Crystal and molecular structure of human plasminogen kringle 4 refined at 1.9-A resolution.

The crystal structure of human plasminogen kringle 4 (PGK4) has been solved by molecular replacement using the bovine prothrombin kringle 1 (PTK1) structure as a model and refined by restrained least-squares methods to an R factor of 14.2% at 1.9-A resolution. The K4 structure is similar to that of PTK1, and an insertion of one residue at position 59 of the latter has minimal effect on the protein folding. The PGK4 structure is highly stabilized by an internal hydrophobic core and an extensive hydrogen-bonding network. Features new to this kringle include a cis peptide bond at Pro30 and the presence of two alternate, perpendicular, and equally occupied orientations for the Cys75 side chain. The K4 lysine-binding site consists of a hydrophobic trough formed by the Trp62 and Trp72 indole rings, with anionic (Asp55/Asp57) and cationic (Lys35/Arg71) charge pairs at either end. With the adjacent Asp5 and Arg32 residues, these result in triply charged anionic and cationic clusters (pH of crystals at 6.0), which, in addition to the unusually high accessibility of the Trp72 side chain, serve as an obvious marker of the binding site on the K4 surface. A complex intermolecular interaction occurs between the binding sites of symmetry-related molecules involving a highly ordered sulfate anion of solvation in which the Arg32 side chain of a neighboring kringle occupies the binding site.     

Structural and chemical complementarity between antibodies and the crystal surfaces they recognize

The sequences of the variable regions of three monoclonal antibodies with different specificities to cholesterol monohydrate and 1,4-dinitrobenzene crystals were determined. The structures of their binding sites were then modeled, based on homology to other antibodies of known structure. Two of these antibodies were previously shown to specifically recognize each one well-defined face of one of the crystals, out of a number of crystal faces of closely related structure. The binding site of the antibody which recognizes the stepped (301) face of the cholesterol crystal is predicted to assume the shape of a step with one hydrophobic and one hydrophilic side, complementary to the corresponding crystal surface. Within the step, the hydroxyl groups of five tyrosines are located such that they can interact with the hydroxyl and water molecules on the cholesterol crystal face, while hydrophobic contacts are made between the cholesterol backbone and hydrophobic amino acid sidechains. In contrast, the modeled binding site of the antibody which recognizes the flat (101) face of 1,4-dinitrobenzene crystals is remarkably flat. It is lined by aromatic and polar residues, that can make favorable contacts with the aromatic ring and nitro groups of the dinitrobenzene molecules, respectively.     

Electrostatic contributions to site specific DNA cleavage by EcoRV endonuclease

Mutational analysis of amino acids at the periphery of the EcoRV endonuclease active site suggests that moderate-range electrostatic effects play a significant role in modulating the efficiency of phosphoryl transfer. Asp36 and Lys38 located on minor-groove binding surface loops approach within 7-9 A of the scissile phosphates of the DNA. While the rates of single-site mutations removing the carboxylate or amine moieties at these positions are decreased 10(3)-10(5)-fold compared to that of wild-type EcoRV, we find that double mutants which rebalance the charge improve catalysis by up to 500-fold. Mutational analysis also suggests that catalytic efficiency is influenced by Lys173, which is buried at the base of a deep depression penetrating from a distal surface of the enzyme. The Lys173 amine group lies just 6 A from the amine group of the conserved essential Lys92 side chain in the active site. Kinetic and crystallographic analyses of the EcoRV E45A mutant enzyme further show that the Glu45 carboxylate group facilitates an extensive set of conformational transitions which occur upon DNA binding. The crystal structure of E45A bound to DNA and Mn2+ ions reveals significant conformational alterations in a small alpha-helical portion of the dimer interface located adjacent to the DNA minor groove. This leads to a tertiary reorientation of the two monomers as well as shifting of the key major-groove binding recognition loops. Because the Glu45 side chain does not appear to play a direct structural role in maintaining the active site, these rearrangements may instead originate in an altered electrostatic potential caused by removal of the negative charge. A Mn2+ binding site on the scissile phosphate is also disrupted in the E45A structure such that inner-sphere metal interactions made by the scissile DNA phosphate and conserved Asp90 carboxylate are each replaced with water molecules in the mutant. These findings argue against a proposed role for Asp36 as the general base in EcoRV catalysis, and reveal that the induced-fit conformational changes necessary for active site assembly and metal binding are significantly modulated by the electrostatic potential in this region.     

