Hemostasis Disorders Caused by Polymer Coated Iron Oxide Nanoparticles

Copyright © 2013 American Scientific PublishersAll rights reservedPrinted in the United States of America

ArticleJournal of

BiomedicalNanotechnology

Vol. 9, 1281–1294, 2013www.aspbs.com/jbn

Hemostasis Disorders Caused by PolymerCoated Iron Oxide Nanoparticles

Lamiaa M. A. Ali1, Martín Gutiérrez2�∗, Rosa Cornudella2, José Antonio Moreno2,Rafael Piñol1, Lierni Gabilondo1, Angel Millán1, and Fernando Palacio1�∗1Instituto de Ciencia de Materiales de Aragón. CSIC – Universidad de Zaragoza, and Departamento de Física de la Materia Condensada.Facultad de Ciencias, 50009 Zaragoza, Spain2Facultad de Medicina, Departamento de Medicina, Universidad de Zaragoza, 50009 Zaragoza, Spain

Background. Superparamagnetic iron oxide nanoparticles (SPIONs) are inorganic nanomaterials gaining strong clinicalinterest due to their increasing number of biological and medical applications. The stabilization of SPIONs in a biocom-patible stable suspension (bioferrofluid) is generally achieved by an adequate polymeric coating. As many applicationsusing these materials are intended for clinical use through intravenous injection, it is of outmost importance to evaluatetheir hemostatic behaviour. Objectives. The aim of this work is to evaluate the hemocompatibility of selected polymercoated bioferrofluids and of their separated components by observing the effects of the bioferrofluid on: the coagula-tion process—by measuring the prothrombin time (PT) and activated partial thromboplastin time (aPTT)–, the completeblood count (CBC)—Erythrocytes, Leucocytes, Platelets, Hemoglobin and hematocrit—and the hemolysis. Methods. ASPIONs/bioferrofluid model consisting of a magnetic core of iron oxide nanoparticles embedded within poly(4-vinyl pyri-dine) (P4VP) and all coated with polyethylene glycol (PEG) has been selected. Results and Conclusions. By increasingthe concentration of the bioferrofluids an inhibitory effect on the intrinsic pathway of blood coagulation is observed, asindicated by significant increase in aPTT in vitro while PT values stay normal. The effect of the coating components on theinhibition of blood coagulation process shows that PEG has no effect on the process while the P4VP-g-PEG copolymercoating has a strong anticoagulant effect indicating that P4VP is at the origin of such effects. The studied bioferrofluidshave no effect on the CBC neither they show in vitro hemolytic effect on blood.

KEYWORDS: Coagulation, Hemocompatibility, Hemostasis, Prothrombin Time, Superparamagnetic Iron Oxide Nanoparticles,

Thromboplastin Time.

INTRODUCTIONSuperparamagnetic iron oxide nanoparticles (SPIONs)a

have been of great interest since the last ten years due totheir important contributions to nanomedicine.1–3 They canbe useful tools for tissue repair,4 detoxification of biologi-cal fluids,5 bioseparation6 and immunization7 and can alsoact as an efficient specific drug and gene delivery tool,8

as contrast agents in magnetic resonance image (MRI),9–14

and in hyperthermia.15 Such a large variety of uses foran inorganic nanomaterial arise from the unique proper-ties coming from its magnetic functionality and from its

∗Authors to whom correspondence should be addressed.Emails: [email protected], [email protected]: 31 August 2012Revised/Accepted: 24 January 2013

aA list of acronyms is given in the Supporting Information.

low toxicity and biodegradability. Thus, SPIONs can bemoved or fixed by magnetic field gradients, they mod-ify the relaxation times of neighbour protons providingstrong contrast enhancement in MRI and they can convertmagnetic energy into heat under the effects of an alternat-ing magnetic field in hyperthermia processes with loweradverse effects than radiotherapy or chemotherapy.16–18

