Search

The neuronal protein α-synuclein, linked to Parkinson’s disease, binds to negatively charged vesicles adopting a partial α-helix structure, but helix arrangement at the vesicle surface is not fully understood. Using linear dichroism spectroscopy (LD), we study the interaction of monomeric α-synuclein with large unilamellar vesicles of 1,2-dioleoyl-sn-glycero-3-phospho-l-serine (DOPS), 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-l-serine (POPS), and 1,2-dioleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (DOPG) under mild shear flow. The LD data of oriented lipid vesicles show that the long axis of the protein helix is oriented preferentially perpendicular to the membrane normal but deviates from a uniform in-plane distribution. Upon initial binding, a fraction of helices are oriented in the direction of least curvature for all ellipsoid-shaped vesicles at a lipid:protein molar ratio of 100. However, at a lower protein concentration the helices distribute uniformly on DOPS and POPS vesicles. In all cases, the α-synuclein helices rearrange with time (minute time scale) in the shear flow and begin to tilt into the vesicle membrane. Faster reorientation kinetics in the presence of flow suggests that modulation of membrane dynamics, by thermal or shear-dynamic activation, may overcome steric barriers by what may be called “flow catalysis”.

The interactions between the stereoisomers of the chiral bis-intercalator [mu -C4(cpdppz)(2)-(phen)(4)Ru-2](4+) and DNA reveal interesting dynamic discrimination properties. The two enantiomers Delta-Delta and Lambda-Lambda both form very strong complexes with calf thymus DNA with similar thermodynamic affinities. By contrast, they display considerable variations in their binding kinetics. The Delta-Delta enantiomer has higher affinity for calf thymus DNA than for [poly(dA-dT)](2), and the association kinetics of the dimer to DNA, as well as to polynucleotides, requires a multiexponential fitting function. The dissociation reaction, on the other hand, could be described by a single exponential for [poly(dA-dT)](2), whereas two exponentials were required for mixed-sequence DNA. To understand the key mechanistic steps of the reaction, the kinetics was studied at varied salt concentration for different choices of DNA and chirality of the threading complex. The enantiomers were found to have markedly different dissociation rates, the Lambda-Lambda enantiomer dissociating about an order of magnitude faster than the Delta-Delta enantiomer. Also. the salt dependence of the dissociation rate constants differed between the enantiomers, being stronger for the Lambda-Lambda enantiomer than for the Delta-Delta enantiomer. Since the dissociation reaction requires unthreading of bulky parts of the bis-intercalator through the DNA helix, a considerable conformational change of the DNA must be involved, possibly defining the rate-limiting step.

In this paper we report the DNA binding proper ties of piperazinecarbonyloxy-2-propyl derivatives of anthracene (2), pyrene (3), and phenylanthracene (4). An intercalative binding mode is found for 2 and 3, while the phenyl group of 4 prevents intercalation and leads to external binding. Preferential binding of the (S)-enantiomers is found for both anthracene 2 and pyrene 3. However, the enantiomeric preference is small, with K-(R)/K-(S) being around 0.5 for both the anthracene and the pyrene compounds. This is interpreted in terms of orientation polarity in the binding, by which any intrinsic enantioselectivity is canceled by averaging of opposite binding orientations. The affinities for poly(dA-dT)(2) (AT) are 10(4) M-1 for anthracene derivative 2, and 5 x 10(5) M-1 for pyrene derivative 2. The affinities for poly(dG-dC)(2) (GC) are I order of magnitude lower than those for AT. This is explained by steric interference of the piperazinium tail with the exocyclic amino groups of guanine in the minor groove of GC, leading to a more shallow intercalation in GC than in AT, as also indicated by significantly less negative reduced linear dichroism of the intercalator absorption bands in the GC complexes. This behavior is consistent with that observed for the previously studied achiral analogues.(1) Binding thermodynamics support the difference in binding mode between AT and GC. The binding enthalpy of the AT complexes is significantly more negative than that of the corresponding GC complexes. This indicates a larger overlap of intercalating moiety and nucleobases in the AT complexes, consistent with the linear dichroism results.

