Treatments for diabetes and obesity based on enteroendocrine hormones are a focus of research interest, partly due to the successes of glucagon-like peptide-1 (GLP-1) mimetic peptides in the treatment of diabetes and the correlation of altered enteroendocrine profiles with the positive metabolic outcomes of gastric bypass surgery. It is thought that simultaneous stimulation of more than one receptor might mimic the superior efficacy of the latter and dual or triple-agonist peptides are under investigation. An important step in developing multiple agonists is to establish the relative pharmacological potency and efficacy of new molecules at its different target receptors, and to optimise the balance of activities to achieve the desired treatment outcome. In a recent issue of the Biochemical Journal, Naylor et al. described how they used CRISPR technology to modulate endogenous receptor density in insulinoma cells to get the balance right for a dual incretin peptide engaging both GLP-1- and glucose-dependent insulinotropic polypeptide-receptors.
Recent years have witnessed a surge of interest in gut hormones, following the dramatic successes of many new medicines that target the intestinal endocrine system. Progress has been most marked in the field of type 2 diabetes and obesity, where treatments based around the gut hormone glucagon-like peptide-1 (GLP-1) have revolutionised therapeutic strategies, as they provide similar improvements in blood glucose control to older conventional anti-diabetic drugs but have substantial benefits such as weight loss and low rates of hypoglycaemic side effects . Most notably, however, the most effective single treatment currently available for type 2 diabetes and obesity is gastric bypass surgery. Initially introduced to restrict food intake and promote malabsorption as a treatment for severe obesity, bypass surgery is now recognised to cause rapid resolution of the majority of cases of type 2 diabetes and to improve long-term survival .
Gastric bypass surgery does not seem to work through a single mechanism, but rather triggers a panoply of physiological changes that include increased levels of several gut hormones, altered bile acid metabolism and negative energy balance . Each of these individual factors probably contributes to the effectiveness of bypass surgery, but none seems able on its own to explain the dramatic resolution of type 2 diabetes. A new wave of anti-diabetic and anti-obesity drug development is, therefore, focussing on trying to mimic bypass surgery by targeting multiple pathways using peptides that target more than one hormonal receptor.
The physiological validity of this approach was first demonstrated in reports that the naturally produced gut hormone oxyntomodulin is an agonist at both GLP-1 and glucagon receptors. Like GLP-1, oxyntomodulin stimulates insulin secretion and reduces appetite, but its glucagon receptor activity additionally endows it with the property to increase energy expenditure . A variety of unimolecular peptides have now been designed that similarly target more than one pathway in a single agent. Starting from the success of the established treatments, many dual agonist peptides have GLP-1 receptor activity at their core. The beneficial actions of GLP-1 include its abilities to amplify glucose-stimulated insulin secretion, to reduce appetite and to inhibit glucagon release, although it also causes gastrointestinal side effects particularly at pharmacological concentrations . In dual agonist peptides GLP-1 has been coupled with sequences that target receptors for glucagon, glucose-dependent insulinotropic polypeptide (GIP), GLP-2, and gastrin [6–10]. More recently, the first triple-agonist peptide, a unimolecular peptide targeting GLP-1, glucagon and GIP has been generated .
The most promising dual agonists currently in development are GLP-1/glucagon and GLP-1/GIP peptides. Rodent in vivo studies have demonstrated the effectiveness of GLP-1/glucagon peptides, which have anti-hyperglycaemic effects due to the GLP-1 component, and enhanced body weight reduction attributed to increased energy expenditure caused by the glucagon moiety . Pharmaceutical interest in the GLP-1/glucagon peptides is increasing, and many different co-agonists are currently in Phase I and II clinical trials. GLP-1/GIP dual peptides are a newer addition to the field, with only one of these dual peptides currently in a Phase II clinical trial .
The concept of combining GLP-1 with GIP is an interesting idea, as both peptides are incretin hormones that increase glucose-stimulated insulin secretion from pancreatic β-cells by activation of predominantly Gαs-coupled G-protein coupled receptors . Despite the therapeutic success of GLP-1-based drugs, GIP-based therapies have not been developed because there have been several concerns surrounding the pharmacology of the GIP receptor, most notably that GIP might promote weight gain as suggested by the observation that mouse GIP receptor knockout models are protected from obesity . However, the obesogenic effect of targeting GIP has recently been contradicted by a study showing that GIP overexpression reduced rather than increased body weight . Indeed studies on a GLP-1/GIP dual peptide have shown synergistic effects of the GLP-1 and GIP components on body weight reduction in rodents, and enhanced insulinotropic effects that translate to monkeys and humans .
