DNA compaction to multi-molecular DNA condensation induced by cationic imidazolium gemini...

transcript

Colloids and Surfaces A: Physicochem. Eng. Aspects 414 (2012) 33– 40

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces A: Physicochemical andEngineering Aspects

jo ur nal homep a ge: www.elsev ier .com/ locate /co lsur fa

NA compaction to multi-molecular DNA condensation induced by cationicmidazolium gemini surfactants

ing Zhoua,b, Guiying Xua,∗, Mingqi Aoa, Yanlian Yangb,∗∗, Chen Wangb

Key Laboratory of Colloid and Interface Chemistry, Ministry of Education, Shandong University, Jinan 250100, PR ChinaNational Center for Nanoscience and Technology, Beijing 100190, PR China

i g h l i g h t s

An evolution from DNA compactionto multi-molecular DNA condensa-tion induced by [Cn-4-Cnim]Br2 isidentified and its mechanism is dis-cussed.[Cn-4-Cnim]Br2 as novel imidazoliumgemini surfactants can interact withDNA via electrostatic, hydrophobicand �–� interaction.The stronger interaction betweenDNA and [Cn-4-Cnim]Br2 with longertails demonstrates the importantcontribution of the hydrophobicinteraction.

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 5 May 2012eceived in revised form 27 August 2012ccepted 29 August 2012vailable online 8 September 2012

eywords:

a b s t r a c t

The compaction and condensation of DNA induced by cationic imidazolium gemini surfactants ([Cn-4-Cnim]Br2, n = 10, 12, 14) at different charge ratios have been investigated by dynamic light scattering(DLS), zeta potential, circular dichroism (CD), and ethidium bromide exclusion assay. Upon additionof [Cn-4-Cnim]Br2, DNA molecules undergo the process from compaction to multi-molecular conden-sation accompanied by conformation change, which could be proved by the DLS and CD results. Thecharge density changes in zeta potential measurements indicated the impact of the electrostatic inter-

NAationic imidazolium gemini surfactants[Cn-4-Cnim]Br2)ondensation

action in DNA–surfactant complex. The comparison between DNA compaction and condensation by[Cn-4-Cnim]Br2 with different tail lengths demonstrated the important contribution of the hydropho-bic interaction. The EtBr exclusion assay indicates the �–� interaction between imidazolium groups of[Cn-4-Cnim]Br2 and DNA aromatic rings also plays a role in the DNA/[Cn-4-Cnim]Br2 complex formation.The impact of the different interactions on the DNA compaction and condensation by gemini surfactants

. Introduction

Gene therapy has been demonstrated as a potential treatmentf both genetic and acquired diseases, while the effective deliv-ry of the therapeutic genes into target cells in vitro and in vivo

∗ Corresponding author. Tel.: +86 531 88365436; fax: +86 531 88564750.∗∗ Co-corresponding author. Tel.: +86 10 82545559.

E-mail addresses: xuguiying@sdu.edu.cn (G. Xu), yangyl@nanoctr.cn (Y. Yang).

is still one of the greatest challenges in gene therapy. It has beenconfirmed that the key parameters for achieving effective genetherapy is the size of the DNA condensates [1,2]. It is also neces-sary to neutralize the negative charges of DNA, because an overallpositive charge significantly improves the docking of the DNA con-densate on the primarily negatively charged cell membranes [3].

3 Physic

stctuetilm[irmrbsfdpleststcDtiotafsica

bWcaaot[oTaitswhvtieodr

4 T. Zhou et al. / Colloids and Surfaces A:

13]. These cationic agents demonstrated their capabilities to com-act, condense, and transfect DNA across the cell membrane forene delivery application.

In recent years, DNA compaction and condensation by cationicurfactants have been investigated intensively due to their mul-ifunctional structures including positive charges to neutralizeharges of DNA phosphate backbone and the hydrophobic tailso interact with DNA bases and cell membrane [14–18]. A clearnderstanding of these driving forces could be very important forxploring and predicting cationic surfactants’ biological applica-ions. Matulis et al. [19] studied the lipid–DNA binding by usingsothermal titration calorimetry, which indicated that lipids withonger aliphatic chains bind to DNA stronger than short ones

ostly due to highly positive hydrophobic entropy. Zhu and Evans20] employed various independent methods to investigate thenteractions of plasmid DNA and cationic surfactants, and theesults indicated the hydrophobic interactions between surfactantolecules and DNA play important roles. The binding isotherm

esults reported by Jadhav et al. [21] showed that hydropho-ic forces were important for 14mer oligonucleotide and cationicurfactants systems. In other aspects, to better comprehend theactors governing the gene delivery abilities of cationic surfactants,etailed biophysical characterization of DNA/surfactants have beenerformed extensively, using fluorescence microscopy [1], dynamic

ight scattering [2,6], and isothermal titration calorimetry [7,19],tc. These studies shed light on the application and design of novelurfactants as DNA condensation agents, such as gemini surfac-ants. As a relatively new class of amphiphilic molecules, geminiurfactants show greatly enhanced surfactant properties relativeo the corresponding monovalent (single chain, single head-group)ompounds such as surface activity, which have shown potential asNA transfection agents [22]. Bacteriophage T4 DNA underwent a

ransition from random coil to globule with an intermediate coex-stence region by using cationic gemini surfactant from a seriesf alkanediyl-˛,ω-bis-(dimethylalkylammonium bromide) surfac-ants, and the influence of spacer length, valency, head-group sizend tail length on the aggregation behavior of DNA/gemini sur-actant system was determined [23]. With more and more geminiurfactants with superior properties synthesized and character-zed, a majority of the new cationic surfactants are shown toondense DNA efficiently and to present very good transfectionctivity in vitro [22].

