Preparation of Fe3O4/poly(styrene-butyl acrylate-[2-(methacryloxy)ethyl]trimethylammonium chloride) by emulsifier-free emulsionpolymerization and its interaction with DNA

Xiaolong Li a, Guoqiang Liu a, Wei Yan a, Paul K. Chu b, Kelvin W.K. Yeung c, Shuilin Wu a,b,c,Changfeng Yi a,b,c, Zushun Xu a,b,c,n

a Ministry-of-Education Key Laboratory for the Green Preparation and Application of Functional Materials, Hubei University, Wuhan, PR Chinab Department of Physics & Materials Science, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, PR Chinac Division of Spine Surgery, Department of Orthopaedics and Traumatology, The University of Hong Kong, Pokfulam, Hong Kong, PR China

a r t i c l e i n f o

Article history:

Received 7 March 2011

Received in revised form

15 November 2011Available online 9 December 2011

Keywords:

Magnetic polymer

Polycations

Emulsifier-free emulsion polymerization

DNA

Fluorescence spectroscopy

a b s t r a c t

Cationic magnetic polymer particles Fe3O4/poly(styrene-butyl acrylate-[2-(methacryloxy)ethyl]tri-

methylammonium chloride), a type of potential gene carrier, were prepared by emulsifier-free

emulsion polymerization with oleic acid modified magnetite Fe3O4, styrene, butyl acrylate and

[2-(methacryloxy)ethyl]trimethylammonium chloride) (METAC). The morphology of the particles was

characterized by transmission electron microscopy and the composites of particles were characterized

by FT-IR spectroscopy, X-ray diffraction. These results showed that magnetic particles were well

dispersed in polymers with the content of about 15%(wt/wt). The composites exhibited super-

paramagnetism and possessed a certain level of magnetic response. The interactions between the

particles with calf-thymus DNA (ct DNA) were confirmed by zeta potential measurement, UV–vis

spectroscopy and fluorescence spectroscopy. The DNA-binding capacity determined by the agarose gel

electrophoresis showed good binding capacity of the emulsion to DNA. These results suggested the

potential of the cationic magnetic polymer emulsion as gene target delivery carrier.

& 2011 Elsevier B.V. All rights reserved.

1. Introduction

Magnetic polymer microspheres consisting of magnetic coreand polymer shell have received much attention due to their widerange of potential applications in the fields such as rapid proteinimmobilization and separation [1–3], drug carrier and delivery[4–6], magnetic resonance imaging (MRI) [7–9]. As the core of thecomposite microsphere, magnetic nanoparticles (MNPs) can bemanipulated by an external magnetic force. In addition, magneticcore can also respond to an alternation magnetic field, leading tothe temperature elevating. Above certain temperature, the cancercells can be effectively killed, which can be applied for cancertreatment. If magnetic particles embedded in temperature sensitivepolymers or liposome, the composites can combine hyperthermiatherapy with drug delivery to provide a synergistic treatmentstrategy [10] .To perform real-time bioapplication, MNPs are oftenencapsulated with functional polymers containing carboxyl [3,11],hydroxide [12,13], epoxy [14,15], amino [16,17] and other functional

groups [18], which are easy to covalent link with various bimole-cular like enzymes, peptides and nucleic acid.

Cationic lipids and cationic polymers have recently gainedincreasing interest in gene therapy as a kind of non-viral transfervectors [19,20]. Typically, interaction of cationic vectors withDNA involves both electrostatic and hydrophobic interactions.The positively charged groups of the vectors are able to combinewith the negatively charged phosphates of DNA via electrostaticinteraction. The hydrophobic interaction comes from the compo-nents of the DNA duplex and polymer or lipid backbone [21].Compared to cationic lipids, cationic polymers are more attractivedues to they are stable, easy to manipulate, more economical andmore tailorable for functionalization such as incorporation oftargeting ligands [22]. A variety of polycations have been pro-posed and investigated for non-viral transfer system, such aspoly-L-lysine (PLL) [23], polyethyleneimine (PEI) [24,25], chitosan[26,27], methacryl oxyethyl trimethylammonium chloride(MOTAC) [28,29]. Furthermore, a number of strategies have beenexplored based on polycations combined with magnetite becauseassociation of polycations with magnetic particles allows thevectors to respond to an applied magnetic force, which enablesto speed up the gene transfer vectors to the target cell surface,shorten the time of gene delivery and allow the use of lower dose

Contents lists available at SciVerse ScienceDirect

journal homepage: www.elsevier.com/locate/jmmm

Journal of Magnetism and Magnetic Materials

0304-8853/$ - see front matter & 2011 Elsevier B.V. All rights reserved.

doi:10.1016/j.jmmm.2011.11.056

n Corresponding author at: Ministry-of-Education Key Laboratory for the Green

Preparation and Application of Functional Materials, Hubei University, Wuhan

430062, PR China. Tel.: þ86 27 88661879; fax: þ86 27 88665610.

E-mail address: zushun25@yahoo.com.cn (Z. Xu).

Journal of Magnetism and Magnetic Materials 324 (2012) 1410–1418

of vectors with high efficiency, thus minimize the toxic riskcaused by high concentration of gene delivery vectors [30]. Mostof them introduced polyethylenimine (PEI) by electrostatic adher-ence [31,32] or utilizing linker [33] onto magnetic particles.However, the stability of electrostatic adherence is easilydestroyed when the biological solution is changed [34]. Thoughthe high molecular weights of PEI ensure the binding capacitywith DNA, they induce insufficient linkage with magnetic parti-cles, cause extensive bridging flocculation and lead the morphol-ogy uncontrollable [18]. Encapsulation magnetic particles withmonomers by in-situ polymerization may solve these drawbacksbecause small molecules can easily adsorb onto the surface ofmagnetic particles without the steric hindrance of long chains andthe morphology of the polymer shell can be more controllable bychanging the process of polymerization.

