C

Ja

b

a

ARRAA

KACDFM

1

echnadeMaapacicfie[

ob

0d

Colloids and Surfaces B: Biointerfaces 79 (2010) 184–190

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

journa l homepage: www.e lsev ier .com/ locate /co lsur fb

ell-specific cytotoxicity of dextran-stabilized magnetite nanoparticles

ing Dinga, Ke Taob, Jiyu Lia,∗∗, Sheng Songb, Kang Sunb,∗

Department of General Surgery, Xinhua Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200092, PR ChinaState Key Lab of Metal Matrix Composites, Shanghai Jiao Tong University, Dongchuan Rd, No. 800, Minhang District, Shanghai 200240, PR China

r t i c l e i n f o

rticle history:eceived 21 September 2009eceived in revised form 23 February 2010ccepted 30 March 2010vailable online 8 April 2010

a b s t r a c t

Cytotoxicity of dextran-hybridized magnetite nanoparticles which were prepared by a novel polyolmethod was evaluated by incubation with four different kinds of cells, including rat liver cells BRL 3A,renal cells NRK, astrocyte and periphery blood mononuclear cells (PBMC). The study was designed notonly to evaluate their cytotoxicity but also to reflect the interaction between nanoparticles and relatedcells in their circulation processes. By fluorescent-activated cell sorting technique, it was found that the

eywords:poptosisytotoxicityextranlow cytometry

cytotoxicity of the nanoparticles is cell-specific. Under the concentrations in our study (0–128 mg/mL),the nanoparticles lead to the apoptosis of PBMC in a concentration-dependant manner, but have almostno influence on the other kinds of cells. TEM images demonstrate that the nanoparticles were endocy-tosed by BRL 3A, NRK and astrocyte, and result in the apoptosis of PBMC without the observation of theuptaking process. The results suggest that the related cells in nanoparticles cycling process should also

toxic

agnetite nanoparticle be concerned for the cyto

. Introduction

As one phase of iron oxide, magnetite nanoparticles (MNPs)xhibit excellent magnetic properties so that makes them potentiallinical reagents applicable to magnetic resonance imaging (MRI),yperthermia, drug-delivery, etc. [1,2]. With the rapid progress ofanobiotechnology, the requirements for the synthesis of MNPsre improved, for example, the size and size distribution control,ispersibility in water or other biological solvents, magnetic prop-rties, etc. More importantly, because of the interaction betweenNPs and bio-issues in the developing biomedical applications,

ssessment of human health and cell cytotoxicity is requireds a premise for their applications [3]. For the biomedical pur-oses, MNPs are often surface-coated to offer them hydrophilicitynd conjugating capability. Although the raw materials for theseoatings are generally biocompatible, as these nanoparticles arententionally engineered to interact with cells, considering theirytotoxicity is of great importance to ensure that the coating is suf-cient to prevent the release of toxic ions, and to ensure that thenhancements from coatings are not causing any adverse effects

4].

Much effort has been devoted to investigating the cytotoxicityf MNPs. Different surface-modified MNPs, including those coatedy polyethylene glycol (PEG) derivatives [5–7], polyvinyl alcohol

∗ Corresponding author. Tel.: +86 21 34202743; fax: +86 21 34202745.∗∗ Corresponding author. Tel.: +86 21 65790000 3080; fax: +86 21 65153984.

E-mail addresses: leejiyu@sina.com (J. Li), ksun@sjtu.edu.cn (K. Sun).

927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfb.2010.03.053

ity evaluation.© 2010 Elsevier B.V. All rights reserved.

(PVA) derivatives [8,9], polysaccharides [10], block copolymer [11],and some commercial products [12], as well as bare MNPs [13],were evaluated. Meanwhile, a series of targeting cell lines for appli-cations, such as fibroblasts, macrophages [14–16], were assessedas well. These studies were designed primarily to ensure the bio-safety of the nanoparticles in biomedical applications in vivo, or todemonstrate the mechanism of the cytotoxicity. However, biomed-ical applications of MNPs as drug-delivery agents, biosensors, orimaging contrast agents usually involve several processes, such asinjection of nanoparticles into the body, circulation with blood,uptake by certain tissues, metabolism by liver, evacuation by kid-ney, etc. Although some coated MNPs have been confirmed safe fortheir targeting cell lines, it is still unclear whether they are harmfulto the related cells in the other processes, which is also impor-tant for their bio-safety as much as cytotoxicity evaluation of thetargeting cells. Moreover, different groups choose to use differentMNPs, which makes direct comparison between the available stud-ies difficult and results in the obstacle in the investigation of thecytotoxicity mechanism [4].

