Compatibility rules for glycosaminoglycan-amyloid interactions

Lead Research Organisation: Lancaster University
Department Name: Chemistry


Several human diseases are associated with plaques which accumulate in tissue and are thought to contribute to organ damage. The most famous example of these so-called amyloidoses is Alzheimer's disease (AD), in which the plaques identified in post-mortem brains are localised at areas of neuronal degeneration. What is less widely known is that similar plaque deposits are found in many other diseased and healthy organs including, for example, the pancreases of individuals with type II ("late-onset") diabetes.

Amyloid was discovered over 150 years ago and the word - literally "starch-like" - was defined in recognition of the early belief that the plaques were made up of starchy carbohydrate sugars. It was later discovered that the amyloid plaques are actually deposits of fibrous proteins, but more recent evidence has vindicated the early medical pioneers and shown that amyloid is, in fact a complex mixture containing both proteins and carbohydrates called glycosaminoglycans - or GAGs. It is now known that GAGs can influence the rate at which amyloid proteins can accumulate and, importantly, can have a profound effect on amyloid toxicity.

In seeking to unravel the causes of amyloid disease, we are striving to understand how normally benign proteins of different chemical composition mis-assemble to form plaques. Thanks to sophisticated analytical techniques we now have detailed all-atom models of the multitude of interactions within amyloid fibrils/plaques. However, how GAGs alter or facilitate amyloid assembly and interact with amyloid fibrils is known, questions that are crucial to amyloid disease formation and propagation. Recent work by the applicants has produced the first experimental evidence unveiling the intimate interactions between a protein and a GAG in a fibrous amyloid deposit. This was made possible using NMR (nuclear magnetic resonance) spectroscopy, to provide atomic-level information of this crucial binding interaction.

Despite this important breakthrough, we are still desperately short of information on how different proteins and GAGs interact with each other in disease-related amyloid. One example is not enough. GAGs, like their amyloid fibril protein partners, are chemically variable and we do not know whether proteins and GAGs in one type of amyloid, such as in AD, interact in the same way as in another type of amyloid, such as in type 2 diabetes. In this application we propose to use NMR and other techniques to elucidate at the molecular level how a series of chemically different naturally occurring GAGs, and model GAGs of defined composition, interact with the two main proteins found in AD plaques and the main protein component of the pancreatic plaques in diabetes. We will ascertain whether each protein favours a particular type of GAG to form plaques, and whether there are general rules dictating the way in which the fibrils interact with GAGs. Furthermore, we will determine whether and how GAGs orchestrate the process of protein assembly, or whether they are just molecular "passengers".

Endeavours to investigate the molecular structures of proteins and other biomolecules have provided information that has been invaluable in the discovery and design of pharmaceuticals and healthcare products. In amyloid research there is an intensive international effort to develop drugs which prevent or alter the way in which proteins assemble into organ-damaging plaques. By revealing how GAGs interact with proteins as they assemble into plaques, and how they bind and stabilise the final fibrillar assembly, the results generated in this proposal will help us to design molecules that mimic these interactions for use as drugs or as agents to help diagnose disease. As many amyloid diseases affect the elderly, our results could have far-reaching consequences for the quality of life of millions of people and the burden on healthcare resources in an ageing population.

Technical Summary

Over 30 polypeptides self-assemble into insoluble amyloid fibrils associated with amyloid disease. As early as the 1850's it was found that amyloid deposits are associated with carbohydrates, later identified and classified as proteoglycans and glycosaminoglycan (GAG) polysaccharides. GAGs are now known to co-localise widely with amyloid assembled from a variety of protein precursors, and to influence fibril growth rate, fibril morphology, fibril stability and the pathogenic properties of the soluble pre-fibrillar intermediates. How the chemical composition of GAGs or their amyloid protein partners influence their molecular compatibility, however, remains unknown.

