NOE data at 500 MHz reveal the proximity of phenyl and tyrosine rings in enkephalin.

Met5-enkephalin-a pentapeptide (Tyr-Gly-Gly-Phe-Met)-can exist in two possible folded arrangements with a rigid two-hydrogen-bonded network. In one arrangement, a Gly 2-Gly 3 beta-bend is formed and in the other a Gly 3-Phe 4 beta-bend. The two conformations are distinguished by the spatial relation of Tyr 1 and Phe 4: in the Gly 2-Gly 3 beta-bend, Tyr 1 and Phe 4 can be brought close to each other while in the Gly 3-Phe 4 beta-bend they are far apart (greater than 5 A). We have utilized one-dimensional (1D) nuclear Overhauser effect (NOE) measurements between the ring protons of Tyr 1 and Phe 4 to determine their proximity. The NOE data clearly show that a pair protons, one each from Tyr 1 and Phe 4, are as close as 3.3 A while other inter-proton distances are beyond 4.5 A. Therefore, we propose the presence of a Gly 2-Gly 3 beta-bend (in which Tyr 1 and Phe 4 are spatially close) for Met5-enkephalin in solution. The structure of Met5-enkephalin in solution is very similar to the single crystal structure of Leu5-enkephalin and tends to explain the biological activity data of several modified enkephalins.

Intramolecular conformation of puromycin in solution as studied by proton magnetic resonance.

The intramolecular conformation of puromycin, a broad spectrum antibiotic, in solution has been investigated by proton magnetic resonance (PMR) spectroscopy. A comparison of the proton chemical shift and proton-proton coupling constant data of puromycin with puromycin aminonucleoside suggests that puromycin in solution exists as an equilibrium blend of extended and folded conformers. These folded conformers are the result of flexibility around the C alpha -C beta bond of the aminoacyl segment of puromycin. One of the folded conformers predicted by PMR is in excellent agreement with the x-ray data.

Self-association of puromycin as studied by proton magnetic resonance spectroscopy.

The self-association of puromycin has been studied using proton magnetic resonance spectroscopy. The concentration, temperature and pH dependence studies of the proton chemical shifts of the adenine protons indicate that puromycin in aqueous solution at pD 7.4 self associates predominantly through adenine-adenine interaction. At this pD, the amino group of the aminoacyl segment of puromycin has been demonstrated to exist in a equilibrium blend of protonated and non-protonated forms. At pD 2.6, PM is found to exist predominantly in the monomeric from in which the methyl groups of the 6N-dimethyladenine are found to be non-equivalent due to hindered rotation about the C6-N6 bond.

Left-handed intercalated DNA double helix: rendezvous of ethidium and actinomycin D in the Z-helical conformation space.

It is now very well recognized that the DNA double helix is conformationally pluralistic and that this flexibility is derived from internal motions due to backbone torsions. But what is less apparent is that such internal motions can occur in a correlated fashion and express themselves in a wide variety of structural motifs and phenomena. For example, flexibility inherent in the DNA molecule can lead to a family of Z-DNA, LZ1 and LZ2 being the two extremes and correlated internal motion can cause LZ1 in equilibrium LZ2 transition. More interestingly, such motions manifest themselves as breathing modes on the DNA lattice resulting in the sequence specific intercalation sites. Following a detailed stereochemical analyses we observed that the intercalation site for ethidium is located at the dCpdG sequence of the intercalated LZ1 helix (LZ1*) while that for actinomycin D is located at the dGpdC sequence of the intercalated LZ2 helix (LZ2*). From the stereochemistry of the drug binding we make experimentally testable predictions which are in fact supported by a few recent experimental studies. These studies also show that a left-handed intercalated B-DNA model is a viable intermediate in the Z to B transition which can hold the drug with binding energy comparable to that of the intercalated right-handed B-DNA.

During B-Z transition there is no large scale breakage of Watson-Crick base pairs. A direct demonstration using 500 MHz 1H NMR spectroscopy.

