Estimating the melting point, entropy of fusion, and enthalpy of fusion of organic compounds via SPARC.

The entropy of fusion, enthalpy of fusion, and melting point of organic compounds can be estimated through three models developed using the SPARC (SPARC Performs Automated Reasoning in Chemistry) platform. The entropy of fusion is modelled through a combination of interaction terms and physical descriptors. The enthalpy of fusion is modelled as a function of the entropy of fusion, boiling point, and flexibility of the molecule. The melting point model is the enthalpy of fusion divided by the entropy of fusion. These models were developed in part to improve SPARC's vapour pressure and solubility models. These models have been tested on 904 unique compounds. The entropy model has a RMS of 12.5 J mol(-1) K(-1). The enthalpy model has a RMS of 4.87 kJ mol(-1). The melting point model has a RMS of 54.4°C.

Development of Monopole Interaction Models for Ionic Compounds. Part I: Estimation of Aqueous Henry's Law Constants for Ions and Gas Phase pKa Values for Acidic Compounds.

The SPARC (SPARC Performs Automated Reasoning in Chemistry) physicochemical mechanistic models for neutral compounds have been extended to estimate Henry's Law Constant (HLC) for charged species by incorporating ionic electrostatic interaction models. Combinations of absolute aqueous pKa values, relative pKa values in the gas phase, and aqueous HLC for neutral compounds have been used to develop monopole interaction models that quantify the energy differences upon moving an ionic solute molecule from the gas phase to the liquid phase. Inter-molecular interaction energies were factored into mechanistic contributions of monopoles with polarizability, dipole, H-bonding, and resonance. The monopole ionic models were validated by a wide range of measured gas phase pKa data for 450 acidic compounds. The RMS deviation error and R(2) for the OH, SH, CO2 H, CH3 and NR2 acidic reaction centers (C) were 16.9 kcal/mol and 0.87, respectively. The calculated HLCs of ions were compared to the HLCs of 142 ions calculated by quantum mechanics. Effects of inter-molecular interaction of the monopoles with polarizability, dipole, H-bonding, and resonance on acidity of the solutes in the gas phase are discussed.

Calculating physical properties of organic compounds for environmental modeling from molecular structure.

Mathematical models for predicting the transport and fate of pollutants in the environment require reactivity parameter values - that is the value of the physical and chemical constants that govern reactivity. Although empirical structure-activity relationships have been developed that allow estimation of some constants, such relationships are generally valid only within limited families of chemicals. The computer program, SPARC, uses computational algorithms based on fundamental chemical structure theory to estimate a large number of chemical reactivity parameters and physical properties for a wide range of organic molecules strictly from molecular structure. Resonance models were developed and calibrated using measured light absorption spectra, whereas electrostatic interaction models were developed using measured ionization pK(a)s in water. Solvation models (i.e., dispersion, induction, H-bonding, etc.) have been developed using various measured physical properties data. At the present time, SPARC's physical property models can predict vapor pressure and heat of vaporization (as a function of temperature), boiling point (as a function of pressure), diffusion coefficient (as a function of pressure and temperature), activity coefficient, solubility, partition coefficient and chromatographic retention time as a function of solvent and temperature. This prediction capability crosses chemical family boundaries to cover a broad range of organic compounds.

Human pancreatic thread protein, an exocrine thread protein with possible implications to Alzheimer's disease: secondary structure in solution at acid pH.

The secondary structure of human pancreatic thread protein (HPTP) in solution at acid pH was derived using Fourier transform infrared (FT-IR) and laser Raman spectroscopic studies. The experimentally derived secondary structure of HPTP was compared with the secondary structure obtained by the Chou-Fasman algorithm. Pancreatic thread protein is a major exocrine secretory protein that in vitro forms filamentous bundles reminiscent of the paired helical filaments of Alzheimer's disease (AD). PTP immunoreactivity in brains afflicted with AD has been demonstrated previously and high levels of its mRNA in the developing human brain have also been reported in the literature. The above studies suggest that AD is associated with enhanced expression of PTP-related transcripts with interneuronal accumulation of PTP-like proteins. The experimentally derived secondary structure of HPTP consists of a significant proportion of beta-sheets and beta-turns and lesser amounts of alpha-helical structures. The beta-sheet component presumably plays an important role in the pH-dependent globule-fibril transformation of HPTP leading to antiparallel beta-sheet structure in the aggregated state. The secondary structure of HPTP and its globule-fibril transformation lend credence to the belief that AD may be viewed as a conformational disease.

