Application of adiabatic fast passage pulses for removal of systematic errors associated with multiple spin-echo sequnces is demonstrated.The adiabatic fast passage pulses faciliate minimization of cumultative pulse errors for all three components of magnetization. It is also shown that off-resonance effects that degrade image quality in mangnetic resonance imaging and introduce systematic errors in measured T2 relaxation time peak amplitudes can be suppressed without any degradation of the overall signal intensity compared to a CMPG sequence built from rectangular hard power pulses. The technique has been tested on the 15N spin-spin relaxation time measurements of p19(INK4d).
Recent advances in heteronuclear NMR with isotope-labeled proteins have allowed relaxation parameters to be measured and dynamics parameters to be calculated for almost all of the backbone NH groups in a protein. We have established the Indiana Dynamics Database (IDD), a database containing these measurements and parameters for many proteins. The IDD is accessible via the World Wide Web (http://pooh.chem.indiana.edu/IDD). The purpose of the database is (1) to serve as a single location for electronic exchange of relaxation and dynamics data among NMR researchers; (2) to facilitate statistical analyses of dynamics data to allow the elucidation of structure-dynamics relationships; and (3) to facilitate the comparison and testing of methods for dynamics data analyses. An overview of the database organization, data submission procedure, search and retrieval tools, and available structural information will be presented.
The ets family of transcription factors regulate gene expression during growth and development through binding to a consensus DNA sequence as a monomer. Using an expressed 94-amino-acid fragment which covers the conserved 85 residues among the family, (short PU.1 or sPU.1), the complete backbone resonances were assigned and the backbone dynamics of the protein were characterized at two concentrations by NMR spectroscopy. 15N relaxation and heteronuclear NOEs were determined using the uniformly 15N-labeled sPU.1 and the dynamic behavior of sPU.1 at rapid time scales was examined according to the laboratory frame relaxation of 15N in combination with Lipari and Szabo's modelfree analysis. The order parameters (S2), internal correlation times(te), and chemical exchage rates (Rex) for each backbone N-H bond as well as the overall rotational correlation time (tm) were derived from the modelfree refinements for samples at two very different concentrations (2.5 mM and 0.3 mM). The DNA-binding loops between helix 2 and helix 3 and between helix 3 and beta-strand 3 present high mobilities when compared to the non DNA-binding portions of the protein. The sample in high concentration showed higher overall rotational correlation time (7.4 ns for the protein with molecular weight of 11 kD), which may indicate that at high concentration, sPU.1 is not monomeric.
Human ubiquitin-conjugating enzyme ,hUBC9, plays an important role in cell differentiation, cell cycle regulation and reponse to stress. In order to better understand the mechanism of action of UBC9 at the molecular level, we have carried out NMR studies of the internal motion of this protein. Backbone resonances have been assigned using 15N labeled, 13C/15N doubly labeled and 2H/13C/15N triply labeled protein samples. Dynamics of the protein has been studied by proton-deuterium exchange and 15N relaxation measurements.
The aspartic protease of the HIV-1 virus is a homodimeric enzyme, containing 99 amino acid residues in each monomer. The protease binds the cyclic diol inhibitor, DMP323 with nanomolar affinity. Previous studies of the structure and dynamics of the HIV-1 protease complex with DMP323 indicated that its flexibility plays a significant role in function (Wlodawer & Erickson, A. Rev. Biochem., 62, 543-585, 1993; Collins et al., Nature Struct. Biol., 2, 334-338, 1995; Yamazaki et al., Protein Sci., 5, 495-506, 1996). Especially, previous 15N rotating frame relaxation measurements indicated the presence of ms-microsec (exchange type) motions at numerous amide sites in the protease/DMP323 complex (Nicholson et al., Nature Struc. Biol., 2, 274-280, 1995). However, the relaxation rate was not observed to depend on the strength of the spin locking field. In order to characterize the slow internal motions in more detail, we have measured nitrogen-15 and proton relaxation in the rotating frame, T1rho, using N15/H2 labeled HIV-1 protease complexed with DMP323 at varying temperatures and spin locking field strengths. Deuterium (H2) labeling (ca. 85%) signficantly reduces proton-proton dipolar interactions of backbone amide and makes the proton relaxation rates more sensitive to other relaxation mechanisms such as chemical exchange. We discuss the complimentary use of 15N and 1H relaxation measurements to probe protein dynamics as well as some puzzling aspects of the proton relaxation measurements.
