- Email: [email protected]
Multiplex binding modes of toluidine blue with calf thymus DNA and conformational transition of DNA revealed by spectroscopic studies
Spectrochimica Acta Part A 74 (2009) 421–426
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Multiplex binding modes of toluidine blue with calf thymus DNA and conformational transition of DNA revealed by spectroscopic studies Juan Wang a,b , Xiurong Yang a,∗ a State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun 130022, PR China b Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China
a r t i c l e
i n f o
Article history: Received 13 April 2008 Received in revised form 30 April 2009 Accepted 12 June 2009 Keywords: Toluidine blue (TB) DNA Partial intercalation Electrostatic interaction Conformational transition
a b s t r a c t It is noteworthy to understand the details of interactions between antitumor drugs and DNA because the binding modes and affinities affect their antitumor activities. Here, The interaction of toluidine blue (TB), a potential antitumor drug for photodynamic therapy of tumor, with calf thymus DNA (ctDNA) was explored by UV–vis, fluorescence, circular dichroism (CD) spectroscopy, UV-melting method and surfaceenhance Raman spectroscopy (SERS). The experimental results suggest that TB could bind to ctDNA via both electrostatic interaction and partial intercalation. The fluorescence quenching of TB by ctDNA was static and due to electron transfer from bases to the excited singlet state of TB. At low [TB]/[DNA] ratio, TB mainly partially intercalated into ctDNA resulting in the slight increase of base stacking degree; at high [TB]/[DNA] ratio, excessive TB externally stacked along the helix surface via coupling with partially intercalated ones, thereby inducing B-A transition of ctDNA. The conformational transition of DNA was confirmed by the obvious improvement of the thermal stability of ctDNA. The SERS spectra suggest that TB could partially intercalate into DNA basepairs with its ring C1 NC1 side buried. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Nucleic acids as the carriers of genetic information provide important targets for many biomacromolecules and small molecules. The interactions between DNA and small molecules are of special importance in chemotherapeutic applications and mutagenesis studies [1–3]. Many small molecular anticancer drugs are known to exert their biological activity through interacting with DNA [1–4], while lots of carcinogens and potent mutagens attack nucleic acids thereby changing DNA replication or reparation and lead to uncontrolled growth of tumor cells [2]. Although a lot of important DNA-binding compounds have been explored in phenotypic, cell-based screens, in vitro studies are needed for comprehensive understanding of the interaction mechanism [1]. The investigations on the nature of interactions between small molecules and nucleic acids, not only shed light on understanding the chemical basis for the antitumor mechanism of drugs and the carcinogenicity of pollutants, but also give chemists and pharmacologists opportunities to discover or design new drug candidates. Phenothiazinium dyes are photocytotoxic in living systems, and even cause photoinduced mutagenic effects when some cells
∗ Corresponding author. E-mail address: [email protected] (X. Yang). 1386-1425/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2009.06.038
survive from damages [5]. DNA is an important target in the phenothiazinium dye-photosensitized biological damage, and guanine residues are particularly accessible [6–8]. Moreover, some phenothiazinium photosensitizers, such as toluidine blue (TB) and methylene blue (MB), have the promising applications in photodynamic therapy of tumor [9–11]. These potential anti-tumor drugs can kill cancer cells through indirect or direct reactions with DNA [5]. TB has been widely used for detection and delineation of some cancers and displays good safety in humans. Furthermore it appears some selectivity for cancer cells in vivo, therefore TB has been explored as a potential anti-tumor drug [11]. It is important to obtain comprehensive understanding of the DNA-binding properties of TB. To our knowledge, there are only a few investigations on the interaction between TB and DNA. Resonance light scattering spectra of TB with the presence of DNA indicated that TB could aggregate along DNA surface [12]. Jiao et al. [13] have studied the interaction of TB with DNA by electrochemistry and proposed the electrostatic binding of TB to DNA. Since TB has the planar ring structure, it is quite possible that TB would bind to DNA via intercalation. However details on the binding mode of TB to DNA and the conformational changes of DNA are still unrevealed, which has great relevance to the bioactivity of TB. In this study, the interaction between TB and ctDNA was investigated by UV–vis, fluorescence, CD, surfaceenhanced Raman spectroscopy (SERS) and UV-melting method.
422
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
This research is helpful for evaluating the antitumor activity of TB. 2. Experimental 2.1. Equipments The absorption spectra were obtained by Cary 50 UV–vis spectrophotometer (Varian, USA). All fluorescence spectra were measured on a PerkinElmer LS-55 luminescence spectrometer. The CD spectra were recorded using a 62A DS CD spectrometer (AVIV, USA). UV-melting profiles of ctDNA were obtained by Cary 500 UV–vis-NIR spectrophotometer (Varian, USA) equipped with a quartz cell of 1 cm path length and a thermoelectrically controlled cell holder. The SERS spectra were measured on a Renishaw 2000 model confocal microscopy Raman spectrometer (Renishaw Ltd., Gloucestershire, U.K.) using radiation of 514.5 nm for the SERS excitation. 2.2. Reagents TB was obtained from Acros Organic Co. (USA) without further purification. Deoxyribonucleic acid sodium salt from calf thymus (type I) was purchased from Sigma Chemical Co. (USA). All reagents used were of analytical grade and aqueous solutions were prepared using doubly distilled deionized water. The TB stock solution of 1 mM was kept away from light to avoid photochemical decomposition and was diluted just before use. The DNA concentration (basepairs) was spectroscopically determined using molar absorption coefficient of 13,200 cm−1 M−1 at 260 nm. The ratio of the absorbance at 260 and 280 nm was larger than 1.8, indicating that ctDNA was sufficiently free of protein. All the experiments involving interaction of TB with ctDNA were carried out in buffer containing 10 mM NaAc-HAc and 50 mM NaCl (pH 5) unless otherwise stated and the ionic strength was modulated by adding an appropriate volume of stock solution of NaCl (5.0 M in doubly distilled deionized water). 2.3. Procedures 2.3.1. Absorption spectroscopy study Absorption spectra were recorded in the range of 500–800 nm using a quartz cell of 1 cm path length at the ambient temperature, unless otherwise stated. The absorbance titrations were performed by keeping the concentration of TB fixed at 0.01 mM while gradually increasing the concentration of ctDNA from 0.002 to 0.03 mM. From the absorption titration data, the binding constant was determined according to [14]. A0 /(A0 − A) = εf /(εf − εa ) + εf /(εf − εa ) × (1/K[DNA]) where A0 and A are the absorbances of drug in the absence and presence of DNA, εf and εa are the absorption coefficients of drug and its complex with DNA, K is the binding constant between drug and DNA, respectively. The plot of A0 /(A − A0 ) versus 1/[DNA] was constructed using the data from the absorbance titrations and a linear fitting of the data yielded the binding constant (i.e. K = intercept/slope). To investigate the influence of temperature on the binding of TB with ctDNA, absorption titrations were carried out at 20, 30 and 40 ◦ C, respectively. 2.3.2. Fluorescence spectroscopy study Fluorescence spectra were obtained in the range of 640–800 nm upon excitation at 634 nm at the ambient temperature, unless otherwise stated. A 1 cm path length quartz cell was used for all the measurements. The fluorescence quenching experiments were car-
Fig. 1. TB molecule structure.