Thus far, several SPIONs preparations have alreadybeen used for clinical practice,2 especially for liver MRI,such as Ferumoxides (i.e., Endorem® in Europe, Feridex®

in the USA and Japan) coated with dextran,19 and Fer-ucarbutran (i.e., Resovist® in Europe and Japan) coatedwith carboxydextran.20 Other preparations proposed forcancer treatments are in clinical trial, such as MagnetofluidNanotherm®.21 However, despite the importance and cov-erage of biomedical applications of metal oxides, notmany studies have been addressed on the effects of these

J. Biomed. Nanotechnol. 2013, Vol. 9, No. 7 1550-7033/2013/9/1281/014 doi:10.1166/jbn.2013.1637 1281

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

materials in the blood. Thus, Ostomel et al. have inves-tigated the hemostatic response due to surface charges ofa variety of metal oxide particles which did not includeSPIONs,22 Li et al. have investigated the influence of nano-TiO2 on erythrocyte,23 and Singh et al. have investigatedthe disturbance in iron homeostasis caused by SPIONs.24

Since SPIONs are developed for intravenous administra-tion, the study of their hematological behaviour is of out-most importance and should be included in the toxicityand compatibility tests to be made in the development ofthese particles.Hemostasis is a very important physiological process

involving blood clotting (thrombus) and thrombus dissolu-tion (fibrinolysis).25 Coagulation occurs after the injury ofblood vessels through two processes: (1) primary hemosta-sis, in which the platelets immediately form a plug at theinjury site, and (2) secondary hemostasis, that has twopathways: (i) the intrinsic pathway (propagation pathway)that is operating with the help of many blood plasmaproteins, called coagulation factors, and (ii) the extrinsicpathway, (the initiation pathway) that is operating with thehelp of tissue factor (TF) upon vascular damage. The effi-ciency of the intrinsic pathway can be evaluated by mea-suring the activated partial thromboplastin time (aPTT)whereas the efficiency of the extrinsic pathway can beevaluated by measuring the prothrombin time (PT). Bothpathways lead to the formation of fibrin strands, whichstrengthen the platelet plug.Biomedical materials designed for contact with blood

including medical devices should be tested in relation totheir possible thrombogenic effect, since in theory theycan activate blood clotting by either of the two pathways,or by inducing platelet aggregation. Regarding nanoparti-cles for biomedical use, the few reported results show thatthey can behave as either pro-coagulant or hypo-coagulantagents depending on size,26 coating material27�28 and elec-tric charge.22�29 Therefore, it is essential to investigate theeffect of both, nanoparticles and coatings, on the bloodcoagulation system.The methods that have been used to investigate

the effects of nanoparticles on the coagulation pro-cess include the global test (PT, aPTT, Fibrinogen),thrombus—elastography, platelet aggregation and degra-dation products of fibrinogen and fibrin. Recently adetailed description of methods has been published,30

which focuses on the global plasma coagulation tests (PT,aPTT, Fibrinogen) and platelet aggregation. Some studieshave evaluated the compatibility of nanoparticles with theblood coagulation system in vivo in rats31 and rabbits,32

while most of the authors have carried them out onlyin vitro.26–28

The in vitro biocompatibility studies should also includeevidence of hemolysis and quantification of leukocytes,erythrocytes and platelets to rule out immediate cyto-toxicity of nanoparticles or contact spontaneous platelet

aggregation.26�27�33 Nanoparticles can cause hemolysis byacting on the membrane of red blood cells. If hemolysis issevere the number of red cells decreases and hemoglobinin plasma can be detected with the naked eye. If hemoly-sis is subtle can only be detected by a laboratory test forfree hemoglobin in plasma. Platelets are very sensitive tocontact with biological substances, like collagen, ADP andepinephrine, and non-biological ones, including ristocetin,solid materials and high shear stress effects. When thisoccurs they tend to produce micro aggregates that are eas-ily detected by modern blood cell counters because theydecrease the number of platelets and alter the size distribu-tion curve. Nanoparticles can potentially damage the cellmembrane and even the cytoplasm since they can penetrateinside cells. Blood cell counters not only quantify bloodcells but also can detect abnormalities in shape, size andhomogeneity.The magnetic nanoparticles studied here are part of

