A newly published study has highlighted how artificial intelligence and robotics-driven computational methods are transforming drug discovery, leading to the identification of a promising anti-obesity peptide.
Study: Prohormone cleavage prediction uncovers a non-incretin anti-obesity peptide. Image Credit: Alones/Shutterstock.com
The researchers used AI-powered analysis and automated systems to uncover BRINP2-related peptide (BRP), a 12-mer peptide that effectively reduces appetite and weight gain in mice and pigs without causing nausea. Unlike existing treatments that target pathways like glucagon-like peptide 1 (GLP-1), BRP operates through an independent mechanism, offering a novel approach to weight management.
Background
Peptide hormones play a vital role in regulating energy balance and have long been explored as potential treatments for obesity. Since the discovery of neuropeptide Y, 12 additional peptides involved in feeding regulation have been identified. However, uncovering new bioactive peptides remains challenging due to their low abundance and difficulty in distinguishing them from degradation byproducts. Traditionally, these peptides were identified from endocrine organs, but systematic efforts to discover novel peptides with therapeutic potential have been limited.
PCSK1 is a key enzyme in peptide processing, with genetic variants linked to obesity in both humans and animals. While peptides like GLP-1 have demonstrated therapeutic potential, the full spectrum of PCSK1-generated peptides remains largely unexplored.
This study aimed to address that gap by developing a computational approach, Peptide Predictor, to systematically identify evolutionarily conserved peptides. Using this method, researchers discovered BRP, a 12-mer peptide with strong appetite-suppressing effects in mice and pigs, independent of GLP-1 signaling. The study underscores the potential of computational techniques in identifying novel bioactive peptides for weight regulation.
Peptide Predictor: A Computational Approach
The research team employed advanced computational and pattern recognition techniques to systematically identify and characterize novel bioactive peptides derived from prohormone processing. Peptide Predictor, a custom algorithm using regular expressions (RegEx), was developed to predict cleavage sites in secreted human proteins. The algorithm analyzed canonical sequences from UniProtKB, identifying 373 proteins with more than four cleavage sites and generating 2683 novel peptides.
To assess bioactivity, the researchers used PeptideRanker and MultiPep classifiers, scoring peptides based on their potential biological activity. Tissue-specific expression of precursor proteins was mapped using data from the Human Protein Atlas, with a circos plot illustrating peptide distribution across different tissues. A synthetic peptide library of 100 peptides was created, prioritizing metabolic tissues and previously uncharacterized functions.
Structural predictions of BRP using AlphaFold provided insights into its molecular conformation, while in vitro studies confirmed BRP’s sequence-specific bioactivity in neuronal and β-cell lines. This research highlights the power of integrating computational tools, machine learning classifiers, and structural prediction methods to uncover novel bioactive peptides with therapeutic applications.
BRP: A Novel Peptide for Appetite Suppression
The study identified BRP, a 12-mer peptide derived from BRINP2, as a potent appetite suppressant with significant anti-obesity effects. By systematically mapping proteolytically processed peptides in the human genome, the researchers pinpointed BRP as a bioactive peptide with strong therapeutic potential.
In vivo studies demonstrated that BRP significantly reduced food intake in both lean and obese mice, producing effects comparable to GLP-1 but without inducing nausea or aversive reactions. BRP’s mechanism of action involves activating the cAMP–PKA–CREB–FOS pathway in hypothalamic neurons, independent of GLP-1, leptin, and MC4R signaling. Pharmacokinetic analysis showed that BRP is rapidly cleared, with peak plasma concentrations occurring within a minute.
Crucially, BRP also reversed obesity in diet-induced obese mice, reducing fat mass and improving glucose tolerance. Its effects were further validated in minipigs, reinforcing its therapeutic potential. Structural analysis pinpointed leucine at position 8 (L8) as a key determinant of BRP’s bioactivity—mutations at this site eliminated its effects. These findings establish BRP as a novel, centrally acting peptide with significant promise for obesity treatment. Ongoing research will focus on identifying BRP’s receptor and exploring its broader physiological functions.
Conclusion
This study highlights the potential of AI and robotics-driven computational methods in bioactive peptide discovery, demonstrated by the identification of BRP using Peptide Predictor. By integrating pattern recognition, machine learning classifiers (PeptideRanker and MultiPep), and structural prediction tools like AlphaFold, researchers systematically mapped over 2600 peptides, ultimately identifying BRP as a potent appetite suppressant.
BRP was found to reduce food intake and reverse obesity in mice and pigs without inducing nausea, acting independently of known pathways. These findings underscore the effectiveness of computational approaches in uncovering new therapeutic peptides for metabolic disorders. Further research will be crucial to fully understand BRP’s mechanisms and its potential applications in weight management treatments.
Journal Reference
Coassolo et al. (2025). Prohormone cleavage prediction uncovers a non-incretin anti-obesity peptide. Nature. DOI:10.1038/s41586-025-08683-y https://www.nature.com/articles/s41586-025-08683-y
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.