Effect of chirality on PVP/drug interaction within binary physical mixtures of ibuprofen, ketoprofen, and naproxen: a DSC study

We report on the thermal behavior of freshly prepared binary drug/polymer physical mixtures that contained ibuprofen, ketoprofen, or naproxen as a drug, and polyvinylpyrrolidone (PVP), hydroxyethylcellulose (HEC), or methylcellulose (MC) as excipient. At 6-10 degrees C/min heating rates the DSC detected a sharp, single endotherm that corresponds to the melting of drug. On heating physical mixtures of PVP and racemic ibuprofen or ketoprofen at lower heating rates, another endotherm was registered in front of the original one. To observe the additional endotherm, specific minimal values of the heating rate and of PVP weight fraction were needed; for ibuprofen and ketoprofen they were 1.5 and 2.0 degrees C/min, and 5 and 15% (w/w), respectively. At greater PVP weight fractions the top temperatures, T(mp), of both peaks were reduced almost linearly indicating strong solid-state interfacial reaction between the drug particles and PVP matrix. The additional endotherm was abolished at greater heating rates (2 degrees C/min for ibuprofen, 3 degrees C/min for ketoprofen), by replacing the racemate with respective S+-enantiomer and by replacing PVP with HEC and MC. Hence, the possible inclusion of enantioselective component within the PVP/drug interaction, responsible for the amorphization of physical mixture over storage, is assumed.     

Visualizing the lipid dynamics role in infrared neural stimulation using stimulated Raman scattering

Infrared neural stimulation (INS) uses pulsed infrared light to yield label-free neural stimulation with broad experimental and translational utility. Despite its robust demonstration, INS’s mechanistic and biophysical underpinnings have been the subject of debate for more than a decade. The role of lipid membrane thermodynamics appears to play an important role in how fast IR-mediated heating nonspecifically drives action potential generation. Direct observation of lipid membrane dynamics during INS remains to be shown in a live neural model system. We used hyperspectral stimulated Raman scattering microscopy to study biochemical signatures of high-speed vibrational dynamics underlying INS in a live neural cell culture model. The findings suggest that lipid bilayer structural changes occur during INS in vitro in NG108-15 neuroglioma cells. Lipid-specific signatures of cell stimulated Raman scattering spectra varied with stimulation energy and radiation exposure. The spectroscopic observations agree with high-speed ratiometric fluorescence imaging of a conventional lipophilic membrane structure reporter, 4-(2-(6-(dibutylamino)-2-naphthalenyl)ethenyl)-1-(3-sulfopropyl)pyridinium hydroxide. The findings support the hypothesis that INS causes changes in the lipid membrane of neural cells by changing the lipid membrane packing order. This work highlights the potential of hyperspectral stimulated Raman scattering as a method to safely study biophysical and biochemical dynamics in live cells.     

Influence of environment on piroxicam polymorphism: vibrational spectroscopic study

FTIR and FT-Raman spectroscopies were used to evaluate the mechanism of transformation of piroxicam into its different forms (alpha, beta, and monohydrate), depending on the environment. These vibrational techniques allowed us to identify the forms of piroxicam that crystallize from different solvents at different cooling rates and the conformation of the drug in some of its derivatives: piroxicam hydrochloride, piroxicam thallium and sodium salt hemihydrates, and piroxicam sodium salt. The usefulness of Raman spectroscopy in characterizing piroxicam:beta-cyclodextrin (PbetaCD) inclusion compounds was described. The Raman spectrum of 1:2 PbetaCD was discussed in comparison with that of the corresponding piroxicam sodium salt containing inclusion compound (1:2 PNabetaCD) in order to study the influence of the piroxicam derivative used on the structure of the inclusion compound. The Raman results showed that in both of the inclusion compounds the piroxicam mainly assumes the zwitterionic structure typical of a monohydrate; therefore, the kind of derivative used does not affect the conformation of the drug in its inclusion compound. The effect of the method of synthesis utilized (freeze-drying or freeze-thaw cycling) to obtain 1:2.5 PbetaCD was investigated. The inclusion compound obtained by freeze-thaw cycling proved to be more crystalline and to contain a higher amount of the beta form than the freeze-dried inclusion compound. Raman spectroscopy proved to be a useful technique for evaluating the effectiveness of the manufacturing process in relation to the pharmaceutical properties of the drug and to the nondestructive and noninvasive on-line quality control of the industrial products.     