Monitoring of the Watson-Crick GNH1 proton in poly(dG-dC).poly(dG-dC) at 500 MHz in 90% H2O:10% D2O at 30 degrees C as a function of NaCl concentration (1.5 to 3.6 M), demonstrates that the bases retain Watson-Crick pairing throughout the transition. This observation unequivocally demonstrates that during the B-Z transition there is no large scale and detectable base pair opening and that macroscopically the phenomenon can be described as a direct helix to helix transition. We present frame by frame, an energetically sound stereodynamical trajectory for this transfiguration from right-handed B-DNA to left-handed Z-DNA.

Determination of handedness of DNA double helices from NOE difference spectra: the structure of poly(dG-dC).poly(dG-dC) in low salt.

Combined use of shielding constant computations, measurements of chemical shifts and NOE studies reveal that poly(dG-dC).(poly)dG-dC) in low salt solutions exist as a right-handed B-DNA double helix described by Gupta, Dhingra, Sarma, Sarma, Rajagopalan and Sasisekharan, J. Biomole. Str. Dyn. 1. 395, 1983. We present a simple and direct method to determine the handedness of DNA double helices from NOE difference spectra. This method takes advantage of the NOE between base protons and the H2'H2" sugar protons; and in the difference NOE spectra in the H2'H2" region the signatures of the right and left-handed helices become imprinted.

Poly(dA-dT).poly(dA-dT) in low salt appears to be a left-handed B-helix combined use of chemical theory, fiber diffraction and NMR spectroscopy.

Poly(dA-dT).poly(dA-dT) can adopt the B- and D- forms in the fibrous state. Theoretical energy calculations and fiber diffraction analyses suggest that there can be three structural models of poly(dA-dT).poly(dA-dT) in each of these two forms viz right and left-handed Watson Crick models and left-handed Hoogsteen--a total of six possible models. Fiber data for the polymer in the B- or the D-form or energy calculations cannot distinguish any one model from the other. However, a comparison of observed proton chemical shifts with the theoretically computed ones and the NOE studies on exchangeable and nonexchangeable protons suggest that poly(dA-dT).poly(dA-dT) in low salt solution exists predominantly in the left-handed B-conformation.

Structure of an anti-HIV-1 hammerhead ribozyme complex with a 17-mer DNA substrate analog of HIV-1 gag RNA and a mechanism for the cleavage reaction: 750 MHz NMR and computer experiments.

The structure of an anti-HIV-1 ribozyme-DNA abortive substrate complex was investigated by 750 MHz NMR and computer modeling experiments. The ribozyme was a chimeric molecule with 30 residues-18 DNA nucleotides, and 12 RNA residues in the conserved core. The DNA substrate analog had 17 residues. The chimeric ribozyme and the DNA substrate formed a shortened ribozyme-abortive substrate complex of 47 nucleotides with two DNA stems (stems I and III) and a loop consisting of the conserved core residues. Circular dichroism spectra showed that the DNA stems assume A-family conformation at the NMR concentration and a temperature of 15 degrees C, contrary to the conventional wisdom that DNA duplexes in aqueous solution populate entirely in the B-form. It is proposed that the A-family RNA residues at the core expand the A-family initiated at the core into the DNA stems because of the large free energy requirement for the formation of A/B junctions. Assignments of the base H8/H6 protons and H1' of the 47 residues were made by a NOESY walk. In addition to the methyl groups of all T's, the imino resonances of stems I and III and AH2's were assigned from appropriate NOESY walks. The extracted NMR data along with available crystallographic data, were used to derive a structural model of the complex. Stems I and III of the final model displayed a remarkable similarity to the A form of DNA; in stem III, a GC base pair was found to be moving into the floor of the minor groove defined by flanking AT pairs; data suggest the formation of a buckled rhombic structure with the adjacent pair; in addition, the base pair at the interface of stem III and the loop region displayed deformed geometry. The loop with the catalytic core, and the immediate region of the stems displayed conformational multiplicity within the NMR time scale. A catalytic mechanism for ribozyme action based on the derived structure, and consistent with biochemical data in the literature, is proposed. The complex between the anti HIV-1 gag ribozyme and its abortive DNA substrate manifests in the detection of a continuous track of A.T base pairs; this suggests that the interaction between the ribozyme and its DNA substrate is stronger than the one observed in the case of the free ribozyme where the bases in stem I and stem III regions interact strongly with the ribozyme core region (Sarma, R. H., et al. FEBS Letters 375, 317-23, 1995). The complex formation provides certain guidelines in the design of suitable therapeutic ribozymes. If the residues in the ribozyme stem regions interact with the conserved core, it may either prevent or interfere with the formation of a catalytically active tertiary structure.

