Binding of BOBO-3 Intercalative Dye to DNA ... - ACS Publications

In this work, we describe the differences in the binding behavior of BOBO-3 with double-stranded DNA (dsDNA) having different base contents. The fluor...
0 downloads 0 Views 2MB Size
J. Phys. Chem. B 2010, 114, 6713–6721

6713

Binding of BOBO-3 Intercalative Dye to DNA Homo-Oligonucleotides with Different Base Compositions Maria J. Ruedas-Rama, Jose M. Alvarez-Pez, Jose M. Paredes, Eva M. Talavera, and Angel Orte* Department of Physical Chemistry, Faculty of Pharmacy, UniVersity of Granada, Campus Cartuja, 18071 Granada, Spain ReceiVed: February 04, 2010; ReVised Manuscript ReceiVed: March 26, 2010

The interactions between trimethine cyanine homodimer dye, BOBO-3 (1,1′-(4,4,7,7-tetramethyl-4,7diazaundecamethylene)-bis-4-[3-methyl-2,3-dihydro-(benzo-1,3-thiazole)-2-methylidene]-pyridinium tetraiodide), and single-stranded homo-oligonucleotides, double-stranded complementary homo-oligonucleotides, and high-molecular-weight double-stranded polyhomonucleotides have been investigated in detail using absorption and both steady-state and time-resolved fluorescence spectroscopy. In this work, we describe the differences in the binding behavior of BOBO-3 with double-stranded DNA (dsDNA) having different base contents. The fluorescence intensity of BOBO-3 interacting with deoxyadenosine-deoxythymidine (dAdT) dsDNA was higher than with the deoxyguanosine-deoxycytidine (dGdC) double helix. However, the BOBO-3 lifetime was longer in dGdC-rich dsDNA than in dsDNA with many dAdT sites. This result was detected at both the ensemble level and the single-molecule level. This behavior is a consequence of the dye’s interacting with dsDNA on two kinds of binding sites. This phenomenon also occurs in natural dsDNA (Ruedas-Rama, M. J.; Orte, A.; Crovetto, L.; Talavera, E. M.; Alvarez-Pez, J. M. J. Phys. Chem. B 2010, 114, 1094-1103). By using a time-resolved fluorescence methodology and the McGhee-von Hippel theory for two overlapping, noncooperative binding modes, we obtained the equilibrium binding constants and the number of occupied sites for each binding mode of BOBO-3 in dAdT and dGdC binding sites. BOBO-3 has a higher affinity for dAdT sites and occupies 4.0 ( 1.0 sites in its primary binding mode, whereas in dGdC-rich double strands, BOBO-3 covers 6.2 ( 1.1 sites and has a lower affinity. These differences in the binding features and spectral properties of BOBO-3 may be used to develop approaches to identify GC- or AT-rich regions within large strands of dsDNA. Introduction Dimeric cyanine dyes are a family of fluorophores used in nucleic acid staining, both because these dyes cause large fluorescence enhancements when bound to double-stranded DNA (dsDNA) and because of their high affinity for dsDNA. They form complexes that remain stable even during electrophoresis. The two more popular cyanine dyes are the oxazole yellow homodimer YOYO-1 and the thiazole yellow homodimer TOTO-11,2 because they have a higher affinity for dsDNA than do other cyanine dyes.3 The fluorescence emission is enhanced by dsDNA binding because rotation between the two heterocyclic, chromophoric groups about the central methine bridge is prevented, decreasing nonradiative relaxations.4-6 These dyes have been used quantitatively to measure the total dsDNA content in different types of samples because the binding is sizeindependent over a broad size range from 600 to 24 000 base pairs.1,2,7,8 A moderate dependence on dsDNA base content has also been found, and the binding strength is thought to be sequence-dependent in small dsDNA fragments and oligonucleotides.9-12 The base content and sequence have different effects on the conformation of the helical structure13-15 and create different chemical environments on the minor and major grooves of dsDNA. These grooves are used by DNA-binding proteins and * Corresponding author. Phone: +34 958 243825. Fax: +34 958 244090. E-mail: [email protected].

enzymes as recognition sites.16 These differences also affect the binding behavior of intercalative dyes. One of the most extensive studies on the base content dependence of emission enhancements, quantum yields, and lifetimes for 10 monomeric and bichromophoric cyanine dyes was reported by Netzel et al.6 The differences in emission enhancement as a function of the base content for cyanine dyes complexed to dsDNA can be caused by different types of binding sites in deoxyadenosine-deoxythymidine (dAdT) and deoxyguanosine-deoxycytidine (dGdC) sequences and by different degrees of flexibility of the double helix.17 Excited-state electron transfer (ET) quenching by nucleosides could also alter the emission enhancement.18 ET between guanosine nucleosides is favored because this nucleoside is the easiest to oxidize.19,20 Netzel and colleagues also showed that the excited-state lifetimes of cyanine dyes are multiexponential and dependent on the composition of the dsDNA.6 Similar findings have been reported by recent singlemolecule fluorescence studies using average decay-time distributions. These studies showed that TOTO intercalated into poly dAdT dsDNA has an average fluorescence lifetime of 1.8 ns, whereas TOTO intercalated into poly dGdC dsDNA shows an average lifetime of 2.2 ns.21 This small lifetime difference has been used at the single-molecule level to estimate the GC content of oligonucleotides with intermediate GC compositions to within a few percent error21 and to determine the relative binding affinity of TOTO for different dsDNA fragments. TOTO’s binding was concluded to be sequence-independent.22

10.1021/jp1010742  2010 American Chemical Society Published on Web 04/23/2010

6714

J. Phys. Chem. B, Vol. 114, No. 19, 2010

Ruedas-Rama et al.

TABLE 1: Sequences of the Tested DNA 5′-AAA AAA AAA AAA AAA AAA AAA AAA AAA AAA-3′ 5′-TTT TTT TTT TTT TTT TTT TTT TTT TTT TTT-3′ 5′-CCC CCC CCC CCC CCC CCC CCC CCC CCC CCC-3′ 5′-III III III III III III III III III III II(I)1-20-3′ 5′-GCC CGC CCG CCC GCC CGC CCG-3′ 5′-CGG GCG GGC GGG CGG GCG GGC-3′ 5′-CTA CTA CAA CTG GAA GAC CGG GAA GTC CGA GAA GTG CAT CTT CTG C-3′ 5′-GCA GAA GAT GCA CTT CTC GGA CTT CCC GGT CTT CCA GTT GTA GTA G-3′