a synthetic platform for SPIONs based on the use ofpolymers. They are made from biocompatible compo-nents, and are stable as a colloidal aqueous fluid in phos-phate buffered saline (PBS) at pH= 7�4 and human bodytemperature.34�35 They consist of:(1) maghemite (�-Fe2O3� cores that are embedded in(2) a hydrophobic, poly(4-vinyl pyridine) (P4VP) matrix,covered by(3) a shell of a second hydrophilic polymer, polyethyleneglycol (PEG), that confers to the material high solubility,stability in aqueous solutions, biocompatibility, and pro-longed blood circulation time, and(4) other physical functionalities, such as fluorescentdyes.4�36

Eventually, a part of the PEG chains are functionalizedwith –COOH groups to provide sites for the anchor-ing of antibodies or peptides to nanoparticle surface sothey can be directed to specific targets. The nanoparti-cles core consist of a small number of maghemite mag-netic nanoparticles, which diameters can be varied from4 nm to 25 nm, all embedded within the P4VP matrix,thus preventing agglomeration. The use of P4VP is animportant part in our synthetic strategy as explained inRefs. [34 and 35]. On the one hand, the polymer is solu-ble in slightly acidic solutions but it is hydrophobic abovepH= 4�5, and pyridine groups are good iron ligands. Bothfeatures are very helpful for the control of particle size inour “in situ” precipitation method for the production ofmaghemite nanoparticles. On the other hand, pyridine isa good Michael donor, which is fundamental in the coat-ing method that is based on a Michael addition to acrylategroups at one end of polyethylene glycol acrylate polymerchains. The core and the polymeric coatings have a finalhydrodynamic diameter that can be varied from 30 nm to160 nm. Recently, stable colloidal aqueous suspensions ofSPIONs of this type have shown excellent relaxometricbehaviour as compared to commercial Endorem.10

1282 J. Biomed. Nanotechnol. 9, 1281–1294, 2013

Ali et al. Hemostasis Disorders Due to Polymer Coated SPIONs

This report focuses on the study of the hematologiceffects of the above-described SPIONs and their sepa-rated components on the coagulation process by measuringboth PT and aPTT. In addition, by measuring erythrocyte,leucocytes, haemoglobin, hematocrit and platelets bloodcount, their hemolytic effects have also been tested.

MATERIALS AND METHODSSPIONs PreparationThe synthesis of the bioferrofluids was performed in twosteps: (1) synthesis of maghemite/P4VP nanocomposites;and (2) synthesis of bioferrofluids in a phosphate bufferedsaline (PBS) medium.Maghemite–P4VP nanocomposites were prepared by

in situ precipitation from iron–P4VP coordination com-pounds, following the procedure described in Refs. [34]and [35]. Briefly, a film of iron-polymer precursor wasobtained by evaporation of a 50% water:acetone solutioncontaining 0.4 g of P4VP (Aldrich, 60 kDa), 0.794 mMof FeBr2 (Aldrich) and 1.584 mM of FeBr3 (Aldrich). Theprecursor film was treated with 20 mL of 1 M NaOH solu-tion for 1 h, washed with water, and dried in open air toobtain a maghemite nanocomposite.Carboxyl end-capped polyethylene glycol monoacrylate

(PEG(1000)A-COOH) was prepared by reacting succinicanhydride with the hydroxyl end group of polyethyleneglycol monoacrylate (PEG(1000)A) following the proce-dure described elsewhere.34 PEG(1000)A is commerciallyavailable from Monomer&Polymer, where 1000 is themolecular weight (Da) of the pendant poly(ethylene gly-col) chain. PEG(1000)A (10 mmol), succinic anhydride(15 mmol), and 4-dimethylaminopyridine (0.5 mmol) weredissolved in 50 mL of dry CH2Cl2 under argon atmo-sphere. The reaction was carried out at room temperaturefor 48 h. The reaction mixture was filtered and concen-trated under vacuum. Product was precipitated three timesin cold diethyl ether from tetrahydrofuran and then driedin vacuum. The molecular structure was confirmed by the1H-NMR spectrum recorded in a BRUKER AV-400 spec-trometer (400 MHz) using CDCl3 as solvent.The bioferrofluids were then prepared by dispersing

