Targeting GLP-1 receptor trafficking to improve therapies for type 2 diabetes

Type 2 diabetes (T2D) kills around 4 million people worldwide every year and causes ill health in many more. A major underlying problem in T2D is that specialised cells in the pancreas known as beta cells start to fail and do not produce or secrete enough insulin, which is needed to control blood sugar. Beta cells can be stimulated to release more insulin by activating a surface protein known as glucagon-like peptide-1 receptor (GLP-1R). GLP-1R treatments, known as GLP-1R agonists, are already used in T2D. We have found that a process known as "receptor trafficking" limits the effectiveness of GLP-1R agonists. When the agonist activates the receptor, it also "internalises" it into the cell where it is hidden and can no longer be activated by agonists outside the cell. We showed this using a modified version of a GLP-1R agonist currently used in T2D (exendin-4, also known as "Exenatide"), which we engineered to reduce the internalisation process. This new agonist called "Ex-phe1" was much more effective than unmodified exendin-4 in mice with T2D, as it allowed the receptor to remain longer at the cell surface and be continually reactivated, leading to more insulin release. Also, despite being more effective in beta cells, Ex-phe1 did not cause more nausea, a well-known side effect of GLP-1R agonists.
This project has two goals. The first is to understand how changes in GLP-1R trafficking modify insulin release from beta cells. The second, to be performed in parallel, is to optimise the compounds we generated in our preliminary work to produce new, better, treatments for T2D. Our approach includes experiments in beta cells and pancreatic tissue ("islets"), which allow us to understand events in molecular detail, and in mice with T2D, which tell us what happens at the whole body level. We place a lot of emphasis on using human islets (provided for research when surplus to requirements for organ donation), as we believe this is essential to understand and treat human disease.
We highlight below some key experiments:
1) We will use a combination of computer modelling and genetic modification of the GLP-1R to identify specific interactions between the receptor and Ex-phe1 or other biased GLP-1R agonists which underpin their different trafficking and therapeutic effects. This knowledge will be critical for the design of even more effective drugs in the future.
2) We will genetically modify human islets to reduce levels of "beta-arrestin-2", a protein in beta cells that we believe is a key link between GLP-1R activation, trafficking, and beneficial insulin release. We will also breed genetically modified mice lacking beta-arrestin-2 in beta cells. We will use these systems to look at complex patterns of protein and gene activation within beta cells in response to GLP-1R agonist treatment. From this we hope to identify molecules that could in the future be targeted as new T2D treatments.
3) We will study the long-term effects of our new, improved GLP-1R agonist Ex-phe1 on human islets transplanted into diabetic mice. Importantly, we will use a new approach in which the islets are actually transplanted in the eye, where changes to islet shape, behaviour and overall health can be directly and repeatedly viewed in a microscope. The mice will be treated with a long-lasting formulation of Ex-phe1 or exendin-4 for several months, and we will monitor changes to blood sugar, allowing us to compare the treatments for their effects on overall islet function. Due to well-known differences between mouse and human islets, we believe that using this "humanised" mice is critical to understanding long-term treatment effects.
In summary, we have identified an exciting new approach based on GLP-1R trafficking for improving the treatment of T2D. In this project we will establish the molecular details of why this approach works, identify new potential drug targets and demonstrate the long-term effects of this new treatment in unique humanised mice.

