Ionic hydrogen bonds are stronger than neutral hydrogen bonds. Ionic hydrogen bonds, with carbon as the donor, are investigated here at the ab initio level. The hydrogen bond energy for systems sp3 carbon as donor  is 9-15 kcal mol^-1 and the C...X distance is 2.75 A, whereas the hydrogen bonds involving sp2 and sp carbons are much stronger and shorter. Biological systems, such as protonated histidine and acetylcholine have been studied. The C-H..O hydrogen bond interaction in zwitterionic glycine was found to be substantial. In addition electrostatic potential studies on the donor systems were done to allow qualitative comparison of hydrogen bond energies.

Return to List Of Publications

The electrostatic potential of valinomycin in various conformations as obtained be the crystal structures (uncomplexed, complexed) and theoretical considerations have been evaluated and compared. The potential energy profiles along the Z-axis of the bracelet-like structures show a systematic variation from the uncomplexed to the complexed structure. This type of conformational change and the potential variation are probably associated with different states of ion transport, like the capture and release of ions by the ionophore. Also, the asymmetry of the molecule due to D-HyIV on one side  and L-Lac on the other side is reflected in the potential values along the z-axis, the magnitude of which, is considerable in the uncomplexed structure. The evaluation of the potential at the ab initio level on smaller fragments indicate that the order of liganding capacity of oxygen is amide > ether > ester. Also, the inductive effect due to alkyl substitution is negligible as evidenced by the potential studies on the substituted amides and esters.

Return to List Of Publications

The mechanism of ion transport by carrier ionophores is investigated. The electrostatic potential is used as index of the binding energy of a cation with valinomycin and enniatin B. The ion binding capacities of these ionophores is studied as functions of conformation and of distance of an approaching ion-complex. The energetics of dimerization and the binding energy profile of an ion in dimers of valinomycin and enniatin B are examined. The binding energy profiles and te electrostatic potential surfaces of valinomycin and enniatin B are compared in relation to their biological activities.

Return to List Of Publications

The chemical groups which take part in the proton transfer reaction in bacteriorhodopsin have been studied by ab initio quantum chemical methods. The various factors such as conjugation with a linear system, electron delocalization of the guanidine type, cis-trans isomerism, geometry distortion and hydrogen bonding with charged groups can influence the properties of a given chemical group. Several systems are studied at 4-31G and STO-3G levels. Some of the Schiff-base analogues and guanidine type molecules are characterized by their molecular orbital diagrams, energy levels and the nature of charge distribution. Also, the effects of the above-mentioned factors on proton affinity are studied. It is hoped that the values thus obtained can be helpful in evaluating various structural models for proton transfer.

Return to List Of Publications

Proton transfer across cationic hydrogen bonds involving Schiff-base, ammonia and related compounds are studied at 4-31G level. Proton transfer characteristics are correlated to the proton affinities of the species involved. Hydrogen bond strengths of these hydrogen bonds are correlated to the difference in the proton affinity of the donor and the acceptor. Influence of a neighboring hydrogen bond on the proton transfer from Schiff-base to ammonia and Schiff-base to water are also studied.

Return to List Of Publications

Return to List Of Publications

Return to List Of Publications

A self-consistent procedure for estimating the secondary structure content from circular dichroism spectra of proteins is presented. In this method the spectrum of the protein to be analyzed is included in the basis set and an initial guess is made for the unknown structure as a first approximation. The resulting matrix equation is solved using the singular value decomposition algorithm and the initial guess is replaced by the solution. The process is repeated until self-consistency is attained. The best features of the variable selection and the locally linearized methods are incorporated in this procedure. We have applied this method to examine the inconsistencies in the CD data, to compare the predictions with different ranges and resolutions of the CD data, and to compare different assignments of secondary structures from x-ray structure analyses in the context of secondary structure predictions. The results are compared using the root mean square differences and correlation coefficients. The results obtained are as good as or better than the previous analyses. For most of the proteins considered the self-consistent solutions obtained with different initial guesses were similar. We find the Kabsch and Sander protein crystal structure analysis to be most suitable for our prediction method.

SELCON program and the related data files are available.