It is essential in the design of dual peptides to consider whether the drug should be similarly potent (balanced) at both receptors, or whether a better physiological outcome might be obtained with a drug that is biased towards one or other of the target receptors. Assessing balanced vs biased agonism of dual peptides has previously been carried out by measuring potencies in separate cell lines overexpressing the receptors of interest . In the current issue of Biochemical Journal, however, Naylor et al.  describe the generation, using CRISPR/CRISPR associated protein 9 (Cas9), of GLP-1R and GIPR knockout INS-1 pancreatic beta cell lines for characterisation of GLP-1/GIP dual peptides.
The CRISPR/Cas9 technique is based on the bacterial protein Cas9 nuclease which, when complexed with a specific guide RNA, can site-specifically introduce a double strand break in the gene of interest. The cell then attempts to repair the double strand break by non-homologous end joining, but in doing so results in a variety of insertions and deletions (indels) that disrupt the open reading frame of the gene of interest. The normal outcome of CRISPR/Cas9-induced gene disruption is a population of clones, each harbouring different indels [16,17]. To confirm GLP-1R and GIPR knockout in INS-1 cells, Naylor et al. screened five potential clones from each line for functionality and performed sequencing of the disrupted genes. Slightly surprisingly, the sequencing results appeared to reveal many homozygous as well as heterozygous clones. Due to the random nature of the non-homologous end-joining reaction, the likelihood of introducing the same indel into both alleles is low, and homozygosity should be rare. It is possible, however, that the apparent homozygosity in the INS-1 lines reflects an artefact in the method used for sequencing and that a more in-depth analysis would reveal the expected heterozygosity within clones. Perhaps because the authors were similarly uncertain about the genetics of the apparently homozygous clones, they chose to work with heterozygous GLP-1R and GIPR knockout clones for the functional characterisation of GLP-1/GIP dual peptides.
Using the engineered knockout INS lines, biased or balanced agonism to the GIPR and GLP-1R was demonstrated for three different GLP-1/GIP dual peptides by cAMP and insulin secretion assays. This approach presents many advantages, most notably that the resultant cell lines are based on a relevant physiological model for insulin secretion and that they express endogenous levels of the receptors under investigation. Relative potencies and cross-reactivities of a molecule can alternatively be assessed using cell lines that heterologously express candidate receptors, although this approach assumes that receptor overexpression in an alternative cell type will not disrupt ligand binding and downstream signalling. Another method is to utilise antagonistic compounds or antibodies selectively to block individual target receptors, but pharmacological antagonists often lack specificity and efficacy. By assessing a physiologically relevant readout (insulin secretion), downstream from activating endogenous receptor expression, it is hoped that the GLP-1R and GIPR knockout INS-1 cells will be more predictive of GLP-1/GIP dual peptide in vivo effects.
The premise for developing GLP-1/GIP dual peptides was that both GLP-1 and GIP moieties would target pancreatic β-cells, and that efficacy on insulin secretion would be achievable using lower doses of GLP-1 receptor ligand, thereby reducing GLP-1-dependent gastrointestinal side effects [6,15]. For most other dual peptides, however, the different agonist activities are designed to target receptors on separate tissues. GLP-1/glucagon peptides, for example, aim to target GLP-1R on beta cells and brain, but glucagon receptors in the liver. Under these circumstances, a single cell line might not be the system of choice for testing both receptor activities, and it may be more relevant to test the components separately on cell lines representing the different physiological target tissues. Interestingly, some dual agonist molecules are in development in which a peptide is chemically fused to a small molecule hormone (e.g. GLP-1/oestrogen , glucagon/thyroxine ). These chemical entities aim to use the peptide hormone to deliver the small molecule specifically to the target tissue of interest, e.g. thyroxine to the liver, and thus reduce side effects of the small molecule hormones on different tissues. The CRISPR/Cas9 knockout approach would be an ideal method to test these molecules in vitro, because as with the GLP-1/GIP peptides, the fusion molecules are designed to target two distinct receptors in a single cell type.
Although activation of both GIP and GLP-1 receptors in healthy β-cells triggers insulin secretion, it has long been of concern that GIP seems to be poorly effective in people with type 2 diabetes . One might, therefore, wonder why it is worth developing GLP-1/GIP dual agonists at all. One of the major reasons for the loss of GIP activity in type 2 diabetes seems to be receptor down-regulation by hyperglycaemia . It has, therefore, been argued that the GLP-1 component of a GLP-1/GIP dual agonist would stimulate insulin secretion and lower blood glucose even at high blood sugar levels, and that the GIP component would come into action once glycaemia had returned closer to the normal range . The additional GIPR agonistic activity should therefore give the dual peptide increased efficacy compared with GLP-1 mono-agonist therapies. The cell lines developed by Naylor and colleagues will facilitate the selection of dual peptides with the predicted preferable balance of agonism on the different target receptors. Further work, however, will be required to shed light on the optimal balance of agonism for GIP/GLP-1 dual peptides to provide therapeutic blood glucose control in people with type 2 diabetes.
The Authors declare that there are no competing interests associated with the manuscript.