Dextran is one kind of water-soluble polysaccharides, whichhas been clinically used for more than half a century [17,18]. Thedextran is a good stabilizer for nanoscale magnetite owing to itsexcellent biocompatibility and water-solubility, and has been uti-lized in both laboratory products and some commercial MNPs.

However, the dextran-stabilized MNPs have confirmed safety forsome cell lines on one hand, they can still cause the cells’ deathon the other hand [6]. In current work, MNPs were stabilized bydextran via a novel one-step method using diethylene glycol asboth solvent and reducer, and the resultant nanoparticles possess

es B: B

gmcwiiMef

2

2

vgttcMifswtft

pctenteosed

2

2

mGapJp

22tbvlcac11mb

J. Ding et al. / Colloids and Surfac

ood water-dispersibility. Related cells in circulation, bio-marking,etabolism and evacuation, namely, periphery blood mononuclear

ells (PBMC), astrocyte, rat liver cells BRL 3A, and renal cells NRKere respectively chosen to evaluate the nanoparticles cytotoxic-

ty. The investigation concerns the bio-safety of MNPs in the wholen vivo bioprocesses and forecasts biomedical applications for the

NPs. The results confirm different cytotoxicity of MNPs for differ-nt cell lines, and suggest the importance of cytotoxicity evaluationor the cell lines other than targeting ones.

. Experimental section

.1. Synthesis and characterization of MNPs

Dextran-stabilized magnetite nanoparticles were synthesizedia a hydrolysis-reducing process of FeCl3 by using diethylenelycol (DEG) as the solvent. Typically, a NaOH/DEG stock solu-ion was prepared by dissolving NaOH (50 mmol) in DEG (20 mL);his solution was heated at 120 ◦C for 1 h under nitrogen, andooled down and kept at 70 ◦C. Then, a mixture of dextran (0.15 g,

w = 40 k), FeCl3 (2 mmol), and DEG (15 mL) were heated to 220 ◦Cn a nitrogen atmosphere for 30 min under vigorous stirring,orming a transparent light-yellow solution. A NaOH/DEG stockolution (2 mL) was injected rapidly into the above hot mixture,hich induced the temperature to drop to 210 ◦C and the reac-

ion solution turned black immediately. The resulting mixture wasurther heated for 1 h to yield water-soluble magnetite nanocrys-als.

Transmission electron microscopy (TEM) samples were pre-ared by putting a drop of the as-prepared suspension on a carbonoated copper grid and then dried in a desiccator. TEM observa-ion was performed at 200 kV on a JEOL JEM-2100F transmissionlectron microscope. The crystallite phase of dextran-Fe3O4anoparticles was identified by recording X-ray diffraction pat-ern (XRD) using a D8 Advance Diffractometer (Bruker, Germany)quipped with a Cu K� radiation source. The surface propertiesf MNPs were characterized by Fourier transform infrared (FTIR)pectroscopy. FTIR were taken on a Bruker EQUINOX55 spectrom-ter. The sample was made by pressed disc method after mixingry nanoparticles with KBr.

.2. Cytotoxicity assay

.2.1. ReagentsDulbeccos Modified Eagles Medium (DMEM)/high glucose

edium, fetal bovine serum (FBS) and trypsin were obtained fromibco Co. Roswell Park Memorial Institute (RPMI) 1640 mediumnd DMEM/F12 medium were obtained from Hyclone Co. Lym-hocytes separation medium was purchased from Shanghai Hua

ing bio-tech Co. Ltd. Annexin-V-FITC apoptosis detection kits wereurchased from Bipec Biopharma Co.

