Improving the accuracy of NMR structures of DNA by means of a database potential of mean force describing base-base positional interactions.

NMR structure determination of nucleic acids presents an intrinsically difficult problem since the density of short interproton distance contacts is relatively low and limited to adjacent base pairs. Although residual dipolar couplings provide orientational information that is clearly helpful, they do not provide translational information of either a short-range (with the exception of proton-proton dipolar couplings) or long-range nature. As a consequence, the description of the nonbonded contacts has a major impact on the structures of nucleic acids generated from NMR data. In this paper, we describe the derivation of a potential of mean force derived from all high-resolution (2 A or better) DNA crystal structures available in the Nucleic Acid Database (NDB) as of May 2000 that provides a statistical description, in simple geometric terms, of the relative positions of pairs of neighboring bases (both intra- and interstrand) in Cartesian space. The purpose of this pseudopotential, which we term a DELPHIC base-base positioning potential, is to bias sampling during simulated annealing refinement to physically reasonable regions of conformational space within the range of possibilities that are consistent with the experimental NMR restraints. We illustrate the application of the DELPHIC base-base positioning potential to the structure refinement of a DNA dodecamer, d(CGCGAATTCGCG)(2), for which NOE and dipolar coupling data have been measured in solution and for which crystal structures have been determined. We demonstrate by cross-validation against independent NMR observables (that is, both residual dipolar couplings and NOE-derived intereproton distance restraints) that the DELPHIC base-base positioning potential results in a significant increase in accuracy and obviates artifactual distortions in the structures arising from the limitations of conventional descriptions of the nonbonded contacts in terms of either Lennard-Jones van der Waals and electrostatic potentials or a simple van der Waals repulsion potential. We also demonstrate, using experimental NMR data for a complex of the male sex determining factor SRY with a duplex DNA 14mer, which includes a region of highly unusual and distorted DNA, that the DELPHIC base-base positioning potential does not in any way hinder unusual interactions and conformations from being satisfactorily sampled and reproduced. We expect that the methodology described in this paper for DNA can be equally applied to RNA, as well as side chain-side chain interactions in proteins and protein-protein complexes, and side chain-nucleic acid interactions in protein-nucleic acid complexes. Further, this approach should be useful not only for NMR structure determination but also for refinement of low-resolution (3-3.5 A) X-ray data.

Sources of and solutions to problems in the refinement of protein NMR structures against torsion angle potentials of mean force.

It is often the case that a substantial number of torsion angles (both backbone and sidechain) in structures of proteins and nucleic acids determined by NMR are found in physically unlikely and energetically unfavorable conformations. We have previously proposed a database-derived potential of mean force comprising one-, two-, three-, and four-dimensional potential surfaces which describe the likelihood of various torsion angle combinations to bias conformational sampling during simulated annealing refinement toward those regions that are populated in very high resolution (< or =1.75 A) crystal structures. We now note a shortcoming of our original implementation of this approach: namely, the forces it places on atoms are very rough. When the density of experimental restraints is low, this roughness can both hinder convergence to commonly populated regions of torsion angle space and reduce overall conformational sampling. In this paper we describe a modification that completely eliminates these problems by replacing the original potential surfaces by a sum of multidimensional Gaussian functions. Structures refined with the new Gaussian implementation now simultaneously enjoy excellent global sampling and excellent local choices of torsion angles.

Crystallography & NMR system: A new software suite for macromolecular structure determination.

A new software suite, called Crystallography & NMR System (CNS), has been developed for macromolecular structure determination by X-ray crystallography or solution nuclear magnetic resonance (NMR) spectroscopy. In contrast to existing structure-determination programs, the architecture of CNS is highly flexible, allowing for extension to other structure-determination methods, such as electron microscopy and solid-state NMR spectroscopy. CNS has a hierarchical structure: a high-level hypertext markup language (HTML) user interface, task-oriented user input files, module files, a symbolic structure-determination language (CNS language), and low-level source code. Each layer is accessible to the user. The novice user may just use the HTML interface, while the more advanced user may use any of the other layers. The source code will be distributed, thus source-code modification is possible. The CNS language is sufficiently powerful and flexible that many new algorithms can be easily implemented in the CNS language without changes to the source code. The CNS language allows the user to perform operations on data structures, such as structure factors, electron-density maps, and atomic properties. The power of the CNS language has been demonstrated by the implementation of a comprehensive set of crystallographic procedures for phasing, density modification and refinement. User-friendly task-oriented input files are available for nearly all aspects of macromolecular structure determination by X-ray crystallography and solution NMR.

Improving the quality of NMR and crystallographic protein structures by means of a conformational database potential derived from structure databases.

A new conformational database potential involving dihedral angle relationships in databases of high-resolution highly refined protein crystal structures is presented as a method for improving the quality of structures generated from NMR data. The rationale for this procedure is based on the observation that uncertainties in the description of the nonbonded contacts present a key limiting factor in the attainable accuracy of protein NMR structures and that the nonbonded interaction terms presently used have poor discriminatory power between high- and low-probability local conformations. The idea behind the conformational database potential is to restrict sampling during simulated annealing refinement to conformations that are likely to be energetically possible by effectively limiting the choices of dihedral angles to those that are known to be physically realizable. In this manner, the variability in the structures produced by this method is primarily a function of the experimental restraints, rather than an artifact of a poor nonbonded interaction model. We tested this approach with the experimental NMR data (comprising an average of about 30 restraints per residue and consisting of interproton distances, torsion angles, 3JHN alpha coupling constants, and 13C chemical shifts) used previously to calculate the solution structure of reduced human thioredoxin (Qin J, Clore GM, Gronenborn AM, 1994, Structure 2:503-522). Incorporation of the conformational database potential into the target function used for refinement (which also includes terms for the experimental restraints, covalent geometry, and nonbonded interactions in the form of either a repulsive, repulsive-attractive, or 6-12 Lennard-Jones potential) results in a significant improvement in various quantitative measures of quality (Ramachandran plot, side-chain torsion angles, overall packing). This is achieved without compromising the agreement with the experimental restraints and the deviations from idealized covalent geometry that remain within experimental error, and the agreement between calculated and observed 1H chemical shifts that provides an independent NMR parameter of accuracy. The method is equally applicable to crystallographic refinement, and should be particular useful during the early stages of either an NMR or crystallographic structure determination and in cases where relatively few experimental restraints can be derived from the measured data (due, for example, to broad lines in the NMR spectra or to poorly diffracting crystals).