the maghemite/P4VP nanocomposites in an acidic solu-tion at pH around 3. The resulting acidic ferrofluid wasmixed with 0.180 mL of PEG(200)A (Monomer&Polymer,MW= 200 Da), and 0.018 g of PEG(1000)A-COOH, andit was heated to 70 �C during 21 h. Then, Na2HPO4was added for a 0.01 M final concentration, the pH wasadjusted to 7.40 by addition of a 0.2 M NaOH solu-tion, and the ionic strength was adjusted to 0.15 by addi-tion of NaCl and KCl to obtain 15 mL of bioferrofluid.This bioferrofluid was further purified by centrifugationat 196.000 G for 30 minutes and redispersed in PBS byultrasounds. Finally, the dispersion was filtered through asterile 0.22 �m membrane filter to obtain a bioferrofluid.

The polymer blank solutions were prepared in a simi-lar way. To obtain 100 mL of P4VP-grafted to polyethy-lene glycol acrylate (PEGA) (P4VP-g-PEGA) solution,1,336 g of P4VP, 1.203 mL of PEG(200)A, and 0.133 gof PEG(1000)A-COOH were dissolved in water at pH =3, and heated to 70 �C during 21 h. Then, Na2HPO4

was added to the solution for a 0.01 M final concen-tration, the pH was adjusted to 7.40 by addition of a0.2 M NaOH solution, and the ionic strength was adjustedto 0.15 by addition of NaCl and KCl, and finally, thevolume adjusted to 100 mL, and then filtered through asterile 0.22 �m membrane filter. To obtain 100 mL ofPEGA solution, the same procedure was followed in theabsence of P4VP. In this way, the concentration of PVP inthe P4VP-g-PEGA blank solution and the total concentra-tions of PEGA (PEG(200)A+PEG(1000)A-COOH) in theP4VP-g-PEGA and PEGA blank solutions were 13,36 g/L,the same as in the bioferrofluid sample.The structures of components of the nanoparticles coat-

ing are shown in Scheme 1. The reaction of P4VP andPEGA is a Michael addition involving the acrylate dou-ble bond and the nitrogen of the pyridine that becomesquaternized and therefore positively charged in this way.The structure of the P4VP-grafted PEGA (P4VP-g-PEGA)graft copolymer is comb-like. The backbone is a polyethy-lene chain with pyridine side groups. The PEGA chainsare linked to some of the pyridine groups by N C bondsbetween the nitrogen of the pyridine and the �-carbon ofthe acrylate PEG ester, thus hanging perpendicular to thecopolymer backbone as depicted in Scheme 1. Therefore,the N of pyridine groups linked to PEGA chains is posi-tively charged and the copolymer is cationic.

SPIONs CharacterizationThe total iron content in the samples was determined byatomic emission in a plasma 40 ICP Perkin–Elmer spec-trometer. The size of the maghemite nanoparticles wasdetermined from transmission electron microscopy (TEM)images in a Phillips CM30 microscope using samples pre-pared by dip coating of carbon coated copper grids. Thehydrodynamic size distribution of the dispersed nanopar-ticles in the bioferrofluids was determined by DynamicLight Scattering (DLS) using a Zetasizer Nano ZS fromMalvern. The superparamagnetic behaviour of SPIONswas verified by means of a MPMS-XL SQUID magne-tometer from Quantum Design.

Coagulation StudiesControl plasma: Blood samples were obtained fromhealthy human volunteers. Samples were collected in cit-rate (0.129 M) vacutainer tubes. The samples were cen-trifuged at 3500 rpm to obtain platelets-poor plasma (PPP).The plasma was processed for the coagulation studies ofaPTT and PT using the coagulometer TOP-ACL fromIL-Instrumentation. The results were within the referencelimits (23–37 s. and 9–14 s. respectively).

J. Biomed. Nanotechnol. 9, 1281–1294, 2013 1283

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

Scheme 1. Structure of polymers and copolymers.

