This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. B

Cite this:DOI: 10.1039/c7tb02361h

Magnetically responsive biopolymeric multilayerfilms for local hyperthermia†

M. Criado,a B. Sanz,bc G. F. Goya, cd C. Mijangosa and R. Hernandez *a

We present a proof of concept on the use of thermomagnetic polymer films (TMFs) as heating devices

for magnetic hyperthermia in vitro. The TMFs were prepared through spray assisted layer-by-

layer assembly of polysaccharides and magnetic iron oxide nanoparticles, yielding an alternate

magnetic-polymer multilayer structure. By applying a remote alternating magnetic field (AMF)

(f = 180 kHz; H = 35 kA m�1) we increased the temperature of the TMFs in a thickness-dependent

way, up to 12 1C within the first 5 minutes. To test our films as heating substrates for magnetic

hyperthermia, a series of in vitro experiments were designed using human neuroblastoma SH-SY5Y

cells, known by their tolerance to thermal stress. The application of two AMF cycles (30 minutes each)

showed that the exogenous magnetic hyperthermia resulted in an 85% reduction of cell viability. This

capacity of the TMFs of hyperthermia-mediated cell killing using a remote AMF opens new options for

the treatment of small and superficial tumor lesions by means of remotely-triggered magnetic

hyperthermia.

1 Introduction

The design of stimuli-responsive materials prepared throughthe incorporation of nanoparticles (NPs) into multilayer polymerfilms has gained increasing attention for the development ofadvanced functional materials employed as optical and humiditysensors,1,2 nanoactuators,3 antibacterial4 or flame retardancymaterials5 and flexible energy and electronic devices6,7 amongother applications. Layer-by-layer (LbL) assembly, which usescomplementary interactions between components to depositmaterials one layer at a time, constitutes one of the mostversatile methodologies for the production of stratified poly-electrolyte films and the incorporation of nanoparticles intothese films.8,9 Specifically, magnetic nanoparticles (NPs) providepolymer films with the ability to respond to a magnetic field,10–12

and this magnetic responsiveness has been already used inbiomedicine for controlled drug release,13 tissue engineering14,15

and magnetic hyperthermia (MHT) therapy.16

The basis of MHT is related to the rise the temperature oftarget tissues or cells up to values E43–46 1C, triggering theapoptotic response of cells. The increase of the temperature

is achieved by using magnetic nanoparticles used as heatingagents under the action of an external AMF.17–20 The specificpower absorption (SPA) is the power absorbed by unit mass ofNPs, which is dissipated and heats the surrounding medium.There is a large number of works on NPs based on iron oxides,mostly Fe3O4 (magnetite) and g-Fe2O3 (maghemite), reportingthat the SPA values of these materials are large enough to heatmacroscopic tumors. The generalized use of iron oxides ismainly due to the lower toxicity of this phases when comparedto other magnetic heating materials such as Co, Ni and theiroxides.21 In local hyperthermia, heat is applied to small andsuperficial tumor areas with the aim of reaching temperaturesup to 42 1C for one hour within the cancer tumor to cause celldeath. Polymer films with embedded gold nanoparticles havebeen successfully employed for tissue laser ablation whenplaced in contact with the tissue to be treated inducing localexogenous hyperthermia. While the photothermal therapyoffers the possibility to control the energy dose delivered tothe target area by both the intensity of the laser beam and theconcentration of NPs, this technique is limited by the shortpenetration depth of the NIR wavelength into the human body(E0.3 cm in breast tissue) even at the known ‘optical window’of water.22

The exogenous heating strategy using magnetic fields thatwe report in this work can deliver energy density values to thetarget area that are lower than the typical values produced byphotothermal therapy. However, as biological tissues essentiallydo not interact with ac magnetic fields of 100–500 kHz, the energycan be delivered essentially into any depth of the human body.

a Instituto de Ciencia y Tecnologıa de Polımeros (ICTP-CSIC), c/Juan de la Cierva,

3, 28006 Madrid, Spain. E-mail: rhernandez@ictp.csic.esb Nanoscience Institute of Aragon, University of Zaragoza, Mariano Esquillor s/n,

Zaragoza, 50018, Spainc nB Nanoscale Biomagnetics S.L., C/Panama 2, Local 1, Zaragoza, 50012, Spaind Department of Condensed Matter Physics, University of Zaragoza,

Pedro Cerbuna 12, Zaragoza, 50009, Spain

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7tb02361h

Received 3rd September 2017,Accepted 16th October 2017

DOI: 10.1039/c7tb02361h

rsc.li/materials-b

Journal ofMaterials Chemistry B

PAPER

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article OnlineView Journal

J. Mater. Chem. B This journal is©The Royal Society of Chemistry 2017

Clinical cases where local hyperthermia is a viable alternativeinclude chest wall recurrences, superficial malignant melanomalesions and head/neck tumors.23,24 In these cases, the applicationof the remote magnetic fields is achieved by adapted superficialapplicators of different shapes and kinds wired to the powergenerators of radiofrequency generators, microwave, ultrasoundor near infrared radiation (NIR).25–27 To the best of our knowl-edge, the potential employment of thermomagnetic films whichcould be more adapted to subcutaneous or deeper applicationshas not been proposed before.

