Structural Biophysics-Guided Computational Design of Semaglutide Analogues to Enhance GLP-1R Activation
This preprint describes a computational study aimed at designing improved semaglutide analogues — variants of the GLP-1 receptor agonist used in weight-loss and diabetes treatment — in the context of the underperforming CagriSema Phase III trial. The researchers used an automated "natural amino acid scanning" approach, systematically introducing single amino acid mutations across the semaglutide peptide backbone. Using the crystal structure of the GLP-1–GLP-1R complex (PDB: 4ZGM) as a structural template, they performed high-throughput computational modeling with Modeller and estimated binding affinities (Kd) using the Prodigy tool. From this pipeline, the study identified 564 computationally designed semaglutide analogues predicted to show improved binding affinity to the extracellular domain (ECD) of GLP-1R. The authors propose a conceptual "interfacial electrostatic scaffold" consisting of four salt bridges at the peptide–receptor interface as a framework for next-generation GLP-1R agonist development, drawing an analogy to the century-long iterative optimization of insulin. Key limitations include the fully computational nature of the study — no experimental validation (biochemical, cellular, or in vivo) is presented — and reliance on a single structural template and computational binding affinity estimators, which may not fully capture dynamic receptor behavior.
Why this grade: This is a purely computational/in silico study with no experimental wet-lab, animal, or human data; all findings are based on structural modeling and predicted binding affinity scores.
CagriSema is a fixed-dose combination of cagrilintide (an amylin analogue) and semaglutide (a GLP-1 receptor agonist), and is currently an experimental obesity drug developed by Novo Nordisk. In March 2025, CagriSema underperformed expectations in a Phase III trial, achieving 15.7% weight loss instead of the anticipated 25%, raising concerns about its efficacy and clinical value. Given its chemical composition, the weight-loss efficacy of CagriSema is inextricably linked to the activations of GLP-1R and amylin receptors (AMYRs). With GLP-1R as an example target here, this study employs a structural biophysics-guided computational approach for the design of semaglutide analogues to enhance the activation of its receptor GLP-1R. To fully harness the therapeutic potential of GLP-1R activation, an experimental structural basis (PDB entry 4ZGM) of the GLP-1-GLP-1R interaction is essential for the design of semaglutide analogues, where site-specific missense mutations are engineered into its peptide backbone to establish additional stabilizing interactions with the extracellular domain (ECD) of GLP-1R. Specifically, this study puts forward an automated systemic natural amino acid scanning of the peptide backbone of semaglutide, where PDB entry 4ZGM was used as the structural template for high-throughput structural modeling by Modeller and ligand-receptor binding affinity (Kd) calculations by Prodigy. To sum up, this article reports a total of 564 computationally designed semaglutide analogues with improved GLP-1R ECD binding affinity. Moreover, this study proposes a concept of an interfacial electrostatic scaffold comprising four salt bridges at the binding interface of GLP-1R ECD and semaglutide analogues. Drawing parallels with the continued optimization in the past century of the history of insulin, this article argues that the interfacial electrostatic scaffold here constitutes a robust framework for the continued development of next-generation GLP-1R agonists, enabling more effective therapies for patients with diabetes and/or obesity.
Educational summary of published research — not medical advice. License: cc by. Full text is shown only where licensing permits.