At present, many approaches have been employed to preparemagnetic polymer microspheres with functional monomers, suchas emulsion polymerization [35,36], seed-emulsion polymerization[37], dispersion polymerization [38,39], emulsifier-free emulsionpolymerization [40], miniemulsion polymerization [41]. Amongthese methods, emulsifier-free emulsion polymerization invokesmore interest in bioapplication because this method without initi-ally adding surfactants can produce monodisperse and ‘‘clean’’particles, avoiding cytotoxic and antibacterial effects of the emulsi-fier. Several methodologies for the preparation of emulsifier-freemagnetic polymer microsphere have been reported. Chiu et al.prepared Fe3O4/PMMA composite particles and studied nucleationmechanism of emulsifier-free polymerization [40]. Xie et al. inves-tigated the emulsifier-free emulsion polymerization of styrene-butylacrylate-methacrylic acid [poly(St-BA-MAA)] in different kinds ofpolar solvent [42].

However, to our knowledge, those studies just focus onemulsifier-free emulsion polymerization with negatively chargedsurface. Reports on cationic magnetic particles are obviously farfewer compared to those on anionic microspheres. Cationicmagnetic microsphere prepared by emulsifier-free polymeriza-tion may be a new way to obtain functional composites in genetherapy as mentioned above. Herein, our work focuses on explor-ing the preparation of cationic magnetic polymer emulsion byemulsifier-free emulsion polymerization and the interactionswith DNA in vitro. The objective is to explore the prospect ofthe application of polymer nanoparticles as gene carriers anddrug delivery. In this study, copolymer shell were comprised ofstyrene, butyl acrylate and a cationic comonomer of METAC,which was selected as a DNA-binding site because it providespermanent high charged to the particles [43]. The final magneticemulsions were characterized by X-ray diffraction (XRD), vibrat-ing sample magnetometer (VSM), Fourier transform infraredspectroscopy (FT-IR), thermogravimetric analysis (TGA), trans-mission electron microscopy (TEM) and Mastersizer particle sizeanalyzers. The physiochemical characteristics of the emulsion/DNA complexes were analyzed by zeta potential measurements,UV–vis spectrophotometry, fluorescent spectroscopy and agarosegel electrophoresis.

2. Material and methods

2.1. Materials

Iron(III) chloride hexahydrate (FeCl3 �6H2O), iron(II) chloridetetrahydrate (FeCl2 �4H2O), oleic acid (OA), a 36% ammoniumhydroxide solution (NH3 �H2O) and ethanol, isopropanol, hexanewere purchased from Sinopharm Chemical Reagent Co. Ltd, China.Styrene (St) and butyl acrylate (BA) were distilled under reducedpressure and stored at 5 1C. All the reagents were AR.

2,2-azobis-(2-amidinopropane) hydrochloride (AIBA) from Aldrichis 97% pure. An aqueous solution (75 wt%) of [2-(methacryloxy)ethyl]trimethylammonium chloride) (METAC) was purchased fromAldrich and used as received and distilled water was used throughoutthe study.

Calf-thymus DNA (ct DNA) was purchased from Sigma (St.Louis, MO, USA).The stock solution of calf-thymus DNA wasprepared by dissolving ct DNA in doubly distilled water andstored at 0–4 1C. The concentration of working solution DNA was64 mg/L. Plasmid DNA (2.5 kbp), loading buffer and other bio-chemicals were supplied by the group of Prof. Ma (Faculty ofBiology Science, Hubei University).

2.2. Preparation of magnetic nanoparticles

Fe3O4 nanoparticles were prepared as the method of our pre-vious work without sodium dodecylsulphate [44], which consists ofFe(III) and Fe(II) coprecipitation in alkaline solution. In detail, 13.5 gFeCl3 �6H2O and 6 g FeCl2 �4H2O were dissolved in 150 mL distilledwater under nitrogen at room temperature, 40 mL NH3 �H2O wasquickly added into the solution with vigorous stirring, the mixturerapidly forming a black precipitate. Then the mixture was heated to60 1C for 1 h. The black precipitate was isolated from the solution bymagnetic separation and washed with distilled water until the pHvalue reached 7.

2.3. Modification of the magnetic nanoparticles

The modification of the magnetic nanoparticles was carriedout in a three-necked flask equipped with a stirring paddle, acondenser and nitrogen inlet. The wet precipitate (50 g), 200 mLdistilled water were added into the reactor and stirred. When themixture was heated to 80 1C, 3 g oleic acid was introduced tomodify the particles. The system was kept at 80 1C for 1 h. Thenthe reaction was ended by cooling the mixture to room tempera-ture, the oleic acid-modified magnetite fluid was separated by amagnet and washed several times with ethanol and hexane untilthe upper solution became transparent. Then the black slurry wasdispersed in hexane.

2.4. Preparation of cationic magnetic composite Fe3O4/poly(St-BA-

METAC) particles

The magnetic slurry (0.5 g), St (2.3 g) and BA (2.0 g) consti-tuted the oil phase, and the METAC (0.5 g), isopropanol (20 mL)and water (80 mL) constituted the aqueous phase. Two differentphases were then put into a 250 mL three-necked flask andhomogenized at 50 1C with vigorous stirring for half an hour.The polymerization was carried out after 0.3 g AIBA was addedinto the flask and the temperature was heated to 80 1C. The air inthe flask was replaced by a stream of nitrogen and the mixturewas kept under nitrogen atmosphere until the polymerizationsustained for 10 h. The resulting composite emulsion was purified byovernight dialysis in distilled water solution to remove isopropanoland monomers. Then the purified emulsion was added dropwise tosaturation calcium chloride/methanol solution to remove unreactedspecies from the composites. The resultant floccus was washed withmethanol and water for several times, and then dried under vacuumfor 12 h at 40 1C. In the end, the powder was obtained for furthercharacterization.

