Structure determination of a DNA octamer in solution by NMR spectroscopy. Effect of fast local motions.

NMR structures of biomolecules are primarily based on nuclear Overhauser effects (NOEs) between protons. For the interpretation of NOEs in terms of distances, usually the assumption of a single rotational correlation time corresponding to a rigid molecule approximation is made. Here we investigate the effect of fast internal motions of the interproton vectors in the context of the relaxation matrix approach for structure determination of biomolecules. From molecular dynamics simulations generalized order parameters were calculated for the DNA octamer d(GCGTTCGC).d(CGCAACGC), and these were used in the calculation of NOE intensities. The magnitudes of the order parameters showed some variation for the different types of interproton vectors. The lowest values were observed for the interresidue base H6/H8-H2" proton vectors (S2 = 0.60), while the cytosine H5-H6 interproton vectors were among the most motionally restricted (S2 = 0.92). Inclusion of the motion of the interproton vectors resulted in a much better agreement between theoretically calculated NOE spectra and the experimental spectra measured by 2D NOE spectroscopy. The interproton distances changed only slightly, with a maximum of 10%; nevertheless, the changes were significant and resulted in constraints that were better satisfied. The structure of the DNA octamer was determined by using restrained molecular dynamics simulations with H2O as a solvent, with and without the inclusion of local internal motions. Starting from A- or B-DNA, the structures showed a high local convergence (0.86 A), while the global convergence for the octamer was ca. 2.6 A.     

Nuclear Overhauser effects as probes of peptide structure. A diagnostic for left-handed helical conformations

The conformational dependence of the interresidue interproton distances in peptides, C alpha H ... Ni + 1 H and NiH ... Ni + 1 H, have been used to identify zones of sterically allowed phi, psi space, where both distances are less than 3A and expected to yield nuclear Overhauser effects (NOEs). L-residues in left-handed helical conformations are expected to yield both interresidue NOEs and also an appreciable intraresidue NiH----C alpha iH NOE. The effect of cutoff distances has been evaluated. Experimental results on three model peptides illustrate the utility of these NOEs in identifying L-residues at the i + 2 position of Type II and I' beta-turns. Simultaneous observation of both interresidue NOEs may also be indicative of conformational heterogeneity in specific cases, as illustrated for a single residue in a decapeptide.     

Nuclear Overhauser effect and cross-relaxation rate determinations of dihedral and transannular interproton distances in the decapeptide tyrocidine A.

The following interproton distances are reported for the decapeptide tyrocidine A in solution: (a) r(phi) distances between NH(i) and H alpha (i), (b) r(psi) distances between NH (i + 1) and H alpha (i), (c) r(phi psi) distances between NH(i + 1) and NH(i), (d) NH in equilibrium NH transannular distances, (e) H alpha in equilibrium H alpha transannular distances, (f) r x 1 distances between H alpha and H beta protons, (g) NH(i) in equilibrium H beta (i) distances, (h) NH (i + 1) in equilibrium H beta (i) distances, (i) carboxamide-backbone protons and carboxamide-side chain proton distances, (j) side chain proton-side chain proton distances. The procedures for distance calculations were: NOE ratios and calibration distances, sigma ratios and calibration distances, and correlation times and sigma parameters. The cross-relaxation parameters were obtained from the product, say, of NOE 1 leads to 2 and the monoselective relaxation rate of proton 2; the NOEs were measured by NOE difference spectroscopy. The data are consistent with a type I beta-turn/ type II' beta-turn/ approximately antiparallel beta-pleated sheet conformation of tyrocidine A in solution and the NOEs, cross-relaxation parameters, and interproton distances serve as distinguishing criteria for beta-turn and beta-pleated sheet conformations. It should be borne in mind that measurement of only r phi and r psi distances for a decapeptide only defines the ( phi, psi)-space in terms of 4(10) possible conformations; the distances b-j served to reduce the degeneracy in possible (phi, psi)-space to one tyrocidine A conformation. The latter conformation is consistent with that derived from scalar coupling constants, hydrogen bonding studies, and proton-chromophore distance measurement, and closely resembles the conformation of gramicidin S.     

