Efficient computation of three-dimensional protein structures in solution from nuclear magnetic resonance data using the program DIANA and the supporting programs CALIBA, HABAS and GLOMSA.

A novel procedure for efficient computation of three-dimensional protein structures from nuclear magnetic resonance (n.m.r.) data in solution is described, which is based on using the program DIANA in combination with the supporting programs CALIBA, HABAS and GLOMSA. The first part of this paper describes the new programs DIANA. CALIBA and GLOMSA. DIANA is a new, fully vectorized implementation of the variable target function algorithm for the computation of protein structures from n.m.r. data. Its main advantages, when compared to previously available programs using the variable target function algorithm, are a significant reduction of the computation time, and a novel treatment of experimental distance constraints involving diastereotopic groups of hydrogen atoms that were not individually assigned. CALIBA converts the measured nuclear Overhauser effects into upper distance limits and thus prepares the input for the previously described program HABAS and for DIANA. GLOMSA is used for obtaining individual assignments for pairs of diastereotopic substituents by comparison of the experimental constraints with preliminary results of the structure calculations. With its general outlay, the presently used combination of the four programs is particularly user-friendly. In the second part of the paper, initial results are presented on the influence of the novel DIANA treatment of diastereotopic protons on the quality of the structures obtained, and a systematic study of the central processing unit times needed for the same protein structure calculation on a range of different, commonly available computers is described.     

Structure determination of membrane-associated proteins from nuclear magnetic resonance data

This Review covers the delineation and optimization of protein-lipid systems for study using solution-state NMR spectroscopy. The first half presents the necessary background for a membrane protein biochemist to initiate collaboration with an NMR spectroscopist. The second half provides guidelines for the spectroscopist on data collection, analysis for obtaining conformational information, and structure generation and assessment. Although the emphasis is on the study of peptides in detergent micelles, methods are outlined for larger membrane-associated proteins and for use of other solubilizing agents.     

Structure determination of biological macromolecules in solution using nuclear magnetic resonance spectroscopy

A detailed understanding of the function of a biological macromolecule requires knowledge of its three-dimensional structure. Most atomic-resolution structures of biological macromolecules have been solved either by X-ray diffraction in single crystals or by nuclear magnetic resonance (NMR) in solution. This review surveys the method of NMR structure determination. First, a brief introduction to NMR and its basic concepts is presented. The main part of the article deals with the individual steps necessary for an NMR structure determination. At the end, the discussion turns to considerations on the influence of the molecular size of the macromolecules on the structure determination by NMR. New techniques are discussed that greatly enhance the possibilities of applying NMR to large molecular systems.     

Secondary structure determination for the Antennapedia homeodomain by nuclear magnetic resonance and evidence for a helix-turn-helix motif.

The homeodomain encoded by the Antennapedia (Antp) gene of Drosophila was studied in aqueous solution by nuclear magnetic resonance (NMR). Sequence-specific resonance assignments have been obtained for the complete polypeptide chain of 68 amino acid residues. The secondary structure determined from nuclear Overhauser effects (NOE) and information about slowly exchanging amide protons includes three helical segments consisting of the residues 10-21, 28-38 and 42-52, respectively. Combination of the presently available NMR data with computer modeling provided preliminary evidence for the presence of a helix-turn-helix motif in the homeodomain. Near the turn, this supersecondary structure appears to be very similar to the DNA binding site in the 434 and P22 c2 repressors, but both helices in the homeodomain include 2-3 additional residues when compared with these prokaryotic DNA-binding proteins.     

