A new method for simultaneously determining the structure of a protein and the mobility and range of motion of its backbone and side chains has been developed by a British research team.
Techniques for studying protein structure and dynamics are already well developed but are generally carried out independently. The new method, dynamic ensemble refinement (DER), marries these techniques much more successfully than has been achieved in previous attempts to do so. DER could aid drug design by making it feasible to study the way small molecules interact with an ensemble of protein conformations that exist in solution—rather than with just a static, averaged structure.
DER combines molecular dynamics and two nuclear magnetic resonance spectrometry (NMR) techniques. Molecular dynamics is a theoretical technique that predicts molecular motions in proteins. NMR order parameters are an experimental measure of the possible range of such motions–that is, the extent to which side chains and other structures flop around in native (folded) proteins. And NMR nuclear Overhauser effects (NOEs) define the average conformations of native proteins. The dynamics and structure of native proteins are usually determined separately, however, making it difficult to accurately assess the entire distribution of conformations that folded proteins adopt in solution.
Research fellow Michele Vendruscolo, professor of chemical and structural biology Christopher M. Dobson, and coworkers at the University of Cambridge have now resolved this dilemma by developing DER, in which information on a native protein’s dynamics and structure are combined synergistically to map the protein's full range of conformational motion [Nature, 433, 128 (2005)].
The technique uses experimental NMR order-parameter and NOE data to rein in the molecular dynamics simulations and constrain them to reality. "It requires that an ensemble of protein conformations generated by molecular dynamics simulations is compatible with structural information, such as NMR NOEs, and also with dynamical data, such as NMR order parameters," Dobson says. The result is an ensemble of protein conformations that represents the structural and dynamic variability of the native protein.
DER could be extended to incorporate other types of experimental measurements, including NMR chemical shifts and residual dipolar couplings and X-ray diffraction data, the researchers note. The NMR order parameters used in the Nature study restricted the results to molecular motions occurring on a picosecond-to-nanosecond timescale, but using other types of measurements could enable DER to define protein motions occurring on longer timescales as well.
Chemistry professor J. Andrew McCammon of the University of California, San Diego, comments that the work"certainly represents a significant advance in structural biology. It confirms that the relatively large amplitude motions seen in molecular dynamics simulations of proteins are correct, while providing a way to improve such simulations to some degree by using experimental data to reduce errors."
NMR spectroscopist Ad Bax of the National Institute of Diabetes & Digestive & Kidney Diseases says the study "appears to represent a very significant breakthrough in that it finally presents a method that gives a quantitative meaning to the distribution of the set of NMR structures&" populated by a native protein. "This has been some sort of grail pursued by many in the past."
Indeed, University of Toronto chemistry professor Lewis E. Kay says his group and others have been trying for some time "to build a bridge between NMR experiment and site-specific contributions to thermodynamics,"but the path hasn't been clear. DER" takes a substantial step forward" toward creating that link, Kay says," and that's the real beauty of the approach."
"This is a pivotal piece of work, opening up new opportunities and insights for NMR, dynamics, protein structure, and folding,"adds biochemistry professor Jane S. Richardson of Duke University.