Four novel piperazinylcarbonyloxyethyl derivatives of anthracene and pyrene have been prepared and investigated with respect to sequence specificity and synergism between hydrophobic and electrostatic effects upon binding to DNA. Linear and circular dichroism spectroscopy was used to assess the orientation of the aromatic chromophores relative to the nucleobases. Anthracene and pyrene derivatives 2a and 3 are both concluded to bind to homo-polynucleotide poly(dA-dT)(2) by intercalation of their aromatic moieties between base pairs, with a binding constant K-AT of 4 x 10(5) M-1 and 2 x 10(6) M-1, respectively. Significantly reduced affinities (K-GC = 3 x 10(4) M-1 and 10(5) M-1, respectively) are observed with poly(dG-dC)(2), due to less favorable interactions of the piperazinium tail in the minor groove. Base pair specificity is reflected in the binding thermodynamics, with the binding to AT being more enthalpically driven than the binding to GC. Phenyl substitution at the quaternary piperazinium site of the anthracene derivative 2b, does not affect the ratio K-AT/K-GC, but reduces the affinity for both AT End GC slightly. Moreover, the phenyl group in the 10-position of 4 prevents intercalation, and apparently, this compound binds externally to both AT and GC duplex polynucleotides. The results are discussed in terms of general features of the interactions of the intercalating and minor-groove binding molecular moieties, and their interplay with each other, with potentials for tuning specificity.

The electronically excited states of the phthalimide chromophore have been studied by means of linear dichroism (LD) of samples partially oriented in poly(vinyl alcohol) films, magnetic circular dichroism (MCD), and circular dichroism (CD) spectroscopy. On the basis of the LD measurements, the low-energy tail (340-320 nm) of the first absorption band is assigned to an out-of-plane polarized pi-->pi* transition (I). At higher energy, the electronic spectrum is resolved into contributions from five pi-->pi* transitions: II(300 nm, long-axis polarized), III (275 nm, short-axis polarized), IV (235 nm, short-axis polarized), V (220 nm, long-axis polarized), and VI (similar to 210 nm, short-axis polarized). The results from semiempirical (INDO/S-CI) and ab initio (CIS/6-31+G(d)) MO calculations compare well with the proposed assignments of the excited states. Degenerate exciton interaction between electric-dipole-allowed transitions of two phthalimide chromophores is observed in the electronic absorption spectra of the achiral bis-phthalimides 2a-c and in the CD spectrum of the chiral bis-phthalimide 3a. For the latter compound, the solid-state geometry has been determined by X-ray diffraction analysis. Good agreement between experimental and computed CD spectra confirms that the coupled-oscillator exciton model provides the basis for a reliable nonempirical method for the assignment of absolute configuration for this class of compounds. Nondegenerate exciton coupling between phthalimide and benzoate or phenyl chromophores is born out in the CD spectra of homochiral molecules 3c and 3d with the rigid cyclohexane skeleton. Finally, the exciton coupling method is used to make stereochemical assignments for the acyclic, conformationally flexible derivatives 4a-c and 5b.

Norfloxacin, a quinolone antibacterial reagent, has been studied with respect to its binding to calf thymus DNA using fluorescence and linear dichroism techniques and unwinding of supercoiled DNA. The fluorescence of norfloxacin is strongly quenched in the presence of DNA and using this decrease in a fluorescence titration the equilibrium constant of the complex formation was established to be 2.8 x 10(3) M-1. The electric transition moments of the norfloxacin chromophore have been analyzed using fluorescence anisotropy, magnetic circular dichroism, and linear dichroism in stretched poly(vinyl alcohol) film and INDO/S calculations. These data are then used to interpret flow linear dichroism results for the norfloxacin-DNA complex. The transition moments for the long-wavelength transitions are found to be oriented at about 65.0-85.0 degrees with respect to the DNA helix axis. A near perpendicular orientation of the norfloxacin chromophore plane makes it possible to exclude classical groove or surface binding modes. The possibility of a classical intercalation binding mode also can be ruled out from unwinding experiments. However, it is shown that the molecular plane of norfloxacin is near perpendicular relative to the DNA helix axis with a possibility of a bending of the DNA helix at the binding site.

The binding of 2,7-diazapyrene (DAP), N-methyl-2,7-diazapyrenium monocation (MDAP), and N,N‘-dimethyl-2,7-diazapyrenium dication (DMDAP) to calf thymus DNA has been studied with respect to molecular geometry and thermodynamics. It is concluded from flow linear dichroism (LD) and induced circular dichroism (CD) spectra that the three diazapyrenes bind by intercalation to alternating AT as well as GC polynucleotide duplexes, as indicated by strong interactions with the transitions of the nucleobases in conjunction with approximately perpendicular orientations of the in-plane symmetry axes relative to the DNA helix axis. The reduced LD (LDr = LD/Aiso) of the DNA complexes is characterized by marked fine structure, decreasing in the order DAP > MDAP > DMDAP. This finding is interpreted in terms of a microscopic heterogeneity associated with rotational mobility of the ligand in a tilted intercalation pocket, with the dication DMDAP having less rotational freedom than the neutral DAP has. Other distinct differe...

