Dotsch / Schmitz Nuclear Magnetic Resonance of Biological Macromolecules, Part A

1. Auflage 2001
ISBN: 978-0-08-049688-7
Verlag: Elsevier Science & Techn.
Format: EPUB
Kopierschutz: 6 - ePub Watermark

E-Book, Englisch, Band Volume 338, 480 Seiten

Reihe: Methods in Enzymology

This volume and its companion, Volume 339, supplement Volumes 176, 177, 239, and 261. Chapters are written with a 'hands-on' perspective. That is, practical applications with critical evaluations of methodologies and experimental considerations needed to design, execute, and interpret NMR experiments pertinent to biological molecules.

Autoren/Hrsg.

Dotsch, Volker

Schmitz, Uli

Inhaltsverzeichnis

1;Front Cover;1
2;Nuclear Magnetic Resonance of Biological Macromolecules;4
3;Copyright Page;5
4;Table of Contents;6
5;Contributors to Volume 338;8
6;Preface;12
7;Volume in Series;14
8;Section I: General Techniques;34
8.1;Chapter 1. Use of Chemical Shifts in Macromolecular Structure Determination;36
8.2;Chapter 2. Cross-Correlated Relaxation for Measurement of Angles between Tensorial Interactions;68
8.3;Chapter 3. Applications of Adiabatic Pulses in Biomolecular Nuclear Magnetic Resonance;115
8.4;Chapter 4. Scalar Couplings across Hydrogen Bonds;144
8.5;Chapter 5. On-Line Cell High-Pressure Nuclear Magnetic Resonance Technique: Application to Protein Studies;167
8.6;Chapter 6. Maximum Entropy Reconstruction: Spectrum Analysis and Deconvolution in Multidimensional Nuclear Magnetic Resonance;192
8.7;Chapter 7. Magnetic Relaxation Dispersion Studies of Biomolecular Solutions;211
8.8;Chapter 8. Nuclear Magnetic Resonance-Based Approaches for Lead Generation in Drug Discovery;235
8.9;Chapter 9. Screening of Compound Libraries for Protein Binding Using Flow-Injection Nuclear Magnetic Resonance Spectroscopy;263
8.10;Chapter 10. The Internet for Nuclear Magnetic Resonance Spectroscopists;280
9;Section II: Nucleic Acids and Carbohydrates;292
9.1;Chapter 11. Solid-Phase Synthesis of Selectively Labeled DNA: Applications for Multidimensional Nuclear Magnetic Resonance Spectroscopy;294
9.2;Chapter 12. Uniform 13C/15N-Labeling of DNA by Tandem Repeat Amplification;316
9.3;Chapter 13. 13C Isotopic Enrichment for Nuclear Magnetic Resonance Studies of Carbohydrates and Glycoconjugates;338
9.4;Chapter 14. Preparation and Use of 2H-Labeled RNA Oligonucleotides in Nuclear Magnetic Resonance Studies;353
9.5;Chapter 15. Investigation of Unusual DNA Motifs;374
9.6;Chapter 16. Resonance Assignment and Structure Determination for RNA;404
9.7;Chapter 17. Solution Nuclear Magnetic Resonance Probing of Cation Binding Sites on Nucleic Acids;433
9.8;Chapter 18. Identification and Characterization of Metal Ion Binding Sites in RNA;454
10;Author Index;478
11;Subject Index;504

