The interaction studies of small molecule with nucleic acids are much-explored research area due to their binding to the nucleic acid structures and their interface in transcription and replication processes. The interaction studies of ligands with the DNA are highly important in the area of chemotherapeutic applications. It is very well documented that anticancer drugs show their biological action through their interaction with the DNA molecule. To understand the molecular mechanism of the DNA-drug interaction and their effect on a particular disease, this spectroscopic study of the comparative analysis of various dyes had been planned and performed. Nucleic acids exhibit remarkable changes in absorbance and fluorescence properties after complex formation.

Thionine acetate (TA), azure A (AA), and azure B (AB) are phenothiazinium dyes with the identical basic phenothiazinium skeleton, with variations in the groups present at 3 and 7 positions (Figure 1).

Figure 1: Chemical structure of (a) Thionine acetate, (b) Azure A, and (C) Azure B.

Thionine (3,7-diamino-5-phenothiazinium) is a planar phenothiazinium dye that consists of a heterocyclic nitrogen atom with two amine groups present at 3 and 7 positions of the aromatic ring [1]. It is widely used in electrochemical and photochemical biosensors [2,3]. It is also a good singlet oxygen generator and used in Photodynamic Therapy (PDT) [4,5]. The interaction studies of thionine acetate with the DNA molecule under various conditions have been well reported [6,7]. It is extensively used in nucleic acid probes [1], decontamination of blood products [8], and against bacteria, viruses, and yeast [9-11]. The interaction studies of thionine with the DNA, by taking advantage of numerous biophysical techniques have already been reported and have shown that the thionine binds to DNA by two different modes, intercalation as well as outside the DNA double helix [12]. They also suggest that thionine does not show the binding affinity towards GC-rich sequences. Later, it was proposed that the thionine binding to GC-rich sequence is preferable in comparison to AT-rich sequences [13]. Brown and Brown have reported that TA has mutagenic activity against eukaryotic cells [14]. It is reported that phenothiazinium dye methylene blue (MB) has the potential to inhibit acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE). The pharmacological action of MB is assisted by one of its metabolite i.e., azure B, which is able to inhibit AChE with twofold lower potency and BuChE with fivefold lower potency as compared to MB [15]. It is also documented that azure B has significance in Alzheimer's disease, where it can inhibit the formation of ß-amyloid protein filament [16]. Azure A and azure B can potentially inhibit the formation of tau filament with IC50 values without affecting the ability of tau protein to interact with microtubules [17]. An effective and accurate LC-MS/MS analytical method was developed by Kim and their coworkers to quantify the methylene blue and its metabolite azure B which is used in the pharmacokinetic studies of phenothiazine dyes [18].

Azure A is a cationic dye belonging to the phenothiazinium group having an asymmetrical structure. It is commonly used as a therapeutic agent due to its potential biological activity. Other biological activity of azure A dye includes anti-malarial activity [19], reducing the extent of arrhythmias [20], and diagnosis of amyloid accumulation related diseases. It is well-known staining dye, used to stain blood smear and also in electrochemical biosensing [21]. In a report, azure A is employed as an agent who is capable of detecting the anionic detergents present in milk [22].

Azure B, monodemethyl methylene blue is very close in structure to azure A, and is also biologically important dye used for staining purpose. It is widely used for staining DNA, proteins, viruses, melanin, keratin fibers, and malaria-infected cells [23]. It has also antiinflammatory and antitumor properties [24]. This dye is used in the treatment of oral cavity infection, nail infection, Alzheimer disease, diabetes, and detection of oral cancer [25].

Although lot of interaction studies of these dyes are reported in the literature, but their comparative analysis is not discussed in detail. Here, we report the intercalation of all the three phenothiazinium dyes with calf thymus DNA using various spectroscopic studies such as circular dichroism, fluorescence spectroscopy, UV-visible absorptions spectroscopy, and UV-thermal melting studies. The basic difference between the three dyes is the presence of methyl group on the functional group which is responsible for increasing the hydrophobicity of the dye. In an attempt of the comparative analysis of binding constant values, it was found that thionine dye can interact with ct-DNA most efficiently among all the three (Thionine acetate, azure A and azure B). The quenching constants were also calculated and it was concluded that all the three dyes are interacting with DNA in the same manner, with difference in their extent of binding.