Encapsulation of Ketoprofen and Ketoprofen Lysinate by Prilling for Controlled Drug Release

In this paper, ketoprofen and ketoprofen lysinate were used as model drugs in order to investigate release profiles of poorly soluble and very soluble drug from sodium alginate beads manufactured by prilling. The effect of polymer concentration, viscosity, and drug/polymer ratio on bead micromeritics and drug release rate was studied. Ketoprofen and ketoprofen lysinate loaded alginate beads were obtained in a very narrow dimensional range when the Cross model was used to set prilling operative conditions. Size distribution of alginate beads in the hydrated state was strongly dependent on viscosity of drug/polymer solutions and frequency of the vibration. The release kinetics of the drugs showed that drug release rate was related with alginate concentration and solubility of the drug. Alginate solutions with concentration higher than 0.50% (w/w) were suitable to prepare ketoprofen gastro-resistant formulation, while for ketoprofen lysinate alginate, concentration should be increased to 1.50% (w/w) in order to retain the drug in gastric environment. Differential scanning calorimetry thermograms and Fourier transform infrared analyses of drug-loaded alginate beads indicated complex chemical interactions between carboxyl groups of the drug and polymer matrix in drug-loaded beads that contribute to the differences in release profile between ketoprofen and ketoprofen lysinate. Total release of the drugs in intestinal medium was dependent on the solubility of the drug and was achieved between 4 and 6 h.     

Pharmmaker: Pharmacophore modeling and hit identification based on druggability simulations

Recent years have seen progress in druggability simulations, that is, molecular dynamics simulations of target proteins in solutions containing drug‐like probe molecules to characterize their drug‐binding abilities, if any. An important consecutive step is to analyze the trajectories to construct pharmacophore models (PMs) to use for virtual screening of libraries of small molecules. While considerable success has been observed in this type of computer‐aided drug discovery, a systematic tool encompassing multiple steps from druggability simulations to pharmacophore modeling, to identifying hits by virtual screening of libraries of compounds, has been lacking. We address this need here by developing a new tool, Pharmmaker, building on the DruGUI module of our ProDy application programming interface. Pharmmaker is composed of a suite of steps: (Step 1) identification of high affinity residues for each probe molecule type; (Step 2) selecting high affinity residues and hot spots in the vicinity of sites identified by DruGUI; (Step 3) ranking of the interactions between high affinity residues and specific probes; (Step 4) obtaining probe binding poses and corresponding protein conformations by collecting top‐ranked snapshots; and (Step 5) using those snapshots for constructing PMs. The PMs are then used as filters for identifying hits in structure‐based virtual screening. Pharmmaker, accessible online at http://prody.csb.pitt.edu/pharmmaker , can be used in conjunction with other tools available in ProDy.     

Local conformational changes induced in B-DNA by ethidium intercalation

Structural effects of binding the intercalating drug ethidium bromide (EtBr) to 160 base pair (bp) fragments of nucleosomal calf thymus DNA have been probed by the method of Raman difference spectroscopy. With the use of a near-infrared (NIR) laser source to excite the Raman spectrum at 752 nm, vibrational signatures of both the EtBr intercalant and DNA target have been identified in spectra of the drug-DNA complexes. Analysis of the results obtained on complexes consisting of 1 EtBr bound/10 bp leads to the following conclusions: (i) Raman markers diagnostic of DNA phosphodiester conformation are converted from the B type to the A type with EtBr binding, commensurate with the proportion of ethidium-bound nucleotides in the complex. (ii) Ethidium binding converts deoxynucleoside sugar puckers from the C2'-endo to the C3'-endo conformation, also consistent with binding stoichiometry. Both pyrimidine and purine deoxynucleoside sugar puckers are perturbed by the phenanthridinium ring intercalation. (iii) Phenanthridinium insertion between bases is accomplished with no apparent change in hypochromicities of purine or pyrimidine Raman markers, indicating that base-phenanthridinium interactions provide compensatory hypochromic effects. (iv) Novel Raman markers of helix unwinding have been identified and assigned primarily to methylene deformation modes of the deoxyribosyl C2'H(2) and C5'H(2) groups. The present study provides new insights into drug-DNA recognition in solution and demonstrates the feasibility of NIR-Raman spectroscopy for structural studies of highly chromophoric DNA complexes.     

