Interaction Between 14mer DNA Oligonucleotide and Cationic

8824

J. Phys. Chem. B 2008, 112, 8824–8831

Interaction Between 14mer DNA Oligonucleotide and Cationic Surfactants of Various Chain Lengths Vaibhav M. Jadhav, Rebecca Valaske, and Souvik Maiti* Institute for Genomics and IntegratiVe Biology, CSIR, Mall Road, Delhi 110 007, India ReceiVed: October 6, 2007; ReVised Manuscript ReceiVed: April 23, 2008

In the recent genomic era, a novel gene silencing approach has been introduced based on the use of small synthetic oligonucleotides, such as antisense RNAs, siRNAs, to inhibit the expression of a specific target gene. Successful implementation of this methodology calls for the development of efficient systems to deliver small oligonucleotides into the cells using various natural and synthetic cationic agents. While extensive studies have focused on the interaction of various natural and synthetic cationic surfactants with long DNA, less attention has been paid to surfactant interaction with small oligonuclotides. In this study, the interaction between 14mer double stranded DNA and alkyltrimethylammonium bromides of C16 (cetyl, CTAB), C14 (tetradecyl, TTAB), and C12 (dodecyl, DTAB) chain lengths was investigated at different charge ratios by gel electrophoresis, ethidium bromide exclusion, circular dichroism, and UV melting. Our gel studies at 1 µM oligonucleotide concentration showed that CTAB, TTAB, and DTAB neutralize the oligonucleotides at a charge ratio (Z() of 1, 14, and 50, respectively. At lower charge ratios, CTAB and TTAB interact with oligonucleotides, and the complexes show electrophoretic mobility shifts in the gel, while such mobility shifts were completely absent in the case of DTAB. UV melting experiments revealed that interaction with all three surfactants increased the thermostability of the oligonucleotide. The extent of thermal stabilization was highest in the case of CTAB, moderate in the case of TTAB, and extremely low in the case of DTAB. Oligonucleotides within fully neutralized complexes denatured at further higher temperatures, and again, stabilization was the highest in the case of CTAB followed by TTAB and DTAB, hence revealing that the oligonucleotides interacted more strongly with CTAB than with the other two surfactants. Ethidium bromide exclusion studies also supported our UV melting studies, confirming that CTAB binds most strongly to the oligonucleotide. CD titrations of oligonucleotides with increasing amounts of surfactants revealed common spectral patterns consisting of the progressive loss of CD signals for native helical DNA conformations. Overall, our results demonstrate that interaction between oligonucleotides and cationic surfactants, although qualitatively similar to long double stranded DNA, shows subtle differences that need to be understood to improve small oligonucleotide delivery into the cells by using common delivery agents that have been used to deliver long pieces of DNA. Introduction An important objective of modern biology is to develop efficient methods for therapeutic gene delivery, allowing the transport of DNA or other genetic material through the cell and its release into the nucleus.1–7 One widely used strategy involves the use of viral vectors but has raised unresolved safety concerns. Another promising approach is to use DNA-cationic agent complexes formed as a result of ionic interactions between the cationic groups of different types of cationic amphiphiles and the negatively charged phosphate groups of DNA.5–7 The interactions between DNA and cationic surfactants have received particular attention because of their easy availability, significance in biomedical applications, as well as for a better understanding of DNA behavior in living cells.8–11 With the previous consideration, the basic understanding of interactions between negatively charged DNA and oppositely charged species becomes important.12–20 Numerous studies have investigated the interactions of cetyltrimethylammonium bromide (CTAB) with DNA, where it was observed that CTAB is an effective DNA compacting agent.19–22 It was demonstrated through the use of * To whom correspondence should be addressed. E-mail: [email protected]; fax: +91 11 27667471.

cationic surfactant-selective electrodes that cationic surfactants bind to the negatively charged DNA macromolecule in a cooperative manner driven by hydrophobic forces.23–26 Studies have shown that DNA chains undergo a coil-globule transition with increasing concentration of the surfactant.13,27 The phenomenon is observed even with very low concentrations of cationic surfactant, well below its critical micelle concentration (cmc).28,29 It also was observed that for surfactant concentrations below the cmc, free/unbound DNA chains coexist with surfactant-saturated DNA chains. A more direct observation of the surfactant-induced conformational changes of a single DNA chain and the coexistence of two states (i.e., free/unbound DNA chains and surfactant-saturated DNA chains) could be obtained using fluorescence microscopy that allows direct visualization of single DNA molecules in solution. The technique earlier was used to study the conformational behavior of high molecular weight DNA in the presence of cosolutes such as surfactants or polyamines.30 Dynamic light scattering (DLS) measurements also provide information on the surfactant-induced conformational changes of DNA molecules based on the change in the translational diffusion coefficient, from an elongated coil state to a compact globule state in bulk.27 Bimodal size distribution obtained from the light scattering intensity correlation functions

10.1021/jp8017452 CCC: $40.75  2008 American Chemical Society Published on Web 06/27/2008

