Peptides corresponding to the N- and C-terminal parts of PEBP are well-structured in solution: new insights into their possible interaction with partners in vivo

Recently, it has been shown that mammalian PEBPs are implicated in several signalling pathways controlling the cellular cycle. In particular, during brain development, the N-terminal part of mammalian PEBP is specifically cleaved and the resulting 11 amino acid peptide stimulates the growth and activity of acetylcholinergic neurons. The crystallographic structure of bovine and human PEBPs has revealed that their N- and C-terminal parts are accessible and exposed to the solvent suggesting that they may be involved in specific interactions with cellular partners. We have chemically synthetized the two peptides corresponding to these terminal parts and studied their structure in solution by circular dichroism and NMR spectroscopies: both of them are well-structured. The N-terminal peptide is composed of a series of turns, leading to a hook conformation. The C-terminal peptide displays a globally helical conformation similar to that observed in the whole protein; it is characterized by an amphipatic feature with a hydrophobic cluster located on one side. These structural features enlighten previous fluorescence and monolayer experiments and give new insights on the roles of both PEBP termini.

Solution structures of the antifungal heliomicin and a selected variant with both antibacterial and antifungal activities

In response to an experimental infection, the lepidopteran Heliothis virescens produces an antifungal protein named heliomicin. Heliomicin displays sequence similarities with antifungal plant defensins and antibacterial or antifungal insect defensins. To gain information about the structural elements required for either antifungal or antibacterial activity, heliomicin and selected point-mutated variants were expressed in yeast as fusion proteins. The effects of mutations, defined by comparing the primary structure of heliomicin with the sequences of members of the insect defensin family, were analyzed using antibacterial and antifungal assays. One of the variants shows significant activity against Gram-positive bacteria while remaining efficient against fungi. The three-dimensional structures of this variant and of the wild-type protein were determined by two-dimensional (1)H NMR to establish a correlation between structure and antibacterial or antifungal activity. Wild-type and mutated heliomicins adopt a similar scaffold, including the so-called cysteine-stabilized alphabeta motif. A comparison of their structures with other defensin-type molecules indicates that common hydrophobic characteristics can be assigned to all the antifungal proteins. A comparative analysis of various structural features of heliomicin mutant and of antibacterial defensins enables common properties to be assessed, which will help to design new mutants with increased antibacterial activity.

Magnetization transfer from laser-polarized xenon to protons located in the hydrophobic cavity of the wheat nonspecific lipid transfer protein

Nonspecific lipid transfer protein from wheat is studied by liquid-state NMR in the presence of xenon. The gas-protein interaction is indicated by the dependence of the protein proton chemical shifts on the xenon pressure and formally confirmed by the first observation of magnetization transfer from laser-polarized xenon to the protein protons. Twenty-six heteronuclear nOes have allowed the characterization of four interaction sites inside the wheat ns-LTP cavity. Their locations are in agreement with the variations of the chemical shifts under xenon pressure and with solvation simulations. The richness of the information obtained by the noble gas with a nuclear polarization multiplied by approximately 12,000 makes this approach based on dipolar cross-relaxation with laser-polarized xenon promising for probing protein hydrophobic pockets at ambient pressure.

The active site of drosomycin, a small insect antifungal protein, delineated by comparison with the modeled structure of Rs-AFP2, a plant antifungal protein

Drosomycin is the first strictly antifungal protein isolated from an insect (Drosophila melanogaster). The solution structure of this 44-residue protein has been reported previously. It involves a three-stranded beta-sheet and an alpha-helix, the protein global fold being maintained by four disulfide bridges. Rs-AFP2 is a plant antifungal protein exhibiting 41% sequence similarity with drosomycin. Mutational analysis of Rs-AFP2 showed the importance of some residues in the antifungal activity of the protein against the fungus target. In order to determine the structural features responsible for antifungal activity in both drosomycin and Rs-AFP2, we modeled the three-dimensional structure of Rs-AFP2, and of other antifungal proteins, using the solution structure of drosomycin as a template. Structure analysis of drosomycin and Rs-AFP2, and comparisons with the other modeled antifungal structures, revealed that the two proteins shared a hydrophobic cluster located at the protein surface in which a lysine residue is embedded. Based on these close structural similarities and the experimental data available for Rs-AFP2 mutants, an antifungal active site of the insect protein is proposed.