In this study, we evaluate the thermal effect and show aproof of concept at the in vitro level of the application in localmagnetic hyperthermia treatment of thermomagnetic polymerfilms constituted of alginate, chitosan and a lab-made magneticferrofluid obtained through spray-assisted LbL from the corres-ponding aqueous polyelectrolyte solutions.28 Spray assisted LbLconstitutes an attractive alternative to dipping or spin-coating LbLbecause it is much faster and easier to adapt at an industriallevel.12 Human neuroblastoma SH-SY5Y cells were selected forproving the proposed application in local magnetic hyperthermiaof the TMFs as a model of human malignant metastatic neuro-blastoma which exhibits a notable resistance to hyperthermicstress.29

2 Experimental section2.1 Materials

Chitosan (Chi) of low molecular weight was supplied by Aldrich(448869, lot SLBG1673V). According to the fabricant, viscositywas 20–300 cps (1 wt% in acetic acid, 25 1C, Brookfield). Thischitosan was purified using the procedure described by Signiniand CampanaFilho.30 The deacetylation degree (DD) was deter-mined by RMN-1H at 70 1C using 2 wt% CD3COOD/D2O assolvent and it was 81%.31 Molecular weight (Mw) determinedby capillary viscosimetry at 25 1C using as solvent acetic acid0.3 M per sodium acetate 0.2 M and applying the equation ofMark–Houwink (k = 74 � 10�5 dL g�1, a = 0.76)32 was 67 000 Da.

Sodium alginate (Alg) was supplied by Sigma-Aldrich (A2158,lot 090M0092V). According to the fabricant, viscosity was136 cps (2% w/v in water at 25 1C). Molecular weight (Mw)determined by capillary viscosimetry at 25 1C using as solventsodium chloride 0.1 M and applying the equation of Mark–Houwink (k = 2 � 10�5 dL g�1, a = 1.0)33 was 166 000 Da.

Poly(ethylenimine) (PEI) with a molecular weight (Mw) of25 000, acetic acid, were supplied by Aldrich and used asreceived. Sodium acetate anhydrous was supplied by Panreacand chloride acid by VWR.

Live/dead viability/cytotoxicity kit, for mammalian cells(Molecular Probes, ThermoFisher scientific, L3224). The kitcontains calcein AM (4 mM in anhydrous DMSO) and ethidiumhomodimer-1 (2 mM in DMSO/H2O 1 : 4 (v/v)).

2.2 Preparation and characterization of lab-made ferrofluid

A lab-made aqueous ferrofluid was synthesized by a coprecipitationmethod or iron salts carried out in a polymer aqueous solution.

Briefly, 0.834 g of iron(II) sulfate heptahydrate (FeSO4�7H2O)and 16.218 g of iron(III) chloride hexahydrate (FeCl3�6H2O) weredissolved in 10 mL of Milli-Q water under N2 atmosphere.Then, the mixture was added to a three-necked flask bubbledcontaining 40 mL of alginate (2.5 mg mL�1) under N2 atmo-sphere and mechanical stirring at 350 rpm using a teflon stirrerand heated at 80 1C for 1 hour. After that, this mixture wasadded drop wise under constant stirring to other three-neckedflask bubbled containing 50 mL of ammonium hydroxide(NH4OH), taking place the color change of the mixture fromyellow orange to black. The mixture was held for 2 hour at 80 1Cunder stirring and N2 atmosphere. Afterwards, the solution wascooled in an ice bath and washed with Milli-Q water until neutralpH giving rise to an alginate based magnetite ferrofluid.34

The determination of iron content of the ferrofluid wascarried out through UV-Vis transmission spectrophotometryusing the thiocyanate complexation reaction and measuringthe absorbance of the iron–thiocyanate complex at 478 nmwavelength.

Fe(aq)3+ + 6SCN(aq)

� - [Fe(SCN)6](aq)3�

Ferrofluid was dissolved in HCl 6 M/HNO3 (65%) for 2 h. Afterthat, potassium thiocyanate was added to the solution to formthe iron–thiocyanate complex.

The ferrofluid is constituted by magnetite (Fe3O4) NPsdispersed in a sodium alginate solution at a concentration of8 mg mL�1 as determined by UV-Vis spectrophotometry with anaverage size of 7.3 � 1.1 nm as determined by transmissionelectron microscopy.34

The heating capacity of the as prepared ferrofluid underalternating magnetic fields (AMF) was measured using a com-mercial magnetic field applicator (DM-100, nB NanoscaleBiomagnetics, Spain) at frequencies 250 o f o 820 kHz andfield amplitudes 0 r H r 24 kA m�1. Calorimetric measure-ments were carried out at thermal insulated conditions using avacuum pump (10�7 mbar). A fibre-optic thermometer integratedinto this equipment allowed to measure the temperature as afunction of time.

In order to evaluate the heating mechanism of NPs immo-bilized in gelatin gels, 500 mL of ferrofluid with a concentrationof 8.0 mg mL�1 were precipitated at 14 500 rpm for 15 min,then the supernatant was removed and NPs were dispersed in1 mL of 1% w/v or 2% w/v gelatin in a vortex for 1 min and keptat �4 1C for 1 hour to form the gel. After that, SPA wasmeasured in the commercial equipment DM-100 (NanoScaleBiomagnetics) at a frequency of 571 kHz and a magnetic fieldamplitude of 24 kA m�1.

2.3 Preparation and characterization of thermomagneticpolymer films

All films were built on glass slides (12 mm +) previouslycleaned with ethanol and rinsed extensively with water.As the glass substrate is negatively charged, a layer of PEI(1 mg mL�1) was deposited as a first layer to get a homogenoussubstrate positively charged by dipping the substrate in asolution of PEI for 5 minutes and then in a solution of distilled

Paper Journal of Materials Chemistry B

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. B

water for 2 minutes. Substrates were dried to begin the experiment.Substrate was inclined 451 in regard to the vertical support to allowthe drainage of the solution. The multilayer films were built by aspray assisted layer-by-layer (LbL) technique of aqueous solutionsof alginate (2.5 mg mL�1), chitosan (1 mg mL�1) and a lab-madeferrofluid containing the NPs (8.0 mg mL�1). The spray time wasfixed at 5 s with a waiting time of 15 s between layers of chitosanand alginate and 30 s between NPs and Chitosan layers. The cyclewas formed by a bilayer Alg/Chi followed by four bilayers NPs/Chiand repeated as many times as necessary until the desired numberof layers was obtained. In order to achieve freestanding films, apolyvinyl alcohol (PVA) layer with a concentration of 100 mg mL�1

was deposited over the films using a pipette and dried for 24 hoursuntil become robust. The resulting multilayer film with the endingPVA layer was peeled from the glass substrate employing tweezers;after that, the resulting freestanding film was immersed in Milli-Qwater and the PVA layer was removed by dissolution in water.