Quantum chemical insight on vibration spectra of silica systems

A set of silicate ions and corresponding lithium salts have been quantum chemically (QC) simulated in a “free molecule” approach. The infrared (IR), inelastic neutron scattering (INS), and Raman spectra have been simulated and fitted to the experimentally registered ones. The complete assignment of the vibrational bands along with the intensities and potential energy distribution has been performed. The applicability of the traditionally used quasimolecule Si–O–Si model to the interpretation of bands near 440–480 cm^? 1 and so-called “Boson” peak near 50 cm^? 1 has been critically discussed.     

Ketoprofen poly(lactide-co-glycolide) physical interaction

The aim of this work was to provide an understanding of the interaction occurring between ketoprofen and poly(lacticco-glycodic acid) (PLGA) that leads to polymer plasticization. Experimental glass transition temperature (Tg) values were fitted with the theoretical ones predicted by the Fox and Gordon-Taylor/Kelley-Bueche equations. PLGA films containing different amounts of ketoprofen (KET) were prepared by solvent casting and characterized by scanning electron microscopy, differential scanning calorimetry, and Fourier transform infrared spectroscopy (FTIR). Differential scanning calorimetry evidenced that KET acted as a plasticizer in a similar biphasic way in both end-capped and uncapped PLGA. At KET contents of 20% to 35%, depending on the investigated polymer, the Tg was around 23°C. Higher KET amounts did not lower further the Tg, and the excess of drug was found to crystallize into the polymeric matrix. Experimental Tg's deviated negatively from the predicted ones probably because of hydrogen bonding. The FTIR spectra of the films, loaded with different amounts of KET, showed a shift to higher wavenumbers for the peaks at 1697 and 1655 cm⁻¹ confirming the presence of some interactions, probably hydrogen bonds between the ketoprofen carboxylic group and the PLGA carbonyl groups along the polymer backbone. The hydrogen bonding between KET and PLGA is probably responsible for KET plasticizing effect. KET behaving as a lubricant may disrupt polymer chain-chain interactions, removing additional barriers to bond rotation and chain mobility. Published: May 11, 2007     

Biocatalytic synthesis and in vitro release of biodegradable linear polyesters with pendant ketoprofen

Enzyme-catalyzed polycondensation for the synthesis of polyester prodrugs of ketoprofen was reported. Lipase acrylic resin from Candida antarctica (CAL-B) was used to synthesize the linear polyesters with pendent ketoprofen groups based on ketoprofen glycerol ester, poly(ethylene glycol), and divinyl sebacate. The products were characterized by GPC and (1)H NMR. The results indicated that the molecular weight and yields of the polyesters depend on experimental conditions such as temperature and feed ratio. The in vitro study showed that the drug release from the polyester was slow under physiological conditions, which indicated that the polyester could be a promising prodrug with extended pharmacological effects by delayed release of ketoprofen.     

Insulin signaling in the liver and uterus of ovariectomized rats treated with estradiol

We used rat hepatic and uterine tissues to examine the impact of estradiol (E2) on insulin (INS) signaling. Ovariectomized (OVX) female Wistar rats were treated with E2 (20 microg/kg b.wt., i.p.) and used for the experiment 6h after E2 administration. To highlight E2 effects on tyrosine phosphorylation of INS receptor (IR) and INS receptor substrates (IRSs) and IRSs association with p85 subunit of phosphatidylinositol 3-kinase (PI3-K) in the context of INS signaling, E2-treated OVX rats were also injected with INS (20 IU, i.p.), 30 min before the experiment. Treatment with E2 did not change the levels of plasma INS and glucose (Glu). However, it significantly decreased the free fatty acid (FFA) level and increased uterine weight. Furthermore, the results show that E2 had no effect on the content of hepatic IR protein, but significantly increased IR protein content in the uterus and decreased IR tyrosine phosphorylation in both the liver and uterus. Compared to the control, hepatic IRS-1 and IRS-2 were significantly decreased and increased, respectively, after E2 treatment. Protein content of both molecules, IRS-1 and IRS-2, was increased in uterine tissue after E2 administration. Protein content of the p85 subunit of PI3-K and that of protein kinase B (Akt) were increased in the uterus, with no changes in the liver. The results suggest that E2 treatment induces tissue-specific changes in INS signaling. The consequences of E2 treatment on INS signaling molecules are more apparent in the uterus, but their physiological relevance for INS action is probably greater in the liver.     