Most of the new gemini surfactants studied for DNA binding areased on dicationic quaternary ammonium compounds [23–26].hile, including aromatic functional groups similar to DNA bases

ould enrich the chemistry of the gemini surfactants and wouldlso bring new thoughts into the DNA-condensing agent inter-ctions, such as �–� interaction and hydrogen bonds. Recently,ur group has explored the synthesis and physicochemical proper-ies of a series of imidazolium gemini surfactants [Cn-4-Cnim]Br227–30]. Imidazolium gemini surfactants are a new generationf amphiphilic molecules exhibiting many promising features.hey are made up of two lipophilic chains and two polar imid-zolium head-groups covalently linked by a spacer. Owing to thenherent ionic nature of imidazolium ionic liquids [31,32] andhe properties of conventional gemini surfactants, they displaytronger self-aggregation tendency and stronger �–� interactionith aromatic rings. Meanwhile, imidazolium gemini surfactantsave lower critical aggregation concentrations (CACs) than con-entional single-chain surfactants with equivalent tail length dueo the hydrophobic contribution [27]. The lower CACs would resultn the decrease of agent amount needed for most applications

ochem. Eng. Aspects 414 (2012) 33– 40

“proton-sponge” effect [33]. The side chain of histidine, imidazolegroup, has a pK ≈ 6.0, can absorb protons in an endosomal envi-ronment to induce endolysosomal escape following endocytosis ofthe DNA condensates for intracellular delivery of DNA. Thus thetwo imidazolium head-groups in an amphiphilic molecule couldincrease the gene transfection efficiency by extra �–� interactionwith DNA bases and the “proton-sponge” effect.

The understanding of the interaction between the imidazoliumgemini surfactants [Cn-4-Cnim]Br2 and DNA is of genuine impor-tance for their application in gene delivery and other biological andmedical applications. Herring sperm DNA is chosen here as a modelgene system for short DNA, microRNA (miRNA), small interferingRNA (siRNA), and so on, which have shown promising gene therapyin clinical trials. In the present work, we investigate the interactionsbetween herring sperm DNA and imidazolium gemini surfactantswith a four-methylene spacer group ([Cn-4-Cnim]Br2, n = 10, 12,and 14) by dynamic light scattering (DLS), zeta potential circulardichroism (CD), and ethidium bromide (EtBr) displacement assay.The aim of this study is to characterize the physicochemical prop-erties of DNA/[Cn-4-Cnim]Br2, and to understand the effect of thehydrophobic chain lengths of [Cn-4-Cnim]Br2 on the DNA conden-sation, which could be beneficial for their potential gene deliveryapplications.

2. Experimental

2.1. Materials

The imidazolium gemini surfactants with a four-methylenespacer group ([Cn-4-Cnim]Br2, n = 10, 12, 14) were prepared andcharacterized according to the procedure reported previously [27].The chemical structures of [Cn-4-Cnim]Br2 are shown in Scheme 1.Deoxyribonucleic acid from herring sperm (sodium salt) was pur-chased from Sigma and used as received. The ratio of the absorbanceof the DNA stock solution at 260 nm to that at 280 nm was found tobe 1.9, suggesting that the DNA solution was protein-free. The fluo-rescent probe, EtBr, was obtained from Alfa Aesar with 98% purity.Water used in the experiments was triply distilled by a quartz waterpurification system. All solutions were prepared in 10 mM Tris–HClbuffer solution (pH 7.4). DNA stock solution was prepared by dis-solving an appropriate amount of the solid in buffer and storedat 4 ◦C for more than 72 h to get homogeneity. The concentrationof DNA solution is calculated by measuring the absorbance of DNAsolution according to Beer–Lambert equation and the molar extinc-tion coefficient of phosphate group at 260 nm (6600 M−1 cm−1)[34]. Stock solutions of surfactants were prepared by simple dis-solution. DNA and [Cn-4-Cnim]Br2 mixed solutions were dilutedto the designed charge ratios (Z+/−) prior to the measurements.The charge ratio (Z+/−) is the molar ratio between the positivelycharged imidazolium groups of the surfactant molecules and thenegatively charged phosphate groups of the DNA molecules. Eventhe net charge of one imidazolium group is 0.74 [28], for simplifi-cation, 2 positive charges in one surfactant molecule are assumedin this manuscript.

2.2. Hydrodynamic radius measurement

DLS measurements were carried out using Dawn Heleos light-scattering spectrometer (Wyatt Corporation) with a maximumwavelength � = 658 nm to investigate the size changes of DNA and

T. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 414 (2012) 33– 40 35

S 2. (A)

la3wcs

[noersaim

bawoftCRtCt4wacido

[C12-4-C12im]Br2, zeta potential measurements were performed atdifferent charge ratios. The zeta potential changes with the increas-ing amounts of [C12-4-C12im]Br2 are shown in Fig. 2 at a fixedDNA concentration (CDNA = 0.01 mmol/L). The zeta potential values

100 10 1 10 2 10 3 10 4

cheme 1. Molecular structures of imidazolium gemini surfactants [Cn-4-Cnim]Brenerated by using Chem3D software.

.3. Zeta potential measurement

Zeta potential measurements were performed by laser Dopplerlectrophoresis using a Zetasizer Nano ZS spectrometer (Malvernnstruments Ltd., U.K.) in a standard DTS1060 C zeta cell. The DNAoncentrations in all the samples for Zeta potential measurementsere held constant (CDNA = 0.01 mmol/L) whereas the contents of

Cn-4-Cnim]Br2 were varied.

