RNA trans-splicing treatment for Duchenne muscular dystrophy

Muscle is a tissue built on a delicate balance of strength and repair. In a healthy body, a massive protein called dystrophin acts as a crucial shock absorber, linking the internal scaffolding of a muscle fiber to its outer membrane. This connection allows muscles to contract and relax without tearing themselves apart. When the gene that provides the instructions for making this protein is broken, the link fails. The muscle fibers begin to rip with every movement, leading to a relentless cycle of damage, scarring, and weakness. This is the reality for people with Duchenne muscular dystrophy, a severe condition that primarily affects boys and leads to the loss of walking ability and, eventually, heart and lung failure. For decades, scientists have tried to fix this by delivering a working copy of the gene using a harmless virus as a delivery truck. However, the virus has a strict size limit; it can only carry a small package. The instructions for making the full-length dystrophin protein are far too large to fit inside a single viral truck, forcing researchers to either send a shortened, less effective version of the protein or try to splice the instructions together inside the body, a process that has often been messy or inefficient.

A team of researchers at the Salk Institute has developed a new way to overcome this size limit, offering a potential path to delivering the full, uncut instructions for making healthy dystrophin. Instead of trying to force a giant package into a small truck, they split the instructions into three smaller pieces and designed a system that allows the cell's own internal machinery to stitch them back together perfectly. They call this system RNA End-Joining, or REJ. In this approach, the genetic instructions are divided into separate segments, each carried by its own viral vector. Once inside a muscle cell, these segments are transcribed into RNA, a temporary copy of the genetic code. The researchers added special molecular tags to the ends of these RNA pieces, acting like a zipper that encourages them to find each other. When the cell's natural splicing machinery encounters these tagged pieces, it recognizes them as a single continuous message and joins them together, creating a complete, flawless set of instructions for making the full-length protein. This method avoids the need for foreign enzymes or bacterial proteins that can trigger immune reactions, relying entirely on the body's own tools to assemble the cure.

To test if this system works in a living animal, the researchers turned to mice that carry the same genetic defect found in humans with Duchenne muscular dystrophy. They designed three different versions of their treatment to see how well the system could handle various therapeutic goals. In one experiment, they used the system to deliver a gene-editing tool called Abe8e, which is designed to correct the broken gene directly. In another, they used it to deliver a slightly shorter but still highly functional version of dystrophin known as Dp253. Finally, they pushed the system to its limit by using three viral vectors to deliver the instructions for the full-length, native dystrophin protein, which is the gold standard for treatment but has never before been deliverable by a single viral vector. The results were striking. In the mice treated with the gene editor, the system successfully restored the production of the full-length protein in a significant portion of muscle fibers. In the mice receiving the Dp253 replacement, the treatment restored protein levels to nearly 72 percent of what is seen in healthy mice. Most impressively, the triple-vector system successfully delivered the full-length protein, achieving levels of about 11.6 percent of normal, which is a substantial amount for such a difficult delivery task.

The researchers did not stop at simply measuring how much protein was made; they looked closely at how the muscle itself changed. Using advanced computer vision and machine learning, they analyzed more than 120,000 individual muscle fibers from treated and untreated mice. This high-tech approach allowed them to see details that human eyes might miss, such as the precise location of cell nuclei and the presence of inflammatory cells. In untreated mice, the muscle fibers were often misshapen, with nuclei floating in the center of the cell rather than tucked against the edge, and the tissue was riddled with signs of inflammation and damage. In the treated mice, these signs of disease vanished. The muscle fibers returned to a healthy shape, the nuclei moved back to their proper positions, and the inflammation subsided. The treated muscles also became stronger, generating significantly more force when stimulated. The mice that received the treatment were able to run much farther on exercise wheels, showing a return of voluntary activity that mirrored the behavior of healthy animals.

One of the most surprising discoveries came from looking at how the treated and untreated fibers interacted with one another. In the mice treated with the gene editor, not every single muscle fiber received the correction; the result was a mosaic of healthy, corrected fibers mixed with uncorrected ones. The researchers found that the uncorrected fibers were healthier when they were surrounded by corrected neighbors. This suggests that a healthy muscle fiber can provide a protective buffer for its damaged neighbors, helping to stabilize the entire tissue even if the repair is not perfect in every single cell. This finding helps explain why some female carriers of the disease, who naturally have a mix of healthy and defective muscle fibers, often remain asymptomatic. The study also confirmed that the system was highly specific, with no evidence of the RNA pieces accidentally sticking to the wrong genetic targets, a safety concern that has plagued other gene therapy approaches.

The researchers compared their new method to existing strategies that try to deliver large genes. Other approaches often rely on bacterial enzymes to cut and paste DNA or RNA, which can leave behind unwanted molecular scars or trigger the immune system. The REJ system, by contrast, uses the cell's own splicing machinery to create a seamless, scar-free protein. While the system requires a higher dose of virus to ensure that enough cells receive all the necessary pieces, the researchers calculated that this dose remains within safe limits for clinical use. The study demonstrates that it is possible to deliver massive therapeutic proteins that were previously thought to be too large for viral delivery. By successfully delivering the full-length dystrophin protein and seeing robust improvements in muscle health and function, the team has shown that this platform could be a viable path forward for treating not just Duchenne muscular dystrophy, but potentially other genetic diseases caused by mutations in very large, complex proteins. The work moves the field closer to a treatment that restores the complete, natural function of the muscle, rather than just a partial fix.