The glucagon-like peptide 1 receptor (GLP-1R) is a class B G-protein coupled receptor that is predominantly expressed on pancreatic β-cells in mice. Endogenous and synthetic ligands for GLP-1R exhibit distinct intracellular signaling cascades and promote insulin secretion to various degrees. Ligand binding typically induces receptor internalization followed by recycling back to the membrane or degradation within the lysosome. We hypothesized that endogenous and synthetic GLP-1R ligands would have distinct rates of internalization and recycling in β-cells. To test this hypothesis, we combined a recently-validated, highly-specific GLP-1R antibody (Glp1r0017) with flow cytometry and applied ligands (GLP-1, exendin-4 [Ex4], glucagon, and tirzepatide [TZP]) to characterize GLP-1R trafficking in freshly-isolated mouse islet cells. Under basal conditions, ~90% of β-cells were GLP1R+, with very low or absent binding in α- and δ-cells. Increasing ligand concentrations (0.01-100 nM) demonstrated distinct rates of internalization between ligands. Most potent was Ex4, followed by the native ligand GLP-1. TZP, a dual-receptor ligand for GLP-1R and GIPR, was more modest in inducing internalization. While showing different potencies, Ex4, GLP-1, and TZP were able to induce maximal internalization. In contrast, maximal treatment with glucagon only resulted in ~70% of β-cells being GLP-1R+. To test the rate of recycling back to the plasma membrane, we treated islet cells with Ex4 or GLP-1 with a dose that induced ~50% internalization, removed the ligand, and then monitored GLP1R+ β-cells over time. Within the first 15 minutes, GLP1R+ β-cells increased linearly and rapidly. This plateaued from 30-60 minutes, resulting in 60-70% of β-cells being GLP1R+ at the end of the experiment. In summary, our assay allows for investigation of GLP-1R trafficking in live β-cells. Understanding into the factors that regulate GLP-1R internalization and recycling may shed light on the insulinotropic properties of these ligands. Disclosure S. M. Gray: None. K. Sloop: Employee; Self; Eli Lilly and Company. P. Ravn: None. J. Campbell: None. D. A. D’alessio: Advisory Panel; Self; Eli Lilly and Company, Sun Pharmaceutical Industries Ltd., Research Support; Self; Eli Lilly and Company, Merck & Co., Inc. Funding National Institutes of Health (F32DK121420, R01DK123075, R01DK101991); Lilly Research Award Program

The ever-growing prevalence of obesity and its associated comorbidities (T2D, NASH/NAFLD) is driving the need to discover new therapies for improving metabolic health. Recently, multi-receptor agonists have offered promise for meeting this need. Here, we characterize LY3437943, a novel single agent tri-agonist at the GIP, GLP-1, and glucagon (Gcg) receptors (R). Pharmacologic analysis of LY3437943 in cAMP assays using recombinant cell lines expressing the individual receptors indicated a potency balance favoring GIPR agonism (1.7- and 2.5-fold less potent at the GLP-1R and GcgR, respectively, but 7-fold more potent at the GIPR; all potencies in relation to the native ligands). In endogenous cells, LY3437943 regulated adipocyte lipolysis and hepatocyte glucose output. In vivo studies demonstrated regulation of multiple metabolic endpoints. Acute treatment with LY3437943 dose-dependently inhibited semi-liquid gastric emptying in mice and enhanced glucose dependent insulin secretion in rat IVGTT experiments. Chronic studies in diet induced obese mice reduced food intake and body weight (45% weight loss primarily via reduced fat mass) superior to other GIPR and GLP-1R agonists. In these experiments, LY3437943 lowered blood glucose and plasma insulin, indicating improved insulin sensitivity. Additionally, chronic administration improved biomarkers of liver health, decreasing both plasma alanine aminotransferase and liver triglycerides. Rodent and cynomolgus monkey PK modeling also suggested the potential for weekly dosing in humans. Taken these findings together, LY3437943 is a novel tri-agonist at the GIPR, GLP-1R, and GcgR, producing superior weight loss and glycemic control compared with other incretin receptor-targeting molecules and offers additional benefit for liver health. These findings prompt evaluation of the potential clinical benefit of LY3437943 in patients with obesity and metabolic diseases. Disclosure