Tertiary structure of human complement component C5a in solution from nuclear magnetic resonance data

The tertiary structure for the region 1-63 of the 74 amino acid human complement protein C5a in solution was calculated from a large number of distance constraints derived from nuclear Overhauser effects with an angular distance geometry algorithm. The protein consists of four helices juxtaposed in an approximately antiparallel topology connected by peptide loops located at the surface of the molecule. The structures obtained for the helices are compatible with alpha-helical hydrogen-bonding patterns, which provides an explanation for the observed slow solvent exchange kinetics of the amide protons in these peptide regions. In contrast to the peptide region 1-63, no defined structure could be assigned to the C-terminal region 64-74, which increasingly acquires dynamic random coil characteristics as the end of the peptide chain is approached. An average root-mean-square deviation of 1.6 A was obtained for the alpha-carbons of the first 63 residues in the calculated ensemble of C5a structures, while the alpha-helices were determined with an average root-mean-square deviation of 0.8 A for the alpha-carbons. A comparison between the solution structure of C5a and the crystal structure of the functionally related C3a protein, as well as inferences for the interaction of C5a with its receptor on polymorphonuclear leukocytes, is discussed.     

Determination of the three-dimensional structure of the Antennapedia homeodomain from Drosophila in solution by 1H nuclear magnetic resonance spectroscopy

The determination of the three-dimensional structure of the Antennapedia homeodomain from Drosophila in solution is described. The techniques used are 1H nuclear magnetic resonance spectroscopy for the data collection, and calculation of the protein structure with the program DISMAN followed by restrained energy minimization with a modified version of the program AMBER. A group of 19 conformers characterizes a well-defined structure for residues 7 to 59, with an average root-mean-square distance from the backbone atoms of 0.6 A relative to the mean of the 19 structures. The structure contains a helix from residues 10 to 21, a helix-turn-helix motif from residues 28 to 52, which is similar to those reported for several prokaryotic repressor proteins, and a somewhat flexible fourth helix from residues 53 to 59, which essentially forms an extension of the presumed recognition helix, residues 42 to 52. The helices enclose a structurally well-defined molecular core of hydrophobic amino acid side-chains.     

A protein structure from nuclear magnetic resonance data. lac repressor headpiece

A procedure is described to determine the three-dimensional structure of biomolecules from nuclear magnetic resonance data. This procedure combines model building with a restrained molecular dynamics algorithm, in which distance information from nuclear Overhauser effects is incorporated in the form of pseudo potentials. The method has been applied to the N-terminal DNA-binding domain or headpiece (amino acid residues 1 to 51) of the lac repressor from Escherichia coli, for which no crystal structure is available. The relative orientation of the three helices of the headpiece is similar to that of the three homologous helices found in the cI repressor of bacteriophage lambda.     

Three-dimensional structure of porcine C5adesArg from 1H nuclear magnetic resonance data

Two-dimensional nuclear magnetic resonance spectra of porcine C5adesArg (73 residues) have been used to construct a list of 34 hydrogen bonds, 27 dihedral angle constraints, and 151 distance constraints, derived from nuclear Overhauser effect data. These constraints were used in restrained molecular dynamics calculations on residues 1-65 of C5a, starting from a folded structure modeled on the crystal structure of a homologous protein, C3a. Forty-one structures have been calculated, which fall into three similar families with few violations of the imposed constraints. Structures in the most populated family have a root-mean-square deviation from the average structure of 1.02 A for the C alpha atoms, with good definition of the internal residues. There is good agreement between the calculated structures and other nuclear magnetic resonance data. The structure is very similar to that recently reported for human C5a [Zuiderweg et al. (1989) Biochemistry 28, 172-185]. Some biological implications of these structures are discussed.     

Determination of the three-dimensional solution structure of barnase using nuclear magnetic resonance spectroscopy

The solution conformation of the ribonuclease barnase has been determined by using 1H nuclear magnetic resonance (NMR) spectroscopy. The 20 structures were calculated by using 853 interproton distance restraints obtained from analyses of two-dimensional nuclear Overhauser spectra, 72 phi and 53 chi 1 torsion angle restraints, and 17 hydrogen-bond distance restraints. The calculated structures contain two alpha-helices (residues 6-18 and 26-34) and a five-stranded antiparallel beta-sheet (residues 50-55, 70-75, 85-91, 94-101, and 105-108). The core of the protein is formed by the packing of one of the alpha-helices (residues 6-18) onto the beta-sheet. The average RMS deviation between the calculated structures and the mean structure is 1.11 A for the backbone atoms and 1.75 A for all atoms. The protein is least well-defined in the N-terminal region and in three large loops. When these regions are excluded, the average RMS deviation between the calculated structures and the mean structure for residues 5-34, 50-56, 71-76, 85-109 is 0.62 A for the backbone atoms and 1.0 A for all atoms. The NMR-derived structure has been compared with the crystal structure of barnase [Mauguen et al. (1982) Nature (London) 297, 162-164].     

