Structural and energetic consequences of mutations in a solvated hydrophobic cavity.

The structural and energetic consequences of modifications to the hydrophobic cavity of interleukin 1-beta (IL-1beta) are described. Previous reports demonstrated that the entirely hydrophobic cavity of IL-1beta contains positionally disordered water. To gain a better understanding of the nature of this cavity and the water therein, a number of mutant proteins were constructed by site-directed mutagenesis, designed to result in altered hydrophobicity of the cavity. These mutations involve the replacement of specific phenylalanine residues, which circumscribe the cavity, with tyrosine, tryptophan, leucine and isoleucine. Using differential scanning calorimetry to determine the relative stabilities of the wild-type and mutant proteins, we found all of the mutants to be destabilizing. X-ray crystallography was used to identify the structural consequences of the mutations. No clear correlation between the hydrophobicities of the specific side-chains introduced and the resulting stabilities was found.

Disordered to ordered folding in the regulation of diphtheria toxin repressor activity.

Understanding how metal binding regulates the activity of the diphtheria toxin repressor protein (DtxR) requires information about the structure in solution. We have prepared a DtxR mutant construct with three additional N-terminal residues, Gly-Ser-His-DtxR(Cys-102 --> Asp), that retains metal-binding capabilities, but remains monomeric in solution and does not bind DNA under conditions that effect dimerization and DNA binding in the functional DtxR(Cys-102 --> Asp) construct. Although the interaction properties of this inactive mutant in solution are very different from that of active repressors, crystallization imposes the same dimeric structure as observed in all crystal forms of the active repressor with and without bound metal. Our solution NMR analyses of active and inactive metal-free diphtheria toxin repressors demonstrate that whereas the C-terminal one-third of the protein is well ordered, the N-terminal two-thirds exhibits conformational flexibility and exists as an ensemble of structural substates with undefined tertiary structure. Fluorescence binding assays with 1-anilino naphthalene-8-sulfonic acid (ANS) confirm that the highly alpha-helical N-terminal two-thirds of the apoprotein is molten globule-like in solution. Binding of divalent metal cations induces a substantial conformational reorganization to a more ordered state, as evidenced by changes in the NMR spectra and ANS binding. The evident disorder to order transition upon binding of metal in solution is in contrast to the minor conformational changes seen comparing apo- and holo-DtxR crystal structures. Disordered to ordered folding appears to be a general mechanism for regulating specific recognition in protein action and this mechanism provides a plausible explanation for how metal binding controls the DtxR repressor activity.

Hyperstable stacked-disk structure of tobacco mosaic virus protein: electron cryomicroscopy image reconstruction related to atomic models.

The stacked disk aggregate of tobacco mosaic virus protein is an intriguing object due to its high degree of stability, in spite of indications that the aggregate is held together to a great extent by water-mediated interactions between adjacent protein rings. Here, we present a set of models that were constructed using the atomic coordinates of the four-layer aggregate, and compare these with a three-dimensional reconstruction of the stacked disk obtained by means of cryoelectron microscopy and helical image processing. The comparison of the four possible models of the stacked disk with the data shows that there is a better correlation of the data with the left-handed model built from the A-A ring pair coordinates than with the two models involving the A-B ring pair, or with the right-handed model of the A-A ring pair. This establishes that the packing of the protein subunits in the stacked disk is different from that previously believed. We also note some differences between the observed structure and A-A ring pair model in the region of the flexible loop at small radius that might be an indication of conformational differences that give rise to the stability of the aggregate.

Solution structure and peptide binding studies of the C-terminal src homology 3-like domain of the diphtheria toxin repressor protein.