A model for the non-specific binding of catabolite gene activator protein to DNA

The binding of E. coli catabolite gene activator protein (CAP) to non-specific sequences of DNA has been modelled as an electrostatic interaction between four basic side chains of the CAP dimer and the charged phosphates of DNA. Calculation of the electrostatic contribution to the binding free energy at various separations of the two molecules shows that complex formation is favored when CAP and DNA are separated by as much as 12 A. Thus, the long range electrostatic interactions may provide the initial energy for complex formation and also the correct relative orientation of CAP and DNA. The non-specific complex does not involve the penetration of amino acid side chains into the major grooves of DNA and permits 'sliding' of the protein along DNA, which would enhance the rate of association of CAP with the specific site as has been proposed previously for lac repressor. We propose that, as it 'slides', CAP is moving in and out of the major grooves in order to sample the DNA sequence. Recognition of the specific DNA site is achieved by a complementarity in structure and hydrogen bonding between amino acids and the edges of base pairs exposed in the major grooves of DNA.     

High-resolution Structures of the Ligand Binding Domain of the Wild-type Bacterial Aspartate Receptor

The high-resolution structures of the wild-type periplasmic domain of the bacterial aspartate receptor have been determined in the absence and presence of bound aspartate to 1.85 and 2.2 Å resolution, respectively. As we reported earlier, in the refined structure of the complexed form of the crosslinked cysteine mutant receptor, the binding of the aspartate at the first site was mediated through four bridging water molecules while the second site showed an occupant electron density that best fit a sulfate group, which was present in the crystallization solution at high concentration. In the wild-type periplasmic domain structure two aspartate residues are bound per dimer, but with different occupancies. There exists a “strong” aspartate-binding site whose binding is again mediated by four water molecules while the second site contains aspartate whoseB-factor is about 10% higher, signifying weaker binding. The interaction between the second, “weaker” aspartate with the three ligand-binding arginine side-chains is slightly different from the first site. The major difference is that there are three water molecules mediating the binding of aspartate at the second site, whereas in the first site there are four bridging water molecules. The fact that aspartate-complexed crystals of the wild-type were grown with a large excess aspartate while the cross-linked crystals were grown with equal molar aspartate may explain the difference in the stoichiometry observed. The conservation of the four bridging water molecules in the strong aspartate site of both the cross-linked and wild-type periplasmic domain may reflect an important binding motif.The periplasmic domain in the apo form is a symmetrical dimer, in which each of the subunits is equivalent, and the two aspartate binding sites are identical. Upon the binding of aspartate, the subunits are no longer symmetrical. The main difference between the aspartate-bound and unbound forms is in a small, rigid-body rotation between the subunits within a dimer. The rotation is similar in both direction and magnitude in the crosslinked and wild-type periplasmic domains. The presence of the second aspartate in the wild-type structure does not make any additional rotation compared to the single-site binding. The conservation of the small angular changein vitrosuggests that the inter-subunit rotation may have relevance to the understanding of the mechanism of transmembrane signal transductionin vivo.     

Formation of AraC-DNA sandwiches.

With the use of special DNA binding sites, but not the natural aral binding site, the dimeric AraC protein can be forced to make sandwich structures in which two DNA molecules are joined by two AraC protein dimers. Apparently one subunit from each dimer contacts each DNA molecule in an extended structure. These sandwich structures form only in the absence of arabinose. This behavior is consistent with the protein's ability to form DNA loops by binding to separated half sites in the absence of arabinose and its preference for binding to adjacent half-sites in the presence of arabinose.     