T. Coskun: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company. F. S. Willard: Employee; Self; Eli Lilly and Company. J. V. Ficorilli: None. O. Cabrera: None. S. Urva: Employee; Self; Eli Lilly and Company. F. Norouziyan cooper: None. L. Guo: Employee; Self; Eli Lilly and Company. J. Alsina-fernandez: None. H. Qu: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company. J. S. Moyers: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company. W. C. Roell: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company. L. O’farrell: None. A. Regmi: Employee; Self; Eli Lilly and Company. X. Ruan: None. A. D. Showalter: None. K. Sloop: Employee; Self; Eli Lilly and Company. D. B. Wainscott: Employee; Self; Eli Lilly and Company, Employee; Spouse/Partner; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company, Stock/Shareholder; Spouse/Partner; Eli Lilly and Company. Funding Eli Lilly and Company

GIP receptor (GIPR) agonism enhances the reduction of food intake and weight loss induced by GLP-1 receptor (GLP-1R) agonism. Recently, GLP-1R agonists have been described that exhibit biased agonism as determined using cells engineered to facilitate measuring the two canonical signaling pathways engaged upon binding the GLP-1R. Such “biased agonists” retain the ability to activate the G alpha S/cyclic AMP (cAMP) pathway to a similar magnitude as native GLP-1 but exhibit markedly weaker ability to induce receptor recruitment of beta-arrestin. The prototype biased GLP-1R agonist Exendin-Phe1 (Ex-Phe1) is reported to exhibit greater weight and glucose control in diet-induced obese (DIO) mice than its unbiased parent Exendin-4 (Ex4). Herein, we investigated whether the enhanced weight loss of Ex-Phe1 erodes the ability of GIPR agonism to further enhance the efficacy of GLP-1R agonism. The peptides were first characterized in vitro to validate the biased nature of Ex-Ph1. In cells expressing either the human or murine GLP-1R, Ex-Phe1 activated cAMP signaling to a similar magnitude as GLP-1 and Ex4 but with approximately 5-fold lower potency than Ex4. The maximum effect of Ex-Phe1 upon human or mouse GLP-1R recruitment of beta-arrestin was 20% compared with full efficacy for Ex4 relative to GLP-1. In this assay Ex-Phe1 was marginally less potent than Ex4. In a 14-day osmotic minipump DIO mice study, Ex4 and Ex-Phe1 each dose dependently reduced food intake and lowered body weight. Ex-Phe1 was 10-fold more potent than Ex4, and at the maximum dose, delivered superior weight loss. Combination treatment with d-Ala-GIP resulted in similar additional body weight loss to that achieved by either Ex-Phe1 or Ex4 alone by enhancing food intake reduction and increasing energy expenditure. D-Ala-GIP alone had no effect on body weight. These findings indicate that GIPR agonism enhances GLP-1R agonism mediated weight loss irrespective of the pathway bias nature of the latter. Disclosure M. P. Coghlan: Employee; Self; Eli Lilly and Company. K. Sloop: Employee; Self; Eli Lilly and Company. T. Coskun: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company. L. O’farrell: None. A. D. Showalter: None. D. B. Wainscott: Employee; Self; Eli Lilly and Company, Employee; Spouse/Partner; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company, Stock/Shareholder; Spouse/Partner; Eli Lilly and Company. C. Stutsman: None. G. Cardona: Employee; Self; Eli Lilly and Company, Stock/Shareholder; Self; Eli Lilly and Company. O. Cabrera: None. J. Alsina-fernandez: None. F. S. Willard: Employee; Self; Eli Lilly and Company. Funding Eli Lilly and Company

Glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide 1 (GLP-1) are the principal incretins connecting nutrient intake to postprandial insulin secretion. Both GIP and GLP-1 directly stimulate insulin secretion in β-cells through cognate receptors (GIPR; GLP-1R) in a glucose-dependent manner. The GIPR is also expressed on α-cells and GIP stimulates glucagon secretion. α- to β-cell communication, manifest as activation of β-cell GLP-1R by proglucagon peptides (PGDP), dictates the magnitude of the insulin response. We hypothesized that GIPR-stimulation of PGDP enhances insulin secretion through α- to β-cell communication, and that GIPR activity in α-cells is nutrient dependent, similar to β-cells. Mouse islets perifused with GIP or alanine alone doubled their