ried out by the similar operation as absorbance titrations. The fluorescence intensity at the maximum emission of TB (672 nm) was determined for the calculation of quenching constant KSV . The data were plotted according to the Stern–Volmer equation [15]: I0 /I = 1 + KSV [DNA], where I0 and I are the fluorescence intensities in the absence and presence of DNA, respectively, KSV is the Stern–Volmer quenching constant, which is a measure of the efficiency of quenching by DNA. The influence of temperature on quenching was investigated at 20, 30 and 40 ◦ C, respectively. 2.3.3. CD spectroscopy study The induced CD spectra of 0.02 mM TB in presence and absence of 0.005 or 0.05 mM ctDNA were recorded in the range of 500–700 nm. The ionic strength was changed by increasing concentration of NaCl in buffer. The CD spectra of 0.05 mM ctDNA in absence and presence of 0.005 and 0.02 mM TB were recorded in the range of 220–320 nm. All CD measurements were carried out at the ambient temperature with a 1 cm path length rectangular quartz cell. 2.3.4. UV-melting profiles measurements Absorbance changes at 260 nm versus temperature were determined with a heating rate of 0.5 ◦ C min−1 for ctDNA melting experiments. The UV-melting profiles of 0.03 mM ctDNA in absence and presence of 0.003, 0.005 and 0.01 mM TB were measured. 2.3.5. SERS study The sodium citrate protected silver colloid for SERS measurements was prepared as follow [16]: 50 mL of 1 mM AgNO3 aqueous solution was boiled, then 1 mL of a 1% trisodium citrate solution was added with vigorous stirring, and the mixture was boiled with refluxing for 1 h. The silver colloid obtained showed a turbid gray aspect and had surface plasmon band around 429 nm indicating spheroidal shape with a diameter from 30 to 60 nm [17]. The solution for SERS measurements was prepared by diluting the stock solution of TB and ctDNA to suitable concentration with silver colloid solution (a mixture of one volume of silver colloid and nine volumes of NaAc-HAc buffer described in Section 2.2). The SERS spectra of 0.001 mM TB in the absence and presence of 0.01 mM ctDNA were recorded in the range of 200–1800 cm−1 (Fig. 1). 3. Results and discussion 3.1. Multiplex binding modes revealed by absorption spectra Fig. 2a shows the absorption spectra of TB after addition of different amounts of ctDNA. With the [DNA]/[TB] ratio shifting from 0.0 to 3.0, The absorption spectra of TB showed moderate hypochromicity (about 16%) at 634 nm and bathochromic shift (about 10 nm). The hypochromicity is a feature common to stacked chromophores. A transitory dipole formed in one chromophore (A) absorbing light induces a dipole in the opposite direction of a neighboring stacked chromophore (B), which partly counteracts the dipole formed in B. Thus the stacked chromophores have less absorbance than the sum
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
423
Fig. 3. Stern–Volmer quenching plots of TB with increasing concentrations of ctDNA at 20 ◦ C (), 30 ◦ C (), 40 ◦ C (). Inset: fluorescence spectra of 0.01 mM TB after addition of ctDNA at 20 ◦ C. Arrow shows the intensity changes upon increasing ctDNA concentration.
Fig. 2. (a) Absorption spectra of 0.01 mM TB in the presence of 0, 5, 10, 15, 20, 25, 30 M ctDNA. Arrow shows the absorbance changes with increasing amount of ctDNA. Inset: enlarged view of the area inside the circle. (b) Absorption spectra of 0.01 mM TB in the presence of 0.03 mM ctDNA with increasing concentrations of NaCl. Dash line: only TB in buffer; solid lines: TB and ctDNA with 50, 100, 150, 200, 250, 300 mM NaCl. Arrow shows the absorbance changes with increasing concentration of NaCl. Inset: enlarged view of the area inside the circle.
of each chromophore, resulting in hypochromicity [18]. The substantial hypochromicity (about 30%) and large bathochromic shift (about 15 nm) have been widely reported for the DNA intercalators [19–21]. Therefore TB should not be a classical intercalator. It has been reported that porphyrins with varying substituents could interact with DNA in different binding modes (intercalation [22–24], outside binding [23] or partial intercalation [24–25]). The presence of bulky non-coplanar groups in the planar core of porphyrin might hinder the classical intercalation [24]. Therefore the substituents of TB molecule might also result in some hinderance to its intercalation into DNA. There was no distinct isosbestic point during the absorption titration of TB with ctDNA (Fig. 2a), indicating multiplex binding modes [20,26,27]. Since TB has a cationic charge, the electrostatic attraction of TB to the negatively charged sugar-phosphate backbone of ctDNA is inevitable [13]. Therefore, both electrostatic interaction and partial intercalation should contribute to the interaction between TB and ctDNA. Fig. 2b displays the absorption spectra of TB in the presence of ctDNA and after five equal additions of NaCl in succession. At low ionic strength, the addition of NaCl induced an obvious reverse of the hypochromicity, however, with increasing ionic strength the addition of NaCl resulted in much smaller reverse of the hypochromicity. This might be because the electrostatic interaction between TB and ctDNA was favourable at low ionic strength, and weakened with increasing ionic strength, while intercalation was less affected and became predominant at high ionic strength [28].
The influence of temperature on the binding of TB to ctDNA was also studied. The binding constants of TB to ctDNA at 20, 30 and 40 ◦ C were 1.63 × 105 , 5.38 × 104 and 3.95 × 104 M−1 , respectively. Increasing temperature is likely to result in decreasing stability of TB-ctDNA complex. At the same [DNA]/[TB] ratio, the higher the temperature was, the smaller the hypochromicity would be, indicating that higher temperature facilitated the decomposition of TB-ctDNA complex. The binding constant of TB to DNA reported by Jiao et al. [13] is 7.16 × 104 M−1 , which is slightly smaller than that obtained from the absorption titration experiment at room temperature. The difference might come from the different methods adopted and different buffer conditions. 3.2. Fluorescence quenching study The fluorescence emission of TB decreased with the addition of ctDNA, meanwhile the emission maximum shifted from 672 to 678 nm (inset of Fig. 3). It indicates that the binding of TB to ctDNA would change the average local environment of TB chromophores [29]. Since the electronic absorption of ctDNA is at much lower wavelength with respect to the emission band of TB, the fluorescence quenching of TB by ctDNA through energy transfer could be ruled out. It has been reported that the quenching constants of the phenazine and phenothiazine family of dyes by four different DNA bases decreased with increasing oxidation potential of the DNA bases, indicating that the fluorescence quenching was due to electron transfer from the DNA bases to the excited dye [7,30]. Kelly et al. [8] have studied the laser flash spectroscopy of MB with nucleic acids and proposed the fluorescence quenching by electron transfer from the guanine base to the excited singlet state of MB [8,31]. Another phenothiazinium dye thionine has also been suggested the same fluorescence quenching process as MB by DNA [6,7,32]. So it is reasonable to suggest that the phenothiazinium dye TB also underwent fluorescence quenching by electron transfer from the DNA bases, especially guanine, to the excited singlet state of TB. The linear Stern–Volmer quenching plots obtained from the fluorescence titrations of TB by ctDNA indicated merely one type of quenching process, namely static or dynamic quenching, under the experimental condition [33] (Fig. 3). The slope of Stern–Volmer quenching plot, namely Ksv , decreased with the increase of temperature indicating that the fluorescence quenching of TB by ctDNA was static [33]. Non-fluorescence complexes formed upon TB binding to ctDNA resulted in the fluorescence quenching of TB. The Ksv of TB by ctDNA at 20, 30 and 40 ◦ C are 4.18 × 104 , 3.23 × 104 and 2.57 × 104 M−1 , respectively. The increase of the incubation tem-
424
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
Fig. 4. (a) Induced CD spectra of 0.02 mM TB (dash line) in the presence of 0.05 mM ctDNA (dash dot line) with the addition of 50 mM NaCl (solid line). (b) Induced CD spectra of 0.02 mM TB (dash line) in the presence of 0.005 mM ctDNA (dash dot line) with the addition of 50 mM NaCl (solid line).