The determination of iron content of the (Alg/Chi)n(NPs/Chi)m

films was carried out following the same procedure as that used forthe ferrofluid.

2.4 Cell culture and adhesion experiments

Direct contact assays to test cell adhesion were carried out on(Alg/Chi)n(NPs/Chi)m films with different number of layers.Films were sterilized under UV light for 5 hours. Humanneuroblastoma SH-SY5Y cells (ATCC CRL-2266) were culturedin Dulbecco’s modified Eagle’s medium (DMEM) and Ham’sF12 (1 : 1) with 15% fetal bovine serum, 100 IU mL�1 penicillin,100 mg mL�1 streptomycin and 2 mM L-glutamine. Cells weremaintained at 37 1C in a saturated humidity atmospherecontaining 95% air and 5% of CO2. At confluence, cells weretrypsinized and seeded on the films surface (1.5 � 105 cells perfilm). Cells were cultured on (Alg/Chi)n(NPs/Chi)m films for48 hours.

2.5 Dual beam (FIB-SEM) analysis

Dual beam FIB/SEM (Nova 200 NanoLab, FEI Company) wascarried out on (Alg/Chi)n(NPs/Chi)m films in order to determine,on the one hand, the inner organization of the polymer filmsand, on the other hand, observe the adhesion of SH-SY5Y cellsonto the surface of the polymer films. Cells adhered onto thefilms were fixed using 2% glutaraldehyde for 2 h at 4 1C andtreated with 2.5% potassium ferrocyanate and 1% osmiumtetroxide. Later, SH-SY5Y cells were dehydrated and coatedwith platinum for imaging using the FIB-SEM. SEM imageswere taken at 5 kV employing a FEG column, and a combinedGa-based 30 kV (10 pA) ion beam was used to cross-sectioningsingle cells. These investigations were completed by energy-dispersive X-ray spectroscopy (EDX) for chemical analysis.

2.6 Determination of the thermomagnetic properties of thefilms

For the (Alg/Chi)n(NPs/Chi)m films, the heating capacitywas measured in an open experimental setup, i.e., no thermalinsulation. The magnetic field was applied using a commercial

induction heating station (EasyHeat, Ambrell B.V., The Netherlands)working at f = 180 kHz and field amplitude of H = 35 kA m�1.The samples were placed at the center of 11-turn solenoid,and thermalized at 37 1C by a water-circulating jacket, whichhas also the function of thermally detach the samples fromthe joule heating generated by the coil. The temperature wasmeasured using a thermal camera (FLIR Systems, USA) ina zenithal position over the film. Before each experiment,the samples were placed inside the solenoid and allowed tothermalize at 37 1C for 5 min before the AMF was turned on(Fig. S1, ESI†). The heating time for each sample was 5 min(Fig. S2 shows representative results obtained for a(Alg/Chi)40(NPs/Chi)160 film, ESI†).

2.7 In vitro hyperthermia treatment of SH-SY5Y cells

Two different experiments were performance to analyze theheating efficiency of (Alg/Chi)n(NPs/Chi)m films, in order totest as a suitable material for magnetic hyperthermia (MHT)applications. In the first experiment, 1.5 � 105 human neuro-blastoma cells (SH-SY5Y) were seeded and cultured on the filmsurface for 48 hours in an Ibidi dish at 37 1C in a saturatedhumidity atmosphere containing 95% air and 5% of CO2. Afterthat, the film was subjected to 1 cycle of MHT ( f = 180 kHz,H = 35 kA m�1, t = 30 min). In addition, several films wereexposed to three cycles of MHT within five hours between them.Cells were maintained at 37 1C in a saturated humidity atmo-sphere 95% air and 5% CO2 between each cycle.

A second experimental protocol consisted on placing athermomagnetic film above 1.5 � 105 human neuroblastomacells (SH-SY5Y) cultured on an Ibidi dish for 48 h at 37 1C in asaturated humidity atmosphere containing 95% air and 5% ofCO2. After that, the Ibidi dish was subjected to two cycles ofMHT ( f = 180 kHz, H = 35 kA m�1, t = 30 min, waiting timebetween cycles = 5 h).

Viability of cells cultured on (Alg/Chi)n(NPs/Chi)m controlfilms was determined using the live/dead viability/cytotoxicitykit. Viability of cells subjected to MHT was measured, asdescribed above, 12 hours after the application of the lastMHT cycle in the first experimental protocol and after 3 hoursin the second experimental protocol. The kit contains calcein AM(4 mM in anhydrous DMSO) and ethidium homodimer-1 (2 mMin DMSO/H2O 1 : 4 (v/v)) which allows staining live cell in greenand dead cells in red, respectively. Cells were washed with 1 mLof PBS, after that, 1 mL of the stain kit was added and kept for30 minutes at room temperature in darkness. The stained cellswere visualized using a Leica SP2 AOBS confocal scanningmicroscope. Images were collected using the microscope insequential mode using a 20� (lens specification, HCPL FLUO-TAR NA 0.50; Leica), a line average of 16 and a format of 1024 �1024 pixels. The confocal pinhole was 1 Airy unit. The cellviability assay was carried out for three different samples foreach measurement. The results shown in this paper correspondto the average value of percentage of cell viability obtained bycomparison of the live cell area of every sample in regard to thelive cell area of the control. The software ImageJ was used tocalculate the area covered by live cells.