Return to List Of Publications

Different possible mechanism for generation of optical activity of hemoproteins in the Soret region are considered. The heme undecapeptide of cytochrome c does not contain aromatic amino acid residues, so its considerable optical activity cannot be due to coupling of heme pp* transitions with those of aromatic residues. CD data for the heme undecapaptide and for ferromyoglobin and some of their complexes with small molecules are presented and critically compared. Symmetrically coordinated imidazole complexes show rotational strengths of the same magnitude as those of corresponding nonsymmetrically coordinated compounds. Inherent chirality in the bond heme is inferred to be a significant source of optical activity in the heme undecapeptide. Theoretical calculations based upon a molecular dynamics simulation support this proposal. Coupled oscillator interactions with the peptide pp* transitions and with the high energy transitions in the peptide groups and their thioehther sulfers, as modeled by polarizabilities, also make significant contributions. These same mechanisms must also be considered in hemoproteins in general.

Return to List Of Publications

It has been demonstrated that the amino acids Asp537, Asp812, Lys631, His811 and Tyr639 are involved in bacteriophage T7 RNS polymerase catalysis. In the present paper, we report kinetic, spectroscopic and calorimetric characterization of the wild-type and mutant T7 RNA polymerases generated at these five loci (D537N, E; K631M, R; Y639F, S, A, W; H811Q, A; D812N, E). The wild-type enzyme has a substantial amount of secondary structure as determined by CD analysis (a-helix, 43%; b-sheet, 14%; b-turn, 25%; unordered, 18%). The CD spectra of 12 mutants at five loci are very similar to that of the wild-type, except for the mutant Y639W. Within experimental error, the thermal transition temperatures measured by CD and DSC aswell as the lmax values of the fluorescence spectra were the same for the wild-type and all of the mutants. Therefore, the overall folding and stability of the mutant enzymes are very similar to those of the wild-type enzyme, although small local conformations cannot be excluded. For the synthesis of the pentamer pppGGACU, the mutants D537E and D812E showed an approximately two- to threefold decrease in (k_cat)_app and an approximately two-to threefold increase in (K_m)_app, relative to the wild-type, in contrast to mutants D537N and D812N which exhibited no detectable activity. The mutant K631R showed a sevenfold reduction in (k_cat)_app and a two- to threefold increase in  (K_m)_app, supporting our earlier observation with the mutant K631M that Lys631 may be involved in phosphodiester bond formation. The mutant Y639S can synthesize the trimer GGA with an approximately 50-fold decrease in(k_cat)_app and a tenfold increase in  (K_m)_app, relative to the wild-type, underlining the importance of the phenyl ring of Tyr639. The mutant H811A, in which the side-chain at position 811 is incapable of forming a hydrogen bond, can synthesize the trimer GGA with an approximately tenfold decrease in (k_cat)_app and approximately 35-fold increse in  (K_m)_app. Thus, either the hydrogen-bonding capacity of this residue is non-essential or some other group can functionally substitute for the His811 side-chain. The wild-type enzyme showed significant effects of the base position in the sequence on the apparent binding constants for the NTPs. The kinetics of GpG-primed trimer, tetramer and pentamer synthesis on three 22 bp templates were investigated for the wild-type and mutant enzymes with measurable activity. The rate of oligonucleotide synthesis depends strongly on the message sequence, and abortive termination occurs more frequently following UMP incorporation. For a given template, the rate of synthesis by the wild-type and all the mutants is trimer > pentamer (run-off transcript) > tetramer. The effect of the 5'-triphosphate in the initiating nucleotide is minimal for the wild-type and mutant enzymes. The apparent pK values obtained from the high-pH side of the pH-activity profiles of the wild-type and the mutants K631R and Y639F are the same within experimental errors, implying that an amino acid(s) other that Lys631 or Tyr639 is responsible for the high-pH side of the pH-activity profile.

