Introduction
The incretin hormone glucagon like peptide-1 (GLP-1) is secreted from enteroendocrine L cells following ingestion of food, and enhances glucose-dependent insulin secretion by activating GLP-1 receptors (GLP1R) on pancreatic beta cells [
1]. GLP-1 mimetics are available in the clinic for the treatment of type 2 diabetes [
2], but several effects of GLP-1 remain unexplained at a mechanistic level, particularly in extrapancreatic tissues.
When considering GLP1R expression, the field of research has been limited by a lack of specificity of commercially available GLP1R antibodies [
3]. GLP1R is a G-protein-coupled receptor (GPCR); thus, generating antibodies is inherently difficult due to the seven transmembrane domains of the receptor. An alternative method for studying receptor expression has been transgenic expression of fluorescent reporters downstream of the
Glp1r promoter [
4]. Aside from this, the monoclonal antibody mAb3F52 with specificity for human and monkey GLP1R has been generated and used in immunostaining [
5]. Combined, these methods have been used to report GLP1R expression across several tissues [
4‐
6].
The seemingly widespread expression of GLP1R suggests that GLP-1 has a number of functions apart from enhancing glucose-induced insulin secretion. Within the pancreas, GLP-1 inhibits glucagon secretion from alpha cells, and stimulates somatostatin secretion from delta cells. Other proposed functions include stimulation of natriuresis in the kidneys, decrease of food intake via signalling in the central nervous system, modulation of heart rate, and cardioprotection in myocardial ischaemia [
7]. Physiological effects of GLP-1 mimetics in the clinic include decreased cardiovascular risk and increased risk of retinopathy, although this varies between studies and may depend on the agonist used [
8‐
10]. Uncertainty surrounds whether these additional effects are mediated directly via GLP1R on affected tissues, indirectly via GLP1R activation in neurons or through GLP1R-independent pathways [
11].
GLP1R antagonists could be used to address some of these functional questions. The objective of this study was to generate and characterise a monoclonal antagonistic antibody for GLP1R that could be used to block GLP1R signalling in vivo. In comparison to the peptide antagonist exendin 9–39, an antibody would provide the advantage of having an extended half-life for use in subchronic functional studies. As off-target effects have also been observed for exendin 9–39 [
12], another major advantage of a GLP1R antagonistic antibody is specificity for GLP1R. Here we developed an antagonistic antibody against GLP1R, and characterised it in a number of in vitro assays and in vivo studies using lean C57/Bl6 mice, which are well established for studying glucose homeostasis in the context of diabetes.
Discussion
A monoclonal antagonistic antibody, Glp1R0017 targeting the GLP1R with nanomolar affinity has been generated using naive phage display. Schild regression analysis showed that Glp1R0017 antagonism of GLP1R is surmountable, as the maximal receptor activity was achieved with increasing competing concentrations of GLP-1. This suggests that Glp1R0017 is a competitive antagonist, although the slope of the Schild plots for Glp1R0017 and exendin 9-39 did not equal 1. This may suggest that equilibrium between antagonist and agonist had not been reached at the time of cell lysis within the cAMP assay, or that the inhibition was not a simple competition for the same binding site and the calculated dissociation constant should only be taken as an estimate.
Glp1R0017 not only inhibited cAMP production from the GLP1R, but also reduced the GLP-1-triggered increase in glucose-stimulated insulin secretion. Of interest, when assessing the effect of Glp1R0017 on insulin secretion from INS-1 832/3 cells, Glp1R0017 also significantly reduced insulin secretion in 8.3 mmol/l glucose without added GLP-1 (1–0.70-fold, p < 0.05). On further investigation, we found that GLP-1 was produced by INS-1 832/3 cells, increasing from 16.0 ± 1.6 pg/ml at 2 mmol/l glucose to 55.3 ± 3.0 pg/ml at 8.3 mmol/l glucose (p < 0.001, n = 9 wells) in the supernatant fraction, which probably explains the observed antibody effect in the absence of added GLP-1. Although GLP-1 secretion in the vicinity of INS-1 832/3 cells is hard to estimate, the observed inhibition suggests that the antibody can compete with very low effective GLP-1 concentrations in the picomolar range. In light of the endogenous GLP-1 secretion by INS-1 832/3 cells, this cell line could not be used to analyse antibody off-target effects on insulin secretion triggered by submaximal concentrations of GIP or other Gs-coupled stimuli.
