Fiber Diffraction Data Research Articles

Solvation Effects on the Conformational Behaviour of Gellan and Calcium Ion Binding to Gellan Double Helices

The contribution of the presence of solvent to the conformations adopted by disaccharide fragments within the repeat unit of gellan have been studied by molecular modelling techniques. Initial conformational energy searches, using a dielectric continuum to represent the solvent, provided starting geometries for a series of molecular dynamics simulations. The solution behaviour from these simulations was subsequently compared to fibre diffraction data of the potassium gellan salt. The present calculations indicate considerable flexibility of the glycosidic linkages, and this is discussed in relation to its effect on gel formation. One of the fragments was solvated with explicit water molecules. These calculations showed the same conformational behaviour as those simulations conducted in implicit solvent.

Pf1 filamentous bacteriophage: refinement of a molecular model by simulated annealing using 3.3 A resolution X-ray fibre diffraction data.

The filamentous bacteriophage Pf1 is structurally similar to the well known Ff (fd, fl, M13) strains, but it gives much better X-ray diffraction patterns, enabling a more detailed analysis of the molecular structure. The 46-residue protein subunit can be closely approximated by a single gently curved stretch of alpha-helix. The axes of the subunits are at a small angle to the virion axis, and several thousand subunits form an overlapping inter-digitated helical array surrounding a DNA core. We have derived a detailed model of the virion based on X-ray data and stereochemical constraints. We have considered potential sources of error in the diffraction data, and used the improved data to study regions where the protein subunit of Pf1 may deviate from a continuous alpha-helix. We use simulated annealing to escape from local minima, and various kinds of electron-density maps to guide the model building. Refinement of the model shows that the first few residues at the N terminus are non-helical, and there is a slight discontinuity in the alpha-helix near the middle of the sequence. The model is consistent both with general structural principles derived from high-resolution analysis of other proteins, and with specific chemical and spectroscopic data about Pf1. We apply the same refinement techniques to an alternative model with a non-helical surface loop between residues 13 and 19. Comparative analysis of models with and without a loop shows that the loop model is not supported by 3.3 A resolution X-ray diffraction data.

Neutron fibre diffraction studies of DNA hydration

Techniques for recording and analysing fibre diffraction data from DNA using the instrument D19 at the Institut Laue-Langevin Reactor Source and the instrument SXD at the Rutherford Appleton Laboratory ISIS Spallation Source are described and compared. Data recorded from a specially prepared sample of deuterated DNA in the A conformation using SXD are compared with previously analysed data from a normally hydrogenated DNA in the same conformation recorded on D19. A preliminary Fourier difference analysis of these two datasets has allowed their consistency to be established.

Normal modes as refinement parameters for the F-actin model

The slow normal modes of G-actin were used as structural parameters to refine the F-actin model against 8-A resolution x-ray fiber diffraction data. The slowest frequency normal modes of G-actin pertain to collective rearrangements of domains, motions that are characterized by correlation lengths on the order of the resolution of the fiber diffraction data. Using a small number of normal mode degrees of freedom (< or = 12) improved the fit to the data significantly. The refined model of F-actin shows that the nucleotide binding cleft has narrowed and that the DNase I binding loop has twisted to a lower radius, consistent with other refinement techniques and electron microscopy data. The methodology of a normal mode refinement is described, and the results, as applied to actin, are detailed.