Return to List Of Publications

Different approaches to improve the analysis of protein secondary structure from circular dichroism spectra are compared. Grouping proteins based on the similarity of their circular dichroism spectra, using cluster analysis methods, was utilized as a new way of implementing variable selection. The performance of three basic methods (neural networks, ridge regression and singular value decomposition) was evaluated in combination with three approaches to improve the predictions, viz., variable selection, cluster analysis and the self-consistent method. Cluster analysis performed on the basis set proteins resulted in three clusters, subanalyses of which provide a new way of performing variable selection. The neural network with two hidden layers performed better than that with one hidden layer and was combined with variable selection. Inclusion of the variable selection principle improved the performance of all three basic methods. While the neural network method performed slightly better than the other two methods at the basic level, the inclusion of variable selection led to similar performance indices for all three methods.

Return to List Of Publications

A method to identify poly-L-proline type (P_II) conformation in crystal structures of globular proteins is presented. Short segments of PII structure were identified in globular protein structures, and these form a significant fraction of the residues which are not assigned to a-helix, b-sheet and turns. The fractions of a-helix, b-sheet, turns, P_II and the remainder, identified in conjunction with the Kabsch and Sander method (Biopolymers, 22, 2577, (1983)), were incorporated in the analysis of CD spectra of proteins. The separation of P_II fraction from the otherwise unassigned fraction resulted in a distinctive P_II CD spectrum and an unusual CD spectrum corresponding to the unassigned structures. The quality of prediction of P_II fraction from CD spectra of proteins was comparable to that of b-sheet and turns.

Return to List Of Publications

The effects of the crystal environment on the electronic spectral parameters in 9-methyladenine and N6-methyladenine have been investigated. We have included the electrostatic effects of the crystal environment, a probable source of discrepancy between the experiment and theory, in the semi-empirical molecular orbital calculations using the INDO/S method. The fields and potentials at atomic centers of the molecules in the crystal were calculated using the ground-state charge distribution and were included in the INDO/S Hamiltonian in an iterative process, until self-consistency was attained. The crystal field polarizes the ground state leading to an increase in the net atomic charges, significantly increasing the magnitude of the ground-state dipole moment in N6-methyladenine, with the direction unaltered. In 9-methyladenine, a significant change was predicted in the direction of the ground-state dipole moment, with only a slight increase in the magnitude. The changes in the ground-state dipole moment depend on the relative directions of the crystal field and the gas phase dipole moment. The predicted gas phase spectra of both molecules are comparable due to the small effects of methyl substitution on the electronic structure. The crystal field introduces mixing of np* and pp* transitions. This leads to a slight red shift in the energy of the transitions, changes in intensities, and rotation of the transition dipole moment directions. The interactions between the excited-states in the crystal were evaluated by a perturbation treatment. Excited-state mixing leads to extensive transfer of intensities and a slight blue shift in the energy of the transitions.  The predicted transition moment directions are in general agreement with experiment, although relative intensities differ in some cases, notably in the two lowest energy transitions.

Return to List Of Publications

By use of an asymmetric Ullmann coupling involving chiral naphthalene oxazolines 1, the title compounds were prepared in good yields and high diastereoselectivity. Hydrolysis of the binaphthanyl oxazolines 2 led to the di-aldehydes 5, which were transformed into the azepine derivative, 6. The latter was treated with appropriate phosphoryl halide to access the chiral HMPA systems 7 and 9. The CD spectra of the chiral azepine 6 and the chiral phosphoramides 7 and 9 were measured and showed strong positive CD couplet near 225 nm, consistent with the P axial chirality (S configuration). Semi-empirical CNDO/S molecular orbital calculations of the CD spectrum of 6 satisfactorily reproduced the major fratures of the spectrum.

