Search

In contrast to DNA replication and transcription where nucleotides are added and matched one by one, homologous recombination by DNA strand exchange tests whole sequences for complementarity, which requires elimination of mismatched yet thermodynamically stable intermediates. To understand the remarkable sequence specificity of homologous recombination, we have studied strand exchange between a 20-mer duplex containing one single mismatch (placed at varied positions) with the matching single strand in presence of poly(ethylene glycol) representing a semi-hydrophobic environment. A FRET-based assay shows that rates and yields of strand exchange from mismatched to matched strands rapidly increase with semi-hydrophobic co-solute concentration, contrasting previously observed general strand exchange accelerating effect of ethyl glycol ethers. We argue that this effect is not caused simply by DNA melting or solvent-induced changes of DNA conformation but is more complex involving several mechanisms. The catalytic effects, we propose, involve strand invasion facilitated by reduced duplex stability due to weakened base stacking ("longitudinal breathing"). Secondly, decreased water activity makes base-pair hydrogen bonds stronger, increasing the relative energy penalty per mismatch. Finally, unstacked mismatched bases (gaps) are stabilized through partly intercalated hydrophobic co-solvent molecules, assisting nucleation of strand invasion at the point of mismatch. We speculate that nature long ago discovered, and now exploits in various enzymes, that sequence recognition power of nucleic acids may be modulated in a hydrophobic environment.

The DNA interaction of derivatives of ellipticine with heterocyclic ring systems with three, four, or five rings and a dimethylaminoethyl side chain was studied. Optical spectroscopy of drug complexes with calf thymus DNA, poly [(dA-dT) · (dA-dT)], or poly [(dG-dC) · (dG-dC)] showed a 10 nm bathochromic shift of the light absorption bands of the pentacyclic and tetracyclic compounds upon binding to the nucleic acids, which indicates binding by intercalation. For the tricyclic compound a smaller shift of 1–3 nm was observed upon binding to the nucleic acids. Flow linear dichroism studies show that the geometry of all complexes is consistent with intercalation of the ring system, except for the DNA and poly [(dG-dC) · (dG-dC)] complexes of the tricyclic compound, where the average angle between the drug molecular plane and the DNA helix axis was found to be 65°. One-dimensional 1H-nmr spectroscopy was used to study complexes between d(CGCGATCGCG)2 and the tricyclic and pentacyclic compounds. The results on the pentacyclic compound show nonselective broadening due to intermediate chemical exchange of most oligonucleotide resonances upon drug binding. The imino proton resonances are in slow chemical exchange, and new resonances with upfield shifts approaching 1 ppm appear upon drug binding, which supports intercalative binding of the pentacyclic compound. The results on the tricyclic compound show more rapid binding kinetics and very selective broadening of resonances. The data suggest that the tricyclic compound is in an equilibrium between intercalation and minor groove binding, with a preference to bind close to the AT base pairs with the side chain residing in the minor groove. © 1994 John Wiley & Sons, Inc.

The Z form of poly [d(G-m5C)2], in presence of Mg2+ ion, is found to be transformed into B form upon interaction with 4',6-diamidino-2-phenylindole (DAPI). The Z --> B transformation is complete at a mixing ratio of about 0.07 DAPI per DNA base pairs, i.e., each DAPI molecule may be related to the conversion of 6-7 base pairs. An interaction between DAPI and poly [d(G-m5C)2] in its Z form at low drug: DNA ratios is suggested from optical dichroism and time-resolved luminescence anisotropy results. The spectroscopic behaviour of DAPI indicates that the Z conformation of DNA does not provide normal binding sites for DAPI, such as groove or intercalation sites, but that the initial association may be of external nature. (C) 1993 John Wiley & Sons, Inc.