The wide binding properties of a wheat nonspecific lipid transfer protein. Solution structure of a complex with prostaglandin B2

The 3D solution structure of wheat nonspecific lipid transfer protein (ns-LTP) complexed with prostaglandin B2, a lipid with both vinyl and hydroxylated groups, has been determined by 1H 2D NMR. The global fold of the protein is close to the previously published structures of wheat, maize, barley and rice ns-LTPs. The ligand is almost completely embedded in the hydrophobic core of the protein. Structure comparisons of free and bound wheat ns-LTP reveal that the binding of prostaglandin B2 hardly affects the global fold of the protein. The structural data on this unusual complex are discussed and compared with other known ns-LTP lipid-complexes.

Solvation study of the non-specific lipid transfer protein from wheat by intermolecular NOEs with water and small organic molecules

Intermolecular nuclear Overhauser effects (NOEs) were measured between the protons of various small solvent or gas molecules and the non-specific lipid transfer protein (ns-LTP) from wheat. Intermolecular NOEs were observed with the hydrophobic pocket in the interior of wheat ns-LTP, which grew in intensity in the order cyclopropane (saturated solution) < methane (140 bar) < ethane (40 bar) < acetonitrile (5% in water) < cyclohexane (saturated solution) < benzene (saturated solution). No intermolecular, NOEs were observed with dioxane (5% in water). The intermolecular NOEs were negative for all of the organic molecules tested. Intermolecular NOEs between wheat ns-LTP and water were weak or could not be distinguished from exchange-relayed NOEs. As illustrated by the NOEs with cyclohexane versus dioxane, the hydrophobic pocket in wheat ns-LTP preferably binds non-polar molecules. Yet, polar molecules like acetonitrile can also be accommodated. The pressure dependence of the NOEs between methane and wheat ns-LTP indicated incomplete occupancy, even at 190 bar methane pressure. In general, NOE intensities increased with the size of the ligand molecule and its vapor pressure. NMR of the vapor phase showed excellent resolution between the signals from the gas phase and those from the liquid phase. The vapor concentration of cyclohexane was fivefold higher than that of the dioxane solution, supporting the binding of cyclohexane versus uptake of dioxane.

Androctonin, a novel antimicrobial peptide from scorpion Androctonus australis: solution structure and molecular dynamics simulations in the presence of a lipid monolayer

Androctonin is a highly cationic antimicrobial peptide from scorpion exhibiting a broad spectrum of activities against bacteria and fungi. It contains 25 amino acids including four cysteine residues forming two disulfide bridges. We report here on the determination of its solution structure by conventional two-dimensional (2D) 1H-NMR spectroscopy and molecular modelling using distance geometry and molecular dynamics methods. The structure of androctonin involves a well-defined highly twisted anti-parallel beta-sheet with strands connected by a more variable positively charged turn. A comparison with the structure of tachyplesin I (horseshoe crab) reveals that the amphiphilic character of the protein surface of this homologous peptide is not observed in androctonin. We have undertaken a 200-ps molecular dynamics simulation study on a system including one androctonin molecule and a monolayer of DMPG (1,2-dimyristoylphosphatidylglycerol) lipids. On the basis of this simulation, the first steps of the membrane permeabilization process are discussed.

Solution structure of a lipid transfer protein extracted from rice seeds. Comparison with homologous proteins

Nuclear magnetic resonance (NMR) spectroscopy was used to determine the three dimensional structure of rice nonspecific lipid transfer protein (ns-LTP), a 91 amino acid residue protein belonging to the broad family of plant ns-LTP. Sequence specific assignment was obtained for all but three HN backbone 1H resonances and for more than 95% of the 1H side-chain resonances using a combination of 1H 2D NOESY; TOCSY and COSY experiments at 293 K. The structure was calculated on the basis of four disulfide bridge restraints, 1259 distance constraints derived from 1H-1H Overhauser effects, 72 phi angle restraints and 32 hydrogen-bond restraints. The final solution structure involves four helices (H1: Cys3-Arg18, H2: Ala25-Ala37, H3: Thr41-Ala54 and H4: Ala66-Cys73) followed by a long C-terminal tail (T) with no observable regular structure. N-capping residues (Thr2, Ser24, Thr40), whose side-chain oxygen atoms are involved in hydrogen bonds with i + 3 amide proton additionally stabilize the N termini of the first three helices. The fourth helix involving Pro residues display a mixture of alpha and 3(10) conformation. The rms deviation of 14 final structures with respect to the average structure is 1.14 +/- 0.16 A for all heavy atoms (C, N, O and S) and 0.72 +/- 0.01 A for the backbone atoms. The global fold of rice ns-LTP is close to the previously published structures of wheat, barley and maize ns-LTPs exhibiting nearly identical pattern of the numerous sequence specific interactions. As reported previously for different four-helix topology proteins, hydrophobic, hydrogen bonding and electrostatic mechanisms of fold stabilization were found for the rice ns-LTP. The sequential alignment of 36 ns-LTP primary structures strongly suggests that there is a uniform pattern of specific long-range interactions (in terms of sequence), which stabilize the fold of all plant ns-LTPs.