3. Characterization

The morphology and structure of the composite emulsion weredetermined by transmission electron microscopy (TEM, Tecnai G20,

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–1418 1411

FEI Corp., USA). The oleic acid-modified magnetite Fe3O4 particleswere dispersed in hexane, the poly (St-BA-METAC) emulsion andFe3O4/poly(St-BA-METAC) composite emulsion were diluted withdistilled water, respectively. Then all of them were dropped ontocooper grids and air-dried for TEM measurement. The TEM imageswere obtained at 25 1C at an electron acceleration voltage of 200 kV.

Fourier transform infrared spectroscopy (FT-IR) analysis of thesamples was taken on a Spectrum One FT-IR spectrometer (Perkin-Bhaskar-Elmer Co., USA). The measurement was carried out usingKBr pellets for powder of oleic acid-modified magnetite and Fe3O4 /poly(St-BA-METAC) powder.

A crystallographic study of magnetic and composite particlepowder was performed on an X-ray diffractometer (D/MAX-IIIC,Japan). The diffraction (XRD) patterns were taken from 20 to 901(2y) using CuKa radiation; scanning speed was kept at 151/min.

The magnetite content of the dried samples was measured bythermogravimetric analysis (TGA-7, Perkin-Elmer), the powderwere heated 10 1C/min from room temperature to 650 1C under anitrogen atmosphere.

The magnetic properties of the powder were measured byusing a vibrating-sample magnetometer (VSM, HH-15, China) at298 K with a 12 kOe applied magnetic field.

The DNA-binding capacity of Fe3O4/poly(St-BA-METAC) emulsionwas examined using Malvern Nano-ZS, Model ZEN3600 (MalvernInstruments, Malvern, UK). DNA and emulsion were dissolved in10mM phosphate buffer with pH 7.3 and mixed to obtain differentproportions of emulsion/DNA samples. The DNA concentration waskept at 64 mg/L before analysis. The surface charge density of cationicmagnetic emulsion was also determined in different pH atmosphereby zeta potential measurement. The analyses were carried out at25 1C and the zeta potential was an average of three measurements.

The cationic magnetic emulsion with ct DNA interaction wasinvestigated by UV–vis spectroscopy. DNA and magnetic emul-sion were dissolved in ultrapure water and then mixed to obtainvarious magnetic emulsion/DNA samples. The UV–vis spectro-scopy was measured with a Perkin-Elmer Lambda 17 UV–visspectrophotometer at room temperature, the measurements weremade in a quartz cuvette (1 cm in width) in the wavelength rangeof 200–400 nm and the absorption at 260 nm for DNA was used.

The fluorescent properties of magnetic emulsion in the pre-sence of DNA were studied on a RF-540 (Hitachi high-technologiescorporation, Tokyo, Japan) spectrometer. 3 mL of magnetic emul-sion (0.32 g/L) and DNA solutions with different concentrations inthe range of 0–3.252�10�6 mol/L in steps of 8.13�10�7 mol/Lwere added to a 1.0 cm quartz cuvette. The fluorescence emissionspectra were recorded in the wavelength range of 300–500 nm byexciting the magnetic emulsion at 270 nm. Both the excitation andemission slits were 5 nm.

The DNA condensation ability of the emulsion was studied bythe means of an agarose gel retardation assay. Emulsion/DNA(2.5 kbp) complexes with different weight ratios were preparedby adding different weights of emulsion to 0.48 mg DNA. Then thecomplexes were eletrophoresed in the 0.7%(w/v) agarose gelusing TAE running buffer at 120 V for 40 min. DNA was detectedby staining with ethidium bromide and visualized with a UV lampon a Vilber Lourmat imaging system (France).

4. Results and discussion

4.1. Characterization of Fe3O4 /poly(St-BA-METAC) particles

4.1.1. FT-IR spectra

Fig. 1 shows the FT-IR spectra of modified Fe3O4 nanoparticles(a) and Fe3O4/poly(St-BA-METAC) particles (b). The peak at578 cm�1 corresponds to the Fe–O vibration of Fe3O4, which

appears both in Fig. 1a and b, demonstrating the existence ofFe3O4 in the composite latexes. As shown in Fig. 1b, the absorp-tion peaks at 3100–3000 cm�1 correspond to the characteristic ofvibration C–H of benzene ring, the bands at 1602 and 1494 cm�1

can be attributed to benzene ring vibrations (gC–C) of polystyreneand peaks 698 and 760 cm�1 correspond to flexural vibrations(dC–H) of single-substitute benzene ring. Moreover, the absorptionpeak at 1453 cm�1 belongs to bending vibration of CH2 in –CH2–Nþ (CH3)3 and the band at 1732 cm�1 is observed to thestretching vibration of C¼O group of BA and METAC. Based onabove observations, the magnetic nanoparticles have been suc-cessfully embedded into the polymers.

4.1.2. Morphology and structure of the magnetic Fe3O4 /poly(St-BA-

METAC) particles

The magnetic emulsion was first analyzed by TEM in order toprovide insights into the structure of composite nanoparticles.Fig. 2 shows the TEM images of oleic acid-modified magnetiteFe3O4 nanoparticles, poly(St-BA-METAC) particles and Fe3O4/poly(St-BA-METAC) particles. As shown in Fig. 2a, the diameter ofoleic acid-modified magnetite Fe3O4 nanoparticles is around10 nm and the particles are so small that some of them stillaggregate. Fig. 2c clearly shows the morphology of the magneticFe3O4/poly(St-BA-METAC) particles. On the whole, the particlesare spherical in shape and the polymers become invisible in corebecause the iron oxide particles absorb the electron beam andappear as dark spots within the latex. At high magnifications, thedetailed structure of the iron oxide within the latex becomesclearer, which is shown in Fig. 2d, with black aggregates ofmagnetic nanoparticles in the core and lighter spots on the edgessignifying St-BA-METAC copolymers. It is evident that the encap-sulation of magnetite particles has been obtained successfully.