The glucagon-like peptide-1 receptor (GLP-1R) is an important target in type 2 diabetes (T2D) as GLP-1R agonists stimulate beta cell insulin release and promote weight loss. The GLP-1R undergoes rapid agonist-mediated endocytosis, leading us to question whether desensitisation limits GLP-1R agonist treatment effectiveness. We have tested this hypothesis by generating modified versions of the GLP-1R agonist exendin-4. One of these, Ex-phe1, displays a sharply reduced propensity for receptor internalisation, is a more potent insulin secretagogue, and provides enhanced glycaemic control in diabetic mice without increasing nausea, a common side effect of GLP-1R agonists. Ex-phe1 exhibits a distinct pharmacological profile to the parent compound, including fast receptor dissociation kinetics and biased signalling, with beta-arrestin-2 (barr2) recruitment selectively reduced. Here, we will elucidate the molecular mechanisms linking biased GLP-1R activation and trafficking to insulin secretion and examine the therapeutic benefits of biased GLP-1R agonism in a humanised T2D model. Specifically, we will: 1) identify the structural basis for biased GLP-1R activation and trafficking using molecular modelling, reciprocal receptor mutagenesis and a GLP-1R conformational biosensor; 2) determine biased agonist-specific kinomic and transcriptomic responses and changes to interactions with endocytic trafficking regulators in beta cells; 3) confirm the role of barr2 in beta cell responses to GLP-1R agonism both in beta cell-specific barr2 knockout mice and by silencing barr2 in human islets; and 4) establish the long-term effects of biased GLP-1R agonism in a humanised mouse model with human islets implanted into diabetic mice eyes, an approach that uniquely allows serial imaging of islet morphology and function. These studies will provide critical molecular insights into the control of GLP-1R trafficking by biased agonists and will potentially identify new targets for T2D treatment.

We identify the following groups as likely beneficiaries from work conducted during this project:

1) Patients with T2D and associated metabolic conditions:

In this project, we will develop new molecules that are expected to display improved therapeutic efficacy and tolerability compared to existing treatments. As well as patients with T2D, patients with non-alcoholic fatty liver disease are another key group who are expected to benefit from treatment with biased GLP-1R agonists. GLP-1R agonism is also an investigational strategy for obesity, cardiovascular disease, and neurological diseases. Full clinical development of new agents is expected to take 5-8 years.

2) The NHS:

The S.R.B. group has considerable experience of "first in man" and phase 1 clinical trials. Molecules developed during this project will in the future be taken forward to human studies. The NHS benefits from these studies, for example through Imperial Academic Health Sciences Centre funding. Pre-clinical studies are likely to begin following successful completion of this project.

3) Imperial College London:

We have filed an International (PCT) Patent Application (PCT/GB2017/051045) for biased GLP-1R agonists developed during pilot work for this project. Via Imperial Innovations, a subsidiary of Imperial College London, we will engage with the pharmaceutical industry regarding acquisition of intellectual property. We have a proven track record in this area, for example with the spin-off company Thiakis, which developed the peptide hormone oxyntomodulin as an anti-obesity agent and was subsequently acquired by Wyeth for $150 million.

4) The UK pharmaceutical industry and UK economy:

An estimated 422 million people have diabetes worldwide. Obesity-related comorbidities, such as non-alcoholic fatty liver disease, are at least as prevalent. There is therefore a potentially enormous market for GLP-1R agonists with improved therapeutic efficacy and tolerability, with clear economic benefits for the UK pharmaceutical industry and associated organisations. G.A.R. enjoys extensive interactions with the UK and European pharmaceutical industry as a Work Package leader in the former EU-EFPIA co-funded Innovative Medicines Initiative in Diabetes (IMIDIA) consortium, and presently in the "Rhapsody" consortium under IMI2. The latter involves close collaborations with Servier, Novo Nordsik, Janssen/J&J and Sanofi Aventis, particularly as part of an initiative to identify new circulating biomarkers for T2D development and progression.

5) The wider scientific community:

During this project we will explore a new technological approach to imaging of ligand binding and receptor trafficking in living islets implanted into the anterior chamber of the eye. This would for the first time allow these cellular events to be monitored at cellular resolution in an in vivo environment. Whilst we will focus on the GLP-1R, if this technique is successful, it is likely to be of significant interest to researchers working within the wider GPCR field. Results will be disseminated at scientific conferences and by publication in peer-reviewed journals.

6) The general public:

Our departments have a strong track record of public engagement. We welcome pre-university students into our laboratory for work experience as part of a drive to widen participation. We are involved in the Imperial College Outreach scheme, which organises science education events for school students and their teachers, introducing them to the science of T2D and obesity. We regularly participate in the Imperial Festival, a public engagement event that allows the public to explore and understand our latest research. We also recently contributed to the BBC Horizon programme, "Why Are We Getting So Fat?".