PPP treated with bioferrofluids: Bioferrofluids weremixed with PPP making serial dilutions from 1:20 to 1:100(final particle concentration, 0.38 to 0.07 g/L of iron oxiderespectively) and processed for the measurement of PT andaPTT.PPP treated with PEGA: To study the effect of the coat-

ing, a PEGA blank solution was mixed with PPP at dilu-tions 1:10 and 1:100.PPP treated with P4VP-g-PEGA: To study the effect

of P4VP-g-PEGA copolymer on blood coagulation tests,a P4VP-g-PEGA blank solution was mixed with PPP atdilutions 1:10 and 1:100.In order to find out the origin of the anticoagulant effect,

the samples where an increase of aPTT was observedwere tested by mixing them with normal plasma and factorquantification.

Complete Blood Counts (CBC) StudiesControl blood: Blood samples were obtained from healthyhuman volunteers. Samples were collected in EDTA K3,1.8 mg/mL vacutainer. The blood samples were processedfor CBC studies using a Coulter LH 780 analyzer fromBeckman Coulter.Blood treated with Bioferrofluids, PEGA and P4VP-g-

PEGA: the investigated materials were mixed with theblood samples at dilutions 1:10 and 1:100 and processedfor blood cell counting.

Hemolysis StudiesBlood samples were obtained from healthy human vol-unteers. Samples were collected in Lithium heparin

17 UI/mL vacutainer tubes. The samples were processedfor the measurement of the free hemoglobin using a doublebeam spectrophotometer Analytic Jena-Specord 205 withwavelength range between 500–630 nm.

RESULTSSPIONs CharacterisationSPIONs Structural CharacterisationThe bioferrofluid stock sample used in this work was com-posed of 7.78 g/L of maghemite, 13.36 g/L of P4VP,12.03 g/L of PEG(200)A and 1.34 g/L of PEG(1000)A-COOH, and the necessary amounts of phosphate, sodium,potassium and chloride ions for a standard PBS solution ofpH= 7�40 and I = 0�15 M. The sample was characterizedby TEM and DLS and the results are shown in Figures 1and 2. Most of the nanoparticles are spherical and quitehomogeneous in size, they consist in a small number ofnon-aggregated magnetic nanoparticles embedded in theirpolymeric P4VP matrix and coated by the PEG, as shownin Figure 1. Their size distribution analysis yields an aver-age size of 9�1± 2�1 nm (Mean± SD) for the magneticcores as shown in Figure 2(A). DLS observations showa single population of particles with an average hydrody-namic diameter of 80 nm, as represented in Figure 2(B).The presence of free polymer, of 10 nm of average size, inthe suspension should be excluded as DSL does not showany peak on this size range. Measurements of zeta poten-tial of the nanoparticles suspension yielded average valuesclose to zero.

1284 J. Biomed. Nanotechnol. 9, 1281–1294, 2013

Ali et al. Hemostasis Disorders Due to Polymer Coated SPIONs

Figure 1. TEM images of maghemite nanoparticles in thebioferrofluid.

SPIONs Magnetic CharacterisationThe temperature dependence of the ac magnetic suscepti-bility of these samples is depicted in Figure 3 and showsthe characteristic behaviour of superparamagnetic systems,where the blocking temperature (TB), position of the max-imum of the out of phase component (� ′′), is a func-tion of the frequency of the applied field (Fig. 3(B)).Below 273 K the ferrofluid is frozen and TB increaseswith increasing the average volume, as predicted by theNéel relaxation model and in agreement with the averagesize estimated from the TEM images. Near to the melt-ing temperature of the ferrofluid a sudden increase of � ′′

is observed. This is attributed to the contribution of theBrownian relaxation mechanism which appears in ferroflu-ids with non-cero � ′′ in this temperature range (Fig. 3(B)).The frequency dependence of � ′′ is explained with theBrownian relaxation model and the position of the maxi-mum is in correspondence with the average hydrodynamicvolume attributed to the beads. These results corroboratethe assumed configuration for the ferrofluids in which themaghemite nanoparticles are bundled in polymeric beadswith hydrodynamic volumes of at least one order of mag-nitude bigger than the average volumes of the magneticcores.