The diphtheria toxin repressor (DtxR) is the best-characterized member of a family of homologous proteins that regulate iron uptake and virulence gene expression in the Gram-positive bacteria. DtxR contains two domains that are separated by a short, unstructured linker. The N-terminal domain is structurally well-defined and is responsible for Fe2+ binding, dimerization, and DNA binding. The C-terminal domain adopts a fold similar to eukaryotic Src homology 3 domains, but the functional role of the C-terminal domain in repressor activity is unknown. The solution structure of the C-terminal domain, consisting of residues N130-L226 plus a 13-residue N-terminal extension, has been determined by using NMR spectroscopy. Residues before A147 are highly mobile and adopt a random coil conformation, but residues A147-L226 form a single structured domain consisting of five beta-strands and three helices arranged into a partially orthogonal, two-sheet beta-barrel, similar to the structure observed in the crystalline Co2+ complex of full-length DtxR. Chemical shift perturbation studies demonstrate that a proline-rich peptide corresponding to residues R125-G139 of intact DtxR binds to the C-terminal domain in a pocket formed by residues in beta-strands 2, 3, and 5, and helix 3. Binding of the proline-rich peptide by the C-terminal domain of DtxR presents an example of peptide binding by a prokaryotic Src homology 3-like protein. The results of this study, combined with previous x-ray studies of intact DtxR, provide insights into a possible biological function of the C-terminal domain in regulating repressor activity.

Disordered water within a hydrophobic protein cavity visualized by x-ray crystallography.

Water in the hydrophobic cavity of human interleukin 1beta, which was detected by NMR spectroscopy but was invisible by high resolution x-ray crystallography, has been mapped quantitatively by measurement and phasing of all of the low resolution x-ray diffraction data from a single crystal. Phases for the low resolution data were refined by iterative density modification of an initial flat solvent model outside the envelope of the atomic model. The refinement was restrained by the condition that the map of the difference between the electron density distribution in the full unit cell and that of the atomic model be flat within the envelope of the well ordered protein structure. Care was taken to avoid overfitting the diffraction data by maintaining phases for the high resolution data from the atomic model and by a resolution-dependent damping of the structure factor differences between data and model. The cavity region in the protein could accommodate up to four water molecules. The refined solvent difference map indicates that there are about two water molecules in the cavity region. This map is compatible with an atomic model of the water distribution refined by using XPLOR. About 70% of the time, there appears to be a water dimer in the central hydrophobic cavity, which is connected to the outside by two constricted channels occupied by single water molecules approximately 40% of the time on one side and approximately 10% on the other.

Structure of the stacked disk aggregate of tobacco mosaic virus protein.

The coat protein of tobacco mosaic virus is known to form three different classes of aggregate, depending on environmental conditions, namely helical, disk, and A-protein. Among the disk aggregates, there are four-layer, six-layer, and long stacks, which can be obtained by varying the ionic strength and temperature conditions during the association process. The four-layer aggregate has been crystallized, and its structure solved to atomic resolution. The stacked disk aggregate had been presumed to be built of a polar two-layer disk related to the crystalline A and B rings. A study using monoclonal antibodies specific to the bottom surface of TMV protein demonstrated that the stacked disk aggregate is bipolar, and suggested that the repeating two-layer unit might be similar to the dihedrally symmetrical A-ring pair in the disk crystal. In this paper we present a three-dimensional reconstruction of the stacked disk aggregate obtained by electron microscopy of ice-embedded samples. After modeling of the structure, we found the ring pairs to have the same quaternary structure as the A-ring pair of the four-layer aggregate. The resolution achieved in the image processing of the electron micrographs is on the order of 9 A in the meridional direction and 12 A in the equatorial. The identification of the structure of the stacked disk with the A-ring pair of the disk crystal provides an explanation of the observation that the axial periodicity of the disk pair, which is approximately 53 A when fully hydrated, can shrink to approximately 43 A in the dry state.

Refined atomic model of the four-layer aggregate of the tobacco mosaic virus coat protein at 2.4-A resolution.

Previous x-ray studies (2.8-A resolution) on crystals of tobacco mosaic virus coat protein grown from solutions containing high salt have characterized the structure of the protein aggregate as a dimer of a bilayered cylindrical disk formed by 34 chemically identical subunits. We have determined the crystal structure of the disk aggregate at 2.4-A resolution using x-ray diffraction from crystals maintained at cryogenic temperatures. Two regions of interest have been extensively refined. First, residues of the low-radius loop region, which were not modeled previously, have been traced completely in our electron density maps. Similar to the structure observed in the virus, the right radial helix in each protomer ends around residue 87, after which the protein chain forms an extended chain that extends to the left radial helix. The left radial helix appears as a long alpha-helix with high temperature factors for the main-chain atoms in the inner portion. The side-chain atoms in this region (residues 90-110) are not visible in the electron density maps and are assumed to be disordered. Second, interactions between subunits in the symmetry-related central A pair have been determined. No direct protein-protein interactions are observed in the major overlap region between these subunits; all interactions are mediated by two layers of ordered solvent molecules. The current structure emphasizes the importance of water in biological macromolecular assemblies.