Estimation of microscopic, zwitterionic ionization constants, isoelectric point and molecular speciation of organic compounds.

Mathematical models based on structure-activity relationships and perturbed molecular orbital theory have been developed to calculate the ionization pK(a)s for a large number of organic molecules. These models include resonance, direct and indirect electrostatic field effects, sigma induction, steric effects, differential solvation and hydrogen bonding. The thermodynamic microscopic ionization constants, pk(i), of molecules with multiple ionization sites and the corresponding complex speciation as a function of pH have been determined using these chemical reactivity models. For a molecule of interest SPARC (SPARC performs automated reasoning in chemistry) calculates all of the microscopic ionization constants and the fraction of each species as a function pH along with the titration (charge) curve. The system has been tested on several biologically and environmentally important compounds.

Non-uniform triple helical structure in chick skin type I collagen on thermal denaturation: Raman spectroscopic study.

The individual chains in the triple helix of collagen occur in a conformation related to polyproline II because of the presence of large number of imino peptide bonds. However, these residues are not evenly distributed in the collagen molecule which also contains many non-imino residues. These non-imino regions of collagen may be expected to show preference for other than triple helical conformations. The appearance of several Raman bands in solution phase at 65 degrees C raises the possibility of non-uniform triple helical structure in collagen. Raman spectroscopic studies on collagen in the solid state and in solution at a temperature greater than its denaturation temperature, reported here suggest that denatured collagen may exhibit an ensemble of conformational states with yet unknown implications to the biochemical interactions of this important protein component of connective tissues.

Mechanism of enolase: the crystal structure of asymmetric dimer enolase-2-phospho-D-glycerate/enolase-phosphoenolpyruvate at 2.0 A resolution.

Enolase, a glycolytic enzyme that catalyzes the dehydration of 2-phospho-d-glycerate (PGA) to form phosphoenolpyruvate (PEP), is a homodimer in all eukaryotes and many prokaryotes. Here, we report the crystal structure of a complex between yeast enolase and an equilibrium mixture of PGA and PEP. The structure has been refined using 29 854 reflections with an F/sigma(F) of >/=3 to an R of 0.137 with average deviations of bond lengths and bond angles from ideal values of 0.013 A and 3.1 degrees , respectively. In this structure, the dimer constitutes the crystallographic asymmetric unit. The two subunits are similar, and their superposition gives a rms distance between Calpha atoms of 0.91 A. The exceptions to this are the catalytic loop Val153-Phe169 where the atomic positions in the two subunits differ by up to 4 A and the loop Ser250-Gln277, which follows the catalytic loop Val153-Phe169. In the first subunit, the imidazole side chain of His159 is in contact with the phosphate group of the substrate/product molecule; in the other it is separated by water molecules. A series of hydrogen bonds leading to a neighboring enolase dimer can be identified as being responsible for ordering and stabilization of the conformationally different subunits in the crystal lattice. The electron density present in the active site suggests that in the active site with the direct ligand-His159 hydrogen bond PGA is predominantly bound while in the active site where water molecules separate His159 from the ligand the binding of PEP dominates. The structure indicates that the water molecule hydrating carbon-3 of PEP in the PEP --> PGA reaction is activated by the carboxylates of Glu168 and Glu211. The crystals are unique because they have resolved two intermediates on the opposite sides of the transition state.

Estimation of the ionization pK(a) of pharmaceutical substances using the computer program Sparc.