secretion of glucagon, but the combination produced ∼10x glucagon secretion, a synergistic effect that was present at both 2.8 and 10 mM glucose. At 10 mM glucose, GIP plus alanine also had a synergistic effect on insulin secretion. This synergy was abolished by the GLP-1R antagonist exendin-(9-39) indicating that amino acid enhancement of GIP-stimulated insulin secretion is dependent on α- to β-cell communication. In vivo, mice with α-cell deletion of GIPR had normal intraperitoneal and oral glucose tolerance, but impaired glycemic control and reduced insulin secretion in response to a mixed nutrient stimulus, a condition that enhances α-cell activity. Together, these results demonstrate that GIPR activity in α-cells is dependent upon amino acid activation and contributes significantly to insulin secretion through α- to β-cell communication. Disclosure K.M. El: None. M. Capozzi: None. S.M. Gray: None. J.L. Brown: None. F.S. Willard: None. P. Emmerson: None. K. Sloop: None. D. D’Alessio: Advisory Panel; Self; Eli Lilly and Company. Consultant; Self; Intarcia Therapeutics. Research Support; Self; Ansh Labs, Eli Lilly and Company, Merck Sharp & Dohme Corp. Other Relationship; Self; Novo Nordisk A/S. J. Campbell: Research Support; Self; Eli Lilly and Company, Novo Nordisk Inc. Speaker’s Bureau; Self; Merck Sharp & Dohme Corp. Funding American Diabetes Association (1-17-JDF-074 to J.C.)

Incretin-based drugs are widely thought to act via a Gαs/adenylate cyclase (AC)/cAMP pathway in β-cells. This model has been challenged by data demonstrating various glucagon-like peptide 1 receptor (GLP-1R) ligands differentially activate multiple signaling pathways including Gαs, Gq, ERK, and β-arrestins. This ligand-mediated signaling bias implicates an array of intracellular mechanisms that regulate GLP-1R action and raises the possibility GLP-1R agonists can be optimized to enhance insulin secretion. We tested the hypothesis that β-arrestin 1 (βarr1), a critical GPCR effector, is necessary for the full insulinotropic action of GLP-1R agonists. The β-cell βarr1 knockout mice (βarr1βcell-/-) used here have reduced Arrb1 expression in the β-cell, but not α- or Δ-cells; Arrb2 expression was unaffected. Islets isolated from βarr1βcell-/- mice have increased glucose stimulated insulin secretion (GSIS) compared to controls. The enhanced GSIS in the knockout islets was ablated by the GLP-1R antagonist exendin-9. In addition, βarr1βcell-/-islets treated with GLP-1 or exendin 4 (Ex4) have increased insulin release compared to controls. Conversely, βarr1 deletion did not impact the insulinotropic actions of glucagon, which signals through both the GLP-1R and the glucagon receptor (GCGR). Moreover, control and βarr1βcell-/- islets produced similar insulin secretion in response to a GCGR specific agonist that does not signaling through GLP-1R and to glucose-dependent insulinotropic polypeptide. Together, these results indicate that βarr1 in β-cells selectively dampens the insulinotropic actions of the GLP-1R, but not the GCGR or GIP receptor. Thus, there is a specific interaction between the GLP-1R and βarr1 that points to divergent regulation between the insulinotropic class B G-protein coupled receptors in β-cells. This data implies that GLP-1R agonists that bias away from βarr1 may be more effective at stimulating insulin secretion. Disclosure J.D. Douros: None. K. Sloop: None. J. Campbell: Research Support; Self; Eli Lilly and Company, Novo Nordisk Inc. Speaker’s Bureau; Self; Merck Sharp & Dohme Corp. D. D’Alessio: Advisory Panel; Self; Eli Lilly and Company. Consultant; Self; Intarcia Therapeutics. Research Support; Self; Ansh Labs, Eli Lilly and Company, Merck Sharp & Dohme Corp. Other Relationship; Self; Novo Nordisk A/S. Funding 1F32DK115031-01