Molecular models and structural comparisons of native and mutant class I filamentous bacteriophages: Ff (fd, f1, M13), If1 and IKe

The filamentous bacteriophages are flexible rods about 1 to 2 microns long and 6 nm in diameter, with a helical shell of protein subunits surrounding a DNA core. The approximately 50-residue coat protein subunit is largely alpha-helix and the axis of the alpha-helix makes a small angle with the axis of the virion. The protein shell can be considered in three sections: the outer surface, occupied by the N-terminal region of the subunit, rich in acidic residues that interact with the surrounding solvent and give the virion a low isoelectric point; the interior of the shell, including a 19-residue stretch of apolar side-chains, where protein subunits interact mainly with each other; and the inner surface, occupied by the C-terminal region of the subunit, rich in basic residues that interact with the DNA core. The fact that virtually all protein side-chain interactions are between different subunits in the coat protein array, rather than within subunits, makes this a useful model system for studies of interactions between alpha-helix subunits in a macromolecular assembly. We describe molecular models of the class I filamentous bacteriophages. This class includes strains fd, f1, M13 (these 3 very similar strains are members of the Ff group), If1 and IKe. Our model of fd has been refined to fit quantitative X-ray fibre diffraction data to 30 A resolution in the meridional direction and 7 A resolution in the equatorial direction. A simulated 3.3 A resolution diffraction pattern from this model has the same general distribution of intensity as the experimental diffraction pattern. The observed diffraction data at 7 A resolution are fitted much better by the calculated diffraction pattern of our molecular model than by that of a model in which the alpha-helix subunit is represented by a rod of uniform density. The fact that our fd model explains the fd diffraction data is only part of our structure analysis. The atomic details of the model are supported by non-diffraction data, in part previously published and in part newly reported here. These data include information about permitted or forbidden side-chain replacements, about the effect of chemical modification, and about spectroscopic experiments.(ABSTRACT TRUNCATED AT 400 WORDS)

Preliminary X-ray Diffraction Studies of Ribgrass Mosaic Virus

Fiber diffraction data were collected from oriented sols of ribgrass mosaic virus and a lead derivative of the virus. Two lead binding sites were found. Two intersubunit carboxyl-carboxylate pairs, different from those in other tobamoviruses, are predicted to control vital assembly and disassembly. One of the carboxyl-carboxylate pairs forms part of a lead binding site.

Refinement of the F-Actin Model against X-ray Fiber Diffraction Data by the Use of a Directed Mutation Algorithm

The F-actin model has been refined by a Directed Mutation Algorithm, a reiterative procedure which combines a Monte-Carlo method of selecting subdomains to be refined at each cycle with a non-linear least-squares routine to get the best fit for the particular selected domains. The G-actin crystal structure was used as a starting model. The experimental data were obtained by X-ray fiber diffraction patterns from oriented F-actin gels. After 250 cycles we were able to obtain an almost perfect fit of the calculated diffraction pattern to the experimental diffraction pattern as well as a reasonable stereochemistry including intermolecular interactions of the actin monomers with an r.m.s. shift in the Cα-positions of 3·2 Å from the crystal coordinates. The stereochemistry of the intersubunit packing was calculated by molecular dynamics using the program X-PLOR. In addition, the binding site of phalloidin, a cyclic heptapeptide from the mushroom Amanita phalloides, could be determined. Furthermore, we were able to determine differences in the structures of F-actin with and without phalloidin. The method proved itself robust and showed a high degree of convergence.

Two-dimensional NMR assignments and conformation of (Pro-Hyp-Gly)10 and a designed collagen triple-helical peptide.

Homonuclear and heteronuclear 2D NMR methods are used to study two triple-helical peptides. One peptide, (POG)10, is considered to be the most stable prototype of a triple helix. The second peptide, (POG)3ITGARGLAGPOG(POG)3 (denoted T3-785), was designed to model an imino acid poor region of collagen and contains 12 residues from near the unique collagenase cleavage site in type III collagen. Both peptides associated as trimers, with melting temperatures of 60 degrees C for (POG)10 and 25 degrees C for the T3-785 peptide. Sequence-specific assignments were made for a tripeptide unit POG in (POG)10, and 80% of the POG triplets are found to be in an equivalent environment. In T3-785, with nonrepeating X-Y-Gly units incorporated in the sequence, the three chains of the homotrimer can be distinguished from one another by NMR. The solution conformation of (POG)10 is very similar to the model derived from X-ray fiber diffraction data, although the peptide contains less ordered regions at the peptide ends. In the trimer from of T3-785, the central residues of the three chains are closely packed, and the data are consistent with a triple-helical model with a one-residue stagger of three parallel chains. For T3-785, in contrast to (POG)10, there are also resonances from a less ordered form, which are probably due to the presence of a small amount of monomer. The similarity of the backbone conformations of T3-785 and (POG)10 suggests that an alternative conformation is not present in the imino acid poor region.

