Crystal structure of a pentamidine-oligonucleotide complex: implications for DNA-binding properties.

The crystal structure of the complex formed between the dodecanucleotide d(CGCGAATTCGCG)2 and the drug pentamidine, which is active against the Pneumocystis carinii pathogen in AIDS patients, has been determined to a resolution of 2.1 A and an R-factor of 19.4%. Analysis of the structure has shown the drug to be bound in the 5'-AATT minor groove region of the duplex, with the amidinium groups H-bonded to adenine N3 atoms in an interstrand manner. The drug molecule adopts an extended conformation, and the immediate binding site spans four base pairs. Structural details of the drug-DNA interactions are discussed, and comparison is made with the dodecamer complex of the structurally similar berenil ligand.     

Sequence-dependent effects in drug-DNA interaction: the crystal structure of Hoechst 33258 bound to the d(CGCAAATTTGCG)2 duplex.

The bis-benzimidazole drug Hoechst 33258 has been co-crystallized with the dodecanucleotide sequence d(CGCAAATTTGCG)2. The structure has been solved by molecular replacement and refined to an R factor of 18.5% for 2125 reflections collected on a Xentronics area detector. The drug is bound in the minor groove, at the five base-pair site 5'-ATTTG and is in a unique orientation. This is displaced by one base pair in the 5' direction compared to previously-determined structures of this drug with the sequence d(CGCGAATTCGCG)2. Reasons for this difference in behaviour are discussed in terms of several sequence-dependent structural features of the DNA, with particular reference to differences in propeller twist and minor-groove width.     

Crystal structure of a berenil-dodecanucleotide complex: the role of water in sequence-specific ligand binding.

The three-dimensional structure of a complex between the dodecanucleotide d(CGCGAATTCGCG) and the anti-trypanocidal drug berenil, has been determined to a resolution of 2.5 A. The structure has been solved by molecular replacement and refined to an R factor of 0.177. A total of 49 water molecules have been located. The drug is bound at the 5'-AAT-3' region of the oligonucleotide. At one end of the drug the amidinium group is in hydrogen-bonded contact with N3 of the adenine base complementary to the thymine of the AAT. The other amidinium group does not make direct interactions with the DNA. Instead, a water molecule mediates between them. This is in hydrogen-bonded contact with an amidinium nitrogen atom, N3 of the 5' end adenine base and the ring oxygen atom of an adjacent deoxyribose. Molecular mechanics calculations have been performed on this complex, with the drug at various positions along the sequence. These show that the observed position is only 0.8 kcal/mol higher in energy than the best position. It is suggested that there is a broad energy well in the AATT region for this drug, and that water molecules as well as the neighbouring sequence, will determine precise positioning. More general aspects of minor groove binding are discussed.     

Interaction of berenil with the tyrT DNA sequence studied by footprinting and molecular modelling. Implications for the design of sequence-specific DNA recognition agents.

We have developed a technique of partially-restrained molecular mechanics enthalpy minimisation which enables the sequence-dependence of the DNA binding of a non-intercalating ligand to be studied for arbitrary sequences of considerable length (greater than = 60 base-pairs). The technique has been applied to analyse the binding of berenil to the minor groove of a 60 base-pair sequence derived from the tyrT promoter; the results are compared with those obtained by DNAse I and hydroxyl radical footprinting on the same sequence. The calculated and experimentally observed patterns of binding are in good agreement. Analysis of the modelling data highlights the importance of DNA flexibility in ligand binding. Further, the electrostatic component of the interaction tends to favour binding to AT-rich regions, whilst the van der Waals interaction energy term favours GC-rich ones. The results also suggest that an important contribution to the observed preference for binding in AT-rich regions arises from lower DNA perturbation energies and is not accompanied by reduced DNA structural perturbations in such sequences. It is therefore concluded that those modes of DNA distortion favourable to binding are probably more flexible in AT-rich regions. The structure of the modelled DNA sequence has also been analysed in terms of helical parameters. For the DNA energy-minimised in the absence of berenil, certain helical parameters show marked sequence-dependence. For example, purine-pyrimidine (R-Y) base pairs show a consistent positive buckle whereas this feature is consistently negative for Y-R pairs. Further, CG steps show lower than average values of slide while GC steps show lower than average values of rise. Similar analysis of the modelling data from the calculations including berenil highlights the importance of DNA flexibility in ligand binding. We observe that the binding of berenil induces characteristic responses in different helical parameters for the base-pairs around the binding site. For example, buckle and tilt tend to become more negative to the 5'-side of the binding site and more positive to the 3'-side, while the base steps at either side of the centre of the site show increased twist and decreased roll.     