.4. Circular dichroism

CD measurements were performed on a Jasco J-810 spectropo-arimeter with a 1.0 cm path quartz cell. Spectra were collectednd typically recorded over the wavelength range from 220 to20 nm at a bandwidth of 1.0 nm, and CDNA = 0.1 mmol/L. CD spectraere expressed as molar ellipticity, [�], in mdeg. The baseline was

orrected by subtracting the spectrum of 10 mM Tris–HCl bufferolution.

.5. Ethidium bromide exclusion assay

Fluorescence spectra of DNA/EtBr in the absence and presence ofCn-4-Cnim]Br2 were performed with a Perkin-Elmer LS-55 Lumi-escence Spectrometer using 1.0 cm quartz cells, and the slit widthsf excitation and emission were fixed at 10.0 nm. The fluorescencemission spectra of DNA solutions with and without EtBr wereecorded with an excitation wavelength of 480 nm. Intensity mea-urements were performed by keeping the concentrations of DNAnd EtBr constant (CDNA = 0.1 mmol/L, CEtBr = 0.01 mmol/L), and var-ous amounts of [Cn-4-Cnim]Br2 were added to the DNA/EtBr

ixture solution.

. Results and discussion

.1. Hydrodynamic radius of DNA/[C12-4-C12im]Br2

Hydrodynamic radii (RH) of DNA/[Cn-4-Cnim]Br2 were obtainedy DLS measurement for investigating the size changes of DNAnd DNA/[Cn-4-Cnim]Br2 complexes. For all the [Cn-4-Cnim]Br2ith different hydrophobic tail lengths, similar trends have been

bserved showing the initial decrease of RH to a minimum valueollowed by an increase. Typical results for [C12-4-C12im]Br2 showhe intensity-weighted size distribution of DNA and DNA/[C12-4-12im]Br2 solutions with Z+/− increasing (Fig. 1). The most probableH value of DNA solution is around 215 nm, which correspondso the translational mode of DNA molecules. Due to the very lowDNA, the contribution of the interactions between DNA moleculeso the RH value could be neglected [2]. With the increase of [C12--C12im]Br2 content, RH reduces to around 115 nm at Z+/− = 0.2,hich is smaller than that of pure DNA solution, which could be

scribed to the compaction of DNA molecules into densely packed

Chemical structure of [Cn-4-Cnim]Br2; (B) molecular structure of [C12-4-C12im]Br2

DNA/[C12-4-C12im]Br2. When Z+/− is further increased to 0.5, RHis greatly decreased to reach a value about 55 nm, which indi-cates the more compacted DNA structures. It should be notedthat, the sizes of the DNA/[C12-4-C12im]Br2 complexes reach theirminimum values at this charge ratio, and the further increase ofZ+/− leads to the increased RH till Z+/− = 1.0. The scattering inten-sity would fluctuate drastically (very high or very low, data notshown) when Z+/− above 1.0, which could be owing to the appear-ance of large aggregates or precipitation. The precipitation mightinduce the decrease of the scattering intensity in bulk solution. Ini-tially the neutralization of some DNA negative charges by bindingwith [C12-4-C12im]Br2 molecules gives rise to the collapse of theextended DNA chains after reaching the optimal charge ratio (0.5),which could be referred to the DNA compaction process. Basedupon our earlier studies on the apparent hydrodynamic radius of[Cn-s-Cnim]Br2 micelles formed in the aqueous solution [29], weobtained that the size of micells are 2–5 nm. Moreover, we havealso investigated the aggregation of pure [C12-4-C12im]Br2 on sili-con wafer by AFM [28], and few aggregates were observed on thesilica surface at the concentration of 3.6 mmol/L. Thus, RH values athigh Z+/− are not the size distribution of micelles or aggregates ofpure [C12-4-C12im]Br2. Further increase of Z+/− would lead to theincrease of the hydrodynamic radius, which could be attributedto the aggregation of [C12-4-C12im]Br2 molecules in the vicin-ity of DNA/[C12-4-C12im]Br2 via the hydrophobic interaction, orthe expansion of the DNA chains due to overcharging, or multi-molecular DNA condensation at the assistance of more imidazoliumgemini surfactant molecules.

3.2. Zeta potential of DNA/[C12-4-C12im]Br2

For gaining insights into the interaction mechanisms of DNA and

RH / nm

Fig. 1. Intensity-weighted distribution of DNA and DNA/[C12-4-C12im]Br2 solutions.The Z+/− values from bottom to top: 0 (DNA solution in the absence of [C12-4-C12im]Br2), 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.7, 0.8, and 1.0.

36 T. Zhou et al. / Colloids and Surfaces A: Physic

0.0 0.4 0.8 1.2 1.6-20

dt[itmdDztcTcn2mia

moDiaaDpoi

Fig. 2. Plots of zeta potential of DNA/[C12-4-C12im]Br2 versus Z+/− .

ecrease initially with the increase of [C12-4-C12im]Br2 concentra-ion till a minimum is reached at Z+/− = 0.4. Owing to the binding ofC12-4-C12im]Br2 to DNA, the moderate change of the DNA foldingnduced by the compacting agent leads to the concentrated nega-ive charges [35], which could be confirmed by the following CD

easurements. Another possible reason is that the original hid-en charges in the DNA chain become exposed on the surfaces ofNA molecules [36]. These would result in the initial decrease of

eta potential. Further increasing the charge ratio Z+/−, zeta poten-ial values increase after passing a minimum, indicating that theharges on DNA are gradually neutralized by [C12-4-C12im]Br2.he zeta potentials of DNA/[C12-4-C12im]Br2 are negative until theharge ratio approaches approximately 1.5. In addition, even theet charge of the imidazolium group is 0.74 [28], for simplification,

positive charges in one surfactant molecule are assumed in thisanuscript, so the charge neutralization of DNA/[C12-4-C12im]Br2

s not reached at Z+/− = 1, while the zeta potential approaches zerot approximately Z+/− = 1.5.

