Applications Of Nmr In Biology

NMR initially grew from a curiosity for physicists into an essential tool for organic chemists, but the more recent explosion in NMR technology and applications has been driven by the use of NMR for structural biology. The drive for higher field strength and greater sensitivity, as well as new technologies (such as gradients and shaped pulses) has made possible the study of ever larger and more complex molecules, breaking the barrier from "small" organic molecules (natural products, peptides, oligosaccharides, etc.) into the realm of proteins, DNA, and RNA: the "informational" biomolecules. The precise three-dimensional (3D) structure of these molecules and the geometry of their noncovalent interactions with small molecules and with other biomolecules are the keys to understanding how biology functions on the molecular level. Formerly the exclusive realm of X-ray crystallography, this structural information is increasingly derived from NMR experiments without the need for crystals. This field is extremely demanding and has not only pushed the development of NMR hardware but also created an explosion of new NMR pulse sequences and techniques.

Biological NMR is a very broad field and we will attempt here only to list some of the applications and then focus on just one: protein NMR spectroscopy in solution. Solid-state NMR has been a very active area for biological molecules—either using fixed samples such as multiple aligned sheets of lipid bilayers to study membrane-bound biomolecules, or using very rapid spinning of powder samples tilted at the "magic angle" (54.74°) to the Bo field. This magic angle spinning (MAS) technique mimics the rapid, isotropic motion of small molecules in solution to give narrow lines similar to solution-state NMR spectra. Isotopic labeling of specific positions in a biological molecule, for example, with one 13 C nucleus in one position and one 15N nucleus in another, allows very precise direct through-space distance measurements. NMR imaging is a very active area of research in the

NMR Spectroscopy Explained: Simplified Theory, Applications and Examples for Organic Chemistry and Structural Biology, by Neil E Jacobsen Copyright © 2007 John Wiley & Sons, Inc.

medical field as well as in animal, plant, and materials studies. Applying gradients during the acquisition of the FID replaces the chemical-shift scale with a distance (position within the sample) scale, so that a physical map of H2O concentration in the human body can be constructed. New methods give contrast in these images from not only water concentration but also flow rate and direction, diffusion, relaxation (T1 or T2), and targeted injectable contrast agents that can "light up" the locations of specific biomolecules in the body. By synchronizing a fast imaging experiment to the patient's heartbeat, a real time sequence of images can be constructed for each stage of the heart's cycle. Microimaging is made possible by very strong field gradients, so that an ordinary wide-bore vertical NMR spectrometer can be used for microscopy of plant or animal tissues. Imaging can also be combined with spectroscopy, so that a separate NMR spectrum (e.g., 31P spectrum) can be obtained for each volume unit ("voxel") of the brain. Dispensing entirely with the imaging component, in vivo spectroscopy can be used to study metabolism in cell cultures. Special NMR tubes and infusion systems have been developed to grow cell cultures directly in the NMR tube, using microimaging to monitor cell growth. Specifically labeled compounds can be introduced and their conversion to metabolites monitored by 13 C or 31P NMR spectroscopy.

In the solution state, NMR allows us to study molecular motion in detail. Pulsed-field gradients allow the measurement of diffusion along any of the three axes (x, y, or z) as well as the direction and speed of flow. Rotation of the entire biomolecule can be observed by its effect on the relaxation parameters (T1, T2, and NOE) of any of the nuclei within the molecule. By focusing on specific pairs of nuclei that are oriented in a fixed relationship to the entire molecule (e.g., 15N-1H), we can measure the rotational diffusion rate about any of the molecular axes. These random rotational motions are on the timescale of \xs to ns, but we can also look at internal motions that are faster than the overall molecular motion (ns and faster). Because we can measure this for every pair (15N-1H or 13C-1H) in the molecule, this "flexibility" can be mapped onto the three-dimensional structure of a molecule to locate disordered regions or regions that become ordered upon binding of other molecules. Slower motions (ms timescale) can also be detected and mapped to specific regions, and these usually correspond to conformational changes that are essential to the function of a biomolecule. In this sense, biological NMR is about biochemical function as much as it is about three-dimensional structure.

