Discovery of the Once-Weekly Glucagon-Like Peptide-1 (GLP-1) Analogue Semaglutide
Lau J, Bloch P, Schäffer L, Pettersson I, Spetzler J, Kofoed J, Madsen K, Knudsen LB
Journal of Medicinal Chemistry (2015)
The molecular design paper for semaglutide — the C18 fatty diacid plus the Aib(8) substitution, paired together to produce week-long albumin binding and full DPP-4 resistance, were the design moves that opened once-weekly GLP-1 dosing as a clinical category.
This 2015 paper from Novo Nordisk's medicinal chemistry group is the foundational design paper for semaglutide and the document that established the molecular pattern most subsequent long-acting GLP-1 analogs followed. The design problem was concrete: native GLP-1 has a half-life of roughly two minutes due to rapid DPP-4 degradation; the existing once-daily liraglutide had solved part of the problem but not all of it. The team's contribution was a peptide engineered to combine three properties — full DPP-4 resistance through an α-aminoisobutyric acid (Aib) substitution at position 8, a long-chain fatty diacid attached via a γ-glutamic acid / OEG linker to Lys26 producing strong albumin binding, and an Arg34 substitution to remove the lysine vulnerability at that position. The two amino-acid substitutions plus the lysine derivatization yielded a molecule with sub-nanomolar GLP-1 receptor potency, near-complete DPP-4 stability, and a plasma half-life supporting once-weekly subcutaneous dosing. The paper presents the structure-activity relationship work behind these choices in detail. The molecule that came out of this design program — semaglutide — is the parent of the modern long-acting GLP-1 class; subsequent designs (tirzepatide, retatrutide) reused the C20 fatty diacid + albumin-binding architecture with different receptor-engagement profiles.
This is a medicinal-chemistry design paper, not a clinical trial. The pharmacological characterization is in vitro and in animal models; the human pharmacokinetic and clinical efficacy data come from subsequent papers (the SUSTAIN program for type 2 diabetes, the STEP program for obesity, and SELECT for cardiovascular outcomes). The structure-activity work the paper presents is necessarily backward-looking — the published account selects a clean design narrative through what was a much messier development process — and reads as a polished design rationale rather than a record of all paths considered. Industry origin (Novo Nordisk) is the operative context. Readers seeking the underlying chemistry will find this paper useful for understanding why semaglutide looks the way it does; readers seeking the clinical translation should follow citations into the SUSTAIN, STEP, and SELECT trial reports.