RNA approach restores CFTR function in preclinical cystic fibrosis models

An experimental RNA-based approach restored production and function of CFTR — the protein that is faulty or missing in cystic fibrosis (CF) — in preclinical models carrying disease-causing nonsense mutations, a study shows.

These mutations introduce an incorrect stop signal in the genetic instructions for making CFTR, causing cells to stop building the protein too early. Researchers developed chemically modified transfer RNA (tRNA) molecules that can bypass these faulty stop signals and allow cells to finish making full-length CFTR. The tRNAs were delivered in tiny fat-based particles called lipid nanoparticles.

Recommended Reading

Aug. 17, 2026 News by Margarida Maia, PhD

CFTR modulators may ease anxiety, depression in children with CF

RNA approach restores CFTR in preclinical models

The approach restored CFTR production and function in lab-grown human airway cells and other preclinical models, including miniature tissue models made from cells donated by a person with CF. The findings point to a potential treatment strategy for people with nonsense mutations, who generally do not benefit from existing CFTR-targeted therapies.

The study, “Nonviral delivery of chemically modified tRNA rescues nonsense mutations in cystic fibrosis,” was published in Science.

CF is a genetic disease caused by mutations in the CFTR gene, which provides instructions for making the CFTR protein. CFTR normally helps control the movement of salt and water into and out of cells. When the protein is missing or doesn’t work properly, thick, sticky mucus builds up in the lungs and other organs, leading to CF symptoms.

CFTR modulators, such as those included in Trikafta, have transformed treatment of CF by improving the function of faulty CFTR. But they are not effective for the roughly one in 10 patients whose disease stems from a nonsense mutation.

Such mutations introduce a stop signal on messenger RNA (mRNA), which carries the genetic instructions used to make a protein. Normally, tRNAs recognize specific three-letter sequences on mRNA and bring the corresponding amino acids — the building blocks of proteins — to the growing protein. But a premature stop signal tells the protein-making machinery to stop too soon, leaving little or no full-length functional CFTR for modulators to act on.

“Think of tRNA as a little car,” Bowen Li, PhD, an associate professor at the University of Toronto and study lead, said in a news story from the University. “A nonsense mutation is like putting a stop sign in the middle of the road. The car has to slam on the brakes, and the protein never gets finished.”

Researchers engineer tRNA to bypass faulty stop signals

A team led by scientists at the University of Toronto sought to overcome that problem by engineering so-called suppressor tRNAs to recognize a premature stop signal and insert the intended amino acid in its place. This allows the protein-making machinery to move past the incorrect signal and continue building a full-length protein.

Turning that idea into a potential therapy presented two major challenges.

First, engineered tRNAs needed to be potent enough to restore meaningful amounts of protein. The team added a specific chemical modification found naturally in tRNAs, making the engineered molecules more active and longer-lasting.

The second challenge was getting the tRNA to disease-relevant tissues. The researchers therefore developed lipid nanoparticles specifically designed to deliver tRNA. After creating and screening about 1,000 different lipids, they identified a candidate for delivering the engineered tRNA to the lungs.

The resulting approach did more than bring CFTR protein back — it also restored the protein’s function.

In lab-grown human airway cell models involving two common nonsense mutations, the tRNA restored CFTR production and function, and the restored protein remained for more than 40 days, according to the researchers. The researchers also saw evidence of restored CFTR production and function in mouse models of the disease.

Patient-derived organoids respond to tRNA plus Trikafta

They then tested the approach in organoids — miniature, lab-grown models of organs or tissues — made from cells donated by a person with CF who had four CFTR mutations, including two nonsense mutations, and had not responded to existing treatments.

Neither the modified tRNA nor Trikafta produced much of a response when used alone. When they were combined, however, the patient’s cells responded. The tRNA enabled the cells to produce full-length CFTR, essentially providing protein for Trikafta to act on.

“That was a great moment for us, where we saw the potential of the therapy,” said Jingan Chen, a PhD candidate at the University of Toronto and study co-lead author.

More studies will be needed before the approach can be tested as a treatment for people with CF. Still, the findings support the therapeutic potential of engineered tRNAs, which the researchers believe could eventually have applications beyond CF.

Nonsense mutations can occur in many different genes, but they produce only three possible premature stop signals. That raises the possibility that one engineered tRNA designed to bypass a particular stop signal could potentially be used across different genetic diseases.

“Our long-term goal is to develop tRNA medicines that recognize these shared stop signals, so that one therapeutic strategy could potentially be applied across many different genetic diseases,” Li said.