Novartis, UNC study suggests mRNA structures are a common, untapped pool of drug targets

In drug discovery, proteins may be king—they are the target of nearly all small-molecule drugs. But there remains an even larger untapped drug target: RNA.

The classic view of RNA is one of a long, floppy molecule, lacking the characteristic, highly ordered structure that makes some proteins druggable with small molecules. Messenger RNA (mRNA), the molecular blueprint for making proteins, has a particularly bad reputation for being nothing more than a spindly courier molecule. Yet a growing cadre of scientists and biotech startups are warming up to the idea that certain RNAs may form druggable structures just like proteins.

Now, a new study from researchers at the University of North Carolina and the Novartis Institutes for Biomedical Research demonstrates that structure in mRNA is more widespread and important than previously imagined. The team boldly suggests that functional structures are a hitherto unappreciated feature found ubiquitously throughout mRNAs, and that thousands more of these structures remain to be discovered.

Furthermore, the study shines a light on the Swiss pharma giant’s intense interest in finding small molecules that have effects on RNAs involved in disease. Novartis says it has spent several years studying the druggability of RNAs with the hope of blocking mRNAs that would normally create disease-causing proteins.

“Textbooks give the impression that mRNAs are linear molecules,” and that protein structure is all that matters, says Anthony Mustoe, a postdoc in Kevin Weeks’s lab at UNC. But for decades, scientists have hypothesized that structure in RNA can affect its function too. “The limiting factor, until recently, is that we’ve only been able to study individual molecules,” he says.

To overcome that limitation, Mustoe, Weeks, and other researchers at UNC and Novartis used a chemical probing technique to map RNA structure across the Escherichia coli transcriptome—the entirety of its RNA. The method, designed by Weeks, combines a chemical reagent with high-throughput sequencing to identify whether regions of RNA are stable or unconstrained. Floppy RNA structures leave individual RNA bases exposed, so they react easily with Weeks’ reagent. In contrast, steady structures’ reactivity is lower (Cell 2018, DOI: 10.1016/j.cell.2018.02.034).

The team then looked at data from the most abundant RNAs in the cells—194 RNA transcripts encoding 400 genes. (Bacteria, unlike humans, sometimes have multiple genes encoded on a single mRNA.) Stable mRNA structures were shockingly pervasive. “This really wowed us,” says Razvan Nutiu, a biologist who led the Novartis portion of the team. “Basically every gene shows some sort of a structure, and all of the structures seem to be very diverse and very different from each other.”

“It is kind of rewriting the textbooks,” says Karissa Sanbonmatsu, an RNA scientist at Los Alamos National Laboratory who was not involved in the study. Individual instances of RNA structure are well-known, “But no one has ever gone out and found so many structures in mRNA.”

To determine if those structures were important, the group tested the protein levels produced from regions of mRNA with varying degrees of structure. They found that when mRNA is highly structured in its ribosome binding site—the place where a cell’s protein-making machinery takes hold of the strand—protein expression levels were much lower. When the ribosome binding site had little structure, ribosomes easily translated the mRNA into many copies of the proteins.

“Every mRNA essentially has its own personality,” Weeks says. “The expression of the proteins from essentially every single gene that we looked at was regulated in a significant, notable, and important way by RNA structure.”

“This is really a tour de force,” says Matt Disney, an RNA chemist at the Scripps Research Institute Florida who was not involved in the study, but recently launched his own company to design small-molecule drugs that bind RNA. “At minimum this paper will start to generate a lot of hypotheses about RNA’s regulatory role, but it will also elucidate some drug targets.”

For Novartis, this experiment is just the beginning. “I think the potential is huge,” says Nutiu, who currently leads a group focused on the chemical biology of RNA at Novartis. “The ultimate goal is definitely to study the structure of RNAs in humans, how the RNA structure influences function, and how this function is linked to disease.”

Nutiu says that Novartis is targeting RNAs involved in cancer, and cardiovascular, musculoskeletal, neurological, and other diseases. In addition to working on mRNA, Novartis is also studying non-coding RNAs, which don’t make proteins, but are involved in the regulation of other genes. “I am extremely excited to see the commitment from Novartis and their openness towards doing things differently,” Nutiu adds. “We are very interested in collaborating and helping the field move this to the next level.”

Although mRNA-targeted drugs are likely years away, the prospects are exciting RNA scientists and helping shine fresh light on the field. “Right now there are only a few obscure groups working on mRNA structure,” Sanbonmatsu says. “But this will really send a big signal out there that it exists and it is worthwhile to look at.”