Molecular dynamics simulations of the mechanical unfolding of proteins
Abstract
A number of proteins perform load-bearing functions in living organisms and
often have unique mechanical properties. In recent years, there has been considerable
effort to understand the relationships between the molecular structure of such proteins
and their mechanical response. Several of them have been studied in great detail through
single molecule mechanical pulling experiments. Interpretation of these experiments
requires the use of atomistic simulations. However typical simulation time scales are
many orders of magnitude shorter than relevant experimental and/or physiological time
scales. In this dissertation, we have developed a simulation methodology that provides a
direct link between experiments and simulations and is capable of predicting the outcome
of single molecule pulling experiments. By using this methodology, we have been able to
understand the relationships between the molecular structure and the mechanical
properties of a number of proteins. I report on our studies of the mechanical unfolding of
the I27 domain of the muscle protein titin, ubiquitin, and protein G and compare them
with the existing experimental data. The distribution of the unfolding force as well as its
dependence on the pulling rate predicted by our simulations is found to be in good
agreement with AFM experiments. We demonstrate that the mechanical unfolding
pathway can be altered by changing the pulling geometry and that the presence of a
hydrogen bonded clamp between terminal parallel strands of these domains is the key
property that is responsible for their high mechanical stability. We have also extended our
studies of single protein domains to protein dimers. Our replica-exchange molecular
dynamics simulation study of the mechanical unfolding of a segment-swapped protein G
dimer suggests that the mechanical resistance of a protein complex may be controlled not
only by the mechanical stability of individual domains but also by the inter-chain
interactions between domains.