Investigations on the structural,vibrational, computational,and molecular docking studies on potential antidiabetic chemical agent Diosmetin

In the present study, the harmonic vibrational frequencies of Diosmetin(5, 7 dihydroxy‐2(3‐hydroxy‐4 methoxyphenyl) chromen‐4‐one) have been investigated by both experimental (FTIR and FT‐Raman) and theoretical (HF and DFT/B3LYP) method. The calculated harmonic vibrational frequencies were compared with experimental data. A detailed interpretation of the vibrational spectra of the compound has been made on the basis of the calculated potential energy distribution (PED). The ¹H, ¹³C NMR chemical shifts and TD‐DFT calculations of the molecule were calculated and compared with the available experimental observations. A study on the molecular electrostatic potential surface (MEP) of the compound was performed, and the electrophilic and nucleophilic reactive sites were identified. Furthermore, the inhibition effect of compound against aldose reductase enzyme has been analyzed by molecular docking method, and the results were compared with the standard drug. The docking study indicates that the investigated compound shows better inhibitory activity toward aldose reductase enzyme than the standard drug, and hence this study may be supportive in the field of drug discovery to design more potential antidiabetic agents.     

Polarized Raman spectra of oriented fibers of A DNA and B DNA: anisotropic and isotropic local Raman tensors of base and backbone vibrations.

Polarized Raman spectra of oriented fibers of calf thymus DNA in the A and B conformations have been obtained by use of a Raman microscope operating in the 180 degrees back-scattering geometry. The following polarized Raman intensities in the spectral interval 200-1800 cm-1 were measured with both 514.5 and 488.0 nm laser excitations: (1) Icc, in which the incident and scattered light are polarized parallel to the DNA helical axis (c axis); (2) Ibb, in which the incident and scattered light are polarized perpendicular to c; and (3) Ibc and Icb, in which the incident and scattered light are polarized in mutually perpendicular directions. High degrees of structural homogeneity and unidirectional orientation were confirmed for both the A and B form fibers, as judged by comparison of the observed Raman markers and intensity anisotropies with measurements reported previously for oligonucleotide single crystals of known three-dimensional structures. The fiber Raman anisotropies have been combined with solution Raman depolarization ratios to evaluate the local tensors corresponding to key conformation-sensitive Raman bands of the DNA bases and sugar-phosphate backbone. The present study yields novel vibrational assignments for both A DNA and BDNA conformers and also confirms many previously proposed Raman vibrational assignments. Among the significant new findings are the demonstration of complex patterns of A form and B form indicator bands in the spectral intervals 750-900 and 1050-1100 cm-1, the identification of highly anisotropic tensors corresponding to vibrations of base, deoxyribose, and phosphate moieties, and the determination of relatively isotropic Raman tensors for the symmetrical stretching mode of phosphodioxy groups in A and B DNA. The present fiber results provide a basis for exploitation of polarized Raman spectroscopy to determine DNA helix orientation as well as to probe specific nucleotide residue orientations in nucleoproteins, viruses, and other complex biological assemblies.     

Dependence of the Raman signature of genomic B-DNA on nucleotide base sequence.

The vibrational spectra of four genomic and two synthetic DNAs, encompassing a wide range in base composition [poly(dA-dT). poly(dA-dT), 0% G + C; Clostridium perfringens DNA, 27% G + C; calf thymus DNA, 42% G + C; Escherichia coli DNA, 50% G + C; Micrococcus luteus DNA, 72% G + C; poly(dG-dC).poly(dG-dC), 100% G + C] (dA: deoxyadenosine; dG: deoxyguanosine; dC: deoxycytidine; dT: thymidine), have been analyzed using Raman difference methods of high sensitivity. The results show that the Raman signature of B DNA depends in detail upon both genomic base composition and sequence. Raman bands assigned to vibrational modes of the deoxyribose-phosphate backbone are among the most sensitive to base sequence, indicating that within the B family of conformations major differences occur in the backbone geometry of AT- and GC-rich domains. Raman bands assigned to in-plane vibrations of the purine and pyrimidine bases-particularly of A and T-exhibit large deviations from the patterns expected for random base distributions, establishing that Raman hypochromic effects in genomic DNA are also highly sequence dependent. The present study provides a basis for future use of Raman spectroscopy to analyze sequence-specific DNA-ligand interactions. The demonstration of sequence dependency in the Raman spectrum of genomic B DNA also implies the capability to distinguish genomic DNAs by means of their characteristic Raman signatures.     