perature resulted in the destabilization of the TB-ctDNA complex leading to smaller quenching constant of TB by ctDNA. It was agreed with absorption spectroscopic results. 3.3. Binding modes at different [TB]/[DNA] ratios confirmed by the induced CD of TB The induced CD (ICD) spectrum of TB in the visible region can provide the direct information for identifying the binding mode of TB with ctDNA. In the solution, TB was free to rotate having no intrinsic CD signal; however, its orientation was restricted after its binding to ctDNA. This resulted in the ICD signal of TB as shown in Fig. 4a and b. Since ctDNA has no absorption in the visible region, the observed CD signal is merely the ICD signal of TB. The [TB]/[DNA] ratio had obvious effects on the ICD spectra of TB suggesting different binding modes at low and high [TB]/[DNA] ratio. At the [TB]/[DNA] ratio of 0.4 the ICD spectrum consisted of one major negative band at the wavelength around 630 nm. It has been reported that an intercalated chromophore centered near the helix axis of DNA should exhibit the negative ICD for a transition polarized perpendicular to the long axis of the basepairs [34]. Therefore the intercalation is a binding mode in the system. When the ionic strength increased, the ICD spectrum slightly changed, however the negative band remained with about the same wavelength and amplitude (Fig. 4a). It suggests that the intercalation might be not affected by ionic strength of the solution. At the [TB]/[DNA] ratio of 4 the ICD spectrum consisted of a positive band at 537 nm and a negative band at 590 nm, an exciton-like CD spectrum [35],
Fig. 5. (a) CD spectra of 0.05 mM ctDNA in the presence of 0, 5, 20 M TB. Inset: enlarged view of the area inside the circle. Arrows show the ellipticity changes upon increasing TB concentration. (b) UV-melting profiles of 0.03 mM ctDNA in the presence of 0, 3, 5, 10 M TB. Arrow shows profiles with increasing concentrations of TB.
which dramatically weakened even vanished with increasing ionic strength (Fig. 4b). This exciton-like CD signal could be explained as one externally bound TB molecule stacked along the helix surface of ctDNA through coupling with one partially intercalated TB molecule to form an aggregating complex [35]. The externally bound TB were attracted to the phosphate backbone of ctDNA through electrostatic interaction [36]. With the increase of NaCl concentration, the electrostatic interaction dramatically weakened leading to disassociation of the complexes, therefore the exciton-like CD signal vanished. 3.4. B-A transition revealed by CD spectra of ctDNA and UV-melting profiles Generally, ctDNA has the characteristic CD signal in the ultraviolet region, one positive band at 274 nm due to the base stacking and one negative band at 245 nm due to the right-handed B-form helicity [35]. The negative band remained nearly unchanged while the positive band increased remarkably with the addition of TB (Fig. 5a). It suggests that the binding of TB to ctDNA would not apparently disturb the right-handed helicity of ctDNA, however dramatically increase the base stacking degree of ctDNA, inducing conformational change of ctDNA. B-form is the most common conformation of double helix DNA in solution. The molecules that compete with the water molecules coating around the grooves and phosphate skeleton of the double helix may dramatically affect the conformation of DNA, which usually induce DNA into more com-
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
425
Table 1 Raman frequencies in cm−1 of the main bands observed in the SERS spectra of TB with a tentative assignment and their changes upon the addition of DNA. Raman peaks (cm−1 )
Tentative assignment
Change upon addition of DNA
1627sa
Ring CC stretch + ring C1 NC1 stretch Ring CC stretch Ring C1 NC1 asymmetric stretch Ring CC stretch Ring CH bend Ring C6 SC6 stretch + NH2 wag C6 SC6 stretch + HN-C stretch Ring CH wag + CH def (CH3 ) Ring C6 SC6 bend
Hypochromicity
1499m 1445m 1388m 1150w 1066w 900w 763w 495w a
Unchanged Hypochromicity Unchanged Unchanged Unchanged Unchanged Unchanged Unchanged
Relative intensities as: s, strong; m, medium; w, weak.
pact form [27,37,38]. In our experiments, ctDNA was prepared in concentration of 0.05 mM and the conformational changes of ctDNA were determined under the [TB]/[DNA] ratio of 0.4, because higher concentration of TB easily precipitated ctDNA. According to the discussion above, at low [TB]/[DNA] ratio TB partially intercalated into ctDNA and moderately increased the base stacking degree. At high [TB]/[DNA] ratio excessive TB externally stacked along the helix surface of ctDNA via coupling with the partially intercalated ones to compete with the water molecules on the helix surface. Therefore the externally bound TB could effectively reduce the electrostatic repulsion of the negatively charged sugar-phosphate backbone of DNA making the conformation of ctDNA more compact, which resulted in B-A transition of ctDNA even the precipitate of DNA [27,37,38]. The conformational transition of ctDNA was further confirmed by the change of the thermal stability of ctDNA in the absence or presence of TB (Fig. 5b). It was obvious that DNA became more stable in the presence of TB, which was consistent to other intercalators [27]. However the improvement of the thermal stability of ctDNA with the presence of TB was not as large as that could be achieved with the presence of mitoxantrone (MTX) [27]. This might be because TB bound to ctDNA through a weaker association than the intercalation between MTX and DNA. 3.5. Molecular detail of TB partial intercalation into ctDNA revealed by SERS The SERS bands of TB in silver colloid were in the same positions as those in aqueous solution and all bands were almost equally enhanced (data not shown). Therefore TB probably adsorbed on the surface of silver colloid through physisorption without special orientation. The tentative assignment [39–42] is shown in Table 1. DNA had no Raman signal under the experimental condition. It has been reported that the adsorption of DNA on the surface of silver colloid neither disturbed the construction of DNA [43–45] nor perturbed the small molecule-DNA interaction [43]. Therefore SERS was used to gain further information on the interaction between TB and ctDNA. The SERS spectra of TB-ctDNA complex displayed nearly identical band positions except the decrease in intensity of certain bands compared to that of TB (Fig. 6, curve a and b). It suggests that TB could bind to ctDNA via intercalation [43,44,46]. Nabiev et al. [43] have reported that only particular SERS bands of MTX displayed the decrease in intensity upon intercalation into DNA indicating that only those bands related components were involved in intercalation. The difference spectrum between the SERS of TB and TB-ctDNA complex (Fig. 6, curve c) showed that only two bands at 1627 and 1445 cm−1 , which are assigned to the ring CC stretch coupling with ring C1 ,NC1 stretch and the ring C1 ,NC1 asymmetric stretch, respec-
Fig. 6. Surface enhanced Raman spectra of 1 M TB (a) in the presence of 10 M ctDNA (b) and their difference spectrum (c).