DNA Oligonucleotide and Surfactant Interaction confirms the coexistence of the coil and globule states in the solution below the cmc at which DNA chains become saturated with the surfactant molecules in bulk.31,32 The effect of the architectural change in lipids upon interactions with DNA is of particular interest, and systematic studies focusing on this aspect are abundant. Seeking optimization of cationic lipid formulation for effective gene delivery, an architectural change in a well-characterized surfactant was introduced. The synthesized surfactants were similar to CTAB but contained systematically varied hydrophobic tail lengths of C14 (tetradecyltrimethylammonium bromide, TTAB) and C12 (dodecyltrimethylammonium bromide, DTAB). Since it was observed that CTAB efficiently condenses DNA, the effects of changing the tail length on the surfactant’s ability to interact with DNA also were identified for high molecular weight DNA.28,29,33 Looking at the conformational behavior of DNA molecules in the presence of cationic surfactants of different chain lengths, it was concluded that a large amount of the short chain length surfactant is needed to induce compaction of DNA macromolecules. It was also observed that the coexistence region for CTAB and DTAB begins from concentrations of 8.0 and 80.0 µM, respectively. These results are in line with results of phase map studies, which showed that lower concentrations of more hydrophobic surfactants are needed to induce precipitation of the system.34 Recently, another study sought to determine the effects of modifying the headgroup of CTAB (e.g., sequential replacement of methyl groups with hydroxyethyl groups). Studies show that there is a stark transition in the DNA chain from extended coils (free chain) to a compact form and that this transition does not depend substantially on the architecture of the headgroup.35 However, the accessibility of ethidium bromide to DNA is retained to a significantly large extent for the more hydrophilic surfactants. These observations suggest that the surfactants with more substitutions have a larger headgroup and therefore form smaller micelles with higher curvature aggregates; these higher curvature aggregates lead to a less efficient, patch-like coverage of DNA.35 While interactions of various natural and synthetic cationic surfactants with long DNA have been extensively characterized, less attention has been paid to surfactant interactions with oligonucleotides. In the recent genomic era, a new gene silencing approach was introduced that is based on the use of small synthetic oligonucleotides, such as antisenses, siRNAs, and miRNA, to inhibit the expression of a specific target gene by inhibiting the function of the corresponding mRNA. These classes of oligonucleotides also represent novel therapeutics for manipulation of gene expression in the treatment of several diseases. The success of these strategies depends on the successful delivery into cells of either small, single, or double stranded DNA, RNA, or RNA/DNA hybrids with high efficiency. Because of the relatively small size of these oligonucleotides, the requirement for vectors designed for their in vitro or in vivo delivery is more stringent than the delivery of long DNA. One of the approaches to deliver these oligonucleotides into the cells is the use of liposomes and cationic polymers including poly-L-lysine, polyamidoamine (PAMAM) dendrimers,polyalkylcyanoacrylatenanoparticles,andpolyethyleneimine.36–39 With this consideration, it becomes a prerequisite to understand the interaction between cationic surfactants and small oligonucleotides. One study that deals with oligonucleotide-surfactant interactions was reported by Pattarkine and Ganesh to identify the cooperativity effect in surfactant binding that leads to duplex stabilization.40 The current study investigates the efficiency of

J. Phys. Chem. B, Vol. 112, No. 29, 2008 8825 three different surfactants with varying chain lengths from C16 to C12, to interact with 14mer double stranded DNA at different mole ratios of surfactant/DNA phosphate. Native-PAGE was used as a preliminary experiment to observe the surfactantoligonucleotide interaction and to identify the amount of surfactant molecules needed to neutralize the oligonucleotide. UV spectroscopy was used to evaluate the thermal stability of the complex, while ethidium bromide exclusion assays involving fluorescence spectroscopy were employed to determine the strength of binding between the interacting molecules. Finally, circular dichroism (CD) was used to analyze and characterize the conformational transition of the oligonucleotides in the presence of the surfactant. Materials and Methods Materials. CTAB, DTAB, and TTAB surfactants of highest purity (>99%) were used. The 14mer oligonucleotide DNA strand 1 (5′-GATGTTCACTCCAG-3′) and strand 2 (5′-CTGGAGTGAACATC-3′) were obtained from SBS-Genetech. The GC content of the oligonucleotide was 50%, as indicated by the manufacturer. The concentration of oligonucleotide was calculated by extrapolation of tabulated values of the monomer bases at 25 °C.41,42 Equimolar concentrations of both oligonucleotides were mixed, and the sample was heated to 95 °C for 10 min and allowed to cool at room temperature for 4-6 h. All experiments were performed in 10 mM sodium phosphate buffer, pH 7.0. The 1 µM oligonucleotide is equivalent to a 26 µM (as 14mer single strand oligonucleotides will have 13 phosphate groups) phosphate charge. Native-PAGE. The electrophoretic mobilities of the surfactant/oligonucleotide complexes at different surfactant/oligonucleotide charge ratios (Z() were determined using 15% Native-PAGE in a buffer consisting of 45 mM Tris-borate and 1 mM EDTA at pH 8.0. The oligonucleotide concentration was 1 µM (26 µM in phosphate unit), and complexes of various Z( were incubated for 2 h. Gels were run at 120 V for 60 min at 4 °C. DNA was visualized under UV illumination after staining the gels with Syber Green (Molecular Probes, Invitrogen) at 4 °C for 20 min. Ethidium Bromide Exclusion Assay. Ethidium bromide and oligonucleotide solutions (one ethidium bromide per base pair) were mixed in 10 mM sodium phosphate buffer and allowed to incubate for 10 min at 20 °C. One milliliter of each of the three different concentrations of oligonucleotides (1, 3, and 5 µM) was taken in the fluorescence cuvette. Increasing amounts of surfactants were added to the oligonucleotide-ethidium bromide mixture and incubated for 10 min before taking each reading. Surfactant stock solutions were prepared in such a way that the maximum dilution due to the addition of surfactant to the oligonucleotide-ethidium bromide complex solution was not more than 10%. Dilution factors were taken into account to calculate the final charge ratios. Fluorescence spectra were recorded from 500 to 700 nm, at an excitation wavelength (λex) of 480 nm using a spectrofluorometer (Jobin Yvon FluoroMax 3) with a peltier thermostat. CD Measurements. CD spectra were recorded in a Jasco spectropolarimeter (model 715) equipped with a thermoelectrically controlled cell holder and a cuvette with a path length of 1 cm. CD spectra were recorded between 220 and 325 nm at 20 °C in 10 mM sodium phosphate buffer, pH 7.0. Various amounts (maximum of 75 µL) of stock solutions of the surfactants (5, 50, and 100 mM for CTAB, TTAB, and DTAB, respectively) were added to 1 mL of 5 µM oligonucleotide solutions. The complexes were then incubated for 10 min before