.3. Circular dichroism spectra of DNA/[C12-4-C12im]Br2

To confirm the effect of [C12-4-C12im]Br2 on the DNA confor-ations, CD measurements of DNA/[C12-4-C12im]Br2 were carried

ut at various charge ratios as shown in Fig. 3. CD spectrum ofNA solution exhibits a typical B-form conformation with a pos-

tive peak at 276 nm corresponding to base stacking with chirality, negative band at 244 nm due to the helicity, and a crossoverround 258 nm [37]. With the addition of [C12-4-C12im]Br2 into

220 240 260 280 300 320

Wavelength / nm

ig. 3. CD spectra of DNA in the presence of various concentrations of [C12-4-12im]Br2. The arrow indicates the increment of Z+/− , Z+/− values from top to bottom:

(DNA solution), 0.01, 0.05, 0.3, 0.5, 0.6, 0.7, 0.8, 1.0, 1.1, 1.3.

progressive loss of CD signals with increasing amounts of [C12-4-C12im]Br2 clearly demonstrate the condensing agent inducedconformational change of DNA. The signal intensity of DNA/[C12-4-C12im]Br2 at 276 nm decreases prominently from ∼4.6 for pureDNA solution to ∼2.8 at a molar charge ratio of 0.5. The changein the intensity of the CD peak centered at 276 nm is associatedwith the alteration of the helix hydration in the vicinity of phos-phate or the ribose ring [38]. It could be reasonable to assumethat [C12-4-C12im]Br2 will exchange with sodium ions present onthe DNA surface, which will in turn lead to changes in the hydra-tion shell near the phosphate group of DNA. As Z+/− is largerthan 0.5, the intensity reduces largely and the crossover is red-shifted significantly. However, when Z+/− reaches 1.0 and above,the positive band at 276 nm and the negative band at 244 nmfade away, which indicates that the addition of [C12-4-C12im]Br2alters the base stacking and helicity of DNA notably, or possibleprecipitation near Z+/− = 1.0 which was hardly visible in our exper-iments. Considering the minimum size of DNA/[C12-4-C12im]Br2at Z+/− = 0.5 and the minimum negative charges at Z+/− = 0.4, thecompaction of DNA molecules could occur by binding with smallamount of surfactants, which results in the smaller size and themoderate loss of CD signals. The slight difference between the crit-ical value Z+/− = 0.4 and 0.5 could be arising from the measurementerrors among different methods. Drastic DNA conformation changeand gradually increased hydrodynamic radius could be associatedwith the [C12-4-C12im]Br2 aggregate formation in the DNA/[C12-4-C12im]Br2 complex in the range of Z+/− = 0.5–1.0. The furtherincrease of hydrodynamic radius, totally denatured conformationand increased zeta potential indicate the multi-molecular DNA con-densation with more amphiphile molecules.

3.4. Study on DNA/[C12-4-C12im]Br2 by ethidium bromideexclusion

The fluorescence emission intensity is enhanced and the emis-sion wavelength changes from 613 nm to 608 nm for EtBr inthe presence of DNA (Fig. S1, Supporting information). The flu-orescence enhancement is associated with the EtBr intercalationwith DNA bases, and the blue-shifted emission indicates themore hydrophobic microenvironment for EtBr molecules [39].The fluorescence emission spectra are overlapped for [C10-4-C10im]Br2/EtBr, [C12-4-C12im]Br2/EtBr, [C14-4-C14im]Br2/EtBr andEtBr for all the measured concentrations of [Cn-4-Cnim]Br2 inour experiments, as shown in Fig. S1 (Supporting information),which indicates no interaction between EtBr and [Cn-4-Cnim]Br2molecules. In other aspects, we can also assume that, amphiphiles[Cn-4-Cnim]Br2 might form molecular aggregates, however, theseaggregates are not large enough for EtBr intercalation or solubili-zation.

The fluorescence emission spectra of EtBr and DNA/EtBr areshown in Fig. 4 in the presence of various amounts of [C12-4-C12im]Br2. The binding affinity of [C12-4-C12im]Br2 to DNA canbe determined by the degree of the fluorescence quenching dueto the displacement of EtBr from DNA/EtBr intercalated complexby formation of DNA/[C12-4-C12im]Br2. The fluorescent intensitiesof DNA/EtBr gradually decrease with increasing [C12-4-C12im]Br2.It is known that DNA base pairs provide the hydrophobic envi-ronment for protecting EtBr from water molecules that mayquench its fluorescence emission. When DNA is compacted or con-densed by cationic agents, some of the intercalated EtBr moleculesare excluded. As a consequence, fluorescence intensity will bequenched because of access of EtBr molecules to the bulk polar

T. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 414 (2012) 33– 40 37

550 600 650 700 7500

Wavelength(nm)

550 600 650 700 7500

Wavelength(nm)

F [C12-40 r).