Journal of Materials Chemistry B Paper

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

J. Mater. Chem. B This journal is©The Royal Society of Chemistry 2017

3 Results and discussion3.1 Preparation and morphological characterization of themultilayer iron oxide-polymer films

Polymer/NPs films can be assembled by alternately sprayinga substrate, coated with a branched polyethylenimine (PEI)primer layer, with solutions of alginate (2.5 mg mL�1), chitosan(1 mg mL�1) and a lab-made aqueous ferrofluid. Alginate andchitosan are known to stablish electrostatic interactionsbetween the protonated amines of chitosan and carboxylategroups of alginate.35 The repeating structural unity wasformed by a bilayer Alg/Chi followed by four bilayers NPs/Chias illustrated in the Fig. 1a. Samples were denoted as(Alg/Chi)n(NPs/Chi)m where n stands for the number of Alg/Chibilayers (n = 2, 4, 8, 12, 20 and 40) and m for the number of NPs/Chi bilayers (m = 8, 16, 32, 48, 80 and 160).

The incorporation of NPs within the multilayer structure offilms was assessed visually (Fig. 1b), as can be observed, filmsbecome darker as the number of NPs layers increases. Themagnetite content (Fe3O4) incorporated within the multilayerfilms was determined by UV-Vis transmission spectrophoto-metry and plotted in Fig. 1c as a function of the number of NPslayers. The magnetite content increases linearly with the numberof NPs layers, that could be fitted by a f (Fe3O4) = �0.44 + 0.052N,where N is the number of NPs layers. The linear trend observedup to the maximum number of NPs layers deposited suggeststhat the final Fe3O4 concentration per unit area could be even

increased, although we observed the formation of cracks on thesurface of films with N 4 200 so that it was only possible to formrobust films until N = 200. However, as discussed below, thedeposition of the NPs layers during the process of build-upof multilayer polymer films was enough to produce a large(i.e., T 4 10 1C) temperature increase during the magnetichyperthermia experiments.

The inner structural organization of (Alg/Chi)n(NPs/Chi)m

multilayer films was characterized through dual-beam (FIB-SEM)images. Representative results corresponding to the cross sectionof sample (Alg/Chi)8(NPs/Chi)32 are shown in Fig. 2a. A magnifica-tion of the SEM image shown in Fig. 2b allows to observe somedegree of nanostructuration in layers of the sample. The thicknessof a film with 32 NPs layers, determined by SEM, is B1 mm(Fig. 2b). Considering a linear growth due to the fact that the ironcontent follows a linear tendency as the number of NPs layersincreases (Fig. 1c), the thickness of a film with 160 NPs layerscould be estimated as B5 mm. An EDX analysis of these cross-sectioned samples revealed the presence of NPs layers (Fig. 2c andd–f) as shown from elementary Fe mapping of the material. Thecarbon is due to the polysaccharides, alginate and chitosan.

3.2 Determination of the heating efficiency of the TMFs

The heating efficiency of single-domain NPs is related totwo main mechanisms of magnetic relaxation, i.e., Brown andNeel relaxation. The former is associated to the relaxation by

Fig. 1 (a) Schematic representation of the spray assisted layer-by-layer procedure and the repeating structural unit of the multilayer films. Notice thatNPs are coated by alginate (pink colour) due to the preparation procedure of the lab made ferrofluid. (b) Image of films assembled on glass substrateswith different number of NPs from 0 to 160 as indicated in the figure. (c) Determination of Fe3O4 content as a function of the number of NPs layersdeposited. Dashed line represents the linear fit of the data.

Paper Journal of Materials Chemistry B

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. B

physical rotation of the NPs against viscosity forces, whereasthe latter corresponds to the rotation of the magnetic momentin regard to the crystal lattice. Therefore Brown relaxation isstalled when NPs are incorporated into the LbL films, makingrelevant to determine how the heating efficiency is changedfrom the as prepared colloids after the NPs are immobilized.To evaluate this, the NPs of the original colloid were immobi-lized in gelatin gel, and the resulting SPA values compared tothe values from the NPs in the as prepared colloids.

The SPA of the as prepared colloids was quite high comparedto most reports from the literature, i.e. SPA = 868.4 � 1.9 W g�1

(see Fig. 3),36,37 although the actual efficiency is determined bythe SPA of blocked NPs as discussed above.38 As expected, theSPA decreased after immobilization in gelatin and this decreaseis higher as gelatin concentration increases. The exposition ofmagnetic NPs to a radio-frequency magnetic field gives rise to asingle domain particle that aligns its magnetic moment M tothe field H to decrease the interaction free energy. There are twodissipation mechanisms, Brownian relaxation and Neel relaxa-tion, which influence the magnetization of ferrofluids whenthey are exposed to external time varying magnetic fields. Brownianrelaxation is due to the physical rotation of the nanoparticlesagainst viscous torques from the surrounding media, i.e., carrierliquid, films, etc., and Neel relaxation is caused by the rotation ofthe magnetic moment inside the magnetic core.39,40

Due to the fact that both mechanisms take place simulta-neously in the dissipation mechanism of the ferrofluid, theimmobilization of NPs in gelatin tends to block the physicalrotation of NPs as gelatin concentration increases and, subse-quently, the Brownian contribution to the SPA value decreasesbeing the Neel relaxation the main dissipation mechanism.Therefore, in our system it is expected that the same physical processwill take place when NPs are encapsulated into the multilayer filmsbased on chitosan and alginate as it will be discussed below.