Return to List Of Publications

A core Y61F mutant of the 87-residue gene 5 single-stranded DNA-binding protein (g5p) of  fd bacterial virus was found by circular dichroism (CD) titrations to have the same binding modes as does wild-type g5p. Taken together with published CD and fluorescence data, our results indicate that these modes are n = 4 and n = 2.5*3 for binding to poly[d(A)], but n = 4, n = 3, and n = 2-2.5 for binding of wild type g5p to fd single-stranded viral DNA (fd ssDNA) at low ionic strength. Binding  of g5p in the terminal n = 2.5-3 or n = 2-2.5 modes involves a partial release of the polynucleotide from the g5p, indicated by a partial reversal of the g5p-induced perturbation of the nucleic acid CD. The only changes observed in the properties of the Y61F g5p were a slight reduction in binding affinity of the n = 4 mode at 0.1-0.2 M NaCl, and an increased tendency of  Y61F complexes to aggregate. Electron microscopy of Y61F g5p complexes in each of the binding modes showed structures indistinguishable from those of the corresponding complexes with wild type g5p. For complexes with the 6408-nucleotide fd ssDNA, progression from binding in the n = 4 to the  n = 3 to the n = 2-2.5 mode was accompanied by an increase in the mean number of helical turns, from (104 to 113 to 123) ± 3, and an increase from (7.7 to 9.5 to 10.4) ± 0.3 g5p dimers per turn, similar to changes in complexes with the wild type g5p. The changes in binding mode thus involve a change in the arrangement of g5p dimers and nucleotides in each helical turn, and hence variations in the g5p dimer-dimer interaction. The 210-240 nm protein CD of a Y34F mutant was unchanged upon binding poly[d(A)], implying that the g5p secondary structure was not significantly altered. Enhanced-resolution protein CD difference spectra comparing wild type and Tyr*Phe replacement mutants for 4 of the 5 tyrosines of the g5p showed that the replacement Phe-61 and Phe-34 residues are relatively immobilized, and that the fifth tyrosine, Tyr-56, probably contributes a strong positive contribution to the protein CD in the near-UV region and is the strongest contributor to the positive 230 nm band. CD calculations on this relatively simple and well-defined system gave poor results for the absolute spectra, but the results for difference spectra in the near- and far-UV were quantitatively correct for several of the mutants. The best results were obtained for Y26F and Y34F mutants. The predictions for Y56F g5p, which has not been studied, are in accord with the CD properties implied by those of the wild-type and the other four mutants studied here. Results for Y61F g5p are less satisfactory and those for Y41F g5p clearly disagree with experiment. These results are discussed in light of the mobility and solvent accessibility of the mutated side-chains.

Return to List Of Publications

A simple approach to estimate the number of a-helical and b-strand segments from protein circular dichroism spectrum is described. The a-helix and b-sheet conformations in globular protein structures, assigned by DSSP and STRIDE algorithms, were divided into regular and distorted fractions by considering a certain number of terminal residues in a given a-helix or b-strand segment to be distorted. The resulting secondary structure fractions for 29 reference proteins were used in the analyses of circular dichroism spectra by the SELCON method. From the performance indices of the analyses we determined that on an average four residues per a-helix and two residues per b-strand may be considered distorted in proteins. The number of a-helical and b-strand segments and their average length in a given protein were estimated from the fraction of distorted a-helix and b-strand conformations determined from the analysis of circular dichroism spectra. The statistical test for the reference protein set shows the high reliability of such a classification of protein secondary structure. The method was used to analyze the circular dichroism spectra of four additional proteins and the predicted structural characteristics agree with the crystal structure data.

SELCON3 Program determines the number of these segments.

Return to List Of Publications

ABSTRACT17
"Calculation of the circular dichroism spectra of nucleic acids." N. Sreerama and R.W. Woody. (in preparation)
 

Circular dichroism spectra of polynucleotides were calculated using the matrix method.  The electronic transition parameters (wavelength, intensity and transition dipole moment direction) of individual bases were from the polarized reflection studies of single crystals of the purine and pyrimidine bases by Clark. CD spectra were computed for different sequences of polynucleotides in A, B and Z conformations and were compared with the experimental CD data. CD spectra were also computed for  three oligonucleotides with known solution structure (Dr. A.N. Lane, MRC Labs, London) and were compared with the experimental CD spectra (Dr. S.R. Martin, MRC Labs, London). The computed spectra reproduced most major features in the spectra, but some of the observed bands in the 260-285 nm range were not predicted or were shifted significantly in wavelength. The np* transitions and environmental effects, which were neglected in these calculations, are expected to influence the long-wavelength region of the spectrum. Calculations based upon transition parameters from semi-empirical MO theory, as modified by the electrostatic fields, have thus far given results substantially poorer than those obtained from the empirical transition parameters. Efforts are under way to improve these calculations.