tively [39–42], exhibited the decrease in the intensity upon TB binding to ctDNA. This suggests that the ring C1 NC1 component might insert into DNA basepairs. On the other hand, the bands at 1499, 1066, 1030, 900, 763, 495, 451 cm−1 , which are mainly related to the ring C6 SC6 component and the substituents [39–42], were nearly unchangeable upon complexation. So those components were not buried into DNA helix. It is reasonable to propose that TB could partially intercalate into DNA with the C1 NC1 component buried into DNA base pairs, while the steric hindrance resulting from the substituents blocked the insertion from the C6 SC6 side of TB. 4. Conclusions In this paper, the interaction between a potential antitumor durg TB and ctDNA was studied by absorption, fluorescence, CD spectroscopy, UV-melting method and SERS. The experimental results suggest that TB could bind to ctDNA via both electrostatic interaction and partial intercalation. At low ionic strength the electrostatic interaction was favorable; while at high ionic strength the partial intercalation was prominent. The fluorescence quenching of TB by ctDNA was static. The binding modes were also affected by the [TB]/[DNA] ratio. At low [TB]/[DNA] ratio TB mainly partially intercalated into ctDNA with the molecular plane preferably inserting into DNA basepairs from the ring C1 NC1 side. At high [TB]/[DNA] ratio some TB molecules partially intercalated into ctDNA and excessive TB stacked along the helix surface inducing B-A transition of ctDNA. The conformational transition of ctDNA was confirmed by the improvement of the thermal stability of ctDNA. The multiplex binding modes of TB with ctDNA and conformational transition of DNA upon complexation were first reported. An overall picture of TB binding to ctDNA is presented, which should be important in deeper insight into the antitumor activity of TB. Acknowledgements This work was supported by the National Nature Science Foundation of China (No. 20475052) and the Project of Chinese Academy of Sciences (No. KJCX2. Y W. H09). References [1] [2] [3] [4] [5]
R. Palchaudhuri, P.J. Hergenrother, Curr. Opin. Biotechnol. 18 (2007) 497–503. L.R. Ferguson, W.A. Denny, Mutat. Res. 623 (2007) 14–23. L. Strekowski, B. Wilson, Mutat. Res. 623 (2007) 3–13. L.H. Hurley, F.L. Boyd, Trends Pharmacol. Sci. 9 (1988) 402–407. E.M. Tuite, J.M. Kelly, J. Photoch. Photobio. B 21 (1993) 103–124.
426
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
[6] C. Dohno, E.D.A. Stemp, J.K. Barton, J. Am. Chem. Soc. 125 (2003) 9586–9587. [7] G.D. Reid, D.J. Whittaker, M.A. Day, D.A. Turton, V. Kayser, J.M. Kelly, G.S. Beddard, J. Am. Chem. Soc. 124 (2002) 5518–5527. [8] J.M. Kelly, W.J. van der Putten, D.J. McConnell, Photochem. Photobiol. 45 (1987) 167–175. [9] R. Rohs, H. Sklenar, R. Lavery, B. Roder, J. Am. Chem. Soc. 122 (2000) 2860– 2866. [10] F. Harris, L.K. Chatfield, D.A. Phoenix, Curr. Drug Target 6 (2005) 615–627. [11] J.F. Tremblay, S. Dussault, G. Viau, F. Gad, M. Boushira, R. Bissonnette, Photochem. Photobio. Sci. 1 (2002) 852–856. [12] L.H. Chen, Y.T. Nie, L.Z. Liu, H.X. Shen, Anal. Lett. 36 (2003) 107–122. [13] K. Jiao, Q.J. Li, W. Sun, Z.J. Wang, Electroanal. 17 (2005) 997–1002. [14] N. Li, Y. Ma, C. Yang, L.P. Guo, X.R. Yang, Biophys. Chem. 116 (2005) 199–205. [15] C.V. Kumar, E.H. Asuncion, J. Am. Chem. Soc. 115 (1993) 8547–8553. [16] P.C. Lee, D. Melsel, J. Phys. Chem. 86 (1982) 3391–3395. [17] S. Sanchez-Cortes, R.M. Berenguel, A. Madejon, M. Perez-Mendez, Biomacromolecules 3 (2002) 655–660. [18] M.E. Hughes, E. Brandin, J.A. Golovchenko, Nano Lett. 7 (2007) 1191–1194. [19] A.H. Shelton, A. Rodger, D.R. McMillin, Biochemistry-us 46 (2007) 9143–9154. [20] H.C. Becker, B. Norden, J. Am. Chem. Soc. 121 (1999) 11947–11952. [21] N.E. Mukundan, G. Petho, D.W. Dixon, M.S. Kim, L.G. Marzilli, Inorg. Chem. 33 (1994) 4676–4687. [22] R.F. Pasternack, E.J. Gibbs, J.J. Villafranca, Biochemistry-us 22 (1983) 5409–5417. [23] M.J. Carvlin, R.J. Fiel, Nucleic Acids Res. 11 (1983) 6121–6139. [24] K.G. Ford, L.H. Pearl, S. Neidle, Nucleic Acids Res. 15 (1987) 6553–6562. [25] G. Dougherty, J. Inorg. Biochem. 34 (1988) 95–103. [26] E. Tuite, J.M. Kelly, Biopolymers 35 (1995) 419–433. [27] Y.S. Babayan, G. Manzini, Molekulyarnaya Biologiya 24 (1990) 1084–1094.
[28] P. Kara, K. Kerman, D. Ozkan, B. Meric, A. Erdem, Z. Ozkan, M. Ozsoz, Electrochem. Commun. 4 (2002) 705–709. [29] A.G. Krishna, D.V. Kumar, B.M. Khan, S.K. Rawal, K.N. Ganesh, Bba-gen. Subjects 1381 (1998) 104–112. [30] S.D. Choudhury, S. Basu, Spectrochim. Acta A 62 (2005) 736–739. [31] B.S. Fujimoto, J.B. Clendenning, J.J. Delrow, P.J. Heath, M. Schurr, J. Phys. Chem. 98 (1994) 6633–6643. [32] C. Hecht, J. Friedrich, T.C. Chang, J. Phys. Chem. B 108 (2004) 10241–10244. [33] W.Y. Zhong, J.S. Yu, W.L. Huang, K.Y. Ni, Y.Q. Liang, Biopolymers 62 (2001) 315–323. [34] R. Lyng, T. Hard, B. Norden, Biopolymers 26 (1987) 1327–1345. [35] X. Jiang, L. Shang, Z.X. Wang, S.J. Dong, Biophys. Chem. 118 (2005) 42–50. [36] Y. Wang, A. Zhou, J. Photoch. Photobio. A: Chem. 190 (2007) 121–127. [37] H.F. Wang, P. Yang, Y.N. Tian, Z.G. Zhang, C.G. Zhao, J. Inorg. Biochem. 68 (1997) 117–121. [38] N.C. Stellwagen, Biochemistry-us 23 (1984) 6311–6319. [39] M. Bolboaca, T. Iliescu, W. Kiefer, Chem. Phys. 298 (2004) 87–95. [40] W. Xu, M. Aydin, S. Zakia, D.L. Akins, J. Phys. Chem. B 108 (2004) 5588–5593. [41] S.H.D. Nicolai, P.R.P. Rodrigues, S.M.L. Agostinho, J.C. Rubim, J. Electroanal. Chem. 527 (2002) 103–111. [42] D.H. Pan, D.L. Phillips, J. Phys. Chem. A 103 (1999) 4737–4743. [43] I. Nabiev, A. Baranov, I. Chourpa, A. Beljebbar, G.D. Sockalingum, M. Manfait, J. Phys. Chem. 99 (1995) 1608–1613. [44] F. Zimmermann, B. Hossenfelder, J.C. Panitz, A. Wokaun, J. Phys. Chem. 98 (1994) 12796–12804. [45] I. Nabiev, I. Chourpa, M. Manfait, J. Phys. Chem. 98 (1994) 1344–1350. [46] W. Chen, C. Turro, L.A. Friedman, J.K. Barton, N.J. Turro, J. Phys. Chem. B 101 (1997) 6995–7000.