The electronic spectrum of the unnatural nucleic acid base 2-aminopurine is important in order to understand the spectroscopy and photophysics of the normal DNA bases. Knowledge of transition moment directions and effective absorption components is also a prerequisite for the interpretation of emission spectra of 2-aminopurine incorporated into DNA as a probe of nucleic acid base motion or excitation energy transfer processes. Using linear dichroism, fluorescence anisotropy, and ordinary and magnetic circular dichroism techniques, the near-UV spectrum of 2-aminopurine is resolved into contributions from five electronic transitions. Four moderately strong pi-->pi* transitions (I, III, IV, and V) are observed, polarized at +53 degrees (I), -72 degrees (III), -77 degrees (IV), and +57 degrees or -67 degrees (V) relative to the short molecular axis. In addition, one weak n --> pi* transition (II) polarized perpendicular to the molecular plane has been identified. Semiempirical INDO/S calculations support these assignments.

Linear dichroism (LD) has been used to probe the binding geometries of the diastereomeric adducts between DNA and the Delta and Lambda enantiomers of [Ru(phen)(2)L](2+) where the chelate ligand L is either dipyrido[3,2-a:2',3'-c]-phenazine (DPPZ) or its benzologue benzodipyrido[a:3,2-h:2'3'-j]phenazine (BDPPZ). By combined use of LD and emission anisotropy excitation spectra measured in highly viscous solution the absorption envelope of the metal-to-ligand charge transfer (MLCT) region has been fur the first time resolved in a complex of this type bound to DNA. The absorption can be described by four effective polarization directions (three almost orthogonal), along one of which any transition is polarized, assignments which are in good agreement with molecular orbital (INDO/S) calculations. The analysis of the LD spectrum provides the orientation of the complex in terms of several angles. Near perpendicularity relative to the DNA helix axis, found for the in-plane long and short axes of BDPPZ and DPPZ, together with extensive hypochromicity of the corresponding intraligand transitions, supports intercalation of the (B)DPPZ ligands between the DNA bases as earlier studies have indicated. MLCT transition moments that make an oblique angle to the molecular plane of the (B)DPPZ chromophores confirm this orientation and, in addition, provide strong evidence for a rotation (roll) of the complex around the pseudo-dyad axis, The roll is small (varying between 5 and 15 degrees), but significant, and has the same sign (clockwise) for both enantiomeric forms of the two complexes studied and shows only minor variations between calf-thymus DNA and alternating GC or AT homopolymer duplexes, It may reflect a property intrinsic of DNA (tilt of bases) or be a result of steric interference of the two phenanthroline ''propeller blades'' with a groove. The roll provides the first example of an angle in a DNA system that has been determined with its sign from LD spectroscopy.

Peptide nucleic acid (PNA) is a DNA analogue in which the negatively charged sugar phosphate backbone has been substituted by uncharged N-(2-aminoethyl)glycine units. The study of a PNA-DNA duplex and the corresponding DNA-DNA duplex gives a unique opportunity to compare two polyelectrolytes with virtually identical geometry but greatly different linear charge density. The results provide a basis for a study of the applicability of the Poisson-Boltzmann (PB) and counterion condensation (CC) theories. UV and circular dichroism spectroscopy as well as isothermal titration calorimetry (ITC) have been used to study the effect of different ions on the stability and conformation of PNA-DNA, PNA-PNA, and DNA-DNA duplexes having the same base sequences. Cations in general destabilize both antiparallel (N/3') and parallel (N/5') PNA-DNA duplexes whereas they stabilize the DNA-DNA duplex. Studies on the effect of monovalent salt such as NaCl on T-m were carried out over a wide range of salt concentrations (0.01 to 5 M). The decrease in the T-m of the N/3' PNA-DNA duplex with increasing ionic strength in the range of concentrations of 0.01 to 0.5 M, where electrostatic effects predominate, is explained in terms of counterion release upon duplex formation in contrast to the counterion association accompanying the formation of a DNA duplex. The uncharged PNA-PNA duplex shows no significant destabilization in this concentration range. The higher stability of the N/3' PNA-DNA compared to the DNA-DNA duplex (Delta Delta G similar to-7 kcal/mol) is ascribed to more favorable entropic contributions consistent with the counterion release that accompanies the PNA-DNA duplex formation. At high salt concentration (>1 M), where electrostatic contributions saturate, similar trends in the decrease in T-m, were observed for the three types of duplexes irrespective of their backbone charges. The destabilizing effects of a series of Na salts with various monovalent anions on N/3' PNA-DNA and PNA-PNA duplexes were found to follow the Hofmeister series, emphasizing the importance of the hydrophobic interaction between nucleobases for the stability of the PNA complexes in high salt concentration.