8826 J. Phys. Chem. B, Vol. 112, No. 29, 2008

Jadhav et al.

each reading was taken. Dilution factors were taken into account to calculate the final charge ratios. The titration was finished when a significant decrease in the CD signal was not observed upon further addition of surfactants. Determination of Melting Curves of DNA-Surfactant. Different volumes of surfactant solutions were separately added to a constant volume of 1 µM (26 µM in terms of phosphate) DNA solution to obtain surfactant-oligonucleotide complexes of different charge ratios. After 2 h of incubation at room temperature, the melting profiles of the complexes were obtained by monitoring the absorbance of the complexes at 260 nm as a function of temperature. The samples were heated from 20 to 95 °C at a scanning rate of 1.0 °C/min. Transition temperatures (Tm) were calculated from the melting curves to gain insight into the helix-coil transition. Results and Discussion Gel Electrophoresis Studies. Gel electrophoresis is a technique that can be used to study the size and charge of DNA. Compacted, globular DNA migrates farther toward the anode during electrophoresis than coiled DNA does. This phenomenon was utilized to study interactions between DNA and oppositely charged ligands.43 As negative charges on DNA are partially or fully neutralized by interacting molecules, the migration of DNA through gel is either slowed or almost completely hindered. The PAGE study was performed to visualize the interaction between the double helix oligonucleotides with cationic surfactants and to study the extent of neutralization of negatively charged DNA by these surfactants, namely, CTAB, TTAB, and DTAB. Solutions of various charge ratios of oligonucleotide (1 µM) with surfactant were prepared, and gel electrophoresis was performed. The resulting image is shown in Figure 1. At low charge ratios of surfactant to oligonucleotide, migration of oligonucleotides was observed, but migration was eventually retarded at and above a certain charge ratio for CTAB and TTAB surfactants. These results indicate that these two surfactants are effectively able to neutralize the oligonucleotide charge but at different charge ratios. CTAB completely neutralizes the negative charge of DNA at a charge ratio (Z() of 1, while TTAB neutralizes DNA at Z( ) 14. In the case of DTAB, no apparent retardation was observed, but the bands corresponding to either free or partially neutralized oligonucleotides (lower band in the gel image) were not visible at and above Z( ) 50. This observation indicates that DTAB neutralizes the negative charge of DNA at a charge ratio (Z() of 50. Another striking observation is that below the charge neutralization (Z( ) 1) point, complexes of oligonucleotides-CTAB with a higher charge ratio move more slowly than the former. This result indicates that below the charge neutralization point (Z( ) 1.0), CTAB binds with the oligonucleotides through electrostatic interactions between the cationic headgroup of CTAB and the negative phosphate groups of the oligonucleotides present in the backbone. Upon binding with CTAB, oligonucleotides become partially neutralized and show a reduced mobility in the gel. The binding strength is strong enough to prevent dissociation of complexes during movement through the gel. Similar but moderate mobility shifts were observed in the case of oligonucleotides-TTAB complexes of charge ratios below the charge neutralization (Z( ) 14) point. Moderate mobility shifts in the case of oligonucleotide-TTAB as compared to the mobility shift observed for oligonucleotide-CTAB complexes indicate that binding of TTAB to oligonucleotides is weaker than binding of CTAB to oligonucleotides. However, no such mobility shift was observed in the case of the complexes of

Figure 1. Electrophoretic mobility shift assay on an acrylamide gel (15%) for complexes of (A) CTAB, (B) TTAB, and (C) DTAB. Complexes were prepared by mixing surfactants and oligonucleotides at different charge ratios. The experiments were run in buffer consisting of 45 mM Tris-borate and 1 mM EDTA at pH 8.0. Experiments were run at 120 V for 60 min at 4 °C. The gels were visualized under UV illumination by staining with SYBER Green I (Invitrogen) at room temperature.

oligonucleotide-DTAB at lower charge ratios. This result indicates that either DTAB does not bind below the charge neutralization point or binding is too weak and complexes dissociate during movement through the gel. The latter case seems to be true, as for all oligonucleotide-DTAB complexes at higher charge ratios, we observed prominent smear formation. Ethidium Bromide (EB) Exclusion Assay. EB is a cationic dye that is widely used to probe DNA-cationic agent interactions. The fluorescence efficiency of EB increases upon intercalation of EB between the base pairs of double helix DNA, the detection of EB fluorescence indicating the presence of native DNA in solution. EB may be removed from DNA by positively charged molecules that bind to DNA with a higher binding affinity, resulting in a quenching of fluorescence. This methodology successfully was utilized to understand the interaction between long DNA chains and different cationic agents such as polypeptides,44,45 polyamines,46 cationic dendrimers,47 cationic polymers,48 and cationic surfactants.35 The quenching of fluorescence intensity can be used to detect the strength of these interactions. A molecule with a higher binding affinity will exclude EB from DNA with a higher efficiency. Figure 2 represents the EB exclusion assay. Traces 1 and 2 in the left panels of Figure 2A-C show typical emission spectra for free EB and oligonucleotides bound with EB. As depicted in Figure 2A-C, the addition of surfactants results in the quenching of fluorescence (indicated by down arrows), with the quenching efficiency decreasing with tail length. CTAB excludes 85% EB at Z( > 1, TTAB excludes 80% EB at Z( > 14, whereas DTAB removes only 50% EB at Z( > 50 when the oligonucleotide concentration was 1 µM. Typical curves of fluorometric titration of solution of the DNA-EB complex with surfactants of different tail lengths are shown in the right panels of Figure 2.