Calf thymus DNA was purchased from S.G Enterprises and used without further purification. Thionine acetate, azure A, and azure B were purchased from Helix Biosciences. The stock solution of both ctDNA and all the ligands were prepared by dissolving in double distilled water and a working solution was prepared by appropriate dilution. The concentration of ctDNA was determined by recording absorption of ctDNA at 260 nm (ε₂₆₀=6600 L.mol^-1.cm^-1). Sodium cacodylate (pH=7.4, containing 0.1 M NaCl and 0.1 mM EDTA) was used as buffer solution. All chemicals were of analytical grade.

UV-Vis Spectroscopic experiments

The UV-absorption spectra was recorded on a UV-1650 PC Shimadzu UV-visible spectrophotometer (TMSPC-8(E)-200) and interfaced with a Pentium IV computer for data collection and analysis. The stoppered quartz cuvette of 1 cm optical path length (1 ml volume) was used for the UV-experiments.

Thermal melting studies

UV-thermal melting experiments of ctDNA were performed at 260 nm in the absence and presence of the dyes on a UV-2450 PC Shimadzu UV-visible spectrophotometer (TMSPC-8(E)-200) and interfaced with a Pentium IV computer for data collection and analysis. The temperature gradient used was 20-95°C at a rate of 0.5°C/ min. The Tm values of ctDNA and their complex were obtained from the midpoint of the transition of melting curves.

Fluorescence measurements

The fluorescence spectra and intensity were recorded on Cary Eclipse spectrofluorimeter (Varian, USA) equipped with a 150W Xenon lamp with a 1 cm quartz cuvette and a thermostat water bath. The pH was measured on a Thermo scientific Orion 2 star pH meter (Singapore). Fluorescence spectra of all the three drugs were obtained at 290 K, 300 K, and 310 K temperature in the range 590-800 nm at an excitation wavelength 595, 625, and 640 nm for TA, AA, and AB respectively, using a slit width of 5 nm.

Circular Dichroism (CD) studies

The CD spectra were recorded on JASCO J-815 spectropolarimeter using a quartz cuvette of 1.0 cm path length, at wavelength 200-700 nm, 1 nm data pitch with a response time of 1 s and averaged of accumulation of three scans at a speed of 100 nm/min.

UV absorption studies

Absorption studies are considered the most common and elementary technique to explore the interaction studies of dyes with the DNA by observing the changes in the absorption spectrum of either DNA or drug molecule [26,27]. Usually, interaction studies are carried out by varying the concentration of calf thymus DNA with a fixed concentration of ligand/proteins. The visible absorption spectra show maxima at 598 nm, 631 nm, 646 nm for thionine acetate, azure A, and azure B respectively. A small hump can also be noticed at 561 nm, 588 nm, and 598 nm for all the three dyes. Figure 2 shows UVabsorption spectra of all the three dyes on increasing concentrations of calf thymus DNA. As the concentration of calf thymus DNA is increased, the absorbance starts decreasing, resulting in hypochromic effect. Apart from this, a bathochromic shift of 7 nm from 631 nm to 638 nm and of 4 nm from 646 nm to 650 nm in maximum absorption peaks of azure A and azure B dyes are observed respectively. While in the case of thionine acetate, a red shift of 6 nm in the absorption maxima from 598 nm to 604 nm is detected. The absorption intensity decreases along with bathochromic shift in the case of intercalation binding mode of the dyes into the base pairs of the DNA [28]. The dye molecule intercalates via coupling of empty π* orbital of the dye molecule with the π* orbital of the DNA base pairs. As a result, π- π* electronic energy decreases which leads to red shift. The hypochromic effect is caused due to the reduced probability of transition because of partially filled empty π* orbital. Sharp isosbestic points at 652 nm, 660 nm, and 611 nm observed for azure A, azure C and thionine acetate respectively, confirming the existence of an equilibrium between the two states of free and bound dye molecules. These results indicate towards the presence of intercalation mode of binding between dyes and the DNA double helix.

Figure 2: UV absorption spectra of Thionine acetate (a), Azure A (b), and Azure B (c) varying with concentrations of ctDNA at pH=7.4 and T=300 K, C_Dye=0 µmol.L^-1, CDNA=0, 100, 120, 140, 160, and 180 µmol.L^-1 for curves 1-8, respectively.