Secondary structure in solution of barwin from barley seed using 1H nuclear magnetic resonance spectroscopy.

Barwin, a basic protein from barley seed of 125 amino acid residues, has been studied by two-dimensional 1H nuclear magnetic resonance spectroscopy. This protein is closely related to the C-terminal domain of proteins whose synthesis is induced by wounding, the so-called win proteins. These proteins may, therefore, have a role in the defense against fungal attack. Full assignment of the 1H nuclear magnetic resonances has been obtained for 104 amino acid residues, and 18 amino acid spin systems were partially assigned. Sequence-specific assignment using nuclear Overhauser spectroscopy has been achieved for 122 of the 125 residues. This has revealed that the secondary structure of the protein is dominated by a large four-stranded antiparallel beta-sheet consisting of the strands Gln2-Thr9, Lys65-Asn71, Gln77-Arg81, and His113-Val121, a small parallel beta-sheet of the strands Trp48-Cys52 and Asp84-Ala87, which together account for a third of the protein. Sequential effects indicate the presence of three small alpha-helices, Tyr30-Lys38, Leu40-Tyr46, and Thr97-Asp103. The secondary structure in other regions of the sequence is characterized mainly by loops and turns and regions where no regular secondary structure arrangement could be identified. A large number of long-range nuclear Overhauser effects has been identified, and these have been used, together with sequential and intranuclear Overhauser effects, for a calculation of the protein's three-dimensional structure.     

Application of nuclear magnetic resonance for the determination of the structure of proteins in solution

Knowledge of three-dimensional structure is a key factor in protein engineering. It is useful, for example, in predicting and understanding the functional consequences of specific substitution of one or more amino acids of the polypeptide chain. It is also necessary for the design of new effectors or analogs of the substrates of enzymes and receptors. X-ray diffraction by crystals of the biomolecule was for a long time the only method of determining three-dimensional structures. In the last 5 years, it has been joined by a new technique, two-dimensional nuclear magnetic resonance (2D NMR), which can resolve the structure of middle-sized proteins (less than 10 kilodaltons). The technique is applied on solutions whose pH, ionic strength, and temperature can be chosen and changed. The two basic measurements, COSY and NOESY, detect respectively the systems of hydrogen nuclei, or protons, coupled through covalent bonds, and those in which the interproton distances are less than 0.5 nm. A systematic strategy leads from resonance assignments of the two-dimensional spectrum to molecular modeling with constraints and finally to the determination of the molecular structure in the solution. Much sophistication is needed even today for the first task, the assignment of the resonances. Each of the COSY and NOESY spectra is a two-dimensional map, where the diagonal line is the one-dimensional spectrum, and the off-diagonal peaks indicate connectives between protons. Peak assignment to a specific type of amino acid is based on the pattern of scalar couplings observed in the COSY spectrum. Next, the amino acids are positioned in the primary sequence, using the spatial proximities of polypeptide chain protons, as observed in the NOESY spectrum. The principal secondary structures (alpha helix, beta sheets, etc.) are then identified by their specific connectivities. The tertiary structure is detected by NOESY connectivities between protons of different amino acids which are far apart in the primary sequence. The distance constraints from the NOESY connectivities also provide the starting point for modeling the tertiary structure. This is then refined using distance geometry and molecular dynamics algorithms. The resolution of the structures obtained with the help of recent algorithmic developments may be comparable to that provided by X-ray diffraction. The COSY measurement can be completed or substituted by other measurements, useful albeit more complex. For example, the HOHAHA experiment, currently in wide use, gives the correlations through multiple covalent bonds. Multiquanta experiments, which select systems of a given number of coupled spins, provide spectral simplification.(ABSTRACT TRUNCATED AT 400 WORDS)     

A polynomial-time algorithm for de novo protein backbone structure determination from nuclear magnetic resonance data.