Formation of a cruciform structure at the simian virus 40 replication origin abolishes T-antigen binding to the origin in vitro.

Heteroduplex DNA molecules were formed by annealing an intact simian virus replication origin-containing fragment to a mutant derivative lacking the indigenous wild-type 27-base-pair (bp) inverted repeat within this structure and containing a nonhomologous 26-bp inverted repeat sequence in its place. Results of restriction enzyme and S1 endonuclease cleavage analyses strongly suggested that a 13-bp stem-loop structure formed at the site of nonhomology between these two DNAs. This structure lies within the boundary of simian virus 40 T-antigen-binding site 2, and its presence inhibited T-antigen binding to that sequence but not to an adjacent higher-affinity binding site (site 1). Therefore, the conformation of sequences within an otherwise intact T-antigen-binding site can have major effects upon T-antigen binding there.     

Targeting DNA with novel diphenylcarbazoles

Double-stranded DNA is a therapeutic target for a variety of anticancer and antimicrobial drugs. Noncovalent interactions of small molecules with DNA usually occur via intercalation of planar compounds between adjacent base pairs or minor-groove recognition by extended crescent-shaped ligands. However, the dynamic and flexibility of the DNA platform provide a variety of conformations that can be targeted by structurally diverse compounds. Here, we propose a novel DNA-binding template for construction of new therapeutic candidates. Four bisphenylcarbazole derivatives, derived from the combined molecular architectures of known antitumor bisphenylbenzimidazoles and anti-infectious dicationic carbazoles, have been designed, and their interaction with DNA has been studied by a combination of biochemical and biophysical methods. The substitutions of the bisphenylcarbazole core with two terminal dimethylaminoalkoxy side chains strongly promote the interaction with DNA, to prevent the heat denaturation of the double helix. The deletion or the replacement of the dimethylamino-terminal groups with hydroxyl groups strongly decreased DNA interaction, and the addition of a third cationic side chain on the carbazole nitrogen reinforced the affinity of the compound for DNA. Although the bi- and tridentate molecules both derive from well-characterized DNA minor-groove binders, the analysis of their binding mode by means of circular and linear dichroism methods suggests that these compounds form intercalation complexes with DNA. Negative-reduced dichroism signals were recorded in the presence of natural DNA and synthetic AT and GC polynucleotides. The intercalation hypothesis was validated by unwinding experiments using topoisomerase I. Prominent gel shifts were observed with the di- and trisubstituted bisphenylcarbazoles but not with the uncharged analogues. These observations, together with the documented stacking properties of such molecules (components for liquid crystals), prompted us to investigate their binding to the human telomeric DNA sequence by means of biosensor surface plasmon resonance. Under conditions favorable to G4 formation, the title compounds showed only a modest interaction with the telomeric quadruplex sequence, comparable to that measured with a double-stranded oligonucleotide. Their sequence preference was explored by DNase I footprinting experiments from which we identified a composite set of binding sequences comprising short AT stretches and a few other mixed AT/GC blocks with no special AT character. The variety of the binding sequences possibly reflects the coexistence of distinct positioning of the chromophore in the intercalation sites. The bisphenylcarbazole unit represents an original pharmacophore for DNA recognition. Its branched structure, with two or three arms suitable to introduce a structural diversity, provides an interesting scaffold to built molecules susceptible to discriminate between the different conformations of nucleic acids.     

Structure of tropomyosin at 9 angstroms resolution.