Coagulation StudiesPPP Treated with BioferrofluidsPT shows no significant difference with dilutions 1:60(0.12 g/L Fe2O3) and above. At dilution 1:60, PT is10�79±0�70 s which compares well with its control value11�30± 0�68 s; At the highest concentrations of biofer-rofluids, for instance at dilution 1:20 (0.38 g/L Fe2O3� PTis 9�18± 0�40 s which is slightly shorter than the controlvalue 10�66± 0�42 s, although it is within our laboratory

(A)

(B)

Figure 2. SPIONs analysis: (A) Size distribution of sphericalmaghemite nanoparticles; (B) DLS size distribution of hydro-dynamic sizes in the bioferrofluid sample.

normal reference range (9–14 s). These results are shownin Figure 4(A).On the other hand, aPTT shows a rapid increase as

the bioferrofluids concentration increases. For instance, atdilution 1:100 (0.07 g/L Fe2O3�, aPTT value is 34�67±4�52 s, significantly different to the control one which is27�76± 2�16 s, and at dilution 1:20 (0.38 g/L Fe2O3�,aPTT value is 64�56± 7�21 s, twice the control value of31�85±1�48 s. The results are shown in Figure 4(B).Data corresponding to PT and aPTT determinations

were analyzed using Mann-Whitney test and are sum-marized in Table I. Mixture of the samples with normalplasma (1:1) did not correct the prolongation of aPTT, thethrombin time (TT) value was normal and the activities ofall factors of the intrinsic pathway were equally decreased.

PPP Treated with PEGATo study the coating effect on the coagulation process,a PEGA blank solution with the same proportion of

J. Biomed. Nanotechnol. 9, 1281–1294, 2013 1285

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

(A) (B)

Figure 3. In phase, (A), and out of phase, (B), components of the ac magnetic susceptibility versus temperature for thebioferrofluid.

(A) (B)

Figure 4. The effect of bioferrofluids on: (A) the prothrombin time (PT) in seconds, (B) the activated partial thromboplastin time(aPTT) in seconds. Bioferrofluids were diluted with PPP, making serial dilutions from 1:20 to 1:100 (final particle concentration,0.38 to 0.07 g/L of iron oxide respectively) and processed for the measurement of PT and aPTT. Values represent mean±SEM(n = 10), (∗) indicates significant differences between bioferrofluids and control.

Table I. Statistical analysis data for the effect of bioferrofluids on PT and aPTT; the number of samples was 10 for eachconcentration; the original bioferrofluid concentration was 7.78 g/L [Fe2O3].

PT mean±SD (s) aPTT mean±SD (s)

Dilution factor Test Control P (Mann-Whitney) Test Control P (Mann-Whitney)

1:20 9�18±0�40 10�66±0�42 0.0002 64�56±7�21 31�85±1�48 0.00021:30 9�47±0�30 10�66±0�42 0.0002 56�06±11�77 31�85±1�48 0.00021:40 9�76±0�35 10�66±0�42 0.0004 46�04±3�96 31�85±1�48 0.00021:50 10�16±0�55 10�88±0�69 0.0156 40�54±4�07 29�65±3�16 0.00021:60 10�79±0�70 11�30±0�68 0.1306 39�19±4�11 27�42±1�88 0.00021:70 10�97±0�82 11�17±0�67 0.4497 37�60±3�88 27�51±1�58 0.00021:80 10�94±0�78 11�28±0�66 0.2899 37�02±4�11 27�88±2�11 0.00021:90 10�27±0�76 10�61±0�80 0.2730 34�03±3�51 27�01±1�87 0.00041:100 10�63±1�07 10�89±0�66 0.3258 34�67±4�52 27�76±2�16 0.0015

1286 J. Biomed. Nanotechnol. 9, 1281–1294, 2013

Ali et al. Hemostasis Disorders Due to Polymer Coated SPIONs

(A) (B)

Figure 5. The effect of PEGA on: (A) prothrombin time (PT), (B) the activated partial thromboplastin time (aPTT), in seconds.PEGA blank solution was mixed with PPP at dilutions 1:10 and 1:100, samples were processed for the measurement of PT andaPTT. Values represent mean±SEM (n = 6), (∗) marks significant differences between PEGA and control.