Homopyrimidine PNA can recognize a purine target of double-stranded DNA by local displacement of the DNA pyrimidine strand and formation of a PNA-DNA-PNA tripler. In order to elucidate the mechanism of this PNA/DNA recognition reaction we have studied the interaction of PNA-T oligomers with poly(da):poly(dT) as a model system. The PNA binding kinetics can be observed in situ by spectroscopy as a change in the circular dichroism spectrum. The overall reaction rate decreases with increasing ionic strength and proceeds in minutes at 50 mM NaCl. From the temperature dependence of the rate constants an activation energy of about 60 kJ/mol (at 50 mM NaCl) was estimated, consistent with the hypothesis that a rate limiting step could be the DNA opening frequency. As expected from the stability of a DNA double helix toward large-scale openings, also the activation energy of the strand displacement increases somewhat with increasing salt concentration. There is a nonlinear dependence of the overall rate constant of tripler formation on the PNA concentration with an exponent of 2 or somewhat higher. With PNA in excess, the CD signal shows an overshoot suggesting the presence of a third species. The CD spectrum of this intermediate indicates an ordered structure. The presence of intercalators ethidium bromide and 9-aminoacridine is found to increase the overall PNA binding rate, whereas minor groove binder DAPI and major groove binder methyl green both decrease the rate. The mechanism of strand invasion of duplex DNA by PNA is discussed in the light of these results.

Fluorescence and absorption spectroscopy, isothermal titration calorimetry, and viscosity measurements have been used to characterize the interaction of Delta and Lambda [Ru(phen)(2)DPPZ](2+) with calf thymus DNA. The method of continuous variations revealed two distinct binding stoichiometries for both Delta- and Lambda-DPPZ, corresponding to 0.7 and 3 mol of base pair/mol of ligand. Binding isotherms were obtained for the two enantiomers, both of which show strong binding to DNA, with K = 3.2 x 10(6) M(-1) bp and 1.7 x 10(6) M(-1) bp for the Delta and Lambda isomers, respectively, at 25 degrees C in solutions containing 50 mM NaCl. Titration calorimetry gave Delta H values of +0.3 kcal mol(-1) for Delta-DPPZ and +2.9 kcal mol(-1) for Lambda-DPPZ for their interaction with DNA. These small positive enthalpies, which were confirmed using thermal difference spectroscopy, indicated that the binding of these compounds to DNA is entropically driven. An enthalpy of +2.5 kcal mol(-1) was obtained for the binding of the parent compound, tris(phenanthroline)-Ru(II), to DNA. Titration of all three compounds into buffer gave a nonnegligible heat of dilution. The salt dependence of the binding constant was examined for both isomers. The slope SK = (delta logK/delta log[Na+]) was found to be 1.9 and 2.1 for the Delta and Lambda isomers, respectively. By using polyelectrolyte theory to interpret the observed salt dependence of the equilibrium constant, it can be shown that there is a significant nonelectrostatic contribution to the binding constant. Relative viscosity experiments showed that both Delta- and Lambda-DPPZ increase the length of rod-like DNA, in a manner consistent with binding by classical intercalation. Fluorescence energy transfer experiments provided additional evidence for the intercalation of Delta- and Lambda-[Ru(phen)(2)DPPZ](2+) into DNA.

Peptide nucleic acid (PNA) oligomers where one of the repeating backbone units is extended with a methylene group to either N-(2-aminoethyl)-beta-alanine or N-(3-aminopropyl)glycine were prepared. Alternatively, the linker to the nucleobase was extended from methylenecarbonyl to ethylenecarbonyl. The thermal stability of the hybrids between these PNA oligomers and complementary DNA oligonucleotides was significantly lower than that of the corresponding complexes involving unmodified PNA. However, the sequence selectivity was retained. Thymidyl decamers with all N-(2-aminoethyl)-beta-alanine or N-(3-aminopropyl)glycine backbones were prepared and shown to be unable to hybridize to the complementary (dA)(10) oligonucleotides, whereas a PNA decamer containing only ethylenecarbonyl linkers between the nucleobases and the N-(2-aminoethyl)glycine backbone showed weak but sequence-specific affinity for complementary DNA. All hybrids involving homopyrimidine PNA oligomers exhibited (PNA)(2)/DNA stoichiometry, whereas mixed-sequence PNA oligomers formed PNA/DNA duplexes. The preferred binding direction between the modified PNA and DNA in the duplex motifs was antiparallel, as previously reported for complexes involving unmodified PNA.