Solution conformation of a model hexapeptide containing RGD sequence.

The solution conformation of a model hexapeptide Asp-Arg-Gly-Asp-Ser-Gly (DRGDSG) containing the RGD sequence has been studied in DMSO-d6 as well as in aqueous solution (H2O:D2O/90:10%) by 1H NMR spectroscopy. The unambiguous identification of spin systems of various amino acid residues and sequence specific assignment of all proton resonances was achieved by a combination of two dimensional COSY and NOESY experiments. The temperature coefficient data of the amide proton chemical shifts in conjunction with the vicinal coupling constants, i.e. 3JNH-C alpha H, NOESY and ROESY results indicate that the peptide in both the solvents exists in a blend of conformers with beta-sheet like extended backbone structure and folded conformations. The folded conformers do not appear to be stabilised by intramolecular hydrogen bonding. Our results are consistent with the flexibility of RGD segment observed in the NMR studies on the protein echistatin containing the RGD motif (references 23-25).

Conformational preference of Leu side chain in melanostatin in DMSO.

The solution conformation of melanostatin (Pro-Leu-Gly-NH2) in the neutral and protonated forms of DMSO has been monitored by one and two dimensional NMR techniques at 500 MHz. The temperature coefficients of the amide proton chemical shifts in conjunction with the observed NOESY spectra suggest that melanostatin in neutral form in DMSO adopts a backbone conformation such that leucine amide proton is buried by the proline ring and the side chain of leucine. Similar observation is made for protonated form of melanostatin in DMSO. The results of the present study are at variance with the earlier NMR studies which proposed a beta-turn structure for both the forms of melanostatin. There is, however, no evidence for the presence of beta-turn structure for both the forms of melanostatin in DMSO. In CDCl3 also Leu NH appears to be buried as evident from the solvent titration with DMSO and NOESY spectra.

Why do nucleic acids have 3'5' phosphodiester bonds?

Details of the stereochemistry of the 2'5' and 3'5' dinucleoside monophosphates of polynucleotides have been delineated in aqueous solution using nuclear magnetic resonance spectroscopy. Incorporation of these experimentally determined geometries into the structure of polynucleotides reveals that the intrinsic spatial configurations of the 2'5' bonds cannot support helical structures whereas the geometries of 3'5' bonds allow the formation of helical configurations for RNA.

Two gamma-bends in the backbone conformation of [D-Ala2]-leucine enkephalin in solution.

The solution conformation of [D-Ala2]-leucine enkephalin in its zwitterionic form in DMSO-d6 has been monitored by one- and two-dimensional proton magnetic resonance spectroscopy at 500 MHz. The resonances from the labile amide protons and the nonlabile protons have been assigned from the shift correlated spectroscopy. The chemical shift of the amide and C-alpha protons are found to vary with temperature but in opposite directions, except the C-alpha proton of the terminal tyrosine residue. This behavior has been explained by the shifting of equilibrium between the zwitterionic and neutral forms of the [D-Ala2]-leucine enkephalin and probably conformational changes accompanying temperature variation. The low values of the temperature coefficients of leucine and glycine amide protons indicate that these protons are either intramolecularly hydrogen bonded or solvent shielded. The observation of sequential cross peaks in the nuclear Overhauser effect spectra obtained at various mixing times, tau m (200-900 ms), indicate an extended backbone, which does not corroborate with the presence of a folded structure, i.e., beta-bend type structure. The estimate of interproton distances in conjunction with the low values of temperature coefficients of the leucine and glycine amide protons and vicinal coupling constants 3JHN-C alpha H have been rationalized by the predominance of two gamma-bends in the backbone conformation of [D-Ala2]-leucine enkephalin. The gamma-bend around the D-Ala residue has phi = 80 degrees and psi = 270 degrees, while the one around Phe it has phi = 285 degrees and psi = 90 degrees.

Side chain-backbone hydrogen bonding contributes to helix stability in peptides derived from an alpha-helical region of carboxypeptidase A.