DNA Oligonucleotide and Surfactant Interaction

J. Phys. Chem. B, Vol. 112, No. 29, 2008 8827

Figure 2. EB exclusion experiments for (A) CTAB, (B) TTAB, and (C) DTAB. Left panels are emission spectra; trace 1 and 2 in each plot are fluorescence emission spectra of free and bound EB. The down arrows indicate the increment of Z(. Right panels are I/I0 vs Z( plots. The oligonucleotide concentration was 1 µM in 10 mM sodium phosphate buffer (pH 7.0).

The EB titration curves of oligonucleotide-surfactant systems display two distinct characteristic Z( values, designated as Z(1 and Z(2, where EB fluorescence changes quite sharply. Z(1 is the low Z( end of the first plateau where negligible quenching is observed and most probably corresponds to the onset of DNAinduced self-assembly of the surfactant molecules. Z(2 is the onset of another plateau in the fluorescence titration curve at higher charge ratios and is considered to be the charge ratio where the oligonucleotide is saturated with surfactant in bulk. Any further surfactant added to the solution cannot bind to the oligonucleotide and, therefore, cannot exclude EB from oligonucleotides. The region between Z(1 and Z(2 describes the width of the coexistence region of unsaturated surfactant and saturated oligonucleotide. The differences in the ethidium bromide displacement presented by the different surfactants can be seen in terms of the strength of interaction as indicated either by Z(1 or Z(2. The Z(1 values for CTAB, TTAB, and DTAB are 0.5, 10, and 45, respectively, when the oligonucleotide concentration is 1 µM as shown by up arrows in the left panels of Figure 2. On the other hand, the Z(2 values for CTAB, TTAB, and DTAB are 1.2, 15, and 55, respectively, when the oligonucleotide concentration is 1 µM as shown by down arrows in the left panels of Figure 2. From these data, it can be concluded that a lower amount of the longer chain length surfactant is needed for DNA-induced self-assembly. These results are in line with the previous results of phase map studies, where it was observed that lower concentrations of the more hydrophobic surfactant are needed to induce precipitation of the system.28,33 Another interesting phenomenon is that the fluorescence variation was more gradual in the DTAB titration curve than in the CTAB titration curve. This is attributed to less cooperative association of hydrophobic moieties of DTAB in the presence of the oligonucleotides, as compared to CTAB. We also observed that the width of the coexistence region is narrower for CTAB and becomes wider for surfactants of shorter hydrophobic tails. A similar observation is found in the literature where the coexistence region for CTAB and DTAB was found to begin from concentrations of 8.0 and 80.0 µM, respectively, and the coexistence region was found to be wider in the case of DTAB.29 This is due to the fact that higher amounts of

Figure 3. EB exclusion experiments at varying concentrations of oligonucleotides. (A) CTAB, (B) TTAB, and (C) DTAB. Panels show I/I0 vs Z( plots. The oligonucleotide concentration was 1 µM (0), 3 µM (O) and 5 µM (∆) in 10 mM sodium phosphate buffer (pH 7.0).

surfactants with shorter chain lengths are required to achieve either DNA-induced self-assembly (corresponds to Z(1) or to saturate oligonucleotides with surfactants in bulk (corresponds to Z(2). As these critical ratios represent the oligonucleotide-induced self-assembly of surfactant molecules or the charge ratio where the oligonucleotide is saturated with surfactant in bulk; these values for any given surfactant should depend on the oligonucleotide concentrations. This aspect was examined by the experiment where oligonucleotide-EB complexes of three different concentrations (1, 3, and 5 µM) were titrated with surfactants (Figure 3). It was observed (Figure 3) that the two critical mixing ratios, Z(1 or Z(2, that define the strength of the interaction do not vary significantly with oligonucleotide concentration in the case of CTAB, whereas for TTAB and DTAB systems the ratio decreased upon an increase in oligonucleotide concentrations, and a perfect plateau was not observed in the case of the DTAB system for the considered concentrations. This is in excellent agreement with phase map studies of DNA and cationic surfactants and of DNA with spermine.34,49,50 It has been found that precipitation of DNA at a polyamine concentration, Cprecip, increases smoothly with DNA concentration and that redissolution in excess of spermine at a concentration, Credissol, nearly is independent of DNA. It has also been observed that above a critical DNA concentration, precipitation of DNA depends on DNA concentration but that the ratio Cprecip/DNA is found to be constant and equal to 0.20 ( 0.02.49 Similarly, inspection of Figure 3 indicates that for CTAB, this concentration already reached at least 1 µM

8828 J. Phys. Chem. B, Vol. 112, No. 29, 2008

Jadhav et al.