Fluorescence spectroscopic experiments

Fluorescence spectroscopy is extensively used to explore the binding mechanism of a ligand to the protein or DNA molecule [29]. An external molecule can potentially quench the fluorescence of any molecule upon binding. Several types of molecular interactions are liable for the quenching of the fluorescence, including complex formation in the ground state, excited-state reactions, molecular rearrangements, energy transfer, and collisional quenching [30]. The fluorescence emission spectra of thionine acetate, azure A, and azure B in the absence and presence of DNA are displayed in Figure 3. Thionine acetate, azure A, and azure B are fluorescent molecules: when excited at 595, 625, and 640 nm, the emission spectra in the range of 600-700 nm with maxima at 623, 653, and 672 nm respectively is observed. Keeping the constant concentration of dye, the DNA concentration was gradually increased, in order to examine the complex formation. The quenching in the fluorescence spectrum is observed with all the three dyes, on each addition of DNA, suggesting the confirmation of the complex formation. The maximum quenching (~69%) in the fluorescence was detected for TA-DNA complex, followed by azure A (~55%) and the least for azure C (~69%). A similar pattern for quenching is already reported; where it is shown that the quenching of azure A is much prominent in comparison to azure B [31]. The obtained fluorescence quenching is the consequence of the successful binding of the dye molecule to the base pairs of DNA.

Figure 3: Fluorescence spectra of (a) TA, (b) AA, and (c) AB in the presence of ctDNA at different concentrations at pH=7.4 and T=300 K, C_Dye=10 µmol.L^-1, CDNA=0, 20, 40, 60, 80, 100, 120, and 140 µmol.L^-1 for curves 1-8, respectively (Inset: The plot of log (F₀-F)/F versus log [DNA]).

Fluorescence intensities of dye samples were rectified for the respective absorbance of dyes at emission and excitation wavelengths of the fluorescence data. Corrections for the inner filter effect were implemented using the following equation [32].

The Binding parameters

The fluorescence quenching due to binding of the dye to ctDNA was further utilized to calculate the binding efficiency (K_b). The binding constant (K_b) at various temperature can be attained by using the following equation [33,34].

Values of K_b were determined by the intercept of the plot of log [(F₀- F)/ F] versus log [Q] (Figure 3). The binding constants for all the three dyes are calculated and shown in Table 1.

Table 1: Binding constant and quenching constant values calculated from fluorescence spectra.

The quenching mechanism

It has been noticed that the fluorescence emission intensity of the dyes is gradually decreased with the increase in the concentration of ctDNA. It means that ctDNA could quench the intrinsic fluorescence of the dyes. The fluorescence quenching constant at different temperatures (290 K, 300 K, and 310 K) was calculated using the Stern- Volmer equation (Figure 4) [35].

Figure 4: Stern-volmer plots for the quenching of (a) TA, (b) AA, and (c) AB by DNA at different temperatures.

Where F₀ and F are the fluorescence intensities of the dyes in the absence and presence of the quencher, respectively, [Q] is the concentration of the quencher and K_SV is the Stern-Volmer quenching constant, which can be determined by the plot of F₀/F against [Q]. The different types of quenching are generally categorized as either static or dynamic quenching. The static and dynamic quenching constants are totally temperature dependent. The K_SV values signify the fluorophore accessibility to the quencher. It was concluded that quenching constant (K_SV) decreases with the increase in temperature, which signifies the presence of static quenching between the dyes and ctDNA [36].

Circular dichroism

Circular Dichroism (CD) is a very sensitive and accurate technique to study the optical behavior of any chiral molecule. It gives the information of the secondary structure of nucleic acid and proteins. It can easily discriminate between various secondary structures formed by DNA (A-, B-, Z- etc.) [37]. All the forms of DNA have a unique CD spectrum with different wavelength maxima and the ellipticity. The Bform of the DNA is comprised of a positive peak around 270-280 nm, due to base stacking and a negative peak at 245 nm, due to righthanded helicity [38]. The conformational changes in the secondary structure of ctDNA have been studied using circular dichroism spectroscopy. The CD spectrum in the range of 200-320 nm was examined to see the effect of dye-induced changes in the DNA secondary structure in the presence of various concentrations of the dyes. Circular dichroism spectra of ctDNA (1 × 10^-4 M) was recorded in the presence of increasing amounts of dye at the following stoichiometric ratios: r_i=[dye]/[DNA]=0.0, 0.10, 0.15 in the wavelength range 200-320 nm, at 298 K temperature (Figure 5).

