Researchers at the Institute for Bioengineering of Catalonia (IBEC) have produced a mutational map showing how mutations in amylin — a hormone that plays a key role in glucose regulation — affect its tendency to form toxic amyloid aggregates in the pancreas. This process is linked to the development of type 2 diabetes. While it was already known that certain mutations could alter this aggregation capacity, understanding of this process was fragmented and based on isolated studies. "For the first time, we can systematically map how thousands of mutations modulate amylin aggregation, bringing human genetics closer to molecular mechanisms," says Benedetta Bolognesi, the principal investigator of the Protein Phase Transitions in Health and Disease group at IBEC, who is also the lead author of the study.
'We have created a map that allows us to anticipate the potential impact of these mutations in the population,' adds Marta Badia, a researcher in the same group and first author of the study. 'We are not assessing toxicity, but rather the protein's intrinsic propensity to form fibres. This is a first step, but an extremely necessary one."
Amyloid aggregates are small clusters of proteins that clump together abnormally, damaging organs and tissues in the process. The first amyloids to be described in human tissue, over a century ago, were amylin aggregates, also known as IAPP. These deposits are now known to damage the beta cells of the pancreas and contribute to the development of type 2 diabetes, a disease that affects more than 100 million people.
Although amylin is very similar across mammals, small differences in its sequence mean that species such as mice or bears do not form amyloids or develop diabetes. In fact, introducing the human version of the protein into a mouse generates pancreatic amyloids and eventually causes the disease to develop.
In order to gain a comprehensive understanding of how the amylin sequence influences its aggregation, the IBEC team analysed 1,916 variants of the protein using a technique known as deep mutational scanning. This methodology involves generating and measuring the behaviour of thousands of slightly different versions of the same protein all at once. This kind of 'mass screening' allows researchers to observe with great precision which mutations promote or hinder amyloid formation. The group had previously used this strategy to map the entire mutational landscape of the Aβ42 peptide, which is implicated in Alzheimer's disease, as well as other proteins involved in neurodegeneration. Using the same approach, the team has now created a comprehensive atlas of amylin mutations.
"Most people are familiar with amyloids because of their link to Alzheimer's disease and other neurodegenerative conditions, but they actually occur in many other diseases too. Our strategy here is very similar to the one we used with Aβ42: studying how mutations affect amyloid formation. Only this time, we're applying it to diabetes," explains Badia.
When certain mutations accelerate a process that is difficult to predict
The study's findings reveal that there is a particularly sensitive region in the amylin sequence where most mutations prevent the formation of fibrils. However, the analysis also identified mutations that accelerate aggregation, a phenomenon which is difficult to predict using conventional methods and that could have clinical relevance, as a faster rate of aggregation might contribute to the early onset of pancreatic damage.
To evaluate this in the context of human health, the team compared their mutational map results with genetic and clinical data from 500,000 individuals collected by the UK Biobank. Although the sample size remains small, they found that variants that reduce aggregation are more prevalent in individuals without diabetes. Badia adds: 'Even so, it is important to make it clear that diabetes is multifactorial and these mutations are just one piece of the puzzle.'
The study also incorporates evolutionary information by comparing human amylin with that of different species. The sequences of animals such as mice and polar bears, which do not develop diabetes despite their extreme diets or prolonged periods of hibernation, are inherently less amyloidogenic. This is a well-known fact, and the new atlas reproduces it with great precision. 'These animals have exactly the changes needed to prevent the protein from forming fibres. Knowing which switches to flip is key to designing new drugs,' explains Bolognesi.
As well as improving our understanding of aggregation mechanisms, the mutational atlas opens up new therapeutic opportunities. It enables the identification of mutations that reduce aggregation, provides a rational basis for designing new amylin analogues and helps anticipate the impact of genetic variants that may emerge in the population as genetic sequencing becomes more widespread.
"Our aim is for this map to serve as a tool for the scientific and clinical communities," Bolognesi concludes.