fbCEatZaZwfzptTafrcsdboasD[fligCiait

ig. 4. Fluorescence emission spectra of DNA/EtBr containing different amounts of.5, 0.6, 0.8, 1.0, 1.1, EtBr. (B) Z+/− from bottom to top: EtBr, 1.1, 1.3, 1.5, 0 (DNA/EtB

ollowed by a sharp increase in fluorescence intensity. It shoulde noted that, in our experimental concentration range, [C12-4-12im]Br2 have no effect on the fluorescence intensity of puretBr solution (Fig. S1, Supporting information). Since the surfactantggregates are not large enough for EtBr solubilization, the con-inuous decrease of the fluorescence intensity are observed before+/− = 1.0, even though the congregation of [C12-4-C12im]Br2 havelready occurred around DNA/[C12-4-C12im]Br2 complex above+/− = 0.5. From CD results, the DNA would be totally denaturedhen Z+/− = 1.0, so some new kinds of condensates should be

ormed to provide hydrophobic microenvironment for solubili-ation of the excluded EtBr. Considering the above DLS, CD, zetaotential results, the EtBr exclusion assay could also indicateshe multi-molecular DNA/[C12-4-C12im]Br2 condensate formation.hese results are different from the interaction between DNA andlkyldimethylammonium gemini surfactant [41]. Zhao et al. [41]ound that the fluorescence intensity change slightly at high chargeatios, meaning that any further surfactant added to the solutionannot exclude EtBr from DNA. As is well known, all double-tranded DNA structures, the nucleobases are in direct �–� vaner Waals contact throughout the stack, and the planes of theases are separated by ca. 3.4 A, corresponding to the thicknessf the � system in an aromatic ring [42]. In the present work,s aromatic compounds (imidazolium cations with �-electronystems) [27,43], [Cn-4-Cnim]Br2 are capable of interacting withNA molecules via �–� stacking between imidazolium group of

Cn-4-Cnim]Br2 and nucleobases of DNA molecule. The increaseduorescence intensity could also be associated with the �–� stack-

ng between bases on denatured DNAs (Fig. 3) and imidazoliumroups of the surfactants within the multi-molecular DNA/[C12-4-12im]Br2 condensates. It should be noted that the fluorescence

ntensity drops but does not reach to the bulk level of EtBr solutionlone when DNA/[C12-4-C12im]Br2 condensates are formed, whichlluminates the incomplete displacement of EtBr molecules fromhe DNA helix.

.5. Comparison of DNA/[Cn-4-Cnim]Br2 (n = 10, 12, 14)

To determine the impact of the hydrophobic tails of [Cn-4-nim]Br2 on the DNA condensation, hydrodynamic radius, zetaotential, CD intensities at 276 nm, and relative fluorescence inten-ity of DNA/[Cn-4-Cnim]Br2 versus charge ratios are systemicallyompared among surfactants with different hydrophobic chainengths (as shown in Fig. 5).

As shown in Fig. 5A, with the increase of [Cn-4-Cnim]Br2

-C12im]Br2. (A) Z+/− from top to bottom: 0 (DNA/EtBr), 0.01, 0.05, 0.1, 0.2, 0.3, 0.4,

(Fig. S2, Supporting information). As observed in Fig. 5B, forDNA/[Cn-4-Cnim]Br2 systems the curves of zeta potential are quitedifferent. For DNA/[C10-4-C10im]Br2 system, the zeta potentialschange slightly, and one discernible minimum is located at aroundZ+/− = 1.0 while the overall charges of aggregates are negative. ForDNA/[C12-4-C12im]Br2 system, the zeta potential decreases ini-tially followed by an increase after passing a minimum at aroundZ+/− = 0.4, and then charge inversion takes place at Z+/− = 1.1. ForDNA/[C14-4-C14im]Br2 system, the initial sharp decrease in zetapotential with increasing [C14-4-C14im]Br2 is followed by an sharpincrease, and the charge inversion is observed at around Z+/− = 0.85.Taking into account the similarity of the charged groups in thesurfactants, the gradual decrease of the minimum charge ratiovalue in the curves could be ascribed to the formation of molec-ular aggregates for [C12-4-C12im]Br2 and [C14-4-C14im]Br2 due tothe increased hydrophobic interaction. CD intensities at 276 nm ofDNA/[Cn-4-Cnim]Br2 decrease upon increment of Z+/−, which areshown in Fig. 5C. When Z+/− is larger than 0.5, the intensities at276 nm are all tend to decrease rapidly, however CD signals ofthe solution containing [Cn-4-Cnim]Br2 with a longer alkyl chaindecrease more rapidly. That is, a smaller amount of the longer tailsurfactant is needed to reduce CD intensity of DNA. With the chargeratio further increasing, DNA molecules are denaturalized by [C14-4-C14im]Br2 and [C12-4-C12im]Br2 at Z+/− = 0.7 and 1.0 respectively(CD signal is 0), while at Z+/− = 1.0 the intensity of CD signal forDNA/[C10-4-C10im]Br2 is still about 1.77 (Fig. S3, Supporting infor-mation). The variations in CD signal for different [Cn-4-Cnim]Br2at higher charge ratios could be associated with the differentDNA compaction or condensation characteristics arising from theincreased hydrophobic interaction. In Fig. 5D, the relative fluo-rescent intensities I/I0, i.e., the fluorescent intensities with (I) andwithout surfactant (I0), change slightly when Z+/− are below 0.05,which could be ascribed to the binding of [Cn-4-Cnim]Br2 to DNAand displacing small amount of EtBr molecules from DNA. Theintensity decreases evidently with increasing Z+/− which indicatesmore EtBr molecules are displaced by [Cn-4-Cnim]Br2. It is evi-dent in Fig. 5D that, the I/I0 values of DNA/EtBr/[Cn-4-Cnim]Br2reach minimum at Z+/− of 1.3, 1.1, and 0.8 for [Cn-4-Cnim]Br2with 10, 12, and 14 carbon chains respectively. At higher chargeratios, I/I0 of DNA/EtBr levels off for [C10-4-C10im]Br2, whichcould be associated with the saturation of cationic agents to dis-place EtBr from DNA [21]. However, the fluorescent intensities forDNA/EtBr/[C12-4-C12im]Br2 and DNA/EtBr/[C14-4-C14im]Br2 solu-tions reach minimum values and increase again with the increase ofcharge ratios (Fig. S4, Supporting information). The variations of CD,

