For decades, obesity medications produced only modest weight loss. But in recent years, Ozempic and related GLP-1 therapies have transformed the field, enabling sustained weight loss of 10 to 15% or more. Despite their success, however, scientists have not fully understood how these drugs work in the brain.

Now, in a new study, Yale researchers have identified an unexpected mechanism of action that challenges a long-held assumption about the brain’s hunger circuitry: that agouti-related peptide (AgRP) neurons, known as drivers of hunger, functioned solely to oppose weight loss. However, the new study shows that GLP-1 therapies like Ozempic instead recruit these neurons to help sustain fat loss.

“This completely changes how we think about the mechanism involved in these medications and provides new insight into the biology underlying their long-term effects, opening an avenue for the development of more efficient drugs,” said Mateus d’Ávila, a Ph.D. candidate in neuroscience working in Tamas Horvath’s lab in the Department of Comparative Medicine at Yale School of Medicine (YSM) and first author of the study.

The study appears in the journal Proceedings of the National Academy of Sciences (PNAS).

Semaglutide, the active ingredient in GLP-1 drugs like Ozempic, has become one of the most effective medications ever developed for obesity. Yet we still don’t fully understand why it works so well. Previous generations of weight-loss drugs suppress appetite almost as effectively as semaglutide, but none produce the same degree of sustained weight loss.

For Yale researchers, that suggested semaglutide was doing something beyond simply reducing appetite. Before their study, one prevailing view was that GLP-1 drugs promote weight loss by reducing the activity of neurons that drive hunger. Although this hypothesis has been widely discussed, the role of AgRP neurons in weight loss during chronic GLP-1 treatment had not been directly tested in vivo.

In the new study, the researchers wanted to uncover that missing biology. By understanding how the brain adapts to treatment, they hoped to reveal new therapeutic targets that could eventually lead to even better obesity medications. For the study, they combined several complementary approaches in a mouse model and monitored body weight, food intake, metabolism, and energy expenditure during semaglutide treatment. They also used genetic methods that allowed them to selectively remove or silence AgRP hunger neurons, enabling them to determine whether those neurons were necessary for the drug’s effects.

When treating mice genetically modified to lack the AgRP neurons, the researchers observed that GLP-1 drugs could no longer sustain weight loss. Further experiments using electron microscopy, molecular biology, and electrophysiology showed that these AgRP neurons were being activated rather than inhibited by semaglutide.

These findings, researchers say, suggest that the brain adapts to the calorie deficit created by GLP-1 treatment by increasing the activity of these AgRP hunger neurons, which also coordinate loss of fat. This reveals a previously unrecognized layer of complexity in how GLP-1 therapies work.

Because the study was performed in mice, additional research is needed before these findings can be translated to people. However, understanding exactly how these medications work in the brain is an important step toward developing future obesity treatments.

“By identifying a previously unrecognized neural mechanism involved in sustaining weight loss, our work provides new biological insights that could eventually help researchers design therapies that are even more effective or have fewer side effects,” d’Ávila said.

Other authors from YSM include Roberto Collado-Pérez, a postdoctoral associate; Zhong-Wu Liu, assistant professor adjunct; Joseph Schlessinger, the William H. Prusoff Professor of Pharmacology; and Horvath, the Jean and David W. Wallace Professor of Comparative Medicine.