Figure 4. CD spectra of oligonucleotides in the presence of varying amounts of (A) CTAB, (B) TTAB, and (C) DTAB. (D) θ/θ0 vs Z( plot for CTAB (0), TTAB (O), and DTAB (∆) in 10 mM sodium phosphate buffer (pH 7.0). The oligonucleotide concentration was 5 µM in 10 mM sodium phosphate buffer (pH 7.0). The down arrows indicate the increment of Z(.

oligonucleotide since binding does not vary significantly with the mixing ratio (and not the DNA concentration). For TTAB, the critical concentration seems to be 3 µM, whereas for DTAB, it is 5 µM or higher. This is due to the fact that CTAB, being the surfactant that binds more strongly, reaches this critical concentration for lower concentration values. CD Measurements. CD spectra allow detection of even slight conformational changes in the double helix of DNA due to possible binding reactions. A positive peak at 275 nm and a negative peak at 245 nm are characteristic of DNA in the B conformation. The CD spectrum can be useful in detecting possible secondary structural changes associated with complex formation of oligonucleotides with cationic surfactants. Figure 4 represents the CD spectra for each of the surfactant-DNA complexes at various charge ratios, and the oligonucleotide concentration was 5 µM. As depicted in Figure 4, the intensity of both peaks, at 275 and 245 nm, decreased upon successive addition of each surfactant. CTAB shows the most rapid and prominent decrease in peak intensity upon binding with oligonucleotide. TTAB shows a similar decrease in peak intensity but at higher charge ratios, and so does DTAB at still higher charge ratios. The original spectrum for the oligonucleotide is altered to a wide, flattened signal for all surfactants upon neutralization of the oligonucleotide. Electrostatic interactions between surfactant and oligonucleotide are apparent with subsequent additions of surfactant. This can be seen from the leveling of peaks at 275 and 245 nm. Figure 4d depicts the change in CD signals at 275 nm with respect to charge ratio, normalized to 1. As observed from Figure 4, CTAB saturates the oligonucleotide at Z( ) 0.8, TTAB achieves saturation at Z( ) 10, while DTAB saturates the oligonucleotide at Z( ) 30. These values were more or less similar to the Z(2 values

Figure 5. UV melting transitions of oligonucleotide at 1 µM concentration and of surfactant-oligonucleotide complexes at different charge ratios (indicated in the panels) for (A and B) CTAB, (C) TTAB, and (D) DTAB. Experiments were performed with 10 mM sodium phosphate buffer (pH 7.0).

obtained in EB exclusion experiments at a 5 µM oligonucleotide concentration. By looking at the conformational behavior of oligonucleotides in the presence of cationic surfactants of different chain lengths (Figure 4), it can be concluded that a larger amount of the shorter chain length surfactant is needed to induce structural changes in oligonucleotides, hence confirming the results obtained by EB exclusion studies. The structural changes in the oligonucleotide are due to compaction of the oligonucleotide chains. Surfactants have only one charge per molecule, but due to their self-assembly properties, they form micellar aggregates in the vicinity of oligonucleotides at a certain critical concentration and thus can act as multivalent ions. Compaction is driven by attractive electrostatic interactions between different parts of oligonucleotides and the multivalent counterions offered by micellar aggregates that are present in the vicinity of oligonucleotides. This is evident from the fact that CTAB, the longer chain surfactant, is more efficient in compacting DNA than the shorter chain surfactant. A narrower coexistence region also was observed for CTAB than was observed for surfactants with shorter chains. Secondary structural changes occurring upon binding of cationic agents to long double helix DNA were studied by CD experiments in the literature. In contrast to the present study, there is no major change reported in the conformation of the DNA structure as cationic agents bind to the long DNA duplex. A slight decrease in the intensity of the positive band was

DNA Oligonucleotide and Surfactant Interaction noticed as the concentrations of cationic agents were increased, while the negative peak in the CD spectrum remained unchanged. A decrease in the magnitude of the positive band also was observed for free DNA with an increase in the salt concentration. Changes in the intensity of the positive CD peak were argued to be associated with the alteration of hydration of the helix in the vicinity of phosphate or the ribose ring as the ionic concentrations are altered. In the present study, a considerable change around both positive and negative bands was observed. It is known that a wide variety of multi- and polyvalent cations condense DNA into compact forms. The characteristics in CD spectra of such condensed DNA are dependent on the size (length) and shape (linear or circular) of the DNA. Damaschun et al.51 showed that the addition of spermidine to 750 bp DNA caused changes in the CD spectrum, which is attributed to the B to Ψ transition; however, no such change was observed for the CD spectrum of unsonicated T7 DNA by Gosule and Schellman.52 Another study focusing on the influence of DNA length on spermidine-induced condensation by electric dichroism measurements demonstrated that 258 and 436 bp DNA condensed into rod-like particles, while 748 bp or longer DNA condensed into torus-shaped particles.53 In the present study, CD titrations of oligonucleotides with increasing amounts of cationic surfactants showed a progressive loss of CD signals that are characteristic of native helical DNA conformations. UV Melting. UV melting experiments were used to measure the stability of oligonucleotide in the absence and presence of different amounts of surfactants. As the temperature of a solution containing a double stranded oligonucleotide is increased, the oligonucleotide begins to denature. This causes an increase in absorbance at 260 nm due to the disruption of base stacking as the oligonucleotide denatures. The temperature at the midpoint of the transition is the melting temperature (Tm), which is used in determining the stability of the oligonucleotide. This method has been widely used for the characterization of other cationic agents/CT-DNA complexes. Figure 5 represents the melting temperature of oligonucleotides at varying charge ratios of surfactant/oligonucleotide. As expected, the free oligonucleotide solution showed a monophasic melting behavior with a melting temperature around 33 °C. In the case of CTAB, before the oligonucleotide’s charge is neutralized, the UV melting curves for each charge ratio (Z( ) 0.25 and 0.50) follow the same general shape as free oligonucleotides but show a shift in Tm from 33 °C for free oligonucleotide to 40 °C for the complexes at Z( ) 0.5 (Figure 5A and 5B). Similar monophasic melting curves were also observed in the case of TTAB-oligonucleotide complexes of Z( ) 4 and 6, but the observed shift in Tm was only 2 °C (Figure 5C). In contrast, DTAB-oligonucleotide complexes of lower charge ratios show no observable change in Tm (Figure 5D). However, melting curves become broader in this case. These observations fit with the argument that below the charge neutralization point, CTAB interacts with oligonucleotides strongly and thus makes the oligonucleotides more stable. As TTAB interacts moderately, it shows a lesser shift in Tm. Interaction between DTAB and oligonucleotides below the charge neutralization point is too weak to show any shift in Tm. These data fairly correlate with the gel electrophoresis data. As the surfactant binding affinity increases, a change in Tm can be seen for corresponding complexes with the mobility shift as seen in the gel study. The stronger binding of CTAB results in both a shift in mobility and a larger increase in Tm. The melting curves for CTAB-oligonucleotide complexes of Z( ) 0.75 and 1 and TTAB-oligonucleotide complexes of