Capitalising on our exciting recent developments which have revealed the first atomic resolution structures of a fibril-GAG interaction, we propose here to determine the molecular determinants of fibril-GAG interactions using both natural and model GAGs and fibrils of Abeta1-40 and Abeta1-42 (associated with Alzheimer's disease), and amylin, the pancreatic islet amyloid polypeptide associated with type II diabetes. Using a combination of solid-state NMR and biochemical and biophysical techniques, we will determine how GAG-amyloid interactions are modulated by protein sequence, fibril architecture, and saccharide substitution patterns. We will establish the relative importance of different GAG family members as potential partners for interacting with amyloid fibrils and elucidate the structural consequences of binding at the detailed level of protein structure. Furthermore, the first details on how GAGs direct the early stages of amyloid assembly will be provided from structural measurements on assembly intermediates in the presence/absence of GAGs. A comprehensive atomistic dissection of these long-overlooked interactions will thus emerge, providing holistic insights into the amyloid assembly process and guidance for the design of therapeutic or diagnostic agents against amyloid assembly and disease.

Planned Impact

Scientific impact: Protein self-aggregation has immense fundamental, biotechnological and medical importance. This project will have wide scientific impact by addressing key questions in the amyloid field: how and why do GAGs bind widely to amyloid fibrils irrespective of sequence and structure; how does GAG binding influence the course of amyloid assembly and how can we harness this information for the development of amyloid diagnostics and therapeutics? This is a challenge of vital biological importance; the results will be of interest to a wide spectrum of scientists, pharmacists and medical researchers in academia and industry.

Industrial impact: Details of amyloid-GAG interactions offer immense benefits for the design of therapeutic strategies targeting several, if not all, amyloid disorders. Amyloid disease is predicted to be the major threat to human health in Western society in the next 50 years. However, the lack of understanding of the origins of molecular recognition between amyloid and GAGs has hindered the development of rationally designed anti-amyloid therapies and diagnostics. The long-term collaboration between the applicants will answer this key question, building on excellent equipment, exemplary infrastructure and a strong record of success in this area. We will also use our close contacts with pharmaceutical companies (e.g, GSK, UCB (Parkinson's), Lilley (Tauopathies), AstraZeneca (AD, Diabetes)) to expedite the design of new anti-amyloid agents based on the results obtained: DAM has already designed novel anti-amyloids by utilizing structural data.

Supporting knowledge and technology translation: DAM, EAY and SER speak regularly at scientific conferences and are well versed in knowledge exchange with academics and industrialists. We anticipate that the high impact of this work will merit spotlight articles in widely read magazines and journals. DAM's work developing amyloid inhibitors has been highlighted in ACS Chemical Biology, SciBX and F1000. SER's research into amyloid fibril toxicity has been highlighted on the front page of the BBSRC website (6.12.2009), and chosen as paper of the week in J. Biol. Chem., illustrating the importance of the research in this area and the pro-active nature of the researchers in promoting its research findings.
There are excellent opportunities to develop the results into therapeutic or diagnostic agents. Systems are in place for patenting and exploiting new materials through the technology transfer arm of the University of Liverpool, Business Gateway, and its business incubator, MerseyBio that provides accommodation for start-up biotechnology companies. ER has active collaborations with UCB, Medimmune and Lilley, which will ensure facile links into these organisations once the project bears fruit. SER is the ACSMB Director, with overall responsibility for linking with Industry though the University of Leeds Industry Hub in pharmaceuticals.

Delivering highly skilled people: The applicants will train two PDRAs in cutting edge technology and ensure their successful career development in academia or industry. Over the past few years, highly skilled post-graduate and post-doctoral fellows from the laboratories of DAM, EAY and SER have gone on to fellowships, lectureships and positions in industry.

Public engagement: This work will have a large impact on the general public, many of whom are affected directly or indirectly by amyloid disease. The Institute of Integrative Biology at Liverpool has an active Public Engagement agenda of substantial breadth, and DAM and EAY will join existing outreach and inreach programs to publicise this work, for example by visiting local colleges and at Institute Open Days. At Leeds, ACSMB members regularly organise workshops and science fairs to inspire school children to study science, student placements in research laboratories, talks at conferences for school teachers and authorship of articles for sixth-formers.
Description This research has focused on the molecular details of amyloid plaques found in the brains of individuals with Alzheimer's, and which accumulate in other organs and blood vessels. Amyloid is formed when certain protein molecules stick together to form insoluble fibres that are individually very small, but which can build up into large plaque-like deposits that clog up arteries and organs and increase the risk of organ failure and cardiovascular disease. Amyloid is also associated with the loss of of brain cells in Alzheimer's disease, although its precise role in this process is not clear.