The acridine dye quinacrine and its interactions with calf thymus DNA, poly(dA-dT).poly(dA-dT), and poly(dG-dC).poly(dG-dC) were studied by light absorption, linear dichroism, and fluorescence spectroscopy. The transition moments of quinacrine give rise to absorption bands polarized along the short axis (400-480-nm band), and the long axis (345-nm and 290-nm bands) of the molecule, respectively. Linear dichroism studies show that quinacrine intercalates into calf thymus DNA as well as into the polynucleotides, displaying fairly homogeneous binding to poly (dA-dT).poly(dA-dT), but more than one type of intercalation site for calf thymus DNA and poly(dG-dC).poly(dG-dC). Fluorescence spectroscopy shows that for free quinacrine the pK = 8.1 between the mono- and diprotonated states also remains unchanged in the excited state. Quinacrine bound to calf thymus DNA and polynucleotides exhibits light absorption typical for the intercalated diprotonated form. The fluorescence enhancement of quinacrine bound to poly (dA-dT).poly(dA-dT) may be due to shielding from water interactions involving transient H-bond formation. The fluorescence quenching in poly (dG-dC).poly (dG-dC) may be due to excited state electron transfer from guanine to quinacrine. (C) 1993 John Wiley & Sons, Inc.

A systematic theoretical study of the CD of [poly (dA-dT)]2 and its complexes with achiral small molecules is presented. The CD spectra of [poly (dA-dT)]2 and of poly (dA):poly(dT) are calculated for various DNA structures using the matrix method. The calculated and experimental spectra agree reasonably well for [poly(dA-dT)]2 but less well for poly(dA):poly(dT). The calculated CD spectrum of [poly(dA-dT)]2 fails to reproduce the wavelength region of 205-245 nm of the experimental spectrum. This discrepancy can be explained by a magnetic dipole allowed transition contributing significantly to the CD spectrum in this region. The induced CD of a transition moment of a molecule bound to [poly(dA-dT)]2 is also calculated. As was the case for [poly(dG-dC)]2, the induced CD of a groove bound molecule is one order of magnitude stronger than that of an intercalated molecule. The calculations also show considerable differences between pyrimidine-purine sites and purine-pyrimidine sites. Both signs and magnitudes of the CD induced into ligands bound in the minor groove agree with experimental observations.

The NaCl-induced transition from B to Z conformation in poly (dG-dC), covalently modified with benzo(a)pyrene diol epoxides (BPDE), has been studied by use of flow linear dichroism (LD), CD, and fluorescence spectroscopy. In contrast to unmodified poly (dG-dC), the BPDE-modified polynucleotide displays a B to Z conversion that is incomplete in 2.7 M NaCl. This and a number of other observations are found consistent with a locally retained BPDE perturbed B-like conformation in the vicinity of the BPDE adduct. In 2.7 M NaCl the polynucleotide orientation in flow is strongly reduced, which could indicate flexibility at the junctions between domains of BPDE perturbed B-like conformation and Z conformation. A virtually stable high-salt B form is found in 2.25 M NaCl at 0°C; its CD spectrum resembles that of the transient so-called B* form, which is observed at an early stage of the B to Z transition. Characteristic B to Z transition times were evaluated by taking this B*-form CD spectrum as the initial state spectrum. The kinetics of the B to Z transition, monitored by the CD signal at 287 nm, shows an increased transition rate with (+)-anti-BPDE modification whereas the (−)-anti- and (±)-syn-BPDE adducts have retarding effects. The difference is discussed against the background of the different binding geometries of the BPDE isomer adducts.

Fluorescence of (+)-anti-benzo (a) pyrene diol epoxide [(+)-anti-BPDE] covalently bound to poly (dG-dC) has been studied with steady-state and time-resolved techniques. Extensive formation of excimers is found, even at small (0.008) BPDE/nucleotide ratios. This indicates favored covalent binding to bases close to already modified guanines. Both fluorescence excitation spectra and lifetime measurements reveal two populations of (+)-anti-BPDE adducts: one that can form excimers and one that cannot. Three excimer lifetimes (4.5, 29, and 83 ns) are observed. Differently shifted monomer and excimer excitation spectra are discussed in terms of pyrene-pyrene exciton interactions, consistent with a distance shorter than 7 A between the excimer-forming BPDE chromophores.