38 T. Zhou et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 414 (2012) 33– 40

(C), a

Acaoa[i

tiifactmpivtmabttslmci

sion of the DNA chains due to overcharging could also lead to theincreasing of complexes diameters (lower part of Scheme 2(A)c).With more and more [Cn-4-Cnim]Br2 molecules binding to DNAchain, considering the above DLS, CD, zeta potential and EtBr

Scheme 2. (A) Schematic illustration of DNA compaction to multi-molecular DNAcondensation progress with the increasing of [Cn-4-Cnim]Br2. (a) The free DNAstrand with loose and flexible conformation. (b) Compacted DNA with minimumsize. (c) Compacted DNA with slightly expanded size. (d) Multi-molecular DNA con-

Fig. 5. Hydrodynamic radius (A), zeta potential (B), CD intensities at 276 nm

s can be seen, the longer the hydrophobic chains, the lower themc value is. Therefore, this would result in the decrease of agentmount needed for DNA condensation for [C14-4-C14im]Br2. More-ver, we could easily get the conclusion that, besides electrostaticttractions, the stronger hydrophobic interaction between DNA andCn-4-Cnim]Br2 with longer tails also play an important role in bind-ng of [Cn-4-Cnim]Br2 to DNA.

.6. Mechanism of the interaction between DNA andCn-4-Cnim]Br2

Based on the results and discussion above, the mechanism ofhe interaction between DNA and [Cn-4-Cnim]Br2 with increas-ng surfactant concentration is depicted in Scheme 2. As shownn Scheme 2(A), the DNA molecule keeps the loose strand-like con-ormation in nature state (Scheme 2(A)a). When [Cn-4-Cnim]Br2re mixed with DNA, the neutralization of some DNA negativeharges by binding with [Cn-4-Cnim]Br2 molecules gives rise tohe collapse of the extended DNA chains after reaching the opti-

al charge ratio, which could be referred to the DNA compactionrocess (Scheme 2(A)b). Further increase of Z+/− would lead to the

ncrease of the size and charge for DNA/[Cn-4-Cnim]Br2. In our pre-ious report [27], imidazolium gemini surfactants were reportedo have lower cmc than conventional single-chain surfactants. That

eans [Cn-4-Cnim]Br2 possess greater ability to form premicellarggregates in the solution. The structure of surfactant dimer coulde described as the interdigitation of hydrophobic tails leavingheir hydrophilic head-groups at opposite ends, and the surfac-ant oligomer could adopt micelle-like structures with granular orpherical morphology, and the existence of this type of premicel-

oderate degree of increasing of the hydrodynamic radius, whichould be attributed to the aggregation of [Cn-4-Cnim]Br2 moleculesn the vicinity of DNA/[Cn-4-Cnim]Br2 via the hydrophobic

nd relative fluorescence intensity I/I0 (D) of DNA/[Cn-4-Cnim]Br2 versus Z+/− .

interaction (upper part of Scheme 2(A)c). In other aspect, the expan-

densation. The purple “n”-shaped structures stand for the [Cn-4-Cnim]Br2 molecules,the green string-like structures for DNA strands. (B) Schematic illustration of thethree types of DNA and [Cn-4-Cnim]Br2 interactions, electrostatic interaction (a),hydrophobic interaction (b), and �–� interaction (c). The blue “Y”-shaped structureswith attached red lines for the zoom-in view of the double-stranded DNA.

Physic

ecsccotcc

DsaDccb(btc

nDbDmwtsFtbh[gtamgt

T. Zhou et al. / Colloids and Surfaces A:

xclude assay results, the multi-molecular DNA/[C12-4-C12im]Br2ondensate (Scheme 2(A)d) could be formed with large complexizes and totally denatured conformation of DNA molecules (pre-ipitation also might happen). The hydrophobic microenvironmentould be formed for solubilization of excluded EtBr by aggregationf surfactant molecules between looped DNA strands. From the sys-ematic characterizations, the DNA-condensing agent interactionsould undergo the process from compaction to multi-molecularondensation with the increase of amount of condensing agents.

The compaction of the DNA molecules and multi-molecularNA condensation are induced by cationic imidazolium gemini

urfactant, which may be mainly attributed to the electrostaticttractions (Scheme 2(B)a) between negative phosphate groups ofNA and positive imidazolium head-groups (an ion exchange ofationic imidazole charge of [Cn-4-Cnim]Br2 with Na+ in the ionondensation region around DNA phosphates), and the hydropho-ic interaction between alkyl tails of surfactant and DNA base pairsScheme 2(B)b), and the �–� stacking (Scheme 2(B)c) betweenases on denatured DNAs (Fig. 3) and imidazolium groups ofhe surfactants within the multi-molecular DNA/[C12-4-C12im]Br2ondensates (Fig. 1).

