Dr. Rafael BrüschweilerGeorge Mathew Edgar Professor
Ph.D. (1991) ETH Zurich, Switzerland
Research InterestStructural Dynamics of Proteins
A protein's function is the result of the subtle interplay of protein structure and dynamics and the interaction with other proteins, nucleic acids, or various ligands. The understanding of these processes at atomic detail lies at the interface of Chemistry, Biology, Physics, and Computer Science. To elucidate their nature our group uses the "four pillars" of modern biophysical chemistry: experiment, theory, computer simulation, and bioinformatics.
On the experimental side, we employ nuclear magnetic resonance spectroscopy (NMR), which is a very powerful tool that measures the strengths, directions, and temporary fluctuations of magnetic interactions at the location of essentially each atom in the protein by means of chemical shifts, spin-spin couplings, and spin relaxation times, respectively. This information, in combination with suitable models, provides unique insight into the structural dynamics of the protein on widely different time scales ranging from seconds to tens of picoseconds. We are developing new protocols for the comprehensive analysis of experimental data using computational and database derived models and apply them to different protein systems. The emerging view of protein dynamics often (but not always) involves motional correlation effects across the protein and thereby also provides access to the protein's conformational entropy, which is one of the dominant driving forces of protein function.
The availability of large amounts of protein sequence data represents a formidable challenge for the structural biology community in terms of determining the corresponding three-dimensional (3D) protein structures by X-ray crystallography or by NMR spectroscopy. In order to use structures that have already been solved by X-ray crystallography in subsequent NMR applications to study ligand binding, protein-DNA interactions, and dynamics, the assignment of nuclear spin resonances to individual atoms is a prerequisite.
Because the assignment is one of the most time consuming and labor intensive steps, even when the 3D protein structure is available, we are developing new strategies based on weighted matching algorithms to speed-up this process that make optimal use of the crystal structure, residual dipolar spin-spin couplings, and chemical shifts. In this way, the complementary strengths of NMR, crystallography, and quantum chemistry can be synergetically used. Related research in our group aims at the rapid protein fold recognition or fold determination using a minimal set of NMR parameters including dipolar spin-spin couplings and chemical shifts in combination with protein data bank (PDB) data mining approaches.
Covariance NMR Spectroscopy
To shorten the NMR measurement time for multi-dimensional NMR spectra and to facilitate their analysis and interpretation, we are developing Covariance NMR Spectroscopy, which represents an NMR spectrum in terms of a covariance matrix calculated from a selected set of one-dimensional NMR spectra. The covariance spectrum can then be subjected to principal component analysis (PCA) that directly provides dominant correlation information. Application to chemical mixtures, for example, allows the easy identification of individual chemical components, which may become useful in the emerging biomedical field of metabolomics.
Quantum Information Processing
Nuclear spins can be employed as quantum-mechanical bits (qubits) to store information and to efficiently perform certain computational tasks. Because NMR spectroscopy operates on very large ensembles of identical molecules, such computations can be performed in parallel by using molecules with different spin-state configurations to represent different data inputs. It can be shown that for an important class of search problems an exponential speed-up is possible over a single-processor computer provided that the available molecular ensemble is sufficiently large. We are developing new applications and practical implementations of this NMR ensemble computing paradigm.
|Long D, Brüschweiler R. Atomistic Kinetic Model for Population Shift and Allostery in Biomolecules. J Am Chem Soc. 2011, Nov 7. [Epub ahead of print]|
|Robinette SL, Brüschweiler R, Schroeder FC, Edison AS. NMR in Metabolomics and Natural Products Research: Two Sides of the Same Coin. Acc Chem Res. 2011, Sep 2. [Epub ahead of print]|
|Brüschweiler R. Protein dynamics: whispering within. Nat Chem. 2011, Aug 23;3(9), 665-6.|
|Bingol K, Brüschweiler R. Deconvolution of chemical mixtures with high complexity by NMR consensus trace clustering. Anal Chem. 2011, Oct 1;83(19), 7412-7|
|Long D, Li DW, Walter KF, Griesinger C, Brüschweiler R. Toward a predictive understanding of slow methyl group dynamics in proteins. Biophys J. 2011, Aug 17;101(4), 910-5|