In a recent paper, we reported an investigation of the photophysics and intercalation properties in small fragments of DNA of the homodimeric cyanine dye BOBO-3 (1,1′-(4,4,7,7tetramethyl-4,7-diazaundecamethylene)-bis-4-[3-methyl-2,3-dihydro-(benzo-1,3-thiazole)-2-methylidene]-pyridinium tetraiodide).23 We chose BOBO-3 because it is a good acceptor for fluorescence resonance energy transfer (FRET)-based hybridization detection methods when the energy donor is a xanthenic dye for labeling DNA.24 As with other members of this family of intercalators, the main photophysical feature of the dye is an extensive enhancement of the fluorescence emission upon intercalation into dsDNA. However, the interaction of BOBO-3 with DNA is not merely reduced to a simple intercalation mode. We have shown that BOBO-3 interacts with single-stranded DNA (ssDNA) in a different binding mode that modifies the spectral properties of the dye. This binding mode coexists and competes with the clamp-like bis-intercalation mechanism. We also developed a time-resolved fluorescence methodology to quantify the contributions of BOBO-3’s interacting with DNA in each binding mode. This method allows the direct assessment of the affinity equilibrium constants and the number of sites covered by each BOBO-3 binding mode according to the McGhee-von Hippel theory.25 The low-affinity mode covers about half as many of the base pairs as the bis-intercalation, supporting the idea of a partial intercalation. In this paper, we explore the base-content dependence of the BOBO-3 photophysical behavior in its interactions with ssDNA and dsDNA homonucleotides of different base contents and lengths. We investigate whether the different affinities of BOBO-3 for AT- or GC-rich regions in dsDNA may provide advantages for the development of base-content determination methodologies based on either ensemble or single-molecule fluorescence. We calculate the fractional population, the affinity equilibrium constants, and the number of sites covered by each BOBO-3 binding mode according to the McGhee-von Hippel theory.25 We investigate the mechanisms underlying the fluorescence enhancement and the relation between the fluorescence quantum yield and the fluorescence lifetimes recovered and address the question of whether these processes are sensitive to the base content of the nucleic acid. Experimental Methods Reagents. All experiments were performed using chemicals of analytical reagent grade and Milli-Q water. Potassium salts of polymeric polycytidylic acid, polyinosinic acid, polyadenylic acid, and polythymidylic acid and the buffer salt of Tris, sodium chloride, and EDTA were obtained from Sigma-Aldrich (Spain). The pH of the solutions and buffers was adjusted using diluted spectroscopic-grade NaOH and HCl (Aldrich, Spain) dissolved in Milli-Q water. All of the chemicals were used as-received without further purification, and the stock solutions were protected from sunlight and kept at about 4 °C in a refrigerator. In all staining procedures, the BOBO-3 iodide stock solution

(dA)30 (dT)30 (dC)30 (dI)30-50 (dCG)21 (dGC)21 dsDNA (46 bp)

(Invitrogen, Carlsbad, CA) was freshly diluted in a buffer solution with a pH of 7.5 that contained 10 mM of Tris, 1 mM of EDTA, and 100 mM of NaCl (TEN buffer). In the intercalation experiments, BOBO-3 was added to the DNA solutions and incubated for 10 min in the dark at 25 °C. Oligonucleotides. The chemically synthesized homonucleotide and other sequences of oligonucleotides were obtained from IBA Technologies (Germany). Polyadenosine, polythymidine, and polycytidine oligonucleotides were synthesized with 30 nucleotides. It is not possible to make, purify, and dissolve polyguanosine oligonucleotides because a guanine higher-order structure, guanine quadruplex, is formed during the synthesis.26,27 To characterize the interactions of deoxyguanosine and the pair dGdC with BOBO-3, a mixed oligonucleotide was ordered. This mixed oligonucleotide was a C(GGGC)5 strand together with its complementary strand. The synthesis of long polyinosine oligonucleotides is also complicated, and the number of inosine residues left after purification ranged between 30 and 50. The sequences of all DNA employed in this work are shown in Table 1. The DNA was purified by double HPLC and dissolved in TEN buffer. All subsequent dilutions were prepared in the same buffer. The ssDNA sequences were annealed with their corresponding complementary sequences to form the respective dsDNA. The hybridization was performed by heating at 75 °C for 5 min, followed by slow cooling to room temperature. The stock ssDNA concentration was verified by absorption measurements at 260 nm. The oligonucleotide solutions used in the measurements of the spectral properties of BOBO-3 ranged between 25 and 500 nM in dsDNA or ssDNA. Instruments. Absorption spectra were recorded at 25 °C using 5- × 10-mm cuvettes with a Perkin-Elmer Lambda 650 UV/vis spectrophotometer equipped with a Peltier temperaturecontrolled cell holder. Steady-state fluorescence emission spectra were collected at 25 °C on a JASCO FP-6500 spectrofluorometer using a 450-W xenon lamp as the excitation source and an ETC273T temperature controller. Fluorescence decay traces were recorded at emission wavelengths of 590, 600, 610, and 620 nm in time-correlated singlephoton counting (TCSPC) mode along 36-ps channels using a FluoTime 200 fluorometer with a TimeHarp 200 collection card (PicoQuant GmbH, Germany). The excitation source was a fiber-coupled 485-nm pulsed laser (LDH-P-C-485 from PicoQuant GmbH) with a 20-MHz repetition rate. The fluorescence was collected at a 90° geometry after crossing a polarizer set at the magic angle (54.7°) and using a 2-nm bandwidth monochromator. Histograms of the instrument response functions (obtained using a LUDOX scatterer solution) and sample decay traces were recorded until they reached 2 × 104 counts in the peak channel. The full width at half-maximum of the laser pulse was ∼80 ps. The four decay traces of each sample were globally analyzed by a least-squares minimization deconvolution method in terms of multiexponential functions using FluoFit software (PicoQuant GmbH). The decay times were treated as shared

Binding of BOBO-3 Intercalative Dye to DNA

J. Phys. Chem. B, Vol. 114, No. 19, 2010 6715

parameters, and the pre-exponential factors were treated as local adjustable parameters. The shortest lifetime was kept fixed to the lifetime of BOBO-3 in solution, 0.173 ns, because this value is close to the instrumental response. The reduced chi-squared value, χ2; the weighted residuals; and the autocorrelation functions were indicators of the goodness of fit. Fluorescence fluctuation traces and decays at the singlemolecule level were recorded with the confocal inverted microscope-based system MicroTime 200 (PicoQuant GmbH). The excitation source was the same fiber-coupled 485-nm pulsed laser directed into an Olympus IX71 inverted microscope by a 510DCXR dichroic filter. The fluorescence was focused onto the sample and collected using an Olympus UPlanSApo 100×/ 1.40-NA oil-immersion objective. After passing through an HQ500LP cutoff filter and a 50-µm pinhole, the fluorescence photons were directed into a single-avalanche photodiode detector (SPCM AQR-14, Perkin-Elmer Optoelectronics) by a 590DCXR dichroic mirror and a D630/30 m band-pass filter. All of the filters and dichroic mirrors were from Chroma Technology (Bellows Falls, VT). Data collection and handling was performed using a TimeHarp200-equipped computer. Measurements were taken in time-tagged time-resolved (TTTR) mode, which records the arrival times of individual photons.28 The TTTR method allows the reconstruction of the whole fluorescence decay traces from fluorescence fluctuation traces, as well as performing burst integrated fluorescence lifetime (BIFL) analysis. BIFL analysis provides estimation of the lifetimes associated with each single-molecule burst, selected by applying an adequate threshold. The data analysis was carried out using SymphoTime software (PicoQuant GmbH). We recorded the fluorescence fluctuation traces of BOBO-3 (10 nM) in the presence of 100 nM of the different dsDNA oligonucleotides after a 10-fold dilution of the stock solution in TEN buffer. Methods of Analysis. Apparent quantum yield (Φ) values of BOBO-3 free in solution and of BOBO-3 intercalated into homonucleotide dsDNA ((dTdA)30, (dCdI)30-50, and (dC-dG)21) were calculated using eq 1 with rhodamine B in ethanol as a reference (Φ ) 0.31),29