A systematic theoretical study of the CD of double-stranded poly(dG-dC) and its complexes with small molecules is presented. The intrinsic CD of the polymer and the induced CD of a transition belonging to a molecule bound to DNA are calculated using the matrix method. The calculations show considerable differences between pyrimidine-purine and purine-pyrimidine binding sites, and we find that the induced CD of a groove bound molecule is one order of magnitude stronger than that of an intercalated molecule. The results form a sound basis for interpreting the CD of ligand-DNA systems in terms of molecular geometry, interactions, and spectroscopy.

We studied the interactions of the substitution-inert inversion-labile complexes Fe(bipy) and Fe(phen) [and the inversion-stable complex Ru(bipy)] with DNA. The association of these complexes to DNA is mainly electrostatic, and Fe(phen) shows a more effective binding to DNA than the two bipyridyl complexes, possibly owing to a different binding mode. The interactions are enantioselective, leading to a Pfeiffer shift in the diastereomeric inversion equilibria and an excess of the Δ-enantiomer of Fe(phen) and Fe(bipy), which is directly monitorable through CD. The partition constants for the inversion equilibrium range from 1.3 to 2.0 for Fe(bipy) and Fe(phen), depending on ionic conditions. From flow LD information about the orientation of the complexes on DNA was obtained: it is consistent with a fit of the Δ-enantiomer in the major groove of the right-handed DNA helix. The mechanisms of interaction are discussed against equilibrium, spectroscopic, and kinetic data.

We have studied the denaturation of DNA by linear dichroism (LD) techniques in the following systems: (1) in aqueous solution at low pH, (2) in 90% (w/w) ethylene glycol and in 90% glycerol solutions, and (3) in a matrix of polyvinyl alcohol (PVA). We report that denatured DNA at low ionic strength can be oriented by shear forces in all these systems and that it exhibits positive LD with markedly varying LD/A over the absorption band centered at 260 nm (A being the ordinary absorbance) as compared to the negative LD with approximately constant LD/A of ordinary native DNA. These results are interpreted in terms of a large tilt of the planes of the DNA bases. DNA adopts denatured forms in (1) aqueous solutions at pH < 3.5, (2) ethylene glycol and glycerol at low Na+ concentrations, and (3) PVA after heating. If the denaturation is done in dry PVA, a complete re-formation of the double helix is obtained after humidifying (60% w/w) the matrix. This shows that the matrix can keep the two strands in a position allowing a perfect fitting on reassociation. Previous attempts to determine any intrinsic LD of denatured DNA have failed, probably due to the fact that, unless a very low ionic strength is used, the flexibility of the single strands prevents the orientation required to give a detectable LD signal even with recently developed high-sensitivity measurement techniques.

The understanding, on a molecular level, of the mechanisms responsible for the improved separation in DNA gel electrophoresis when using modulated electric fields requires detailed information about conformational distribution and dynamics in the DNA/gel system. The orientational order due to electrophoretic migration ("electrophoretic orientation") is an interesting piece of information in this context that can be obtained through linear dichroism spectroscopy [M. Jonsson, B. Akerman, and B. Norden, (1988) Biopolymers 27, 381-414]. The technique permits measurement of the orientation factor S of DNA (S = 1 corresponds to perfect orientation) within an electrophoretic zone in the gel during the electrophoresis. It is reported that the degree of orientation of T2 DNA [170 kilo base pairs (kpb)] is considerable (S = 0.17 in 1% agarose at 10 V/cm) compared to relatively modest orientations of short fragments found earlier (for 23-kbp DNA, S = 0.03 in 1% agarose at 10 V/cm), showing that large DNA coils are substantially deformed during the migration. Growth and relaxation dynamics of the orientational order of the T2 DNA are also reported, as functions of gel concentration (0.3-2%), electric field strength (0-40 V/cm), and pulse characteristics. The rise profile of the DNA orientation, when applying a constant field, is a nonmonotonic function that displays a pronounced overshoot, followed by a minor undershoot, before it reaches steady-state orientation (after 12 s in 1% agarose, 9 V/cm). The orientational relaxation in absence of field shows a multiexponential decay in a time region of some 10 s, when most of the DNA anisotropy has disappeared. A surprising phenomenon is a memory over minutes of the DNA/gel system to previous pulses: with two consecutive rectangular pulses (of the same polarity), the orientational overshoot and undershoot as a response to the second pulse are significantly reduced compared to the first pulse. The time required to recover 90% of their amplitudes is typically 1200 s (1% agarose, 9 V/cm), which may be compared to the time required to relax 90% of the DNA orientation, which is only 6 s. The major part of the over- and undershoot recovery is thus a reorganization of a system in which DNA is already randomly oriented. The different response amplitudes and relaxation times, including the amplitude and recovery time of the overshoot, of the orientational order of DNA in the electrophoretic gel have been studied as functions of gel concentration and field strength. The results are discussed against relevant theories of polymer dynamics.

