An arm injection that makes the heart produce its own repair drug lasted four weeks in animals

After a heart attack, cardiologists can reopen blocked arteries and restore blood flow within hours. But the muscle cells that died during the blockage are gone permanently. The heart is one of the body's least regenerative organs, and the scar tissue that replaces dead muscle does not contract, does not conduct electrical signals, and progressively weakens the heart's ability to pump blood.

A study published March 5 in Science by Ke Cheng and colleagues at Columbia Engineering describes a therapy designed to change that equation - not by delivering drugs to the heart directly, but by turning the patient's own arm muscle into a pharmaceutical factory.

What newborn hearts can do that adult hearts cannot

In the first days of life, many mammals have a brief window during which their hearts can regenerate muscle cells. A hormone called atrial natriuretic peptide (ANP) plays a central role, encouraging new blood vessel growth, calming inflammation, and reducing scar formation. As the organism matures, ANP production declines substantially, and the regenerative capacity disappears.

Cheng's team confirmed this experimentally. After a heart attack in newborn mice, the gene that produces ANP's precursor - called Nppa - ramped up more than 25 times its normal level. In adult mice, it rose only about 10-fold. When the researchers knocked out Nppa in newborn mice, those hearts lost much of their ability to heal.

The insight was straightforward: if adult hearts cannot make enough ANP on their own, supplement it. But ANP breaks down within minutes in the body, making conventional drug delivery impractical. And the heart lacks the natural accumulation mechanisms that make liver- or lung-targeted therapies relatively straightforward.

A two-step delivery trick

The team's solution was to stop trying to deliver the drug to the heart altogether. Instead, they developed a two-phase approach.

First, RNA-lipid nanoparticles encoding the Nppa gene are injected into skeletal muscle - an arm or thigh. The muscle cells read the RNA instructions and produce a precursor molecule called pro-ANP. This precursor is biologically inert and circulates harmlessly through the entire bloodstream.

Second, a specific enzyme called Corin converts pro-ANP into active ANP. Corin is roughly 60 times more abundant in the heart than in any other organ. So the inactive precursor circulates everywhere, but is activated almost exclusively in the one organ that needs it.

The targeting is elegant: no direct cardiac injection, no catheterization, no chest surgery. The heart's own enzymatic machinery selects it as the activation site.

Self-amplifying RNA extends the effect

Standard mRNA therapies produce protein for a limited time before the RNA degrades. To extend the therapeutic window, Cheng's team used self-amplifying RNA (saRNA), which replicates itself inside cells. A single injection sustained pro-ANP production for at least four weeks in animal models - long enough to cover the critical post-heart-attack healing period without requiring repeated hospital visits.

Testing across diverse conditions

Before a therapy can move toward human trials, it needs to demonstrate efficacy beyond young, healthy lab mice. Cheng's team deliberately ran the experiment in aged mice, in animals genetically prone to atherosclerosis, in mice with diet-induced type 2 diabetes, and in large animals. They also tested delayed treatment - waiting a full week after the heart attack before administering the injection, by which point significant damage had already occurred.

The therapy worked consistently across all conditions. Scarring was significantly reduced and heart function improved in every model tested. The delayed-treatment results are particularly relevant clinically, since many heart attack patients do not receive specialized care immediately.

From heart to other organs

The RNA-to-prodrug-to-organ-specific-activation strategy is not limited to the heart. Torsten Vahl, a co-author and cardiologist at Columbia University Irving Medical Center, noted that cell damage affects many organs. If the approach can regenerate cardiac cells in a clinical setting, the same principle could potentially be adapted for kidney disease, high blood pressure, and preeclampsia - conditions where ANP or similar targeted molecules might offer benefit.

What stands between the lab and the clinic

These are animal results. The saRNA platform has not been tested in humans for this application. Safety questions - including immune responses to self-amplifying RNA, off-target Corin activity in non-cardiac tissues, and long-term effects of sustained ANP production - require formal toxicology and phase-one clinical studies.

Manufacturing saRNA nanoparticles at clinical scale is also nontrivial, though Cheng notes that Columbia's Initiative in Cell Engineering and Therapy has in-house manufacturing capabilities. He hopes to conduct a phase-one safety trial at Columbia University Irving Medical Center.

The broader significance is the delivery paradigm itself. Using one tissue as a factory to produce a prodrug that is activated by a different tissue's unique enzymes is a generalizable strategy. If it works for ANP and Corin, the same logic could apply to other organ-specific enzymes and their corresponding therapeutic molecules - a modular approach to precision drug delivery that bypasses the hardest targeting problems in medicine.