What is Psyncretin: the AI-designed weight-loss molecule your muscles would manufacture

Ozempic proved the point. Wegovy, Zepbound, and Mounjaro drove it home: put the body’s own metabolic hormones to work, and the fat comes off (and stays off) in a way decades of diet drugs never managed. Semaglutide and tirzepatide were built for diabetes and ended up rewriting weight-loss medicine. So the obvious question is what comes next. One early answer has an awkward name and an unusual origin story: psyncretin.

Psyncretin, stylized pSynCretin, is short for plasmid-encoded synthetic consensus incretin. It is not a molecule anyone found in nature. It was designed. A team at the Weiner Lab at Philadelphia’s Wistar Institute used AI to build it and described the work in Trends in Biotechnology in June 2026. Like tirzepatide, it hits two gut-hormone receptors at once: GLP-1 and GIP.

Why bother with two receptors?

GLP-1 is the workhorse. Drugs that mimic it dial down appetite, slow how fast the stomach empties, and sharpen the body’s insulin response. GIP is the more surprising partner. On its own it looked like a dud for years. Pair it with GLP-1, though, and the combination appears to do more for both weight and blood sugar than either does alone. Tirzepatide was the first drug to exploit that; a “twincretin,” in the trade’s slightly unfortunate coinage. And it tended to outperform the GLP-1-only crowd in trials. Psyncretin plays the same two-receptor game.

Designed, not discovered

Most hormone drugs are copies, or near-copies, of something the body already makes. Psyncretin’s designers went the other way. Using AI-guided structural modeling and an approach called synthetic consensus design, they pulled out the structural features that GLP-1, GIP, and today’s incretin drugs share, then folded those features into a single engineered protein meant to grab both receptors. It worked, at least at the bench: the molecule stuck to the GLP-1 receptor more tightly than tirzepatide does.

Your muscles as a drug factory

But the molecule isn’t even the boldest part. Delivery is.

Today’s GLP-1 therapies mean a needle every week. Psyncretin skips that. It arrives as plasmid DNA, a small loop of genetic instructions injected once into muscle and nudged into the cells with a brief electrical pulse. From there, the cells make the drug themselves, for weeks on end. The team also clipped an antibody fragment onto the molecule to keep the body from chewing through it too quickly, a chronic headache with peptide drugs.

In mice, the numbers were striking. A single dose kept incretin levels detectable for up to 70 days, with weight and blood glucose staying down the whole time. Head-to-head against semaglutide, the DNA-treated mice held onto their weight loss after the study wrapped; the semaglutide mice started creeping back up the moment the drug stopped. All told, the effect ran up to ten times longer than a standard incretin drug. (We walked through the mechanics of this “muscle cell as GLP-1 factory” approach separately, in “A single shot turns muscle cells into a months-long GLP-1 factory.”)

The bigger picture

Psyncretin sits at the crossroads of three things happening at once in obesity research:

• More receptors. Triple agonists are already in the works, adding glucagon to the mix to burn more energy while curbing appetite.
• Smarter design. AI lets researchers build molecules for human physiology from scratch, rather than reverse-engineering nature. Potentially a faster path to new drugs.
• Longer-lasting delivery. Oral peptides, gene-based methods, implants: all chasing the same goal of dosing patients far less often.

A large caveat

None of this is a drug yet. Psyncretin is early-stage research, and the results are from mice. Approved GLP-1 drugs come with years of human trial data behind them; psyncretin comes with none. Safety, durability, manufacturing, how it behaves in an actual human body are all open questions. The graveyard between a promising mouse study and a clinic-ready therapy is enormous, and most candidates never make the crossing.

Still, the direction is hard to miss. GLP-1 drugs settled the argument that metabolic hormones are worth targeting. Psyncretin sketches what the next round might look like: several hormone signals hit at once, molecules drawn up by AI, and a single shot that could keep working for months.

Sources

• Wistar Institute press release, Single-Dose DNA Method for Delivering Long-Acting Weight Loss and Diabetes Drugs (2026): wistar.org
• Gary et al., Engineering single-dose plasmid DNA for sustained in vivo delivery of designer incretins, Trends in Biotechnology (2026): cell.com
• Nauck & Meier, review of dual GIP/GLP-1 receptor agonism (twincretins): PMC