.2.2. Cell lines

.2.2.1. PBMC. SD rats were purchased from Shanghai Experimen-al Animal Center, Chinese Academy of Sciences. The peripherallood was collected after sacrificing, and was added by an equalolume of phosphate buffered saline (PBS) containing anticoagu-ant (heparin). After being centrifuged at 600 × g for 20 min, theells were harvested from the interface between the plasma layernd lymphocyte separation medium using a Pasteur pipette. The

ells were washed twice with PBS, and were cultured in RPMI-640 media with 100 U/mL penicillin, 100 U/mL streptomycin, and0% heat-inactivated fetal bovine serum (FBS). Then the cells wereaintained in a humidified atmosphere of 95% air and 5% CO2 incu-

ator at 37 ◦C.

iointerfaces 79 (2010) 184–190 185

2.2.2.2. Astrocyte. Astrocytes were derived from brains of postna-tal days 2 SD rats. The cerebral cortex of SD rats was trituratedand trypsinized, so as to become the mixed glial Single-cell sus-pension. Then it was plated onto tissue culture T-75 flasks at thedensity of 1 × 106/mL in DMEM/F12 media with 100 U/mL peni-cillin, 100 U/mL streptomycin and 10% heat-inactivated FBS. Theculture media was replaced every 3 days. About 7–9 days later, thesuspension layer was removed and adherent cells were harvestedvia a rotating shaker at 220 rpm for 22 h at 37 ◦C. The harvestedcells were cultured in RPMI-1640 media with 100 U/mL penicillin,100 U/mL streptomycin, and 10% heat-inactivated FBS and weremaintained in a humidified atmosphere of 95% air and 5% CO2incubator at 37 ◦C.

2.2.2.3. BRL3A. BRL 3A immortalized rat liver cells were usedbetween passages 10 and 20. BRL 3A cells were grown in RPMI-1640culture media with 100 U/mL penicillin, 100 U/mL streptomycin,and 10% fetal bovine serum (FBS). Cells were plated at a density in6- or 24-well plates for exposures by 36–48 h when dosing withnanoparticles was initiated. The cells were maintained in a 5% CO2incubator at 37 ◦C.

2.2.2.4. NRK. NRK immortalized rat renal cells were purchasedfrom Cell Bank of Type Culture Collection of Chinese Academy ofSciences (Shanghai, China). NRK cells were grown in DMEM/Highglucose media containing 100 U/mL penicillin, 100 U/mL strepto-mycin, and 10% heat-inactivated FBS. The cells were maintained ina humidified atmosphere of 95% air and 5% CO2 incubator at 37 ◦C.

2.2.3. MethodCells were plated at a density of 1 × 106 cells/mL in 6-well plates

for exposures by 24 h when dosing with nanoparticle was initiated.The concentration of MNPs was controlled as 0 �g/mL, 32 �g/mL,64 �g/mL and 128 �g/mL, respectively. Cells were stained withAnnexin-V (25 �g/mL) and PI (50 �g/mL) and analyzed with a flowcytometer (Cytomics FC500, Beckman Counter Inc., CXP software).10,000 events were analyzed for each sample. For TEM observation,the cells were washed with PBS and then fixed with 2% glutaralde-hyde and 1% osmium tetroxide for 2 h. The cells were dehydratedusing a graded ethanol series (30%, 50%, 70% with 3% uranyl acetate,80%, 95%, and 100% for 10 min each), followed by two changes in100% propylene oxide. After infiltration and embedment in epoxyresins at 60 ◦C for 48 h, the sections were stained with lead cit-rate and analyzed by transmission electron microscopy (CM-120,Philips) at 80 kV.

3. Results and discussion

Fig. 1a shows the TEM image of as-prepared nanoparticles.The size of nanoparticles is in the range of 4–6 nm, showing nar-row size distribution. Clear lattice fringe in high-resolution TEMimage (Fig. 1b) demonstrates the well crystalline nature of resultantnanoparticles, and interplanar distance is about 0.26 nm, which isconsistent with the (3 1 1) plain of Fe3O4. XRD pattern (Fig. 1c) fur-ther demonstrates the phase and composition of magnetite becausethe position and relative intensity of peaks are same to those of purephase (JCPDS PDF19-0629).