The modes of binding of the phenothiazinium dye methylene blue (1) to alternating and nonalternating polynucleotides and to calf thymus (CT) DNA have been characterized using linear dichroism (LD) and circular dichroism (CD) spectroscopy. With the polynucleotide [poly(dG-dC)](2) the interaction at low binding ratios is shown to be purely intercalative and the binding mode is insensitive to changes in ionic strength. The observed CD spectrum is bisignate, which may be due to intercalation at the different base-pair steps (5'G-C3' and 5'C-G3'), giving rise to CD signals of different sign and shape, By contrast, a single intercalative binding mode with the alternating AT polynucleotide [poly(dA-dT)](2) is likely only at very low ionic strength; at high ionic strength (200 mM phosphate, pH 6.9), a second binding mode is also manifest which is attributed to groove binding of the dye. The absorption and linear dichroism spectroscopic features of the methylene blue/CT-DNA (42% GC) complex reflect those of complexes of the dye with both [poly(dA-dT)](2) and [poly(dG-dC)](2); the circular dichroism spectrum of the methylene blue/CT-DNA complex and its variation with ionic strength reflect the complexity of even this simple system where numerous possible binding sites exist. Comparative binding to the nonalternating polynucleotides poly(dA).poly(dT) and poly(da).poly(dC), which each possess only one base-pair step, was also examined. On the basis of the combined LD and CD evidence, it is proposed that the dye is loosely bound with poly(dA).poly(dT), probably in the major groove, and intercalated with poly(dG).poly(dC).

Complexes of 4',6-diamidino-2-phenylindole (DAPI) with [poly(dG-dC)]2 and [poly(dI-dC)]2 were studied by optical spectroscopic techniques including linear dichroism (LD), circular dichroism (CD), and fluorescence measurements. The aim was to investigate the importance of the exocyclic amino group of guanine that protrudes into the minor groove of [poly(dG-dC)]2 but is absent in [poly(dI-dC)]2. When bound to [poly(dG-dC)]2, DAPI exhibits a negative, and strongly wavelength-dependent, reduced linear dichroism (LDr) in the DAPI absorption region, a weak positive CD, and a fluorescence behavior that is similar to that of free DAPI with accessibility to quenching by the aqueous solvent. These spectroscopic properties have been interpreted in terms of a major-groove binding geometry by Kim et al. (Kim, S. K.; Eriksson, S.; Kubista, M.; Norden, B. J. Am. Chem. Soc. 1993,115,3441-3447). By contrast, when bound to [poly(dI-dC)]2, DAPI exhibits a strong positive CD in the 300-420-nm region, a positive (wavelength-independent) LD(r), a strong increase of the fluorescence intensity, and shielding to added quencher. These spectroscopic properties closely resemble those of the DAPI-[poly(dA-dT)]2 complex, in which DAPI is situated deep in the minor groove of the polynucleotide. We conclude that the major groove binding geometry of DAPI when complexed with [poly(dG-dC)]2 is a result of a decreased affinity to the minor groove of [poly(dG-dC)]2 due to steric hindrance and decreased electronegative attraction caused by the amino group of guanine.

We have characterized the interaction of 4',6-diamidino-2-phenylindole (DAPI) with a series of DNA oligonucleotides containing a varying number of contiguous alternating AT base pairs in the center of a GC-stretch. The binding affinity of DAPI to the oligonucleotides increases with the number of contiguous AT base pairs. When entropy effects are taken into account, the microscopic affinity reaches a maximum value when the oligonucleotides contain three AT base pairs. Three is also the minimum number of base pairs that provides a binding site with all the characteristics of binding observed with DNA and [poly(dA-dT)]2 at low binding ratios. With none or only a single AT base pair in the center of a GC dodecamer, the spectral properties of bound DAPI are very similar to those of DAPI bound to [poly(dG-dC)]2. When DAPI is complexed with an oligonucleotide containing two AT base pairs, the absorption and fluorescence properties of DAPI closely resemble those of DAPI complexed with [poly(dG-dC)]2, while the circular dichroism spectrum resembles that of DAPI bound to [poly(dA-dT)]2.