PEG(200)A+PEG(1000)A-COOH as the one used for thenanoparticles coating (13.36 g/L) was used. No alterationin the coagulation profile has been observed, except thePT at the dilution 1:10 (1.33 g/L PEGA), the PT valuefor the test sample was 12�40± 0�69 s, which is slightlydifferent than the control value 11�47±0�65 s. The resultsare shown in Figure 5, while statistical analysis data areshown in Table II.

PPP Treated with P4VP-g-PEGAThe P4VP-g-PEGA blank sample used in this work wasprepared under the same conditions and concentrationsapplied for the bioferrofluid preparation. The PT valuedoes not change at dilution 1:10 (11�78± 1�17 s as com-pared to control value 11�47±0�65 s) and becomes shorterthan the control value at dilution 1:100 (10�28±0�57 s and11�30± 0�48 s, respectively) and dilution 1:1000 (9�98±0�98 s and 11�30± 0�48 s, respectively). On the contrary,the aPTT shows a large increase at a dilution of 1:10and 1:100, greater than 120 seconds, the time limit of theequipment; further dilution at 1:1000 still shows signifi-cant differences for aPTT value with respect to the controlone (39�63±11�17 s and 31�45±2�01 s, respectively). The

Table II. Statistical analysis data for the effect of PEGA on PT and aPTT; the number of samples was 6 for each concentration;the original PEGA concentration was 13.36 g/L.

PT Mean±SD (s) aPTT Mean±SD (s)

Dilution factor Test Control P (Mann-Whitney) Test Control P (Mann-Whitney)

1:10 12�40±0�69 11�47±0�65 0.04 29�35±1�49 29�80±1�47 0.421:100 10�78±0�31 10�88±0�41 0.63 29�17±2�36 29�18±3�07 0.87

results are shown in Figure 6 while statistical analysis dataare summarized in Table III.

Complete Blood Counts (CBC) StudiesBlood Treated with BioferrofluidsNo significant differences were detected in CBC betweencontrol and treated blood with bioferrofluids in both the1:10 and 1:100 dilutions, as well as no significant differ-ence in hemoglobin and hematocrit were either observed.The spectrophotometric study of hemoglobin in plasmademonstrates the absence of hemolysis for dilution 1:100.For dilution 1:10 the high concentration of particles inter-fere the spectrophotometric measurements; however, nakedeye inspection does not indicate hemolysis. Altogetherthese data show the safety of nanoparticles in relation toerythrocytes. Platelets and leukocytes do not show signif-icant differences between the control sample and the twodilutions (1:10, 1:100) tested neither the instrument showflags indicating morphologic alterations and aggregates inany of the studied cell series, demonstrating the safety ofnanoparticles on these two cell lines. The only hematolog-ical abnormality is in the coagulation tests, with an unex-pected increase aPTT, discussed in detail later. Statistical

J. Biomed. Nanotechnol. 9, 1281–1294, 2013 1287

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

(A) (B)

Figure 6. The effect of P4VP-g-PEGA on: (A) Prothrombin time (PT), (B) the activated partial thromboplastin time (aPTT), inseconds. P4VP-g-PEGA blank solution was mixed with PPP at dilutions 1:10, 1:100 and 1:1000, samples were processed for themeasurement of PT and aPTT Values represent mean±SEM (n = 6), (*) indicates significant differences between P4VP-g-PEGAand control.

analysis data are summarised in Table IV and results areshown in Figure 7.