Solution structure of thanatin, a potent bactericidal and fungicidal insect peptide, determined from proton two-dimensional nuclear magnetic resonance data

Thanatin is the first inducible insect peptide that has been found to have, at physiological concentrations, a broad range of activity against bacteria and fungi. Thanatin contains 21 amino acids including two cysteine residues that form a disulfide bridge. Two-dimensional (2D) 1H-NMR spectroscopy and molecular modelling have been used to determine its three-dimensional (3D) structure in water. Thanatin adopts a well-defined anti-parallel beta-sheet structure from residue 8 to the C-terminus, including the disulfide bridge. In spite of the presence of two proline residues, there is a large degree of structural variability in the N-terminal segment. The structure of thanatin is quite different from the known structures of other insect defence peptides, such as antibacterial defensin and antifungal drosomycin. It has more similarities with the structures of various peptides from different origins, such as brevinins, protegrins and tachyplesins, which have a two-stranded beta-sheet stabilized by one or two disulfide bridges. Combined with activity test experiments on several truncated isoforms of thanatin, carried out by Fehlbaum et al. [Fehlbaum, P., Bulet, P., Chernysh, S., Briand, J. P., Roussel, J. P., Letellier, L., Hétru, C. & Hoffmann, J. (1996) Proc. Natl Acad. Sci. USA 93, 1221-1225], our structural study evidences the importance of the beta-sheet structure and also suggests that anti-Gram-negative activity involves a site formed by the Arg20 side-chain embedded in a hydrophobic cluster.

Comparison of solution and crystal structures of maize nonspecific lipid transfer protein: a model for a potential in vivo lipid carrier protein

The three-dimensional solution structure of maize nonspecific lipid transfer protein (nsLTP) obtained by nuclear magnetic resonance (NMR) is compared to the X-ray structure. Although both structures are very similar, some local structural differences are observed in the first and the fourth helices and in several side-chain conformations. These discrepancies arise partly from intermolecular contacts in the crystal lattice. The main characteristic of nsLTP structures is the presence of an internal hydrophobic cavity whose volume was found to vary from 237 to 513 A3 without major variations in the 15 solution structures. Comparison of crystal and NMR structures shows the existence of another small hollow at the periphery of the protein containing a water molecule in the X-ray structure, which could play an important structural role. A model of the complexed form of maize nsLTP by alpha-lysopalmitoylphosphatidylcholine was built by docking the lipid inside the protein cavity of the NMR structure. The main structural feature is a hydrogen bond found also in the X-ray structure of the complex maize nsLTP/palmitate between the hydroxyl of Tyr81 and the carbonyl of the lipid. Comparison of 12 primary sequences of nsLTPs emphasizes that all residues delineating the cavities calculated on solution and X-ray structures are conserved, which suggests that this large cavity is a common feature of all compared plant nsLTPs. Furthermore several conserved basic residues seem to be involved in the stabilization of the protein architecture.

Solution structure of Ace-AMP1, a potent antimicrobial protein extracted from onion seeds. Structural analogies with plant nonspecific lipid transfer proteins

The three-dimensional solution structure of Ace-AMP1, an antifungal protein extracted from onion seeds, was determined using 1H NMR spectroscopy and molecular modeling. This cationic protein contains 93 amino acid residues and four disulfide bridges. Its structure was determined from 1260 NOE-derived distance restraints and 173 dihedral restraints derived from NOEs and 3JCaHNH coupling constants. The global fold involves four helical segments connected by three loops and a C-terminal tail without regular secondary structures, except for a 3(10)-helix turn and a beta-turn. The most striking feature is the absence of any continuous cavity running through the whole molecule as found in recently determined structures of nonspecific transfer proteins extracted from wheat and maize seeds, although their global folds are very similar. Consistent with the absence of a cavity in the core of Ace-AMP1, it was found that this protein, in contrast to ns-LTPs, does not bind fluorescently labeled phospholipids in solution. On the other hand, Ace-AMP1 is able to interact with phospholipid membranes as shown by the release of carboxyfluorescein from the lumen of artificial liposomes and by the induction of alterations in fluorescence polarization of fluorescently labeled phospholipids embedded in artificial liposomes.