To assess the efficiency of the thermomagnetic films pro-duced, a series of (Alg/Chi)n(NPs/Chi)m films with differentnumber of NPs layers were heated under a fixed AMF amplitudeand frequency. The temperature increase, DT, measured from thebasal value T0 = 37 1C, is plotted as a function of time (Fig. 4a)and as a function of the NPs layers (Fig. 4b) for the differentsamples. The heating of the films, from the basal value T0 = 37 1C,as a function of the time by action of the AMF was followed for5 minutes for the different samples (Fig. 4a). As can be observed,temperature increases during the first two minutes, then aplateau is reached. It can be seen from Fig. 4b that DT increaseslinearly with the number of NPs layers (i.e., films with 80 or160 NPs layers yield DT of 6 and 12 1C, respectively). Notice that

Fig. 2 Dual beam FIB-SEM micrographs corresponding to the cross-sectional view of sample (Alg/Chi)8(NPs/Chi)32 (a) scale bar 1 mm and(b) scale bar 400 nm (c) EDX spectrum and area of individual componentsof EDX spectrum, carbon (d), iron (e) and oxygen (f).

Fig. 3 SPA of NPs as prepared in water (concentration of 8 mg mL�1) andblocked in gelatin (red bars, left scale) with their corresponding heatingrates (black bars, right scale).

Fig. 4 (a) Temperature increase of (Alg/Chi)n(NPs/Chi)m films measuredas a function of the number NPs layers and (b) temperature increase withthe time of application of the magnetic field for films with different numberof NPs layers: (’) 16, ( ) 32, ( ) 80 and ( ) 160.

Journal of Materials Chemistry B Paper

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

J. Mater. Chem. B This journal is©The Royal Society of Chemistry 2017

the magnetic heating of the ferrofluid with a magnetite concen-tration of 8.0 mg mL�1 gave rise to a temperature increase of25 1C in 1.5 min (Fig. S3, ESI†) and nanocomposite films with80 and 160 NPs layers give rise to a temperature increase of5.8 and 11.2 1C in 1.5 min, respectively (Fig. 4a). The immobiliza-tion of NPs into a multilayer Alg/Chi film through spray LbLproduces a similar effect to this observed for the immobilizationof NPs into a gel-like material decreasing the temperatureincrease (DT) in regard to the ferrofluid.

3.3 In vitro cell studies

As (Alg/Chi)n(NPs/Chi)m films are intended to be employed asadhesive pads for exogenous magnetic hyperthermia treatment,the first step in order to optimize the proposed application wasto ascertain the cell adhesion onto the surface of the thermo-magnetic films. Fig. 5a shows a general view of neuroblastomacells cultured for 24 h on a (Alg/Chi)20(NPs/Chi)80 film wherecell adhesion and proliferation can be observed along the filmsurface. A magnification of the SEM image (Fig. 5b) allowsobserving an individual cell adhered onto the film surface.

Further information on cell adhesion can be obtained from thecross section of the film. The cell adhered completely to thefilm surface as can be observed in a lamella cross section ofthe cell shown (Fig. 5c). The analysis of the EDX spectrumcorresponding to this area (Fig. 5d) with the different elements,carbon in green (Fig. 5e), iron in red (Fig. 5f), oxygen in grey(Fig. 5g) and silicon in blue (Fig. 5h) allows to clearly differ-entiate the iron component embedded on the film (red color)from the nucleus of the cell (black color) marked by the arrow.Note that the presence of silicon is due to the fact that the filmis supported on a glass substrate. The composition of the filmcan be quantitatively determined from the EDX spectrum(Fig. 5i) where the major component is iron (64.67%), followedby O (28.69%) and C (6.64%).

The effect of magnetic hyperthermia (MHT) on the cellviability was evaluated using neuroblastoma cells cultured onfilms with 80 and 160 NPs layers and subjected to 1 cycle ofMHT (H = 35.1 kA m�1 and f = 180 kHz; 30 min) and 3 cycles ofMHT giving a waiting time of 5 h between cycles. The resultswere compared to those obtained for neuroblastoma cellscultured on a control film not subjected to MHT treatment.A schematic representation of the experimental procedure isshown in Fig. 6a.

Cell adhesion on the surface of biomaterials is influencedby different factors such as the substrate topography, rigidity,anisotropy, surface charge, and wettability. Taken into accountthis, the cellular response to specific surfaces constitutesa complex phenomenon that includes a wide range of physicalexternal stimuli generated at the interface between the interfacebetween cells and the biomaterial surface.41,42 The discussionof the mechanism of adhesion of SH-SY5Y cells on the multilayerfilms under study goes far beyond the scope of this paper;however some qualitative results can be extracted from the resultsshown in control experiments depicted in Fig. 6b and c. Filmswith 80 NPs layers and not subjected to MHT present a lowernumber of live cells (stained in green) adhered to the films thatthat observed for cells cultured on films with 160 NPs (Fig. 6c).This is in general agreement with the results found in literatureregarding cell adhesion on multilayer films that show that,generally speaking, cells tend to adhere more on films whichhave a higher number of layers.43 Other studies have alsoshown that the incorporation of nanoparticles into polymerfilms improved cell adhesion due to the increase in surfaceroughness.44,45 In particular, human neuroblastoma cell line(SH-SY5Y) presents a remarkable response to surface nano-topography, and a surprising sensitivity to variations of fewnanometers.41 For multilayer films based on Alg and Chi withNPs layers, the elastic moduli determined through PeakForcequantitative nanomechanical mapping atomic force microscopy(PF-QNM AFM) increased with the number of NPs layers, beinghigher than the corresponding to multilayer films of Alg and Chiwithout NPs. For example, in the case of multilayer films withN = 40, the elastic modulus for a film with NPs was 26.2 �3.9 GPa whereas the elastic modulus for the same film withoutNPs was 9.2 � 1.6 GPa.28 Therefore, for the samples under studyhere, the increase of cell adhesion observed with the number of

Fig. 5 SEM micrographs corresponding to neuroblastoma cells culturedon a (Alg/Chi)20(NPs/Chi)80 film for 24 hours taken at two differentmagnifications: (a) 50 mm and (b) 5 mm, (c) FIB-SEM image correspondingto the cross-sectional view of an individual neuroblastoma cell adheredonto a (Alg/Chi)20(NPs/Chi)80 film, (d) EDX spectrum and area of individualcomponents of EDX spectrum delimited by a black rectangle, carbon (e),iron (f), oxygen (g) and silicon (h) and EDX spectrum (i).