Return to List Of Publications

Improved descriptions of the lowest energy excited states of tyrosine and phenylalanine side chains have been developed in order to extend the capabilities of calculating the CD spectra of proteins. Four transitions (Lb, La, Bb and Ba) for each of the side chain chromophores were considered, and the transition monopoles were obtained from a CNDO/S calculation on models representing the individual groups. Monopoles at midpoints of the bonds, corresponding to the maximum transition charge densities in the Lb band, and monopoles representing the vibronic coupling with the B transitions for the La transition were also included. The aromatic transitions were combined with the peptide transitions (np*, pp*, n’p*, and p+p*) in the framework of the origin-independent matrix method to compute the CD spectra of different crystal forms and Y??L and F??L mutant BPTI. The structures of the mutants were obtained by replacing the appropriate tyrosine or phenylalanine residue by leucine in the wild type crystal structure, and energy minimizing the structure. The CD spectrum calculated for the form II crystal structure of BPTI showed the best agreement with experiment. In the far UV, the calculated and experimental CD spectra agree to various extents for the wild-type and mutant BPTI. Among the mutants, Y10L shows the best agreement while Y21L and F22L, the two residues interacting with most aromatic groups, show the worst agreement. In the near UV, the negative bands predicted for the wild type and mutant BPTI have much less intensity than the experiment.

Return to List Of Publications

Strutural changes in T7 RNA polymerase induced by temperature and urea have been studied in an extensive range of solvent conditions to obtain information about the stability and the molecular organization of the protein. Calorimetric studies of the conformational transitions in T7 RNAP induced by temperature show that the enzyme consists of three compact cooperative units that can be regarded as structural domains. Interaction between structural domains and their stability strongly depend on solvent conditions. The unfolding of T7 RNAP under different solvent conditions may induce a stable intermediate state that contains residual secondary structure. CD measurements show that thermal unfolding in the absence of urea leads to a state in which the fractions of b-sheet and unordered conformation are enhanced at the expense of a-helix. Isothermal unfolding by urea at 25^oC reveals a two-step unfolding process. A transition centered near 3M urea leads to a plateau from about 4-5M urea, followed by a second transition near 6.5M urea. ......

Return to List Of Publications

A significant fraction of the so-called “random coil” residues in globular proteins exists in the left-handed poly(Pro)II conformation. In order to compare the behavior of this secondary structure with that of the other regular secondary structures, molecular dynamics simulations, with the GROMOS suite of programs, of an alanine octapeptide in water, in a-helix, b-strand and left-handed poly(Pro)II conformations, have been performed. Our results indicate a limited flexibility for the a-helix conformation and a relatively larger flexibility for the b-strand and poly(Pro)II conformations.  The behavior of oligopeptides with a starting configuration of b-strand and poly(Pro)II conformations, both lacking interchain hydrogen bonds, were similar. The (f, y) angles reflect a continuum of structures including both b and PII conformations, but with a preference for local PII regions. Differences in the network of water molecules involved in hydrogen bonding with the backbone of the polypeptide were observed in local regions of b and PII conformations. Such water bridges help stabilize the PII conformation relative to the b conformation.

Return to List Of Publications

Circular dichroism spectra for 23 proteins have been calculated using transition parameters from experiments on model amides and semi-empirical MO calculations. The results are substantially better than those reported by Hirst. The improvements result primarily from using distributed dipoles (monopoles), rather than the point dipoles used by Hirst.

Return to List Of Publications

We have expanded our reference set of proteins for the estimation of protein secondary structure from CD spectra, from 29 proteins to 37 proteins, including CD spectra of three globular proteins with known x-ray structure and five denatured proteins. The secondary structure corresponding to the denatured CD spectra were approximated to be 90% unordered. We have also modified the self-consistent method for analyzing protein CD spectra, SELCON3, by including a new selection criterion developed by W.C. Johnson Jr. The expanded reference set, examined with SELCON3, gave slight improvements in the performance indices for practically all types of secondary structures. We examined the thermal denaturation of  ribonuclease T1 by CD using both the original and expanded reference sets, and obtained better results with the expanded set. This can be interpreted as evidence for spectral equivalence of random coil structure of denatured proteins and unordered structure of the native proteins in the reference set. An important application of this work will be for analyzing CD spectra of proteins with significant unordered structure content and/or in the course of denaturation.