Refined solution structure of the anti-mammal and anti-insect LqqIII scorpion toxin: comparison with other scorpion toxins

The solution structure of the anti-mammal and anti-insect LqqIII toxin from the scorpion Leiurus quinquestriatus quinquestriatus was refined and compared with other long-chain scorpion toxins. This structure, determined by 1H-NMR and molecular modeling, involves an alpha-helix (18-29) linked to a three-stranded beta-sheet (2-6, 33-39, and 43-51) by two disulfide bridges. The average RMSD between the 15 best structures and the mean structure is 0.71 A for C alpha atoms. Comparison between LqqIII, the potent anti-mammal AaHII, and the weakly active variant-3 toxins revealed that the LqqIII three-dimensional structure is closer to that of AaHII than to the variant-3 structure. Moreover, striking analogies were observed between the electrostatic and hydrophobic potentials of LqqIII and AaHII. Several residues are well conserved in long-chain scorpion toxin sequences and seem to be important in protein structure stability and function. Some of them are involved in the CS alpha beta (Cysteine Stabilized alpha-helix beta-sheet) motif. A comparison between the sequences of the RII rat brain and the Drosophila extracellular loops forming scorpion toxin binding-sites of Na+ channels displays differences in the subsites interacting with anti-mammal or anti-insect toxins. This suggests that hydrophobic as well as electrostatic interactions are essential for the binding and specificity of long-chain scorpion toxins.

Distortions of the DNA double helix induced by 1,3-trans-diamminedichloroplatinum(II)-intrastrand cross-link: an internal coordinate molecular modeling study

A trans-diamminedichloroplatinum(II) (trans-DDP) intrastrand adduct within the sequence d(TCTG*TG*TC).d(GACACAGA) (where G* represents a platinated guanine) is modeled on the basis of qualitative experimental data concerning global unwinding and curvature as well as information on base pairing. Modeling is performed using the internal coordinate JUMNA program, specific to nucleic acids, and modified to include the possibility of covalently bound ligands. Calibration of the energy functions representing the Pt-N7 bond with guanine is described. The platinum atom and the platinum-nitrogen bonds are parameterized for use in the Hückel Del Re method to calculate monopoles at each atom. These monopoles are consistent with the Flex force field included in Jumna. By developing an appropriate minimization protocol we are able to generate stable, distorted three-dimensional structures compatible with the experimental data and including an unusually high global unwinding. No a priori geometric assumptions are made in generating these structures.

Solution structure of the wild-type HMGI a box by 1 h and 15n NMR and molecular modelling

Abstract:

The solution structure of the wild-type A box of rat HMG1 has been determined by heteronuclear three-dimensional NMR and molecular modelling. This structure adopts a "L" shape structure globaly similar to that described for a mutant A box and other HMG boxes. However the wild-type A box contains two Cys residues which could form intramolecular bridge in an oxidative environment. This should be into competition with a contribution of the thiol groups in the binding ofHMG proteins to cisplatin-modified DNA.

Homology modelling of an antimicrobial protein, Ace-AMP1, from lipid transfer protein structures

BACKGROUND

Plant nonspecific lipid transfer proteins (ns-LTPs) are small basic proteins that facilitate lipid shuttling between membranes in vitro. The function of ns-LTPs in vivo is still unknown. It has been suggested, in relation to their lipid binding ability, that they may be involved in cutin formation. Alternatively, they may act in the plant defence system against pathogenic agents. Ace-AMP1 is an antimicrobial protein extracted from onion seed that shows sequence homology with ns-LTPs but that is unable to transfer lipids. We have recently determined the three-dimensional structure of wheat and maize ns-LTPs. In order to compare the structural features of Ace-AMP1 and ns-LTPs, we have used the comparative modelling software MODELLER to predict the structure of Ace-AMP1.