Paper Journal of Materials Chemistry B

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. B

NPs layers could be attributed to a combination of both factors,a higher surface roughness and a higher stiffness of films asnumber of NPs layers increases.

For cells cultured onto films with 80 NPs layers (Fig. 6b),the application of one and three cycles of MHT does notsignificantly change the area covered by live cells in regard tothe control experiment. In contrast, Fig. 6c shows that theapplication of one cycle of MHT on cells cultured on films with160 NPs layers results in a decrease of area covered by live cellpopulation in regard to the control film, being this decreasemuch more noticeable after the application of 3 cycles of MHT.It is worth to note that the percentage of dead cells stained in red(ethidium homodimer-1 in Fig. 6b and c) does not correspond tothe total number of dead cells due to the fact that cell death resultsin the loss of cell adhesion and hence detachment from the film.

As can be observed in Fig. 6d, the cell viability data (%)obtained from the quantification of the area covered by livecells stained in green cultured on films with 80 NPs layersdemonstrates that the application of MHT results in negligiblecytotoxic effects after the application of one (96% viable cells)or three MHT cycles (94% viable cells). However, the cellviability data (%) corresponding to neuroblastoma cells culturedon films with 160 NPs layers decreases to 69% in regard to thecontrol experimental by the application of one cycle of MHT andto 21% by the application of three cycles of MHT. At this point, itis important to take into account the final temperature reachedby the TMFs by the application of AMF which is related to thenumber of NPs layers as shown in Fig. 4a. TMFs with 80 NPslayers reach a final temperature of 43 1C, which under theexperimental protocol employed, is not high enough to inducea cytotoxic effect on neuroblastoma cells whereas the tempera-ture reached by TMFs with 160 NPs layers increases to 49 1Cwhich induces a decrease of cell viability to 69% in regard to thecontrol experiment. These results demonstrate the importance ofcontrolling the concentration of NPs and hence, the temperaturegenerated during the magnetic field application in order toproduce a measurable magnetic hyperthermia effect. The tem-perature range for activating apoptotic cell mechanisms dependson the cellular type involved, but it can be generalized that above43–45 1C different metabolic pathways are activated to reverse theeffects of the temperature.18,46 For our experiments, the temperaturegradient was expected to be large due to the bidimensional natureof the samples (and thus large surfaces for exchanging heat).

It can be observed from Fig. 6 that the films with 80magnetic layers did not produce enough thermal effect to affectthe cell viability even after three application cycles. However,for those films with 160 magnetic layers the application ofseveral cycles increased the cell mortality through MHT. Thisis consistent with previous studies showing an increase incell death in cell cultures subjected to several expositionsto AMF.47,48 The stability of our films remained essentiallyunchanged after the magnetic hyperthermia experiments. Indeed,we would expect that excessive heating could result in damageof the polymers. However it is known that chitosan/alginatemultilayers retain their integrity up to 200 1C,49 which is muchhigher than target temperatures in any clinical application.

Fig. 6 (a) Experimental set up to measure in vitro magnetic hyperthermia.Images corresponding to the live/dead test carried out on neuroblastomacells cultured onto (b) (Alg/Chi)20(NPs/Chi)80 and (c) (Alg/Chi)40(NPs/Chi)160 films subjected to 0 cycles of MHT (control), 1 cycle of MHT(30 min) and 3 cycles of MHT (30 min, 5 h waiting time between cycles).Note that cells are labeled with calcein (live cells in green) and EthD-1(dead cells in red). (d) Determination of % cell viability obtained from theimages corresponding to live cells stained in green.

Journal of Materials Chemistry B Paper

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

J. Mater. Chem. B This journal is©The Royal Society of Chemistry 2017

Because of the potential therapeutic applications requirethese polymer films to be placed on the surface of the treatedarea, we conducted further in vitro MHT experiments aimed totest the cell viability of neuroblastoma cells grown underneaththe thermomagnetic films mimicking the in vivo conditions.To that aim, human neuroblastoma cells were cultured on thesurface of an Ibidi dish for 48 hours and a film with 80 NPslayers was placed above cells as shown in Fig. 7a. As controlexperiment, neuroblastoma cells were cultured in an Ibidi dishwith a film and not subjected to MHT. Cells were subjected totwo cycles of MHT (30 min and a waiting time of 5 hoursbetween cycles; H = 35 kA m�1, f = 180 kHz) which causes asignificant decrease in the calcein signal that corresponds tolive cells stained in green. In addition, it is possible to observethat cellular death takes place only within the area in contactwith the thermomagnetic film which proves the local effect ofthe magnetic hyperthermia (Fig. 7b).

In a recent study reported on the same SH-SY5Y cell linestudied here, it was demonstrated that MHT required a lowertemperature than an exogenous heating source (i.e., water bath)to provoke a given level of cell death.50 The results depicted in theinset of Fig. 7b show a much higher efficiency of the TMFs understudy in regard to exogenous heating provided by a thermalizedbath. The quantification of the calcein signal after the applicationof 2 MHT cycles yields a cell viability of 15% which represents asignificant decrease in regard to the control film. It is important tonote that the temperature achieved by the TMF after each cycle was43 1C (see Fig. 4). In the case of the application of hyperthermiathrough a thermalized water bath, the temperature needed toachieve a similar decrease in cell viability was 52 1C.50