Structure of cubic insulin crystals in glucose solutions.

X-ray structures of cubic insulin crystals in high concentrations of glucose at different pH levels and temperatures have been refined to high resolution. We have identified one glucose-binding site near the N-terminus of the A-chain whose occupancy is pH dependent. The effects of reduced water activity on the ordered protein and solvent structures have been examined. Our analysis showed no notable conformational changes in the ordered protein structures or ordered solvent molecules near the protein surface, but the presence of glucose does have a significant effect on the overall density distribution of the bulk solvent in the solvent-accessible volume. We compared the structure of cubic insulin at room temperature and liquid-nitrogen temperature, under identical solvent conditions, using glucose as a cryoprotectant. In this case, we found that the average temperature factor of the protein is reduced and more water molecules can be identified, but there are no significant changes in the protein conformation.

Five-fold symmetry in crystalline quasicrystal lattices.

To demonstrate that crystallographic methods can be applied to index and interpret diffraction patterns from well-ordered quasicrystals that display non-crystallographic 5-fold symmetry, we have characterized the properties of a series of periodic two-dimensional lattices built from pentagons, called Fibonacci pentilings, which resemble aperiodic Penrose tilings. The computed diffraction patterns from periodic pentilings with moderate size unit cells show decagonal symmetry and are virtually indistinguishable from that of the infinite aperiodic pentiling. We identify the vertices and centers of the pentagons forming the pentiling with the positions of transition metal atoms projected on the plane perpendicular to the decagonal axis of quasicrystals whose structure is related to crystalline eta phase alloys. The characteristic length scale of the pentiling lattices, evident from the Patterson (autocorrelation) function, is approximately tau 2 times the pentagon edge length, where tau is the golden ratio. Within this distance there are a finite number of local atomic motifs whose structure can be crystallographically refined against the experimentally measured diffraction data.

Problems in simulating macromolecular movements.

Insufficient sampling of conformational sub-states by current molecular dynamics simulations accounts for deficiencies in representations of the fluctuating interatomic separations in macromolecules.

Neutron diffraction analysis of the solvent accessible volume in cubic insulin crystals.

The average contact distance between protein and solvent surface atoms in cubic insulin crystals has been determined from two sets of 15 A resolution neutron diffraction data. A contact distance between the water hydrogen sites and the protein surface that is significantly shorter than the average protein-water oxygen contact distance implies that many water molecules are oriented with hydrogen atoms pointed towards the protein surface. The shape of the protein/solvent interface is consistent with the protein envelope obtained from atomic co-ordinates.

Stereospecific dihaloalkane binding in a pH-sensitive cavity in cubic insulin crystals.

Crystallographic analysis at 2-A resolution of the selective binding of dihalogenated methane, ethane, and ethylene compounds in the cavity on the cubic insulin dimer axis provides a model for anesthetic-protein interactions. At pH 6-11, 1,2-dichloroethane binds isomorphically in the right-handed cis-conformation, displacing four water molecules from the invariant cavity. Lowering the pH to 5.7 in 1 M Na2SO4 without dihaloalkanes induces a cooperative structural transition in which the dyad cavities between B13 glutamate pairs are constricted, and SO4(2-) ions are bound by rearranged triads of B1 NH+3 groups. In the presence of dichloroethane at pH 5-5.5, the equilibrium is shifted to a mixture of the ligand-bound and ligand-excluding cavity structures, with half-occupancy of the sulfate sites, exemplifying how a volatile anesthetic can act as an allosteric effector. Measurements at pH 9 of the occupancies of structurally similar dihaloalkanes demonstrate a high degree of binding selectivity. Induced polarization of the ligand and bound water by the charge distribution in the binding cavity apparently provides the selective electrostatic interactions that discriminate between dihaloalkanes of comparable size and polarity.

Electron diffraction of helical particles.