Molecular dynamics in refinement against fiber diffraction data

The molecular dynamics (MD) method has been adapted for refinement of the structures of helical macromolecular aggregates aginst X-ray fiber diffraction data. To test the effectiveness of the method, refinements of the tobacco mosaic virus structure were carried out against a set of simulated fiber diffraction intensities using the MD method as well as the conventional restrained least-squares (RLS) method. The MD refinement converged to a very low R factor and produced a structure with generally satisfactory sterochemistry, while the RLS refinemnt was trapped at a local energy minimum with a larger R factor. Results suggest that the effective experimental radius of convergence of the MD method is significantly greater than that of the RLS method. Even when the initial structure is too far from the true structure to allow direct refinement, the MD method is able to find local minima that resemble the true structure sufficiently to allow improved phasing and thus lead to interpretable difference maps for model rebuilding.

Deconvolution of disoriented fiber diffraction data using iterative convolution and local regression.

Computationally efficient procedures are described for the deconvolution of disoriented fiber diffraction data to the resolution limit of measurable intensity in the patterns. The methods can be applied to diffraction data from imperfectly parallel arrays of one-dimensionally periodic rods or two-dimensionally periodic sheets, randomly rotated about their unique axes, to derive a representation of the intensity distribution corresponding to perfectly parallel orientation. With use of angular convolution and local angular regression, a set of uniform cylindrically averaged squared structure factors are iteratively adjusted, subject to a minimum-wavelength constraint, until they produce a disoriented pattern that fits the observed diffraction data. The results from this deconvolution provide a measure of the properly scaled cylindrically averaged squared structure factors, which can be used with other structural information to construct a physically plausible trial model suitable for further refinement. Sample deconvolutions of simulated X-ray patterns from partially oriented gap junction membranes are presented and the results from point-model deconvolutions are compared to those from constrained deconvolutions that began with the transform of a physically plausible trial model.

Computer processing and analysis of X-ray fibre diffraction data

The processing and analysis of fibre diffraction data is often cumbersome. Presented here is a compact interactive computer program, PROFIDA (processing of fibre diffraction data), for fast processing of fibre diffraction data from biological molecules without many of the difficulties. Interactive rotation and centring of the diffraction pattern allow four-quadrant averaging and background subtraction. Furthermore, film coordinates may be transformed into reciprocal-space coordinates for any orientation of the specimen. Specimen tilt towards the incident-beam direction may be determined and the appropriate correction made. In addition, a routine that scales and merges data sets from a film pack is available. Data analysis is supported by an integration routine that allows layer-line profiles to be examined and integrated and layer-line plots to be produced under control of the cursor. The program is written in C to run on the Silicon Graphics Iris workstation with a button-and-dial box using the Iris graphics library.

Helical structure of P pili from Escherichia coli: Evidence from X-ray fiber diffraction and scanning transmission electron microscopy