Contents lists available at ScienceDirect
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa
Multiplex binding modes of toluidine blue with calf thymus DNA and conformational transition of DNA revealed by spectroscopic studies Juan Wang a,b , Xiurong Yang a,∗ a State Key Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun 130022, PR China b Graduate School of the Chinese Academy of Sciences, Beijing 100039, PR China
a r t i c l e
i n f o
Article history: Received 13 April 2008 Received in revised form 30 April 2009 Accepted 12 June 2009 Keywords: Toluidine blue (TB) DNA Partial intercalation Electrostatic interaction Conformational transition
a b s t r a c t It is noteworthy to understand the details of interactions between antitumor drugs and DNA because the binding modes and affinities affect their antitumor activities. Here, The interaction of toluidine blue (TB), a potential antitumor drug for photodynamic therapy of tumor, with calf thymus DNA (ctDNA) was explored by UV–vis, fluorescence, circular dichroism (CD) spectroscopy, UV-melting method and surfaceenhance Raman spectroscopy (SERS). The experimental results suggest that TB could bind to ctDNA via both electrostatic interaction and partial intercalation. The fluorescence quenching of TB by ctDNA was static and due to electron transfer from bases to the excited singlet state of TB. At low [TB]/[DNA] ratio, TB mainly partially intercalated into ctDNA resulting in the slight increase of base stacking degree; at high [TB]/[DNA] ratio, excessive TB externally stacked along the helix surface via coupling with partially intercalated ones, thereby inducing B-A transition of ctDNA. The conformational transition of DNA was confirmed by the obvious improvement of the thermal stability of ctDNA. The SERS spectra suggest that TB could partially intercalate into DNA basepairs with its ring C1 NC1 side buried. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Nucleic acids as the carriers of genetic information provide important targets for many biomacromolecules and small molecules. The interactions between DNA and small molecules are of special importance in chemotherapeutic applications and mutagenesis studies [1–3]. Many small molecular anticancer drugs are known to exert their biological activity through interacting with DNA [1–4], while lots of carcinogens and potent mutagens attack nucleic acids thereby changing DNA replication or reparation and lead to uncontrolled growth of tumor cells [2]. Although a lot of important DNA-binding compounds have been explored in phenotypic, cell-based screens, in vitro studies are needed for comprehensive understanding of the interaction mechanism [1]. The investigations on the nature of interactions between small molecules and nucleic acids, not only shed light on understanding the chemical basis for the antitumor mechanism of drugs and the carcinogenicity of pollutants, but also give chemists and pharmacologists opportunities to discover or design new drug candidates. Phenothiazinium dyes are photocytotoxic in living systems, and even cause photoinduced mutagenic effects when some cells
∗ Corresponding author. E-mail address: [email protected] (X. Yang). 1386-1425/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2009.06.038
survive from damages [5]. DNA is an important target in the phenothiazinium dye-photosensitized biological damage, and guanine residues are particularly accessible [6–8]. Moreover, some phenothiazinium photosensitizers, such as toluidine blue (TB) and methylene blue (MB), have the promising applications in photodynamic therapy of tumor [9–11]. These potential anti-tumor drugs can kill cancer cells through indirect or direct reactions with DNA [5]. TB has been widely used for detection and delineation of some cancers and displays good safety in humans. Furthermore it appears some selectivity for cancer cells in vivo, therefore TB has been explored as a potential anti-tumor drug [11]. It is important to obtain comprehensive understanding of the DNA-binding properties of TB. To our knowledge, there are only a few investigations on the interaction between TB and DNA. Resonance light scattering spectra of TB with the presence of DNA indicated that TB could aggregate along DNA surface [12]. Jiao et al. [13] have studied the interaction of TB with DNA by electrochemistry and proposed the electrostatic binding of TB to DNA. Since TB has the planar ring structure, it is quite possible that TB would bind to DNA via intercalation. However details on the binding mode of TB to DNA and the conformational changes of DNA are still unrevealed, which has great relevance to the bioactivity of TB. In this study, the interaction between TB and ctDNA was investigated by UV–vis, fluorescence, CD, surfaceenhanced Raman spectroscopy (SERS) and UV-melting method.
422
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
This research is helpful for evaluating the antitumor activity of TB. 2. Experimental 2.1. Equipments The absorption spectra were obtained by Cary 50 UV–vis spectrophotometer (Varian, USA). All fluorescence spectra were measured on a PerkinElmer LS-55 luminescence spectrometer. The CD spectra were recorded using a 62A DS CD spectrometer (AVIV, USA). UV-melting profiles of ctDNA were obtained by Cary 500 UV–vis-NIR spectrophotometer (Varian, USA) equipped with a quartz cell of 1 cm path length and a thermoelectrically controlled cell holder. The SERS spectra were measured on a Renishaw 2000 model confocal microscopy Raman spectrometer (Renishaw Ltd., Gloucestershire, U.K.) using radiation of 514.5 nm for the SERS excitation. 2.2. Reagents TB was obtained from Acros Organic Co. (USA) without further purification. Deoxyribonucleic acid sodium salt from calf thymus (type I) was purchased from Sigma Chemical Co. (USA). All reagents used were of analytical grade and aqueous solutions were prepared using doubly distilled deionized water. The TB stock solution of 1 mM was kept away from light to avoid photochemical decomposition and was diluted just before use. The DNA concentration (basepairs) was spectroscopically determined using molar absorption coefficient of 13,200 cm−1 M−1 at 260 nm. The ratio of the absorbance at 260 and 280 nm was larger than 1.8, indicating that ctDNA was sufficiently free of protein. All the experiments involving interaction of TB with ctDNA were carried out in buffer containing 10 mM NaAc-HAc and 50 mM NaCl (pH 5) unless otherwise stated and the ionic strength was modulated by adding an appropriate volume of stock solution of NaCl (5.0 M in doubly distilled deionized water). 2.3. Procedures 2.3.1. Absorption spectroscopy study Absorption spectra were recorded in the range of 500–800 nm using a quartz cell of 1 cm path length at the ambient temperature, unless otherwise stated. The absorbance titrations were performed by keeping the concentration of TB fixed at 0.01 mM while gradually increasing the concentration of ctDNA from 0.002 to 0.03 mM. From the absorption titration data, the binding constant was determined according to [14]. A0 /(A0 − A) = εf /(εf − εa ) + εf /(εf − εa ) × (1/K[DNA]) where A0 and A are the absorbances of drug in the absence and presence of DNA, εf and εa are the absorption coefficients of drug and its complex with DNA, K is the binding constant between drug and DNA, respectively. The plot of A0 /(A − A0 ) versus 1/[DNA] was constructed using the data from the absorbance titrations and a linear fitting of the data yielded the binding constant (i.e. K = intercept/slope). To investigate the influence of temperature on the binding of TB with ctDNA, absorption titrations were carried out at 20, 30 and 40 ◦ C, respectively. 2.3.2. Fluorescence spectroscopy study Fluorescence spectra were obtained in the range of 640–800 nm upon excitation at 634 nm at the ambient temperature, unless otherwise stated. A 1 cm path length quartz cell was used for all the measurements. The fluorescence quenching experiments were car-
Fig. 1. TB molecule structure.