The glucagon-like peptide 1 receptor (GLP-1R) is a class B G-protein coupled receptor that is insulinotropic in β-cells. Multiple endogenous and synthetic ligands exist for the GLP-1R and each ligand interacts with the receptor differently to produce specific activation of intracellular signaling pathways. Ligand binding typically induces receptor internalization, leading to either i) receptor degradation or ii) recycling to the cell membrane. We hypothesized that each specific GLP-1R ligand would produce a unique trafficking profile in primary β-cells. To test this hypothesis, we developed flow cytometry-based assay that can be utilized in primary dispersed mouse islets. First, we validated a GLP-1R antibody (GLP-1R-APC) for flow cytometry application. The GLP-1R-APC stained β-cells, but not α-cells, aligning with the reported expression pattern in islets. Post-sort analysis confirmed GLP-1R expression only in the GLP-1R-APC+ population. Finally, GLP-1R-APC did not stain any cell populations in GLP1R-/- islets. Next, we used this assay to quantify GLP-1R internalization in wild type β-cells. To induce internalization, we treated dispersed islets with either vehicle or a GLP-1R ligand for 30 minutes prior to antibody staining. After vehicle treatment, 90% of the β-cells stained GLP-1R-APC+. Treatment with either GLP-1 or exendin-4 reduced the number of GLP-1R-APC+ β-cells by 97 and 89%, respectively, suggesting near complete internalization of all GLP-1Rs. In contrast, treatment with the GLP-1R antagonist exendin-9 produced staining similar to PBS controls. Finally, we used image-stream technology, a combination of brightfield microscopy and flow cytometry, to visually confirm GLP-1R internalization. Here, vehicle-treated cells produced GLP-1R staining only on the cell membrane, which was nearly ablated by the treatment with a ligand. In conclusion, we have developed an assay that allows for quantification of GLP-1R trafficking in primary mouse islets. Disclosure S.M. Gray: None. P. Ravn: Employee; Self; AstraZeneca. K. Sloop: None. J. Campbell: Research Support; Self; Eli Lilly and Company, Novo Nordisk Inc. Speaker’s Bureau; Self; Merck Sharp & Dohme Corp. D. D’Alessio: Advisory Panel; Self; Eli Lilly and Company. Consultant; Self; Intarcia Therapeutics. Research Support; Self; Ansh Labs, Eli Lilly and Company, Merck Sharp & Dohme Corp. Other Relationship; Self; Novo Nordisk A/S.

The glucagon-like peptide 1 receptor (GLP-1R) is insulinotropic in β-cells and the target of multiple classes of diabetes drugs. People with T2D secrete normal amounts of GLP-1, but are less responsive to its actions; the mechanisms for this are unknown. Studies have shown that metabolic stress decreases GLP-1R in islets, however, it is unknown if this occurs in all, or just a subset, of β-cells. We hypothesized that GLP-1R expression is heterogeneous in β-cells, such that not all β-cells express the GLP-1R, and that metabolic stress would decrease the number of GLP-1R+ β-cells. To test this, we first performed single-cell RNA sequencing (scRNAseq) in mouse and human islet cells and found that the majority of β-cells were GLP-1R negative, supporting significant heterogeneity. To validate the scRNAseq results, we generated a GLP-1R reporter mouse by crossing GLP-1R-Cre with mTmG mice. Using FACS to separate and qPCR to characterize positive (GFP+) from negative (tdTomato+) dispersed islet cells, we found GLP-1R+ cells were enriched for Ins2 and Sst, while GLP-1R- cells were enriched for Gcg, suggesting GLP-1R promoter activity is localized to β- and d-cells. In addition, β-cell markers were not present in the tdTomato+ population, suggesting very few GLP-1R negative β-cells. Next, we stained dispersed islets from the reporter mice with a validated GLP-1R antibody and found >90% congruence with GFP+ cells and no staining in tdTomato+ cells. In dispersed islets from wild type mice, ∼90% of β-cells were GLP-1R+; very few of the α- or d-cells were GLP-1R+. Metabolic stress induced through chronic high-fat feeding or using multiparous models did not alter the number of GLP-1R+ β-cells. In conclusion, assessment of GLP-1R expression and GLP-1R protein revealed the majority of β-cells are GLP-1R+, a finding that conflicts with the interpretation of scRNAseq data. These findings suggest caution in interpreting the results of scRNAseq for low abundant transcripts such as the GLP-1R. Disclosure S.M. Gray: None. B.M. Chazotte: None. E.C. Ross: None. B. Svendsen: None. P. Ravn: Employee; Self; AstraZeneca. K. Sloop: None. J. Campbell: Research Support; Self; Eli Lilly and Company, Novo Nordisk Inc. Speaker’s Bureau; Self; Merck Sharp & Dohme Corp. D. D’Alessio: Advisory Panel; Self; Eli Lilly and Company. Consultant; Self; Intarcia Therapeutics. Research Support; Self; Ansh Labs, Eli Lilly and Company, Merck Sharp & Dohme Corp. Other Relationship; Self; Novo Nordisk A/S.