J. Phys. Chem. B, Vol. 112, No. 29, 2008 8829

Figure 6. Derivative of absorbance at 260 nm with respect to the temperature (∆A/∆T) vs temperature for DTAB at charge ratios of 0 (0), 40 (O), and 100 (∆).

Z( ) 14 show a biphasic behavior. This biphasic melting behavior was observed in earlier studies where the melting of long DNA chains was studied in the presence of cationic surfactants including CTAB. The first transition appears for the helix f coil transition of partially neutralized oligonucleotides. The second transition occurs as a result of the melting of completely neutralized DNA and, thus, appears at higher temperatures due to the increased stability of the surfactantbound oligonucleotide. Figure 6 depicts that as the surfactant binding affinity decreases, the melting profile shows a decline in biphasic properties. CTAB shows two clear melting temperatures, while the second Tm in TTAB appears less evident. DTAB does not show obvious biphasic behavior. Above the neutralizing charge ratio, the second Tm increases further with an increase in chain length of the surfactants. While the second transition for TTAB-oligonucleotide complexes appears between 70 and 80 °C, the melting profile of CTAB-oligonucleotide exhibited a loss of biphasic behavior. This can be attributed to the significantly high binding affinity of CTAB to the respective oligonucleotide. Above the neutralization ratio, the oligonucleotide existed in a completely complexed form with CTAB. The existence of a single but highly stabilized population of the complex results in a monophasic curve with a significantly high Tm > 80 °C. The higher melting temperature for CTAB-oligonucleotide complexes than the melting temperature for TTAB-oligonucleotide complexes reveals that TTAB-oligonucleotide complexes are less stable than the former. A detailed examination of melting profiles for DTABoligonucleotide complexes below the neutralization ratio indicates the existence of only a single melting domain corresponding to melting of the duplex only. However, above the neutralization point, complex formation with DTAB led to a broadening of the melting curve, which indicates the presence of more than one species. Because of weak binding, DTAB stabilizes the duplex moderately. As a result, the second transition melting domain appeared very close to the first, leading to broadening of the peak. Hence, biphasic behavior is more obvious in Figure 6, where the melting data are plotted as the derivative of the absorbance at 260 nm with respect to the temperature. From this derivative curve, it is apparent that the melting transition of oligonucleotides in DTAB-oligonucleotide complexes occurs around 65 °C. Overall, UV melting studies indicate that the oligonucleotides are strongly stabilized upon interaction with CTAB and moderately stabilized upon interaction with TTAB, whereas no observable stabilization occurs when they interact with the DTAB surfactant. The binding of monovalent and multivalent (e.g., Na+, Mg2+, etc.) counterions to oligonucleotides reduces the electrostatic repul-

8830 J. Phys. Chem. B, Vol. 112, No. 29, 2008 sion among the phosphate groups of the oligonucleotide backbone and induces the stability of the helical structures of oligonucleotides.54 As a result, the helix-coil transition occurs at higher temperatures at high salt concentrations. Thus, the binding of cationic surfactant to oligonucleotide phosphate groups should induce stability in a similar way. It was observed that the melting temperature of poly-[dA - dT] increased by 6 °C when the medium was changed from 1 M NaCl to 1 M Me4NCl.55 This was interpreted in terms of increased stability of the helix due to stronger binding of the tetraalkylammonium ions as compared to the Na+ duplex at 1 M concentration of the salt. Since the cationic units of all the surfactants studied here are tetraalkylammonium ions, it is quite likely that CTAB and TTAB also stabilize oligonucleotides to a higher extent due to stronger binding of the cationic units of CTAB and TTAB to the phosphate groups of the oligonucleotide backbone at the concentration range of the surfactants studied here. While DTAB has weaker binding than the counterion (16 mM sodium ions are available from 10 mM sodium phosphate buffer), it showed no such strong stabilization effects. The electrostatic interaction between CTAB and oligonucleotide phosphate groups is stronger than that formed by TTAB because of better hydrophobicity of CTAB as compared to TTAB. The well-separated biphasic melting behaviors in the case of CTAB and TTAB surfactants indicate the coexistence of completely saturated and partially saturated oligonucleotides. Above the charge neutralization point, the oligonucleotide melts at >80 °C when it interacts with CTAB, whereas it melts between 70-80 and 60-70 °C when it interacts with TTAB and DTAB, respectively. A higher melting temperature of oligonucleotides when they interact with CTAB again re-emphasizes that CTAB binds with oligonucleotides more strongly than the other surfactants studied here. Because of the hydrophobic interactions between the cationic surfactant molecules that are bound to the oligonucleotides, these will self-assemble and act as a multivalent ion. CTAB being a more hydrophobic surfactant produces stronger multivalent ions around the oligonucleotide chains, thus stabilizing the oligonucleotides most effectively. Taking all of the previous results for the oligonucleotidecationic surfactant complex system into consideration, it can be argued that surfactant monomers bind to negatively charged phosphate groups of oligonucleotide chains by electrostatic attraction below a certain charge ratio (Z(1) and produce partially neutralized oligonucleotide-surfactant complexes. At this stage, the hydrocarbon chains project out from the surface of the oligonucleotides into an aqueous environment. Since the mean distance between ions condensed on oligonucleotides (∼3.4 Å) is larger than the diameter of the surfactant headgroup (CnTA+), the electrostatics will dominate the interactions between surfactant monomers and oligonucleotides. At Z(1, oligonucleotides are fully neutralized by bound surfactants. A further increase in surfactant concentration results in new surfactant molecules binding to these neutralized complexes, due to hydrophobic interactions, to avoid any unfavorable interaction of hydrophobic surfactant chains with water molecules. This association is again dictated by the hydrocarbon chain lengths and as its chain length is longer, CTAB shows stronger interactions. With a further increase in surfactant concentration, the oligonucleotide is saturated with surfactant in bulk at a certain charge ratio (Z(2). The region between Z(1 and Z(2 values describes the width of the coexistence region of the surfactant with unsaturated and saturated oligonucleotides. During this saturation process, there is a possibility of the formation of DNA-induced self-assembly. However, our data

Jadhav et al. do not provide a definite conclusion regarding structural aspects of complexes at different stages of binding and the effect of surfactant chain length on them. Structural studies by different techniques along with aggregation number determinations are necessary to unambiguously solve this problem. Conclusion Binding of cationic surfactants to 14mer oligonucleotides studied here occurs at concentrations well below the respective cmc of the individual surfactant, and the binding isotherms have a sigmoidal shape that demonstrates cooperative binding. The binding isotherms were shown to be strongly dependent on the surfactant chain length, suggesting that hydrophobic interactions are important for binding and that it was analogous to the formation of micelles. At a 1 µM oligonucleotide concentration, CTAB, TTAB, and DTAB neutralize the oligonucleotides at charge ratios (Z() of 1, 14, and 50, respectively. At charge ratios below the neutralizing point, CTAB and TTAB interact with oligonucleotides through electrostatic interactions, and complexes show a lower mobility in the gel and a higher melting temperature. Overall, the data show that a smaller amount of the longer chain length surfactant is needed to achieve DNAinduced self-assembly. Progressive loss of the CD signature indicates the alteration of the normal B-type DNA structure, but detailed studies are necessary to determine the final structure of the oligonucleotide-surfactant complexes. Acknowledgment. Financial support for this work (CSIR young scientist award project) from the Council of Scientific and Industrial Research, Government of India, New Delhi is gratefully acknowledged. R.V. acknowledges a Hewlett-Packard fellowship from the Institute of Genomics and Integrative Biology, Delhi. Dr. Prasanta Kumar Das, The Indian Association for the Cultivation of Science, Kolkata, India is thanked for providing the surfactants used in this study. References and Notes (1) Mulligan, R. C. Science (Washington, DC, U.S.) 1993, 260, 926. (2) Anderson, W. F. Science (Washington, DC, U.S.) 1992, 25, 808. (3) Hanania, E. G.; Kavanagh, J.; Hortobagyi, G.; Giles, R. E.; Champlin, R.; Deisseroth, A. B. Am. J. Med. 1995, 99, 537. (4) Nishikawa, M.; Huang, L. Hum. Gene Ther. 2001, 12, 861. (5) Luo, D.; Saltzman, W. M. Nat. Biotechnol. 2000, 18, 33. (6) Zuber, G.; Dauty, E.; Nothisen, M.; Belguise, P.; Behr, J. P. AdV. Drug DeliVery ReV. 2001, 52, 245. (7) Maurer, N.; Fenske, D. B.; Cullis, P. R. Expert Opin. Biol. Ther. 2001, 1, 923. (8) Peterson, H.; Kunath, K.; Martin, A. L; Stolnick, S.; Roberts, C. J.; Davies, M. C.; Kissel, T. Biomacromolecules 2002, 3, 926. (9) Bonincontro, A.; La Mesa, C.; Proietti, C.; Risuleo, G Biomacromolecules 2007, 8, 1824. (10) Marques, E. F.; Dias, R.; Miguel, M.; Khan, A.; Lindman, B. In Polymer Gel Networks; Osada, Y., Khokhlov, A. R., Eds.; Marcel Dekker: New York, 2002; p 67. (11) Ganguli, M.; Jayachandran, K. N.; Maiti, S. J. Am. Chem. Soc. 2004, 126, 26. (12) Leal, C.; Moniri, E.; Pegado, L.; Wennerstrom, H. J. Phys. Chem. B 2007, 111, 5999. (13) Marchetti, S.; Onori, G.; Cametti, C. J. Phys. Chem. B 2006, 110, 24761. (14) Zhu, D.-M.; Evans, R. K. Langmuir 2006, 22, 3735. (15) Rosa, M.; Dias, R.; da Graca Miguel, M.; Lindman, B. Biomacromolecules 2005, 6, 2164. (16) Nakanishi, H.; Tsuchiya, K.; Okubo, T.; Sakai, H.; Abe, M. Langmuir 2007, 23, 345. (17) Hsu, W.-L.; Li, Y.-C.; Chen, H.-L.; Liou, W.; Jeng, U.-S.; Lin, H.-K.; Liu, W.-L.; Hsu, C.-S. Langmuir 2006, 22, 7521. (18) Kawashima, T.; Sasaki, A.; Sasaki, S. Biomacromolecules 2006, 7, 1942. (19) Izumrudov, V. A.; Zhiryakova, M. V.; Goulko, A. A. Langmuir 2002, 18, 10348.

DNA Oligonucleotide and Surfactant Interaction (20) Zhou, S.; Liang, D.; Burger, C.; Yeh, F.; Chu, B. Biomacromolecules 2004, 5, 1256. (21) Hsu, W.-L.; Chen, H.-L.; Liou, W.; Lin, H.-K.; Liu, W.-L. Langmuir 2005, 21, 9426. (22) Bhattacharaya, S.; Mandal, S. S. Biochim. Biophys. Acta 1997, 1323, 29. (23) Hayakawa, K.; Santerre, J.; Kwak, J. J. Biophys. Chem. 1983, 17, 175. (24) Nishio, T.; Shimizu, T.; Kwak, J. C. T.; Minakata, A. Biophys. Chem. 2003, 104, 501. (25) Shirahama, K.; Takashima, K.; Takisawa, N. Bull. Chem. Soc. Jpn. 1987, 60, 43. (26) Mel’nikov, S. M.; Sergeyev, V. G.; Yoshikawa, K. J. Am. Chem. Soc. 1995, 117, 9951. (27) Marchetti, S.; Onori, G.; Cametti, C. J. Phys. Chem. B 2005, 109, 3676. (28) Miguel, M. G.; Pais, A. A. C. P.; Dias, R. S.; Leal, C.; Rosa, M.; Lindman, B. Colloids Surf., A 2003, 228, 45. (29) Dias, R, S.; Pais, A. A. C. P.; Miguel, M. G.; Lindman, B. Colloids Surf., A 2004, 250, 115. (30) Baigl, D.; Yoshikawa, K. Biophys. J. 2005, 88, 3486. (31) Mel’nikova, Y. S.; Mel’nikov, S. M.; Lofroth, J. E. Biophys. Chem. 1999, 81, 125. (32) Dias, R. S.; Innerlohinger, J.; Glatter, O.; Miguel, M. G.; Lindman, B. J. Phys. Chem. B 2005, 109, 10458. (33) Hayakawa, K.; Santerre, J. P.; Kwak, J. C. T. Biophys. Chem. 1983, 17, 175. (34) Dias, R.; Mel’nikov, S.; Lindman, B.; Miguel, M. G. Langmuir 2000, 16, 9577. (35) Dasgupta, A.; Das, P. K.; Dias, R. S.; Miguel, M. G.; Lindman, B.; Jadhav, V. M.; Gnanamani, M.; Maiti, S. J. Phys. Chem. B 2007, 111, 8502. (36) Spagnou, S.; Miller, A. D.; Keller, M. Biochemistry 2004, 43, 13348. (37) Segura, T.; Hubbell, J. A. Bioconjugate Chem. 2007, 18, 736.

J. Phys. Chem. B, Vol. 112, No. 29, 2008 8831 (38) Chaltin, P.; Margineanu, A.; Marchand, D.; Van Aerschot, A.; Rozenski, J.; De Schryver, F.; Herrmann, A.; Mullen, K.; Juliano, R.; Fisher, M. H.; Kang, H.; De Feyter, S.; Herdewijn, P. Bioconjugate Chem. 2005, 16, 827. (39) Remaut, K.; Lucas, B.; Braeckmans, K.; Sanders, N. N.; Demeester, J.; De Smedt, S. C. Biochemistry 2006, 45, 1755. (40) Pattarkine, M. V.; Ganesh, K. N. Biochem. Biophys. Res. Commun. 1999, 263, 41. (41) Marky, L. A.; Bloomfield, K. S.; Kozlowski, S.; Breslauer, K. J. Biopolymers 1983, 9, 1247. (42) Kaur, H.; Arora, A.; Wengel, J.; Maiti, S. Biochemistry 2006, 45, 7347. (43) Fried, M.; Crothers, D. M. Nucleic Acids Res. 1981, 9, 6505. (44) Plank, C.; Tang, M. X.; Wolfe, A. R.; Szoka, F. C. Hum. Gene Ther. 1999, 10, 319. (45) Wyman, T. B.; Nicol, F.; Zelphati, O.; Scaria, P. V.; Plank, C.; Szoka, F. C. Biochemistry 1997, 36, 3008. (46) Tang, M. X.; Szoka, F. C. Gene Ther. 1997, 4, 823. (47) Chen, W.; Turro, N. J.; Tomalia, D. A. Langmuir 2000, 16, 15. (48) Nisha, C. K.; Sukara, V. M.; Ganguli, M.; Maiti, S.; Kizhakkedathu, J. N. Langmuir 2004, 20, 2386. (49) Raspaud, E.; de la Cruz, M. O.; Sikorav, J. L.; Livolant, F. Biophys. J. 1998, 74, 381. (50) Raspaud, E.; Chaperon, I.; Leforestier, A.; Livolant, F. Biophys. J. 1999, 77, 1547. (51) Damaschun, H.; Damaschun, G.; Becker, M.; Buder, E.; Misselwitz, R. Nucleic Acids Res. 1978, 10, 3801. (52) Gosule, L. C.; Schellman, J. A. J. Mol. Biol. 1978, 121, 311. (53) Marquet, R.; Wyart, A.; Houssier, C. Biochem. Biophys. Acta 1987, 909, 165. (54) Wolf, B.; Hanlon, S. Biochemistry 1975, 14, 1661. (55) Marky, L. A.; Patel, D.; Breslauer, K. J. Biochemistry 1981, 20, 1427.

JP8017452