AI helps Stanford scientists discover a “natural Ozempic” with a surprising advantage

The molecule, known as BRP, works differently from existing treatments. Rather than affecting several organs throughout the body, it appears to act more selectively on a part of the brain responsible for controlling hunger and metabolism. This more focused approach could eventually help people lose weight without experiencing problems such as nausea, constipation, or substantial muscle loss.

Researchers discovered BRP with the help of artificial intelligence, which allowed them to search through thousands of possible biological compounds. The findings were promising enough that the scientists began preparing for the next major step: determining whether the molecule is safe and effective in humans.

“The receptors targeted by semaglutide are found in the brain but also in the gut, pancreas and other tissues,” said assistant professor of pathology Katrin Svensson, PhD. “That’s why Ozempic has widespread effects including slowing the movement of food through the digestive tract and lowering blood sugar levels. In contrast, BRP appears to act specifically in the hypothalamus, which controls appetite and metabolism.”

A Different Approach to Weight Loss

The hypothalamus is a small but essential region deep within the brain. Among its many responsibilities, it helps regulate hunger, energy use, body temperature, and other functions needed to maintain the body’s internal balance. Because it plays such a central role in appetite, it is an important target for researchers developing obesity treatments.

Semaglutide works by imitating a hormone called glucagon-like peptide 1, or GLP-1. This hormone helps regulate blood sugar and signals that influence hunger and feelings of fullness. However, GLP-1 receptors are distributed throughout several parts of the body, which helps explain why medications that activate them can produce effects beyond appetite suppression.

BRP appears to take another route. Although it influences some of the same broad metabolic functions as semaglutide, it activates different populations of brain cells. Scientists believe this distinction could make it possible to control hunger more precisely while reducing unwanted effects elsewhere in the body.

Svensson, who served as senior author of the study, published in Nature, has also cofounded a company to advance BRP toward human clinical trials. The research was led by senior research scientist Laetitia Coassolo, PhD.

How Artificial Intelligence Uncovered a Hidden Molecule

Finding BRP required solving a problem that has long complicated the search for new hormones. The human body produces enormous numbers of small protein fragments, but only a fraction have important biological functions. Identifying the ones that influence appetite or metabolism can be extremely difficult.

Many hormones begin as larger, inactive proteins known as prohormones. Specialized enzymes cut these proteins into smaller pieces called peptides. Some of those peptides become chemical messengers that help coordinate processes such as energy metabolism, appetite, and organ function.

A single prohormone can be cut in several different places, potentially producing numerous peptides with distinct effects. The challenge is determining which fragments are biologically useful and which are simply leftovers from normal protein processing. The important peptide hormones (which are relatively rare) can be difficult to distinguish from the much larger collection of ordinary protein fragments.

Traditional approaches often involve extracting proteins from biological tissues and analyzing them with techniques such as mass spectrometry. This method identifies molecules by measuring characteristics related to their mass, but the sheer number of possible peptides makes searching for new hormones a demanding process.

The Stanford team decided to approach the problem differently. They concentrated on an enzyme called prohormone convertase 1/3, which cuts prohormones at particular amino acid sequences and has previously been linked to human obesity. Amino acids are the molecular building blocks of proteins, and their arrangement helps determine where enzymes can cut them.

One of the hormones produced through this process is GLP-1, the same hormone whose effects semaglutide mimics. The researchers reasoned that other peptides generated through similar processing might also help regulate appetite and energy use.

AI Searches 20,000 Human Genes for New Hormones

To accelerate the search, the scientists created a computer algorithm called Peptide Predictor. Instead of physically isolating and examining hundreds of thousands of peptides, the program analyzed the roughly 20,000 genes that encode human proteins, looking for sequences where the enzyme was likely to make cuts.

The team then refined the results by concentrating on genes associated with proteins released outside cells, a common characteristic of hormones. They also required each candidate protein to contain at least four potential cutting sites.

These filters reduced an enormous collection of possibilities to 373 prohormones, making experimental testing much more practical.

“The algorithm was absolutely key to our findings,” Svensson said.

Peptide Predictor estimated that the enzyme could produce 2,683 distinct peptides from those 373 proteins. From this collection, Svensson and Coassolo selected candidates that appeared particularly likely to influence brain activity.

The researchers eventually tested 100 peptides, including GLP-1, using nerve cells grown in the laboratory. Their goal was to find substances capable of stimulating neuronal activity, an early indication that a peptide might influence important brain processes.

A Tiny Peptide Produces a Surprising Result

GLP-1 performed strongly in the laboratory tests, increasing neuronal activity to three times the level observed in untreated control cells.