RESULTS

The global fold of Ace-AMP1 is very similar to those of ns-LTPs, involving four helices and a C-terminal tail without secondary structure elements. The structure of maize and wheat ns-LTP is characterized by the existence of a tunnel-like hydrophobic cavity in which a lipid molecule can be inserted. In the Ace-AMP1 structure, this cavity is blocked by a number of bulky residues. Similarly, the electrostatic potential contours of ns-LTPs show some common features that were not observed in Ace-AMP1.

CONCLUSIONS

Although Ace-AMP1 displays a similar global fold to ns-LTPs, it does not present a hydrophobic cavity, which may explain why Ace-AMP1 cannot shuttle lipids between membranes in vitro. The large differences in the electrostatic properties of Ace-AMP1 and ns-LTPs suggest a different mode of interaction with membranes.

Molecular modelling study of the netropsin complexation with a nucleic acid triple helix

A detailed molecular mechanical study has been made on the complexes of netropsin with the double stranded oligonucleotide (dA)12.(dT)12 and with the triple helix (dA)12.(dT)12.(dT)12. The complexes were built using computer graphics and energy refined using JUMNA program. In agreement with circular dichroism experiments we have shown that 3 netropsins can bind the minor grooves of the triple helix and of the double helix. The groove geometry in the duplex and in the triplex is very similar. However a detailed analysis of the energetic terms shows, in agreement with thermal denaturation studies, that the affinity of netropsin toward the double helices is larger than towards triple helices.

Solution structure and lipid binding of a nonspecific lipid transfer protein extracted from maize seeds

The three-dimensional solution structure of a nonspecific lipid transfer protein extracted from maize seeds determined by 1H NMR spectroscopy is described. This cationic protein consists of 93 amino acid residues. Its structure was determined from 1,091 NOE-derived distance restraints, including 929 interresidue connectivities and 197 dihedral restraints (phi, psi, chi 1) derived from NOEs and 3J coupling constants. The global fold involving four helical fragments connected by three loops and a C-terminal tail without regular secondary structures is stabilized by four disulfide bridges. The most striking feature of this structure is the existence of an internal hydrophobic cavity running through the whole molecule. The global fold of this protein, very similar to that of a previously described lipid transfer protein extracted from wheat seeds (Gincel E et al., 1994, Eur J Biochem 226:413-422) constitutes a new architecture for alpha-class proteins. 1H NMR and fluorescence studies show that this protein forms well-defined complexes in aqueous solution with lysophosphatidylcholine. Dissociation constants, Kd, of 1.9 +/- 0.6 x 10(-6) M and > 10(-3) M were obtained with lyso-C16 and -C12, respectively. A structure model for a lipid-protein complex is proposed in which the aliphatic chain of the phospholipid is inserted in the internal cavity and the polar head interacts with the charged side chains located at one end of this cavity. Our model for the lipid-protein complex is qualitatively very similar to the recently published crystal structure (Shin DH et al., 1995, Structure 3:189-199).

1H-NMR-derived secondary structure and the overall fold of the potent anti-mammal and anti-insect toxin III from the scorpion Leiurus quinquestriatus quinquestriatus

We describe the secondary structure and the overall fold of toxin III from the venom of the scorpion Leiurus quinquestriatus quinquestriatus determined using two-dimensional-1H-NMR spectroscopy. This protein, which contains 64 amino acids and 4 disulfide bridges, belongs to the long-chain toxin category and is highly toxic to both mammals and insects. The overall fold was determined on the basis of 1208 inter-proton-distance restraints derived from NOE measurements and 90 psi, phi dihedral-angle restraints derived from NOE connectivities and 3JNH-alphaH coupling constants using the HABAS program. This fold, which mainly consists of an alpha-helix packed against a small antiparallel three-stranded beta-sheet, and of several turns and loops, is similar to that of other long-chain scorpion toxins. Aromatic and non-polar residues form several patches on the surface of the protein which alternate with patches of charged and polar residues. Such a topology should be important in the interactions of toxin III with sodium channels in membranes. Two weakly constrained loops introduce some flexibility to the structure which could be related to the activity of this toxin. The central core of toxin III is compared with the cysteine-stabilized alpha beta motif (an alpha-helix connected to a beta-sheet through two disulfide bridges) found in insect defensins and plant thionins. Defensins and thionins are small proteins (approximately 40--50 amino acid residues) containing three or four disulfide bridges, respectively. This comparison confirms that the cysteine-stabilized alpha beta motif is a common core to a number of small proteins from different origins and having different activities.

