Structural insights into multiplexed pharmacological actions of tirzepatide and peptide 20 at the GIP, GLP-1 or glucagon receptors

Abstract

Glucose homeostasis, regulated by glucose-dependent insulinotropic polypeptide (GIP), glucagon-like peptide-1 (GLP-1) and glucagon (GCG) is critical to human health. Several multi-targeting agonists at GIPR, GLP-1R or GCGR, developed to maximize metabolic benefits with reduced side-effects, are in clinical trials to treat type 2 diabetes and obesity. To elucidate the molecular mechanisms by which tirzepatide, a GIPR/GLP-1R dual agonist, and peptide 20, a GIPR/GLP-1R/GCGR triagonist, manifest their multiplexed pharmacological actions over monoagonists such as semaglutide, we determine cryo-electron microscopy structures of tirzepatide-bound GIPR and GLP-1R as well as peptide 20-bound GIPR, GLP-1R and GCGR. The structures reveal both common and unique features for the dual and triple agonism by illustrating key interactions of clinical relevance at the near-atomic level. Retention of glucagon function is required to achieve such an advantage over GLP-1 monotherapy. Our findings provide valuable insights into the structural basis of functional versatility of tirzepatide and peptide 20.

Fig. 1. Cryo-EM structures of tirzepatide and peptide 20-bound GIPR, GLP-1R and GCGR in complex with G_s.

a Multi-targeting peptides tirzepatide and peptide 20 possess distinct combinatorial agonism at GIPR, GLP-1R and GCGR. b Cryo-EM maps (left) and structural models (right) of tirzepatide-bound GIPR (top) and GLP-1R (bottom) in complex with G_s. The sharpened cryo-EM density map at the 0.243 threshold shown as light gray surface indicates a micelle diameter of 10 nm. The colored cryo-EM density map is shown at the 0.424 threshold. The tirzepatide is shown in salmon, GIPR in yellow green, GLP-1R in dodger blue, Gα_s in yellow, Gβ subunit in cyan, Gγ subunit in navy blue and Nb35 in gray. c Cryo-EM maps (left) and structural models (right) of peptide 20-bound GIPR (top), GLP-1R (middle) and GCGR (bottom) in complex with G_s. The sharpened cryo-EM density map at the 0.228 threshold shown as light gray surface indicates a micelle diameter of 11 nm. The colored cryo-EM density map is shown at the 0.576 threshold. The peptide 20 is shown in orange, GIPR in forest green, GLP-1R in blue, GCGR in hot pink, Gα_s in yellow, Gβ subunit in cyan, Gγ subunit in navy blue and Nb35 in gray.

Fig. 2. Structural comparison of GIPR, GLP-1R and GCGR bound by mono-, dual and triple agonists.

a Structural comparison of GIP–GIPR–G_s, tirzepatide–GIPR–G_s and peptide 20–GIPR–G_s. Receptor ECD and G protein are omitted for clarity. b Comparison of residue interactions employed by GIPR to recognize GIP, tirzepatide and peptide 20, described by fingerprint strings encoding different interaction types of the surrounding residues in each peptide. Color codes are listed on the top panel. Residues that show no interaction with ligands are displayed as white circles. c Structural comparison of GLP-1–GLP-1R–G_s, tirzepatide–GLP-1R–G_s and peptide 20–GLP-1R–G_s. Receptor ECD and G protein are omitted for clarity. d Comparison of residue interactions that GLP-1R employed to recognize GLP-1, tirzepatide and peptide 20, described by fingerprint strings encoding different interaction types of the surrounding residues in each peptide. e Structural comparison of GCG–GCGR–G_s, peptide 15–GCGR–G_s and peptide 20–GCGR–G_s. Receptor ECD and G protein are omitted for clarity. f Comparison of residue interactions that GCGR employed to recognize GCG, peptide 15 and peptide 20, described by fingerprint strings encoding different interaction types of the surrounding residues in each peptide.

Fig. 3. Molecular recognition of tirzepatide by GIPR and GLP-1R.

a Structural comparison of tirzepatide–GIPR–G_s and tirzepatide–GLP-1R–G_s. Receptor ECD and G protein are omitted for clarity. b Interactions between tirzepatide (salmon) and the TMD of GIPR (yellow green). Residues involved in interactions are shown as sticks. c Interactions between tirzepatide (light salmon) and the TMD of GLP-1R (dodger blue). Residues involved in interactions are shown as sticks. d–e Effects of receptor mutations on tirzepatide-induced cAMP accumulation. Data shown are means ± S.E.M. of three independent experiments (n = 3) performed in quadruplicate. f The peptide recognition modes are described by fingerprint strings encoding different interaction types of the surrounding residues in each receptor. Residues that show no interaction with receptors are displayed as white circles. Color codes are listed on the top panel. WT, wild-type. Source data are provided as a Source Data file.