ried out by the similar operation as absorbance titrations. The fluorescence intensity at the maximum emission of TB (672 nm) was determined for the calculation of quenching constant KSV . The data were plotted according to the Stern–Volmer equation [15]: I0 /I = 1 + KSV [DNA], where I0 and I are the fluorescence intensities in the absence and presence of DNA, respectively, KSV is the Stern–Volmer quenching constant, which is a measure of the efficiency of quenching by DNA. The influence of temperature on quenching was investigated at 20, 30 and 40 ◦ C, respectively. 2.3.3. CD spectroscopy study The induced CD spectra of 0.02 mM TB in presence and absence of 0.005 or 0.05 mM ctDNA were recorded in the range of 500–700 nm. The ionic strength was changed by increasing concentration of NaCl in buffer. The CD spectra of 0.05 mM ctDNA in absence and presence of 0.005 and 0.02 mM TB were recorded in the range of 220–320 nm. All CD measurements were carried out at the ambient temperature with a 1 cm path length rectangular quartz cell. 2.3.4. UV-melting profiles measurements Absorbance changes at 260 nm versus temperature were determined with a heating rate of 0.5 ◦ C min−1 for ctDNA melting experiments. The UV-melting profiles of 0.03 mM ctDNA in absence and presence of 0.003, 0.005 and 0.01 mM TB were measured. 2.3.5. SERS study The sodium citrate protected silver colloid for SERS measurements was prepared as follow [16]: 50 mL of 1 mM AgNO3 aqueous solution was boiled, then 1 mL of a 1% trisodium citrate solution was added with vigorous stirring, and the mixture was boiled with refluxing for 1 h. The silver colloid obtained showed a turbid gray aspect and had surface plasmon band around 429 nm indicating spheroidal shape with a diameter from 30 to 60 nm [17]. The solution for SERS measurements was prepared by diluting the stock solution of TB and ctDNA to suitable concentration with silver colloid solution (a mixture of one volume of silver colloid and nine volumes of NaAc-HAc buffer described in Section 2.2). The SERS spectra of 0.001 mM TB in the absence and presence of 0.01 mM ctDNA were recorded in the range of 200–1800 cm−1 (Fig. 1). 3. Results and discussion 3.1. Multiplex binding modes revealed by absorption spectra Fig. 2a shows the absorption spectra of TB after addition of different amounts of ctDNA. With the [DNA]/[TB] ratio shifting from 0.0 to 3.0, The absorption spectra of TB showed moderate hypochromicity (about 16%) at 634 nm and bathochromic shift (about 10 nm). The hypochromicity is a feature common to stacked chromophores. A transitory dipole formed in one chromophore (A) absorbing light induces a dipole in the opposite direction of a neighboring stacked chromophore (B), which partly counteracts the dipole formed in B. Thus the stacked chromophores have less absorbance than the sum
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
423
Fig. 3. Stern–Volmer quenching plots of TB with increasing concentrations of ctDNA at 20 ◦ C (), 30 ◦ C (), 40 ◦ C (). Inset: fluorescence spectra of 0.01 mM TB after addition of ctDNA at 20 ◦ C. Arrow shows the intensity changes upon increasing ctDNA concentration.
Fig. 2. (a) Absorption spectra of 0.01 mM TB in the presence of 0, 5, 10, 15, 20, 25, 30 M ctDNA. Arrow shows the absorbance changes with increasing amount of ctDNA. Inset: enlarged view of the area inside the circle. (b) Absorption spectra of 0.01 mM TB in the presence of 0.03 mM ctDNA with increasing concentrations of NaCl. Dash line: only TB in buffer; solid lines: TB and ctDNA with 50, 100, 150, 200, 250, 300 mM NaCl. Arrow shows the absorbance changes with increasing concentration of NaCl. Inset: enlarged view of the area inside the circle.
of each chromophore, resulting in hypochromicity [18]. The substantial hypochromicity (about 30%) and large bathochromic shift (about 15 nm) have been widely reported for the DNA intercalators [19–21]. Therefore TB should not be a classical intercalator. It has been reported that porphyrins with varying substituents could interact with DNA in different binding modes (intercalation [22–24], outside binding [23] or partial intercalation [24–25]). The presence of bulky non-coplanar groups in the planar core of porphyrin might hinder the classical intercalation [24]. Therefore the substituents of TB molecule might also result in some hinderance to its intercalation into DNA. There was no distinct isosbestic point during the absorption titration of TB with ctDNA (Fig. 2a), indicating multiplex binding modes [20,26,27]. Since TB has a cationic charge, the electrostatic attraction of TB to the negatively charged sugar-phosphate backbone of ctDNA is inevitable [13]. Therefore, both electrostatic interaction and partial intercalation should contribute to the interaction between TB and ctDNA. Fig. 2b displays the absorption spectra of TB in the presence of ctDNA and after five equal additions of NaCl in succession. At low ionic strength, the addition of NaCl induced an obvious reverse of the hypochromicity, however, with increasing ionic strength the addition of NaCl resulted in much smaller reverse of the hypochromicity. This might be because the electrostatic interaction between TB and ctDNA was favourable at low ionic strength, and weakened with increasing ionic strength, while intercalation was less affected and became predominant at high ionic strength [28].
The influence of temperature on the binding of TB to ctDNA was also studied. The binding constants of TB to ctDNA at 20, 30 and 40 ◦ C were 1.63 × 105 , 5.38 × 104 and 3.95 × 104 M−1 , respectively. Increasing temperature is likely to result in decreasing stability of TB-ctDNA complex. At the same [DNA]/[TB] ratio, the higher the temperature was, the smaller the hypochromicity would be, indicating that higher temperature facilitated the decomposition of TB-ctDNA complex. The binding constant of TB to DNA reported by Jiao et al. [13] is 7.16 × 104 M−1 , which is slightly smaller than that obtained from the absorption titration experiment at room temperature. The difference might come from the different methods adopted and different buffer conditions. 3.2. Fluorescence quenching study The fluorescence emission of TB decreased with the addition of ctDNA, meanwhile the emission maximum shifted from 672 to 678 nm (inset of Fig. 3). It indicates that the binding of TB to ctDNA would change the average local environment of TB chromophores [29]. Since the electronic absorption of ctDNA is at much lower wavelength with respect to the emission band of TB, the fluorescence quenching of TB by ctDNA through energy transfer could be ruled out. It has been reported that the quenching constants of the phenazine and phenothiazine family of dyes by four different DNA bases decreased with increasing oxidation potential of the DNA bases, indicating that the fluorescence quenching was due to electron transfer from the DNA bases to the excited dye [7,30]. Kelly et al. [8] have studied the laser flash spectroscopy of MB with nucleic acids and proposed the fluorescence quenching by electron transfer from the guanine base to the excited singlet state of MB [8,31]. Another phenothiazinium dye thionine has also been suggested the same fluorescence quenching process as MB by DNA [6,7,32]. So it is reasonable to suggest that the phenothiazinium dye TB also underwent fluorescence quenching by electron transfer from the DNA bases, especially guanine, to the excited singlet state of TB. The linear Stern–Volmer quenching plots obtained from the fluorescence titrations of TB by ctDNA indicated merely one type of quenching process, namely static or dynamic quenching, under the experimental condition [33] (Fig. 3). The slope of Stern–Volmer quenching plot, namely Ksv , decreased with the increase of temperature indicating that the fluorescence quenching of TB by ctDNA was static [33]. Non-fluorescence complexes formed upon TB binding to ctDNA resulted in the fluorescence quenching of TB. The Ksv of TB by ctDNA at 20, 30 and 40 ◦ C are 4.18 × 104 , 3.23 × 104 and 2.57 × 104 M−1 , respectively. The increase of the incubation tem-
424
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
Fig. 4. (a) Induced CD spectra of 0.02 mM TB (dash line) in the presence of 0.05 mM ctDNA (dash dot line) with the addition of 50 mM NaCl (solid line). (b) Induced CD spectra of 0.02 mM TB (dash line) in the presence of 0.005 mM ctDNA (dash dot line) with the addition of 50 mM NaCl (solid line).