We have used molecular replacement followed by a highly parameterized refinement to determine the structure of tropomyosin crystals to a resolution to 9 A. The shape, coiled-coil structure and interactions of the molecules in the crystals have been determined. These crystals have C2 symmetry with a = 259.7 A, b = 55.3 A, c = 135.6 A and beta = 97.2 degrees. Because of the unusual distribution of intensity in X-ray diffraction patterns from these crystals, it was possible to solve the rotation problem by inspection of qualitative aspects of the diffraction data and to define unequivocally the general alignment of the molecules along the (332) and (3-32) directions of the unit cell. The translation function was then solved by a direct search procedure, while electron microscopy of a related crystal form indicated the probable location of molecular ends in the asymmetric unit, as well as the anti-parallel arrangement. The structural model we have obtained is much clearer than that obtained previously with crystals of extraordinarily high solvent content and shows the two alpha-helices of the coiled coil over most of the length of the molecules and establishes the coiled-coil pitch at 140(+/- 10) A. Moreover, the precise value of the coiled-coil pitch varies along the molecule, probably in response to local variations in the amino acid sequence, which we have determined by sequencing the appropriate cDNA. The crystals are constructed from layers of tropomyosin filaments. There are two molecules in the crystallographic asymmetric unit and the molecules within a layer are bent into an approximately sinusoidal profile. Molecules in consecutive layers in the crystal lie at an angle relative to one another as found in crystalline arrays of actin and myosin rod. There are three classes of interactions between tropomyosin molecules in the spermine-induced crystals and these give some insights into the molecular interactions between coiled-coil molecules that may have implications for assemblies such as muscle thick filaments and intermediate filaments. In interactions within a layer, the geometry of coiled-coil contacts is retained, whereas in contacts between molecules in adjacent layers the coiled-coil geometry varies and these interactions instead appear to be dominated by the repeating pattern of charged zones along the molecule.     

An unexpected major groove binding of netropsin and distamycin A to tRNA(phe)

Crystalline complexes of yeast tRNA(phe) and the oligopeptide antibiotics netropsin and distamycin A were prepared by diffusing drugs into crystals of tRNA. X-ray structure analyses of these complexes reveal a single common binding site for both drugs which is located in the major or deep groove of the tRNA T-stem. The netropsin-tRNA complex is stabilized by specific hydrogen bonds between the amide groups of the drug and the tRNA bases G51 O(6), U52 O(4) and G53 N(7) on one strand, and is further stabilized by electrostatic interactions between the positively charges guanidino side chain of the drug and the tRNA phosphate P53 on the same strand and the positively charged amidino propyl side chain and the phosphates P61, P62 and P63 on the opposite strand of the double helix. These results are in contrast to the implicated minor groove binding of these drugs to non-guanine sequences in DNA. The binding to the GUG sequence in tRNA implies that major groove binding to certain DNA sequences is possible.     

Intermolecular interactions and water structure in a condensed phase B-DNA crystal

By controlled dehydration, the unit cells of dodecamer DNA–drug crystals have been shrunk from 68 000 (normal state) to 60 000 (partially dehydrated intermediate state) to 51 000 ų (fully dehydrated state), beyond which no further solvent loss occurs. The total solvent content in the normal crystals is ~40% by volume, reducing to ~20% in the fully dehydrated phase. The 25% reduction in cell volume induced a dramatic enhancement in the resolution of the X-ray diffraction data (from 2.6 to beyond 1.5 Å). We have determined the structures of the normal, partially dehydrated and fully dehydrated crystals. Details of the ligand binding have been presented in the preceding article. The present paper describes the unique features of the structure of the fully dehydrated phase. This structure was refined with 9015 unique observed reflections to R = 14.9%, making it one of the most reliable models of B-form DNA available. The crystals exist as infinite polymeric networks, in which neighbouring dodecamer duplexes are crosslinked through phosphate oxygens via direct bonding to magnesium cations. The DNA is packed so tightly that there is essentially only a single layer of solvent between adjacent molecules. The details of the crystal packing, magnesium bridging, DNA hydration and DNA conformation are described and compared with other experimental evidence related to DNA condensation.     