To evaluate their surface chemistry, FTIR spectra of pure dextranand dextran-stabilized MNPs were recorded respectively, as shownin Fig. 2. It can be found that some peaks in coated MNPs should

be ascribed to dextran. For example, the peak located at 1650 cm−1

is attributed to the asymmetrical stretching vibration of O–C–O indextran, and broad peak at 3420 cm−1 should be ascribed to the–OH in dextran [19]. Meanwhile, some peaks are not consistentwith dextran, which may be caused by the coating of DEG. Because

186 J. Ding et al. / Colloids and Surfaces B: Biointerfaces 79 (2010) 184–190

attern

otnNt

Ft

Fig. 1. TEM images (a and b) and XRD p

f the well stabilization of dextran and perhaps DEG, the resul-

ant nanoparticles possess good dispersibility in water. When theanoparticles dispersed in water, they can resist the attraction ofbFeB magnet for about ten hours, as shown in Fig. 3. As a con-

rast, the nanocrystals would be attracted to the wall of bottle in

ig. 2. FTIR spectra of pure dextran and dextran-stabilized nanoparticles, respec-ively.

(c) of dextran-stabilized nanoparticles.

seconds when they dispersed in acetone, demonstrating their mag-netic property. Furthermore, dry nanoparticles powder prepared bythis method can be easily redispersed in water by slightly shaking,indicating the convenience for their storage and transportation.

Two strategies were generally used to modify the MNPs surfacefor the demand of applications in recent studies. The first one isbased on the coprecipitation of Fe2+ and Fe3+ ions in a basic aque-ous medium with the existence of coating macromolecules, whichis generally proceeded at a relatively low temperature [24]. Theother one is based on modifying the surface of pre-synthesizedhydrophobic MNPs [25]. In that case, besides the complicated oper-ation of synthesizing and modifying, the feasible kinds of modifyingmolecules are limited. In current work, the synthetic approachinvolves hydrolysis and reduction of FeCl3 by using DEG as both sol-vent and reducing agent [26]. Comparing with existing approaches,the current method combines the synthesis and surface modifica-tion of MNPs into one facile step, and the relative high reactiontemperature at about 220 ◦C provides MNPs with good crystallinity.Furthermore, resultant MNPs that simultaneously coated by dex-tran and DEG exhibit extraordinary water-dispersibility, which isof great importance for biomedical applications. In addition, this

method is general for most of the water-soluble macromoleculesto coat on, such as carboxymethyl-dextran, polyvinyl pyrrolidone(PVP), and polyethylene glycol (PEG). Owing to convenience of theapproach and water-dispersiblity of the products, macromoleculescoated nanoparticles prepared by the current method is suitable

J. Ding et al. / Colloids and Surfaces B: Biointerfaces 79 (2010) 184–190 187

e (a) a

fr

trtctfgDciirdti1idisn

Fc

Fig. 3. Photos of dextran-stabilized nanoparticles in aceton

or different cells’ cytotoxicity evaluation, which could make theesults comparable and systematic.

Classical MTT assay was preceded for astrocyte and NRK cellso evaluate their viability. As shown in Fig. 4, the cell viabilityemains above 90% even if the concentration of MNPs increaseso 500 �g/mL, demonstrating that the influence of MNPs on theells’ proliferation is ignorable. Fluorescent-activated cell sortingechnique (FACS) by flow cytometry utilizes a laser beam that dif-erentiates cells based on their size and density to determine theenotoxic potential of nanoparticles by examining the extent ofNA damage [20]. Using DNA intercalating dyes, the cellular DNAontent can be used to determine the proportion of cells undergo-ng apoptosis [21]. The FACS technique has been used intensivelyn studying the effect of nanoparticles exposure [22,23]. In cur-ent work, the cytotoxicity of various concentrated MNPs on fourifferent kinds of cells was evaluated by using FACS to detectheir apoptosis via a fluorescein annexin-V-FITC/PI double label-ng. The results of astrocyte with 0 �g/mL, 32 �g/mL, 64 �g/mL and28 �g/mL MNPs were respectively shown in Fig. 4 as an example,

n which the area C4 representing the apoptosis in early stage is the

irect indicator of the cytotoxicity of nanoparticles [21]. Compar-

ng with control experiment (without MNPs, Fig. 5a), MNPs did nothow obvious effect on the apoptosis of astrocyte cells, with cells’umber in area C4 being lower than 4%. Apoptosis in the early stage

ig. 4. Cytotoxicity profiles of astrocyte and NRK cell, when incubated with variousoncentrations of MNPs as determined by MTT assay.

nd water (b), with NbFeB magnet by the side of the bottle.

of other kinds of cells was summarized in Fig. 6. The results showedthe as-prepared MNPs show no obvious cytotoxicity to astrocyte,BRL 3A and NRK cells up to the concentration of 128 �g/mL. How-ever, 14.1% and 76.1% PBMC cells were located in area C4 with theconcentration of 64 �g/mL and 128 �g/mL, respectively, indicatingthat PBMC is hard to stand up to the high concentration of MNPs.