Compared to the St-BA-METAC particles in Fig. 2b, the size ofthe composite particles (Fig. 2c) are larger than the size of St-BA-METAC particles and the shell of the composite particles are not

Fig. 1. FT-IR spectra of (a) acid-modified magnetic nanoparticles (b) Fe3O4/

poly(St-BA-METAC) nanoparticles.

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–14181412

clean and regular as traditional emulsifier-free particles. It maycontribute to the effects of magnetic nanoparticles during thepolymerization.The mechanism of magnetite-containing emulsi-fier-free polymerization was proposed in Scheme 1. In thismechanism, when a dose of AIBA was added to the reaction, themonomers were initiated to form oligomeric radical cations. Theactive species carrying radicals undergoes further propagation. Asthe hydrophobic monomers and cationic monomers reacted withthe active species, the length of hydrophobic backbone becamelonger and electric density of the chain increased, and then theembryos formed. Meanwhile, oleic acid modified magnetic nano-particles would effectively embed into the embryos because theiron oxide particles coated with oleic acid possess very weak

negative charge and hydrophobic surface. After high molecularweight of chains had been obtain, the hydrophilic and hydrophobicmoieties of the chains could nucleate to form micelles whichserved as a locus of polymerization for the emulsion polymeriza-tion. Significant coagulation of magnetite particles also occurreddue to their nano-size effects, so magnetite particles acceleratedthe formation of micelles and the size of the micelles are largerthan that of emulsifier-free polymerization without magnetiteparticles. During the polymerization, the residual magnetite parti-cles and monomers incorporated into micelles along with growingof polymer particles. However, the inherent steric hindrance of themagnetite particles impeded transportation of monomers, causingirregular thickness of polymeric shell on the surface of the particle.

Fig. 2. TEM micrographs of (a) acid-modified magnetic nanoparticles, (b) poly(St-BA-METAC) particles, (c) magnetic Fe3O4/poly(St-BA-METAC) particles, (d) higher-

magnification image of magnetic Fe3O4/poly(St-BA-METAC) particles.

Scheme1. The proposed mechanism of composite particle nucleation and growth.

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–1418 1413

4.1.3. Magnetic properties

Fig. 3 shows a typical magnetization curve of modified magnetitenanoparticles and polymer latexes encapsulated magnetite particlesmeasured by VSM at room temperature. The results of VSM analysisshow the saturation magnetization of the acid-modified magneticnanoparticles and Fe3O4/poly(St-BA-METAC) particles are approxi-mately 33.78 emu/g and 7.69 emu/g, respectively. The loss ofmagnetization compared to bulk magnetite, which is 84 emu/g,may be due to the oxidation processes and a little crystal destroyduring modification and polymerization. Also the shell of polymerreduces the total magnetization. The ratio of coercivity to negligibleremnant magnetization (Mr/Ms¼1.6%) indicates that the compositeparticles exhibit a superparamagnetic behavior.

4.1.4. XRD characterization

The X-ray diffraction analysis (XRD) patterns of (a) modifiedmagnetic Fe3O4 nanoparticles, (b) magnetic Fe3O4/poly(St-BA-METAC) particles are shown in Fig. 4. The XRD pattern of Fe3O4

and magnetic Fe3O4/poly(St-BA-METAC) show main peaks at 2y of301, 351, 431, 571, 621, 741corresponding to (2 2 0), (3 1 1), (4 0 0),(5 1 1), (4 4 0), (5 3 3) phase of the face-centered cubic(fcc) Fe3O4

crystal structure. It is clearly seen from Fig. 4b that their peaks areidentical although the broad diffusion pattern in the range of low 2ybecause of the existence of amorphous state of copolymers inthe magnetic composite particles, indicating that the structure ofmagnetic nanoparticles keep essentially unchanged during thepolymerization and a portion of the Fe3O4 diffused into thecomposite latexes.

Magnetic particle size is an important factor that determinesthe superparamagentic behavior of the magnetite. The particleswith the size less than 15 nm may exhibit superparamagneticbehavior [45]. The average diameter of the crystal was calculatedaccording to the Scherrer equation: D¼k l/b cosy. Here, the X-raywavelength l is 0.154 nm; k is the shape factor, which is assigneda value of 0.89; D is the average diameter of the crystals, y is theBragg angle in degrees and b is full-width at half-height of themain peaks. The calculated result was approximately 8 nm basedon the widths of the main peak at 2y¼351, which was consistentwith the results of TEM and VSM analysis.

4.1.5. Thermogravimetric analysis

In general, TGA was used to measure the magnetite content incomposite particles. Fig. 5 shows weight loss curves of (a) oleic acidmodified magnetic particles, (b) Fe3O4/poly(St-BA-METAC) particles,(c) poly(St-BA-METAC) particles obtained from a thermogravimetric(TG) analyzer. In Fig. 5a, the weight loss in the range of 240–370 1Cis related to the degradation of oleic acid on Fe3O4 particles. Theamount of residue is 91 wt% at 650 1C, proving the high content ofmagnetite after oleic acid encapsulation. The TGA curve of thecomposite particles (Fig. 5b) shows that the weight loss stage from210 1C to 350 1C, which dues to decomposition of small organicmolecules. It could be observed that the weight loss is initiated atlower temperature than that of oleic acid-modified magneticparticles, which illustrates magnetite catalyzes the decompositionof the small molecules. When the temperature reaches up to 350 1C,rapid weight loss occurs responding to the decomposition ofpolymer shell. After the temperature rising to 650 1C, the weightof hybrid particles and pure polymers keep a constant content about20.8 wt% and 1.6 wt% (Fig. 5c), respectively. Calculation results showthat the magnetic content of composite particles is about 19.2 wt%.

4.1.6. Zeta potential characterization

To be non-viral gene carrier, the emulsion should have a gooddispersive stability in aqueous media and contain positively charged

Fig. 3. Magnetic hysteresis loop of (a) acid-modified magnetic nanoparticles

(b) magnetic Fe3O4/poly(St-BA-METAC) particles.