The incretin peptides glucose-dependent insulinotropic polypeptide (GIP) and glucagon-like peptide 1 (GLP-1) stimulate insulin secretion in β-cells by binding to cognate receptors. The GIPR and GLP-1R are family B G-protein coupled receptors generally thought to mediate signaling through G-αs and generation of cAMP. However, the GLP-1R has also been reported to signal through alternative pathways beyond Gαs, and the relative contributions of Gαs vs. non-Gαs are unknown. The GLP-1R is essential for basal levels of cAMP and normal β-cell tone. We hypothesized that elimination of Gαs would render mouse islets insensitive to GLP-1R agonists. Induction of β-cell specific Gnas knockout with tamoxifen inducible cre caused a >90% reduction in Gαs mRNA in purified β-cells (Gnasβcell-/-). Perifusion of control and knockout islets demonstrated reduced glucose-stimulated insulin secretion, but persistence of an insulin response to GLP-1R agonists that was ∼40% of control. Similarly, insulin secretion in response to either GIP or glucagon was only partially reduced in Gαs knockout islets. A Gq inhibitor partially reduced insulin secretion in control islets, but abolished GLP-1 stimulated insulin secretion in Gnasβcell-/- islets. These results demonstrate that the incretins can signal through both Gαs and non-Gαs mechanisms. Interestingly, Gnasβcell-/- mice were severely hyperglycemic (>400 mg/dl) and had no reduction of glycemia in response to GLP-1R or GIPR agonists. Thus, non-Gαs signaling by either GLP-1R or GIPR is unable to stimulate sufficient insulin secretion to maintain euglycemia in vivo. Disclosure M. Capozzi: None. S.M. Gray: None. P. Emmerson: None. F.S. Willard: None. K. Sloop: None. D. D’Alessio: Advisory Panel; Self; Eli Lilly and Company. Consultant; Self; Intarcia Therapeutics. Research Support; Self; Ansh Labs, Eli Lilly and Company, Merck Sharp & Dohme Corp. Other Relationship; Self; Novo Nordisk A/S. J. Campbell: Research Support; Self; Eli Lilly and Company, Novo Nordisk Inc. Speaker’s Bureau; Self; Merck Sharp & Dohme Corp. Funding American Diabetes Association (1-17-JDF-074 to J.C.); National Institutes of Health (R01DK123075, F32DK116542)

In August 2016, several leaders in glucagon biology gathered for the European Association for the Study of Diabetes Hagedorn Workshop in Oxford, England. A key point of discussion focused on the need for basal insulin to allow for the therapeutic benefit of glucagon blockade in the treatment of diabetes. Among the most enlightening experimental results presented were findings from studies in which glucagon receptor-deficient mice were administered streptozotocin to destroy pancreatic beta cells or had undergone diphtheria toxin-induced b cell ablation. This article summarizes key features of the discussion as a consensus was reached. Agents that antagonize glucagon may be of great benefit for the treatment of diabetes; however, sufficient levels of basal insulin are required for their therapeutic efficacy.