Internal mobility in a double-stranded B DNA hexamer and undecamer. A time-dependent proton-proton nuclear Overhauser enhancement study

The internal mobility of the deoxyribose H2'-H2" and base C(H5)-C(H6) and T(CH3)-T(H6) vectors has been investigated by means of time-dependent nuclear Overhauser enhancement (NOE) measurements in a B DNA hexamer and undecamer. Cross-relaxation rates between these proton pairs are determined from the initial slopes of the time development of the NOEs, and, as the interproton distances between these proton pairs are fixed, apparent correlation times for the 3 interproton vectors are calculated from the cross-relaxation rate data. It is shown that there is little residue to residue variation in the cross-relaxation rates of the interproton vectors within each oligonucleotide, that the mean apparent correlation times of the C(H5)-C(H6) and T(CH3)-T(H6) vectors are approximately equal and significantly greater than that of the H2'-H2" vectors, and that the data for the H2'-H2" vectors of both oligonucleotides and the C(H5)-C(H6) and T(CH3)-T(H6) vectors of the undecamer cannot be accounted for by isotropic tumbling alone. The data are analysed in terms of a two motion model with isotropic tumbling and a single internal motion. The relaxation time of the internal motion at 23 degrees C is less than or equal to 1 ns for the H2'-H2" vectors of both oligonucleotides and less than or equal to 3 ns for the C(H5)-C(H6) and T(CH3)-T(H6) vectors of the undecamer. In the case of the H2'-H2" vectors, however, the amplitude of the internal motion is found to be too large to be compatible with the known stereochemistry of DNA. This finding can only be explained by invoking additional degrees of internal freedom with a larger number of internal motions of small amplitude of the type deduced from the analysis of crystallographic thermal factors [(1984) J. Mol. Biol. 173, 361-388].     

The determination of the conformational properties of nucleic acids in solution from NMR data

A program, NUCFIT, has been written for simulating the effects of conformational averaging on nuclear Overhauser enhancement (NOE) intensities for the spin systems found in nucleic acids. Arbitrary structures can be generated, and the NOE time courses can be calculated for truncated one-dimensional NOEs, two-dimensional NOE and rotating frame NOE spectroscopy (NOESY and ROESY) experiments. Both isotropic and anisotropic molecular rotation can be treated, using Woessner's formalism (J. Chem. Phys. (1962) 37, 647-654). The effects of slow conformational averaging are simulated by taking population-weighted means of the conformations present. Rapid motions are allowed for by using order parameters which can be supplied by the user, or calculated for specific motional models using the formalism of Tropp (J. Chem. Phys. (1980) 72, 6035-6043). NOE time courses have been simulated for a wide variety of conformations and used to determine the quality of structure determinations using NMR data for nucleic acids. The program also allows grid-searching with least-squares fitting of structures to experimental data, including the effects of spin-diffusion, conformational averaging and rapid internal motions. The effects of variation of intra and internucleotide conformational parameters on NOE intensities has been systematically explored. It is found that (i) the conformation of nucleotides is well determined by realistic NOE data sets, (ii) some of the helical parameters, particularly the base pair roll, are poorly determined even for extensive, noise-free data sets, (iii) conformational averaging of the sugars by pseudorotation has at most second-order influence on the determination of other parameters and (iv) averaging about the glycosidic torsion bond also has, in most cases, an insignificant effect on the determination of the conformation of nucleotides.     

The effect of selective deuteration on magnetization transfer in larger proteins

Summary The effects of selective deuteration on calculated NOESY intensities have been analyzed for the structure of theE. coli trp aporepressor, a 25 kDa protein. It is shown that selectively deuteratedtrp aporepressor proteins display larger calculated NOESY intensities than those for the same interproton distances in the natural abundance protein. The relatively larger magnetization transfer is demonstrated by a comparison of the NOE build-up curves for specific proton pairs, and for the calculated NOE intensities of short-range NOEs to backbone amide protons. This increase in intensity is especially pronounced for the NH¹–NH¹⁺¹ cross peaks in the -helical regions, and particularly for amide protons of two sequential deuterated residues. The effect is shown to be further intensified for longer mixing times. It is also shown that in all cases, each amide proton exhibits stronger NOEs to its own side chain, with an enhanced effect for deuterated derivatives. This theoretical analysis demonstrates that an evaluation of the relative NOE intensities for different selectively deuterated analogs may be an important tool in assigning NMR spectra of large proteins. These results also serve as a guide for the interpretation of NOEs in terms of distances for structure calculations based on data using selectively deuterated proteins.     

