PNAS study finds semaglutide needs the brain's hunger neurons to keep fat off

Semaglutide, the active ingredient in Ozempic and Wegovy, may not work in the brain the way scientists long assumed, according to a new study from Yale researchers published August 4 in the Proceedings of the National Academy of Sciences ^[2]. The study, done in mice, found that agouti-related peptide (AgRP) neurons, brain cells long known for driving hunger, are not shut down by the drug. Instead, they appear to be activated and are required for the drug's fat-loss effects to last ^[1].

Researchers had generally believed GLP-1 drugs caused weight loss mainly by suppressing the activity of hunger-promoting neurons like AgRP cells ^[1]. But this idea had not been directly tested in animals given long-term GLP-1 treatment. The Yale team, led by scientists in Tamas Horvath's lab, gave mice semaglutide while tracking body weight, food intake, metabolism and energy use. They also used genetic tools to eliminate or silence AgRP neurons in some animals ^[1].

Mice that lacked these neurons kept eating less on semaglutide, but they did not sustain fat loss the way mice with intact AgRP neurons did ^{[1] [2]}. Additional lab work using electron microscopy, molecular biology and electrophysiology showed the neurons were activated by the drug rather than suppressed, and that they also help coordinate the body's mobilization of fat ^[1]. Disrupting a related fat-mobilization signaling pathway also weakened sustained fat loss, according to the summary of the findings ^[2].

The researchers caution that the effect depended on sex, diet and the specific method used to disrupt the neurons, and the study was conducted only in female mice ^[2]. Mateus d'Ávila, the study's first author and a Ph.D. candidate at Yale School of Medicine, said the discovery "completely changes how we think about the mechanism involved in these medications" ^[1]. He added that identifying this previously unrecognized pathway could help researchers eventually design obesity therapies that are more effective or carry fewer side effects ^{[1] [2]}.

Why it matters for patients

For people currently taking semaglutide or considering it, this study does not change how the drug is prescribed or used today. It is a mouse study, and the researchers themselves say more work is needed to determine whether the same brain mechanism operates in humans ^{[1] [2]}.

What it does change is scientists' basic understanding of why these drugs produce such durable weight loss compared with older obesity medications, which could suppress appetite about as well but did not sustain weight loss to the same degree ^[1]. Michael Russo, a bariatric surgeon quoted in the Forbes Health account of the study, put it bluntly: "These drugs are very mysterious" and "we don't fully understand how GLP-1s affect weight loss in the brain" — even though their effectiveness is well established ^[2].

That gap in understanding matters because it could shape future drug development. If hunger neurons are actually being recruited to help mobilize fat rather than simply turned off, that could point toward new drug targets aimed at that fat-mobilization pathway specifically ^[1]. Whether that leads to different side-effect profiles or different levels of effectiveness in people is not yet known.

What happens next

The study appeared in PNAS on August 4, 2026, and was covered by Yale's own news office and by outside outlets in mid-August ^{[1] [2]}. The authors say follow-up research is needed to see whether the AgRP neuron mechanism identified in female mice also applies to male mice and, eventually, to humans ^{[1] [2]}. No timeline for human studies was given in the available sources. Separately, other obesity drugs in development, including Eli Lilly's triple-hormone agonist retatrutide, continue to move toward potential FDA filing, which the company has said it expects in early 2027, though that drug is unrelated to this specific brain-mechanism finding ^[2].