Refined three-dimensional solution structure of insect defensin A

BACKGROUND

Insect defensin A is a basic 4 kDa protein secreted by Phormia terranovae larvae in response to bacterial challenges or injuries. Previous biological tests suggest that the bacterial cytoplasmic membrane is the target of defensin A. The structural study of this protein is the first step towards establishing a structure-activity relationship and forms the basis for understanding its antibiotic activity at the molecular level.

RESULTS

We describe a refined model of the three-dimensional structure of defensin A derived from an extensive analysis of 786 inter-proton nuclear Overhauser effects. The backbone fold involves an N-terminal loop and an alpha-helical fragment followed by an antiparallel beta-structure. The helix and the beta-structure are connected by two of the three disulphide bridges present in defensin A, forming a so-called 'cysteine-stabilized alpha beta' (CS alpha beta) motif. The N-terminal loop, which is locally well defined, can occupy different positions with respect to the other moieties of the molecule.

CONCLUSIONS

The CS alpha beta motif, which forms the core of the defensin A structure, appears to be a common organization for several families of small proteins with toxic properties. The distribution of amino acid side chains in the protein structure creates several hydrophobic or hydrophilic patches. This leads us to propose that the initial step in the action of positively charged defensin A molecules with cytoplasmic membranes may involve interactions with acidic phospholipids.

Solution structure of oligonucleotides covalently linked to a psoralen derivative

Psoralen (pso) was attached via its C-5 position to the 5'-phosphate group of an oligodeoxynucleotide d(TAAGCCG) by a hexamethylene linker (m6). Complex formation between pso-m6-d(TAAGCCG) and the complementary strands d(CGGCTTA)[7-7mer] or d(CGGCTTAT)[7-8mer] was investigated by nuclear magnetic resonance in aqueous solution. Structural informations derived from DQF-COSY and NOESY maps, revealed that the mini double helix adopts a B-form conformation and that the deoxyriboses preferentially adopt a C2'-endo conformation. The nOe connectivities observed between the protons of the bases or the sugars in each duplex, and the protons of the psoralen and the hexamethylene chain, led us to propose a model involving an equilibrium between two conformations due to different locations of the psoralen. Upon UV-irradiation, the psoralen moiety cross-linked the two DNA strands at the level of 5'TpA3' sequences. NMR studies of the single major photo-cross-linked duplex pso-m6-d(TAAGCCG) and d(CGGCTTA) were performed. The stereochemistry of the diadduct is indeed cis-syn at both cyclobutane rings. In addition, the effects of this diadduct on the helical structure are analyzed in detail.

Three-dimensional structure in solution of a wheat lipid-transfer protein from multidimensional 1H-NMR data. A new folding for lipid carriers

Two-dimensional and three-dimensional 1H-NMR experimental data [Simorre, J. P., Caille, A., Marion, D., Marion, D. & Ptak, M. (1991) Biochemistry 30, 11600-11608] were used to build models of the three-dimensional structure of a non-specific wheat lipid-transfer protein (LTP) by using distance geometry, simulated annealing, energy minimization and molecular dynamics techniques. A first set of 881 distance constraints derived from NOE cross-peak intensities was used to generate 74 initial structures. One family of topological mirror images of the protein structure was eliminated by considering helical secondary-structure organization and steric requirements. Back calculations of NOE intensities led us to introduce 535 additional distance constraints. Finally, 21 structures were selected as representative of the structure of the protein. The polypeptide backbone folds into a simple and original right-handed winding. It is composed of a bundle of four helices linked by flexible loops, which is packed against a C-terminal fragment forming a non-standard saxophone-like shape. The folded protein is stabilized by hydrophobic interactions and the four disulfide bridges combined by pairs on each side of the protein. An hydrophobic cleft, formed by residues located in the second half of the protein could be a potential site for the binding of lipids.

Comparison of solution structure of free and complexed lac operator by molecular modelling with NMR constraints

The structure difference between the free operator of the lac system d(GCTCACAAT).d(ATTGTGAGC) and the same operator complexed to the headpiece of the lac repressor has been investigated by 2-D-1H NMR spectroscopy in conjunction with molecular modelling in internal coordinates (JUMNA). The free and complexed operator adopt both a right-handed B helical conformation, but a more detailed analysis of the conformational parameters using the Curves program shows striking differences in the groove geometries, the rises, the twists and the total bending.