We describe an efficient algorithm for protein backbone structure determination from solution Nuclear Magnetic Resonance (NMR) data. A key feature of our algorithm is that it finds the conformation and orientation of secondary structure elements as well as the global fold in polynomial time. This is the first polynomial-time algorithm for de novo high-resolution biomacromolecular structure determination using experimentally recorded data from either NMR spectroscopy or X-ray crystallography. Previous algorithmic formulations of this problem focused on using local distance restraints from NMR (e.g., nuclear Overhauser effect [NOE] restraints) to determine protein structure. This approach has been shown to be NP-hard, essentially due to the local nature of the constraints. In practice, approaches such as molecular dynamics and simulated annealing, which lack both combinatorial precision and guarantees on running time and solution quality, are used routinely for structure determination. We show that residual dipolar coupling (RDC) data, which gives global restraints on the orientation of internuclear bond vectors, can be used in conjunction with very sparse NOE data to obtain a polynomial-time algorithm for structure determination. Furthermore, an implementation of our algorithm has been applied to six different real biological NMR data sets recorded for three proteins. Our algorithm is combinatorially precise, polynomialtime, and uses much less NMR data to produce results that are as good or better than previous approaches in terms of accuracy of the computed structure as well as running time.     

Three-dimensional structure of proteolytic fragment 163-231 of bacterioopsin determined from nuclear magnetic resonance data in solution.

546 NOESY cross-peak volumes were measured in the two-dimensional NOESY spectrum of proteolytic fragment 163-231 of bacterioopsin in organic solution. These data and 42 detected hydrogen bonds were applied for determining the peptide spatial structure. The fold of the polypeptide chain was determined by local structure analysis, a distance geometry approach and systematic search for energetically allowed side-chain rotamers which are consistent with experimental NOESY cross-peak volumes. The effective rotational correlation time of 6 ns for the molecule was evaluated from optimization of the local structure to meet NOE data and from the dependence on mixing time of the NiH/Ci alpha H cross-peak volumes of the residues in alpha-helical conformation. The resulting structure has two well defined alpha-helical regions, 168-191 and 198-227, with root-mean-square deviation 44 pm and 69 pm, respectively, between the backbone atoms in 14 final energy refined conformations. The alpha-helices correspond to transmembrane segments F and G of bacteriorhodopsin. The segment F contains proline 186, which introduces a kink of about 25 degrees with a disruption of the hydrogen bond with the NH group of the following residue. The segments are connected by a flexible loop region 192-197. Torsion angles chi 1 are unequivocally defined for 62% of side chains in the alpha-helices but half of them differ from electron cryo-microscopy (ECM) model of bacteriorhodopsin, apparently because of the low resolution of ECM. Nevertheless, the F and G segments can be packed as in the ECM model and with side-chain conformations consistent with all NMR data in solution.     

A high quality nuclear magnetic resonance solution structure of peptide deformylase from Escherichia coli: Application of an automated assignment strategy using GARANT

The NMR structure of the peptide deformylase (PDF) (1–150) from Escherichia coli, which is an essential enzyme that removes the formyl group from nascent polypeptides and represents a potential target for drug discovery, was determined using 15N/13C doubly labeled protein. Nearly completely automated assignment routines were employed to assign three-dimensional triple resonance, 15N-resolved and 13C-resolved NOESY spectra using the program GARANT. This assignment strategy, demonstrated on a 17 kDa protein, is a significant advance in the automation of NMR data assignment and structure determination that will accelerate future work. A total of 2302 conformational constraints were collected as input for the distance geometry program DYANA. After restrained energy minimization with the program X-PLOR the 20 best conformers characterize a high quality structure with an average of 0.43 Å for the root-mean-square deviation calculated from the backbone atoms N, C and C, and 0.81 Å for all heavy atoms of the individual conformers relative to the mean coordinates for residues 1 to 150. The globular fold of PDF contains two -helices comprising residues 25–40, 125–138, six -strands 57–60, 70–77, 85–88, 98–101, 105–111, 117–123 and one 310 helix comprising residues 49–51. The C-terminal helix contains the HEXXH motif positioning a zinc ligand in a similar fashion to other metalloproteases, with the third ligand being cysteine and the fourth presumably a water. The three-dimensional structure of PDF affords insight into the substrate recognition and specificity for N-formylated over N-acetylated substrates and is compared to other PDF structures.     