Φ ) ΦR

I ODR n2 IR OD n2

(1)

R

where I is the integrated intensity, OD is the optical density, and n is the refractive index. The subscript R stands for the corresponding known values of the reference fluorophore. The mole fractions, xi, of BOBO-3 in each binding mode and free in solution were obtained from the pre-exponential factors, Ai, recovered from fits of the decay traces. The values of xi were determined by applying conversion factors that account for the differential excitation rates of the distinct dye species and the different contributions at each emission wavelength of the three species (free dye, intercalated dye, and externally bound dye). These mole fractions (xi) are given by

xi )

Ai /(kiλem × εiλex)

∑ Ai/(kiλ

em

× εiλex)

(2)

i

where ελi ex is the extinction coefficient at the excitation wavelength λex, and kλi em stands for the relative steady-state fluorescence signal of species i at the emission wavelength λem. The

procedure to estimate ελi ex and kλi em for each form of BOBO-3 and the obtained values is given in the Supporting Information section and Tables SI 1 and SI 2. The recovered mole fractions, xi, allow the determination of the binding affinity constants, Ki, and the number of binding sites occupied for each mode, ni, for different DNA compositions. This information was used to study the differences in the interaction behavior of BOBO-3 with dAdT and dCdG pairs. The determination of Ki and ni was performed by means of a global nonlinear regression model via minimization of the weighted residuals. This method was based on the McGhee-von Hippel equations25,30 and was coded using MathCad 14.0 (PTC, Needham, MA). A more detailed explanation of the code is provided elsewhere,23 and more details of the fitting procedure are described in the Supporting Information. We also performed a time-gated BIFL analysis on the singlemolecule fluorescence fluctuation traces. First, the overall fluorescence decay histogram was reconstructed from the whole fluorescence fluctuation trace. Second, a time gate was applied. Only photons that arrived at the detectors between 1.3 and 12.3 ns after the excitation pulse were used to rebuild the fluorescence trace. This ensures that only the fluorescence photons are accounted for, discarding photons from scattered light, background noise, and fluorescence from free BOBO-3. Third, we set a threshold of 10 times the background noise for burst selection. Finally, the BIFL analysis extracts the lifetime(s) associated with each fluorescence burst on the basis of a maximum likelihood estimation method. The frequency histograms of the recovered lifetimes allow exploration of lifetime distributions from individual single-molecule fluorescence bursts. Results and Discussion Absorption and Fluorescence Spectroscopy. The absorbance of BOBO-3 in Tris buffer with a pH of 7.5 showed an asymmetric band at 534 nm and a weak vibronic shoulder on its red edge. These data suggest the formation of intermolecular dimers by self-association that was previously reported in aqueous buffers.6 In the presence of double-stranded homonucleotides of different base compositions, the absorption band of BOBO-3 was red-shifted. The absorbance spectra of BOBO-3 bound to (dTdA)30, (dCdI)30-50, and (dCdG)21 have nearly identical bandshapes with maxima around 573 nm, which is also the absorption maximum of BOBO-3 intercalated into heteronucleotide dsDNA (46 bp) (Figure 1A). None of the bandshapes for these BOBO-3/dsDNA complexes show doublebanded structures. These bandshapes look like monomeric absorbances,6 consistent with each end of the dye binding separately to the DNA duplex. At high cyanine dye/dsDNA bp ratios, external binding becomes noticeable,23,31 and therefore, for these experiments, the dye/dsDNA bp ratios were kept no larger than 1:10. As previously reported, BOBO-3 interacts with ssDNA, generally by showing a diminished extinction coefficient. Nevertheless, the observed shifts of the absorption bands of strands with different base compositions were noticeably dependent on the nucleoside component. For instance, BOBO3/(dA)30 ssDNA mixtures showed an absorption maximum at 560 nm and a large shoulder at the maximum of the free dye (Figure 1B). In addition, a small band appeared at 458 nm. Similarly, when BOBO-3 was mixed with poly-dThymidine ((dT)30), the resultant spectra showed a higher contribution of free BOBO-3 and a shoulder around 560 nm. These features indicate that the interactions of BOBO-3 with (dA)30 or (dT)30 ssDNA were almost negligible, and the BOBO-3 was present primarily in solution.

6716

J. Phys. Chem. B, Vol. 114, No. 19, 2010

Ruedas-Rama et al.

Figure 1. The normalized (A) absorbance and (B) emission spectra of BOBO-3 free in solution (black), BOBO-3/(dTdA)30 dsDNA (red), BOBO3/(dCdI)30-50 dsDNA (blue), BOBO-3/(dCdG)21 dsDNA (green), and BOBO-3/dsDNA (46bp) (pink). All of the mixtures were made in a ratio of 2 BOBO-3 to 1 dsDNA. (C) The normalized absorbance spectra of (dA)30 (black), (dT)30 (red), (dC)30 (blue), and (dI)30-50 ssDNA (green). (D) The emission intensity at 600 nm of BOBO-3 bound to the following homonucleotide dsDNA (5 × 10-7 M) at different ratios of BOBO-3/bp: (dTdA)30 (black), (dCdI)30-50 (red), and (dCdG)21 (green). The excitation wavelength was 573 nm.

A different behavior was found with the other homooligonucleotides. When the added strand was polyinosine ((dI)30-50 ssDNA), the maximum absorption of the dye shifted to 563 nm, and the band shape became narrower and more similar to the band shape of BOBO-3 bound to dsDNA. This result shows a lower contribution of the dye in solution than in the presence of (dA)30 or (dT)30 ssDNA. The most striking case was BOBO-3 in the presence of (dC)30 ssDNA. The absorption spectra of these mixtures showed almost no shift of the free BOBO-3 absorption maximum, but a large new band with a maximum at 464 nm was observed together with a large decrease in the extinction coefficient of the dye. This new band is similar to the one detected in the BOBO-3/(dA)30 system, but this band is much more prominent. The formation and features of this absorption band are currently under investigation and will be reported elsewhere. This band was not detected in dsDNA and did not show any fluorescence properties. Therefore, this band is not relevant in the present study. The BOBO-3 dye strongly fluoresces upon interaction with dsDNA, like other members of the family, because of an intercalation mechanism.3,23 The BOBO-3 in solution showed a low emission intensity with a weak band at 590 nm (quantum yield around 0.001). However, a remarkable enhancement of the fluorescence was detected upon addition of (dTdA)30, (dCdI)30-50, and (dCdG)21 dsDNA, and a concomitant redshift of the emission maximum occurred at 597, 599, and 602 nm, respectively (Figure 1C). The BOBO-3/(dCdG)21 dsDNA was the only complex that had the same emission maximum as the BOBO-3 intercalated into heteronucleotide dsDNA (46 bp). For