Recently, Presta and Rose proposed that a necessary condition for helix formation is the presence of residues at the N- and C-termini (called NTBs and CTBs) whose side chains can form hydrogen bonds with the initial four amides and the last four carbonyls of the helix, which otherwise lack intrahelical hydrogen bonding partners. We have tested this hypothesis by conformational analysis by circular dichroism (CD) of a synthetic peptide corresponding to a region (171-188) of the protein carboxypeptidase A; in the protein, residues 174 to 186 are helical and are flanked by NTBs and CTBs. Since helix formation in this peptide may also be stabilized by electrostatic interactions, we have compared the helical content of the native peptide with that of several modified peptides designed to enable dissection of different contributions to helix stability. As expected, helix dipole interactions appear to contribute substantially, but we conclude that hydrogen bonding interactions as proposed by Presta and Rose also stabilize helix formation. To assist in comparison of different peptides, we have introduced two concentration-independent CD parameters which are sensitive probes of helix formation.

Spatial configuration of deoxyribotrinucleoside diphosphates in aqueous solution.

The detailed conformational features and dynamics of the naturally occurring deoxyribotrinucleoside diphosphates d-TpTpT and d-TpTpC have been investigated at 20 degrees C and 80 degrees C in aqueous solution by nuclear magnetic resonance spectroscopy. The observed NMR parameters indicate that the conformational properties of the trimers are very similar to those of the constituent dimers and monomers, i.e., the monomers and dimers conserve their intrinsic conformational features when they become incorporated into oligomers. Model building indicate that the distant shieldings can originate from spatial configurations in which the central nucleotidyl unit is bulged out and the w'1w1, w'2w2 occupy /g+g+, g+g+/ domains.

Stereodynamics of dimer segments of RNA in aqueous solution.

Arguments are presented which show that conformations II and III proposed by Lee and Tinoco [Lee, C.H., and Tinoco, I., Jr. (1977), Biochemistry 16, 5403] for ribodinucleoside monophosphates in aqueous solution are untenable. It has been shown that ribodinucleoside monophosphates exist in aqueous solution as an equilibrium blend of the classically recognized right-handed stack (g-g-), loop stack (g+g+), skewed (g+t), and extended arrays. In order to determine the effect of epsilonA base on the conformer distribution in the equilibrium blend, detailed ring-current calculations were performed and the isoshielding curves for epsilonA were derived. Use of these curves vis-a-vis dimerization shift data indicates that introduction of epsilonA perturbs the equilibrium blend which causes an increase in the population of skewed (g+t) arrays.

Left handed double helices: effect of sequence on the spatial configuration of high anti nucleic acids.

The conformational properties of purine-pyrimidine and pyrimidine-purine dinucleoside monophosphates in which the glycosidic torsion is fixed to congruent to 120 degree by the formation of a covalent link between the base and the sugar ring are explored by 1H NMR spectroscopy in order to obtain information about the spatial configuration of high anti nucleic acids. The intramolecular stack of the high anti dimers were found to be left handed, in contrast to that (right handed) for natural oligomers, which are low anti. Even though both the high anti pyrimidine-purine and purine-pyrimidine dimers have similar backbone torsion angles, they display widely different relative geometry between the bases; thus in the former there is extensive base-base overlap in the stack, and in the latter there is negligible intramolecular base-base overlap. In addition it was found that purine-pyrimidine systems form miniature double helices in which there is substantial interstrand purine-purine interaction; on the other hand the pyridine-purine high anti dinucleosides have no proclivity to form such base-paired complexes in solution. Mathematical polymerization of the conformation of the high anti purine-pyrimidine dinucleoside monophosphates generates a left handed helix for high anti polynucleotides. This also means that the double helix for high anti-nucleic acids containing purine-pyrimidine repeated units may also be left handed, as had been suggested [Sundaralingam, M., & Yathindra, N. (1977) Int. J. Quantum Chem., Quantum Biol. Symp. 4, 285]. It is suggested that the plasticity in the structure of genomic DNA is such that, if under certain conditions of interactions the sugar-base torsion of certain domains assume high anti values, that domain will become left handed, and this in turn can be a mechanism for the control of expression by genomic DNA.