perature resulted in the destabilization of the TB-ctDNA complex leading to smaller quenching constant of TB by ctDNA. It was agreed with absorption spectroscopic results. 3.3. Binding modes at different [TB]/[DNA] ratios confirmed by the induced CD of TB The induced CD (ICD) spectrum of TB in the visible region can provide the direct information for identifying the binding mode of TB with ctDNA. In the solution, TB was free to rotate having no intrinsic CD signal; however, its orientation was restricted after its binding to ctDNA. This resulted in the ICD signal of TB as shown in Fig. 4a and b. Since ctDNA has no absorption in the visible region, the observed CD signal is merely the ICD signal of TB. The [TB]/[DNA] ratio had obvious effects on the ICD spectra of TB suggesting different binding modes at low and high [TB]/[DNA] ratio. At the [TB]/[DNA] ratio of 0.4 the ICD spectrum consisted of one major negative band at the wavelength around 630 nm. It has been reported that an intercalated chromophore centered near the helix axis of DNA should exhibit the negative ICD for a transition polarized perpendicular to the long axis of the basepairs [34]. Therefore the intercalation is a binding mode in the system. When the ionic strength increased, the ICD spectrum slightly changed, however the negative band remained with about the same wavelength and amplitude (Fig. 4a). It suggests that the intercalation might be not affected by ionic strength of the solution. At the [TB]/[DNA] ratio of 4 the ICD spectrum consisted of a positive band at 537 nm and a negative band at 590 nm, an exciton-like CD spectrum [35],
Fig. 5. (a) CD spectra of 0.05 mM ctDNA in the presence of 0, 5, 20 M TB. Inset: enlarged view of the area inside the circle. Arrows show the ellipticity changes upon increasing TB concentration. (b) UV-melting profiles of 0.03 mM ctDNA in the presence of 0, 3, 5, 10 M TB. Arrow shows profiles with increasing concentrations of TB.
which dramatically weakened even vanished with increasing ionic strength (Fig. 4b). This exciton-like CD signal could be explained as one externally bound TB molecule stacked along the helix surface of ctDNA through coupling with one partially intercalated TB molecule to form an aggregating complex [35]. The externally bound TB were attracted to the phosphate backbone of ctDNA through electrostatic interaction [36]. With the increase of NaCl concentration, the electrostatic interaction dramatically weakened leading to disassociation of the complexes, therefore the exciton-like CD signal vanished. 3.4. B-A transition revealed by CD spectra of ctDNA and UV-melting profiles Generally, ctDNA has the characteristic CD signal in the ultraviolet region, one positive band at 274 nm due to the base stacking and one negative band at 245 nm due to the right-handed B-form helicity [35]. The negative band remained nearly unchanged while the positive band increased remarkably with the addition of TB (Fig. 5a). It suggests that the binding of TB to ctDNA would not apparently disturb the right-handed helicity of ctDNA, however dramatically increase the base stacking degree of ctDNA, inducing conformational change of ctDNA. B-form is the most common conformation of double helix DNA in solution. The molecules that compete with the water molecules coating around the grooves and phosphate skeleton of the double helix may dramatically affect the conformation of DNA, which usually induce DNA into more com-
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
425
Table 1 Raman frequencies in cm−1 of the main bands observed in the SERS spectra of TB with a tentative assignment and their changes upon the addition of DNA. Raman peaks (cm−1 )
Tentative assignment
Change upon addition of DNA
1627sa
Ring CC stretch + ring C1 NC1 stretch Ring CC stretch Ring C1 NC1 asymmetric stretch Ring CC stretch Ring CH bend Ring C6 SC6 stretch + NH2 wag C6 SC6 stretch + HN-C stretch Ring CH wag + CH def (CH3 ) Ring C6 SC6 bend
Hypochromicity
1499m 1445m 1388m 1150w 1066w 900w 763w 495w a
Unchanged Hypochromicity Unchanged Unchanged Unchanged Unchanged Unchanged Unchanged
Relative intensities as: s, strong; m, medium; w, weak.
pact form [27,37,38]. In our experiments, ctDNA was prepared in concentration of 0.05 mM and the conformational changes of ctDNA were determined under the [TB]/[DNA] ratio of 0.4, because higher concentration of TB easily precipitated ctDNA. According to the discussion above, at low [TB]/[DNA] ratio TB partially intercalated into ctDNA and moderately increased the base stacking degree. At high [TB]/[DNA] ratio excessive TB externally stacked along the helix surface of ctDNA via coupling with the partially intercalated ones to compete with the water molecules on the helix surface. Therefore the externally bound TB could effectively reduce the electrostatic repulsion of the negatively charged sugar-phosphate backbone of DNA making the conformation of ctDNA more compact, which resulted in B-A transition of ctDNA even the precipitate of DNA [27,37,38]. The conformational transition of ctDNA was further confirmed by the change of the thermal stability of ctDNA in the absence or presence of TB (Fig. 5b). It was obvious that DNA became more stable in the presence of TB, which was consistent to other intercalators [27]. However the improvement of the thermal stability of ctDNA with the presence of TB was not as large as that could be achieved with the presence of mitoxantrone (MTX) [27]. This might be because TB bound to ctDNA through a weaker association than the intercalation between MTX and DNA. 3.5. Molecular detail of TB partial intercalation into ctDNA revealed by SERS The SERS bands of TB in silver colloid were in the same positions as those in aqueous solution and all bands were almost equally enhanced (data not shown). Therefore TB probably adsorbed on the surface of silver colloid through physisorption without special orientation. The tentative assignment [39–42] is shown in Table 1. DNA had no Raman signal under the experimental condition. It has been reported that the adsorption of DNA on the surface of silver colloid neither disturbed the construction of DNA [43–45] nor perturbed the small molecule-DNA interaction [43]. Therefore SERS was used to gain further information on the interaction between TB and ctDNA. The SERS spectra of TB-ctDNA complex displayed nearly identical band positions except the decrease in intensity of certain bands compared to that of TB (Fig. 6, curve a and b). It suggests that TB could bind to ctDNA via intercalation [43,44,46]. Nabiev et al. [43] have reported that only particular SERS bands of MTX displayed the decrease in intensity upon intercalation into DNA indicating that only those bands related components were involved in intercalation. The difference spectrum between the SERS of TB and TB-ctDNA complex (Fig. 6, curve c) showed that only two bands at 1627 and 1445 cm−1 , which are assigned to the ring CC stretch coupling with ring C1 ,NC1 stretch and the ring C1 ,NC1 asymmetric stretch, respec-
Fig. 6. Surface enhanced Raman spectra of 1 M TB (a) in the presence of 10 M ctDNA (b) and their difference spectrum (c).