We have calculated the uv linear dichroism for the A- and B-forms of DNA using π-π* transition moments and band components determined from the free DNA bases. The reduced dichroism (LDR) as a function of wavelength is estimated is in the 220–300-nm region, for both the oriented-gas model and a simple exciton model. For B-form DNA, LDR is obtained to −1.48S (S being the orientation factor) over the whole wavelenth region by both models. For A-form DNA, LDR is not constant, but changes monotonically from about −1.15S at 220 nm to about −1.35S to −1.45S at 300 nm, depending on base combination and degree of interaction (−1.35S for the oriented gas). It is emphasized that a common assumption of a single “effective” transition moment of the principal band at 260 nm may not generally be made because of the extensive overlap of differently polarized bands. The possibility of using the reduced dichroism curve for characterizing the secondary structure of DNA is discussed.

The uv linear dichroism of calf thylus DNA has been studied at different degrees of orientation both in flow-oriented ethanol–water solutions and in a stretched aqueous host of poly(vinyl alcohol) (PVA). The reduced dichroism (LDR) curves in the region 250–290 nm for DNA in PVA films at 75 and 100% relative humidity (r.h.) are in fair agreement with the curves calculated for the A- and B-forms of DNA, based on the fiber structures and the π-π* transitions of the free bases. This suggests that DNA adopts its A and B conformations in PVA at 75 and 100% r.h. In ethanol, on the other hand, a deviation from the A-form spectrum shows that the conformation of DNA in the solution can differ from the fiber structure. At shorter wavelenghts, a positive contribution to LDR is explained in terms of an out-of-plane polarized n-π* transition.

The interaction of silver(I) with DNA has been studied with uv LD in aqueous solution and in a humid anisotropic poly(vinyl alcohol) host (B-DNA) and also in 80% ethanolic solution (A-DNA). Addition of silver ions has a pronounced effect on the dichroic spectra of DNA, indicating that the DNA structure is significantly altered. By correlation with calculated reduced LD spectra, using intensities and moments of the corresponding electronic transitions of the DNA bases, the experimental spectra of DNA at high silver content may be interpreted in terms of tilt and roll angles of the bases in the double helix. In ethanolic A-DNA solution there is a pronounced decrease in the orientation by flow of DNA, suggesting that the complexation of DNA to silver may be accompanied by the formation of compact tertiary structures.

We have studied the linear dichroism (LD) of rat liver chromatin oriented by flow. Soluble chromatin, prepared by brief nuclease digestion, is found to exhibit a positive LD at low ionic strength (1 mM NaCl), with a constant LD/A over the absorption band centered at 260 nm (A, isotropic absorbance). Several previous dichroism studies on soluble chromatin have been performed on sonicated materials and have given negative LD, probably due to the presence of uncoiled DNA. The positive dichroism can be interpreted in terms of a supercoil of DNA in chromatin with a pitch angle larger than 55°, and is, for example, consistent with a model where the cylindrical nucleosome core particles are stacked face to face in the chromatin filament. In contrast to the nuclease-digested chromatin, sonicated chromatin was confirmed to exhibit negative LD. This difference can be attributed to a partial uncoiling of the linker regions between the nucleosomes due to the shearing. The structural transition of chromatin to a compact form can be observed as a reduction of the positive LD of the nuclease-digested chromatin to almost zero in 0.1 M NaCl or in 0.1 mM MgCl2. This transition is due to a decreased electrostatic repulsion between negative phosphate groups on the DNA chain. In the case of Na+, this can be explained as a screening effect due to the bulk concentration of Na+. With Mg2+ a considerably stronger effect may indicate a more localized binding to the phosphates. At ionic strengths higher than 0.5M NaCl, the dissociation of the histones from DNA leads to uncoiling of chromatin. The change in LD during this process shows that histone H1 contributes only to a small degree to the coiling of the DNA chain, whereas histones H3 and H4 play the major role in the coiling.