Fig. 4. XRD spectrum of (a) oleic acid-modified magnetic particles, (b) Fe3O4/

poly(St-BA-METAC) particles.

Fig. 5. TG curves of (a) oleic acid-modified magnetic particles, (b) Fe3O4/poly(St-

BA-METAC) particles, (c) poly(St-BA-METAC) particles.

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–14181414

surface which can electrostatically adsorb negatively charged nucleicacid. The zeta potential characterization of the vector’s surface chargeprovides indication of the long-term stability. Cationic particles withzeta potential more than þ30 mV are normally considered stable.Therefore, the zeta potential of the Fe3O4/poly(St-BA-METAC) emul-sion at pH value of 7.3 was measured first. The zeta potential of thepure Fe3O4/poly(St-BA-METAC) emulsion is 45.2 mV, which indicatesthe good stability of the emulsion. Fig. 6 shows the zeta potentialchanged of Fe3O4/poly(St-BA-METAC) emulsion at different ranges ofpH values (from 2 to 12) and the density of positive charge of Fe3O4/poly(St-BA-METAC) emulsion decreases with the pH increasing. Itreveals that Fe3O4/poly(St-BA-METAC) emulsion has highly positivesurface charge in a wide range of pH value, which indicates emulsionhas a good ability to bind DNA in a wide range.

4.2. Characterization of emulsion/DNA particles

4.2.1. Zeta potential of emulsion/DNA

To confirm the binding ability of Fe3O4/poly(St-BA-METAC)emulsion with DNA, zeta potentials of emulsion/DNA in differentconcentration (pH¼7.3) were recorded as shown in Fig. 7. Asdemonstrated in Fig. 7, the zeta potentials of original emulsionand the pure ct DNA are 45.2 mV and –76.7 mV, respectively. Thezeta values decreased with increasing concentration of ct DNA,account for the adsorption of the negatively charged DNA onto thesurface of particles. When the ratio of emulsion/DNA reduces to4.5:1(wt/wt), the zeta potential of the complexes turns to negativecharge. The zeta values becomes more negative with further decreas-ing concentration of emulsion. It may be the reason that excessnegatively charged DNA chains loaded on the surface of the emulsion.

4.2.2. Particle size measurement

The hydrodynamic particle size (Dh) and polydispersity indexof emulsion and emulsion/DNA complexes with different ratioswere also measured on Malvern Nano-ZS instrument. As shown inTable 1, the average particle size of the pure magnetic cationicemulsion was 316.9 nm. The particle size was increasing with themore DNA was added, which also suggests the absorption of DNAonto the surface of particles. Besides, the polydispersity index ofthe complex still kept less than 0.2, which indicates that thecomplexes kept well homogeneous size and stability.

4.2.3. UV–vis spectroscopy

Usually, the UV–vis spectra are employed to determine theinteraction between molecules and DNA. To confirm the interac-tion between magnetic emulsion with DNA, the magnetic emul-sion was tested first. As shown in Fig. 8a, there is no absorption at260 nm for magnetic emulsion. When DNA was added, there is anew absorption at 260 nm, which is the specific absorption ofDNA in solution. Fig. 8b displays the UV spectral change of DNA at260 nm when the emulsion was added. With the addition of theemulsion, the light absorption of DNA obviously increases. Thereare three kinds of binding modes between the molecules andDNA: intercalative mode, static electronic mode and groovebinding [46]. The composite particle contains cationic surface insolution, which is easy to adsorb negatively charged phosphategroup of DNA onto its surface. The hyperchromic effect mayattributes to the electrostatic force cause aggregation of nucleicacid on the composite particles.

4.2.4. Fluorescence spectroscopy of DNA/emulsion complexes

The fluorescence spectra of magnetic emulsion in differentconcentrations of DNA are shown in Fig. 9. It can be seen fromFig. 9a that the emulsion has fluorescence emission peaks at362.5 nm and 471 nm after being excited with a wavelength of270 nm. The fluorescence emission peak at 362.5 nm is stronger.The fluorescence may come from the chromophore of the phenylring on each repeat unit of polystyrene [47,48]. When a fixedconcentration of emulsion was titrated with increasing amounts ofDNA, it is observed that a remarkable intrinsic fluorescence ofemulsion is decreased. Furthermore, there is a slight blue shift at

Fig. 6. Effect of pH values on the zeta potentials of emulsion.

Fig. 7. Zeta potentials of emulsion/ct DNA complexes with different weight ratios

(ct DNA concentration was kept at 64 mg/L): (a) pure emulsion (b) 25:1 (c) 12.5:1,

(d) 5:1, (e) 4.5:1, (f) 4:1, (g) 3:1, (h) 2.5:1, (i) pure ct DNA.

Table 1Hydrodynamic diameters and polydispersity of emulsion and emulsion/DNA with different ratios (v/v).a

Emulsion/DNA ratio (v/v) Pure emulsion 2:1 1:1 1:2 1:3 1:4 1:5

Particle size (nm) 316.9 340.6 342.5 380 406 561.1 853

Polydispersity index 0.065 0.117 0.137 0.153 0.15 0.195 0.108

a The working solutions of DNA and emulsion was kept at 64 mg/L and 1.6 mg/mL, respectively.

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–1418 1415

the maximum wavelength of emulsion fluorescence emission whenthe solution of DNA was added. The results indicate that DNA couldquench the intrinsic fluorescence of magnetic emulsion and theinteraction between emulsion and DNA indeed exists.