a BOBO-3/dsDNA bp ratio of 1:10, the fluorescence enhancement was 50-fold for (dTdA)30, 45-fold for (dCdI)30-50, and 20fold for (dCdG)21 dsDNA. The spectral features of the complexes between the BOBO-3 and the different dsDNA types are shown in Table 2. The fluorescence enhancement of the BOBO-3 in the (dCdG)21 dsDNA was smaller than in (dTdA)30 and (dCdI)30-50 dsDNA. The differences in the gradient of fluorescence enhancement were detected by performing fluorimetric titrations of each dsDNA system ((dTdA)30, (dCdI)30-50 and (dCdG)21 at 5 × 10-7 M) in TEN buffer at a pH of 7.5 and with increasing concentrations of BOBO-3 (Figure 1D). A linear increase of the emission intensity was detected at low dye concentrations, and a plateau was gradually reached. This trend is similar to the curves previously described for natural dsDNA.23 At low mixing ratios of BOBO-3/dsDNA, almost all of the dye present in solution is intercalated into the DNA double helix. The intercalation protects the dye from the solvent quenching and provides a higher fluorescence quantum yield. Low ratios were used for our calculations of the quantum yields to favor this binding mode (Table 2). Figure 1D shows very similar curves for the (dTdA)30 and (dCdI)30-50 systems and a reduced fluorescence intensity for BOBO-3 in the presence of (dCdG)21. An approximation of the saturation point of BOBO-3 can be obtained from the titrations. The saturation point was around a ratio of 0.15 BOBO-3/dsDNA bp in each case. However, the saturation number estimated using this method shows a dependence on the total dsDNA concentration.23 Expressing the number of sites covered by an intercalator dye in these terms

Binding of BOBO-3 Intercalative Dye to DNA

J. Phys. Chem. B, Vol. 114, No. 19, 2010 6717

TABLE 2: Summary of Some Spectral Characteristic of BOBO-3 in Solution and Interacting with Different Types of dsDNA

max absorbance (nm) max emission (nm) quantum yielda lifetimes (ns)

BOBO-3 in solution

(dTdA)30 dsDNA/ BOBO-3

(dCdI)30-50 dsDNA/ BOBO-3

(dCdG)21 dsDNA/ BOBO-3

dsDNA (46bp)/ BOBO-3

534 590 (1.0 ( 0.1) × 10-3 0.173 ( 0.009

573 597 0.074 ( 0.003 2.99 ( 0.09 1.86 ( 0.10 0.173b

573 599 0.055 ( 0.006 3.32 ( 0.09 1.71 ( 0.07 0.173b

572 602 0.037 ( 0.004 4.02 ( 0.06 1.82 ( 0.13 0.173b

572 602 0.072 ( 0.006 4.01 ( 0.04 1.65 ( 0.04 0.173b

a Calculated quantum yields represent average values. They were determined for different mixtures BOBO-3/dsDNA within the range 1:50-1:1 (where the intercalated form was favored). b Because the value of the shortest lifetime is close to the instrumental response, it was kept fixed for the fitting to the value experimentally obtained for free BOBO-3 in solution (in the absence of DNA).

requires a different approach, such as the method based on the McGhee-von Hippel theory,25,30 which uses data from timeresolved fluorescence measurements. This analysis is described in the following section. Similar fluorometric titrations were performed using singlestranded homonucleotide DNA ((dA)30, (dT)30, (dC)30, and (dI)30-50) to test the interactions detected in the absorption spectra. The BOBO-3 added to (dC)30 ssDNA had the same spectral features as free BOBO-3. In contrast, the BOBO-3 with (dA)30 or (dT)30 ssDNA showed a very low enhancement of the emission intensity and had an emission maximum at 605 and 597 nm, respectively (Supporting Information Figure SI 1). This enhancement was very small as compared with the enhancement with (dTdA)30. When BOBO-3 was added to (dI)30-50 ssDNA, the emission intensity gradually increased to a maximum at 601 nm. This change was not as significant as that observed with (dCdI)30-50 dsDNA. The spectral features of different mixtures of BOBO-3/ssDNA are compiled in Table SI 3 in the Supporting Information. These results indicate that BOBO-3 does not intercalate into (dA)30, (dT)30, or (dC)30 ssDNA to form enhanced fluorescent species but may interact by partial intercalation in (dI)30-50 ssDNA. The above results contrast with the behavior of BOBO-3 bound to single strands of DNA composed of the four nucleosides. This bound BOBO-3 showed a redshift of both the absorption and fluorescence spectra maxima and a detectable fluorescence enhancement.23 This result is due to intercalation of BOBO-3 into the self-hybridized strands induced by the presence of the dye. This phenomenon has been reported for other intercalating dyes,32 including cyanine dyes.33 Intercalating dyes have a stabilizing effect on the partial hybridization of DNA strands, on the formation of regional stem loops,34 and on increasing melting temperatures.32,33 However, in homonucleotide ssDNA, the bases cannot interact with each other to form the hydrogen bonds required to produce the doublestranded structure. Therefore, homonucleotide ssDNA cannot locate the highly fluorescent intercalated fluorophore. Only inosines are capable of interacting with themselves35 and partially allowing the internal binding of BOBO-3. The results indicate different interactions between BOBO-3 dye and ssDNA and dsDNA and demonstrate that those interactions are dependent on the base composition. BOBO-3 interacts with any type of dsDNA undergoing a relaxation of the excited state, and the result is a redshift of the absorption maximum. The magnitude of the emission enhancement and the redshift of the maximum of fluorescence differ for each base pair sequence and support differences in the fluorophore environment. The most notable difference is the reduced quantum yield in CG-rich sequences as compared with AT sequences. A moderate dependence of the emission enhancement on the dsDNA base content has been detected for other cyanine dyes.6

Figure 2. Fluorescence decay traces of BOBO-3/DNA mixtures (ratio 5:1): BOBO-3 free in solution (black); BOBO-3/(dA)30 ssDNA (red); BOBO-3/(dTdA)30 dsDNA (blue); and BOBO-3/(dCdG)21 dsDNA (green).