The computer program sparc was used to calculate the pK(a) values of some important pharmaceutical substances. The sparc models proved to be suitable for estimating the pK(a) values of beta-adrenergic blocking agents and benzodiazepine drugs. Ionization macroconstants, microconstants, zwitterionic equilibria, speciation curves as a function of the pH and the isoelectric points of a semi-essential amino acid, arginine, and an anti-inflammatory, niflumic acid, were calculated.

Fluorimetric determination of aminoglycoside antibiotics using lanthanide probe ion spectroscopy.

The application of probe ion fluorimetry has succeeded in the microdetermination of six aminoglycoside antibiotics: neomycin, streptomycin, gentamicin, tobramycin, amikacin and kanamycin as sulfate salts in pure form and in some pharmaceutical preparations. The method is based on the reaction of Eu(3+) ions with aminoglycosides through amino and hydroxy groups. Such interactions enhance the intensity of the 616 nm fluorescence emission of the Eu(3+) ion. The fluorescence at 592 nm comes from a non-hypersensitive transition and is not affected by the ligand which is bound to the probe ions. The intensity ratio R, defined as I (592)I (616) was used to determine the amount of free and bound europium ions. A linear relationship between bound europium ions and aminoglycoside was found within the concentration ranges 20-100 ppm for neomycin, 5-60 ppm for streptomycin, and 10-70 ppm for gentamicin, tobramycin, amikacin, and kanamycin as sulfate salts. The percentage recoveries ranged from 99.22 to 101.07, with standard deviations ranging from +/- 1.5 to +/- 4.38. The relative stability constants ranged from 5 x 10(3) to 2 x 10(4). The optimum reaction conditions were studied and the results obtained compared favourably with the fluorimetric method using fluorescmine reagent.

Europium(III) ion probe spectrofluorometric determination of diclofenac sodium.

A new method has been devised for the determination of diclofenac sodium in bulk and in pharmaceutical preparations using Eu3+ ions as the Fluorescent probe. The technique was built around the hypersensitive property of the transitions of the fluorescent probe ion, Eu3+, at 616 nm. This is normally a forbidden transition, but the interaction with diclofenac sodium, which contains a carboxylic group, makes the transition allowed and enhances the intensity of its fluorescence emission. The Eu3+ fluorescence emission at 592 nm comes from a non-hypersensitive transition and is not affected by ligation. The intensity ratio, R, defined as I592/I616, was used as a measure of the percentage of bound probe ions. Diclofenac and Eu(III) forms a (1:1) molar complex. The relative stability constant of the complex was found to be 10(5). A linear relationship between bound Eu3+ and the concentration of diclofenac sodium was found for concentrations from 10 to 200 micrograms ml-1, with a recovery percentage of 100.22 +/- 2.27. The method shows a good agreement with a spectrophotometric method.

Calculated effects of the chemical environment of 2-phospho-D-glycerate on the pKa of its carbon-2 and correlations with the proposed mechanism of action of enolase.

The mechanism of action of enolase requires drastic lowering of the pK of the alpha carbon of the substrate to enable removal of the alpha hydrogen. Calculations suggest this is done primarily by rotation of the carboxyl ("carbon-1") of the substrate around the carbon-1 carbon-2 bond by 90 degrees so that its oxygen atoms lie in the plane formed by the three carbon atoms and secondarily by effectively neutralizing the negative charges on the carboxylate and phosphate. These predictions are in very good agreement with the proposed mechanism of action of enolase which is based on X-ray crystallographic data.

Synthesis and biological properties of 4-norleucine-neuropeptide Y; secondary structure of neuropeptide Y.

Neuropeptide Y (NPY) is a 36 amino acid peptide amide isolated from porcine brain. The NPY analog, 4-norleucine-NPY was synthesized by a solid-phase method and purified to homogeneity in 20% yield by reverse-phase chromatography. Investigation of the biological properties indicated that the analog is an agonist of NPY. Secondary structural analyses revealed that NPY and the analog exhibited predominantly alpha-helical and beta-sheet structures, respectively; however, experiments in trifluoroethanol indicated that the analog has the potential of assuming an alpha-helical structure. Based on circular dichroism (CD), Raman spectroscopy and Chou-Fasman analyses, a model has been proposed for the secondary structure of NPY.