The structure of the P pili from Escherichia coli has been studied using X-ray fiber diffraction and scanning transmission electron microscopy (STEM). Analysis of the fiber diffraction data indicates that the pili are constituted largely of structural subunits arranged helically with approximately 33 subunits in 10 turns in an axial repeat of 244·5 ± 1·8 Å. Radial electron density distributions calculated from equatorial diffraction data and STEM data indicate that the pili are about 65 Å in diameter with a small central cavity roughly 15 Å across. The principal protein component of the pili is PapA, which has a molecular weight of 16·5 kDa. Assuming that each subunit consists of a single PapA molecule, the mass-per-unit-length of the pili predicted from the X-ray data is 2·23 kDa/Å. Measurements of mass-per-unit-length were also made through the analysis of STEM images. These measurements indicate a value of 2·13 ± 0·14 kDa/Å. STEM images demonstrated the presence of thin, thread-like structures emerging from the ends of pili and spanning breaks in the pili structure. These structures, which have been observed under other conditions, have been termed fibrillae. In the STEM images the fibrillae appear about 20 Å in diameter. The mass-per-unit-length of the fibrillae was estimated using the STEM data to be 0.4 kDa/Å. These data are consistent with the fibrillae representing an unwound or unraveled form of the pili proteins overstretched to about five times the length they would have in the intact pili.

Characterization of a putative calcium-binding site in tobacco mosaic virus.

Lead has been used as a substitute for calcium binding to tobacco mosaic virus (TMV). The high atomic number of lead has allowed us to use difference maps from X-ray fiber diffraction data to characterize a calcium-binding site in the virus. The metal ligands are slightly different from those previously believed to bind calcium to TMV, although the binding site is very close to one previously described. Two acetate groups are also bound to the lead atom. There is no significant backbone conformational change in the protein as a result of metal binding; the binding is accomplished by means of relatively small movements in amino acid side chains.

22] Modeling DNA structures: Molecular mechanics and molecular dynamics

Model building studies may be used to supplement structurally low resolution experimental data with detailed three-dimensional hypothetical atomic models. Because of the strong relation between structure and function in biological molecules such models may give a consistent, integral view of a wealth of experimental data. In most cases such models will predict the outcome of certain experiments. The outcome of these experiments will often either confirm the model may be used for further refinement or even demand a major revision of the model. Coordinates obtained from X-ray fiber diffraction data or in special cases single-crystal data may provide the elements for DNA or RNA model building. Local and nonlocal optimization may be used to refine these structures and to evaluate their statistical significance as estimated by a chosen force field. Appreciable progress using nonlocal optimization procedures can only be expected if the dimensionality of the problem can be reduced sufficiently to the relevant degrees of freedom. Taking advantage of structural symmetries may critically improve the convergence while refining the target molecule or its building blocks. Monte Carlo and molecular dynamics methods allow one to calculate averaged quantities. In addition, molecular dynamics provides time evolutions of certain averages. During the simulation of certain physical properties of molecules a huge amount of data will be generated. They will provide many answers, but these answers may not always apply to the original question. So what type of questions will be reliably answered by a force field? Relatively safe answers concern the local geometry of the molecules. If a conformation leads to strong distortions of bond distances or angles or to close van der Waals contacts, this conformation can safely be rejected. Optimizing such unfavorable structures energetically may lead to structures showing how to avoid such distortions. More difficult are energetical questions: which of two conformers is more stable, or what is the free energy of the substrate in the active site? One cannot always be sure that the force field provides the correct answer. Therefore, one should pose only those questions which can be checked experimentally. Because of the many possible answers, the experiment may benefit by starting with a choice proposed by the simulation. The application of this procedure to curved DNA and the DNA four-way junction was successful.

Effects of disorder on fibre diffraction patterns

Diffraction patterns from oriented polycrystalline specimens of fibrous polymers sometimes contain features indicating that the constituent microcrystallites are disordered. Such disorder affects the relationship between the crystal structure and the diffracted intensities. Therefore, the effects of disorder must be considered when determining structures from fibre diffraction data. Theory is developed here that describes the intensities diffracted by disordered polycrystalline fibres, and is used to calculate diffraction patterns from simulated specimens with various kinds of disorder. The results show how Bragg and continuous diffraction are distributed on such diffraction patterns, and implications for structure determination are discussed.