Several processes could cause differential emission enhancements for cyanine dyes complexed to dsDNA with different base contents. On one hand, AT sequences could produce binding sites of nature different from what CG sequences could produce. The different stiffness of the CG-rich sites compared to the AT-rich regions can cause differences in accessibility or wobbling of the intercalated dye.17 On the other hand, certain DNA nucleosides could cause excited-state ET quenching.18 Further fluorescence enhancement upon binding to AT sites is consistent with dsDNA groove binding. BOBO-3 should bind more readily to the narrow minor grooves of AT sequences than to the wider minor grooves of GC sequences, which may not provide the snug fit required for strong binding.11 Time-Resolved Fluorescence Spectroscopy at the Ensemble and Single-Molecule Level. Time-resolved fluorescence experiments were performed to investigate the temporal behavior of BOBO-3 emission and the changes to the BOBO-3 in its lifetime with interactions with different base pairs. For bulk experiments, we prepared solutions of BOBO-3 in the presence of the three different dsDNA types ((dTdA)30, (dCdI)30-50, and (dCdG)21) and high-molecular-weight double-stranded polyhomonucleotides (poly(A)-poly(dT) and poly(C)-poly(I)) at distinct BOBO-3/DNA bp ratios, and we recorded fluorescence decay traces as described in the Experimental Methods. We collected fluorescence decay traces of BOBO-3 in the presence of (dA)30, (dT)30, (dC)30, (dI)30-50, (dCG)21, and (dGC)21 ssDNA at different BOBO-3/nucleotide ratios. Figure 2 shows an example of the fluorescence decay traces of BOBO-3 free in solution and in the presence of singlestranded and double-stranded DNA of different base compositions ((dTdA)30 and (dCdG)21). Table 2 shows the average values of the lifetimes for different interactions of BOBO-3 with the

6718

J. Phys. Chem. B, Vol. 114, No. 19, 2010

different types of dsDNA. We also show in the Supporting Information, Tables SI 4 and SI 5, some examples of global fits, including the recovered decay times and pre-exponential factors. For all of the dsDNA samples, the best fits required a sum of the following three exponential decay functions: a short lifetime corresponding to BOBO-3 in solution (0.173 ( 0.009 ns);23 a second decay time around 1.7 ns with different weights, depending on the experimental conditions (see Supporting Information Tables SI 4 and SI 5); and a third, longer lifetime. The long lifetime showed significant differences for different types of dsDNA and ranged between 3 ns for AT pairs and 4 ns for CG pairs (Table 2). The lifetimes were fairly invariable over a large range of BOBO-3/dsDNA ratios, and a gradual decrease was observed at high ratios with the concomitant quenching of the steady-state fluorescence signal. This result is consistent with self-quenching or nonradiative resonance energy transfer between bound fluorophores caused by the close proximity between the fluorophores. This phenomenon has been observed for other intercalating dyes, such as Ethidium Bromide.36 The high-molecular-weight homonucleotide DNA polymers (double-stranded poly(dA)-poly(dT) and poly(dC)-poly(dI)) showed slightly different lifetimes from the corresponding shorter oligonucleotides (see Supporting Information Table SI 6). The longest lifetime is ∼1 ns higher for both polymers than the corresponding oligonucleotide. This fact suggests the dye is in different environments in each system. The perfect hybridization of complementary strands of high-molecularweight polymers is unlikely because the high probability of annealing between nonconsecutive bases causes the formation of a net instead of a perfect double helix.37 The normalized preexponential factors, which indicate the fraction of BOBO-3 in each state, are similar for short oligonucleotides and polynucleotides of the same composition (Tables 2 and SI 6). The experiments performed at the single-molecule level showed consistent results. The fluorescence fluctuation traces of BOBO-3 in the presence of (dTdA)30 and (dCdG)21 dsDNA provided reconstructed fluorescence decays by using the TTTR collection mode (Supporting Information Figure SI 2). Once the fluorescence decay histograms were reconstructed, we applied a time window of 11 ns starting 1.3 ns after the excitation pulse to rebuild the fluorescence fluctuation traces. We performed a BIFL analysis on the gated traces by setting a threshold of 10 times the background. The BIFL analysis extracts one or two lifetimes for each fluorescence burst based on a maximum likelihood estimation fitting. The frequency histograms of the lifetimes obtained in the single-molecule fluorescence bursts are shown in Figure 3. Two different broad populations are clearly visible for BOBO-3 in the presence of both (dTdA)30 and (dCdG)21 dsDNA. The fluorescence bursts of BOBO-3 in the presence of (dTdA)30 dsDNA showed the major population centered at 3.26 ns and a second peak centered at 1.50 ns. In contrast, the two populations were centered at 3.77 and 1.69 ns for BOBO-3 in the presence of (dCdG)21 dsDNA. Most of the fluorescence bursts presented both lifetimes simultaneously in each case, whereas 8.1 ( 1.3% of the bursts showed only the short lifetime, and 16.0 ( 0.6% showed only the long lifetime. These results confirm that BOBO-3 presents heterogeneity in the binding modes, also revealed by singlemolecule fluorescence burst analysis. The bulk experiments also revealed that BOBO-3 shows bior triexponential fluorescence decay traces when interacting with different homonucleotide ssDNA (Supporting Information Table SI 3). The decay traces of BOBO-3 mixed with (dA)30, (dT)30,

Ruedas-Rama et al.

Figure 3. Lifetime distributions from time-gated BIFL analysis of single-molecule fluorescence fluctuation traces of BOBO-3/dsDNA mixtures (ratio 1:10): (A) BOBO-3/(dTdA)30 dsDNA and (B) BOBO3/(dCdG)21 dsDNA. Two main distributions can be observed: at short lifetimes (gray bars) and at long lifetimes (red line). The distributions were fitted to Gaussian functions (blue and green lines) to obtain the central position and the width of the distributions.

and (dC)30 ssDNA were fit to biexponential functions. The major contribution was from the lifetime of the free dye, and the longer lifetime had a smaller contribution. Mixtures of the dye with (dI)30-50, (dCG)21, and (dGC)21 ssDNA showed triexponential decays and decay times similar to those found in dsDNA. In these cases, the weighted pre-exponentials of the long decay time, which was around 4 ns, were remarkable and in agreement with the higher emission intensity detected in them. These results are in good agreement with the mechanism described in the previous section. BOBO-3 in the presence of (dA)30, (dT)30, and (dC)30 ssDNA showed only the two lifetimes corresponding to a major contribution of free dye and a small population at the secondary interaction mode, confirming the absence of bis-intercalation. Bis-intercalation is absent because, as dicussed above, selfhybridization is not possible for these homonucleotide strands. However, when (dI)3050, (dCG)21, or (dGC)21 ssDNA is present, BOBO-3 shows triexponential decay traces similar to those found in dsDNA (Supporting Information Table SI 3). In these three strands, partial self-hybridization is possible and highly stabilized by the intercalating dye.23,33,34 The way that the strands (dCG)21 and (dGC)21 are sequenced (to avoid the difficulty of synthesizing several adjacent dGs) allows self-hybridization between adjacent strands and the formation of transient contacts. Inosines are capable of interacting with themselves to form duplexlike structures35 that may allow intercalation of BOBO-3. Quantization of Binding Constants and Binding Sites of BOBO-3 in (dTdA)30 and (dCdG)21 dsDNA. The global analysis of fluorescence decay traces and the recovery of preexponential weights for each decay time allow us to calculate the fraction of dye that exists in each microenvironment or state.23 We have obtained the mole fraction of BOBO-3 in each form (Figure 4) in the presence of homonucleotide dsDNA of the different base compositions. The three forms of BOBO-3 are the following: free in solution (corresponding to the shortest decay time), the partially quenched BOBO-3 bound to the secondary sites (corresponding to the intermediate lifetime), and the BOBO-3 intercalated into the dsDNA (corresponding to the longest decay time). The correspondence of the different decay times of the dye to different environments supports the presence of at least two interaction modes of the dye with the DNA. The presence of multiple interaction modes has been reported for BOBO-3 and