In term of dextran-stabilized MNPs, cytotoxic results in ourwork on astrocyte, NRK and BRL3A are comparable to a numberof studies focusing on its effects on cell viability and apopto-sis. For instance, Ferumoxtran-10 (a dextran-stabilized ultrasmallsuperparamagnetic iron oxide particle) was not toxic to humanhematopoietic progenitor cells after 2 h at a concentration of0.25 mg/mL [27]. Similarly, dextran-stabilized MNPs having ahydrodynamic diameter of about 20 nm were not toxic to phago-cytic C6 cells at concentrations of 0.73 M Fe for up to 10 days [28].In addition, cytotoxicity of the coated MNPs is also comparable tothe reported results of different surface-modified MNPs to BRL 3A[29] or astrocyte cells [30].

The dose-dependent cytotoxicity of MNPs for PBMC is similar tosome reports of adverse effects on cells in culture. Human dermalfibroblasts incubated with dextran-stabilized magnetite nanopar-ticles demonstrated an increase in cell death and apoptosis after48 h at a concentration of 0.05 mg/mL [31]. Ferumoxides inducedsignificant apoptosis in human monocytes after 4 h at concentra-tions of 0.5 mg/mL and above [32]. Labeling of mesenchymal stemcells with 0.1–0.25 mg/mL Ferumoxides for 48 and 72 h resulted inincreased cell death [33]. However, it is worth noting that althoughthe MNPs exert obvious cytotoxic effect on PBMC when the con-centration is up to 128 �g/mL and maybe therefore harmful to theimmune system, it is still safe when the concentration is below64 �g/mL. The concentration of commercial MNPs used in clini-cal trial, for example, generally around 20–35 �g/mL for MRI [34],is much lower than the highest concentration used in our experi-ments. Considering these data together with the lack of toxicity invitro, it seems likely that dextran-stabilized MNPs prepared by thecurrent method are not toxic to cells in vivo.

According to our results, it can be concluded that the cytotoxicityof dextran-stabilized MNPs is cell-specific. Actually, Brunner et al.[35] also found a cell-specific response to bare iron oxide nanopar-ticle exposure. 3T3 cells remained proliferative with the additionof up to 30 ppm iron oxide; however, human mesothelioma cells

exhibited significant reduction in cell viability at only 3.75 ppm ironoxide. They attributed the observed toxicity to iron-induced freeradical production.

For further evaluating the cell-specific cytotoxicity, TEM imageswere taken for different kinds of cells. The whole internalization

188 J. Ding et al. / Colloids and Surfaces B: Biointerfaces 79 (2010) 184–190

F strocy6

potttas(ww

Fffl

ig. 5. Fluorescein annexin-V-FITC/PI double labeling flow cytometry results for a4 �g/mL (c) and 128 �g/mL (d).

rocess of MNPs into BRL 3A and NRK cells can be respectivelybserved, as shown in Fig. 7. When the cells are exposed to MNPs,he particles are endocytosed from the extracellular fluid. A por-ion of the plasma membrane is then invaginated and pinched offo form a membrane-bound vesicle, as the arrows pointed in Fig. 7bnd e, respectively. As a result, the MNPs were enclosed in lyso-

omes, and the number of lysosomes increases. However, for PBMCshown in Fig. 8), the endocytosis process from extracellular fluidas not observed, and the number of lysosomes did not increaseith all concentrations of MNPs. Instead, most of the PBMC cells

ig. 6. Summarizing of initial apoptosis of different cells according to the resultsrom flow cytometry when incubated with various concentrations. Inset shows theow cytometry result for PBMC with MNPs at the concentration of 128 �g/mL.

te cells incubated with MNPs at the concentration of 0 �g/mL (a), 32 �g/mL (b),

was found apoptotic, as the example was shown in Fig. 8b, con-taining several apoptotic bodies.