Fluorescence quenching refers to a process in which the fluores-cence intensity from a sample diminishes. As for the quenchingeffect of DNA on the magnetic emulsion, the intensity reduction isdescribed by the well-known Stern–Volmer equation [49]:

F0F

¼ 1þKsv Q½ � ¼ 1þkqt0 Q½ � ð1Þ

where F0 and F are the steady-state fluorescence intensities in theabsence and presence of quencher (DNA), Ksv. ithe Stern–Volmerquenching constant, ½Q�. the quencher concentration, kq. the bimo-lecular quenching constant,t0. the lifetime of the average fluoro-phore in the absence of quencher which is 10�8 s [50]. Fig. 9billustrates that the results are well with the Stern–Volmer equationand the value of kq. 4.012�1012 L s�1 mol�1. The value of kq. fargreater than 2.0�1010 L s�1 mol�1, the maximum diffusion colli-sion quenching rate constant of various quenchers with the biopo-lymer [51], which indicates that the quenching mechanism ofemulsion by DNA is most likely to be a static quenching procedure.

4.2.5. Gel electrophoresis of the DNA/emulsion complexes

To further confirm the binding interaction between the emulsionand DNA molecules, the DNA retardation assay by agarose gelelectrophoresis is conducted. As shown in Fig. 10, the pure DNAindicates maximum migration and all DNA molecules move towardsthe anode (lane 1). When the increased amounts of the compositeemulsion (1.2–9.6 mg) were added to a fixed amount of plasmid DNA(0.48 mg), the amount of plasmid DNA is retained more at the anode,which contributes to more DNAwas neutralized and adsorbed by thecationic composite emulsion. The capacity of retarding DNA indicatesthat the composite emulsions can efficiently bind to DNA.

5. Conclusion

Magnetic cationic Fe3O4/poly(St-BA-METAC) emulsion wassuccessfully synthesized by emulsifier-free polymerization. Theobtained particles have high positive charged surface and are

Fig. 8. UV–vis spectra of magnetic emulsion/DNA complex. (a) Magnetic emulsion:

0.286mg/L; DNA (1–4): 0, 0.86 mg/L, 1.72 mg/L, 2.58 mg/L; (b) DNA: 8.6 mg/L;

magnetic emulsion: (1–7) 0, 0.14 mg/L, 0.4 mg/L, 0.572 mg/L, 1.114mg/L, 1.686mg/L.

Fig. 9. (a) Fluorescence spectra of magnetic emulsion in the presence of DNA at

different concentrations (pH¼7.4, T¼298 K, lex¼270 nm). c(magnetic emulsion¼0.32 g/L); c(DNA)/(�10�7 mol/L), (1–5): 0, 8.13, 16.26, 24.39, 32.52, respectively.

(b) The Stern–Volmer plots for the quenching of magnetic emulsion by DNA

(pH¼7.4, lex¼270 nm, lem¼362.5 nm).

Fig. 10. Agarose gel electrophoresis retardation assay: lane 1, naked DNA (2.5 kbp,

0.48 mg); lanes 2–5: 0.48 mg DNA complexed with 1.2 mg, 2.4 mg, 4.8 mg, 9.6 mg of

emulsion.

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–14181416

stable in a wide range of pH values. The results of zeta potential,UV spectra and fluorescence spectroscopy, agarose electrophor-esis indicate that the emulsion could interact with DNA in anelectrostatic bonding mode. These results suggest the potential ofmagnetic cationic emulsion as target gene carrier.

Acknowledgments

The authors are grateful to the support from SpecializedResearch Fund for the Doctoral Program of Higher Education ofChina (No. 20094208110002).

References

[1] Q.-Q. Gai, F. Qu, Z.-J. Liu, R.-J. Dai, Y.-K. Zhang, Superparamagnetic lysozymesurface-imprinted polymer prepared by atom transfer radical polymerizationand its application for protein separation, Journal of Chromatography A 1217(2010) 5035–5042.

[2] D.F.M. Neri, V.M. Balc~ao, F.O.Q. Dourado, J.M.B. Oliveira, L.B. Carvalho Jr,J.A. Teixeira, Galactooligosaccharides production by [beta]-galactosidaseimmobilized onto magnetic polysiloxane–polyaniline particles, Reactive andFunctional Polymers. 69 (2009) 246–251.

[3] W. Zheng, F. Gao, H. Gu, Carboxylated magnetic polymer nanolatexes:preparation, characterization and biomedical applications, Journal of Mag-netism and Magnetic Materials 293 (2005) 199–205.

[4] V. Zavisova, M. Koneracka, M. Muckova, P. Kopcansky, N. Tomasovicova,G. Lancz, M. Timko, B. Patoprsta, P. Bartos, M. Fabian, Synthesis andcharacterization of polymeric nanospheres loaded with the anticancer drugpaclitaxel and magnetic particles, Journal of Magnetism and MagneticMaterials 321 (2009) 1613–1616.

[5] S. Purushotham, R.V. Ramanujan, Thermoresponsive magnetic compositenanomaterials for multimodal cancer therapy, Acta Biomaterialia 6 (2010)502–510.

[6] K. Cai, Z. Luo, Y. Hu, X. Chen, Y. Liao, L. Yang, L. Deng, Magnetically triggeredreversible controlled drug delivery from microfabricated polymeric multi-reservoir devices, Advanced Materials 21 (2009) 4045–4049.

[7] C. Wilhelm, F. Gazeau, Universal cell labelling with anionic magneticnanoparticles, Biomaterials 29 (2008) 3161–3174.

[8] R.Y. Hong, B. Feng, L.L. Chen, G.H. Liu, H.Z. Li, Y. Zheng, D.G. Wei, Synthesis,characterization and MRI application of dextran-coated Fe3O4 magneticnanoparticles, Biochemical Engineering Journal 42 (2008) 290–300.

[9] H. Yang, J. Zhang, Q. Tian, H. Hu, Y. Fang, H. Wu, S. Yang, One-pot synthesis ofamphiphilic superparamagnetic FePt nanoparticles and magnetic resonanceimaging in vitro, Journal of Magnetism and Magnetic Materials 322 (2010)973–977.