. Conclusions

Cationic imidazolium gemini surfactants ([Cn-4-Cnim]Br2, = 10, 12, 14) as novel amphiphilic molecules can interact withNA via attractive electrostatic interaction, strong hydropho-ic force and �–� interaction. Upon addition of [Cn-4-Cnim]Br2,NA molecules undergo the process from compaction to multi-olecular condensation accompanied by conformation change,hich could be evidenced by the DLS and CD results. The impact of

he electrostatic interaction can be deduced from the charge den-ity and charge polarity changes by zeta potential measurements.rom the comparison between DNA compaction and condensa-ion by [Cn-4-Cnim]Br2 (n = 10, 12, 14), the stronger interactionetween DNA and [Cn-4-Cnim]Br2 with longer tails indicates thatydrophobic interaction also play an important role in binding ofCn-4-Cnim]Br2 to DNA. The �-� interaction between imidazoliumroups of surfactants and DNA aromatic rings also contributes inhe DNA/[Cn-4-Cnim]Br2 complex formation from EtBr exclusionssay. The demonstrated process for DNA compaction and multi-olecular condensation could be beneficial for the application of

emini surfactants as DNA condensing agents in gene delivery andransfection.

cknowledgments

We acknowledge financial support from the Natural Scienceoundation of China (20833010, 20973043). Financial support fromhe National Basic Research Program of China (2009CB9301000) islso gratefully acknowledged. The authors thank Prof. Xia Wu fromhandong University for providing herring sperm DNA.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/.colsurfa.2012.08.060.

eferences

[1] S.M. Mel’nikov, V.G. Sergeyev, K. Yoshikawa, Discrete coil-globule transition

[2] R.S. Dias, J. Innerlohinger, O. Glatter, M.G. Miguel, B. Lindman, Coil-globuletransition of DNA molecules induced by cationic surfactants: a dynamic lightscattering study, J. Phys. Chem. B109 (2005) 10458–10463.

ochem. Eng. Aspects 414 (2012) 33– 40 39

[3] M.B. Bally, P. Harvie, F.M.P. Wong, S. Kong, E.K. Wasan, D. Reimer, Biologicalbarriers to cellular delivery of lipid-based DNA carriers, Adv. Drug. Deliv. Rev.38 (1999) 291–315.

[4] C.H. Spink, J.B. Chaires, Thermodynamics of the binding of a cationic lipid toDNA, J. Am. Chem. Soc. 119 (1997) 10920–10928.

[5] F.M.P. Wong, D.L. Reimer, M.B. Bally, Cationic lipid binding to DNA: character-ization of complex formation, Biochemistry 35 (1996) 5756–5763.

[6] S. Marchetti, G. Onori, C. Cametti, DNA condensation induced by cationic sur-factant: a viscosimetry and dynamic light scattering study, J. Phys. Chem. B109(2005) 3676–3780.

[7] V. Jadhav, S. Maiti, Effect of the head-group geometry of amino acid-basedcationic surfactants on interaction with plasmid DNA, Biomacromolecules 9(2008) 1852–1869.

[8] X. Guo, B. Cui, H. Li, Z.Y. Gong, R. Guo, Facilitation effect of oligonucleotideon vesicle formation from single-chained cationic surfactant-dependences ofoligonucleotide sequence and size and surfactant structure, J. Polym. Sci. Polym.Chem. 47 (2009) 434–449.

[9] M.C. Morán, M.G. Miguel, B. Lindman, DNA gel particles from single and double-tail surfactants: supramolecular assemblies and release characteristics, SoftMatter 7 (2011) 2001–2010.

10] T.M. Reineke, Poly(glycoamidoamine)s: cationic glycopolymers for DNA deliv-ery, J. Polym. Sci. Polym. Chem. 44 (2006) 6895–6908.

11] M.L. Ainalem, T. Nylander, DNA condensation using cationic dendrimers-morphology and supramolecular structure of formed aggregates, Soft Matter 7(2011) 4577–4594.

12] H. Chen, Q.C. Zou, H. Yu, M. Peng, G.W. Song, J.Z. Zhang, S.G. Chai, Y.H. Zhang, C.E.Yan, Study on interaction between cationic polystyrene nanoparticles and DNA,and the detection of DNA by resonance light scattering technology, Microchim.Acta 168 (2010) 331–340.

13] M. Tecle, M. Preuss, A.D. Miller, Kinetic study of DNA condensation by cationicpeptides used in nonviral gene therapy: analogy of DNA condensation to pro-tein folding, Biochemistry 42 (2003) 10343–10347.

14] P.S. Kuhn, M.C. Barbosa, Y. Levin, Complexation of DNA with cationic surfactant,Physica A 269 (1999) 278–284.

15] J.P. Clamme, S. Bernacchi, C. Vuilleumier, G. Duportail, Y. Mély, Gene transfer bycationic surfactants is essentially limited by the trapping of the surfactant/DNAcomplexes onto the cell membrane: a fluorescence investigation, Biochim. Bio-phys. Acta 1467 (2000) 347–361.

16] D. Llères, J.P. Clamme, E. Dauty, T. Blessing, G. Krishnamoorthy, G. Dupor-tail, Y. Mély, Investigation of the stability of dimeric cationic surfactant/DNAcomplexes and their interaction with model membrane systems, Langmuir 18(2002) 10340–10347.

17] C. Leal, E. Moniri, L. Pegado, H. Wennerstrom, Electrostatic attraction betweenDNA and a cationic surfactant aggregate. The screening effect of salt, J. Phys.Chem. B111 (2007) 5999–6005.

18] R.L. Sheng, T. Luo, Y.D. Zhu, H. Li, A.M. Cao, Interactions of new synthesized flu-orescent cationic amphiphiles bearing pyrene hydrophobe with plasmid DNA:binding affinities, aggregation and intracellular uptake, Macromol. Biosci. 10(2010) 974–982.

19] D. Matulis, I. Rouzina, V.A. Bloomfield, Thermodynamics of cationic lipid bind-ing to DNA and DNA condensation: roles of electrostatics and hydrophobicity,J. Am. Chem. Soc. 124 (2002) 7331–7342.

20] D.M. Zhu, R.K. Evans, Molecular mechanism and thermodynamics study of plas-mid DNA and cationic surfactants interactions, Langmuir 22 (2006) 3735–3743.

21] V.M. Jadhav, R. Valaske, S. Maiti, Interaction between 14mer DNA oligonu-cleotide and cationic surfactants of various chain lengths, J. Phys. Chem. B112(2008) 8824–8831.