Detection of internal motions in oligosaccharides by 1H relaxation measurements at different magnetic fields.

The effect of internal motions on proton relaxation data in oligosaccharides has been investigated experimentally. 1H steady-state and transient NOEs together with 13C T1's have been measured at two magnetic field strengths. The existence of internal motions leads to additional modulations of the dipolar interaction between proton pairs, thus producing a range of spectral density functions for these interactions. As a result, it is possible to show that protons relaxing through fixed distances have a different ratio of relaxation parameters, acquired at 500 and 300 MHz, compared to those relaxing through fluctuating distances. This approach has been used to unequivocally establish for two disaccharides the existence of internal motions on the time scale of the overall tumbling.     

Conformation and dynamics of an Fab'-bound peptide by isotope-edited NMR spectroscopy.

The dynamics and conformation of the peptide antigen MHKDFLEKIGGL bound to the Fab' fragment of the monoclonal antipeptide antibody B13A2, raised against a peptide from myohemerythrin, have been investigated by isotope-edited NMR techniques. The peptides were labeled with 15N (98%) or 13C (99%) at the backbone of individual amino acid residues. Well-resolved amide proton and nitrogen backbone resonances were obtained and assigned for eight of the 12 residues of this bound peptide. Significant resonance line width and chemical shift differences were observed. The 15N and 1H line width variations are attributed to differential backbone mobilities among the bound peptide residues which are consistent with the previously mapped epitope of this peptide antigen. Local structural information was obtained from isotope-directed NOE studies. The approximate distances associated with the experimental NOEs were estimated on the basis of a theoretical NOE analysis involving the relative integrated intensities of the NOE and source peaks. In this way, the sequential NH-NH NOEs obtained for seven of the Fab'-bound peptide residues were shown to correspond to interproton separations of approximately 3 A or less. Such short distances indicate that the backbone dihedral angles of these residues are in the alpha rather than the beta region of phi,psi conformational space; the peptide most likely adopts a helical conformation from F5 to G11 within the antibody combining site. The significance of these results with respect to the type and extent of conformational information obtainable from studies of high molecular weight systems is discussed.     

Effects of internal motions on the development of the two-dimensional transferred nuclear Overhauser effect

Summary In this paper we address the influence of internal motions on the development of the transferred nuclear Overhauser effect in a ligand undergoing chemical exchange between a free and a bound state. We examine the effects of varying the effective correlation time as well as the motional order parameter for methyl group and phenyl ring rotations in the free and bound ligand conformations. The effect of decreasing the motional order for a proton pair on a methyl group or phenyl ring is to decrease the effective correlation time of the internuclear vector, and thus to decrease the cross-relaxation rate between the proton pair. This functions to dampen the effects of spin diffusion, especially in the bound ligand- where cross-relaxation rates are much faster than in the free ligand. The effect of changing the effective correlation time for methyl group motions has little effect on the build-up behaviour of the transferred nuclear Overhauser effect for small values of fraction bound, but a larger effect on how fast it decays. This effect is greater for internal motions in the free peptide than it is for internal motions in the bound peptide.Dedicated to the memory of Professor V.F. Bystrov     

A Monte Carlo simulation study of the influence of internal motions on the molecular conformation deduced from two-dimensional NMR experiments

Monte Carlo methods have been used to simulate internal motions of aromatic protons of an oligonucleotide at the nanosecond time scale. Each proton is allowed to fluctuate about its equilibrium position. The longitudinal cross-relaxation rates of such a system of spins have been determined by computing the appropriate correlation functions. Then the interproton distances have been deduced according to the procedure generally used in two-dimensional nmr techniques (nuclear Overhauser effect spectroscopy--NOESY) and compared to the true values. The influence of the amplitude A and of the internal rotational diffusion constant Dint characterizing the dynamics of the system has been checked for in-phase and for uncorrelated motions. It is shown that for the investigated models the distances deduced from NOESY experiments may be under- or overestimated, depending strongly on the values of A and Dint. Furthermore, the cross-relaxation rate of a couple of protons is very sensitive to the correlation level of the motions of both protons.     