Fig. 4. Molecular recognition of peptide 20 by GIPR, GLP-1R and GCGR.

a Structural comparison of peptide 20–GIPR–G_s, peptide 20–GLP-1R–G_s and peptide 20–GCGR–G_s. Receptor ECD and G protein are omitted for clarity. b–d Interactions between peptide 20 and the TMDs of GIPR (forest green), GLP-1R (blue), and GCGR (hot pink). Residues involved in interactions are shown as sticks. e–g Surface representations of the receptor for each of the peptide-receptor complex, with the peptides shown as ribbon and sticks. The receptor is shown in surface representation and colored from dodger blue for the most hydrophilic region, to white, to orange red for the most hydrophobic region. h The peptide recognition modes are described by fingerprint strings encoding different interaction types of the surrounding residues in each receptor. Color codes are listed on the top panel. Residues that show no interaction with receptors are displayed as white circles.

Fig. 5. Structural and functional feature of lipidated K10^P of peptide 20.

a–c Close-up views of the crevices between TM1 and TM2 displayed by cryo-EM maps of peptide 20-bound GIPR a, GLP-1R b, and GCGR c. Continuous electron densities connected to K10 in peptide 20 were observed in the three peptide 20-bound receptor–G_s complexes. d–f Interactions between lipidated K10^P and the TM1-TM2 crevice of GIPR d, GLP-1R e, and GCGR f, with interacting residues shown in sticks. Hydrogen bonds are shown with dashed lines. g Effects of receptor mutations on peptide 20-induced cAMP accumulation. Data shown are means ± S.E.M. of at least three independent experiments (n = 3–9) performed in quadruplicate. h Effects of K10 lipidation on peptide 20-induced cAMP accumulation. The bar graph represents the average pEC₅₀ (that is, −logEC₅₀) and data are presented as means ± S.E.M. of four independent experiments (n = 4) performed in quadruplicate. Statistically significant differences were determined with a two-tailed Student’s t test. **P < 0.01 and ****P < 0.0001. WT, wild-type. Source data are provided as a Source Data file.

Fig. 6. G protein coupling of multi-targeting agonist-bound GIPR, GLP-1R and GCGR.

a Comparison of G protein coupling among GIPR, GLP-1R and GCGR^{, ,} . The Gα_s α5-helix of the Gα_s Ras-like domain inserts into an intracellular crevice of receptor’s TMD. The receptors and G proteins are colored as the labels. b Comparison of ICL2 conformation in the peptide 20-bound GIPR, GCGR and GLP-1R. c Comparison of F257^3.60b conformation in the GLP-1R bound by GLP-1, tirzepatide and peptide 20. d Comparison of E262^ICL2 conformation in the GLP-1R bound by GLP-1, tirzepatide and peptide 20. e Comparison of E253^ICL2 conformation in the GIPR bound by tirzepatide and peptide 20. Residues involved in interactions are shown as sticks. Polar interactions are shown as black dashed lines.

Fig. 7. Structure-basis of receptor selectivity demonstrated by tirzepatide, peptide 20 and GLP-1 analogs.

a Amino acid sequences of endogenous agonists, multi-targeting agonists and approved GLP-1 analogs including semaglutide. Residues are colored according to sequence conservation among GIP, GLP-1 and GCG. Aib, aminoisobutyric acid. Semaglutide and tirzepatide are conjugated by a C20 fatty diacid moiety via a linker connected to the lysine residue at position 20, while peptide 20 is covalently attached by a 16-carbon acyl chain (palmitoyl; 16:0) via a γ-carboxylate spacer at K10^P. b Receptor signaling profiles of endogenous agonists, multi-targeting agonists and approved drug GLP-1 analogs including semaglutide. Data shown are means ± S.E.M. of four independent experiments (n = 4) performed in quadruplicate. Source data are provided as a Source Data file. c Receptor binding profiles of endogenous agonists, multi-targeting agonists and approved GLP-1 analogs. Data shown are means ± S.E.M. of three independent experiments (n = 3) performed in duplicate.