The development of low-dose electron cryo-microscopy has provided the means to see structural details to better than 10 A resolution in helical structures. The application of techniques of image analysis to micrographs can yield accurate phases, but not amplitudes with which to generate three-dimensional maps of the structure. Electron diffraction can provide reliable amplitudes, which can be combined with the phases from the images. In order to collect amplitude data, two problems have to be overcome: the pattern should be obtained from a large well ordered sample of particles, and the inelastic background should be properly subtracted. In this paper, we present three simple methods to produce rafts of helical particles. Using these methods we have obtained electron diffraction patterns from TMV (with data out to 0.28 nm), TMV protein stacked disks (with data out to 0.3 nm) and bacterial flagellar filaments (with data out to 0.5 nm). In addition, we describe the algorithms used to extract the amplitudes from the diffraction patterns.

Thallium counterion distribution in cubic insulin crystals determined from anomalous x-ray diffraction data.

To determine the distribution of monovalent cations around a protein we have measured anomalous scattering diffraction data from Tl-containing cubic insulin crystals at pH 8 and pH 10.5. The differences between Bijvoet reflection pairs within each set of data were used to calculate anomalous scattering difference maps. Both maps show the same six Tl+ sites, which include two well-ordered Tl+ ions previously identified from isomorphous exchange experiments. The other four sites constitute a second class of cations, which, while much more mobile than the protein atoms, are associated with particular ligating groups. Three of the six Tl+ sites are created exclusively by protein main and side chain carbonyl dipoles rather than negatively charged groups. All of the Tl+ ions are positioned so as to interact with both protein atoms and water molecules. The Tl+ occupancies appear to depend in a complex way on interactions with each other and flexibility in the protein structure. The combined occupancies of these cations are slightly less than is required to neutralize the net protein charge of approximately -2e at pH 8 but account for only about half of the approximately -5e protein charge at pH 10.5. Thus, more disordered counterions, not seen in these Bijvoet anomalous scattering difference maps, are more numerous at higher protein net charge.

Structure and selectivity of a monovalent cation binding site in cubic insulin crystals.

Cubic insulin crystals contain a binding site for monovalent cations in a cavity of the crystal dyad in which the bound cation is ligated by protein atomic dipoles and water molecules. These types of interaction are analogous to interactions that occur in small cation-selective carrier and channel molecules. X-ray diffraction data collected from cubic insulin crystals containing Li+, Na+, K+, NH4+, Rb+, and Tl+ show that (i) the differences in cation size do not cause any large alteration in the protein structure around the cation, and (ii) the bound cation is co-ordinated by one or two water molecules, depending on its ionic radii. The relative binding affinities for cations at this dyad site were obtained from an x-ray diffraction analysis of competition experiments in which crystals were dialyzed in mixtures of Tl+ with Li+, Na+, NH4+, Rb+, or Cs+. These data show that this site provides very little discrimination between Na+, K+, Rb+, and Tl+, some selectivity against the small Li+ and the tetrahedrally shaped NH4+, and stronger selectivity against the larger Cs+. The capacity of this site to bind monovalent cations of different sizes may be accounted for by the small number of protein ligating groups and a change from two ligating waters with Li+ and Na+ to one ligating water with the larger cations.

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.

Conformational changes in cubic insulin crystals in the pH range 7-11.

To determine the effect of variations in the charge distribution on the conformation of a protein molecule, we have solved the structures of bovine cubic insulin over a pH range from 7 to 11 in 0.1 M and 1 M sodium salt solutions. The x-ray data were collected beyond 2-A resolution and the R factors for the refined models ranged from 0.16 to 0.20. Whereas the positions of most protein and well-ordered solvent atoms are conserved, about 30% of residues alter their predominant conformation as the pH is changed. Conformational switching of A5 Gln and B10 His correlates with the pH dependence of monovalent cation binding to insulin in cubic crystals. Shifts in the relative positions of the A chain NH2-terminal and B chain COOH-terminal groups are probably due to titration of the A1 alpha-amino group. Two alternative positions of B25 Phe and A21 Asn observed in cubic insulin at pH 11 are similar to those found in two independent molecules of the 2Zn insulin dimer at pH 6.4. The conformational changes of the insulin amino acids appear to be only loosely coupled at distant protein sites. Shifts in the equilibrium between distinct conformational substates as the charge distribution on the protein is altered are analogous to the electrostatically triggered movements that occur in many functional protein reactions.