Leseproben

[1] Use of Chemical Shifts in Macromolecular Structure Determination
David S. Wishart; David A. Case Introduction
Chemical shifts are the universal language of nuclear magnetic resonance (NMR). They communicate in a simple way detailed molecular information that almost every chemist can understand. Although chemical shifts have long been used as tools for covalent structural analysis, it is important to remember that they can also provide detailed information about noncovalent structure, solvent interactions, ionization constants, ring orientations, hydrogen bond interactions, and other phenomena. One major force driving a growing interest in chemical shifts has been the rapid growth in the number of macromolecular NMR assignments in repositories such as the BioMagResBank (BMRB). Another important factor has been improvements in computer hardware and software that now make accurate chemical shift calculations on fragments of biological macromolecules almost routine.1–3 The purpose of this article is to provide readers with practical advice on how to use chemical shifts as an aid in understanding, generating or refining macromolecular structures. Some attempt will be made to update material presented an earlier volume of this series.4,5 However, no attempt will be made to extensively survey the theory and history of macromolecular chemical shifts, as several excellent reviews already exist.3,6,7 Chemical Shift Referencing
Chemical shifts are among the most precisely measurable but least accurately measured spectral parameters in all of NMR spectroscopy. The issue of accuracy, as opposed to precision, lies at the heart of how reliably chemical shifts can be used for structural interpretation and analysis. Systematic errors in chemical shift measurements as small as 0.05 ppm for 1H shifts, 0.3 ppm for 13C shifts, or 0.5 ppm for 15N shifts can make a significant difference in the identification of secondary structure,4,8–11 the measurement of hydrogen bond lengths,12,13 the determination of dihedral angle restraints,14–16 the convergence of a chemical shift refinement17–19 or the development of empirical chemical shift “surfaces.”14,15,20–22 It is important to remember that chemical shifts are relative frequency measurements, not absolute measurements. In organic chemistry and for compounds dissolved in organic solvents, TMS (tetramethylsilane) has been the de facto 1H and 13C chemical shift standard since the 1970s.23 However, it has only been recently that a set of universal standards for chemical shift referencing in aqueous solutions has been proposed24,25 and adopted26 by the IUPAC and IUBMB. Prior to that decision, more than a dozen different chemical shift standards had been in use at different times by different labs around the world.4 The specific IUPAC recommendations for biological molecules are that internal DSS (2,2-dimethyl-2-silapentane-5-sulfonic acid), a water-soluble, pH-insensitive form of TMS, should be the standard used for 1H and 13C referencing. In addition, external anhydrous liquid ammonia should be used for 15N referencing, external 100% trifluoroacetic acid should be used 19F referencing,25 and internal 10% trimethylphosphate is recommended for 31P referencing.26 Because of the difficulties in working with some of these reference compounds, an alternative indirect referencing procedure has been strongly advocated.24,26,27 In particular, by using predetermined nucleus-specific frequency ratios (called ? or the Greek letter xi) derived for DSS (13C), liquid ammonia (15N), trifluoroacetic acid (19F), and trimethylphosphate (31P), it is possible to determine the zero point reference for these compounds (and hence these nuclei) using the absolute 1H frequency of internal DSS. Some of the more commonly used ? values are presented in Table I. A more extensive list is available at the BioMagResBank (www.bmrb.wisc.edu/bmrb). Table I IUPAC/IUBMB Recommended ? (Xi) Ratios for Indirect Chemical Shift Referencing in Biomolecular NMRa Nucleus Compound ? Ratio 1H DSS 1.000 000 000 13C DSS 0.251 449 530 15N Liquid NH3 0.101 329 118 19F CF3COOH 0.940 867 196 31P (CH3)3PO4 0.404 808 636 a Relative to DSS. As an example, let us assume one wished to reference the 15N dimension of an 15N-1H HSQC experiment. First, the sample must contain a detectable amount of dissolved DSS (say 100 µM). Prior to collecting the spectrum, determine the 1H carrier frequency (say 500,000,087.2 Hz) of the spectrometer. Second, determine the 1H DSS frequency relative to the carrier frequency (assume it is 2521.2 Hz upfield of the carrier). This implies the absolute DSS frequency is 500,0,087.2 – 2521.2 = 499,997,566.0 Hz. Third, multiply this DSS 1H frequency by the 15N ? ratio found in Table I (the result is 50,664,312.4 Hz). This value corresponds to the hypothetical 15N resonance frequency of external liquid ammonia, which by definition is 0 ppm. If the 15N carrier (decoupler) frequency is also known or measured (say it is 50,670,450.8 Hz), then the 15N chemical shift scale, in ppm, can be fully determined. Because magnetic fields drift and spectrometer frequencies vary over time, this indirect referencing procedure must generally be repeated each time a new sample is placed in a spectrometer. For most spectrometers it is possible to write a simple computer program to perform this referencing task routinely. A macro called XREF for indirect referencing is available over the Web (see Table V for more information). Regardless of whether one chooses to use the direct or indirect referencing procedures, properly referenced spectra are absolutely key to obtaining meaningful chemical shift information. Given the long-standing problems in the biomolecular NMR community concerning chemical shift referencing, one might ask: “How accurate are the protein or nucleic acid chemical shifts that have been published over the past 20 years?” The answer to this question is important because it affects our ability to compare, extract, and reproduce data from other laboratories. One of us (DSW) has undertaken a long-term retrospective survey of the protein chemical shift data deposited in the BMRB to attempt to identify and correct potentially misreferenced data sets. The process involves predicting protein 1H, 13C, and 15N chemical shifts using X-ray or NMR coordinate data and then comparing those predictions to the measured values of corresponding proteins found in the BMRB. The program, called SHIFT-COR (Table V), was originally developed and tested on a series of proteins known to be correctly referenced via IUPAC conventions. Although the residue specific shift predictions for this program are often less than perfect, the global averages for 1H, 13Ca, 13Cß, 13CO, and 15N shifts (calculated over all residues in a given protein) have been found to be quite sensitive to chemical shift referencing errors. Of the more than 60 proteins surveyed to date it appears that (1) there are essentially no detectable problems with 1H shift referencing; (2) approximately 20% of the proteins deposited in the BMRB have 13Ca 13CO and 13Cß shifts that are misreferenced (> 0.5 ppm); and (3) approximately 30% of the proteins deposited in the BMRB have 15N shifts that are misreferenced (> 1 ppm). Other investigators have reported similar problems with 15N and 13C shifts over smaller protein data sets.15,16 This result is quite worrisome—particularly for those who would like to use experimental chemical shift data in structural analysis and comparison. Databases of “re-referenced” protein chemical shifts are available in the TALOS program16 and in RefDB, which is composed of those proteins that have already gone through the SHIFTCOR filter, and is available over the Web (Table V). Interestingly, 13CO data seem to have their own particular chemical shift referencing problems, which are distinct from 13Ca and 13Cß referencing issues. Indeed, it is not uncommon to find a protein with correctly referenced 13Ca and 13Cß shifts but with 13CO shifts that are systematically off by 1 or 2 ppm. Evidently this problem may be related to the tendency of some spectroscopists to use offset-synthesized pulses to excite 13CO resonances while still keeping their carrier at the 13Ca frequency. When this is done, the resulting spectrum will be folded many times and may often show resonances at reasonable (but incorrect) 13CO ppm values. Essentially, the referencing error arises when one uses the ppm frequency of the offset synthesized pulse as the carrier frequency in the 13CO dimension (A. Bax, personal communication). Random Coil Shifts
Random coil shifts are defined as the characteristic chemical shifts of amino acid residues or nucleic acid...