4 Conclusions

We have successfully prepared thermomagnetic polymericfilms by alternate deposition of chitosan and alginate layersand magnetic nanoparticles using a spray assisted layer-by-layer assembly. The layered structure, together with the ordereddistribution and intercalation of NPs between layers, providedthe expected mechanical stability and thermal responsiveness.Indeed, the concentration of NPs (up to E65% in mass) intothe multilayer structure of these films was enough to produce atemperature increase of up to 12 1C above the control samplesunder external AMF. The resulting polymeric films showedgood adhesion properties for neuroblastoma cells cultured onthe film surface. By remotely heating the films using AMFs wewere able to induce a reduction in cell viability down to 69%and 21% of the original population after one and three heatingcycles, respectively. The results in experimental conditionsmimicking in vivo applications yielded a reduction of 85% ofthe cell viability only within the contact area with the film aftertwo cycles of MHT.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support from MINECO (MAT2014-53437-C2-1-P andMAT2013-42551) is gratefully acknowledged. Rebeca Hernandezthanks MEC for a Ramon y Cajal contract and Miryam Criadothanks MEC for a FPU fellowship and Tecnicas Reunidas S.A.for financial support in Residencia de estudiantes (CSIC). Theauthors are indebted to Dr T. Torres and LMA-INA for his helpduring the Dual Beam (FIB-SEM) experiments. Technical supportfrom SAI-UZ is also acknowledged for confocal microscopymeasurements (Centro de Investigaciones Biomedicas deAragon, CIBA). The COST action TD1402 (RADIOMAG) is alsoacknowledged.

Notes and references

1 G. Machado, A. F. Feil, P. Migowski, L. Rossi, M. Giovanela,J. D. S. Crespo, L. Miotti, M. A. Sortica, P. L. Grande,M. B. Pereira and R. R. B. Correia, Nanoscale, 2011, 3,1717–1723.

2 S. Taccola, F. Greco, A. Zucca, C. Innocenti, C. de JulianFernandez, G. Campo, C. Sangregorio, B. Mazzolai andV. Mattoli, ACS Appl. Mater. Interfaces, 2013, 5, 6324–6332.

3 F. Hua, T. Cui and Y. M. Lvov, Nano Lett., 2004, 4, 823–825.4 G. Xu, D. Pranantyo, L. Xu, K.-G. Neoh, E.-T. Kang and

S. L.-M. Teo, Ind. Eng. Chem. Res., 2016, 55, 10906–10915.5 F. Carosio, A. Di Blasio, F. Cuttica, J. Alongi, A. Frache and

G. Malucelli, Ind. Eng. Chem. Res., 2013, 52, 9544–9550.6 E. Ahn, T. Lee, M. Gu, M. Park, S. H. Min and B.-S. Kim,

Chem. Mater., 2017, 29, 69–79.

Fig. 7 (a) Experimental set up to measure in vitro magnetic hyperthermia.(b) Live/dead test carried out on neuroblastoma cells cultured onto an Ibididish, covered by a (Alg/Chi)20(NPs/Chi)80 film and subjected to 2 cycles ofMHT. Neuroblastoma cells cultured on an Ibidi dish and not subjected toMHT are employed as control. Cells are labeled with calcein (live cells ingreen) and EthD-1 (dead cells in red). The inset in Fig. 7b shows thedetermination of % cell viability obtained from the images correspondingto live cells stained in green.

Paper Journal of Materials Chemistry B

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online

This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. B

7 F. Ç. Cebeci, D. J. Schmidt and P. T. Hammond, ACS Appl.Mater. Interfaces, 2014, 6, 20519–20523.

8 T. Boudou, T. Crouzier, K. Ren, G. Blin and C. Picart, Adv.Mater., 2010, 22, 441–467.

9 J. o. Borges and J. F. Mano, Chem. Rev., 2014, 114, 8883–8942.10 B. P. Pichon, P. Louet, O. Felix, M. Drillon, S. Begin-Colin

and G. Decher, Chem. Mater., 2011, 23, 3668–3675.11 L. G. Paterno, F. J. Fonseca, G. B. Alcantara, M. A. G. Soler, P. C.

Morais, J. P. Sinnecker, M. A. Novak, E. C. D. Lima, F. L. Leiteand L. H. C. Mattoso, Thin Solid Films, 2009, 517, 1753–1758.

12 P. Schaaf, J.-C. Voegel, L. Jierry and F. Boulmedais, Adv.Mater., 2012, 24, 1001–1016.

13 I. Bruvera, R. Hernandez, C. Mijangos and G. Goya, J. Magn.Magn. Mater., 2015, 377, 446–451.

14 S. Gil, J. M. Silva and J. F. Mano, ACS Biomater. Sci. Eng.,2015, 1, 1016–1025.

15 S. K. Samal, M. Dash, T. Shelyakova, H. A. Declercq,M. Uhlarz, M. Banobre-Lopez, P. Dubruel, M. Cornelissen,T. Herrmannsdorfer, J. Rivas, G. Padeletti, S. De Smedt,K. Braeckmans, D. L. Kaplan and V. A. Dediu, ACS Appl.Mater. Interfaces, 2015, 7, 6282–6292.

16 V. Zamora-Mora, M. Fernandez-Gutierrez, J. San Roman,G. Goya, R. Hernandez and C. Mijangos, Carbohydr. Polym.,2014, 102, 691–698.

17 M. P. Calatayud, L. Asin, A. Tres, G. F. Goya and M. R. Ibarra,Curr. Nanosci., 2016, 12, 372–377.

18 A. Hervault and N. T. K. Thanh, Nanoscale, 2014, 6, 11553–11573.19 M. J. Sailor and J.-H. Park, Adv. Mater., 2012, 24, 3779–3802.20 V. Zamora-Mora, M. Fernandez-Gutierrez, A. Gonzalez-

Gomez, B. Sanz, J. S. Roman, G. F. Goya, R. Hernandezand C. Mijangos, Carbohydr. Polym., 2017, 157, 361–370.

21 L. Cristofolini, K. Szczepanowicz, D. Orsi, T. Rimoldi,F. Albertini and P. Warszynski, ACS Appl. Mater. Interfaces,2016, 8, 25043–25050.