Heterogeneous DNA binding modes of berenil.

Isothermal titration calorimetry (ITC) profiles of berenil bound to different DNAs show that, despite the strong preference of berenil for AT-rich regions in DNA, it can bind to other DNA sequences significantly. The ITC results were used to quantify the binding of berenil, and the thermodynamic profiles were obtained using natural DNAs as well as synthetic polynucleotides. ITC binding isotherms cannot be simply described when a single set of identical binding sites is considered, except for poly[d(A-T)2]. Ultraviolet melting of DNA and differential scanning calorimetry were also used to quantify several aspects of the binding of berenil to salmon testes DNA. We present evidence for secondary binding sites for berenil in DNA, corresponding to G+C rich sites. Berenil binding to poly[d(G-C)2] is also observed. Circular dichroism experiments showed that binding to GC-rich sites involves drug intercalation. Using a molecular modeling approach we demonstrate that intercalation of berenil into CpG steps is sterically feasible.     

NMR studies of the interaction of the antibiotic nogalamycin with the hexadeoxyribonucleotide duplex d(5'-GCATGC)2

1H resonance assignments in the NMR spectra of the self-complementary hexadeoxyribonucleoside pentaphosphate d(5'-GCATGC)2 and its complex with the antibiotic nogalamycin, together with interproton distance constraints obtained from two-dimensional nuclear Overhauser effect (NOE) spectra, have enabled us to characterize the three-dimensional structure of these species in solution. In the complex described, two drug molecules are bound per duplex, in each of two equivalent binding sites, with full retention of the dyad symmetry. Twenty-eight NOE distance constraints between antibiotic and nucleotide protons define the position and orientation of the bound drug molecule. Nogalamycin intercalates at the 5'-CA and 5'-TG steps with the major axis of the anthracycline chromophore aligned approximately at right angles to the major axes of the base pairs. The nogalose sugar occupies the minor groove of the helix and makes many contacts with the deoxyribose moieties of three nucleotides along one strand of the duplex in the 5'-TGC segment. The charged dimethylamino group and hydroxyl functions of the bicyclic sugar lie in the major groove juxtaposed to the guanine base, the bridging atoms of the bicyclic sugar making contacts with the methyl group of the thymine. Thus the antibiotic is not symmetrically disposed in the intercalation site but is in close contact in both grooves with atoms comprising the 5'-TGC strand. The intercalation cavity is wedge-shaped, the major axes of the base pairs forming the site being tilted with respect to one another. All base-pair hydrogen-bonding interactions are maintained in the complex, and there is no evidence for Hoogsteen pairing. The free duplex adopts a regular right-handed B-type conformation in which all glycosidic bond angles are anti and all sugar puckers lie in the C2'-endo range. In the complex the glycosidic bond angles and the sugar puckers deviate little from those observed for the duplex alone. The presence of two bound nogalamycin molecules substantially slows the "breathing" motions of the base pairs forming the intercalation cavity, and the observation of two downfield-shifted resonances in the 31P NMR spectrum of the complex suggests a pronounced local helix unwinding at the drug binding site. The footprinting data of Fox and Waring [Fox, K.R., & Waring, M.J. (1986) Biochemistry 25, 4349-4356] imply that the highest affinity binding sites of nogalamycin have the sequence 5'-GCA (or 5'-TGC).(ABSTRACT TRUNCATED AT 400 WORDS)     

H-NMR studies on d(GCTTAAGC)2 and its complex with berenil.

