Skip to content

Back

Robert Best

Dr Robert Best

Dr Robert Best

Research Fellow

Interests and expertise (Subject groups)

Grants awarded

Understanding biomolecular dynamics using molecular simulations

Scheme: University Research Fellowship

Organisation: University of Cambridge

Dates: Oct 2012-Sep 2012

Value: £296,100.04

Summary: This project summary is not available for publication.

Understanding biomolecular dynamics using molecular simulations

Scheme: University Research Fellowship

Organisation: University of Cambridge

Dates: Oct 2007-Sep 2012

Value: £493,807.20

Summary: I have highlighted here one particular aspect of my work: Usually, proteins are required to fold to a very specific three-dimensional structure in order to perform their function. Occasionally, however, proteins 'misfold' to form incorrect structures. These misfolded species in many cases are toxic to cells, because they may lead to the formation of large-scale protein aggregates. One may expect this to be a particular problem in "multidomain proteins", i.e. proteins in which two very similar globular domains are expressed side-by-side. The proximity of similar structures and sequences would appear to favour aggregation. In recent work, we have investigated protein misfolding in multidomain proteins in the context of a model system. Starting with a case where two adjacent domains are identical (both titin I27 domains), our experimental collaborators have demonstrated the presence of a misfolded species. However, from their single molecule experiments alone, they were unable to determine the structure of the misfolded state - if any. By using coarse-grained molecular models, with a fairly simple set of assumptions, we were able to predict the misfolded structure. Our prediction explains both the lifetime of the misfolded species and the structural properties monitored in single-molecule FRET experiments, as well as observations in single molecule pulling experiments with the atomic force microscope. The predictions of the simulations were further validated by studies on a construct consisting of three, rather than two, adjacent identical domains. Finally, by studying domains which are not identical, we show that such misfolding can occur even with just 42% sequence identity between adjacent domains. This work should serve as a basis for future quantitative understanding of misfolding of adjacent domains in multidomain proteins.

Was this page useful?
Thank you for your feedback
Thank you for your feedback. Please help us improve this page by taking our short survey.