The solution structure of human thioredoxin complexed with its target from Ref-1 reveals peptide chain reversal.

BACKGROUND: Human thioredoxin (hTRX) is a 12 kDa cellular redox protein that has been shown to play an important role in the activation of a number of transcriptional and translational regulators via a thiol-redox mechanism. This activity may be direct or indirect via another redox protein known as Ref-1. The structure of a complex of hTRX with a peptide comprising its target from the transcription factor NF kappa B has previously been solved. To further extend our knowledge of the recognition by and interaction of hTRX with its various targets, we have studied a complex between hTRX and a Ref-1 peptide. This complex represents a kinetically stable mixed disulfide intermediate along the reaction pathway. RESULTS: Using multidimensional heteronuclear edited and filtered NMR spectroscopy, we have solved the solution structure of a complex between hTRX and a 13-residue peptide comprising residues 59-71 of Ref-1. The Ref-1 peptide is located in a crescent-shaped groove on the surface of hTRX, the groove being formed by residues in the active-site loop (residues 32-36), helix 3, beta strands 3 and 5, and the loop between beta strands 3 and 4. The complex is stabilized by numerous hydrogen-bonding and hydrophobic interactions that involve residues 61-69 of the peptide and confer substrate specificity. CONCLUSIONS: The orientation of the Ref-1 peptide in the hTRX-Ref-1 complex is opposite to that found in the previously solved complex of hTRX with the target peptide from the transcription factor NF kappa B. Orientation is determined by three discriminating interactions involving the nature of the residues at the P-2' P-4 and P-5 binding positions. (P0 defines the active cysteine of the peptide, Cys65 for Ref-1 and Cys62 for NF kappa B. Positive and negative numbers indicate residues N-terminal and C-terminal to this residue, respectively, and vice versa for NF kappa B as it binds in the opposite orientation.) The environment surrounding the reactive Cys32 of hTRX, as well as the packing of the P+3 to P-4 residues are essentially the same in the two complexes, despite the opposing orientation of the peptide chains. This versatility in substrate recognition permits hTRX to act as a wide-ranging redox regulator for the cell.

Solution structure of the DNA binding domain of HIV-1 integrase.

The solution structure of the DNA binding domain of HIV-1 integrase (residues 220-270) has been determined by multidimensional NMR spectroscopy. The protein is a dimer in solution, and each subunit is composed of a five-stranded beta-barrel with a topology very similar to that of the SH3 domain. The dimer is formed by a stacked beta-interface comprising strands 2, 3, and 4, with the two triple-stranded antiparallel beta-sheets, one from each subunit, oriented antiparallel to each other. One surface of the dimer, bounded by the loop between strands beta 1 and beta 2, forms a saddle-shaped groove with dimensions of approximately 24 x 23 x 12 A in cross section. Lys264, which has been shown from mutational data to be involved in DNA binding, protrudes from this surface, implicating the saddle-shaped groove as the potential DNA binding site.

Fast folding of a prototypic polypeptide: the immunoglobulin binding domain of streptococcal protein G.

The folding of the small (56 residues) highly stable B1 immunoglobulin binding domain (GB1) of streptococcal protein G has been investigated by quenched-flow deuterium-hydrogen exchange. This system represents a paradigm for the study of protein folding because it exhibits no complicating features superimposed upon the intrinsic properties of the polypeptide chain. Collapse to a semicompact state exhibiting partial order, reflected in protection factors for ND-NH exchange up to 10-fold higher than that expected for a random coil, occurs within the dead time (< or = 1 ms) of the quenched flow apparatus. This is followed by the formation of the fully native state, as monitored by the fractional proton occupancy of 26 backbone amide groups spread throughout the protein, in a single rapid concerted step with a half-life of 5.2 ms at 5 degrees C.

High-resolution solution structure of the beta chemokine hMIP-1 beta by multidimensional NMR.

The three-dimensional structure of a member of the beta subfamily of chemokines, human macrophage inflammatory protein-1 beta (hMIP-1 beta), has been determined with the use of solution multidimensional heteronuclear magnetic resonance spectroscopy. Human MIP-1 beta is a symmetric homodimer with a relative molecular mass of approximately 16 kilodaltons. The structure of the hMIP-1 beta monomer is similar to that of the related alpha chemokine interleukin-8 (IL-8). However, the quaternary structures of the two proteins are entirely distinct, and the dimer interface is formed by a completely different set of residues. Whereas the IL-8 dimer is globular, the hMIP-1 beta dimer is elongated and cylindrical. This provides a rational explanation for the absence of cross-binding and reactivity between the alpha and beta chemokine subfamilies. Calculation of the solvation free energies of dimerization suggests that the formation and stabilization of the two different types of dimers arise from the burial of hydrophobic residues.