Small interfering RNA (siRNA) drugs take advantage of RNA interference (RNAi) to “silence” genes with major advantages: They can target any gene without changing its underlying DNA. But designing these drugs has been difficult because so many different sequences of siRNA must be tested to find out which one works best to silence the target. So far, drug developers have been in the dark about why certain siRNA sequences work so much better than others, or how to predict which ones will be successful.
Ian MacRae, a biologist at The Scripps Research Institute, wanted to start to crack open these questions and help contribute to better drug design from the start. MacRae’s team recently published the first high-resolution structural images of the human protein essential to the RNAi process, Argonaute 2, in action while it cuts and degrades mRNA. Their findings revealed that Argonaute 2 doesn’t just act as a passive clamp to bind an siRNA — instead, the shape of the RNA duplex, which only takes form after an siRNA engages its target mRNA, becomes incredibly important.
“To become catalytically active, the RNA duplex has to bend, widen, and compress in very specific ways inside the protein,” MacRae told DDN. “I was surprised by how much the shape of the RNA duplex matters.”
A new view
The new work changes the field’s understanding from a simple model in which the guide pairing with the target activates cleavage to one that recognizes the importance of pairing “plus the right RNA shape,” MacRae explained.
To make that discovery, the scientists used cryo-electron microscopy (cryo-EM) to capture Argonaute 2 in action right before it cuts the RNA target. Their new insider’s view also allowed them to uncover two new catalytic residues, which are the amino acids that actually perform the cutting of target mRNA.
“Argonaute has been studied for decades, so we thought the chemistry of the active site was basically a solved problem. Instead, we found two mobile amino acids that respond to RNA binding and are essential for efficient catalysis,” said MacRae.
The newly revealed amino acids, Lysine 709 and Arginine 710, can also help explain why some siRNA drugs work better than others. Lysine 709 doesn’t enter its cutting position until the duplex is deformed in the correct way, while Arginine 710 monitors the position of pyrimidine-rich sites and favors them for cutting.
Towards rational drug design
The study offers a new blueprint to design better siRNA drugs — getting developers closer to rational drug design. “Before, we knew some sequences and chemical modifications performed better than others, but we didn't have a clear picture of what a productive guide-target duplex looked like inside Argonaute,” said MacRae. “Now, we can start asking how sequence and chemistry influence the way the RNA fits into the protein and whether it adopts the geometry associated with efficient cleavage.”
In future work, MacRae’s team plans to further investigate how different siRNA sequences and chemical modifications impact Argonaute 2’s activity. Their hope is that this knowledge will contribute even more to the design of siRNA drugs based on how they will interact with Argonaute 2 and skip through much of the trial-and-error aspects.
“We’re also interested in working with therapeutic RNA companies, because this is exactly the kind of basic mechanism that could feed directly into better drug design,” said MacRae.