Binding of BOBO-3 Intercalative Dye to DNA

J. Phys. Chem. B, Vol. 114, No. 19, 2010 6719

Figure 4. The mole fraction of BOBO-3 in each state, calculated from the corrected pre-exponential factors of τ1 (long decay time, black), τ2 (intermediate decay time, red), and τ3 (short decay time, blue) at different ratios of BOBO-3 to (A) (dTdA)30, (B) (dCdG)21, and (C) (dCdI)30-50 dsDNA.

other cyanine derivatives.23,31,38 In the first of these sites, the dye is highly protected and has a longer lifetime. The second site involves a complex with less fluorescence than the other sites. These results are in agreement with first binding mode corresponding to bis-intercalation. The bis-intercalation binding assumes a clamp-like structure by which the two moieties of the cyanine monomer, the benzothiazole and the pyridinium, are intercalated by parallel stacking between the nucleotide bases, and the linker lies on the major groove of the dsDNA. This model of the bis-intercalation interaction has been experimentally demonstrated by high-resolution correlation NMR spectroscopy for TOTO-339 and other derivatives.40 The secondary binding mode of BOBO-3 may arise from either electrostatic interactions between the positive charge of BOBO-3 and the negative charge of the phosphate groups of DNA or a partial intercalation by stacking interactions.41-43 The linker in BOBO-3 between the two cyanine monomers is a polyamine, with two positive charges, similar to molecules such as spermine. It is known that spermine and other polyamines interact with dsDNA.44 Therefore, this part of BOBO-3 promotes the interaction with dsDNA primarily due to electrostatic attraction. In addition, the corresponding monomeric cyanine dye of BOBO3, so-called BOPRO-3, is also a dsDNA intercalative dye.3 Thus, it is plausible to think that the secondary interaction mainly corresponds to a half-intercalation mode in which only one of the two cyanine moieties is intercalated. This kind of halfintercalation model has been previously proposed for other dyes, such as Thiazole Orange and Cyan 13.45 Although we do not have experimental evidence of the structural arrangement of the BOBO-3 in the dsDNA, our experimental data are in agreement with other members of the family for which structural insights

were previously published, supporting our structural considerations on the two interaction modes of BOBO-3. Figure 4 shows some examples of recovered mole fractions of BOBO-3 in each state in the presence of different amounts of (dTdA)30, (dCdI)30-50, and (dCdG)21 dsDNA. For the systems with (dTdA)30 and (dCdG)21 dsDNA, the main contribution corresponds to the intercalated BOBO-3 (with small proportions of BOBO-3 bound to secondary sites), even at low BOBO-3 concentrations. At much higher BOBO-3 concentrations, the contribution of the shortest lifetime increases drastically when all of the binding sites are occupied, and the excess dye remains free in solution. In contrast, the major contribution in the (dCdI)30-50 dsDNA system corresponds to the dye bound to the secondary sites (Figure 4C). This result is in agreement with the interpretation that in this system, the strand of polyinosine is longer than the strand of polycytidine. Some single-stranded regions are left to preferentially interact with BOBO-3 through electrostatic interactions, increasing the secondary site population. The calculation of the mole fraction of each form of BOBO-3 interacting with any type of dsDNA provided the necessary information to further quantify these interactions according to the McGhee-von Hippel theory.25,30 The methodology was developed elsewhere using dsDNA with a mixed base composition.23 This methodology uses the equations for two overlapping, noncooperative binding modes and allows the recovery of binding constants, K1 and K2, for each binding mode and the number of sites (base pairs), n1 and n2, covered by a dye molecule bound in each mode. This methodology is based on iterative global fitting of a surface of mole fraction values obtained from time-resolved fluorescence measurements under different conditions using BOBO-3 and each homonucleotide

6720

J. Phys. Chem. B, Vol. 114, No. 19, 2010

Ruedas-Rama et al.

TABLE 3: Results from the McGhee-von Hippel Theory-Based Global Fitting of BOBO-3 Interacting with (dTdA)30 and (dCdG)21 dsDNA K1 (M-1) (dTdA)30 dsDNA (dCdG)21 dsDNA

(1.2 ( 0.5) × 10 (5.0 ( 1.0) × 105 6

dsDNA system. The application of this approach and further details are described in the Supporting Information section and are reported in a previous paper.23 We applied this methodology to BOBO-3’s binding to (dTdA)30 and (dCdG)21 dsDNA and compared the behavior of AT sites versus GC binding sites. The McGhee-von Hippel theory-based global fitting provided the results shown in Table 3. These results have three major implications. First, BOBO-3 has a higher affinity for AT sites in both modes than for GC sites. This result is suggested by the doubled binding constants, which are probably caused by different chemical environments on the AT and GC grooves.16 Second, the secondary interaction mode covers a similar number of sites in both cases, and third, the bis-intercalation mode covers about six GC base pairs but only four AT base pairs. The latter result may be caused by the characteristic heteronomous helical structure formed by poly(dA)-poly(dT).13-15 This structure has a narrower minor groove and a more positive propeller twist than the conventional DNA double helix. The special arrangement of BOBO-3 in multiple AT sites may allow a greater accessibility of the solvent46 and cause a partial quenching of the dye, as mentioned earlier (Table 2). When BOBO-3 binds a random sequence including the four bases (52% of GC content), about six base pairs are covered in the primary mode.23 That situation is similar to the GC-rich oligonucleotide studied here because poly(GC) dsDNA presents the usual B-double helix. This structure is typical of random, mixed sequences and contrasts with the heteronomous B′ double helix of the poly(dA)-poly(dT) mentioned above. The 46-bp dsDNA sequence included not more than two consecutive AT pairs, preventing the formation of a complete AT-rich binding region. The binding constants obtained here are in good agreement with those previously published for 46 bp dsDNA, including the four bases.23 The previously published binding constants are approximately the arithmetic average of the corresponding binding constants obtained here for AT and GC sites but are about 4-6 orders of magnitude smaller than the constants of other dyes in the cyanine family.1,31 These results also explain the differences in the average quantum yields observed in Table 2. BOBO-3 bound to (dCdG)21 has a higher lifetime than BOBO-3 bound to (dTdA)30. This difference also occurs in other cyanine dyes.6,21 However, the average quantum yield is higher in (dTdA)30 dsDNA (Figure 1 and Table 2) because of the higher affinity of BOBO-3 for (dTdA)30. This high affinity for (dTdA)30 dsDNA results in less free nonfluorescent BOBO-3 in solution, and more fluorescent dye stays bound than in (dCdG)21 dsDNA. This situation results in a higher fluorescence emission of BOBO-3 in (dTdA)30 dsDNA. Conclusions We are interested in developing sensitive methods of DNA hybridization in homogeneous media by using DNA intercalators as energy acceptors in FRET-based approaches. We studied the photophysical behavior of BOBO-3 and its interaction with single- and double-stranded DNA. In our previous work, we

n1

K2 (M-1)