Conformational features of dynorphin A-(1-13), Laser Raman spectroscopic studies.

Conformational features of dynorphin A-(1-13) were examined by laser Raman spectroscopy. Dynorphin A-(1-13) appears to have a mixture of extended beta-pleated sheet and "random" structure.

Conformational states of Leu5- and Met5-enkephalins in solution.

The molecular conformations of Leu5- and Met5-enkephalins in aqueous and DMSO solutions were investigated by FT-IR and laser Raman spectroscopic methods. The amide I, II, and III regions in the FT-IR spectra of Leu5- and Met5-enkephalins in aqueous solution were analyzed by performing Fourier self-deconvolution of the bands. Leu5-enkephalin in aqueous solution is found to exist in both type II beta-turn and beta-sheet structures, whereas Met5-enkephalin has a lesser tendency to form beta-turn structure in aqueous solution. It is likely that these different conformers of enkephalins might bind to different receptor types.

Studies of activating and nonactivating metal ion binding to yeast enolase.

Measurements of binding of certain divalent cations to yeast apoenolase were made using a pH-meter, chromatography, a divalent cation electrode, and ultrafiltration. The binding of the activating metal ions Mg2+ and Co2+ and the nonactivator Ca2+ were studied as functions of the presence or absence of substrate/product, phosphate, and fluoride or level of Tb3+. The data suggest phosphate and fluoride increase Mg2+ binding but not Ca2+ binding. Substrate/product appears to increase Ca2+ binding as well as that of Mg2+ and Co2+. In the presence of substrate, Co2+ binding was 5-6 mol/mol dimer. In the absence of substrate/product, Tb3+ reduced Co2+ binding from 4 mol/mol to 2. These data are interpreted in terms of binding to "conformational," "catalytic" (substrate/product dependent), and "inhibitory" sites. Measurements of Tb3+ fluorescence quenching by Co2+ suggested that the distance between "conformational" sites on the two subunits was large, while the distance between "conformational" and "inhibitory" sites was ca. 17 +/- 4 A. Potentiometric titrations of apoenzyme with Ca2+ and Mg2+ showed that the metal ions produced the same proton release in the presence or absence of substrate/product. If phosphate and fluoride were present, then more protons were released if Ca2+ was the titrant rather than Mg2+, suggesting a difference in ionization state in the complex with the activating metal. Electron paramagnetic resonance studies of Co2+ binding to the various sites in the enzyme are presented. The Co2+ bound to all three sites appears to be high spin, consistent with a preponderance of oxyligands in an octahedral environment. Substrate, citrate, and a strongly binding substrate analogue strongly enhance the hyperfine structure of conformational Co2+. This is interpreted as the result of a change in interaction of an axial ligand to conformational Co2+ produced by carbon-3 of substrate or analogue.

Raman spectra of flavin bound in flavodoxins and in other flavoproteins. Evidence for structural variations in the flavin-binding region.

The resonance coherent anti-Stokes Raman scattering (CARS) spectra for a number of flavoproteins are found to be fingerprints for the particular type of flavoprotein. One group studied were the bacterial flavodoxins: Desulfovibrio vulgaris, Desulfovibrio desulfuricans, Azotobacter vinelandii, Megasphaera elsdenii, Clostridium kluyverii and Clostridium formicoaceticum. The other examples were the enzymes lactate monooxygenase and glucose oxidase. FMN complexed to Vibrio harveyi luciferase, and a partially characterized non-fluorescent flavoprotein from Photobacterium leiognathi. In the frequency range 1700-1100 cm-1, differences in the frequency positions and relative intensities of the prominent bands are reflections of the interactions of the isoalloxazine ring with the protein. Based on tentative assignment of the vibrational modes in flavin models, the spectra are interpreted in terms of hydrogen bonding between the amino acid residues of the binding site and particular atoms of the isoalloxazine ring.

Effects of growth substrate and respiratory chain composition on bioenergetics in Acinetobacter sp. strain HO1-N.