Two-dimensional (2D) 1H-NMR spectroscopy has been used to analyze the structure of d(GCTTAAGC)2 and its interaction with berenil in solution. Nuclear Overhauser enhancement connectivities enabled sequential assignments of nearly all proton resonances in the self-complementary octamer duplex and demonstrated that the oligonucleotide is primarily in a B-type conformation. No major conformational changes were observed by the addition of berenil, but proton resonances of the two adenosine nucleotides shifted substantially. Intermolecular nuclear Overhauser effects between berenil and the DNA duplex revealed that the drug binds via the minor groove of d(GCTTAAGC)2 in the A.T-base-pair region. At 18 degrees C the twofold symmetry of the duplex is preserved on berenil binding. However, strongly shifted proton resonances broadened significantly. A model is proposed for the berenil-d(GCTTAAGC)2 complex involving fast exchange of berenil between two equivalent symmetry-related binding sites, which span the 5'-TAA-3' region and are asymmetrically disposed with respect to the dyad axis of the duplex. These results are compared with previous studies on the berenil-d(GCAATTGC)2 complex.     

The crystal structure of the DNA-binding drug berenil: molecular modelling studies of berenil-DNA complexes.

The crystal structure of the DNA minor-groove DNA-binding drug berenil has been determined. Molecular-modelling techniques have been used to establish plausible binding modes of the structure to A-T sequences. These have shown that specific hydrogen bonds are possible between the amidine groups of the drug molecule and 02 atoms of thymine, although global energy minimisations tended to emphasise electrostatic interactions with phosphate groups rather than these hydrogen bonds with bases.     

Structure of a bis-amidinium derivative of hoechst 33258 complexed to dodecanucleotide d(CGCGAATTCGCG)2: the role of hydrogen bonding in minor groove drug-DNA recognition.

The crystal structure is reported of a complex between the dodecanucleotide sequence d(CGCGAATTCGCG)2and an analogue of the DNA binding drug Hoechst 33258, in which the piperazine ring has been replaced by an amidinium group and the phenol ring by a phenylamidinium group. The structure has been refined to an R factor of 19.5% at 2.2 A resolution. The drug is held in the minor groove by five strong hydrogen bonds, together with bridging water molecules at both ends. There are few other contacts with the floor of the groove, indicating a lack of isohelicity with the groove and suggesting (i) that the observed high DNA affinity of this drug is primarily due to the array of hydrogen bonds and (ii) that these more than compensate for its poor isohelicity.     

Crystal structure of the 2:1 complex between d(GAAGCTTC) and the anticancer drug actinomycin D.

The crystal structures of the 2:1 complex of the self-complementary DNA octamer d(GAAGCTTC) with actinomycin D has been determined at 3.0 A resolution. This is the first example of a crystal structure of a DNA-drug complex in which the drug intercalates into the middle of a relatively long DNA segment. The results finally confirmed the DNA-actinomycin intercalation model proposed by Sobell & co-workers in 1971. The DNA molecule adopts a severely distorted and slightly kinked B-DNA-like structure with an actinomycin D molecule intercalated in the middle sequence, GC. The two cyclic depsipeptides, which differ from each other in overall conformation, lie in the minor groove. The complex is further stabilized by forming base-peptide and chromophore-backbone hydrogen bonds. The DNA helix appears to be unwound by rotating one of the base-pairs at the intercalation site. This single base-pair unwinding motion generates a unique asymmetrically wound helix at the binding site of the drug, i.e. the helix is loosened at one end of the intercalation site and tightened at the other end. The large unwinding of the DNA by the drug intercalation is absorbed mostly in a few residues adjacent to the intercalation site. The asymmetrical twist of the DNA helix, the overall conformation of the two cyclic depsipeptides and their interaction mode with DNA are correlated to each other and rationally explained.     