tively [39–42], exhibited the decrease in the intensity upon TB binding to ctDNA. This suggests that the ring C1 NC1 component might insert into DNA basepairs. On the other hand, the bands at 1499, 1066, 1030, 900, 763, 495, 451 cm−1 , which are mainly related to the ring C6 SC6 component and the substituents [39–42], were nearly unchangeable upon complexation. So those components were not buried into DNA helix. It is reasonable to propose that TB could partially intercalate into DNA with the C1 NC1 component buried into DNA base pairs, while the steric hindrance resulting from the substituents blocked the insertion from the C6 SC6 side of TB. 4. Conclusions In this paper, the interaction between a potential antitumor durg TB and ctDNA was studied by absorption, fluorescence, CD spectroscopy, UV-melting method and SERS. The experimental results suggest that TB could bind to ctDNA via both electrostatic interaction and partial intercalation. At low ionic strength the electrostatic interaction was favorable; while at high ionic strength the partial intercalation was prominent. The fluorescence quenching of TB by ctDNA was static. The binding modes were also affected by the [TB]/[DNA] ratio. At low [TB]/[DNA] ratio TB mainly partially intercalated into ctDNA with the molecular plane preferably inserting into DNA basepairs from the ring C1 NC1 side. At high [TB]/[DNA] ratio some TB molecules partially intercalated into ctDNA and excessive TB stacked along the helix surface inducing B-A transition of ctDNA. The conformational transition of ctDNA was confirmed by the improvement of the thermal stability of ctDNA. The multiplex binding modes of TB with ctDNA and conformational transition of DNA upon complexation were first reported. An overall picture of TB binding to ctDNA is presented, which should be important in deeper insight into the antitumor activity of TB. Acknowledgements This work was supported by the National Nature Science Foundation of China (No. 20475052) and the Project of Chinese Academy of Sciences (No. KJCX2. Y W. H09). References [1] [2] [3] [4] [5]
R. Palchaudhuri, P.J. Hergenrother, Curr. Opin. Biotechnol. 18 (2007) 497–503. L.R. Ferguson, W.A. Denny, Mutat. Res. 623 (2007) 14–23. L. Strekowski, B. Wilson, Mutat. Res. 623 (2007) 3–13. L.H. Hurley, F.L. Boyd, Trends Pharmacol. Sci. 9 (1988) 402–407. E.M. Tuite, J.M. Kelly, J. Photoch. Photobio. B 21 (1993) 103–124.
426
J. Wang, X. Yang / Spectrochimica Acta Part A 74 (2009) 421–426
[6] C. Dohno, E.D.A. Stemp, J.K. Barton, J. Am. Chem. Soc. 125 (2003) 9586–9587. [7] G.D. Reid, D.J. Whittaker, M.A. Day, D.A. Turton, V. Kayser, J.M. Kelly, G.S. Beddard, J. Am. Chem. Soc. 124 (2002) 5518–5527. [8] J.M. Kelly, W.J. van der Putten, D.J. McConnell, Photochem. Photobiol. 45 (1987) 167–175. [9] R. Rohs, H. Sklenar, R. Lavery, B. Roder, J. Am. Chem. Soc. 122 (2000) 2860– 2866. [10] F. Harris, L.K. Chatfield, D.A. Phoenix, Curr. Drug Target 6 (2005) 615–627. [11] J.F. Tremblay, S. Dussault, G. Viau, F. Gad, M. Boushira, R. Bissonnette, Photochem. Photobio. Sci. 1 (2002) 852–856. [12] L.H. Chen, Y.T. Nie, L.Z. Liu, H.X. Shen, Anal. Lett. 36 (2003) 107–122. [13] K. Jiao, Q.J. Li, W. Sun, Z.J. Wang, Electroanal. 17 (2005) 997–1002. [14] N. Li, Y. Ma, C. Yang, L.P. Guo, X.R. Yang, Biophys. Chem. 116 (2005) 199–205. [15] C.V. Kumar, E.H. Asuncion, J. Am. Chem. Soc. 115 (1993) 8547–8553. [16] P.C. Lee, D. Melsel, J. Phys. Chem. 86 (1982) 3391–3395. [17] S. Sanchez-Cortes, R.M. Berenguel, A. Madejon, M. Perez-Mendez, Biomacromolecules 3 (2002) 655–660. [18] M.E. Hughes, E. Brandin, J.A. Golovchenko, Nano Lett. 7 (2007) 1191–1194. [19] A.H. Shelton, A. Rodger, D.R. McMillin, Biochemistry-us 46 (2007) 9143–9154. [20] H.C. Becker, B. Norden, J. Am. Chem. Soc. 121 (1999) 11947–11952. [21] N.E. Mukundan, G. Petho, D.W. Dixon, M.S. Kim, L.G. Marzilli, Inorg. Chem. 33 (1994) 4676–4687. [22] R.F. Pasternack, E.J. Gibbs, J.J. Villafranca, Biochemistry-us 22 (1983) 5409–5417. [23] M.J. Carvlin, R.J. Fiel, Nucleic Acids Res. 11 (1983) 6121–6139. [24] K.G. Ford, L.H. Pearl, S. Neidle, Nucleic Acids Res. 15 (1987) 6553–6562. [25] G. Dougherty, J. Inorg. Biochem. 34 (1988) 95–103. [26] E. Tuite, J.M. Kelly, Biopolymers 35 (1995) 419–433. [27] Y.S. Babayan, G. Manzini, Molekulyarnaya Biologiya 24 (1990) 1084–1094.
[28] P. Kara, K. Kerman, D. Ozkan, B. Meric, A. Erdem, Z. Ozkan, M. Ozsoz, Electrochem. Commun. 4 (2002) 705–709. [29] A.G. Krishna, D.V. Kumar, B.M. Khan, S.K. Rawal, K.N. Ganesh, Bba-gen. Subjects 1381 (1998) 104–112. [30] S.D. Choudhury, S. Basu, Spectrochim. Acta A 62 (2005) 736–739. [31] B.S. Fujimoto, J.B. Clendenning, J.J. Delrow, P.J. Heath, M. Schurr, J. Phys. Chem. 98 (1994) 6633–6643. [32] C. Hecht, J. Friedrich, T.C. Chang, J. Phys. Chem. B 108 (2004) 10241–10244. [33] W.Y. Zhong, J.S. Yu, W.L. Huang, K.Y. Ni, Y.Q. Liang, Biopolymers 62 (2001) 315–323. [34] R. Lyng, T. Hard, B. Norden, Biopolymers 26 (1987) 1327–1345. [35] X. Jiang, L. Shang, Z.X. Wang, S.J. Dong, Biophys. Chem. 118 (2005) 42–50. [36] Y. Wang, A. Zhou, J. Photoch. Photobio. A: Chem. 190 (2007) 121–127. [37] H.F. Wang, P. Yang, Y.N. Tian, Z.G. Zhang, C.G. Zhao, J. Inorg. Biochem. 68 (1997) 117–121. [38] N.C. Stellwagen, Biochemistry-us 23 (1984) 6311–6319. [39] M. Bolboaca, T. Iliescu, W. Kiefer, Chem. Phys. 298 (2004) 87–95. [40] W. Xu, M. Aydin, S. Zakia, D.L. Akins, J. Phys. Chem. B 108 (2004) 5588–5593. [41] S.H.D. Nicolai, P.R.P. Rodrigues, S.M.L. Agostinho, J.C. Rubim, J. Electroanal. Chem. 527 (2002) 103–111. [42] D.H. Pan, D.L. Phillips, J. Phys. Chem. A 103 (1999) 4737–4743. [43] I. Nabiev, A. Baranov, I. Chourpa, A. Beljebbar, G.D. Sockalingum, M. Manfait, J. Phys. Chem. 99 (1995) 1608–1613. [44] F. Zimmermann, B. Hossenfelder, J.C. Panitz, A. Wokaun, J. Phys. Chem. 98 (1994) 12796–12804. [45] I. Nabiev, I. Chourpa, M. Manfait, J. Phys. Chem. 98 (1994) 1344–1350. [46] W. Chen, C. Turro, L.A. Friedman, J.K. Barton, N.J. Turro, J. Phys. Chem. B 101 (1997) 6995–7000.