Progress in multidimensional NMR investigations of peptide and protein 3-D structures in solution. From structure to functional aspects

2-D and 3-D NMR techniques were used to investigate the conformations in solution of several peptides and proteins for which crystalline structures are not available yet. Insect defensin A is a small (40 aa) antibiotic protein exhibiting a characteristic 'loop-helix-beta-sheet' structure. A striking analogy was found with charybdotoxin, a scorpion toxin in which a CSH (cysteine stabilized alpha-helix) motif is also present. Wheat phospholipid transfer protein (PLTP) (90 aa) has a 3-D structure resulting from the packing of four helices and of a C-terminal less well-defined fragment. Preliminary results show that PLTP forms a complex with lyso-PC and that such an interaction results in a conformational change affecting principally the C-terminal half of the protein. A last example is given with surfactin, a lipopeptide biosurfactant from bacterial origin. Its protonated form shows a very compact structure in which the two acidic residues located on the top of a 'horse saddle' topology face each other, whereas the ionized form could adopt a more extended conformation. A common property of these compounds is their capacity to interact with lipids. The present structural data open the way for a further establishment of structure-activity relationships.

Distortions induced in DNA by cis-platinum interstrand adducts

A 22 base pair double-stranded oligonucleotide containing a unique interstrand adduct resulting from chelation of the two guanine residues within the central sequence d(TGCT/AGCA) by a cis-platinum residue has been studied by means of gel electrophoresis, chemical probes, and molecular mechanics. The anomalously slow electrophoretic mobility of the multimers of the platinated and ligated oligomers suggests that the platinated oligonucleotide is bent. The two cytosine residues (complementary to the platinated guanines) are hyperreactive to hydroxylamine, indicating a large exposure of the two bases to the solvent. The adduct does not induce a local denaturation within the flanking sequences since the adenine residues are not reactive with diethyl pyrocarbonate. This is confirmed by the nonreactivity of the complementary T residues with osmium tetraoxide. These results and the molecular mechanics modeling suggest that the interstrand adduct bends the double helix by approximately 55 degrees toward the major groove, that the double helix conserves its average twist angle, and that the distortion induced by the adduct is localized at the platinated sequence d(GC/CG).

Sequence dependent hydration of DNA

The transitions between the different helical conformations of DNA depend on the base sequence and the ambient conditions such as humidity and counter-ion concentration. In this study energy minimization techniques have been used to locate water molecule sites around nucleotides especially those which form hydrogen bonds between two or more nucleotide atoms and thus form solvent mediated bridges. We have studied several sequences and find that those which are known not to exist in the low hydration 'A' form have very similar number of bridging sites in both 'A' and 'B' conformations. Those sequences which are found in the 'A' conformation have considerably more bridging sites in this low hydration form than in the 'B' conformation. Sequence related solvent effects for a given conformation have also been analysed.

2D NMR studies and 3D structure of the parallel-stranded duplex oligonucleotide Acrm5-alpha-d(TCTAAACTC)-beta-d(AGATTTGAG) via complete relaxation matrix analysis of the NOE effects and molecular mechanics calculations

The three-dimensional structure of the duplex formed by the association of the unnatural oligonucleotide alpha-d(TCTAAACTC) covalently linked to an acridine derivative (m5Acr) with its natural and parallel complementary sequence beta-d(AGATTTGAG) was investigated by nuclear magnetic resonance spectroscopy and constrained molecular mechanics calculations. All the nonexchangeable and exchangeable resonances were assigned in this duplex. The structure was refined by using interproton distances determined by NOE measurements. The NOE values were converted into distances by using the complete 190 x 190 relaxation matrix. The unnatural duplex Acrm5-alpha-d(TCTAAACTC)-beta-d(AGATTTGAG) forms a parallel right-handed helix with Watson-Crick base pairing; the alpha and beta deoxyriboses adopt a 3'-exo conformation. The acridine moiety was found stacked up the C9-G9 base pair. The structure of the first seven base pairs of this duplex was found similar to that of the duplex alpha-d(TCTAAAC)-beta-d(AGATTTG), which we had already investigated [Lancelot, G., et al. (1989) Biochemistry 28, 7871-7878]. Since these structures were generated by using experimental NOE values obtained independently on macromolecules whose global correlation time was different (3.8 and 2.2 ns), we conclude that this comparison is a good test of the viability of our method to generate three-dimensional structures of oligonucleotides in solution. Starting from different initial conformations, we show that the NOE constraints allow one to reach the same final restrained conformation, taking into account implicitly the solvent effect.