Improved distance analysis in RNA using network-editing techniques for overcoming errors due to spin diffusion

Multispin magnetization transfer, or spin diffusion, is a significant source of error in NOESY-derived distance measurements for the determination of nucleic acid solution structures. The BD-NOESY and CBD-NOESY experiments, which allow the measurement of interproton distances with greatly reduced contributions from spin diffusion, have been adapted to structural analysis in RNA oligonucleotides. The techniques are applied to a lead-dependent ribozyme (LZ2). We demonstrate the measurement of both aromatic proton–aromatic proton NOEs free of spin diffusion involving the intervening ribose moieties and aromatic proton–ribose proton NOEs free of the efficient cross-relaxation within the ribose ring. In LZ2, the accuracy and precision of the resulting distances are significantly improved. We also find that, by allowing the use of longer mixing times with greater sensitivity, the experimental attenuation of spin diffusion in RNA increases the distance range of interactions that can be analyzed. This effect permits measurement of important long-range distances in LZ2 that are not accessible with standard techniques. Thus, these techniques allow the simultaneous optimization of the number, accuracy, and precision of distance constraints used for RNA structure determinations.     

Intermolecular relaxation has little effect on intra-peptide exchange-transferred NOE intensities

Exchange-transferred nuclear Overhauser enhancement (etNOE) provides a useful method for determining the 3-dimensional structure of a ligand bound to a high-molecular-weight complex. Some concern about the accuracy of such structures has arisen because indirect relaxation can occur between the ligand and macromolecule. Such indirect relaxation, or spin diffusion, would lead to errors in interproton distances used as restraints in structure determination. We address this concern by assessing the extent of intermolecular spin diffusion in nineteen peptide-protein complexes of known structure. Transferred NOE intensities were simulated with the program CORONA (Calculated OR Observed NOESY Analysis) using the rate-matrix approach to include contributions from indirect relaxation between protein-ligand and intraligand proton pairs. Intermolecular spin diffusion contributions were determined by comparing intensities calculated with protonated protein to those calculated with fully deuterated protein. The differences were found to be insignificant overall, and to diminish at short mixing times and high mole ratios of peptide to protein. Spin diffusion between the peptide ligand and the protein contributes less to the etNOE intensities and alters fewer cross peaks than the well-studied intramolecular spin diffusion effects. Errors in intraligand interproton distances due to intermolecular relaxation effects were small on average and can be accounted for with the restraint functions commonly used in NMR structure determination methods. In addition, a rate-matrix approach to calculate distances from etNOESY intensities using a volume matrix comprising only intraligand intensities was found to give accurate values. Based on these results, we conclude that structures determined from etNOESY data are no less accurate due to spin diffusion than structures determined from conventional NOE intensities.     

Study of the spatial structure of the duplex (Phn-NH(CH2)NH)pd(CCAAACA).pd(TGTTTGGC) with covalently bound 10-(2-hydroxyethyl)phenazine by in aqueous solution by 2D-(1)H-NMR and by limited molecular mechanics]

Detailed investigation of the spatial structure of duplex (Phn-NH(CH2)2NH) x pd(CCAAACA).pd(TGTTTGGC) having a covalently linked N-(2-hydroxyethyl)-phenazine in aqueous solution was continued by means of one- and two-dimensional 1H-NMR spectroscopy. Distances between the protons of the oligonucleotides as well as distances between the phenazinium and the nearest nucleotide groups protons were determined from the series of one-dimensional NOE experiments. The effective correlation time tau c determined for some proton pairs shows the phenazinium fragment to have greater internal motion than the heterocyclic bases. The deoxyribose protons coupling constants show the sugars to be in 2'-endo-conformation. The restrained molecular mechanics have yielded a possible structure of duplex in the aqueous solution fitting the experimental set of interproton distances.     

Proton and tritium NMR relaxation studies of peptide inhibitor binding to bacterial collagenase: conformation and dynamics.