Ultrasonic Footprinting

Ligand binding influences the dynamics of the DNA helix in both the binding site and adjacent regions. This, in particular, is reflected in the changing pattern of cleavage of complexes under the action of ultrasound. The specificity of ultrasound-induced cleavage of the DNA sugar-phosphate backbone was studied in actinomycin D (AMD) complexes with double-stranded DNA restriction fragments. After antibiotic binding, the cleavage intensity of phosphodiester bonds between bases was shown to decrease at the chromophore intercalation site and to increase in adjacent positions. The character of cleavage depended on the sequences flanking the binding site and the presence of other AMD molecules bound in the close vicinity. A comparison of ultrasonic and DNase I cleavage patterns of AMD–DNA complexes provided more detail on the local conformation and dynamics of the DNA double helix in both binding site and adjacent regions. The results pave the way for developing a novel approach to studies of the nucleotide sequence dependence of DNA conformational dynamics and new techniques to identify functional genome regions.     

Quantitative measurement of water diffusion lifetimes at a protein/DNA interface by NMR

Hydration site lifetimes of slowly diffusing water molecules at the protein/DNA interface of the vnd/NK-2 homeodomain DNA complex were determined using novel three-dimensional NMR techniques. The lifetimes were calculated using the ratios of ROE and NOE cross-relaxation rates between the water and the protein backbone and side chain amides. This calculation of the lifetimes is based on a model of the spectral density function of the water-protein interaction consisting of three timescales of motion: fast vibrational/rotational motion, diffusion into/out of the hydration site, and overall macromolecular tumbling. The lifetimes measured ranged from approximately 400 ps to more than 5 ns, and nearly all the slowly diffusing water molecules detected lie at the protein/DNA interface. A quantitative analysis of relayed water cross-relaxation indicated that even at very short mixing times, 5 ms for ROESY and 12 ms for NOESY, relay of magnetization can make a small but detectable contribution to the measured rates. The temperature dependences of the NOE rates were measured to help discriminate direct dipolar cross-relaxation from chemical exchange. Comparison with several X-ray structures of homeodomain/DNA complexes reveals a strong correspondence between water molecules in conserved locations and the slowly diffusing water molecules detected by NMR. A homology model based on the X-ray structures was created to visualize the conserved water molecules detected at the vnd/NK-2 homeodomain DNA interface. Two chains of water molecules are seen at the right and left sides of the major groove, adjacent to the third helix of the homeodomain. Two water-mediated hydrogen bond bridges spanning the protein/DNA interface are present in the model, one between the backbone of Phe8 and a DNA phosphate, and one between the side chain of Asn51 and a DNA phosphate. The hydrogen bond bridge between Asn51 and the DNA might be especially important since the DNA contact made by the invariant Asn51 residue, seen in all known homeodomain/DNA structures, is critical for binding affinity and specificity.     

The DNA sequence at echinomycin binding sites determines the structural changes induced by drug binding: NMR studies of echinomycin binding to [d(ACGTACGT)]2 and [d(TCGATCGA)]2

The complexes formed between the cyclic octadepsipeptide antibiotic echinomycin and the two DNA octamers [d(ACGTACGT)]2 and [d(TCGATCGA)]2 have been investigated by using one- and two-dimensional proton NMR spectroscopy techniques. The results obtained for the two complexes are compared to each other, to the crystal structures of related DNA-echinomycin complexes, and to enzymatic and chemical footprinting results. In the saturated complexes, two echinomycin molecules bind to each octamer by bisintercalation of the quinoxaline moieties on either side of each CpG step. Binding of echinomycin to the octamer [d(ACGTACGT)]2 is cooperative so that only the two-drug complex is observed at lower drug-DNA ratios, but binding to [d(TCGATCGA)]2 is not cooperative. At low temperatures, both the internal and terminal A.T base pairs adjacent to the binding site in the [d(ACGTACGT)]2-2 echinomycin complex are Hoogsteen base paired (Gilbert et al., 1989) as observed in related crystal structures. However, as the temperature is raised, the internal A.T Hoogsteen base pairs are destabilized and are observed to be exchanging between the Hoogsteen base-paired and an open (or Watson-Crick base-paired) state. In contrast, in the [d(TCGATCGA)]2-2 echinomycin complex, no A.T Hoogsteen base pairs are observed, the internal A.T base pairs appear to be stabilized by drug binding, and the structure of the complex does not change significantly from 0 to 45 degrees C. Thus, the structure and stability of the DNA in echinomycin-DNA complexes depends on the sequence at and adjacent to the binding site. While we conclude that no single structural change in the DNA can explain all of the footprinting results, unwinding of the DNA helix in the drug-DNA complexes appears to be an important factor while Hoogsteen base pair formation does not.     

