transcription, by generating guide RNAs to positions within the GAL10 ncRNA. Targeting a site 116 nucleotides downstream of the GAL10 ncRNA start site completely eliminated GAL10 ncRNA transcription. Encouragingly, in the presence of galactose and the guide RNA, there was no effect on GAL10 transcription. To study the effect of GAL10 ncRNA on GAL10 activation, the scientists grew cells in raffinose or glucose with or without the guide RNAs then added galactose and examined GAL10levels by Northern blot. Blocking GAL10 ncRNA caused faster GAL10 induction in cells grown in glucose but not raffinose. Similar results were observed for GAL1, which shares an upstream activating sequence withGAL10. This increased activation was due to increased sensitivity to galactose. When GAL10 ncRNA was inhibited and cells were grown in mixed glucose and galactose, more cells switched to galactose metabolism, demonstrating that GAL10 ncRNA is involved in modulating the relative preference for galactose over glucose by modulating the GAL activation threshold.

The researchers hypothesized that, because GAL10 ncRNA regulates the expression of GAL1 and GAL10, removing the ncRNA would increase basal levels of GAL1 and GAL10 expression. Since Gal1 protein is a sensor for galactose, its increase would heighten sensitivity to this sugar. To test the idea, they used quantitative RT-PCR on cells grown in raffinose to monitor GAL10 levels. Inhibiting GAL10 ncRNA with guide RNAs led to a five-fold increase in GAL10 expression. Similar results were obtained from cells grown in glucose and from cells in which GAL10 ncRNA was silenced by mutating transcription factor binding sites, indicating a direct role of GAL10 ncRNA in repressing transcription from the GAL locus. As further support, the investigators found that overexpression of GAL1 also increased sensitivity to galactose. Perhaps somewhat surprisingly, the increased sensitivity to galactose in the absence of GAL10 ncRNA led to decreased cell fitness, particularly in cells grown in mixtures of low glucose and galactose. These results suggest that the selective advantage for maintaining GAL10 ncRNA is in ensuring yeast cells use glucose when it is available.

The investigators then wondered whether GAL10 ncRNA could act at additional points within the galactose regulatory network. Mathematical modeling of a single point of regulation agreed with their observed data, and genome-wide expression analysis revealed that controlling the transcription of the GAL locus is only role ofGAL10 ncRNA. Taken together, these studies demonstrate that GAL10 ncRNA regulates the yeast galactose response by limiting inappropriate transcription from the GAL locus and that the same ncRNA can be functional or not depending on the cellular context.

Summary Posted: Sun, 11/01/2015


Lenstra TL, Coulon A, Chow CC, Larson DR. Single-Molecule Imaging Reveals a Switch between Spurious and Functional ncRNA Transcription. Molecular Cell. November 5, 2015. PubMed Link