Binding of an antitumor drug to DNA, Netropsin and C-G-C-G-A-A-T-T-BrC-G-C-G

The antitumor antibiotic netropsin has been co-crystallized with a double-helical B-DNA dodecanucleotide of sequence: C-G-C-G-A-A-T-T-BrC-G-C-G, and the structure of the complex has been solved by X-ray diffraction at a resolution of 2.2 A. The structure has been refined independently by Jack-Levitt and Hendrickson-Konnert least-squares methods, leading to a final residual error of 0.257 by the Jack-Levitt approach (0.211 for two-sigma data) or 0.248 by the Hendrickson-Konnert approach, with no significant difference between refined structures. The netropsin molecule displaces the spine of hydration and fits snugly within the minor groove in the A-A-T-T center. It widens the groove slightly and bends the helix axis back by 8 degrees, but neither unwinds nor elongates the double helix. The drug molecule is held in place by amide NH hydrogen bonds that bridge adenine N-3 and thymine O-2 atoms, exactly as with the spine of hydration. The requirement of A X T base-pairs in the binding site arises because the N-2 amino group of guanine would demand impermissibly close contacts with netropsin. It is proposed that substitution of imidazole for pyrrole in netropsin should create a family of "lexitropsins" capable of reading G X C-containing base sequences.     

Structure of a eukaryotic decoding region A-site RNA

The aminoglycoside antibiotics target a region of highly conserved nucleotides in the aminoacyl-tRNA site (A site) of 16 S RNA on the 30 S subunit. The structures of a prokaryotic decoding region A-site oligonucleotide free in solution and bound to the aminoglycosides paromomycin and gentamicin C1A have been determined. Here, the structure of a eukaryotic decoding region A-site oligonucleotide has been determined using homonuclear and heteronuclear NMR spectroscopy, and compared to the unbound prokaryotic rRNA structure. The two structures are similar, with a U1406-U1495 base-pair, a C1407-G1494 Watson-Crick base-pair, and a G1408-A1493 base-pair instead of the A1408-A1493 base-pair of the prokaryotic structure. The two structures differ in the orientation of the 1408 position with respect to A1493; G1408 is rotated toward the major groove, which is the binding pocket for aminoglycosides. The structures also differ in the stacking geometry of G1494 on A1493, which could have slight long-range conformational effects.     

Variability in DNA minor groove width recognised by ligand binding: the crystal structure of a bis-benzimidazole compound bound to the DNA duplex d(CGCGAATTCGCG)2.

An analogue of the DNA-binding compound Hoechst 33258, in which the piperazine ring has been replaced by an imidazoline group, has been cocrystallized with the dodecanucleotide sequence d(CGCGAATTCGCG)2. The structure has been solved by X-ray diffraction analysis and has been refined to an R-factor of 19.7% at a resolution of 2.0 A. The ligand is found to bind in the minor groove, at the central four AATT base pairs of the B-DNA double helix, with the involvement of a number of van der Waals contacts and hydrogen bonds. There are significant differences in minor groove width for the two compounds, along much of the AATT region. In particular this structure shows a narrower groove at the 3' end of the binding site consistent with the narrower cross-section of the imidazole group compared with the piperazine ring of Hoechst 33258 and therefore a smaller perturbation in groove width. The higher binding affinity to DNA shown by this analogue compared with Hoechst 33258 itself, has been rationalised in terms of these differences.     

Structure of cobalt carbonic anhydrase complexed with bicarbonate.

The three-dimensional structure of a complex between catalytically active cobalt(II) substituted human carbonic anhydrase II and its substrate bicarbonate was determined by X-ray crystallography (1.9 A). One water molecule and two bicarbonate oxygen atoms are found at distances between 2.3 and 2.5 A from the cobalt ion in addition to the three histidyl ligands contributed by the peptide chain. The tetrahedral geometry around the metal ion in the native enzyme with a single water molecule 2.0 A from the metal is therefore lost. The geometry is difficult to classify but might best be described as distorted octahedral. The structure is suggested to represent a water-bicarbonate exchange state relevant also for native carbonic anhydrase, where the two unprotonized oxygen atoms of the substrate are bound in a carboxylate binding site and the hydroxyl group is free to move closer to the metal thereby replacing the metal-bound water molecule. A reaction mechanism based on crystallographically determined enzyme-ligand complexes is represented.     

