Molecular basis of Amyloid Disease : In silico modelling of Beta-Amyloid (Aβ) aggregation

Khurshid, Beenish and Anwar, Jamshed (2020) Molecular basis of Amyloid Disease : In silico modelling of Beta-Amyloid (Aβ) aggregation. PhD thesis, Lancaster University.

Text (2020beenishPhd)
2020beenishPhd.pdf - Published Version
Available under License Creative Commons Attribution-NonCommercial-NoDerivs.
Download (20MB)

Abstract

Whilst the life expectancy in the world is increasing, there is an epic increase in the proportion of people with Alzheimer's disease (AD). Statistics show that over 35 million people all over the world and 0.8 million in the United Kingdom living with AD. his number is anticipated to double by 2030. AD is characterized by an immense loss of neurons leading to a loss of cognitive functions. Despite much effort and advances in scientific research and technology, we still don't know the exact mechanism of AD and as a result, there is still no cure for this detrimental disease. According to the Amyloid hypothesis, AD is caused by the aggregation of neurotoxic Aβ which is cleaved from the C-terminal of a very large protein, the amyloid precursor protein (APP). A molecular-level understanding is critical to understanding the development and progression of AD. Whilst molecular-level experimental techniques such as NMR spectroscopy, X-ray crystallography, or atomic force microscopy have advanced the field considerably, the early stages of Aβ aggregation largely remain inaccessible by experiment. An effective alternative approach is the use of molecular dynamics simulation, which has been employed here. The thesis aims to advance our molecular-level understanding of Aβ aggregation and how it is modulated by selected endogenous and exogenous chemical co-factors. There are 4 aspects to the study: (i) a systematic comparison of the popular protein force fields to see whether they could reproduce various hierarchical structures of Aβ; (ii) selfassembly of Aβ using both an all-atom and coarse-grained (CG) approach; (iii) interaction of the endogenous glycosaminoglycan heparan sulfate with Aβ; and (iv) interaction of curcumin (a predominant component of the spice turmeric) with Aβ –curcumin is an inhibitor of Aβ aggregation and also has potential for being a diagnostic for detecting amyloid. The systematic comparison of protein force fields to see whether they could reproduce various hierarchical structures of Aβ revealed that the choice depends on the size and complexity of the Aβ structure and the nature of the problem being studied. For example, the GROMOS family of force fields is known and was demonstrated here to stabilize β-sheets. So, for studying α-helix to β-sheet conversion during Aβ oligomerization, GROMOS would be the preferred choice. Similarly, for larger more complex systems, our study shows that OPLS-aa is a better choice as it achieves a good balance between α-helix and β-sheet secondary structures and gives significantly less deviation (RMSD) for the larger Aβ structures. Heparan sulfate is endogenous and is known to co-exist within Aβ deposits and can enhance fibril formation, suggesting a pathological connection. Here we have shown that it has a strong interaction with Aβ and indeed forms a composite structure with Aβ. Its flexibility is considered to be essential and we have also rationalized its molecular weight and concentration effects. Curcumin is a natural fluorescent dye and has a potential for serving as a non-invasive, highly specific diagnostic agent, along with its possible therapeutic role of inhibiting Aβ aggregation. The simulations explain the nature of the high specificity of curcumin for Aβ including the variation in contrast between the various Aβ morphologies.