Fiber diffraction analysis of cucumber green mottle mosaic virus using limited numbers of heavy-atom derivatives

The structure of cucumber green mottle mosaic virus has been determined from fiber diffraction data to a resolution of 5 A, using only two derivatives and without making any specific assumptions about the molecular structure of the virus. Because of the cylindrical averaging of fiber diffraction data, large numbers of heavy-atom derivatives have been required in previous structure determinations, but it is shown that simplifying assumptions about the overlapping intensities in fiber diffraction are sufficient for structure determination at this resolution using only two derivatives.

A comparison of the linked-atom least squares and PS79 systems for determining polymer structures from X-ray fibre diffraction data

When determining polymer structures using X-ray fibre diffraction, the diffraction data must be supplemented by stereochemical information. The structure obtained depends on the flexibility of structural models, the types of restraints and constraints used, the relative weighting of different terms, and the balance between the stereochemical and X-ray components. This affects not only final models, but also the important step of rejecting classes of incorrect models. We report a comparison of determinations of a structure using the two most widely used refinement systems (the linked-atom least squares and PS79 systems). Discrimination between classes of model and stereochemistry of the final structures are compared in detail. Implications for polymer structure determination using X-ray fibre diffraction are discussed.

Triple-strand formation in the homopurine:homopyrimidine DNA oligonucleotides d(G-A)4 and d(T-C)4.

Interest in triple and quadruple DNA helices built from homopurine and homopyrimidine strands has recently intensified principally because such structures may occur in vivo but also because of the potential use of triplexes both in forming highly sequence-specific complexes for use in chromosome mapping and in repressing transcription. From fibre diffraction data, models for triplex structures with poly(U).poly(A).poly(U) and poly(dT).poly(dA).poly(dT) have been proposed, in which the purine and one pyrimidine strand are Watson-Crick paired in an A' helix, and the other pyrimidine strand is Hoogsteen base-paired parallel to the purine strand along the major groove. A similar base-pairing scheme involving G and C would require protonation of C for Hoogsteen base-pair formation, and models for such triplexes have been proposed by analogy to the single-sequence fibre diffraction data. To date, however, there have been no single crystal or NMR structural data on DNA triplexes, and no direct observation of the protonated C in such a context. We present here the first NMR evidence for triplex formation in DNA from the homopurine d(G-A) and homopyrimidine d(T-C) oligonucleotides, and report direct observation of imino protons from protonated cytosines in the triplex.

A re-examination of the crystal structure of A-DNA using fiber diffraction data.

Classical A-DNA helices with h = 0.25 nm may represent the greatest mass per unit length attainable by polynucleotide duplexes. The X-ray diffraction pattern from polycrystalline and well-oriented fibers of calf thymus DNA in its A-form has been carefully re-examined. Indexing on the basis of a C-face-centered monoclinic unit cell of dimensions a = 2.170 nm, b = 3.990 nm, c = 2.803 nm and beta = 96.82 degrees is superior to alternatives that have been proposed. Two right-handed. Watson-Crick base-paired, helical DNA chains with 2 X 11 nucleotides per 2.803 nm pitch, each carrying C3'-endo furanose rings, pass through the unit cell. The crystallography requires the two chains in the duplex to be antiparallel and conformationally identical but the 11 nucleotides in each pitch may be distinct. However, a secondary structure with a mononucleotide asymmetric unit provides as good an X-ray agreement as one with 11 distinct nucleotides. This relative lack of variability is quite different from what is observed in fibrous B-DNAs.

The probability distributions of X-ray intensities in fiber diffraction: largest likely values for fiber diffraction R factors.

R factors in fiber diffraction are generally lower than in conventional crystallography, because of the cylindrical averaging of fiber diffraction data. The probability distributions for fiber diffraction intensities, analogous to Wilson's distributions for crystal diffraction intensities, are derived, and from these the largest likely values of R are estimated. These values depend on the size and symmetry of the diffracting particle and on the resolution of the analysis, and range from 0.586 for systems for very high symmetry (as in crystal diffraction) to much lower values for systems of low symmetry.