Over the past two decades scientists have discovered much about how these certain proteins assemble into the fibrous deposits that accumulate in the brain and other human tissue. Using techniques such as nuclear magnetic resonance (NMR) spectroscopy, it has been possible to build up a picture of these fibrous amyloid deposits at the atomic level, and this information is helping to design new drugs that combat amyloid disease and new imaging agents that help to diagnose amyloid disease. Until recently, however, an important part of the picture has been missing. It has been known for some time that long sugar-like molecules called glycosaminoglycans (GAGs) are found together with amyloid plaques throughout the body, and we also know that these molecules can speed up the formation of amyloid plaques and prevent their removal by the body's natural defence mechanisms. Now, using NMR and molecular biology techniques, our work has revealed how these sugar molecules interact with the amyloid fibres. Our key findings so far have been as follows:

1. To show that glycosaminoglycan molecules interact strongly with the outer surface of Alzheimer's amyloid fibres to form a tight complex (Hughes, et al., J. Am. Chem. Soc., 2016). This information was gained using NMR spectroscopy together with a chemically-synthesised GAG containing nuclear isotope labels to "light up" the molecule when associated with the fibres. This information will help to identify agents that block this interaction and either prevent the fibrils from forming or help the body's defences to break up the fibres more rapidly.

2. We showed exactly which parts of the Alzheimer's amyloid protein interact with GAGs. By changing the amino acids in the protein one-by-one, we were able to identify which parts of the protein are recognized by the GAG molecule and which parts of the protein are not important (Stewart et al., J. Mol. Biol., 2017). This information again could help to identify therapeutic and diagnostic molecules.

3. We turned our attention to another protein, called apoA-I, which forms amyloid plaques in blood vessels and contributes to the risk of heart disease. Using similar methods to those described above, we discovered that a certain type of GAG called heparin, which is also a widely-used anticoagulant, speeds up the formation of apoA-I amyloid. However, when the amyloid-heparin fibres were treated with a molecule called EGCG, a major component of green tea, the fibres dissolved into much smaller structures (Townsend et al., J. Biol. Chem., 2018). These results attracted international press coverage because of the potential of a natural dietary product (green tea) to alleviate the risks associated with amyloid disease. However, much further work is required to substantiate these initial findings.

4. In the course of the work above, we developed new techniques that provide unprecedented insights into the interactions of therapeutic or diagnostic molecules with amyloid fibrils. For the first time, we were able to show exactly how a diagnostic molecule is orientated on the surface of Alzheimer's amyloid fibrils (Townsend et al., J. Phys. Chem. Lett., 2018). This work opens the door to a host of further applications to design and optimise diagnostic ligands for detecting amyloid disease.
5. Further work has been done to identify which parts of the GAG molecule interact with amyloid proteins, focussing on another protein called tau associated with Alzheimer's (Townsend et al., Biochemistry, 2020).

This project has been highly successful and has generated new and important insights into a class of human disease that is becoming more prevalent in an ageing society. Although our work is at a fundamental level, we expect the outcomes of this research to inform the search for new disease therapies and diagnostics.
Exploitation Route Glycosaminoglycan polysaccharides associate with amyloid ubiquitously. Our work has given unprecedented detail on the interaction between heparin-derived molecules and amyloid, which will be of value to medicinal chemists seeking to design heparin-derived inhibitors or diagnostic molecules for amyloid disorders.
Sectors Healthcare,Pharmaceuticals and Medical Biotechnology

Description Compatibility rules for glycosaminoglycan amyloid interactions 
Organisation University of Leeds
Country United Kingdom 
Sector Academic/University 
PI Contribution Structural analysis of glycosaminoglycan interactions with amyloid fibrils using solid-state NMR spectroscopy.
Collaborator Contribution Preparation of amyloid fibrils and characterising the binding of glycosaminoglycans using biochemical methods.
Impact Publications: