Influence of the cyclodextrin nature on the decompaction of dimeric cationic surfactant-DNA complexes

Colloids and Surfaces A 555 (2018) 133–141

Contents lists available at ScienceDirect

Colloids and Surfaces A journal homepage: www.elsevier.com/locate/colsurfa

Influence of the cyclodextrin nature on the decompaction of dimeric cationic surfactant-DNA complexes

T

Manuel López-Lópeza, Pilar López-Cornejob, Carmen González-Cortésa, Daniel Blanco-Arévalob, ⁎ David Pérez-Alfonsob, Cristina Mozo-Muleroa, Jaime Oviedob, María Luisa Moyáb, a

Science and Technology Research Centre, Department of Chemical Engineering, Physical Chemistry and Materials Science, University of Huelva, Avda. de las Fuerzas Armadas s/n, 21071 Huelva, Spain b Department of Physical Chemistry, University of Seville, C/ Profesor García González 1, 41012 Seville, Spain

G R A P H I C A L A B S T R A C T

Cyclodextrin nature affects the decompaction of dimeric cationic surfactant-DNA complexes.

A R T I C LE I N FO

A B S T R A C T

Keywords: Dimeric surfactants DNA Charge inversion Compaction Cyclodextrins Decompaction

In this work, the influence of the cyclodextrin, CD, nature on the decompaction of positively charged compacted dimeric surfactant-DNA complexes was investigated. First, the condensation of calf thymus DNA by addition of three cationic dimeric surfactants with different spacer groups was studied by fluorescence, zeta potential, circular dichroism and atomic force microscopy measurements. Electromotive force experiments provided quantitative information about the influence of the spacer group on the DNA surfactant compaction efficiency. Cytotoxicity was evaluated to determine the biocompatibility of the cationic lipids. Subsequently, the decompaction of the surfactant-DNA complexes was achieved by adding α-, β-, and γ-cyclodextrin, the experimental observations being analogous for the three surfactants investigated. α- and β-cyclodextrin were found to behave similarly. These CDs completely hinder the interactions between the surfactant and the nucleic acid and provoke the DNA morphological change from a globular to an elongated form. A concentration of γ-CD higher than those of α- or β-CD is necessary in order to decompact the nucleic acid. Besides, zeta potential measurements show that in the presence of γ-CD, surfactant-DNA interactions are only partially hindered, some of the surfactant molecules remaining bound to the DNA.

⁎

Corresponding author. E-mail address: [email protected] (M.L. Moyá).

https://doi.org/10.1016/j.colsurfa.2018.06.066 Received 30 May 2018; Received in revised form 24 June 2018; Accepted 25 June 2018 Available online 28 June 2018 0927-7757/ © 2018 Published by Elsevier B.V.

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

1. Introduction

was investigated. Subsequently, the decompaction of the dimeric surfactant-DNA complexes by α-, β-, and γ-cyclodextrin has been studied by using several techniques. The spacer group of the dimeric surfactants was varied with the goal of examining its influence on the compaction and decompaction process. The results in this work will provide valuable information regarding how to choose the most appropriate CD in order to optimize the surfactant-CD complexes decompaction.

Gene therapy is the therapeutic delivery of genetic material for the treatment of a disease that enables the patient’s cells to produce the lacking proteins [1,2]. Usually the genetic material is combined with a delivery vehicle, or vector, which helps the nucleic acids to go into the cells achieving the therapeutic effect. Vectors make possible that the nucleic acid crossed barriers such as cell and nucleus membranes [3–5] since the anionic character of DNA limits its permeation across negatively charged cell membranes, and the steric restriction inside the cell prevents its transportation to the nucleus [6]. The use of vectors avoids these problems by condensing the nucleic acids into a neutral o positively charged complex with a compact globular conformation of reduced size [7]. Nonetheless, if DNA compaction is necessary for protecting the nucleic acid from degradation by nucleases as well as to facilitate cell uptake, once the nucleic acids have reached the cell nucleus, decompaction is indispensable for the release of the genetic material inside the cells. This allows the recovering of its properties and permits to proceed to the following transcription [8]. For this reason the interest of many researchers have been attracted by the decompaction process of DNA. Different reagents, such as non-ionic, anionic and zwitterionic surfactants, β-cyclodextrin, or electrolytes, among others [9–13], have been investigated as decompacting agents. A large variety of reagents has been used as non-viral vectors: cationic lipids, surfactants, polymers, etc. [3,14–21]. Among them, cationic surfactants have been shown to be efficient DNA compacting agents [16,15–21]. In particular, the use of dimeric cationic surfactants is of special interest not only for the unique relation they show between the surfactant structure and its activity [22,23], but also because several dimeric surfactants have been found to be efficient non-viral vectors [17,15–21,24]. Cyclodextrins, CDs, are cyclic oligosaccharides formed by six to eight (α-, β-, and γ-CD) α(1-4) ether linkages of glucopyranoside units [25]. Their shape is like a truncated cone and the internal cavity has a relatively hydrophobic character. This favors the formation of hostguest inclusion complexes with hydrophobic species of adequate size. The hydrophobic tails of surfactants have a strong tendency to intercalate into the hydrophobic CDs cavity forming highly stable inclusion complexes [26–33]. Therefore, CDs are able to strip the bound surfactant molecules from the surfactant-DNA complexes, this resulting in the DNA decompaction process. Most of the investigations about decompaction of cationic surfactant-DNA complexes have involved single-chain cationic surfactants and only few of them studied the influence of the CD nature on the process [12,26,34–36]. To the authors’ knowledge the influence of the CD nature on the decompaction of cationic dimeric surfactant-DNA complexes has not been investigated yet. For this reason, in the present work the compaction of calf thymus DNA caused by the addition of the dimeric surfactants 1,3-bis-(N,N’-dimethyl-N-docecylammonium) isobutylene dichloride, 12-ib-12,2Cl−, 1,3-bis-(N,N’-dimethyl-N-docecylammonium) 2-butylene dichloride, 12-bt-12,2Cl−, and 1,2-bis(dodecyldimethylammonium) dichloride, 12-2-12,2Cl− (see Scheme 1)

2. Experimental methods 2.1. Materials Most of materials were purchased from Sigma-Aldrich and Fluka, of the highest purity available, and used without further purification. α-, β-, and γ-cyclodextrin (> 99% purity) were kept under vacuum. The dimeric surfactants 1,3-bis-(N,N’-dimethyl-N-docecylammonium) isobutylene dichloride, 12-ib-12,2Cl−, and 1,3-bis(N,N’-dimethyl-N-docecylammonium) 2-butylene dichloride, 12-bt12,2Cl−, were synthesized and kindly provided by Prof. Laschewsky [37]. The 1,2-bis-(dodecyldimethylammonium) dichloride, 12-212,2Cl−, was synthesized from an ion-exchange of the corresponding dibromide [38], which was prepared following the method in the literature [39]. 12-2-12,2Cl− was characterized by 1H NMR, 13CNMR, and elemental analysis, with the results being in agreement with those previously reported. The calf thymus DNA concentration (given by phosphate groups) was estimated spectrophotometrically at 260 nm (molar absorptivity of 6600 M−1 cm−1) [40]. An agarose gel electrophoresis test, using ethidium bromide, EB, yielded an average number of base pairs per polynucleotide molecule of 10,000 bp [41]. The ratio Absorbance260nm/ Absorbance280nm of the DNA solutions varied between 1.7 and 1.8, suggesting the absence of proteins [42]. All solutions were prepared in HEPES 40 mM, at pH 7.4. Deionized distilled water Super Q Millipore (resistivity > 18 MΩ cm) was used.

2.2. UV–vis absorption spectroscopy Absorbance was measured using a CARY 1E UV–vis spectrophotometer (Varian), connected to a water flow Lauda cryostat. A standard quartz cell of 1 cm path length was used. Temperature was maintained at 298.0 ± 0.1 K. The stability of the surfactant-DNA solutions was investigated keeping a fixed DNA concentration ([DNA] = 3 × 10−5 M) and varying the surfactant concentration.

2.3. Conductivity measurements Conductivity was measured with a Crison GLP31 conductimeter as described in Ref. [43].

Scheme 1. Structure of the surfactants used in this work. 134

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

(adenocarcinomic human alveolar basal epithelial cell line), H358 (human lung cancer cell line), HepG2 (human liver cancer cell line), LS180 (adenocarcinomic human colonic epithelial cell line), and MCF7 (breast cancer cell line), and the normal cell line RPE-1. 96 well cell plates, at a density of 3000 cells per plate, were used. After 24 h, different amounts of the dimeric surfactants were added to the wells they were incubated for 4 additional days. Different culture media were used for the different cell lines: RPMI + 1% penicillin/streptomycin for H358, and DMEM + 10%FSB+1%penicillin/streptomycin for A549, MCF7, LS180, and HepG2. After incubation, they were pulsed with 3(4,5-dimethylthiazol-2-yl)5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS, ROCHE). Cell viability was measured using luminometry. Experiments for each surfactant concentration were done in triplicate.

2.4. Zeta potential experiments A Zetasizer Nano ZS Malvern Instrument Ltd. (UK) was used for the zeta potential measurements at 298.0 K. All the experiments were carried out at a fixed DNA concentration (3 × 10−5 M) and varying surfactant concentration. At least six zeta potential measurements were done at each surfactant concentration. 2.5. Competitive binding experiments A Hitachi F-2500 spectrofluorimeter was used for the recording of the fluorescence emission spectra. A water flow Lauda cryostat, connected to the cell compartment, maintained the temperature at 298.0 ± 0.1 K. The concentrations used in the emission fluorescence experiments were [EB] = 4.9 × 10−6 M, [DNA] = 3 × 10-5 M, and varying surfactant concentrations. The excitation and emission slits were 5 and 5 nm, respectively, and the scan rate was selected at 60 nm/ min. The excitation wavelength was 480 nm and the emission spectra of EB were registered in the range 500–700 nm. All solutions were buffered in 40 mM HEPES at pH = 7.4.

3. Results and discussion 3.1. DNA compaction Before studying the binding of the dimeric surfactants to DNA, their aggregation behavior was investigated by conductivity measurements. Fig. S1 (Supplementary Material) shows the dependence of the specific conductivity on 12-bt-12,2Cl− concentration. The critical micellar concentrations, cmc, were estimated by using Carpena’s method [47], the results being 2.0 mM, 2.1 mM, and 1.6 mM for 12-ib-12,2Cl−, 12-bt12,2Cl−, and 12-2-12,2Cl, respectively, in agreement with literature data [37]. Before carrying out the study of the binding of the surfactants to DNA, the stability of the buffered surfactant-DNA solutions was checked by UV–vis spectroscopy. Solutions were found to be stable for more than six hours (see Fig. S2, Supplementary Material). No turbidity was observed in the solutions within the charge ratio cationic surfactant/ DNA, Z+/−, range investigated. In all the solutions studied the surfactant concentration was well below the cmc. The interactions between the dimeric surfactants and DNA were first studied using the ethidium bromide, EB, competitive binding assay. EB is a cationic fluorescent dye which intercalates between the DNA base pairs [48] and it can be displaced from the DNA to the solution when other species associate to the nucleic acid [49]. Fig. 1 shows the dependence of I/Io on the charge ratio Z+/−, where Io and I are the fluorescence emission intensities of the dye in the absence and in the presence of surfactant, respectively. I/Io shows a sigmoidal decrease upon increasing Z+/− for the three surfactants investigated. This observation can be explained by the association of the dimeric surfactants to the nucleic acid that can cause the DNA compaction. As a consequence, insufficient available space for the accommodation of EB will

2.6. Circular dichroism experiments Circular dichroism experiments were carried out using a Biologic MOS-450 spectropolarimeter. A thermostatic bath kept the temperature at 298.0 ± 0.1 K. A cuvette of 1 cm path length was used, with a scan speed of 50 nm min−1. Scans were done in the range 230–300 nm and the spectrum for each solution was the average of 10 runs, with a 5 min equilibration before the scan. It was checked that the dimeric surfactant as well as the HEPES buffer do not contributed to the absorption bands observed. In the solutions the concentration of the polynucleotide was kept at 3 × 10−5 M and the dimeric surfactant concentration was varied. All solutions were buffered in 40 mM HEPES, pH = 7.4. 2.7. Atomic force microscopy, AFM AFM images were obtained with a Molecular Imaging PicoPlus2500 AFM (Agilent Technologies) as described in a previous work [44]. DNA concentration in the solutions was kept at 0.750 μM and the dimeric surfactant concentration was varied. All solutions were buffered with HEPES 40 mM, pH = 7.4. 2.8. Potentiometric measurements Electromotive forces were measured with a custom-built electrometric amplifier previously described [45]. The preparation of the selective membrane electrodes of the cationic dimeric surfactants was done following the method reported in Ref. [44]. Scheme 2 shows the cells used. The accuracy of the measurements was ± 0.1 mV. Temperature was kept at 298.0 ± 0.1 K by using a thermostatted glass cell connected to a Lauda cryostat. The membrane gave a Nernstian response for the three dimeric surfactants investigated, with a slope value close to 30 mV. The concentration of the free surfactant present in the solution was estimated using the calibration curves of the electrodes. 2.9. In vitro cytotoxicity assays The MTT assay [46] was used in order to estimate the cytotoxicity of the surfactants used in this work. The cell lines used were A549

Fig. 1. Plot of I/Io as a function of the charge ratio Z+/− (40 mM HEPES at pH = 7.4, [DNA] = 3.0 × 10−5 M, [EB] = 3.9 × 10-6 M). T = 298 K.

Scheme 2. Cells used in the potentiometric measurements. 135

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

be left, this resulting in its displacement from the DNA to the solution. Electrostatic as well as hydrophobic interactions play a major role in the association of cationic surfactants to DNA. Initially, electrostatic attraction between the negatively charged DNA and the cationic surfactants results in the exchange of the bound Na+ DNA counterions by the surfactant molecules, the main driving force being the entropy gain resulting from the counterion release. As binding advanced, the hydrophobic interactions between the incoming cationic surfactant and the surfactant molecules bound to the nucleic acid surface play a key role in the association process and are responsible for the high degree of cooperativity observed in the binding isotherms [13]. The importance of the hydrophobic interactions could explain that the spacer nature hardly affects the surfactant interactions with DNA. In the DNA compaction process surfactant aggregates, rather than surfactant monomers, are present at the nucleic acid surface [50]. The hydrophobic contribution to the binding Gibbs energy of the surfactants to DNA mainly depends on the number of carbon atoms in the alkyl tails, which is the same for the three surfactants. The rather similar cmc’s of the three surfactants also indicate that the hydrophobic Gibbs energy contribution to the Gibbs energy of micellization are analogous for all the surfactants studied since this Gibbs energy term mainly controls the self-aggregation process. Nonetheless other Gibbs energy terms related to the spacer nature will also be operative. Zeta-potential, ξ, measurements provide information on the effect of the surfactant binding to DNA on the nucleic acid charge. Fig. 2 shows the dependence of ξ on Z+/− for the surfactants used in this work. The charge inversion of DNA observed for the three surfactants is the result of the surfactant association to the DNA and is related to the nucleic acid condensation. This process is promoted by an entropy increase due to the release of the bound Na+ DNA counterions, and by the charge fractionalization mechanism, suggested by Shklovskii [51]. Fig. 2 shows that dependence of ξ on Z+/− is similar for 12-ib-12,2Cl−, 12-bt12,2Cl− and 12-2-12,2Cl−, in agreement with the fluorescence data. The efficiency of the dimeric surfactants to associate to DNA can be evaluated by potentiometric titrations, utilizing ion-selective membrane electrodes [13,52,53]. Using the experimental design described in the experimental section, calibration curves were obtained for each dimeric surfactant (see Fig. S3, Supplementary Material). A Nernstian response was observed in all cases, with slope values equal to 0.030 V, within experimental errors. In the potentiometric measurements the dimeric surfactant concentration was kept constant at 2 × 10−5 M and the DNA concentration varied within the range 1 × 10-6 M-1 × 10-4 M. All solutions were buffered with HEPES 40 mM, pH = 7.4. Higher nucleic acid concentrations could not be used because the solutions

Fig. 3. Binding isotherms of the dimeric surfactants-DNA system in 40 mM HEPES buffer pH = 7.4 at 298 K. The experimental values are expressed as the mean ± SD (n = 3).

became turbid. The dimeric surfactant concentration bound to DNA was estimated using eq. 1: [Surfactant]bound = [Surfactant]Total − [Surfactant]free

(1)

Here [Surfactant]free is the free surfactant concentration estimated from the calibration curves. A binding isotherm can be derived from the potentiometric data in terms of the binding degree parameter, β, which can be written as:

β=

[Surfactant ]bound Ca

(2)

Ca being the DNA concentration expressed in base pairs present in the solution. Fig. 3 shows the binding isotherms obtained for the three surfactants. A sigmoidal trend was observed, in agreement with the cooperative character of the dimeric surfactants binding to DNA. Data corresponding to high free surfactant concentrations could not be obtained because the electromotive force readings were out of the calibration curves. Fig. 3 shows that 12-ib-12,2Cl−, 12-bt-12,2Cl−, and 122-12,2Cl− present analogous tendencies to associate to DNA, in agreement with the fluorescence and zeta potential results. Condesation of DNA is accompanied by morphological changes in the polynucleotide. These changes can be studied by using circular dichroism measurements [54–56]. Fig. 4 shows the circular dichroism

Fig. 2. Dependence of the zeta potential on the charge ratio Z+/−. Conditions: [DNA] = 3.0 × 10−5 M, 40 mM buffer HEPES at pH = 7.4 and T = 298 K. The experimental values are expressed as the mean ± SD (n = 6).

Fig. 4. Circular dichroism spectra of 12-ib-12,2Cl−-DNA solutions in the absence and presence of CDs (40 mM HEPES at pH = 7.4, T = 298 K). 136

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

Fig. 5. AFM topographic images of 12-ib-12,2Cl−-DNA solutions, HEPES 40 mM at pH 7.4, adsorbed on APTES modified mica surface. (A) Pure DNA; (B) Z+/ − = 0.5; (C) Z+/− = 4; (D) Z+/− = 8; (E) Z+/− = 12; (F) Z+/− = 4 and [β-CD]/[Surfactant] = 40.

137

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

spectra of 12-ib-12,2Cl−-DNA solutions at different Z+/- values. In the absence of surfactant, the typical spectrum corresponding to the B conformation is observed, with a positive band at 278 nm, due to the stacking interactions between the bases pairs, and a negative one at 248 nm, caused by the right-handed DNA helicity [54]. Fig. 4 shows that an increase in Z+/-leads to a large decrease in the intensity of the positive and negative bands, with a bathochromic shift in their positions. These changes can be taken as indicative of a morphological change from an elongated structure to a globular one caused by the DNA compaction and partial denaturation of the double strand [54,57]. Similar changes in the DNA circular dichroism spectra were observed upon increasing Z+/− for 12-bt-12,2Cl− and 12-2-12,2Cl− (see Fig. S4, Supplementary Material). In order to visualize the DNA morphological changes caused by the formation of dimeric surfactants-DNA complexes, AFM measurements were performed at different Z+/- ratios. Fig. 5 shows the AFM topographic images of 12-ib-12,2Cl−-DNA solutions. Different structures are observed depending on the Z+/− value. Fig. 5A shows the coil conformation corresponding to the double strand of pure DNA. At low Z+/ratios, DNA molecules are mainly in an elongated state, although small globules are also observed. When Z+/− increases, the number of globular forms also increases. For high Z+/− values, only globular structures are observed (Fig. 5D and E). Similar results were found for 12-bt-12,2Cl− and 12-2-12,2Cl−, as is shown in Fig. S4 in Supplementary Material. From the AFM topographic images one can say that the association of the dimeric surfactants to DNA results in the compaction of the nucleic acid from an elongated form to a globular one. The potential toxic effects that the dimeric surfactants exert on

different cancer lines as well as on a normal cell line were investigated using the MTT essay. The results are shown in Fig. 6. The toxicity displayed by the surfactants follows the trend 12-ib-12,2Cl− > 12-bt12,2Cl− > 12-2-12,2Cl−. For the three surfactants, the adenocarcinomic human colonic epithelial cell line, LS180, is the less affected and the human liver cancer cell line, HepG2, can be considered the most affected. The toxicity data point out that none of the three dimeric surfactants could be used in gene transfection since at the surfactant concentrations needed to cause DNA charge inversion and compaction its toxicity is too high. Nonetheless, Fig. 6 shows that 12-2-12,2Cl− is substantially less toxic than 12-ib-12,2Cl− and 12-bt-12,2Cl−.

3.2. DNA decompaction by cyclodextrins As was mentioned in the Introduction, decompaction of cationic surfactant-DNA complexes using different reagents has been previously studied in the literature. Among the decompacting agents, CDs are a good choice for therapeutical applications because they are resistant to degradation by human enzymes and do not cause any immune response [34,58,59]. The internal cavities of α-, β-, and γ-CD range from 5 to 9 Å [60], which are of suitable size to form stable host-guest complexes with surfactants [26]. The use of CDs as decompacting agents in this work is based on their capacity to form dimeric surfactants-CD inclusion complexes. The effect of the addition of different amounts of α-, β-, and γ-CD on the fluorescence emission intensity of EB in buffered dimeric surfactants-DNA solutions at a charge ratio Z+/- = 2 ([Surfactant] = 6.0 × 10−5 M) is shown in Fig. S6 (Supplementary

Fig. 6. Cytotoxicity of the dimeric surfactants for different cancer cell lines after 4 days of incubation. 138

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

measurements. Fig. 7 shows the changes in ξ upon varying the molar ratio [CD]/[Surfactant], at a charge ratio Z+/- = 2. One can see that for α-CD and β-CD a molar ratio [CD]/[Surfactant]≥10 is enough for completely hindering the dimeric surfactant-DNA interactions, since the zeta potential reaches the value corresponding to pure DNA buffered solutions (−43 mV, see Fig. 2). In the case of γ-CD, ξ decreases upon increasing the CD concentration, but even at high γ-CD concentrations ξ remains equal to −19 mV, clearly higher than −43 mV. It is important to note that the CDs are not expected to associate to DNA. In fact, the zeta potential of solutions containing 3 × 10−5 M of DNA and 5 × 10−3 M of any of the CDs used in this work is similar to that of pure DNA. This observation is in agreement with previous results [12,34]. The experimental results shown in Fig. 7 can be explained by considering the formation of dimeric surfactant-CD inclusion complexes which leads to the stripping of the bound surfactant molecules from the nucleic acid. The dimeric surfactants can form 1:1 and 1:2 surfactantCD inclusion complexes with α-, β-, and γ-CD [26]. However, as pointed out by Valente and Soderman [26], there are not many studies on the formation of dimeric surfactant-CD inclusion complexes and the influence of the CD cavity size on the binding equilibrium constants does not follow a general trend. It depends on the surfactant nature [61–63]. For single chain surfactants, the equilibrium binding constants usually follow the trend K(α-CD) > K(βCD) > > K(γ-CD) [26]. It is worth noting that an increase in the CD concentration in the medium will favor the formation of 2:1 inclusion complexes [26]. Results in Fig. 7 indicate that the interactions between the surfactants and α- and β-CD are strong enough to remove practically all the surfactant molecules bound to the DNA, since ξ reaches the value observed for the pure nucleic acid. In contrast, the surfactant-γ-CD interactions must be weaker given that the minimum zeta potential value observed was higher than −43 mV. Therefore, in the case of γ-CD, part of the DNA bound surfactant molecules remain associated to the nucleic acid, although they could be, at the same time, bound to γ-CD. At this point it is worth noting the complexity of the systems investigated. Several equilibria have to be considered: equilibrium between DNA and both, surfactant and CD-surfactant inclusion complexes; equilibrium between CD and both, surfactant and surfactant bound to DNA; and so forth. For α- and β-CD the formation of the CD-surfactant inclusion complexes in solution would affect the thermodynamic equilibrium in such a way that surfactant molecules are no longer bound to the nucleic acid. That is not the case for γ-CD. The spacer nature does not influence the decompaction process of the dimeric surfactant-DNA complexes investigated. This can be taken as indicative that the equilibrium constants of the inclusion complexes are not much different for the three surfactants studied. This seems reasonable since all surfactants have two dodecyl chains and the hydrophobic interactions are the main contribution to the host-guest complex formation [26,61]. The stripping effect of CDs, both total or partial, is expected to be accompanied by morphological changes in the DNA. In order to investigate this issue circular dichroism was used. Figs. 4 and S5 (Supplementary Material) show that addition of α-, β-, and γ-CD to the dimeric surfactant-DNA solutions results in the recovery of the spectrum corresponding to the native state of DNA. That is, a decompaction of the nucleic acid from a globular form to an elongated one occurs. In the case of γ-CD, results indicate that in spite of surfactant molecules remaining bound to the DNA, the morphology of the nucleic acid corresponds to the elongated B form. AFM images can also shed light on the decompaction process. Fig. 5F shows the AFM topographic image of 12-ib-12,2Cl−-DNA with [β-CD]/[Surfactant] = 40 solutions, adsorbed onto an APTES modified mica surface. One can see that the observed structure corresponds to that of pure double stranded DNA, showing an extended coil conformation. The same structure was observed for the other surfactants and CDs. That is, a decompaction of the condensed DNA takes place in all cases.

Fig. 7. Dependence of the zeta potential on the molar ratio [CD]/[Surfactant] at Z+/− = 2. a)α-CD; b)β-CD; c)γ-CD. Conditions: [DNA] = 3.0 × 10−5 M, [Surfactant] = 6.0 × 10−5 M and varying cyclodextrin concentrations, 40 mM buffer HEPES at pH = 7.4 and T = 298 K. The experimental values are expressed as the mean ± SD (n = 6).

Material). For all the surfactants and CDs investigated, if enough CD is added to the solution, I/Io increases up to the value observed in the absence of surfactant; that is, to that observed when EB is intercalated between the base pairs of the elongated double helix DNA. A similar behavior is observed for different Z+/− values. Fig. S6 also shows that the molar ratio [CD]/[Surfactant] needed to reach this I/Io value, ([CD]/[Surfactant])plateau, depends on the CD nature. ([CD]/ [Surfactant])plateau is somewhat lower for α-CD than for β-CD, but it is much higher for γ-CD than for α- and β-CD. The changes following the addition of the CDs in the buffered dimeric surfactant-DNA solutions were also investigated by ξ-potential 139

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

4. Conclusions

delivery: a mechanistic study, J. Gen. Med. 3 (2001) 135–144. [9] C.P. Corbyn, P.D.I. Fletcher, R. Gemici, R.S. Dias, M.G. Miguel, Re-dissolution and de-compaction of DNA-cationic surfactant complexes using non-ionic surfactants, Phys. Chem. Chem. Phys. 11 (2009) 11568–11576. [10] T. Chaudhuri, A. Pan, S. Das, S.P. Moulik, Ratimetric interactions of anionic surfactants with calf thymus DNA bound to cationic surfactants: study II, J. Surfact. Deterg. 21 (2018) 127–137. [11] L. Feng, L. Xu, J. Hao, S. Dong, Controlled compaction and de-compaction of DNA by zwitterionic surfactants, Colloid Surf. A 501 (2016) 65–74. [12] A. González-Pérez, R.S. Dias, T. Nylander, B. Lindman, Cyclodextrin-surfactant complex: a new route in DNA decompaction, Biomacromolecules 9 (2008) 772–775. [13] S.M. Mel´nikov, V.G. Sergeyev, K. Yoshikawa, Transition of double-stranded DNA chains between random coil and compact globule states induced by cooperative binding of cationic surfactant, J. Am. Chem. Soc. 117 (1995) 9951–9956. [14] S.D. Li, L. Huang, Non-viral is superior to viral gene delivery, J. Control. Release 123 (2007) 181–183. [15] S.Y. Wong, J.M. Pelet, D. Putnam, Polymer systems for gene delivery–past, present, and future, Prog. Polym. Sci. 32 (2007) 799–837. [16] I. Zuhorn, J.B.F.N. Engberts, D. Hoekstra, Gene delivery by cationic lipid vectors: overcoming cellular barriers, Eur. Biophys. J. 36 (2007) 349–362. [17] M.A. Mintzer, E.E. Simanek, Nonviral vectors for gene delivery, Chem. Rev. 109 (2008) 259–302. [18] T. Ahmed, A.O. Kamel, S.D. Wettig, Interactions between DNA and gemini surfactant: impact on gene therapy: parts I and II, Nanomedicine 11 (2016) 289–306 and 403–420. [19] E. Aicart, E. Junquera, Recent progress in gene therapy to deliver nucleic acids with multivalent cationic vectors, Adv. Colloid Interface Sci. 233 (2016) 161–175. [20] K. Kumar, A.L. Barrán-Berdón, S. Datta, M. Múñoz-Úbeda, C. Aicart-Ramos, P. Kondaiah, E. Junquera, E. Aicart, A delocalizable cationic head group together with an oligo-oxyethylene spacer in gemini cationic lipids improves their biological activity as vectors of plasmid DNA, J. Mater. Chem. 2 (2015) 1495–1506. [21] A. Zinchenko, O. Pyshkina, A. Lezov, V. Sergeyev, K. Yoshikawa, R.S. Dias, B. Lindman (Eds.), DNA Interactions With Polymers and Surfactants, Wiley, Hoboken, NJ, 2008pp. 59–88. [22] F.M. Menger, J.N. Keiper, Gemini surfactants, Angew. Chem. Int. Ed. 39 (2000) 1906–1920. [23] R. Zana, Dimeric and oligomeric surfactants. Behavior at interfaces and in aqueous solutions: a review, Adv. Colloid Interface Sci. 97 (2002) 205–253. [24] M. Martínez-Negro, A.L. Barrán-Berdón, C. Aicart-Ramos, M.L. Moyá, C. Tros de Ilarduya, E. Aicart, E. Junquera, Transfection of plasmid DNA by nanocarriers containing a gemini cationic lipid with an aromatic spacer or its monomeric counterpart, Colloid Surf. B 161 (2018) 519–527. [25] G. Crini, Review: a history of cyclodextrins, Chem. Rev. 114 (2014) 10940–10975. [26] A.J.M. Valente, O. Söderman, The formation of host-guest complexes between surfactants and cyclodextrins, Adv. Colloid Interface Sci. 205 (2014) 156–176. [27] R.A. Carvalho, H.A. Correia, A.J.M. Valente, O. Södderman, M. Nilsson, The effect of the head-group spacer length in the host-guest association with β-cyclodextrin, J. Colloid Interface Sci. 354 (2011) 725–732. [28] C.M.C. Faustino, A.R.T. Calado, L. García-Río, Complexation of β-cyclodextrin and an amino acid-based gemini surfactant derived from cyteine, J. Colloid Interface Sci. 367 (2012) 286–292. [29] M. Benko, L.A. Kyrály, S. Puskás, Z. Kyrály, Coomplexation of β-cyclodextrin with a gemini surfactant stuided by isothermal titration calorimetry and surface tensiometry, Langmuir 30 (2014) 6756–6762. [30] R.H. Lu, J.C. Hao, H.Q. Wang, L.H. Tong, Determination of association constants for cycodextrin-surfactant inclusion complexes: a numerical method based on surface tension measurements, J. Colloid Interface Sci. 192 (1997) 37–42. [31] D.J. Dobe, R.E. Verall, E. Junquera, E. Aicart, Effects of surfactant beta-cyclodextrin complex-formation on the surfactant monomer-micelle exchange-rate in aqueous solutions of decyltrimethylammonium bromide, J. Phys. Chem. 97 (1993) 1243–1248. [32] N. Funasaki, S. Ishikawa, S. Hirota, Chemical shifts as a novel measure of interactions between two binding sites of symmetric dialkyldimethylammonium bromides to alpha-cyclodextrin, Anal. Chim. Acta 555 (2006) 278–285. [33] C. Cabaleiro-Lago, M. Nilsson, A.J.M. Valente, M. Bonini, O. Söderman, NMR diffusometry and conductometry study of the host-guest association between betacyclodextrin and dodecane a,a2-bis(trimethylammonium bromide), J. Colloid Interface Sci. 300 (2006) 782–787. [34] M. Cao, M. Deng, X.L. Wang, Y. Wang, Decompaction of cationic gemini surfactantinduced DNA condensates by β-cyclodextrin or anionic surfactant, J. Phys. Chem. B 112 (2008) 13648–13654. [35] K. Nagaraj, S. Arunachalam, Binding of a double-chain surfactant-cobalt(III) to CT DNA:effect of β-cyclodextrin in the medium, Int. J. Biol. Macromol. 62 (2013) 273–280. [36] L. Jiang, Y. Yan, J. Huang, Versatility of cyclodextrinas in self-assembly systems of amphiphiles, Adv. Colloid Interface Sci. 169 (2011) 13–25. [37] A. Laschewsky, K. Lunkenheimer, Rh. Rakotoaly, L. Wattebled, Spacer effects in dimeric cationic surfactants, Colloid Polym. Sci. 283 (2005) 469–479. [38] K. Esumi, J. Hara, N. Ahira, K. Usui, K. Torigore, Preparation of anisotropic gold nanoparticles using a gemini surfactant template, J. Colloid Interface Sci. 208 (1998) 578–581. [39] F.M. Menger, J.S. Keiper, B.N.A. Mbadugha, K.L. Caran, L.S. Romsted, Interfacial composition of gemini surfactant micelles determined by chemical trapping, Langmuir 16 (2002) 9095–9098. [40] Y.J. Liu, H. Chao, L.F. Tan, Y.X. Yuan, L.N. Ji, Ruthenium(II) mixed-ligand

We have shown in this work that the dimeric surfactants 12-ib12,2Cl−, 12-bt-12,2Cl−, and 12-2-12,2Cl− can compact DNA molecules, showing a similar compaction efficiency. Addition of appropriate amounts of α-, β-, and γ-CD to dimeric surfactant-DNA buffered solutions is followed by the decompaction of DNA from a condensed globular form to an elongated one. This process is caused by the formation of surfactant-CD inclusion complexes which results in the stripping of the bound surfactant molecules from the nucleic acid. The ability of the CDs to hinder the surfactant-DNA interactions is due to the stronger and more specific surfactant-CD hydrophobic interactions. The spacer nature of the dimeric surfactants has no influence on the decompaction process. This result can be taken as indicative that the equilibrium association constants of the surfactant-CD inclusion complexes are similar for the three surfactants studied. The CD nature determines the cyclodextrin concentration needed to decompact the DNA. α- and β-CD behave similarly, a relation [CD]/[Surfactant]∼10 being enough to remove all the DNA bound surfactant molecules (see Fig. 7). In the case of γ-CD, even at a molar ratio [CD]/[Surfactant] as high as 80, surfactant molecules remained associated to the DNA. In spite of this, addition of γ-CD causes the DNA decompaction, as is shown by circular dichroism and AFM images. The experimental results show that the use of cyclodextrins as decompacting agents is a convenient procedure when dimeric surfactants have been used as condensing agents. Since it is always desirable to use a low decompating agent concentration, the nature of the cyclodextrin is an essential factor to be considered. In most of the previous studies on the use of CDs to decompact DNA condensed by surfactants, β-CD has been used. However, for single-chain surfactant-DNA complexes, and taking into account that surfactant-α-CD inclusion complexes are substantially more stable than the surfactant-β-CD ones for most of the single chain surfactants [26], α-CD is doubtless a better choice than βCD as decompacting agent. In the case of dimeric surfactants α-CD could be a better choice in some cases. Further studies using other types of cyclodextrins to decompact DNA will be necessary. Acknowledgements This work was financed by the Consejería de Innovación, Ciencia y Empresa de la Junta de Andalucía (P12-FQM-1105), FQM-274 and FQM-206, University of Seville (2017/1004 and 2018/500). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.colsurfa.2018.06.066. References [1] T. Wirth, N. Parker, S. Ylä-Herttuala, History of gene therapy, Gene 525 (2013) 162–169. [2] D. Ibraheem, A. Elaissari, H. Fessi, Gene therapy and DNA delivery systems, Int. J. Pharm. 459 (2014) 70–83. [3] A.J. Kirby, P. Camilleri, J.B.F.N. Engberts, M.C. Feiters, R.J.M. Nolte, O. Söderman, M. Bergsma, P.C. Bell, M.L. Fielden, C.L. García Rodríguez, P. Guédat, A. Kremer, C. McGregor, C. Perrin, G. Ronsin, M.C.P. van Eijk, Gemini surfactants: new synthetic vectors for gene transfection, Angew. Chem. Int. Ed. 42 (2003) 1448–1457. [4] C. Chittimalla, L. Zammut-Italiano, G. Zuber, J. Behr, Monomolecular DNA nanoparticles for intravenous delivery of genes, J. Am. Chem. Soc. 127 (2005) 11436–11441. [5] I.M. Verma, N. Somia, Gene therapy-promises, problems and prospects, Nature 389 (1997) 239–242. [6] I.M. Verma, M.D. Weitzman, Gene therapy: twenty-first century medicine, Ann. Rev. Biochem. 74 (2005) 711–738. [7] K.K. Ewert, C.E. Samuel, C.R. Safinya, Lipid–DNA interactions: structure–function studies of nanomaterials for gene delivery, in: R. Dias, B. Lindman (Eds.), DNA Interactions With Polymers and Surfactants, John Wiley & Sons, Inc., Boston, MA, USA, 2007, pp. 377–404. [8] A. Kichler, C. Leborgne, E. Coeytaux, O. Danos, Polyethylenimine-mediated gene

140

Colloids and Surfaces A 555 (2018) 133–141

M. López-López et al.

[41]

[42] [43]

[44]

[45]

[46] [47]

[48]

[49]

[50]

[51]

Top. 60 (1990) 5802–5811. [52] S.Z. Bathaie, A.A. Moosavi-Movahediu, Saboury, energetic and biding properties of DNA upon interaction with dodecyltrimethylammonium bromide, Nucleic Acids Res. 4 (1999) 1001–1005. [53] S. Maulik, D.K. Chattoraj, S.P. Moulik, Biopolymer-surfactant interaction: 3. Kinetics of binding of cetyltrimethylammonium bromide to deoxyribonucleic acid, Colloid Surf. B 11 (1998) 57–65. [54] S. Neidle, Nucleic Acid Structure and Recognition, Oxford University Press, New York, 2002. [55] L. Wang, Y. Wang, X. Sun, G. Zhang, S. Dong, J. Hao, Versatile self-assembly and biosensing applications of DNA and carcon quantum dots coordinated cerium ions, Chem. Eur. J. 23 (2017) 10413–14122. [56] L. Wang, Y. Wang, J. Hao, S. Dong, Magnetic fullerene-DNA/hialluronic acid nanovehicles with magnetism/reduction dual-responsive triggered release, Biomacromolecules 18 (2017) 1029–1038. [57] W. Andrzejewska, M. Wilkowska, M. Cheabaszczewska, M. Kozak, The study of compexation between dicationic surfactants and the DNA duplex using structural and spectroscopic methods, RSC Adv. 7 (2017) 26006–26018. [58] M.E. Davis, M.E. Brewster, Cyclodextrin-based pharmaceutics: past, present and future, Nat. Rev. Drug. Discov. 3 (2004) 1023–1035. [59] R.H. Lu, J.C. Hao, H.Q. Wang, L.H. Tong, Determination of association constants for cyclodextrin–surfactant inclusion complexes: a numerical method based on surface tension measurements, J. Colloid Interface Sci. 192 (1997) 37–42. [60] W.R. Saenger, J. Jacob, K. Gessler, T. Steiner, D. Hoffmann, H. Sanbe, K. Koizumi, S.M. Smith, T. Takaha, Structures of the common cyclodextrins and their larger analogues—beyond the doughnut, Chem. Rev. 98 (1998) 1787–1802. [61] L. Liang, Y. Yan, J. Huang, Versatility of cyclodextrins in self-assembly systems of amphiphiles, Adv. Colloid Interface Sci. 169 (2011) 13–25. [62] A. Guerrero-Martínez, G. González-Gaitano, M.H. Viñas, G. Tardajos, Inclusion complexes between β-cyclodextrin and a gemini surfactant in aqueous solution: an NMR study, J. Phys. Chem. B 110 (2006) 13819–13828. [63] D.-Z. Sun, X.-M. Qiu, L. Li, X.-L. Wei, B.-L. Yin, A study of α-cyclodextrin with a group of cationic gemini surfactants utilizing isothermal titration calorimetry and NMR, J. Chem. Thermodyn. 38 (2006) 773–777.

complexes containing 2-(6-methyl-3-chromonyl)imidazo[4,5-f][1,10]-phenanthroline: synthesis, DNA-binding and photocleavage studies, Inorg. Chim. Acta 359 (2006) 3807–3814. F. Secco, M. Venturini, T. Biver, F. Sanchez, R. Prado-Gotor, E. Grueso, Solvent effects on the kinetics of the interaction of 1-pyrenecarboxaldehyde with calf thymus DNA, J. Phys. Chem. B 114 (2010) 4686–4691. W. Saenger, Principles of Nucleic Structure, Springer-Verlag, New York, 1984. M. López-López, P. López-Cornejo, V.I. Martín, F.J. Ostos, C. Checa-Rodríguez, R. Prados-Carvajal, J.A. Lebrón, P. Huertas, M.L. Moyá, Importance of the hydrophobic interactions in the single-chained cationic surfactant-DNA complexation, J. Colloid Interface Sci. 521 (2018) 197–205. B. Sarrión, E. Bernal, V.I. Martín, M. López-López, P. López-Cornejo, M. GarcíaCalderón, M.L. Moyá, Binding of 12-s-12 dimeric surfactants to calf thymus DNA: evaluation of the spacer length influence, Colloid Surf. B 144 (2016) 311–318. C. Salvador, M.S. Mesa, E. Durán, J.L. Álvarez, J. Carbajo, J.D. Mozo, Open ISEmeter: an open hardware high-impedance interface for potenciometric detection, Rev. Sci. Instrum. 87 (2016) 055111. J. van Meerloo, G.J.L. Kaspers, J. Cloos, Cell sensitivity assays: the MTT assay, Methods Mol. Biol. 731 (2011) 237–245. P. Carpena, J. Aguiar, P. Bernaola-Galván, C. Carnero Ruiz, Problems associated with the treatment of conductivity-concentration data in surfactant solution: simulations and experiments, Langmuir 18 (2002) 6054–6058. E. De Pauw, F. Rosu, Interaction between antitumor drugs and a double-stranded oligonucleotide studied by electrospray ionization mass spectrometry, J. Mass. Spectrom. 34 (1999) 1328–1337. G. Cosa, K.S. Focsaneau, J.R.N. Mc Lean, J.P. Mac Namee, J.C. Scaiano, Photophysical properties of fluorescent DNA-dyes bound to single- and doublestranded DNA in aqueous buffered solutions, Photochem. Photobiol. 73 (2001) 585–599. D. Matulis, I. Rouzina, V. Bloomfield, Thermodynamics of cationic lipid binding to DNA and DNA condensation: roles of electrostatics and hydrophobicity, J. Am. Chem. Soc. 124 (2002) 7331–7342. B.I. Shklovskii, Screening of a macroion by multivalent ions: correlation-induced inversion of charge, Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip.

141