22] A.J. Kirby, P. Camilleri, J.B.F.N. Engberts, M.C. Feiters, R.J.M. Nolte, et al., Geminisurfactants: new synthetic vectors for gene transfection, Angew. Chem. Int. Ed.42 (2003) 1448–1457.

23] L. Karlsson, M.C.P. Eijk, O.J. Soderman, Compaction of DNA by gemini sur-factants: effects of surfactant architecture, J. Colloid Interface Sci. 252 (2002)290–296.

24] X.D. Chen, J.B. Wang, N. Shen, Y.H. Luo, L. Li, M.H. Liu, R.K. Thomas, Geminisurfactant/DNA complex monolayers at the air–water interface: effect of sur-factant structure on the assembly, stability, and topography of monolayers,Langmuir 18 (2002) 6222–6228.

25] C.Z. Wang, X.F. Li, S.D. Wetting, I. Badea, M. Foldvari, R.E. Verrall, Investiga-tion of complexes formed by interaction of cationic gemini surfactants withdeoxyribonucleic acid, Phys. Chem. Chem. Phys. 9 (2007) 1616–1628.

26] X.L. Wang, X.H. Zhang, M.W. Cao, H.Z. Zheng, B. Xiao, Y.L. Wang, M. Li, Geminisurfactant-induced DNA condensation into a beadlike structure, J. Phys. Chem.B113 (2009) 2328–2332.

27] M.Q. Ao, G.Y. Xu, Y.Y. Zhu, Y. Bai, Synthesis and properties of ionic liquid-type gemini imidazolium surfactants, J. Colloid Interface Sci. 326 (2008)490–495.

28] M.Q. Ao, G.Y. Xu, J.Y. Pang, T.T. Zhao, Comparison of aggregation behaviorsbetween ionic liquid-type imidazolium gemini surfactant [C12-4-C12im]Br2and its monomer [C12mim]Br on silicon wafer, Langmuir 25 (2009) 9721–9727.

29] M.Q. Ao, P.P. Huang, G.Y. Xu, X.D. Yang, Y.J. Wang, Aggregation and thermo-

30] M.Q. Ao, G.Y. Xu, W.L. Kang, L.W. Meng, H.J. Gong, T. Zhou, Surface rheologi-cal behavior of gelatin/ionic liquid-type imidazolium gemini surfactant mixedsystems, Soft Matter 7 (2011) 1199–1206.

4 Physic

0 T. Zhou et al. / Colloids and Surfaces A:

31] T. Inoue, H. Ebina, B. Dong, L.Q. Zheng, Electrical conductivity study on micelleformation of long-chain imidazolium ionic liquids in aqueous solution, J. ColloidInterface Sci. 314 (2007) 236–241.

32] Y.H. Ding, L. Zhang, J. Xie, R. Guo, Binding characteristics and molecular mech-anism of interaction between ionic liquid and DNA, J. Phys. Chem. B114 (2010)2033–2043.

33] W. Yu, K.F. Pirollo, B. Yu, A. Rait, L.M. Xiang, W.Q. Huang, Q. Zhou, G.Ertem, E.H. Chang, Enhanced transfection effciency of a systemically deliveredtumor-targeting immunolipoplex by inclusion of a pH-sensitive histidylatedoligolysine peptide, Nucleic Acids Res. 32 (2004) e48.

34] E. Chargaff, R. Lipshitz, Composition of mammalian desoxyribonucleic acids, J.Am. Chem. Soc. 75 (1953) 3658–3661.

35] A.F. Jorge, R.S. Dias, J.C. Pereira, A.A.C.C. Pais, DNA condensation by pH-responsive polycations, Biomacromolecules 11 (2010) 2399–2406.

36] X.F. Zhao, Y.Z. Shang, H.L. Liu, Y. Hu, Complexation of DNA with cationic geminisurfactant in aqueous solution, J. Colloid Interface Sci. 314 (2007) 478–483.

37] J. Kypr, I. Kejnovská, D. Renciuk, M. Vorlícková, Circular dichroism and con-formational polymorphism of DNA, Nucleic Acids Res. 37 (2009) 1713–1725.

38] M.C. Mercado, M. Tomasz, Circular dichroism of mitomycin–DNA complexes.Evidence for 26 a conformational change in DNA, Biochemistry 16 (1977)2040–2046.

39] R.F. Pasternack, M. Caccam, B. Keogh, T.A. Stephenson, A.P. Williams, E.J. Gibbs,Long-range fluorescence quenching of ethidium ion by cationic porphyrins inthe presence of DNA, J. Am. Chem. Soc. 113 (1991) 6835–6840.

40] A. Rodríguez-Pulido, E. Aicart, E. Junquera, Electrochemical and spectroscopicstudy of octadecyltrimethylammonium bromide/DNA surfoplexes, Langmuir25 (2009) 4402–4411.

41] X.F. Zhao, Y.Z. Shang, J. Hu, H.L. Liu, Y. Hu, Biophysical characterization of com-plexation of DNA with oppositely charged gemini surfactant 12-3-12, Biophys.Chem. 138 (2008) 144–149.

42] E.T. Kool, Replacing the nucleobases in DNA with designer molecules, Acc.Chem. Res. 35 (2002) 936–943.

43] L. Cardoso, N.M. Micaelo, DNA molecular solvation in neat ionic liquids,ChemPhysChem 12 (2011) 275–277.

44] L.D. Song, M.J. Rosen, Surface properties, micellization, and premicellar aggre-gation of gemini surfactants with rigid and flexible spacers, Langmuir 12 (1996)1149–1153.

DNA compaction to multi-molecular DNA condensation induced by cationic imidazolium gemini...

Documents