The binding of an eightmer of peptide nucleic acid, H-T8-Lys-NH2 (=PNA-T8), to a polynucleotide, poly(dA), was studied by flow linear dichroism (LD) and circular dichroism (CD) spectroscopy. Whereas the single stranded DNA, due to its high flexibility, does not display any measurable LD signal when subjected to shear flow, the complex with PNA does. A titration shows that saturation occurs at a stoichiometry of two PNA thymine bases per DNA adenine base, indicating the formation of a triplex PNA2-DNA complex. The persistence length of the adduct remains small up to relatively high stoichiometries (above 1:1 T:A) indicating that no significant amounts of PNA:DNA duplex are formed. Instead triplex stretches seem to form surrounded by flexible parts of single stranded poly(dA). Upon approaching the stoichiometry 2:1 of T:A the LD increases dramatically demonstrating that the stiffness of the PNA-DNA triplex arises from base-base contacts preventing bending of the chain. It is also inferred that the main stiffness of duplex DNA very probably has a similar origin and is not primarily a result of the increased phosphate-phosphate repulsion. Circular dichroism spectra support the conclusion that a triplex is formed as the only PNA-DNA complex and that it is a right-handed helix. The wavelength dependence of the reduced linear dichroism shows that the inclination of the bases from perpendicularity relative to the helix axis is small. The base conformation of the poly(dA)[PNA-T8]2 triplex is very similar to that of the conventional poly(dA)[poly(dT)]2 triplex.

Linear dichroism (LD) spectroscopy and steady-state as well as time-resolved luminescence spectroscopy have been used to investigate the interaction of the DELTA and LAMBDA enantiomers of Ru(phen)2DPPZ2+ (phen = 1,10-phenanthroline; DPPZ = dipyrido[3,2-a:2',3'-c]phenazine) with DNA. The pure enantiomers, which were difficult to separate by traditional resolving methods, were synthesized via a chiral precursor. Changes in luminescence, isotropic absorption and excited state lifetimes upon binding, and the LD observed in flow-oriented DNA systems provide detailed information about the DNA binding of the enantiomers. Flow LD shows that both enantiomers bind to DNA in a well-defined manner with an orientation of the dipyridophenazine chromophore consistent with intercalation of this moiety between base-pairs. Both enantiomers are found to show luminescence in the presence of DNA to which they bind very strongly (K almost-equal-to 10(8) M-1); however, the relative luminescence quantum yield of the bound DELTA enantiomer is 6-10 times larger than that of the bound LAMBDA enantiomer. Furthermore, for each enantiomer two distinct excited state lifetimes are found in varying proportions depending on the binding ratio. The large difference in luminescence quantum yield between the enantiomers is interpreted in terms of slightly different intercalation geometries of the dipyridophenazine ligand, resulting in different protections from quenching by solvent water and diastereomeric differences in the interactions between enantiomers bound in contigue on DNA.

The interactions of DAPI with poly[d(G-C)2] and poly[d(G-m5C)2] are characterized by optical spectroscopic methods. DAPI is found to bind to poly[d(G-C)2] with its molecular long-axis roughly perpendicular but with its plane more parallel to the belix axis of the polynucleotide, probably being bound in a non-intercalated way in the major groove of the polynucleotide. In poly[d(G-m5C)2], DAPI is bound with its molecular plane perpendicular to the helix axis, probably partially intercalated from the major groove. The conclusions are supported by fluorescence quenching results. Together with the earlier established minor groove binding mode of DAPI in the AT regions, our results demonstrate a heterogeneous nature of the interaction of a small ligand with DNA. We conclude that the conventional classification of a ligand as being either intercalator, minor groove binder, or major groove binder may not be adequate since the mode of binding clearly depends also on the base sequence.

Covalent adducts of the carcinogenic polycyclic aromatic hydrocarbon (+)-anti-benzo[a]pyrene 7,8-dihydrodiol 9,10-epoxide ((+)-anti-BPDE) in polynucleotides have been studied by fluorescence spectroscopy. The pyrenyl chromophores of the BPDE adducts, linked by the C10 atom to the exocyclic nitrogen of guanine, interact in the photoexcited state, as evidenced by excimer fluorescence. Strong BPDE excimer fluorescence is observed in the alternating poly(dGdC).poly(dGdC) sequence, whereas it is weak in the homopolymeric poly(dG).poly(dC) and in calf thymus DNA. No excimer fluorescence is observed for the BPDE adducts in poly(dAdC).poly(dGdT) or poly(dAdG).poly(dCdT). It is concluded that the formation of BPDE excimers in polynucleotides requires binding to guanines on different strands on consecutive basepairs. The experimental results are supported by graphics modeling and energy minimization of BPDE adducts in various oligonucleotide sequences. The results show that the most favorable arrangement for excimer formation of the BPDE-dG adducts is in a 5'(dCdG-BPDE).5'(dCdG-BPDE) sequence, where the pyrenyl chromophores interact in the minor groove.