4.0 ( 1.0 6.2 ( 1.1

(3.0 ( 1.0) × 10 (1.2 ( 0.2) × 105

n2 5

2.3 ( 1.3 2.3 ( 0.7

investigated the intercalating features of BOBO-3 in DNA with mixed base compositions. The main characteristic of the intercalated dye is a remarkable enhancement of the fluorescence emission. This enhancement occurs in other members of the family of cyanine dyes. In the present work, we found that BOBO-3 behavior was highly dependent on the base composition. For instance, BOBO-3 showed a higher affinity for AT sites, but the fluorescence lifetime was higher when BOBO-3 was intercalated into CG sites. This result was detected at both the ensemble level and the single-molecule level. These differences may provide a single-molecule fluorescence method to extract the GC content in dsDNA, which has been previously tried using other cyanine derivatives.21 The differences in the spectral features BOBO-3 may also be used to develop approaches to identify GC-rich or AT-rich regions within large strands of DNA. Acknowledgment. This work was supported by Grants CTQ2007-61619/BQU from the Ministerio Espan˜ol de Educacio´n y Ciencia, and P07-FQM-3091 from the Consejerı´a de Innovacio´n, Ciencia y Empresa (Junta de Andalucı´a). A.O. acknowledges a European Reintegration Grant from the seventh EU Framework. Supporting Information Available: Detailed description of additional methods and supporting Tables SI 1-6, and Figure SI 1. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Rye, H. S.; Yue, S.; Wemmer, D. E.; Quesada, M. A.; Haugland, R. P.; Mathies, R. A.; Glazer, A. N. Stable Fluorescent Complexes of Double-Stranded DNA with Bis-Intercalating Asymmetric Cyanine Dyes: Properties and Applications. Nucleic Acids Res. 1992, 20, 2803–2812. (2) Glazer, A. N.; Rye, H. S. Stable Dye-DNA Intercalation Complexes As Reagents for High-Sensitivity Fluorescence Detection. Nature 1992, 359, 859–861. (3) Haugland, R. P. Handbook of Fluorescent Probes and Research Products; Molecular Probes, Inc.: Eugene, OR, 2002. (4) Sundstro¨m, V.; Gillbro, T. Viscosity Dependent Radiationless Relaxation Rate of Cyanine Dyes. A Picosecond Laser Spectroscopy Study. Chem. Phys. 1981, 61 (3), 257–269. (5) Momicchioli, F.; Baraldi, I.; Berthier, G. Theoretical Study of Trans-Cis Photoisomerism in Polymethine Cyanines. Chem. Phys. 1988, 123 (1), 103–112. (6) Netzel, T. L.; Nafisi, K.; Zhao, M.; Lenhard, J. R.; Johnson, I. BaseContent Dependence of Emission Enhancements, Quantum Yields, and Lifetimes for Cyanine Dyes Bound to Double-Strand DNA: Photophysical Properties of Monomeric and Bichromomphoric DNA Stains. J. Phys. Chem. 1995, 99 (51), 17936–17947. (7) Rye, H. S.; Dabora, J. M.; Quesada, M. A.; Mathies, R. A.; Glazer, A. N. Fluorometric Assay Using Dimeric Dyes for Double- and SingleStranded DNA and RNA with Picogram Sensitivity. Anal. Biochem. 1993, 208 (1), 144–150. (8) Rye, H. S.; Yue, S.; Quesada, M. A.; Haugland, R. P.; Mathies, R. A.; Glazer, A. N. Picogram Detection of Stable Dye-DNA Intercalation Complexes with Two-Color Laser-Excited Confocal Fluorescence Gel Scanner. Methods Enzymol. 1993, 217, 414–431. (9) Hirons, G. T.; Fawcett, J. J.; Crissman, H. A. TOTO and YOYO: New Very Bright Fluorochromes for DNA Content Analyses by Flow Cytometry. Cytometry 1994, 15, 129–140. (10) Jacobsen, J. P.; Pedersen, J. B.; Hansen, L. F.; Wemmer, D. E. Site Selective Bis-Intercalation of a Homodimeric Thiazole Orange Dye in DNA Oligonucleotides. Nucleic Acids Res. 1995, 23 (5), 753–760.

Binding of BOBO-3 Intercalative Dye to DNA (11) Kumart, C. V.; Turner, R. S.; Asuncion, E. H. Groove Binding of a Styrylcyanine Dye to the DNA Double Helix: the Salt Effect. J. Photochem. Photobiol. A: Chem. 1993, 74, 231–238. (12) Bunkenborg, J.; Stidsen, M. M.; Jacobsen, J. P. On the Sequence Selective Bis-Intercalation of a Homodimeric Thiazole Orange Dye in DNA. Bioconjugate Chem. 1999, 10 (5), 824–831. (13) Leslie, A. G. W.; Arnott, S.; Chandrasekaran, R.; Ratliff, R. L. Polymorphism of DNA Double Helices. J. Mol. Biol. 1980, 143, 49–72. (14) Arnott, S.; Chandrasekaran, R.; Hall, I. H.; Puigjaner, L. C. Heteronomous DNA. Nucleic Acids Res. 1983, 11 (12), 4141–4155. (15) Chuprina, V. P. Regularities in Formation of the Spine of Hydration in the DNA Minor Groove and Its Influence on the DNA Structure. FEBS Lett. 1985, 186, 98–102. (16) Watson, J. D.; Baker, T. A.; Bell, S. P.; Gann, A.; Levine, M.; Losick, R. The Structures of DNA and RNA. In Molecular Biology of the Gene, 5th ed.; Benjamin Cummings: New York, 2004. (17) Shaw, A. K.; Pal, S. K. Fluorescence Relaxation Dynamics of Acridine Orange in Nanosized Micellar Systems and DNA. J. Phys. Chem. B 2007, 111 (16), 4189–4199. (18) Kemnitz, K.; Yoshihara, K.; Tani, T. Short and ExcitationIndependent Fluorescence Lifetimes of J-Aggregates Adsorbed on Silver(I) Bromide and Silica. J. Phys. Chem. 1990, 94 (7), 3099–3104. (19) Steenken, S.; Telo, J. P.; Novais, H. M.; Candeias, L. P. OneElectron-Reduction Potentials of Pyrimidine Bases, Nucleosides, and Nucleotides in Aqueous Solution. Consequences for DNA Redox Chemistry. J. Am. Chem. Soc. 1992, 114 (12), 4701–4709. (20) Dunn, D. A.; Lin, V. H.; Kochevar, I. E. Base-Selective Oxidation and Cleavage of DNA by Photochemical Cosensitized Electron Transfer. Biochemistry 1992, 31 (46), 11620–11625. (21) Bowen, B. P.; Enderlein, J.; Woodbury, N. W. Single-Molecule Fluorescence Spectroscopy of TOTO on Poly-AT and Poly-GC DNA. Photochem. Photobiol. 2003, 78 (6), 576–581. (22) Bowen, B. P.; Woodbury, N. W. TOTO Binding Affinity Analysis Using Single-Molecule Fluorescence Spectroscopy. Photochem. Photobiol. 2003, 78 (6), 582–586. (23) Ruedas-Rama, M. J.; Orte, A.; Crovetto, L.; Talavera, E. M.; Alvarez-Pez, J. M. Photophysics and Binding Constant Determination of the Homodimeric Dye BOBO-3 and DNA Oligonucleotides. J. Phys. Chem. B 2010, 114, 1094–1103. (24) Ruedas-Rama, M. J. unpublished results. (25) McGhee, J. D.; von Hippel, P. Theoretical aspects of DNA-Protein interactions: Co-operative and Non-co-operative Binding of Large Ligands to a One-Dimensional Homogeneous Lattice. J. Mol. Biol. 1974, 86, 469– 489. (26) Sˇtefl, R.; Cheatham, T. E.; Sˇpacˇkova´, N. a.; Fadrna´, E.; Berger, I.; Kocˇa, J.; Sˇponer, J. Formation Pathways of a Guanine-Quadruplex DNA Revealed by Molecular Dynamics and Thermodynamic Analysis of the Substates. Biophys. J. 2003, 85, 1787–1804. (27) Read, M. A.; Neidle, S. Structural Characterization of a GuanineQuadruplex Ligand Complex. Biochemistry 2000, 39, 13422–13432. (28) Wahl, M. Time Tagged Time-Resolved Fluorescence Data Collection. PicoQuant Technical Note 2004, http://www.picoquant.com/ technotes/technote_tttr.pdf (accessed 2010). (29) Magde, D.; Rojas, G. E.; Seybold, P. G. Solvent Dependence of the Fluorescence Lifetimes of Xanthene Dyes. Photochem. Photobiol. 1999, 70 (5), 737–744. (30) Gaugain, B.; Barbet, J.; Capelle, N.; Roques, B. P.; Le Pecq, J. B. DNA Bifunctional Intercalators. 2. Fluorescence Properties and DNA