In the 1980s the foundations were laid for a series of solution NMR techniques that allow the determination of three-dimensional structure (conformation) of biomolecules in solution. These methods were pioneered by Kurt Wiithrich at ETH in Zurich, using small proteins such as bovine pancreatic trypsin inhibitor (BPTI). This approach has been applied to increasingly more complex proteins and nucleic acids (DNA and RNA) and more recently to the structure of molecular complexes (protein-protein, protein-DNA, drug-DNA, etc.). Because NMR relies on local "reporters" (specific 1H, 13C, and 15N nuclei within a molecule) it can be used to "map" the binding of small molecules or other biomolecules onto the surface of a protein by observing perturbations of chemical shifts. This technique, called "SAR by NMR" (structure-activity relationships by NMR), is now used for screening of large libraries of small molecules in the drug discovery process. In this book, we will be limited to outlining the general process of structure determination of proteins, starting with small proteins at natural abundance and moving to newer techniques involving uniformly 15N- and/or 13C-labeled proteins.

There are two kinds of solution-state NMR problems and up to now we have been concerned entirely with the first: determining the covalent structure or bonding network of a molecule. We are confronted with an unknown or incompletely known covalent structure of a "small" molecule, and we need to find out which atoms are connected to which and in what stereochemical orientation (trans, cis, E, Z, equatorial, axial, a, f, etc.). These problems come up in organic synthesis (confirmation of an expected product structure), synthetic methods (ratios of products with different regiochemistry and stereochemistry), in the oxidation or metabolism of natural products or drugs (known starting materials), and in the "blue sky" or "white powder" pursuit of new natural products isolated from plants and microorganisms. The tools for this pursuit are the experiments we have discussed in the previous chapters: HSQC, HMBC, COSY, TOCSY, NOESY/ROESY, and the 1D selective NOE and TOCSY experiments. From these experiments we get basic information about chemical shifts, 1, 2, and 3 bond relationships between XH and 13 C, 2-3 bond relationships between protons, 1H-1H distances, and 1H-1H and 1H-13C dihedral angles.

The second type of structural problem in NMR is the subject of this chapter: determining the conformation or specific three-dimensional fold of a molecule with a known covalent structure. The conformation must be relatively rigid, held together by a large number of noncovalent interactions and hydrophobic forces, and specific with little or no heterogeneity of structure. While conformation is sometimes of interest with small molecules, this type of problem is found mostly in the area of biopolymers—large molecules composed of a specific covalent sequence of unlike monomer building blocks and "folded" into a specific three-dimensional shape. These systems include carbohydrates, peptides, glycopeptides, proteins, double-stranded DNA, RNA, and the noncovalent complexes of any pair of these molecules. The techniques include all of the 2D NMR tools used for small molecules (except HMBC) and a number of new methods we will describe in this chapter: 1H/2H exchange, uniform 15N,13 C and/or 2H labeling, 3D NMR expanding our homonuclear 2D experiments (TOCSY and NOESY) into a third dimension using the 15N or 13C chemical shift, and 3D and higher-dimensional "triple-resonance" experiments that rely on doubly-labeled (15N and 13C) protein samples. We will also briefly mention two new techniques that extend the size limit of molecules we can study (TROSY) and cross the boundary with solid-state NMR by measuring the direct through-space dipolar (dipole-dipole) interaction, which is normally averaged to zero by rapid isotropic tumbling in the solution state. The information we obtain from these experiments includes sequence-specific chemical-shift assignments for all spins (1H, 15N, and 13C) in the molecule, chemical shift deviations as indicators of secondary structure (a-helix or f-sheet), degree of protection of amide N-H groups from solvent, thousands of proton-proton dihedral angles and distances, sequence-specific dynamics (order vs. flexibility), and in the case of the newest experiments the orientation of N-H and C-H vectors relative to the rest of the molecule. This vast and heterogeneous store of information is used in the process of structure calculation, which attempts to arrive at a single three-dimensional structure (conformation) of the molecule that is most consistent with all of these measurements.


X-ray diffraction (crystallography) has been around for much longer than NMR and has been used to determine the precise three-dimensional structure of biomolecules as large as viruses (molecular weight in the tens of millions!). This technique requires a high-quality crystal and calculates a three-dimensional map of electron density for the molecule. The molecular structure (atoms and bonds) is then fit to this electron density map to generate the precise coordinates of all atoms, which is what we call a "structure." Because hydrogen does not contribute much to the electron density, the X-ray technique effectively only sees the "heavy" atoms (N, C, O, P, S, etc.). In contrast, NMR determines the molecular structure (conformation) directly in the solution state in the native environment for biomolecules (water) by detecting primarily the hydrogens and measuring the short-range distances (<5 A) and dihedral angles (3-bond relationships) between them. Until recently, solution NMR did not offer any way to determine global relationships (geometry of the molecule relative to the outside world).

0 0

Post a comment