Measurements of UV linear dichroism on purine and three methyl derivatives partially oriented in poly(vinyl alcohol) matrix gave direct evidence for the assignment of the first singlet npi* state. Intensity distributions and moment directions for the first three pi --> pi* transitions were also determined. The pi --> pi* transitions in purine were found to be polarized at (angles, relative to the pseudo-symmetry long axis, counted positive in the N7 direction): -31-degrees +/- 5-degrees (II at 265 nm), +38-degrees +/- 5-degrees (III at 244 nm), and +36-degrees +/- 10-degrees (IV at 214 nm). The transition energies and moment directions were not markedly perturbed by methyl substitution at the sixth, seventh, or ninth position. Therefore, these methyl substituents could be used as orientational perturbers to resolve a sign ambiguity problem regarding transition moment directions. The orientation were determined by infrared dichroic measurements using both in-plane and out-of-plane polarized vibrational transitions. In addition, the phosphorescence spectra were studied, including phosphorescence anisotropy, phosphorescence lifetimes, and quantum yields, for the purines in an organic glass at 80 K. Based on these measurements, the lowest triplet state is concluded to have effectively pipi* character, and its emission allowedness appears to originate from spin-orbit interactions primarily with singlet sigmapi* states but also with singlet pipi* states via vibronic mixing. The phosphorescence emission spectra of purine and 6-methylpurine are complex, compared to 7-methylpurine and 9-methylpurine, with emission wavelength-dependent lifetimes and excitation spectra. This is ascribed to a prototropic tautomeric equilibrium between the 7H and 9H forms of purine and 6-methylpurine, a ground-state heterogeneity that we believe has caused confusion in earlier studies and, e.g., led to an assignment of the phosphorescence origin of purine.

The interaction of the uranyl(VI) ion (UO22+) with DNA and its light-induced cleavage of DNA has been studied using flow-linear dichroism and P-32-endlabeled oligonucleotides. It was found that binding of uranyl ion to DNA is a prerequisite for photocleavage; from run-off experiments the binding constant was estimated to be of the order of 10(10) M-1 at pH 4. The angular orientation of the [O=U=O]2+ chromophore is consistent with binding by bridging phosphate groups on opposite strands of the minor groove of DNA; at higher DNA concentration aggregation indicates intermolecular bridging as well. The uranyl-mediated photocleavage of DNA is not influenced by the presence of O2, is more efficient at low pH (

In this work we examine the trapping and conversion of visible light energy into chemical energy using a supramolecular assembly. The assembly consists of a light-absorbing antenna and a porphyrin redox center, which are covalently attached to two complementary 14-mer DNA strands, hybridized to form a double helix and anchored to a lipid membrane. The excitation energy is finally trapped in the lipid phase of the membrane as a benzoquinone radical anion that could potentially be used in subsequent chemical reactions. In addition, in this model complex, the hydrophobic porphyrin moiety acts as an anchor into the liposome positioning the DNA construct on the lipid membrane surface. The results show the suitability of our system as a prototype for DNA-based light-harvesting devices, in which energy transfer from the aqueous phase to the interior of the lipid membrane is followed by charge separation.

Binuclear ruthenium complexes that bind DNA by threading intercalation have recently been found to exhibit an exceptional kinetic selectivity for long polymeric adenine-thymine (AT) DNA. A series of oligonucleotide hairpin duplexes containing a central tract of 6-44 alternating AT base pairs have here been used to investigate the nature of the recognition mechanism. We find that, above a threshold AT tract length corresponding to one helix turn of B-DNA, a dramatic increase in threading intercalation rate occurs. In contrast, such length dependence is not observed for rates of unthreading. Intercalation by any mechanism that depends on the open end of the hairpin was found not to be important in the series of oligonucleotides used, as verified by including in the study a hairpin duplex cyclized by a copper-catalyzed "click" reaction. Our observations are interpreted in terms of a conformational pre-equilibrium, determined by the length of the AT tract. We finally find that mismatches or loops in the oligonucleotide facilitate the threading process, of interest for the development of mismatch-recognizing probes.