The relationship between respiratory chain composition and efficiency of coupling phosphorylation to electron transport was examined in Acinetobacter sp. strain HO1-N. Cells containing only cytochrome o as a terminal oxidase displayed the same stoichiometries of adenosine 5'-triphosphate synthesis and proton extrusion as cells which contained both cytochromes o and d as terminal oxidases. In addition, CO inhibition and photo-relief of cytochromes o or d did not alter the efficiency of energy coupling. These findings indicate that adenosine 5'-triphosphate synthesis is coupled to electron transport through both cytochromes o and d in Acinetobacter.

Binding of terbium (III) to yeast enolase.

Several independent criteria indicate 2 mol of terbium (III) bind to yeast enolase in the absence of substrate-fluorescence titrations of enzyme and metal, effects on thermal stability and published ultrafiltration and inhibition experiments. These measurements also suggest the terbium binding sites are the same as those normally occupied by "conformational" magnesium. Terbium binds much more strongly than magnesium, however, and measurements of the kinetics of the absorbance change in the terbium-enzyme on adding excess EDTA suggest the terbium-enzyme dissociation constant is about 1/500 that of the magnesium-enzyme. Measurements of enzyme activity as a function of substrate concentration show that terbium permits no enzymatic activity. However, magnesium competes more effectively with the lanthanide if the substrate analogue 3-aminoenolpyruvate 2-phosphate (AEP) is present. The fluorescence of the lanthanide is not readily observed on exciting the terbium-enzyme at 280 nm, indicating the absence of tyrosines or tryptophans in the coordination sphere of the metal. Excitation of terbium using 488 nm radiation from an argon ion laser shows the fluorescence of the metal is enhanced by binding to the enzyme. EDTA and carbonate have similar effects. This suggests carboxyl groups are involved in binding metal at the conformational sites of yeast enolase. Measurements of lifetimes of enzyme-bound terbium in the presence and absence of D2O indicated three moles of water remained on each of the bound metals, independently of the buffer used. If enzyme-bound terbium is assumed to be nine-coordinate, the metal must bind to six groups from the enzyme. The presence of substrate does not markedly affect the emission spectrum of the bound terbium or the number of water molecules remaining on the metal, but calorimetric measurements show that substrate binds to the terbium enzyme.

Protein-ligand interactions in lumazine protein and in Desulfovibrio flavodoxins from resonance coherent anti-Stokes Raman spectra.

The resonance coherent anti-Stokes Raman technique was used to obtain vibrational spectra of flavin in flavodoxins from Desulfovibrio gigas and Desulfovibrio vulgaris and of the simpler 6,7-dimethyl-8-ribityllumazine chromophore in the blue fluorescence lumazine protein from the bioluminescent bacterium Photobacterium phosphoreum. In the region examined, 1100-1700 cm-1, the Raman spectrum of the lumazine is less crowded than that of the flavin and this facilitates assignment of observed frequencies to particular vibrational modes. Certain modes are not affected by chromophore binding to the protein, but others are changed in frequency or intensity in a way that can be rationalized by expected interactions of the chromophore with the amino acid residues of the binding site. For example, a tentative assignment of change in hydrogen bonding at N(5) is suggested as the cause of the frequency shift for the chromophore in both flavodoxins (free-bound, 1582-1572 cm-1) and for C(4)=O in glucose oxidase (1359-1364 cm-1) and lumazine protein (1359-1362 cm-1). Shifts of the C(2)-N(3) mode in D2O may arise from hydrogen-bonding changes at C(2)=O in lumazine protein (1284-1291 cm-1), flavodoxin (1300-1280 cm-1), and glucose oxidase (1297-1287 cm-1). Bonding at N(3)-H may be the origin of changes in the frequency or intensity of the amide III mode in riboflavin binding protein and glucose oxidase. A stacking interaction is suggested for the change in a pyrimidine ring mode in FAD (1508 cm-1) since this mode is found at 1504 cm-1 in 30% Me2SO/H2O, where the adenine and pyrimidine are unstacked. The results clearly indicate different interactions in the binding sites of those proteins studied to date.