Ph.D. University of Sheffield (UK), 1980
Our research interests are in protein conformational dynamics, protein-solvent interactions, and the characterization of dynamically distinct substructures in proteins and their role in protein function and stability. We employ a number of techniques to explore protein dynamics including 2H NMR, and hydrogen isotope exchange measured by 1H 2D COSY NMR and Electrospray Ionization Mass Spectrometry (ESI-MS). A second area of interest is the development of protein analytical methods including the development of novel chromatographic stationary phases for the separation of peptides, proteins, and metabolites, as well as the development of proteomics approaches to monitor protein chemical modification, oxidation and changes in protein stability. We also conduct proteomics studies of Multiple Sclerosis. Proteomics approaches include Surface Enhanced Laser Desorption and Ionization Mass Spectrometry (SELDI-MS), 2D HPLC, peptide mass fingerprint analysis, and various electrophoresis techniques including western blotting.
Protein molecules are not rigid but undergo a variety of internal motions with amplitudes that range from a few hundredths of Angstroms to tens of Angstroms and over time scales that vary from picoseconds to seconds. The types of motion include high frequency, small amplitude vibrations of bonded atoms, rotations of amino acid side-chains, as well as a rich spectrum of slower, more collective rigid body motions, including helix and loop displacements and motions of protein domains.
Protein conformational dynamics is important for a number of protein functions including ligand binding, enzyme catalysis and regulation, signal transduction, and the interaction of proteins with other macromolecules. The dynamic properties of proteins are also important for the protein folding process itself as well as for protein stability.
The dynamic properties of proteins are strikingly similar to the dynamic properties of many other complex systems, including glass forming liquids and synthetic polymers. For example, hydrated proteins display a dynamical transition at ~220 K and like synthetic polyamides are plasticized by water.
Figure 1: Location of the slowest exchanging proteins (blue and green) in lysozyme determined from a 1H 2D COSY NMR spectrum of the protein following kinetic labeling by deuterium-hydrogen exchange. The spectrum was obtained on a 500 MHz Varian NMR spectrometer by the method of Redfield et al [Biochemistry, 27, 122, (1988)].
Globular proteins appear to be dynamically heterogeneous, consisting of rigid, solid-like regions embedded in a more mobile liquid-like matrix. These dynamically distinct regions were first identified in hydrogen isotope exchange experiments and we continue to characterize them and explore their role in protein function and stability.
We are currently studying the dynamic heterogeneity of the protein interior by a combination of hydrogen isotope exchange and deuterium NMR spectroscopy in collaboration with Dr. Mahinda Gangoda. Protein samples are kinetically labeled by hydrogen isotope exchange so as to select regions of the protein interior with different hydrogen isotope exchange rates. The location of deuterium labels within the protein structure and the hydrogen/deuterium occupancy of each site are established by 1H 2D COSY NMR (Figure 1). We then acquire 2H NMR powder spectra using a quadrupole echo pulse sequence and employ the Lipari-Szabo model-free formalism to analyze the spin lattice relaxation times (Lipari, G.; Szabo, A. J. Am. Chem. Soc. 104, 4546, 1982).
Spin-lattice relaxation times of dry partially 2H labeled lysozyme are found to correlate with the average solution exchange rate of the labeled sites, which presumably reflects the fact that both exchange rates and motional correlation times depend on the same intrinsic features of the protein structure, such as local packing density and the number of residue contacts about the N-2H sites. These initial studies with lysozyme are being extended to other enzymes (ribonuclease A and adenylate kinase), where our interest is in exploring how protein conformational dynamic properties are perturbed by ligand binding and as the enzyme cycles through its catalytic reactions.
Figure 2:B-factor plot for sialidase (3sil.pdb). Upper plot: Contributions to the calculated B-factors from the TLS isotropic translation term (black), the TLS libration-screw terms (blue), and the Gaussian Network model (red) as well as corrections for lattice contacts (green). Lower Plot: Experimental (blue) and calculated (red) B-factors and residuals (green) (corr. coeff. 0.83).
Another source of information about protein flexibility is the Debye-Waller factor or B-factor determined by X-ray crystallography. The isotropic atomic B-factor is given by:
Bi (82/3)< -Ri2>
where < -Ri2> is the atomic mean square displacement of the ith atom. We are interested in modeling the atomic B-factors and in understanding the various contributions to B-factors from internal protein motions, rigid-body motions of protein molecules in the crystal, as well as the effects of crystal lattice contacts on B-factors. We are currently developing models (Figure 2) based on the Gaussian Network Model (GNM) [Haliloglu et al Phys. Rev. Lett., 79, 3090, (1997)] and Local Density Model (LDM) [Halle, PNAS, 99, 1274, (2002)] to account for internal motions in combination with the TLS model [Schomaker & Trueblood, Acta. Cryst., B24, 63, (1968)] to account for rigid-body motions of the protein within the crystal lattice.
A second area of interest is the development of protein analytical methods including the development of novel chromatographic stationary phases for the separation of peptides, proteins, and metabolites, such as 4-propylaminomethyl benzoic acid bonded silica (Figure 3), the development of multi-functional chromatographic stationary phases based on a reactive aldehyde silica platform, as well as the development of proteomics approaches to monitor protein chemical modification, particularly oxidation and changes in protein stability.
We employ SELDI-MS, 2D-LC and gel electrophoresis techniques in proteomics studies of Multiple Sclerosis (MS) and Experimental Autoimmune Encephalomyelitis (EAE) in collaboration with Dr. Jennifer McDonough (Dept. of Biological Sciences) and Dr. Ernie Freeman (Dept. of Biological Sciences). Oxidative damage plays an important role in a number of neurodegenerative diseases and appears to be important in MS where it may lead to mitochondrial dysfunction. We are attempting to identify proteins that are differentially expressed or chemically modified in MS and EAE brain tissue relative to controls. We have developed antibody pull-down methods coupled to SELDI-MS to identify protein peaks in the mass spectra and to identify oxidized proteins. These approaches are more rapid and convenient than current western blotting techniques.
Figure 3:4-propylaminomethyl benzoic acid (PAMBA) bonded silica - a versatile zwiterionic HPLC stationary phase which operates with high efficiency in cationic and anionic separation modes shown here separating (1) histidine, (2) lysine, and (3) arginine.