Immunostaining of the mouse pancreas with Glp1R0017 can be compared with that recently reported with 7F38A2 [
24]. Both antibodies immunostained the beta cells of the islets of Langerhans in a GLP1R-dependent fashion. The Glp1R0017 cross-species reactivity, shown by the cAMP HTRF assay, suggests that immunostaining tissue from different species should be possible with Glp1R0017. This would complement immunostaining with MAb 3F52, which shows GLP1R localisation in monkey and human tissue [
5]. MAb 3F52 has been characterised as an antagonistic antibody that directly blocks the GLP-1-binding site of human GLP1R [
25]; characterisation of 7F38A2 antagonistic activity has, however, not yet been published. Glp1R0017 adds to these antibodies, providing an additional tool for studying GLP1R physiology in rodents and the other species (cynomolgus monkey, dog and human) with which it was cross-reactive. The specificity of Glp1R0017, demonstrated across a range of assays, provides a major benefit when compared with the peptide antagonist exendin 9-39, which has been shown to have off-target effects [
12], although we cannot exclude potential interactions with unknown targets.
The in vivo ability of Glp1R0017 to inhibit GLP1R is a key finding of this study. Based on the pharmacokinetic study, it is estimated that the biological half-life of Glp1R0017 is longer than 120 h. Thus, with repeated doses, Glp1R0017 could be used to investigate chronic inhibition of GLP1R. The availability of both a GLP1R antagonistic antibody Glp1R0017, and a GIPR antagonistic antibody Gipg013 [
16], enables investigation of the enteroinsular axis in vivo. The OGTT results are in line with GTT data from
Glp1r
−/− and
Gipr
−/− mice as, in both cases, only mild effects on glucose tolerance were observed in the individual knockout mice and single-antibody-treated animals [
23,
26], whereas the combination of GLP1R and GIPR antibodies had a marked effect on glucose levels following the GTTs. These results support the idea that the two incretin hormones can compensate for each other, at least in part. A similar conclusion was reached in studies using single or double incretin receptor knockout mice [
27]. Our finding that a single dose of Gipg013 did not significantly affect oral glucose tolerance, whereas Glp1R0017 caused a small but significant increase in plasma glucose levels after the OGTT, suggests that GLP-1 may be more important than GIP for the incretin effect in this mouse model, contrasting with studies using single and double incretin receptor knockout mice, which concluded that GIP was the more important incretin hormone [
27].
The IPGTTs demonstrated that Glp1R0017 was able to block the effect of exogenously dosed liraglutide. This property of Glp1R0017 could be applied to the study of GLP-1 containing dual or triple agonists, which are emerging for the treatment of type 2 diabetes [
28,
29]. These agonists are based on the concept that gastric bypass surgery, which commonly resolves the symptoms of type 2 diabetes, does not solely target one molecular pathway [
30]. The balance of the different components within multi-agonists is important in development [
14]; thus, Glp1R0017 may be useful for investigating the importance of the GLP-1 component within these agonists.
The ability of Glp1R0017 to block GLP1R in vivo opens up the possibility of studying extrapancreatic effects of GLP-1 in multiple species due to its cross-reactivity. There are a number of questions over how GLP-1 analogues exert the beneficial cardiovascular effects observed in clinical trials [
8‐
10]. In summary, the GLP1R antagonistic antibody Glp1R0017, developed and extensively characterised here, provides a novel tool for investigating the GLP-1 component of unimolecular dual agonists for the treatment of type 2 diabetes, and for further understanding the physiology of GLP1R in vitro and in vivo. Glp1R0017 provides a new method to block GLP-1 receptors over several days in a range of species. It complements the use of
Glp1r knockout mice by enabling transient and/or age-restricted GLP1R blockade, and may have advantages over exendin 9–39 when it is important to exclude potential cross-reactivity with related GPCRs.
Acknowledgements
C. B. Newgard (Duke University, Durham, NC, USA) kindly supplied the INS-1 832/3 cells. Glp1r
−/− animals were acquired under license from D. Drucker (Lunenfeld-Tanenbaum Research Institute, Toronto, ON, Canada). The pharmacokinetic experiment was carried out with the assistance of D. Corkill (MedImmune, Cambridge, UK) and the MedImmune Cambridge BSU team. The Meso Scale Discovery rat insulin assay was performed by K. Burling and his team at the Core Biochemical Assay Laboratory (Cambridge, UK). The authors wish to thank all of the above. Some of the data were presented as a poster at the 53rd EASD Annual Meeting in Lisbon, 2017.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.