The interaction of succinyl-Pro-Ala, a competitive inhibitor of Achromobacter iophagus collagenase, with the enzyme was studied by longitudinal proton and tritium relaxation. Specific deuterium and tritium labeling of the succinyl part at vicinal positions allowed the measurement of the cross-relaxation rates of individual proton or tritium spin pairs in the inhibitor-enzyme complex as well as in the free inhibitor. Overall correlation times, internuclear distances, and qualitative information on the internal mobility in Suc1 (as provided by the generalized order parameter S2) could be deduced by the comparison of proton and tritium cross-relaxation of spin pairs at complementary positions in the -CH2- CH2- moiety as analyzed in terms of the model-free approach by Lipari and Szabo. The conformational and motional parameters of the inhibitor in the free and enzyme-bound state were directly compared by this method. The measurement of proton cross-relaxation in the Ala residue provided additional information on the inhibitor binding. The determination of the order parameter in different parts of the inhibitor molecule in the bound state indicates that the succinyl and alanyl residues are primarily involved in the interaction with the enzyme activity site. The succinyl moiety, characterized in solution by the conformational equilibrium among the three staggered rotamers--i.e., trans: 50%; g+: 20%; g-: 30%--adopted in the bound state the unique trans conformation.     

The octamer motif in immunoglobulin genes: extraction of structural constraints from two-dimensional NMR studies.

Phase-sensitive two-dimensional nuclear Overhauser enhancement (2D NOE) and double-quantum-filtered correlated (2QF-COSY) spectra were recorded at 500 MHz for the DNA duplex d(CATTTGCATC).d(GATGCAAATG), which contains the octamer element of immunoglobulin genes. Exchangeable and nonexchangeable proton resonances including those of the H5' and H5" protons were assigned. Overall, the decamer duplex adopts a B-type DNA conformation. Scalar coupling constants for the sugar protons were determined by quantitative simulations of 2QF-COSY cross-peaks. These couplings are consistent with a two-state dynamic equilibrium between a minor N- and a major S-type conformer for all residues. The pseudorotation phase angle P of the major conformer is in the range 117-135 degrees for nonterminal pyrimidine nucleotides and 153-162 degrees for nonterminal purine nucleotides. Except for the terminal residues, the minor conformer comprises less than 25% of the population. Distance constraints obtained by a complete relaxation matrix analysis of the 2D NOE intensities with the MARDIGRAS algorithm confirm the dependence of the sugar pucker on pyrimidine and purine bases. Averaging by fast local motions has at most small effects on the NOE-derived interproton distances.     

Conformation of secretin in dimethyl sulfoxide solution. NMR studies and restrained molecular dynamics

The solution conformation of the 27-residue polypeptide hormone secretin in dimethyl sulfoxide has been determined on the basis of 1H-NMR measurements. The experimental data set used in the structure determination consisted of 98 nuclear-Overhauser-enhancement-derived interproton and dihedral angle restraints from coupling constants. The NH-NH and H alpha-NH NOEs were determined from build-up rates, while the remaining distances were classified in a qualitative manner. The structure calculations consisted of two phases. First, dynamical simulated annealing calculations were carried out to find conformations of the peptide which satisfy NOE and phi dihedral restraints. The convergence of ten calculated structures was good except for those regions of the molecule where NOE data were not unambiguous. From the calculated set another initial structure was built which was again minimized in several 5-ps calculations now employing the full empirical energy function. The resulting structures of secretin reveal conformationally well-defined regions, but not a single uniform secondary structure. The structure is different from the calculated structure from trifluoroethanol/water measurements.     

Solution structure of a LewisX analogue by off-resonance 1H NMR spectroscopy without use of an internal distance reference