22 E. Redolfi Riva, A. Desii, E. Sinibaldi, G. Ciofani, V. Piazza,B. Mazzolai and V. Mattoli, ACS Nano, 2014, 8, 5552–5563.

23 H. P. Kok and J. Crezee, Int. J. Hyperthermia, 2016, 1–9, DOI:10.1080/02656736.2016.1268726.

24 S. Gao, M. Zheng, X. Ren, Y. Tang and X. Liang, Oncotarget,2016, 7, 57367–57378.

25 P. Wust, B. Hildebrandt, G. Sreenivasa, B. Rau, J. Gellermann,H. Riess, R. Felix and P. M. Schlag, Lancet Oncol., 2002, 3,487–497.

26 Current cancer treatment-novel beyond conventional approaches,ed. O. Ozdemir, In Tech, Istanbul, pp. 455–474.

27 R. W. Y. Habash, R. Bansal, D. Krewski and H. T. Alhafid,Chem. Rev. Biomed. Eng., 2006, 34, 491–542.

28 M. Criado, E. Rebollar, A. Nogales, T. A. Ezquerra,F. Boulmedais, C. Mijangos and R. Hernandez, Biomacro-molecules, 2017, 18, 169–177.

29 L. Cheng, D. J. Smith, R. L. Anderson and P. Nagley, Int.J. Hyperthermia, 2011, 27, 415–426.

30 R. Signini and S. C. Filho, Polym. Bull., 1999, 42, 159–166.31 M. Lavertu, Z. Xia, A. Serreqi, M. Berrada, A. Rodrigues,

D. Wang, M. Buschmann and A. Gupta, J. Pharm. Biomed.Anal., 2003, 32, 1149–1158.

32 M. Rinaudo, M. Milas and P. Le Dung, Int. J. Biol. Macromol.,1993, 15, 281–285.

33 K. King, Food hydrocolloids, 1994, 8, 83–96.34 M. Criado, C. Mijangos and R. Hernandez, Material ferro-

fluidico acuoso, procedimiento de obtencion y sus usos, ES Pat.,201631442, 2016.

35 G. Lawrie, I. Keen, B. Drew, A. Chandler-Temple, L. Rintoul,P. Fredericks and L. Grøndahl, Biomacromolecules, 2007, 8,2533–2541.

36 C. Martinez-Boubeta, K. Simeonidis, A. Makridis, M. Angelakeris,O. Iglesias, P. Guardia, A. Cabot, L. Yedra, S. Estrade, F. Peiro,Z. Saghi, P. A. Midgley, I. Conde-Leboran, D. Serantes andD. Baldomir, Sci. Rep., 2013, 3, 1652.

37 P. Guardia, R. Di Corato, L. Lartigue, C. Wilhelm,A. Espinosa, M. Garcia-Hernandez, F. Gazeau, L. Mannaand T. Pellegrino, ACS Nano, 2012, 6, 3080–3091.

38 B. Sanz, M. P. Calatayud, E. De Biasi, E. Lima, M. V.Mansilla, R. D. Zysler, M. R. Ibarra and G. F. Goya, Sci.Rep., 2016, 6, 38733.

39 B. S. Sague, Universidad de Zaragoza, 2015.40 R. Kotitz, W. Weitschies, L. Trahms and W. Semmler,

J. Magn. Magn. Mater., 1999, 201, 102–104.41 V. Brunetti, G. Maiorano, L. Rizzello, B. Sorce, S. Sabella,

R. Cingolani and P. P. Pompa, Proc. Natl. Acad. Sci. U. S. A.,2010, 107, 6264–6269.

42 Y. N. Sergeeva, T. Huang, O. Felix, L. Jung, P. Tropel,S. Viville and G. Decher, Biointerphases, 2016, 11, 019009.

43 A. I. Neto, N. L. Vasconcelos, S. M. Oliveira, D. Ruiz-Molinaand J. F. Mano, Adv. Funct. Mater., 2016, 26, 2745–2755.

44 L. Ventrelli, T. Fujie, S. D. Turco, G. Basta, B. Mazzolai andV. Mattoli, J. Biomed. Mater. Res., Part A, 2014, 102, 2652–2661.

45 A. M. Lipski, C. J. Pino, F. R. Haselton, I. W. Chen andV. P. Shastri, Biomaterials, 2008, 29, 3836–3846.

46 P. I. P. Soares, I. M. M. Ferreira, R. AGBN Igreja, C. M. M.Novo and J. P. M. R. Borges, Recent Pat. Anticancer DrugDiscovery, 2012, 7, 64–73.

47 A. Ito, K. Tanaka, H. Honda, S. Abe, H. Yamaguchi andT. Kobayashi, J. Biosci. Bioeng., 2003, 96, 364–369.

48 S. Balivada, R. S. Rachakatla, H. Wang, T. N. Samarakoon,R. K. Dani, M. Pyle, F. O. Kroh, B. Walker, X. Leaym,O. B. Koper, M. Tamura, V. Chikan, S. H. Bossmann andD. L. Troyer, BMC Cancer, 2010, 10, 119.

49 D. Moura, S. G. Caridade, M. P. Sousa, E. Cunha, H. C.Rocha, J. F. Mano, M. C. Paiva and N. M. Alves, J. Mater.Chem. B, 2016, 4, 7718–7730.

50 B. Sanz, M. P. Calatayud, T. E. Torres, M. L. Fanarraga,M. R. Ibarra and G. F. Goya, Biomaterials, 2017, 114,62–70.

Journal of Materials Chemistry B Paper

Publ

ishe

d on

16

Oct

ober

201

7. D

ownl

oade

d by

UN

IVE

RSI

DA

D D

E Z

AR

AG

OZ

A o

n 26

/10/

2017

08:

51:1

7.

View Article Online