The magnetic field dependence of the nuclear spin-lattice relaxation rate, also called the nuclear magnetic relaxation dispersion (MRD), provides a direct characterization of the Fourier transform of the correlation function that describes the fluctuations that drive magnetic relaxation. Therefore, the MRD profile is a very powerful approach to characterizing both inter and intra molecular dynamics. We report MRD profiles for simple ionic solutes in aqueous phases containing proteins and phospholipid vesicles which permit characterization of the binding lifetimes of the ionic species and the surface translational diffusion coefficient for the ion adjacent to the biomolecular surfaces. The dipole-dipole spin relaxation induced in lithium ion by a symmetric paramagnetic center such as manganese(II) ion reflects the character of the transient complex formation between these solutes. The shape of the lithium ion relaxation profiles suggest complexes between aquolithium ion and hexaaquomanganese(II) ion with lifetimes as long or longer than 100 ps.
The magnetic field dependence of the water-proton spin-lattice relaxation rate provides a characterization of the water molecule binding to the protein. The proton experiment may be complicated by proton exchange of labile protein protons, cross-relaxation between the protein protons and the water or solvent protons, and exchange of water or solvent molecules between specific bound environments and the bulk solvent. We have investigated a series of proteins with different numbers of potential water binding cavities identified by the x-ray structure to explore what correlations there are between the water-proton relaxation efficiency and the number of potential water binding cavities in the protein. Although the relaxation may be complicated by protein aggregation at the high protein concentrations used, the results do not support a trivial relationship between apparent availability of binding sites and the relaxation efficiency. We discuss the high field tail of the relaxation dispersion profile in terms of the distribution of water molecule lifetimes on the protein.
Self-diffusion of proteins is of interest both because it is required for molecular transport in biological systems and also as a measure of molecular size and shape. NMR can be used to make quick and accurate measurements of self-diffusion and offers also the possibility to make multicomponent measurements of single system. It is possible to use self-diffusion measurements to study changes in the aggregated state, conformation and the state of hydration of proteins. Whereas aggregation has received some study previously, little work has been carried out concerning the use of self-diffusion to determine the hydration state of a protein. Water plays a vital role in the function of proteins and a method for determining the amount of the hydration offers interesting possibilities. These include measuring the change in hydration as a function of pH, and thus the titration of charged groups. If, as is common in proteins, there is a charged group within the active site, this could titrate at a different pH compared to non-active site residues of the same type. We may, therefore, be able to measure the hydration of a single amino acid in a protein. The amount of hydrated water can be calculated because bound water will diffuse much slower than the bulk water. The measured self-diffusion coefficient will be a weighted average of the self-diffusion coefficients of the bulk and bound water. We have used our technique to measured the amount of hydration of lysozyme as a function of pH. In practice the changes observed in the self-diffusion coefficients of the protein and the water may depend on changes in aggregated states, conformation as well as hydration state. We have considered all these effects whilst measuring the change in hydration of lysozyme with titration of charged groups and can quantify changes in hydration and conformation.
The backbone dynamics of the B1 IgG-binding domain of streptococcal protein G have been characterized at 6 different temperatures using inverse-detected two-dimensional 1H-15N NMR spectroscopy. Longitudinal (T1) and transverse (T2) 15N relaxation time constants and steady-state {1H}-15N NOEs were measured, at a spectrometer proton frequency of 500 MHz, for 55 of 56 protonated backbone nitrogens. The relaxation parameters were analyzed using a model-free formalism that characterizes the dynamics of the N-H bond vectors in terms of generalized order parameters and effective correlation times. The temperature dependence of backbone motions is studied to elucidate its relationship to temperature-induced unfolding. Dynamics parameters for different temperatures are compared to provide insight into thermodynamic stability of protein G.
Heteronuclear relaxation data for proteins may be analyzed in terms of the Lipari-Szabo model free formalism (or extended Lipari-Szabo formalisms) to yield an order parameter describing the degree of spatial restriction of internal motions on a sub-nanosecond time scale. Such data have now been reported for the motions of NH groups in more than forty proteins. However the structural features that influence the order parameters remain unknown. The Indiana Dynamics Database (http://pooh.chem.indiana.edu) was established to consolidate many of these data, allowing us to perform statistical analyses to probe structure-dynamics relationships. Each residue in each protein has been classified according to amino acid type, amino acid sequence context (the identities of neighboring residues), secondary structure, and surface accessibility. These features have been correlated with the NMR-derived order parameters for backbone NH groups. The results of these studies will be presented and the implications for the design of proteins with specific motional properties will be discussed. Relationships between protein molecular weights and molecular rotational correlation times and conclusions regarding protein hydration will also be presented.