The orientation and conformation of the cell-penetrating peptide "penetratin" associated with phospholipid vesicle membranes has been determined using polarized-light spectroscopy. The magnitude of orientation of penetratin is unprecedented for a solute in our membrane system, which we believe indicates a strong, specific interaction with the membrane. To validate the spectroscopic technique for studying the orientation of the two tryptophan residues of penetratin, we applied tryptophan octyl ester as a model compound. It is found to be incorporated in the membrane and preferentially oriented with its hydrophobic benzene edge of the indole chromophore pointing into the membrane and its hydrophilic groups oriented toward the water. For penetratin, the results indicate that a central alpha-helical part of the peptide is aligned parallel with the membrane surface, while the ends of the peptide adopt a planar structure. The planes of the two tryptophan side chains show a preferred orientation parallel with the membrane surface, indicating that they are not inserted into the membrane.

Detergent sequestration using micelles as a hydrophobic sink for dissociated drug molecules is an established technique for determination of dissociation rates. The anionic surfactant molecules are generally assumed not to interact with the anionic DNA and thereby not to affect the rate of dissociation. By contrast, we here demonstrate that the surfactant molecules sodium dodecyl sulfate (SDS), sodium decyl sulfate, and sodium octyl sulfate all induce substantial rate enhancements of the dissociation of intercalators from DNA. Four different cationic DNA intercalators are studied with respect to surfactant-induced dissociation. Except for the smallest intercalator, ethidium, the dissociation rate constants increase monotonically with surfactant concentration both below cmc and (more strongly) above cmc, much more than expected from electrostatic effects of increased counterion concentration. The rate enhancement, most pronounced for the bulky, multicationic, hydrophobic DNA ligands in this study, indicates a reduction of the activation energy for the ligand to pass out from a deeply penetrating intercalation site of DNA. The discovery that surfactants enhance the rate of dissociation of cationic DNA-intercalators implies that rate constants previously determined by micelle-sequestered dissociation may have been overestimated. As an alternative, more reliable method, we suggest instead the addition of excess of dummy DNA as an absorbent for dissociated ligand.

We here report a remarkably slow rearrangement of binding modes for a binuclear ruthenium(II) complex upon interaction with DNA. It has been previously shown that Delta,Delta-[mu-(11,11'-bidppz)(phen)4Ru2]4+ binds to DNA in one of the grooves. However, we find that this is only an initial, metastable, binding mode, which is extremely slowly reorganized into an intercalative binding geometry. The slow rearrangement and dissociation, revealed by flow linear dichroism and fluorescence spectroscopy, are concluded to be a result from the complex being threaded through the DNA, with one of the bridging aromatic dppz ligands intercalated between the base pairs of the DNA, placing one metal center in the minor groove and one in the major groove. A negative LD, a high luminescence quantum yield, and long luminescence lifetimes, similar to the intercalating complex Delta-[Ru(phen)2dppz]2+, indicate intercalation of the bidppz moiety. The unique slow dissociation of the complex in its final DNA-binding mode suggests that this class of threading, partially intercalated binuclear complexes may be interesting in the context of cancer therapy. Also, their unique optical and photophysical properties could make such complexes, either alone or scaffolded by DNA structures, of interest for the development of nanometer-sized molecular optoelectronic devices.

The interaction between DNA and a series of mono-, bis-, tris-, tetrakis-, and hexakis-intercalating 9-aminoacridines has been studied with flow linear dichroism (LD), circular dichroism (CD), electric orientation relaxation (EOR) techniques, and with viscometry and equilibrium analyses. The orientation of the 9-aminoacridine ligand relative to the average orientation of the DNA bases, measured by LD, shows that with both 9-aminoacridine and the bis(acridines) the in-plane short axes of the acridine ligands are oriented perfectly parallel to the planes of the DNA bases, as expected for classical intercalation, whereas the long axes are found to be significantly tilted. This is supported by the DNA lengthening measured by EOR, which for 9-aminoacridine is 1.5 base-pair units, compared with 1.0 for ethidium bromide. Also in case of the tris(acridines) LD, CD, viscometry, and equilibrium data indicate that all acridine ligands are intercalated. The binding analysis shows an increasing degree of cooperativity in the sequence 9-aminoacridine < bis(acridines) < tris(acridines), and the corresponding binding densities, 4, 8, and 11–14, respectively, are in good agreement with those expected from the nearest-neighbor exclusion principle. The LD and CD measurements show that the tetrakis- and hexakis(acridines), despite long and flexible links, bind to DNA with only three of the acridine ligands intercalated.