Poly(3-hydroxybutyrate)/chitosan/ketoprofen or piroxicam composite microparticles: Preparation and controlled drug release evaluation

Poly(3-hydroxybutyrate)/chitosan/piroxicam or ketoprofen composite microparticles were prepared by the solid-in-water-in-oil emulsion-solvent evaporation technique with the aim of reducing the burst effect and controlling the drug release. Reservoir-type microparticles, composed of poly(3-hydroxybutyrate) microspheres embedded in a chitosan matrix were prepared. The size and morphological characteristics of the composite microparticles were evaluated in relation to the chitosan concentration and cross-linking with glutaraldehyde. Reservoir-type composite microparticles were obtained using 2.0% and 3.0% w/v chitosan solutions. A significant reduction in the burst effect and prolonged drug release were observed, particularly when higher chitosan and glutaraldehyde concentrations were used.     

Chromophore interactions in flavocytochrome c552: a resonance Raman investigation

Resonance Raman spectra with both Soret and visible excitation have been obtained for Chromatium flavocytochrome c552 and its isolated diheme subunit under varying conditions of pH and inhibitor binding. The spectra are generally consistent with previously established classification schemes for porphyrin ring vibrations. The presence of covalently bound flavin in the protein was apparent in the fluorescent background it produced and in flavin-mediated photoeffects observed in heme Raman spectra obtained at high laser power. No flavin modes were present in the Raman spectra, nor was any evidence of direct heme-flavin interaction found by using this technique; however, a systematic perturbation of heme B1g vibrational frequencies was found in the oxidized holoprotein. The heme vibrational frequencies of c552 are compared to those of the diheme peptide and of other c-type cytochromes. They are consistent with an interpretation that involves pH-dependent changes in axial ligation and treats the hemes and flavin as isolated chromphores communicating via protein-mediated interactions.     

Raman spectroscopic study of the conformational changes of thyroxine induced by interactions with phospholipid

Comparison of the Raman spectra of thyroxine ( L-3,3',5,5'-tetraiodothyronine) in the pure state and in a 1:5 mixture with phosphatidylcholine reveals spectral differences that reflect structural changes of thyroxine induced by interactions with the phospholipid. These structural changes could be localized in specific parts of the thyroxine molecule on the basis of a vibrational analysis that was carried out by density functional calculations with the B3LYP hybrid functional applying the SDD effective core potential basis set. The calculated (and subsequently scaled) frequencies reveal a good agreement with the experimental data, which together with calculated IR and Raman intensities allow a plausible assignment of most of the IR and Raman bands. Thus, it is found that modes localized in the aromatic beta-ring and in the ether group as well as the C-I stretching modes of ring alpha are affected upon lipid interactions, indicating that thyroxine interacts with the phosphatidylcholine bilayer via penetration of the hydrophobic part of the molecule.     

Bio-Dis and the Paddle Dissolution Apparatuses Applied to the Release Characterization of Ketoprofen from Hypromellose Matrices

The purposes of this work were: (1) to comparatively evaluate the effects of hypromellose viscosity grade and content on ketoprofen release from matrix tablets, using Bio-Dis and the paddle apparatuses, (2) to investigate the influence of the pH of the dissolution medium on drug release. Furthermore, since direct compression had not shown to be appropriate to obtain the matrices under study, it was also an objective (3) to evaluate the impact of granulation on drug release process. Six formulations of ketoprofen matrix tablets were obtained by compression, with or without previous granulation, varying the content and viscosity grade of hypromellose. Dissolution tests were carried out at a fixed pH, in each experiment, with the paddle method (pH 4.5, 6.0, 6.8, or 7.2), while a pH gradient was used in Bio-Dis (pH 1.2 to 7.2). The higher the hypromellose viscosity grade and content were, the lower the amount of ketoprofen released was in both apparatuses, the content effect being more expressive. Drug dissolution enhanced with the increase of the pH of the medium due to its pH-dependent solubility. Granulation caused an increase in drug dissolution and modified the mechanism of the release process.Key words: apparatus 3, Bio-Dis, dissolution, hypromellose matrix, ketoprofen     

Vibrational analysis and molecular force field of hypoxanthine as determined from ultraviolet resonance Raman spectra of native and deuterated species