[10] K.L. Ang, S. Venkatraman, R.V. Ramanujan, Magnetic PNIPA hydrogels forhyperthermia applications in cancer therapy, Materials Science and Engi-neering C 27 (2007) 347–351.

[11] J. Hong, D. Xu, P. Gong, H. Ma, L. Dong, S. Yao, Conjugation of enzyme onsuperparamagnetic nanogels covered with carboxyl groups, Journal ofChromatography B 850 (2007) 499–506.

[12] X. Zhang, X. Tang, X. Huang, Synthesis of novel magnetic phosphazene-containing polymer microspheres with active hydroxyl groups, Journal ofMagnetism and Magnetic Materials 321 (2009) 2966–2970.

[13] Y. Bai, B. Teng, S. Chen, Y. Chang, Z. Li, Preparation of magnetite nanoparticlescoated with an amphiphilic block copolymer: a potential drug carrier with acore–shell–corona structure for hydrophobic drug delivery, MacromolecularRapid Communications 27 (2006) 2107–2112.

[14] A. Tocchio, D. Horak, M. Babic, M. Trchova, M. Veverka, M.J. Benes, M. Slouf,A. Fojtık, Magnetic poly(glycidyl methacrylate) particles prepared in thepresence of surface-modified g-Fe2O3, Journal of Polymer Science Part A:Polymer Chemistry. 47 (2009) 4982–4994.

[15] Y. Xu, H. Xu, H. Gu, Controllable preparation of epoxy-functionalizedmagnetic polymer latexes with different morphologies by modified mini-emulsion polymerization, Journal of Polymer Science Part A: PolymerChemistry. 48 (2010) 2284–2293.

[16] J. Gunn, H. Wallen, O. Veiseh, C. Sun, C. Fang, J. Cao, C. Yee, M. Zhang,A. Multimodal Targeting, Nanoparticle for selectively labeling T cells, Small 4(2008) 712–715.

[17] R. Veyret, T. Delair, C. Pichot, A. Elaissari, Amino-containing magneticnanoemulsions: elaboration and nucleic acid extraction, Journal of Magnet-ism and Magnetic Materials 295 (2005) 155–163.

[18] B. Steitz, H. Hofmann, S.W. Kamau, P.O. Hassa, M.O. Hottiger, B. vonRechenberg, M. Hofmann-Amtenbrink, A. Petri-Fink, Characterization ofPEI-coated superparamagnetic iron oxide nanoparticles for transfection: sizedistribution, colloidal properties and DNA interaction, Journal of Magnetismand Magnetic Materials 311 (2007) 300–305.

[19] S.M. Kwon, H.Y. Nam, T. Nam, K. Park, S. Lee, K. Kim, I.C. Kwon, J. Kim,D. Kang, J.H. Park, S.Y. Jeong, In vivo time-dependent gene expression of

cationic lipid-based emulsion as a stable and biocompatible non-viral genecarrier, Journal of Controlled Release 128 (2008) 89–97.

[20] A. Schallon, V. Jerome, A. Walther, C.V. Synatschke, A.H.E. Muller, R. Freitag,Performance of three PDMAEMA-based polycation architectures as genedelivery agents in comparison to linear and branched PEI, Reactive andFunctional Polymers. 70 (2010) 1–10.

[21] Q.-H. Xu, B.S. Gaylord, S. Wang, G.C. Bazan, D. Moses, A.J. Heeger, Time-resolved energy transfer in DNA sequence detection using water-solubleconjugated polymers: the role of electrostatic and hydrophobic interactions,Proceedings of the National Academy of Sciences of the United States ofAmerica 101 (2004) 11634–11639.

[22] J. Dai, S. Zou, Y. Pei, D. Cheng, H. Ai, X. Shuai, Polyethylenimine-graftedcopolymer of poly(l-lysine) and poly(ethylene glycol) for gene delivery,Biomaterials 32 (2011) 1694–1705.

[23] X. Wang, J. Sun, J. Ji, pH modulated layer-by-layer assembly as a newapproach to tunable formulating of DNA within multilayer coating, Reactiveand Functional Polymers 71 (2011) 254–260.

[24] H. Cheng, J.-L. Zhu, X. Zeng, Y. Jing, X.-Z. Zhang, R.-X. Zhuo, Targeted genedelivery mediated by Folate-polyethylenimine-block-poly(ethylene glycol)with receptor selectivity, Bioconjugate Chemistry 20 (2009) 481–487.

[25] T.-i. Kim, S.W. Kim, Bioreducible polymers for gene delivery, Reactive andFunctional Polymers 71 (2011) 344–349.

[26] S. Mao, W. Sun, T. Kissel, Chitosan-based formulations for delivery of DNAand siRNA, Advanced Drug Delivery Review 62 (2010) 12–27.

[27] L. Liu, A. Shi, S. Guo, Y.e. Fang, S. Chen, J. Li, Preparation of chitosan-g-polylactide graft copolymers via self-catalysis of phthaloylchitosan and theircomplexation with DNA, Reactive and Functional Polymers. 70 (2010)301–305.

[28] T. Andersson, S. Holappa, V. Aseyev, H. Tenhu, Effect of polycation length onits complexation with DNA and with poly(oxyethylene-block-sodium metha-crylate), Biomacromolecules 7 (2006) 3229–3238.

[29] S.K. Wiedmer, T. Andersson, M. Sundermann, M.-L. Riekkola, H. Tenhu,Cationic poly(methacryl oxyethyl trimethylammonium) and its poly(ethy-lene glycol)-grafted analogue as capillary coating materials in electrophor-esis, Journal of Polymer Science Part B: Polymer Physics 45 (2007)2655–2663.

[30] H.P. Song, J.Y. Yang, S.L. Lo, Y. Wang, W.M. Fan, X.S. Tang, J.M. Xue, S. Wang,Gene transfer using self-assembled ternary complexes of cationic magneticnanoparticles, plasmid DNA and cell-penetrating Tat peptide, Biomaterials 31(2010) 769–778.