Figure 5: Circular Dichroism spectra of DNA(1×10^-4 M) in sodium cacodylate buffer (20 mM, pH 7.4) in the presence of increasing amounts of dyes (a) TA, (b) AA, and (c) AB at the following stoichiometric ratios: ri=[dye]/[DNA]=0.0, 0.1, 0.15, 0.2.

The CD spectrum of ctDNA consisted of a positive peak at 275 nm and a negative peak at 244 nm, which is characteristic of B-form of DNA. An increase in the positive peak at 275 nm is observed on addition of dye to it, which is in turn result of intercalation of dye into the DNA base pairs [39]. An increment in the intensity at 245 nm has also been noticed, giving an indication of stabilization in the righthanded more compact B-form of DNA [40]. All the dyes used in this study (TA, AA, and AB) as well as ctDNA do not show optical behavior around 400 to 700 nm range individually, while their complexes (DNA-dyes) have shown optical active nature, which is attributed to induced circular dichroism [41,42]. Figure 6 depicts the induced circular Dichroism spectral changes of TA, AA and, AB upon binding to ctDNA which quite clearly confirms the the formation of the DNA-dye complex in all cases.

Figure 6: Circular Dichroism spectra of Dyes (10 μM) (a) TA, (b) AA, and (c) AB in sodium cacodylate buffer (20 mM, pH 7.4) in the presence of increasing amounts of DNA.

Thermal-denaturation studies

The thermal stability of DNA can be determined by employing UVthermal denaturation technique. The absorbance increases sigmoidally, when DNA sample is heated and the temperature at which half of the DNA gets unwind is known as the melting temperature (Tm). It helps in the detection of stabilization or destabilization of a DNA structure on interaction with a ligand [43,44]. Experimental evidences suggest that intercalative mode of binding is attributed to increase in TM value up to 3-8°C, whereas no significant change is observed in TM value for non-intercalative interaction [45]. Figure 7 represents the melting curves of calf thymus DNA at 260 nm in the absence or presence of dyes. A melting temperature of 84°C was observed for calf thymus DNA in the absence of dyes under physiological conditions [46], however the melting temperature in the presence of thionine acetate, azure A, and azure B were found to be 89°C, 86.5°C, and 86°C respectively. As can be seen from the results, the interaction with all the three dyes causes a significant increase in the denaturation temperature, which strongly recommends the intercalation mode of binding of dyes. The binding constant calculated by the fluorescence spectroscopy manifested that the TA could bind to DNA more efficiently in comparison to other two dyes. A similar pattern is observed from the thermal melting profiles as well. The melting temperature of TA-DNA complex was found to 4°C more than the alone DNA melting temperature, while in the case of AA and AB, it was found only around 1°C or 2°C. The thermal melting analysis is also in accordance with the fluorescence data.

Figure 7: Melting curves of DNA in the absence and presence of TA, AA, and AB in 20 mM sodium cacodylate buffer (pH 7.4). C_Dye=10 μmol.L^-1 and CDNA=100 μmol.L^-1.

In the present work, the spectroscopic analysis of the interaction of the three phenothiazinium dyes (Thionine acetate, azure A, Azure B) with ctDNA was carried out using biophysical techniques such as UVabsorption, fluorescence, circular dichroism and thermal melting analysis. In UV-spectra of the dyes, a hypochromic and red shift was detected with the increasing concentrations of ctDNA, suggesting the intercalation mode of binding. The binding constant and quenching constant was calculated using fluorescence spectroscopy. The K_b value decreases with the rise in temperature, due to a decrease in the association between the dyes and ctDNA. CD results indicated the stabilization of B-form of DNA on dye binding. An increase in the melting temperature in case of all the three dyes indicates the intercalation binding mode. Thionine acetate had shown the maximum DNA stabilization the DNA double helix among all the studied dyes. The order in which these dyes stabilize the DNA double helix is TA > AA > AB, which was confirmed by the binding constants value and their melting profiles. All the results presented here are indicative of intercalation mode of binding between the dyes and ctDNA. This report provides insights on the interaction of ligands with DNA and adds to the knowledge towards findings of more efficient molecules for DNA binding.

The authors would like to thank the University of Delhi for its R&D grants. Financial support by Council of Scientific & Industrial Research to Mohan Kumar is gratefully acknowledged.