Solution structure of the parallel-stranded duplex oligonucleotide alpha-d(TCTAAAC)-beta-d(AGATTTG) via complete relaxation matrix analysis of the NOE effects and molecular mechanics calculations

The solution structure of the duplex formed by the association of the unnatural oligonucleotide alpha-d(TCTAAAC) with its natural and parallel complementary sequence beta-d(AGATTTG) was investigated by nuclear magnetic resonance spectroscopy and constrained molecular mechanics calculations. The structure was refined on the basis of interproton distances determined by NOE measurements for a series of mixing times. The NOE values were converted to distances by using the complete 134 x 134 relaxation matrix including all proton dipole-dipole interactions and spin diffusion. The computation of the relaxation matrix requires the Cartesian coordinates of the oligonucleotide, which are not known, a priori. To avoid this ambiguity, we used an iterative procedure in which the new distance constraints are obtained by using the complete relaxation matrix calculated from the previous structure. After three iterations, the process converged. The unnatural duplex alpha-d(TCTAAAC)-beta-d(AGATTTG) adopts in solution a right-helical structure with Watson-Crick base pairing, an anti conformation on the glycosyl linkage on the beta-strand, a syn conformation on the alpha-strand, and a 3'-exo conformation of the deoxyriboses for both sugar anomers. The three-dimensional structure obtained allowed us to describe the local heterogeneity of the duplex.

Biomolecular energy calculations using transputer technology

We report an application of current parallel processing transputer technology which has readily achieved a 25-fold reduction in computational time of peptide-solvent interactions.

Structure of the intrastrand cis-[Pt(NH3)2(d(GpCpG))] adduct in a dodecanucleotide duplex: II. A molecular mechanics modeling study

Molecular mechanics modeling has been carried out for the intrastrand cis-[Pt(NH3)2(d(pGpCpG))] adduct in a dodecanucleotide duplex. In a first step, an analysis of the conformation of a platinated trinucleotide shows the tendency of the two chelating guanines to take respectively anti-syn or syn-anti positions and the role of phosphate-platinum amino group interactions. In a second step, duplex structures are examined. The distortions induced by platination are especially analysed for kinked and unkinked forms of the dodecanucleotide. In all models the central C.G pair in the platination site is disrupted and the double helix is more disturbed on the 5' side of this site than on the 3' side. Most of the structural features arising from the modeling are in agreement with the conclusions of an n.m.r. study reported in the preceding paper.

Solvent bridging sites in A- and B-DNA helices

Nucleotide hydration is important for the understanding of the stability of and the transitions between the different helical conformations of DNA. We have used energy minimization and geometric criteria in order to look for possible sites for solvent which can bridge more than one polar or charged atomic group on a nucleotide. Such bridging sites between phosphate groups have been seen experimentally and used to explain the A to B transition. We show that these phosphate bridging sites occur at energy minima around A-DNA but do not occur around B-DNA. We also find that there are further low energy bridging sites which depend on sequence and which enable the more economical hydration of the A form.

Computer simulation of aqueous biomolecular systems

Computer simulation techniques are increasingly being used to predict structural and thermodynamic properties of large heterogeneous macromolecule and solvent assemblies. We discuss, with examples from our own studies, some problems we and others have experienced in using these techniques, which were originally devised for simple liquids. In particular, we consider the problems which arise from the large size and heterogeneity of macromolecule water systems, comparisons with experimental data and equilibrium and sampling procedures.

A fractional charge model for empirical calculations of peptide-water interactions

The properties of an empirical model of interaction between a water molecule and polar groups of peptides or small peptides are explored. The H2O molecule is represented by a four-point charges distribution. In electron donor groups, a point charge is located on the axis of the lone pairs orbitals in order to introduce some directionality in hydrogen bonds. The effective potential is approximated by the sum of the coulombic interactions between point charges distribution and of a 6--12 atom-atom potential. The coefficients of this last potential are first adjusted by simulating the geometry of the water dimer. Equilibrium configurations of associated polar molecules and H2O predicted by the model are found to be in good agreement with those resulting from more sophisticated ab initio SCF calculations. Interactions between H2O and the side-chains of the cyclic dipeptide C(L-Thr-L-His) are then calculated. It is shown that internal bridging by water is an essential effect of the solvent. The experimental position of the H2O molecule is reproduced, stability of which depends also on intermolecular interactions.