The crystal structure of the ligand binding module of axonin-1/TAG-1 suggests a zipper mechanism for neural cell adhesion

We have determined the crystal structure of the ligand binding fragment of the neural cell adhesion molecule axonin-1/TAG-1 comprising the first four immunoglobulin (Ig) domains. The overall structure of axonin-1(Ig1-4) is U-shaped due to contacts between domains 1 and 4 and domains 2 and 3. In the crystals, these molecules are aligned in a string with adjacent molecules oriented in an anti-parallel fashion and their C termini perpendicular to the string. This arrangement suggests that cell adhesion by homophilic axonin-1 interaction occurs by the formation of a linear zipper-like array in which the axonin-1 molecules are alternately provided by the two apposed membranes. In accordance with this model, mutations in a loop critical for the formation of the zipper resulted in the loss of the homophilic binding capacity of axonin-1.     

An Unexpected Major Groove Binding of Netropsin and Distamycin A to tRNAphe

Abstract

Crystalline complexes of yeast tRNA^phe and the oligopeptide antibiotics netropsin and distamycin A were prepared by diffusing drugs into crystals of tRNA. X-ray structure analyses of these complexes reveal a single common binding site for both drugs which is located in the major or deep groove of the tRNA T-stem. The netropsin-tRNA complex is stabilized by specific hydrogen bonds between the amide groups of the drug and the tRNA bases G51 0(6), U52 0(4) and G53 N(7) on one strand, and is further stabilized by electrostatic interactions between the positively charges guanidino side chain of the drug and the tRNA phosphate P53 on the same strand and the positively charged amidino propyl side chain and the phosphates P61, P62 and P63 on the opposite strand of the double helix. These results are in contrast to the implicated minor groove binding of these drugs to non-guanine sequences in DNA. The binding to the GUG sequence in tRNA implies that major groove binding to certain DNA sequences is possible.     

Solution structure of the HMG protein NHP6A and its interaction with DNA reveals the structural determinants for non-sequence-specific binding

NHP6A is a chromatin-associated protein from Saccharomyces cerevisiae belonging to the HMG1/2 family of non-specific DNA binding proteins. NHP6A has only one HMG DNA binding domain and forms relatively stable complexes with DNA. We have determined the solution structure of NHP6A and constructed an NMR-based model structure of the DNA complex. The free NHP6A folds into an L-shaped three alpha-helix structure, and contains an unstructured 17 amino acid basic tail N-terminal to the HMG box. Intermolecular NOEs assigned between NHP6A and a 15 bp 13C,15N-labeled DNA duplex containing the SRY recognition sequence have positioned the NHP6A HMG domain onto the minor groove of the DNA at a site that is shifted by 1 bp and in reverse orientation from that found in the SRY-DNA complex. In the model structure of the NHP6A-DNA complex, the N-terminal basic tail is wrapped around the major groove in a manner mimicking the C-terminal tail of LEF1. The DNA in the complex is severely distorted and contains two adjacent kinks where side chains of methionine and phenylalanine that are important for bending are inserted. The NHP6A-DNA model structure provides insight into how this class of architectural DNA binding proteins may select preferential binding sites.