Ultraviolet resonance Raman scattering spectra from aqueous solutions of hypoxanthine and its deuterated species (C8-deuterated, N-deuterated and C8-, N-deuterated derivatives) have been collected and reported in the spectral region between 400 and 1800 cm^–1. The laser excitation wavelengths at 281 nm and 257 nm correspond to preresonance and pure resonance conditions, respectively, with the purine strongly allowed ^* electronic transition: thus the observed experimental Raman features mainly correspond to inplane vibrational modes. The latter were then assigned according to the Wilson GF method by using an empirical harmonic valence force field. Normal mode calculations are based on a non-redundant set of internal coordinates. The calculated vibrational mode wavenumbers and their isotopic shifts upon selective deuterations are in good agreement with the experimental data. The present normal mode analysis rests on the transferability of the guanine and adenine force constants proposed in recent works based on resonance Raman spectroscopy and neutron inelastic scattering data from these major purine bases. Correspondence to: M. Ghomi     

Development and Optimization of Micro/Nanoporous Osmotic Pump Tablets

Micro/nanoporous osmotic pump tablets coated with cellulose acetate containing polyvinylpyrolidone (PVP) as pore formers were fabricated. Propranolol hydrochloride was used as a model drug in this study. Formulation optimization based on USP 31 requirements was conducted following a central composite design using a two-level factorial plan involving two membrane variables (pore former and coating levels). Effect of molecular weight of pore former (PVP K30 and PVP K90) was also evaluated. Responses of drug release to the variables were analyzed using statistical software (MINITAB 14). Scanning electron microscopy and atomic force microscopy showed that the pores formed by PVP. The drug release was dependent on the molecular weight and concentration of PVP and the level of coating. The results showed that acceptable 12-h profile could be achieved with only specific range of PVP K30-containing membrane at the defined membrane thickness. However, satisfactory 24-h profile could be accomplished by both PVP K30 and PVP K90-containing membrane at the range and membrane thickness tested. Preparation and testing of the optimized formulation showed a good correlation between predicted and observed values.     

Raman spectroscopy of DNA-metal complexes. I. Interactions and conformational effects of the divalent cations: Mg, Ca, Sr, Ba, Mn, Co, Ni, Cu, Pd, and Cd.

Interactions of divalent metal cations (Mg2+, Ca2+, Ba2+, Sr2+, Mn2+, Co2+, Ni2+, Cu2+, Pd2+, and Cd2+) with DNA have been investigated by laser Raman spectroscopy. Both genomic calf-thymus DNA (> 23 kilobase pairs) and mononucleosomal fragments (160 base pairs) were employed as targets of metal interaction in solutions containing 5 weight-% DNA and metal:phosphate molar ratios of 0.6:1. Raman difference spectra reveal that transition metal cations (Mn2+, Co2+, Ni2+, Cu2+, Pd2+, and Cd2+) induce the greatest structural changes in B-DNA. The Raman (vibrational) band differences are extensive and indicate partial disordering of the B-form backbone, reduction in base stacking, reduction in base pairing, and specific metal interaction with acceptor sites on the purine (N7) and pyrimidine (N3) rings. Many of the observed spectral changes parallel those accompanying thermal denaturation of B-DNA and suggest that the metals link the bases of denatured DNA. While exocyclic carbonyls of dT, dG, and dC may stabilize metal ligation, correlation plots show that perturbations of the carbonyls are mainly a consequence of metal-induced denaturation of the double helix. Transition metal interactions with the DNA phosphates are weak in comparison to interactions with the bases, except in the case of Cu2+, which strongly perturbs both base and phosphate group vibrations. On the other hand, the Raman signature of B-DNA is largely unperturbed by Mg2+, Ca2+, Sr2+, and Ba2+, suggesting much weaker interactions of the alkaline earth metals with both base and phosphate sites. A notable exception is a moderate perturbation by alkaline earths of purine N7 sites in 160-base pair DNA, with Ca2+ causing the greatest effect. Correlation plots demonstrate a strong interrelationship between perturbations of Raman bands assigned to ring vibrations of the bases and those of bands assigned to exocyclic carbonyls and backbone phosphodiester groups. However, strong correlations do not occur between the Raman phosphodioxy band (centered near 1092 cm-1) and other Raman bands, suggesting that the former is not highly sensitive to the structural changes induced by divalent metal cations. The structural perturbations induced by divalent cations are much greater for > 23-kilobase pair DNA than for 160-base pair DNA, as evidenced by both the Raman difference spectra and the tendency toward the formation of insoluble aggregates. In the presence of transition metals, aggregation of high-molecular-weight DNA is evident at temperatures as low as 11 degrees C.(ABSTRACT TRUNCATED AT 400 WORDS)