J. Phys. Chem. B, Vol. 114, No. 19, 2010 6721 Binding Interaction of an Ethidium Homodimer and an Acridine Ethidium Heterodimer. Biochemistry 1978, 17, 5078–5088. (31) Larsson, A.; Carlsson, C.; Jonsson, M.; Albinsson, B. Characterization of the Binding of the Fluorescent Dyes YO and YOYO to DNA by Polarized Light Spectroscopy. J. Am. Chem. Soc. 1994, 116, 8459–8465. (32) Monis, P. T.; Giglio, S.; Saint, C. P. Comparison of SYTO9 and SYBR Green I for Real-time Polymerase Chain Reaction and Investigation of the Effect of Dye Concentration on Amplification and DNA Melting Curve Analysis. Anal. Biochem. 2005, 340 (1), 24–34. (33) Bjorndal, M. T.; Fygenson, D. K. DNA Melting in the Presence of Fluorescent Intercalating Oxazole Yellow Dyes Measured with a Gel-Based Assay. Biopolymers 2002, 65 (1), 40–44. (34) Mao, F.; Leung, W.-Y.; Xin, X. Characterization of EvaGreen and the Implication of Its Physicochemical Properties for qPCR Applications. BMC Biotechnol. 2007, 7 (1), 76. (35) Rossi, C.; Picchi, M.; Tiezzi, E.; Corbini, G.; Corti, P. Conformational and Dynamic Investigation in Solution of Inosine and Its Molecular Complex, Inosiplex, by Proton and Carbon NMR Spectroscopy. Magn. Reson. Chem. 1990, 28 (4), 348–354. (36) Heller, D. P.; Greenstock, C. L. Fluorescence Lifetime Analysis of DNA Intercalated Ethidium Bromide and Quenching by Free Dye. Biophys. Chem. 1994, 50, 305–312. (37) Chamberlin, M. J.; Patterson, D. L. Physical and Chemical Characterization of the Ordered Complexes Formed between Polyinosinic Acid, Polycytidylic Acid and Their Deoxyribo-Analogues. J. Mol. Biol. 1965, 12, 410–428. (38) Johansen, F.; Jacobsen, J. P. H-1 NMR Studies of the Bisintercalation of a Homodimeric Oxazole Yellow Dye in DNA Oligonucleotides. J. Biomol. Struct. Dyn. 1998, 16, 205–222. (39) Spielmann, H. P.; Wemmer, D. E.; Jacobsen, J. P. Solution Structure of a DNA Complex with the Fluorescent Bis-Intercalator TOTO Determined by NMR Spectroscopy. Biochemistry 1995, 34 (27), 8542–8553. (40) Petersen, M.; Jacobsen, J. P. Solution Structure of a DNA Complex with the Fluorescent Bis-Intercalator TOTO Modified on the Benzothiazole Ring. Bioconjugate Chem. 1998, 9 (3), 331–340. (41) Rye, H. S.; Glazer, A. N. Interaction of Dimeric Intercalating Dyes with Single-stranded DNA. Nucleic Acids Res. 1995, 23, 1215–1222. (42) Tyutyulkov, N.; Fabian, J.; Mehlhom, A.; Dietz, F.; Tadjer, A., Molecular and aggregate structures of polymethines. In Polymethine Dyes. Structure and Properties. St. Kliment Ohridski University Press: Sofia, 1991; pp 89-121. (43) Pritchard, N. J.; Blake, A.; Peacocke, A. R. Modified Intercalation Model for the Interaction of Amino Acridines and DNA. Nature 1966, 212 (5068), 1360–1361. (44) Kral, T.; Hof, M.; Langner, M. The Effect of Spermine on Plasmid Condensation and Dye Release Observed by Fluorescence Correlation Spectroscopy. Biol. Chem. 2002, 338, 331–335. (45) Nygren, J.; Ogul’chansky, T. Y.; Losytskyy, M. Y.; Kovalska, V. B.; Yashchuk, V. M.; Yarmoluk, S. M. Interactions of Cyanine Dyes with Nucleic Acids. XXIV. Aggregation of Monomethine Cyanine Dyes in Presence of DNA and Its Manifestation in Absorption and Fluorescence Spectra. Spectrochim. Acta A 2001, 57, 1525–1532. (46) Chalikian, T. V.; Sarvazyan, A. P.; Plum, G. E.; Breslauer, K. J. Influence of Base Composition, Base Sequence, and Duplex Structure on DNA Hydration: Apparent Molar Volumes and Apparent Molar Adiabatic Compressibilities of Synthetic and Natural DNA Duplexes at 25 °C. Biochemistry 2002, 33 (9), 2394–2401.

JP1010742