[31] S. Seino, Y. Matsuoka, T. Kinoshita, T. Nakagawa, T.A. Yamamoto, Dispersi-bility improvement of gold/iron-oxide composite nanoparticles by polyethy-lenimine modification, Journal of Magnetism and Magnetic Materials 321(2009) 1404–1407.

[32] O. Mykhaylyk, D. Vlaskou, N. Tresilwised, P. Pithayanukul, W. Moller,C. Plank, Magnetic nanoparticle formulations for DNA and siRNA delivery,Journal of Magnetism and Magnetic Materials 311 (2007) 275–281.

[33] K.H. Bae, K. Lee, C. Kim, T.G. Park, Surface functionalized hollow manganeseoxide nanoparticles for cancer targeted siRNA delivery and magnetic reso-nance imaging, Biomaterials 32 (2011) 176–184.

[34] A. Petri-Fink, B. Steitz, A. Finka, J. Salaklang, H. Hofmann, Effect of cell mediaon polymer coated superparamagnetic iron oxide nanoparticles (SPIONs):colloidal stability, cytotoxicity, and cellular uptake studies, Eur. J. Pharm.Biopharm. 68 (2008) 129–137.

[35] C. Wang, J. Yan, X. Cui, D. Cong, H. Wang, Preparation and characterization ofmagnetic hollow PMMA nanospheres via in situ emulsion polymerization,Colloids and Surfaces A: Physicochemical and Engineering Aspects 363(2010) 71–77.

[36] E. Pollert, K. Knızek, M. Marysko, K. Zaveta, A. Lancok, J. Bohacek, D. Horak,M. Babic, Magnetic poly(glycidyl methacrylate) microspheres containingmaghemite prepared by emulsion polymerization, Journal of Magnetismand Magnetic Materials 306 (2006) 241–247.

[37] S. Lu, R. Qu, J. Forcada, Preparation of magnetic polymeric compositenanoparticles by seeded emulsion polymerization, Materials Letters 63(2009) 770–772.

[38] H. Liu, J. Guo, L. Jin, W. Yang, C. Wang, Fabrication and functionalization ofdendritic poly(amidoamine)-immobilized magnetic polymer compositemicrospheres, Journal of Physical Chemistry B 112 (2008) 3315–3321.

[39] S. C- akmak, M. Gumusderelioglu, A. Denizli, Biofunctionalization of magneticpoly(glycidyl methacrylate) microspheres with protein A: characterizationand cellular interactions, Reactive and Functional Polymers. 69 (2009)586–593.

[40] P.-C. Wang, W.-Y. Chiu, C.-F. Lee, T.-H. Young, Synthesis of iron oxide/poly(methyl methacrylate) composite latex particles: nucleation mechanismand morphology, Journal of Polymer Science Part A: Polymer Chemistry. 42(2004) 5695–5705.

[41] I. Csetneki, M.K. Faix, A. Szilagyi, A.L. Kovacs, Z. Nemeth, M. Zrinyi, Prepara-tion of magnetic polystyrene latex via the miniemulsion polymerizationtechnique, Journal of Polymer Science Part A: Polymer Chemistry 42 (2004)4802–4808.

[42] G. Xie, Q. Zhang, Z. Luo, M. Wu, T. Li, Preparation and characterization ofmonodisperse magnetic poly(styrene butyl acrylate methacrylic acid) micro-spheres in the presence of a polar solvent, Journal of Applied Polymer Science87 (2003) 1733–1738.

[43] T. Andersson, S. Holappa, V. Aseyev, H. Tenhu, Complexation of linear andpoly(ethylene oxide)-grafted poly(methacryl oxyethyl trimethylammonium

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–1418 1417

chloride) with poly(ethylene oxide-block-sodium methacrylate), Journal ofPolymer Science Part A: Polymer Chemistry. 41 (2003) 1904–1914.

[44] J. Huang, H. Pen, Z. Xu, C. Yi, Magnetic Fe3O4/poly(styrene-co-acrylamide)composite nanoparticles prepared by microwave-assisted emulsion poly-merization, Reactive and Functional Polymers. 68 (2008) 332–339.

[45] A.K. Gupta, M. Gupta, Synthesis and surface engineering of iron oxide nano-particles for biomedical applications, Biomaterials 26 (2005) 3995–4021.

[46] H. Hussain, K. Busse, J. Kressler, Poly(ethylene oxide)- and poly(perfluor-ohexylethyl methacrylate)-containing amphiphilic block copolymers: asso-

ciation properties in aqueous solution, Macromolecular Chemistry andPhysics 204 (2003) 936–946.

[47] R.W. Sandoval, D.E. Williams, J. Kim, C.B. Roth, J.M. Torkelson, Critical micelleconcentrations of block and gradient copolymers in homopolymer: effects of

sequence distribution, composition, and molecular weight, Journal of Poly-mer Science Part B: Polymer Physics 46 (2008) 2672–2682.

[48] M.K. Mundra, C.J. Ellison, P. Rittigstein, J.M. Torkelson, Fluorescence studiesof confinement in polymer films and nanocomposites: glass transitiontemperature, plasticizer effects, and sensitivity to stress relaxation and localpolarity, The European Physical Journal - Special Topics 141 (2007) 143–151.

[49] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Springer, New York,2006.

[50] S. Bi, H. Zhang, C. Qiao, Y. Sun, C. Liu, Studies of interaction of emodin andDNA in the presence of ethidium bromide by spectroscopic method, Spectro-chimica Acta Part A 69 (2008) 123–129.

[51] W.R. Ware, Oxygen quenching of fluorescence in solution: an experimentalstudy of the diffusion process, Journal of Physical Chemistry A 66 (1962)455–458.

X. Li et al. / Journal of Magnetism and Magnetic Materials 324 (2012) 1410–14181418