Semaglutide is a GLP-1 receptor agonist sold as Ozempic and Wegovy and used in the treatment of type 2 diabetes and obesity. Three recent pieces of research approach it from different directions: one on how it acts in the brains of mice, one on pairing it with an experimental metabolic compound in mice, and one on laboratory redesign of its peptide structure. None of the three is a human trial.

Brain circuits in mice

A Yale team reported in PNAS that agouti-related peptide (AgRP) neurons, long associated with stimulating hunger, were activated rather than suppressed during semaglutide treatment in mice. In animals genetically engineered to lack AgRP neurons, the researchers said GLP-1 drugs were no longer able to sustain weight loss.

The authors describe this as a reversal of a widely discussed explanation for how the drugs work. The experiments were conducted in mice, and the researchers say further work is needed to determine whether the same mechanism operates in humans.

Combination with an experimental compound

UC Berkeley researchers reported in Science Advances that an old compound called TOFA increased energy use by as much as 18% in obese mice and reduced fat without a significant reduction in lean muscle mass. When TOFA was combined with semaglutide or tirzepatide, the team reported larger improvements in body weight, glucose control, insulin levels and triglycerides than either treatment produced alone.

Senior author Anders Naar said the team views TOFA as complementary rather than a replacement. The researchers emphasize that TOFA has so far been studied only in animals and that its safety and effectiveness in humans remain unknown.

Redesigning the peptide itself

Separately, chemists described building 108 stapled peptide candidates from a short semaglutide-derived GLP-1R-engaging segment, of which 35 were synthesized and characterized. Most of the stapled analogues showed improved serum and proteolytic stability relative to semaglutide, the authors reported, with three candidates showing the most favorable stability profiles.

The work is chemistry and modeling. The authors present it as a basis for subsequent functional optimization, not as evidence of any effect in an animal or a person.

What Isn't Established

None of these three reports involved human participants. The Yale mechanism was observed in mice, and the paper title specifies female mice. The TOFA combination results were also in mice, with human safety and efficacy explicitly described as unknown. The stapled peptide analogues were assessed for stability and predicted receptor interactions, not for glucose control, weight or safety.

State of the evidence

Semaglutide itself is an approved and widely used medication, but this particular cluster of research sits upstream of the clinic. Each of the three groups frames its finding as a starting point: a mechanism to test in humans, a compound that has not entered human testing, and a scaffold awaiting functional optimization.