Direct determination of a membrane-peptide interface using the nuclear magnetic resonance cross-saturation method

Membrane-peptide interactions are involved in many crucial biological and pharmacological activities. To clarify the interaction mode of membrane-peptide complexes, it is important to analyze both the dynamic properties and the contact residues of the membrane-bound peptide. In this study, we investigated the dynamic properties of a peptide bound to a lipid bilayer, using relaxation and amide-water exchange analyses, and directly determined the membrane-peptide interface, using the cross-saturation method. For the models of a lipid bilayer and a peptide, isotropic bicelles and mastoparan were used, respectively. The results indicate that mastoparan had a heterogeneous distribution of motion over various timescales and interacted with the lipid bilayer by using its hydrophobic side; the molecule was located within the lipid bilayer rather than on the surface, as thought previously. This study shows that the cross-saturation method is useful for determining the interface of not only protein-protein but also membrane-peptide complexes.     

Three-dimensional solution structure of the HIV-1 protease complexed with DMP323, a novel cyclic urea-type inhibitor, determined by nuclear magnetic resonance spectroscopy.

The three-dimensional solution structure of the HIV-1 protease homodimer, MW 22.2 kDa, complexed to a potent, cyclic urea-based inhibitor, DMP323, is reported. This is the first solution structure of an HIV protease/inhibitor complex that has been elucidated. Multidimensional heteronuclear NMR spectra were used to assemble more than 4,200 distance and angle constraints. Using the constraints, together with a hybrid distance geometry/simulated annealing protocol, an ensemble of 28 NMR structures was calculated having no distance or angle violations greater than 0.3 A or 5 degrees, respectively. Neglecting residues in disordered loops, the RMS deviation (RMSD) for backbone atoms in the family of structures was 0.60 A relative to the average structure. The individual NMR structures had excellent covalent geometry and stereochemistry, as did the restrained minimized average structure. The latter structure is similar to the 1.8-A X-ray structure of the protease/DMP323 complex (Chang CH et al., 1995, Protein Science, submitted); the pairwise backbone RMSD calculated for the two structures is 1.22 A. As expected, the mismatch between the structures is greatest in the loops that are disordered and/or flexible. The flexibility of residues 37-42 and 50-51 may be important in facilitating substrate binding and product release, because these residues make up the respective hinges and tips of the protease flaps. Flexibility of residues 4-8 may play a role in protease regulation by facilitating autolysis.     

Proton resonance assignments and three-dimensional solution structure of the ragweed allergen Amb a V by nuclear magnetic resonance spectroscopy.

Essentially complete assignment of the proton resonances in the allergenic protein Amb a V has been made by analysis of two-dimensional NMR experiments. Conformational constraints were obtained in three forms: interproton distances derived from NOE cross-peak intensities of NOESY spectra, torsion angle constraints derived from J-coupling constants of COSY and PE-COSY spectra, and hydrogen bond constraints derived from hydrogen-exchange experiments. Conformations of Amb a V with low constraint violations were generated using dynamic simulated annealing in the program XPLOR. The refined structures are comprised of a C-terminal alpha-helix, a small segment of antiparallel beta-sheet, and several loops. A hydrophobic core exists at the interface of the alpha-helix and beta-sheet. The derived structure accounts for the several anomalous proton chemical shifts that are observed. The structure determined here for Amb a V is topologically similar to the structure determined previously for the homologous allergenic protein Amb t V [Metzler, W. J., Valentine, K., Roebber, M., Friedrichs, M. S., Marsh, D., & Mueller, L. (1992) Biochemistry 31, 5117-5127]; however, significant differences exist in the packing of side chains in the hydrophobic core of the molecules. Comparison of the detailed structural features of these two proteins will allow us to suggest surface substructures for the Amb V allergens that are likely to participate in B cell epitopes.     

