47
Modeling of Biomolecular Systems Interactions, Dynamics, and Allostery Session IX Abstracts
Restraining Molecular Dynamics and Modeling with Ion-Mobility Mass Spectrometry
Erik Marklund
, Matteo Degiacomi, Carol Robinson, Andrew Baldwin, Justin Benesch.
University of Oxford, Oxford, United Kingdom.
The dynamical variability of protein assemblies is a major stumbling block for structural biology
and biophysical tools that provide only an ensemble average. Recent developments in ion
mobility coupled to mass spectrometry (IM-MS) have rendered it an attractive new approach for
studying the structure and dynamics of biomolecular complexes.
From drift times measured by IM-MS the collisional cross sections (CCS) of proteins can be
inferred, a single measurement reporting on all mass-separated species in a sample, such as
multiple oligomeric states of polydisperse proteins. Since IM-MS separation takes place within
milliseconds, the observed CCS distribution accurately represents the solution conditions.
Furthermore, with the ability to perform time-resolved experiments, IM-MS is well suited for
studying protein interactions and dynamics. As shown in this study, CCS hold structural
information that is distinct from other geometric parameters of proteins, such as gyration radii.
Making use of IM-MS data however requires computational modeling, for example comparison
of CCS from candidate structures with those observed in experiments. This approach has been
limited by the demanding calculation of macromolecular CCS. To overcome this limitation we
have developed a new algorithm (IMPACT) for estimating CCS of structure models at
unprecedented speed, with potential to transform the IM-MS-based modeling and for the first
time allowing for directly restraining molecular dynamics simulations with IM-MS data. We
have analyzed conformational ensembles and show how modern IM instrumentation is capable
of assessing not only the structure of proteins, but also their conformational dynamics.
Our analysis of all biological assemblies in the Protein Data Bank and other structural databases
furthermore reveals how protein complexes of similar mass frequently have sufficiently different
cross sections to be distinguishable by IM, which translates to proteome-wide applicability of
IM-MS for the study of biomolecular structure and dynamics.