Summary A recent ¹H NMR method has been applied to the determination of the solution structure and internal dynamics of a synthetic mixed C/O trisaccharide related to sialyl Lewis^x. Varying the rf field offset in ROESY-type experiments enabled the measurement of longitudinal and transverse dipolar cross-relaxation rates with high accuracy. Assuming that for each proton pair the motion could be represented by a single exponential autocorrelation function, it was possible to derive geometrical parameters (r) and dynamic parameters _cp. With this assumption, 224 cross-relaxation rates have been transformed into 30 interproton distance constraints and 30 dipolar correlation times. The distance constraints have been used in a simulated-annealing procedure. This trisaccharide exhibits a structure close to the O-glycosidic analogue, but its flexibility seems highly reduced. On the basis of the determined structure and dynamics, it is shown that no conformational exchange occurs, the molecule existing in the form of a unique family in aqueous solution. In order to assess the quality of the resulting structures and to validate this new experimental procedure of distance extraction, we finally compare these solution structures to the ones obtained using three different sets of distances deduced from three choices of internal reference. It appears that this procedure allows the determination of the most precise and accurate solution.Abbreviations COSY correlation spectroscopy - NOE nuclear Overhauser enhancement - NOESY nuclear Overhauser enhancement spectroscopy; rmsd, root-mean-square deviation - ROESY rotating frame Overhauser enhancement spectroscopy - SLe^x sialyl Lewis^x - TOCSY total correlation spectroscopy     

A nuclear Overhauser enhancement study on the imino proton resonances of a DNA pentadecamer comprising the specific target site of the cyclic AMP receptor protein in the ara BAD operon

A 500 MHz 1H-NMR study on a synthetic DNA pentadecamer comprising the specific target site of the cAMP receptor protein in the ara BAD operon is presented. Using pre-steady state NOE measurements, unambiguous assignments of all the imino proton resonances and associated adenine (H2) resonances are obtained. From the NOE data interbase pair interproton distances involving the imino and adenine (H2) protons are determined. It is shown that these distances are very similar to those expected for classical B DNA (RMS difference of 0.5 A), but are significantly different from those expected for classical A DNA (RMS difference of 1.1 A).     

Refinement of the spatial structure of the gramicidin A ion channel]

The spatial structure of the gramicidin A (GA) transmembrane ion-channel was refined on the base of cross-peak volumes measured in NOESY spectra (mixing time tau m = 100 and 200 ms). The refinement methods included the comparison of experimental cross-peak volumes with those calculated for low-energy GA conformations, dynamic averaging of the low-energy conformation set and restrained energy minimization. Accuracy of the spatial structure determination was estimated by the penalty function Fr defined as a root mean square deviation of interproton distances corresponding to the calculated and experimental cross-peak volumes. As the initial conformation we used the right-handed pi 6,3 LD pi 6,3 LD helix established on the base of NMR data regardless of the cross-peak volumes. The conformation is in a good agreement with NOE cross-peak volumes (Fr 0.2 to 0.5 A depending on NOESY spectrum). For a number of NOEs formed by the side chain protons, distances errors were found as much as 0.5-2.0 A. Restrained energy minimization procedure had little further success. However some of these errors were eliminated by the change in torsional angle chi 2 of D-Leu12 and dynamic averaging of the Val7 side chain conformations. Apparently, majority of deviations of the calculated and experimental cross-peak volumes are due to the intramolecular mobility of GA and cannot be eliminated within the framework of rigid globule model. In summary the spatial structure of GA ion-channel can be thought as a set of low-energy conformations, differing by the side chain torsion angles chi 1 Val7 and chi 2 D-Leu4 and D-Leu10 and the orientation of the C-terminal ethanolamine group. Root mean square differences between the atomic coordinates of conformations are in the range of 0.3-0.8 A.     

Determination of three-dimensional protein structures from nuclear magnetic resonance data using fragments of known structures

A method to build a three-dimensional protein model from nuclear magnetic resonance (NMR) data using fragments from a data base of crystallographically determined protein structures is presented. The interproton distances derived from the nuclear Overhauser effect (NOE) data are compared to the precalculated distances in the known protein structures. An efficient search algorithm is used, which arranges the distances in matrices akin to a C alpha diagonal distance plot, and compares the NOE distance matrices for short sequential zones of the protein to the data base matrices. After cluster analysis of the fragments found in this way, the structure is built by aligning fragments in overlapping zones. The sequentially long-range NOEs cannot be used in the initial fragments search but are vital to discriminate between several possible combinations of different groups of fragments. The method has been tested on one simulated NOE data set derived from a crystal structure and one experimental NMR data set. The method produces models that have good local structure, but may contain larger global errors. These models can be used as the starting point for further refinement, e.g., by restrained molecular dynamics or interactive graphics.