Proton nuclear magnetic resonance assignments and secondary structure determination of the ColE1 rop (rom) protein

The complete resonance assignment of the ColE1 rop (rom) protein at pH 2.3 was obtained by two-dimensional (2D) proton nuclear magnetic resonance spectroscopy (1H NMR) at 500 and 600 MHz using through-bond and through-space connectivities. Sequential assignments and elements of regular secondary structure were deduced by analysis of nuclear Overhauser enhancement spectroscopy (NOESY) experiments and 3JHN alpha coupling constants. One 7.2-kDa monomer of the homodimer consists of two antiparallel helices connected by a hairpin loop at residue 31. The C-terminal peptide consisting of amino acids 59-63 shows no stable conformation. The dimer forms a four-helix bundle with opposite polarization of neighboring elements in agreement with the X-ray structure.     

Biochemical characterization and nuclear magnetic resonance structure of novel alpha-conotoxins isolated from the venom of Conus consors.

Two novel alpha-conotoxins were purified and characterized from the venom of the fish-hunting cone snail Conus consors. These peptides were identified by screening HPLC fractions of the crude venom and by binding experiments with Torpedo nicotinic acetylcholine receptor. The toxins named alpha-CnIA and alpha-CnIB exhibited sequences of 14 and 12 amino acids, respectively. The alpha-CnIA represents the main alpha-conotoxin contained in the venom, whereas alpha-CnIB is present in a relatively small amount. Chemical synthesis of alpha-CnIA was carried out using the Fmoc methodology by selective disulfide bond formation. The biological activity of the toxin was assessed in fish and mice. The alpha-CnIA inhibited the fixation of iodinated alpha-bungarotoxin to Torpedo nicotinic acetylcholine receptors with an IC50 of 0.19 microM which can be compared to the IC50 of 0.31 microM found for the previously characterized alpha-MI isolated from the piscivorous Conus magus. The synthetic alpha-CnIA blocked spontaneous and evoked synaptic potentials in frog and mouse isolated neuromuscular preparations at sub-micromolar concentrations. Solution NMR of this toxin indicated a conformational heterogeneity with the existence of different conformers in solution, at slow and intermediate exchange rates relative to the NMR chemical shift time scale, similar to that reported for alpha-GI and alpha-MI. NMR structures were calculated for the major NMR signals representing more than 80% of the population at 5 degrees C.     

Protein--DNA contacts in the structure of a homeodomain--DNA complex determined by nuclear magnetic resonance spectroscopy in solution.

The 1:1 complex of the mutant Antp(C39----S) homeodomain with a 14 bp DNA fragment corresponding to the BS2 binding site was studied by nuclear magnetic resonance (NMR) spectroscopy in aqueous solution. The complex has a molecular weight of 17,800 and its lifetime is long compared with the NMR chemical shift time scale. Investigations of the three-dimensional structure were based on the use of the fully 15N-labelled protein, two-dimensional homonuclear proton NOESY with 15N(omega 2) half-filter, and heteronuclear three-dimensional NMR experiments. Based on nearly complete sequence-specific resonance assignments, both the protein and the DNA were found to have similar conformations in the free form and in the complex. A sufficient number of intermolecular 1H-1H Overhauser effects (NOE) could be identified to enable a unique docking of the protein on the DNA, which was achieved with the use of an ellipsoid algorithm. In the complex there are intermolecular NOEs between the elongated second helix in the helix-turn-helix motif of the homeodomain and the major groove of the DNA. Additional NOE contacts with the DNA involve the polypeptide loop immediately preceding the helix-turn-helix segment, and Arg5. This latter contact is of special interest, both because Arg5 reaches into the minor groove and because in the free Antp(C39----S) homeodomain no defined spatial structure could be found for the apparently flexible N-terminal segment comprising residues 0-6.