The mechanism for MNPs cytotoxicity, when it does occur, hasbeen proposed as responding to reactive oxygen species (ROS) pro-duction via Fenton or Haber–Weiss reactions, which is caused bythe release of free iron ions and can result in lipid peroxidation,DNA damage, and protein oxidation [4]. Because the coating or sta-bilizing status and the release of iron are the same for all four kindsof cells, based on the above mechanism, cell-specificity should notbe observed. Another suggestion to the cytotoxicity is that MNPsshould be taken up by the cells as a result of endocytosis and pro-moting apoptosis due to weak cell adhesive interactions with thenanoparticles [6]. This hypothesis may be appropriate for our result.Due to a relative strong adhesion with astrocyte, NRK and BRL3A,MNPs would be uptaken by cells, sequestered in digestive vacuoles,and metabolized in a normal path, which would not lead to theapoptosis [4]. However, the MNPs are hardly to be uptaken by thePBMC, and the nanoparticles are tended to be attracted towardsthe cells surface [6,31]. Because the adhesion of the MNPs on thecells membrane would strongly affect the metabolism of normalcells, when the concentration of nanoparticles is high, the cellsare at high risk of apoptosis from overload with particles. Accord-ing to our results, the interaction of different cells with MNPs canbe proposed as follows: for NRK, BRL3A and astrocyte, when thecells are exposed to MNPs, the nanoparticles are endocytosed fromthe extracellular fluid. A portion of the plasma membrane is theninvaginated and pinched off to form a membrane-bound vesicle.The vesicle containing the MNPs fuses with lysosomes to form

secondary lysosomes, which leads to the increase of lysosome num-bers. This process is consistent with proposed mechanism of MNPsinternalization into breast cancer cell [36]. On the contrary, if theMNPs can hardly be uptaken by the cells such as PBMC, the adhesionof the MNPs on the cells membrane would affect the metabolism of

J. Ding et al. / Colloids and Surfaces B: Biointerfaces 79 (2010) 184–190 189

Fig. 7. TEM images for BRL 3A (a–c) and NRK (d–f) cells. (a and d) control cells, (b and e) MNPs were endocytosing into the cell. (c and f) MNPs have been endocytosed intothe cells. Scale bar is 2000 nm.

Fig. 8. TEM images for PBMC after incubating with 32 �g/mL MNPs, (a) alive cell, and (b) apoptotic cell.

1 es B: B

nbccct

4

nMscdaMwMccPPas

A

3Rtaf

R

[

[

[

[

[

[

[

[[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

90 J. Ding et al. / Colloids and Surfac

ormal cells and result in the sudden death of cells. Furthermore,ecause NRK and BRL 3A cells are from solidary organs but PBMCsome from periphery blood circulation, combining with the endo-ytosis proceeded by NRK and BRL 3A cells rather than by PBMC, itan be suggested that the MNPs potentially possess the capabilityo be used as MRI contrast agent or drug carrier for solidary organs.

. Conclusion

Highly water-dispersible Dextran-stabilized magnetiteanoparticles were synthesized, providing an appropriate kind ofNPs for further biomedical applications. Cytotoxicity of dextran-

tabilized magnetite nanoparticles was evaluated for four kinds ofells, and the results showed that they affected different cells inifferent manners. It was confirmed that three kinds of cells, i.e.strocyte, NRK and BRL 3A, would not obviously be affected by theNPs up to the concentration at 128 �g/mL. However, apoptosisas significantly observed for PBMC with high concentrationNPs (64 �g/mL and 128 �g/mL), demonstrating the cell-specific

ytotoxicty. Based on the TEM observation, the apoptosis of PBMCells was ascribed to the forced adhesion of the MNPs towardsBMC surface which may influenced the normal processes ofBMC’s metabolism. The results also indicated that the MNPsre possible to be used as MRI contrast agent or drug carrier forolidary organs.

cknowledgements

This work was financially supported by NSFC (Project No.0872630 and no. 30600598) and was sponsored by Shanghaiising-Star Program (Project No. 06QA14034). We thank Instrumen-al Analysis Center of SJTU for the assistance on measurements. Welso thank Shanghai Sunny New Technology Development Co., Ltd.or their support.

eferences

[1] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, Applications of magneticnanoparticles in biomedicine, J. Phys. D: Appl. Phys. 36 (2003) R167–R181.

