The role of biased agonism in the treatment of obesity

Almost one billion people are obese, including one third of the UK population. People who are obese are at high risk of developing diabetes, heart disease, joint problems and cancer. Obesity is an area of high priority for the MRC and for the NHS.

Many people find it difficult to lose weight naturally. The only truly effective treatment is weight loss surgery, such as the "gastric bypass" operation, but this can lead to complications and is not popular with all patients. As a result, there is enormous interest in producing medicines to help people lose weight. A group of medicines known as "glucagon-like peptide-1 receptor (GLP-1R) agonists" has recently been approved for weight loss. These mimic the effect of a natural hormone (GLP-1) by activating its target, or "receptor", in the brain to reduce hunger. Other hunger-reducing hormones, including "amylin", are also under investigation as potential weight loss treatments. However, none of these are as effective as gastric bypass surgery because the maximum dose is limited by nausea and/or vomiting.

I have recently generated new, improved GLP-1R agonists that behave differently to existing versions on the market as they activate GLP-1R in an unusual way, referred to as "biased agonism". Biased GLP-1R agonists appear very promising as they produce more weight loss but less nausea. Interestingly, they also work particularly well in combination with amylin-based treatments (amylin analogues), causing larger reductions in hunger even at low doses.

The overall aim of my research is to understand why biased GLP-1R agonists work so well for weight loss, both on their own and when combined with with amylin analogues. Fully understanding their mechanism of action will help develop this approach into a better treatment for obesity. I will address three main research questions:

1. Which neurons in the brain respond to biased GLP-1R agonists and amylin analogues - are they the same or different? This will be done by measuring the effect of different GLP-1 and amylin analogues on feeding behaviour and changes to brain tissue in mice.
2. How do neurons respond to biased GLP-1R agonists and amylin analogues - what is happening inside the neurons themselves that allows them to reduce hunger? To answer this, I will generate a genetically modified mouse in which the receptor itself has a fluorescent molecule attached, meaning individual receptors and how they behave can be directly observed using a microscope.
3. How do biased GLP-1R agonists and amylin analogues physically interact with their receptors, and how is this affected by natural genetic variations in receptor structure found in different people? To do this I will develop a new technique called "deep mutational scanning", which allows thousands of receptor variations to be tested in in a single experiment, which is much faster than the normal method based on testing each variation individually.

This work will be done at Imperial College London but involve collaborations with other researchers in the UK and internationally. The results will be an important step to developing a new weight loss treatment, and will also show how genetic differences can influence how well the drugs work, meaning that in the future we may be able to individualise treatments based on genetic testing.

The overall design of the experiments should also be applicable to other hormone targets, meaning it could be used to develop other obesity treatments in the future.

Almost one billion people worldwide are obese. The leading pharmacotherapy approved for obesity, Wegovy (Semaglutide), is based on the hormone glucagon-like peptide-1 (GLP-1) and suppresses appetite to produce 12% weight loss on average. However, health benefits of weight loss continue well above this point, meaning that many people taking Wegovy would benefit from a better weight loss treatment.

The effectiveness of GLP-1 receptor agonists (GLP-1RAs) is limited by 1) nausea at high doses and 2) GLP-1R desensitisation due to recruitment of beta arrestins. My preliminary data show that redesigning GLP-1RAs to abolish arrestin recruitment allows the receptor to signal for longer, increasing weight loss. At the same time, moderately reducing G protein signalling limits nausea. Interestingly, these ligands (G protein-biased, partial GLP-1RAs) also strongly potentiate the effects of other anorectic agents e.g. the dual amylin-calcitonin receptor agonist (DACRA) Cagrilintide. The mechanisms underpinning these effects are poorly understood.

I will address 3 questions:

1. Which neuronal populations respond to biased GLP-1RAs, and do synergistic effects with DACRAs arise from signalling crosstalk in the same cells?
2. What are the receptor behaviours and signalling profiles of biased GLP-1RAs in neurons, and how do these influence DACRA responses?
3. How does receptor coding variation influence GLP-1RA and DACRA responses?

Aims 1 and 2 will use mouse models to target neurons with the relevant receptor for gene knockdown or marker expression, and will include metabolic testing, NuTRAP transcriptomics, and live cell imaging of receptor dynamics and signalling. Aim 3 will use deep mutational scanning to determine the effects of thousands of GLP-1R/CALCR variants in parallel.

These studies will 1) reveal how biased GLP-1RAs enhance weight loss in combination with DACRAs, and 2) develop new approaches that can be applied to other receptors in the future.