One previously unrecognized peptide, however, produced an even more striking response. Consisting of just 12 amino acids, it increased neuronal activity tenfold compared with the controls.

The researchers named the molecule BRP after the larger protein from which it originates, BPM/retinoic acid inducible neural specific 2, or BRINP2 (BRINP2-related-peptide).

That unusually powerful response made BRP a promising candidate for further investigation. The next question was whether stimulating laboratory-grown brain cells would translate into meaningful changes in appetite and body weight in living animals.

Food Intake Drops by Up to 50% in Animal Experiments

The team tested BRP in lean mice and minipigs (which more closely mirror human metabolism and eating patterns than mice do). Before the animals were fed, researchers administered BRP through an injection into the muscle.

The results were striking. During the following hour, the treated animals consumed up to 50% less food than expected, suggesting that BRP could strongly influence hunger.

The scientists then examined whether repeated treatment could produce sustained changes in body weight. Obese mice received daily BRP injections for 14 days. By the end of the experiment, the treated mice had lost an average of 3 grams, while untreated control mice had gained approximately 3 grams.

Importantly, nearly all the weight lost by the BRP-treated mice came from body fat rather than muscle. Preserving muscle is an important goal in obesity treatment because muscle tissue supports strength, mobility, and metabolic health.

The treated animals also showed improvements in glucose and insulin tolerance. These measurements reflect how effectively the body handles sugar and responds to insulin, the hormone that helps move glucose from the bloodstream into cells.

Together, the findings suggest that BRP may influence both appetite and metabolic health, although it remains unknown whether people would experience comparable benefits.

Could BRP Avoid Common Ozempic Side Effects?

One of the most intriguing aspects of the discovery is the possibility that BRP could provide weight loss benefits while avoiding some unwanted effects associated with semaglutide and similar medications.

To investigate potential side effects, the researchers monitored the behavior and physical responses of the treated mice and pigs. They found no significant differences in movement, water consumption, anxiety-related behavior, or fecal output compared with control animals.

The absence of changes in fecal production was particularly interesting because constipation is a known side effect of medications such as Ozempic. The experiments also showed that BRP promoted fat loss without the substantial muscle loss that can accompany some weight reduction treatments.

Nausea is another common concern with existing GLP-1 medications, although animal experiments cannot fully establish whether BRP would prevent that symptom in people. Human testing will be essential to determine the molecule’s actual side effect profile.

Additional investigations into brain activity and other physiological responses reinforced the idea that BRP works differently from semaglutide. The peptide activated distinct neuronal and metabolic pathways rather than simply copying the effects of GLP-1.

This difference may explain why the researchers observed strong appetite suppression without several of the behavioral and digestive changes they were monitoring.

What Comes Next for the Potential Ozempic Alternative?

Despite the encouraging findings, BRP is still at an early stage of development. Researchers need to understand more about exactly how the molecule communicates with cells and how its effects might be sustained over longer periods.

One priority is identifying the cell surface receptors that recognize BRP. Receptors are specialized proteins that respond to particular chemical signals, allowing cells to detect messages and trigger appropriate biological responses. Determining which receptors BRP activates could help explain its effects on hunger and metabolism.

The team also wants to investigate ways of extending the molecule’s activity inside the body. Peptides can be broken down relatively quickly, potentially requiring frequent injections. A longer-lasting version of BRP could make treatment more convenient if it eventually proves effective in people.

The ultimate challenge will be establishing whether its apparent advantages in animals translate into meaningful benefits for patients. Clinical trials will need to evaluate its safety, effectiveness, appropriate dosage, and potential side effects before it could become a practical obesity treatment.

“The lack of effective drugs to treat obesity in humans has been a problem for decades,” Svensson said. “Nothing we’ve tested before has compared to semaglutide’s ability to decrease appetite and body weight. We are very eager to learn if it is safe and effective in humans.”

Research Collaborators and Funding

Scientists from the University of California, Berkeley; the University of Minnesota; and the University of British Columbia also participated in the research.

Financial support came from the National Institutes of Health (grants R01DK125260, P30DK116074, K99AR081618 and GM113854), the SPARK Translational Research Program at Stanford, Stanford Bio-X, the Stanford Maternal and Child Health Research Institute, the American Heart Association, a Stanford Medicine Dean’s Fellowship Award, the Carlsberg Foundation, and the Wu Tsai Human Performance Alliance.

Svensson and Coassolo are named inventors on patents involving the use of BRP peptides to treat metabolic disorders. Svensson is also a cofounder of Merrifield Therapeutics.