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

[3] K. Soto, K.M. Garza, L.E. Murr, Cytotoxic effects of aggregated nanomaterials,Acta Biomater. 3 (2007) 351–358.

[4] N. Lewinski, V. Colvin, R. Drezek, Cytotoxicity of nanoparticles, Small 4 (2008)26–49.

[5] A.K. Gupta, S. Wells, Surface-modified superparamagnetic nanoparticles fordrug delivery: preparation, characterization, and cytotoxicity studies, IEEETrans. Nanobiosci. 3 (2004) 66–73.

[6] A.K. Gupta, M. Gupta, Cytotoxicity suppression and cellular uptake enhance-ment of surface modified magnetic nanoparticles, Biomaterials 26 (2005)1565–1573.

[7] F. Hu, K. Neoh, L. Cen, E. Kang, Cellular response to magnetic nanoparti-cles “PEGylated” via surface-initiated atom transfer radical polymerization,Biomacromolecules 7 (2006) 809–816.

[8] A. Petri-Fink, M. Chastellain, L. Juillerat-Jeanneret, A. Ferrari, A. Hofmann,Development of functionalized superparamagnetic iron oxide nanopar-ticles for interaction with human cancer cells, Biomaterials 26 (2005)2685–2694.

[9] F. Cengelli, D. Maysinger, F. Tschudi-Monnet, F. Montet, C. Corot, A. Petri-Fink,H. Hofmann, L. Juillerat-Jeanneret, Interaction of functionalized superparam-

agnetic iron oxide nanoparticles with brain structures, J. Pharmacol. Exp. Ther.318 (2006) 108–116.

10] A. Jordan, R. Scholz, P. Wust, H. Schirra, T. Schiestel, H. Schmidt, R. Felix, Endo-cytosis of dextran and silan-coated magnetite nanoparticles and the effect ofintracellular hyperthermia on human mammary carcinoma cells in vitro, J.Magn. Magn. Mater. 194 (1999) 185–196.

[

iointerfaces 79 (2010) 184–190

11] W.W. Yu, E. Chang, C.M. Sayes, R. Drezek, V.L. Colvin, Aqueous dispersion ofmonodisperse magnetic iron oxide nanocrystals through phase transfer, Nan-otechnology 17 (2006) 4483–4487.

12] K. Muller, J. Skeppera, M. Posfaib, R. Trivedi, S. Howarth, C. Corot, E. Lancelot,P. Thompson, A. Brown, J. Gillard, Effect of ultrasmall superparamagnetic ironoxide nanoparticles (Ferumoxtran-10) on human monocyte-macrophages invitro, Biomaterials 28 (2007) 1629–1642.

13] F.Y. Cheng, C.H. Su, Y.S. Yang, C.S. Yeh, C.Y. Tsai, C.L. Wu, M.T. Wu, D.B.Shieh, Characterization of aqueous dispersions of Fe3O4 nanoparticles and theirbiomedical applications, Biomaterials 26 (2005) 729–738.

14] A. Ito, K. Ino, T. Kobayashi, H. Honda, The effect of RGD peptide-conjugated mag-netite cationic liposomes on cell growth and cell sheet harvesting, Biomaterials26 (2005) 6185–6193.

15] Y. Zhang, N. Kohler, M. Zhang, Surface modification of superparamagneticmagnetite nanoparticles and their intracellular uptake, Biomaterials 23 (2002)1553–1561.

16] H. Lee, E. Lee, D.K. Kim, N.K. Jang, Y.Y. Jeong, S. Jon, Antibiofouling polymer-coated superparamagnetic iron oxide nanoparticles as potential magneticresonance contrast agents for in vivo cancer imaging, J. Am. Chem. Soc. 128(2006) 7383–7389.

17] L. Thoren, The dextrans-clinical data, Dev. Biol. Stand. 48 (1981) 157–167.18] R. Mehvar, Dextrans for targeted and sustained delivery of therapeutic and

imaging agents, J. Control. Rel. 69 (2000) 1–25.19] C.W. Jung, Surface properties of superparamagnetic iron oxide MR contrast

agents: ferumoxides, ferumoxtran, ferumoxil, Magn. Reson. Imag. 13 (1995)675–691.

20] M. King, Detection of dead cells and measurement of cell killing by flow cytom-etry, J. Immunol. Methods 243 (2000) 155–166.

