In vivo and in vitro characterization of GL0034, a novel long-acting glucagon-like peptide-1 receptor agonist

Abstract

Aims: To describe the in vitro characteristics and antidiabetic in vivo efficacy of the novel glucagon-like peptide-1 receptor agonist (GLP-1RA) GL0034.

Materials and methods: Glucagon-like peptide-1 receptor (GLP-1R) kinetic binding parameters, cyclic adenosine monophosphate (cAMP) signalling, endocytosis and recycling were measured using HEK293 and INS-1832/3 cells expressing human GLP-1R. Insulin secretion was measured in vitro using INS-1832/3 cells, mouse islets and human islets. Chronic administration studies to evaluate weight loss and glycaemic effects were performed in db/db and diet-induced obese mice.

Results: Compared to the leading GLP-1RA semaglutide, GL0034 showed increased binding affinity and potency-driven bias in favour of cAMP over GLP-1R endocytosis and β-arrestin-2 recruitment. Insulin secretory responses were similar for both ligands. GL0034 (6 nmol/kg) led to at least as much weight loss and lowering of blood glucose as did semaglutide at a higher dose (14 nmol/kg).

Conclusions: GL0034 is a G protein-biased agonist that shows powerful antidiabetic effects in mice, and may serve as a promising new GLP-1RA for obese patients with type 2 diabetes.

Keywords: GLP-1 analogue; antidiabetic drug; antiobesity drug; beta-cell function; drug development; incretin therapy.

FIGURE 1

Pharmacological characterization of GL0034 in comparison to semaglutide. (A) Amino acid sequences of glucagon‐like peptide‐1 [GLP‐1](7‐37), semaglutide and GL0034. (B) Binding of exendin(9‐39)‐ fluorescein isothiocyanate (Ex9‐FITC) to HEK293‐SNAP‐ glucagon‐like peptide‐1 receptor (GLP‐1R) cells with and without coaddition of indicated concentration of GL0034 or semaglutide (Sema), n = 4. (C) cAMP production and GLP‐1R internalization measured in parallel HEK293‐SNAP‐GLP‐1R cells, with three‐parameter logistic curve fitting of pooled data from n = 5 repeats. (D) β‐arrestin‐2 recruitment in PathHunter‐GLP‐1R‐EA‐βarr2 cells, n = 5, (E) subtracted pEC₅₀ estimates from (C) and (D), compared by one‐way ANOVA with Tukey's test. (F) Representative images showing GLP‐1R internalization in INS‐1‐SNAP‐GLP‐1R cells stimulated by 100 nM agonist. (G) Quantification of internalized GLP‐1R from n = 4 repeats of experiment shown in (F). (H) Quantification of exendin‐4‐TMR uptake by INS‐1‐SNAP‐GLP‐1R cells after prior treatment with 100 nM semaglutide or GL0034, indicating re‐emergence of internalized GLP‐1R over time, from n = 5 experiments and comparison by two‐way repeated‐measures ANOVA with Sidak's test. *P < 0.05. Data represented as mean ± SEM

FIGURE 2

Comparable in vitro insulin secretory effects of GL0034 and semaglutide. (A) Insulin secretion from mouse islets stimulated for 30 minutes with 1 nM, 10 nM or 100 nM at 11 mM glucose, represented as fold increase versus 11 mM glucose alone, n = 3‐5, paired t‐tests. (B) As for (A) but with human islets, n = 4. (C) Insulin secretion from INS‐1832/3 beta cells stimulated at 11 mM glucose with a range of agonist concentrations for 16 hours, n = 5. Data represented as mean ± SEM, with individual replicates shown where possible

FIGURE 3

GL0034 has powerful antidiabetic effects in vivo. (A) Plasma GL0034 concentration following single subcutaneous injection (1 mg/kg) in male CD‐1 mice (n = 5 per timepoint). See also Table 2. (B) Cumulative food intake during a 28‐day study in db/db mice (n = 8/group, average body weight at study initiation = 48.3 ± 0.7 g) receiving alternative day subcutaneous injection of the indicated agonist. (C) As for (B) but showing body weight reduction. (D) Total body weight loss at the end of the study, with comparison by one‐way ANOVA with Holm Sidak's test to compare groups. (E) As for (D) but for plasma glucose concentration. (F) As for (D) but insulin concentration. (G) As for (D) but HOMA‐B as a measure of beta‐cell function. (H) As for (D) but % glycated haemoglobin (HbA1c). (I) As for (D) but plasma glucagon concentration. (J) As for (D) but plasma triglyceride concentration. *P < 0.05 versus vehicle, £ P < 0.05 versus semaglutide by indicated statistical test. Data represented as mean ± SEM, with individual replicates shown where possible. Dula, dulaglutide; Sema, semaglutide; Veh, vehicle

FIGURE 4

Effects of GL0034 and semaglutide in high‐fat, high‐sucrose‐fed (HFHS) mice. (A) Intraperitoneal glucose (3 g/kg) tolerance test (IPGTT) in HFHS mice 2 hours after subcutaneous administration of GL0034 (6 nmol/kg; n = 7), semaglutide (14 nmol/kg; n = 7) or vehicle (veh; n = 6), with two‐way repeated measures ANOVA and Tukey's test. (B) Area under the curve (AUC) analysis from (A), one‐way ANOVA with Tukey's test. (C) Body weight change over 4 weeks with subcutaneous administration of GL0034 (6 nmol/kg) or semaglutide (14 nmol/kg) every 48 hours, two‐way repeated‐measures ANOVA and Tukey's test. (D) Body weight loss at the end of the study, one‐way repeated‐measures ANOVA and Tukey's test. (E) IPGTT at 14 days, performed 48 hours after agonist dosing, with two‐way repeated measures ANOVA and Tukey's test. (F) As for (E) but at 28 days. (G) AUC analysis from (E) and (F), one‐way ANOVA with Tukey's test. (H) Plasma insulin concentrations from IPGTT shown in (F), two‐way repeated measures ANOVA and Tukey's test. (I) Fasting blood glucose from (F), one‐way ANOVA with Tukey's test. (J) Fasting plasma insulin from (H), one‐way ANOVA with Tukey's test. (K) HOMA1‐B calculated from (I) and (J), one‐way ANOVA with Tukey's test. (L) Oral glucose tolerance test (3 g/kg) performed at the end of the study, 48 hours after agonist dosing, with two‐way repeated‐measures ANOVA and Tukey's test. (M) Plasma amylase concentration from the end of the study, n = 3. (N) Representative immunohistochemical images from pancreata stained for insulin (green) or glucagon (red); scale bar = 100 μm. (O) Quantification of alpha (α) and beta (β)‐cell mass from (N), n = 3. * P < 0.05 versus vehicle (with colour coding: Blue = semaglutide, red = GL0034), £ P < 0.05 GL0034 versus semaglutide, by indicated statistical test. Data represented as mean ± SEM, with individual replicates shown where possible. Sema, semaglutide; Veh, vehicle