A team of researchers from the Scripps Institute and Novartis this month reported new details about how small RNAs such as siRNAs and microRNAs interact with the RNA-induced silencing complex.
Specifically, the investigators found that highly complementary targets of small RNAs can accelerate the release of guide RNA from Argonaute2 by destabilizing RISC. This “unloading” activity can be further enhanced by mismatches between the target and the 5’ end of the RNA guide strand, but attenuated by mismatches to the 3’ end.
While the findings, which appeared in Molecular Cell, stem from in vitro experiments, evidence suggests that the observed effect of target RNA on Ago2/guide RNA interactions holds true in living cells, and may ultimately help with the design of more effective RNA therapeutics.
According to senior author and Scripps researcher Ian MacRae, the work detailed in Molecular Cell was an outgrowth of a previous effort in his lab to identify the crystal structure of Ago2.
“We were trying to crystalize [Ago2] and we noticed that … it always came bound with small RNAs, and it’s very hard to get those things out,” which raised the question of how — or even if — this occurs in nature, he explained to Gene Silencing News.
“We set up an assay to measure small RNA dissociation from Argonaute, and we found right away that dissociation is extremely slow,” MacRae said. “We couldn’t even really measure it because it was [on a] time scale that our assay wouldn’t let us monitor well.”
During these experiments, the researchers noticed that while guide RNA was “tightly bound” to Ago2, guide RNAs that did not have a 5’ phosphate dissociated from the complex, leading to the conclusion that the Ago-2/guide RNA complex is extremely stable and that a 5’ phosphate on the small RNA is “required for long-lived association” with the protein, they wrote in their paper.
Further study showed that binding to Ago2 stabilized small RNAs, protecting them from nuclease degradation.
In light of evidence showing that heterologous expression of miRNA targets can alter levels of their corresponding mature miRNAs in mammalian cells, MacRae and his colleagues next looked at the effect the introduction of target RNAs would have on Ago2/guide RNA interactions.
Surprisingly, they discovered that the complementary RNA induced dissociation of the guide strand.
“We had that result for about a year and didn’t know what to do with it because it was so strange and so counterintuitive,” he said. “But it was totally reproducible and seemed to be sequence-specific,” so they ultimately decided to investigate further.
“We found that mismatches on the 3’ end of the guide would slow it down and, strangely, mismatches to the 5’ end … would a lot of the time accelerate” dissociation.
“In the end, we decided that we’re probably seeing duplex loading in reverse,” MacRae said. When bound together, guide RNA and Ago form an “extremely tight complex,” he added. “I think by putting in complementary RNAs, we’re tapping into the loading process intrinsic to the protein, but … just going backwards.”
The researchers wrote that, given the protective effects of binding to Argonaute, “all miRNA degradation pathways have to contend with the high-afﬁnity interaction between Argonaute and its bound miRNA,” and that this barrier is lowered significantly upon binding to a highly complementary target.
“We therefore hypothesize that some miRNA decay mechanisms may capitalize on this effect,” they stated.
MacRae pointed out that there are many examples where some mature miRNAs are much less stable than others in certain cellular contexts, most notably members of the miR-16 family, which are constitutively unstable in mouse 3T3 cells, allowing for dynamic transcriptional control of the family during the cell cycle.
While the mechanisms behind this instability are not yet known, the data reported in Molecular Cell suggest that it may be based on this newly described unloading activity.
“Our results also have implications for design of … chemical antagonists of miRNAs and synthetic siRNAs,” the study’s authors wrote.
“What you could do, if you got all the rules down and appreciate all these effects, is design siRNAs with different lifetimes,” allowing for RNAi drugs with controlled durations, MacRae said.
At the same time, he said that it might be possible to develop an miRNA antagonist that would remove the guide strand from RISC for degradation by ribonucleases, but stay intact and active. “That way, [the antagonist] could be catalytic and you’d have to deliver many fewer molecules per cell,” which could help address issues with delivery and toxicity, MacRae said.
Such real-world applications of his team’s findings, however, require further research, he cautioned.