21] Z. Darzynkiewicz, S. Bruno, G. Del Bino, W. Gorczyca, M.A. Hotz, P. Lassota, F.Traganos, Features of apoptotic cells measured by flow cytometry, Cytometry13 (1992) 795–808.

22] K. Kostarelos, L. Lacerda, G. Pastorin, W. Wu, S. Kieckowski, J. Luangsivilay, S.Godefroy, D. Pantarotto, J. Briand, S. Muller, M. Prato, A. Bianco, Cellular uptakeof functionalized carbon nanotubes is independent of functional group and celltype, Nat. Nanotechnol. 2 (2007) 108–113.

23] N. Kam, T. Jessop, P. Wender, H. Dai, Nanotube molecular transporters: inter-nalization of carbon nanotube-protein conjugates into mammalian cells, J. Am.Chem. Soc. 126 (2004) 6850–6851.

24] K. Tao, H. Dou, K. Sun, Combined investigation of experimental characterizationand theoretic calculation on the structure of dextran-Fe3O4 clusters, ColloidSurf. A 290 (2006) 70–76.

25] A.B. Bourlinos, A. Bakandritsos, V. Georgakilas, D. Petridis, Surface modificationof ultrafine magnetic iron oxide particles, Chem. Mater. 14 (2002) 3226–3228.

26] J. Ge, Y. Hu, M. Biasini, C. Dong, J. Guo, W.P. Beyermann, Y. Yin, One-step syn-thesis of highly water-soluble magnetite colloidal nanocrystals, Chem. Eur. J.13 (2007) 7153–7161.

27] H.E. Daldrup-Link, M. Rudelius, R.A.J. Oostendorp, M. Settles, G. Pionteck, S.Metz, Targeting of hematopoietic progenitor cells with MR contrast agents,Radiology 228 (2003) 760–767.

28] E. Schulze, J.T. Ferrucci, K. Poss, L. Lapointe, A. Bogdanova, R. Weissleder, Cellu-lar uptake and trafficking of a prototypical magnetic iron oxide label in vitro,Investig. Radiol. 10 (1995) 604–610.

29] S.M. Hussain, K.L. Hess, J.M. Gearhart, K.T. Geiss, J.J. Schlager, In vitro toxicityof nanoparticles in BRL 3A rat liver cells, Toxicol. In Vitro 19 (2005) 975–983.

30] G. Fleige, C. Nolte, M. Synowitz, F. Seeberger, H. Kettenmann, C. Zimmer, Mag-netic labeling of activated microglia in experimental gliomas, Neoplasia 3(2001) 489–499.

31] C.C. Berry, S. Well, S. Charles, G. Aichison, A.S.G. Curtis, Cell response todextran-derivatised iron oxide nanoparticles post internalization, Biomaterials25 (2004) 5405–5413.

32] S. Metz, G. Bonaterra, M. Rudelius, M. Settles, E.J. Rummeny, H.E. Daldrup-Link,Capacity of human monocytes to phagocytose approved iron oxide MR contrastagents in vitro, Eur. Radiol. 14 (2004) 1851–1858.

33] A.S. Arbab, L.A. Bashaw, B.R. Miller, E.K. Jordan, J.W.M. Bulte, J.A. Frank, Intracy-toplasmic tagging of cells with ferumoxides and transfection agent for cellularmagnetic resonance imaging after cell transplantation: methods and tech-niques, Transplantation 76 (2003) 1123–1130.

34] A.S. Arbab, G.T. Yocum, H. Kalish, E.K. Jordan, S.A. Anderson, A.Y. Khakoo, E.J.Read, J.A. Frank, Efficient magnetic cell labeling with protamine sulfate com-plexed to ferumoxides for cellular MRI, Blood 104 (2004) 1217–1223.

35] T. Brunner, P. Wick, P. Manser, P. Spohn, R. Grass, L. Limbach, A. Bruinink, W.

Stark, In vitro cytotoxicity of oxide nanoparticles: comparison to asbestos,silica, and the effect of particle solubility, Environ. Sci. Technol. 40 (2006)4374–4381.

36] N. Kohler, C. Sun, J. Wang, J. Zhang, Methotrexate-modified superparamagneticnanoparticles and their Intracellular uptake into human cancer cells, Langmuir21 (2005) 8858–8864.