A new crystal form for the dodecamer C-G-C-G-A-A-T-T-C-G-C-G: symmetry effects on sequence-dependent DNA structure

The dodecanucleotide d(CGCGAATTCGCG) has been crystallised in the space group P3(2)12, representing a new crystal form for this sequence. The structure has been solved by molecular replacement and refined at 1.8 A resolution. The present structure contrasts with previous ones for this sequence since it is situated on a crystallographic 2-fold axis, and the crystal symmetry reflects the palindromic nature of this sequence. Some features accord with previous observations, notably that the minor groove is hydrated with a continuous spine of solvent. There is no evidence of alkali metal ions within this spine. The minor groove retains its narrow width, although it is now symmetric and extends over the A/T tract. Various base and base-pair morphological parameters have been examined. Their values do not show significant correlations with earlier reports, suggesting that crystal packing effects on them are more dominant than has been hitherto realised.     

Base recognition mechanism of bleomycin: solution structure of d(GGGGAGCTCCCC)2 based on 1H-NMR and restrained molecular dynamics.

The self-complementary d(GGGGAGCTCCCC) dodecanucleotide cleavage by bleomycin occurs at G(4)pA(5) site rather than G(6)pC(7) site which is known as the preferential sequence. The molecular structure of this dodecamer is observed as the distorted B-DNA by the CD spectra and proton NMR measurements. Three-dimensional structures in solution evaluated by molecular dynamics calculations with inter proton distance restraints have the distorted form of the minor groove structure at G(6)pC(7) site. These evidences offer the recognition mechanism model of the DNA base sequence by bleomycin.     

Molecular structure of the B-DNA dodecamer d(CGCAAATTTGCG)2. An examination of propeller twist and minor-groove water structure at 2.2 A resolution.

The crystal structure of the dodecanucleotide duplex d(CGCAAATTTGCG)2 has been solved to 2.2 A resolution and refined to an R-factor of 18.1% with the inclusion of 71 water molecules. The structure shows propeller twists of up to -20 degrees for the A.T base-pairs, although there is probably only one (weak) three-centre hydrogen bond in the six base-pair AT narrow minor-groove region. An extensive ribbon of hydration has been located in this groove that has features distinctive from the classic "spine of hydration". Solvation around phosphate groups is described, with several instances of water molecules bridging between phosphates.     

The binding of berenil to Escherichia coli ribosome.

The binding of the nonintercalating dye berenil to the 70S ribosome of Escherichia coli has been demonstrated by spectrophotometric measurements and gel filtration through Biogel P100 column. The berenil spectrum is gradually shifted towards the red region with the increasing amount of ribosome added, the isosbestic point being at 375 nm. There is positive cooperativity in the binding of berenil to the ribosome as demonstrated by the equilibrium dialysis. On binding with berenil, the ribosome is degraded faster by RNase I especially at low Mg++ concentration and its capacity to inhibit RNase I catalysed hydrolysis of ribopolymers is decreased. These indicate the unfolding of the structure of the ribosome.     

The structure of the zinc sites of Escherichia coli DNA-dependent RNA polymerase.

X-ray absorption spectroscopy is ideally suited for the investigation of the electronic structure and the local environment (approximately 5 A) of specific atoms in biomolecules. While the edge region provides information about the valence state of the absorbing atom, the chemical identity of neighboring atoms, and the coordination geometry, the extended x-ray absorption fine structure region contains information about the number and average distance of neighboring atoms and their relative disorder. The development of sensitive detection methods has allowed studies using near physiological concentrations (as low as approximately 100 microM). RNA polymerase from Escherichia coli contains two zinc atoms: one tightly bound in the beta' subunit, the subunit that participates in template binding, and the other loosely bound in the beta subunit, the subunit that participates in substrate binding. X-ray absorption studies of these zinc sites in the native protein and of the zinc site in the beta' subunit after removal of the zinc in the beta subunit site by p-(hydroxymercuri)benzenesulfonate (Giedroc, D. P., and Coleman, J. E. (1986) Biochemistry 25, 4969-4978) indicate that both zinc sites have octahedral coordination. The zinc in the beta' subunit site has four sulfur ligands at an average distance of 2.36 +/- 0.02 A and two oxygen (or nitrogen) ligands at an average distance of 2.23 +/- 0.02 A. The beta subunit zinc site has five sulfur ligands at an average distance of 2.38 +/- 0.01 A and one histidine nitrogen ligand at 2.14 +/- 0.02 A. These results are in general agreement with earlier biochemical and spectroscopic studies.