Circular RNAs (circRNAs), closed loops of noncoding RNA, drift through the cytoplasm like so many ring buoys floating on the sea. Precisely what the circRNAs are doing there hasn’t been clear. Because they are especially abundant in brain cells, circRNAs have been thought to have a role in brain function, possibly via the regulation of gene expression. This possibility looks more likely than ever, now that a new study indicates that circRNAs, which are stable and long-lived, help preserve relatively fragile microRNAs (miRNAs), and thereby influence brain function.
Scientists based at the Max Delbrück Center (MDC) deliberately depleted cells of circRNA to see what would happen. Specifically, these scientists, led by Nikolaus Rajewsky, Ph.D., deployed CRISPR/Cas9 technology to deprive mice of the gene for Cdr1as, a circRNA known to be predominantly expressed in excitatory neurons. Then the scientists assessed how miRNA expression, electrophysiology, and behavior might be affected.
The expression of most miRNAs remained unperturbed. However, miR-7 was downregulated and miR-671 upregulated. These changes were post-transcriptional, consistent with the idea that Cdr1as usually interacts with these miRNAs in the cytoplasm.
The changes in miRNA concentration had dramatic effects on the messenger RNA (mRNA) and proteins produced by nerve cells, especially for a group called “immediate early genes.” They are part of the first wave of responses when stimuli are presented to neurons. Also affected were mRNAs that encode proteins involved in the maintenance of the animals' sleep–wake cycles.
Interestingly, the researchers observed increased spontaneous neurotransmitter vesicle release and decreased synaptic responses to two consecutive stimuli. Additional behavioral analyses showed that while mice lacking Cdr1as demonstrated normal social behavior, unaffected anxiety levels, unperturbed locomotor activity, and no deficits in recognition memory, they exhibited impairments in sensory and cognitive processing (called sensorimotor gating)—a deficiency associated with neuropsychiatric disorders.
The MDC scientists detailed their findings August 10 in the journal Science, in an article entitled “Loss of a Mammalian Circular RNA Locus Causes miRNA Deregulation and Affects Brain Function.” This article presents data that—for the first time—link a circRNA to brain function.
“Expression of immediate early genes such as Fos, a direct miR-7 target, was enhanced in Cdr1as-deficient brains, providing a possible molecular link to the behavioral phenotype,” wrote the article’s authors. “Our data indicate an in vivo loss-of-function circRNA phenotype and suggest that interactions between circRNAs and miRNAs are important for normal brain function.”
RNA is much more than the mundane messenger between DNA and the protein it encodes. Indeed, there are several different kinds of noncoding RNA molecules. In the past 20 years, scientists have discovered some two dozen RNA varieties that form intricate networks within the molecular microcosm. The most enigmatic among them are circRNAs, an unusual class of RNAs whose heads are connected to their tails to form a covalently closed ring. These structures had for decades been dismissed as a rare, exotic RNA species. In fact, the opposite is true. Current RNA-sequencing analyses have revealed that they are a large class of RNA, which is highly expressed in brain tissues.
In 2013, two pioneering studies that characterized circRNAs appeared in the journal Nature, one of them by Rajewsky and his team. Intriguingly, most circRNAs are unusually stable, floating in the cytoplasm for hours and even days on end. The systems biologists proposed that—at least sometimes— circRNAs serve gene regulation. Cdr1as, a large single-stranded RNA loop that is 1500 nucleotides around, might act as a “sponge” for miRNAs. For example, it offers more than 70 binding sites for a miRNA called miR-7. miRNAs are short RNA molecules that typically bind to complementary sequences in mRNAs, thereby controlling the amounts of specific proteins produced by cells.
Additionally, Rajewsky and his collaborators mined databases and discovered thousands of different circRNAs in nematode worms, mice, and humans. Most of them were highly conserved throughout evolution. “We had found a parallel universe of unexplored RNAs,” said Rajewsky. “Since publication, the field has exploded; hundreds of new studies have been carried out.”
For the current paper in Science, the systems biologists teamed up with the lab of Carmen Birchmeier, Ph.D., at the MDC to reconsider Cdr1as. “This particular circle can be found in excitatory neurons but not in glial cells,” noted Monika Piwecka, Ph.D., one of the first authors of the paper and coordinator of most of the experiments. “In brain tissues of mice and humans, there are two miRNAs called miR-7 and miR-671 that bind to it.”
In a next step, Rajewsky and his collaborators selectively deleted the circRNA Cdr1as in mice using CRISPR/Cas9. It was at this point that the MDC team found that miR-7 was downregulated and miR-671 upregulated.
“This indicates that Cdr1as usually stabilizes or transports miR-7 in neurons by sponging them up, while miR-167 might serve to regulate levels of this particular circRNA,” asserted Rajewsky. If miRNA floated in the cytoplasm without binding anywhere, it would get broken down as waste. The circle would prevent that and also carry it to new places like the synapses. “Maybe we should think about Cdr1as not as a 'sponge' but as a 'boat,'” he suggested. “It prevents its passengers from drowning and also moves on to new ports.”
Using single-cell electrophysiology, Charité-researcher Christian Rosenmund, Ph.D., observed that spontaneous vesicle release at the synapse happened twice as often. The synaptic responses to two consecutive stimuli were also altered. Additional behavioral analyses performed at the MDC mirrored these findings. Even though the mice appeared normal in many ways, they were unable to tune down their responses to external signals such as noises. Similar disruptions in prepulse inhibition have been noted in patients suffering from schizophrenia or other psychiatric diseases.
It is an everyday experience how much we depend on this filtering function: When a loud noise suddenly disturbs the quiet atmosphere of a library, you cannot avoid being alarmed. The same bang, however, will seem much less threatening next to a construction site. In this instance, the brain has had the chance to process previous noises and filter out unnecessary information. Therefore, the startle reflex is dampened (prepulse inhibition). This basic brain function that allows healthy animals and people to temporarily adapt to a strong stimulus and avoid information overload has now been linked to Cdr1as.
“Functionally, our data suggest that Cdr1as and its direct interactions with miRNAs are important for sensorimotor gating and synaptic transmission,” explained Rajewsky. “More generally, since the brain is an organ with exceptionally high and diverse expression of circRNAs, we believe that our data suggest the existence of a previously unknown layer of biological functions carried out by these circles.”