A method for in situ study of orientation of DNA during gel electrophoresis has been developed. Linear dichroism spectra measured by this phase-modulation technique can sensitively and selectively detect orientation of DNA during electrophoretic migration in gel. [Measurement of “electrophoretic orientation” was first reported in 1985 by B. Akerman, M. Jonsson, and B. Norden (1985) (J. Chem. Soc. Chem. Commun. 422–423)]. Restriction fragments of duplex DNA of lengths in the ranges of 300–2319 base pairs (bp) and 4361–23130 bp have been studied in 5% polyacrylamide and 1% agarose gels, respectively. The fragments become preferentially oriented with the DNA helix axis parallel to the migration direction. In agarose the orientation is found to increase sigmoidally, and in polyacrylamide, linearly, with the electric field strength, within the field ranges accessible to measurement (0–40 and 5–40 V/cm, respectively). In both types of gels a considerable increase in orientation with length of DNA was observed. Compared to dipole orientation in electric fields, the electrophoretic orientation is high: orientation factor S = 0.027 in agarose for 23130 bp at 10 V/cm and S = 0.004 in polyacrylamide for 2319 bp at 10 V/cm. In addition to orientation of DNA, the electrophoresis also leads to orientation effects in the gel structure owing to Joule heating. In agarose there is also an effect that is associated with the migrating DNA zones and that produces different orientations of the gel at the front and rear parts of a zone. Evidence is presented that this effect is due to a DNA-induced electroosmotic flow causing a contraction of the gel in the front of the zone and an expansion in the rear. The experimental results on DNA orientation are compared with the reptation theories for gel electrophoresis. The theory of Lumpkin et al. [O. J. Lumpkin, P. Dejardin, and B. H. Zimm (1985) Biopolymers24, 1573–1593] predicts no orientation length dependence, but it does predict a shape of the field dependence that resembles the shape observed in agarose. The theory of Slater and Noolandi [G. W. Slater and J. Noolandi (1986) Biopolymers25, 431–454] predicts an orientational length dependence that is an order of magnitude less than the experimental one, and a field dependence that agrees neither with the sigmoidal shape observed in agarose nor with the linear dependence in polyacrylamide.

Bradley et al. [(1972) Biopolymers11, 645–652] used electro-optical measurements to show that methylene blue (MB), like acridine orange, in its DNA complex is oriented more or less perpendicular to the helix axis as expected if intercalated. High-precision flow linear dichroism (LD) here confirms that MB is coplanar with the DNA bases at low dye/DNA binding ratios and low ionic strengths. CD shows two origins of induced optical activity for the transition of lowest energy (polarized parallel to the long-axis of the dye molecule): at low binding ratios (r 300 mM Na+ or >0.4 mM Mg2+). Nondegenerate coupled-oscillator theory can explain the CD in terms of interactions of transition moments of the dye and the nearest nucleotide bases and indicates a change between two intercalation geometries: a Lerman type of mode, denoted γ−, and an orthogonal mode, denoted γ+. This rotation of MB in the base-pair pocket is accomplished at Na+ and Mg2+ concentrations when the phosphates are effectively screened and the result suggests a more localized bonding of Mg2+ than is expected from simple polyelectrolyte models. The exciton effect at high binding ratios, observed both in CD and in LD, can be interpreted in terms of an interaction between an intercalated and a nonintercalated MB. The geometry of this “accidental” dimer is consistent with a location of the nonintercalated MB in the minor groove, bridging the strands by the positively charged amino groups directed towards phosphate groups. The dihedral angle of the MB pairs, corresponding to a left-handed helix, is opposite to that with acridine orange and proflavine on DNA, indicating that the latter ligands bind to DNA in a different way.