Blood Treated with PEGA and with P4VP-g-PEGAThe CBC (erythrocytes, leukocytes, platelets, hemoglobinand hematocrit) corresponding to blood treated withPEGA[PEG(200)A+PEG(1000)A-COOH] shows a minordifference with control sample only for erythrocytes,hemoglobin and hematocrit at dilution 1:100 (P = 0�04).The CBC corresponding to blood treated with P4VP-g-PEGA at dilutions 1:10 and 1:100 did not show in eithercase significant differences from control blood. The resultsare shown, respectively, in Figures 8 and 9 while statisticalanalysis data are summarised in Table V.

Hemolysis StudiesThe bioferrofluids were diluted with whole blood at dilu-tion 1:10 and 1:100, 0.77 and 0.07 g/L Fe2O3 respectively,then centrifuged at 2500 rpm for 5 min and the plasma pro-cessed for the measurement of the free hemoglobin. Bloodtreated with bioferrofluids at dilution 1:100 (0.07 g/LFe2O3) did not show any hemolytic effect neither observedwith naked eyes or spectrophotometric measurements. Thisresult is confirmed by the obtained results from complete

Table III. Statistical analysis data for the effect of P4VP-g-PEGA copolymer on PT and aPTT; the number of samples was 6 foreach concentration.

PT mean±SD (s) aPTT mean±SD (s)

Dilution factor Test Control P (Mann-Whitney) Test Control P (Mann-Whitney)

1:10 11�78±1�17 11�47±0�65 0.74 > 120 29�80±1�47 −1:100 10�28±0�57 11�30±0�48 0.02 > 120 31�45±2�01 −1:1000 9�98±0�98 11�30±0�48 0.04 39�63±11�17 31�45±2�01 0.03

Table IV. Statistical analysis data for the effect of bioferroflu-ids on CBC; the number of samples was 4 for each concen-tration; the original bioferrofluid concentration was 7.78 g/L[Fe2O3].

CBC Mean±SD

Dilution factor Test Control P (Mann-Whitney)

Erythrocytes (106/�L)1:10 4�65±0�31 4�66±0�27 >0.991:100 4�69±0�36 4�61±0�30 0.77

Leukocytes (103/�L)1:10 6�13±2�28 6�28±2�20 >0.991:100 6�79±2�21 6�88±2�21 0.77

Platelets (103/�L)1:10 180�95±26�48 193�50±25�72 0.381:100 176�25±34�72 180�25±34�22 0.77

Hemoglobin (g/dL)1:10 14�22±1�13 14�00±0�95 0.561:100 14�39±1�04 14�23±0�91 0.88

Hematocrit (%)1:10 40�92±2�97 40�68±1�81 0.771:100 42�70±2�09 41�80±1�32 0.38

blood count (CBC) where there is no significant decreasein erythrocytes, hemoglobin, and hematocrite. However,this assay is not suitable for hemolysis detection at highconcentration of bioferrofluids (i.e., dilution 1:10), since

1288 J. Biomed. Nanotechnol. 9, 1281–1294, 2013

Ali et al. Hemostasis Disorders Due to Polymer Coated SPIONs

(A) (B)

(C) (D)

(E)

Figure 7. The effect of bioferrofluids (0.07 and 0.77 g/L Fe2O3� on: (A) Erythrocytes, (B) Leukocytes, (C) Platelets, (D) Hemoglobinand (E) Hematocrit. Bioferrofluids were diluted with blood at dilutions 1:10 and 1:100 and processed for blood cell counting.Values represent mean±SEM (n = 4).

their high concentration could interfere with the spec-trophotometic measurements. In the control sample theplasma hemoglobin was been analyzed after blood cen-trifugation at 2500 rpm for 5 minutes.

DISCUSSIONA prolonged aPTT suggests either deficiency of one ormore coagulation factors or the presence of an anticoagu-lant in plasma. The usual procedure to elucidate which of

J. Biomed. Nanotechnol. 9, 1281–1294, 2013 1289

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

(A) (B)

(C)

(E)

(D)

Figure 8. The effect of PEGA at different dilutions 1:100 and 1:10 (0.13 and 1.33g/L) on: (A) Erythrocytes, (B) Leukocytes, (C)Platelets, (D) Hemoglobin and (E) Hematocrit. PEGA was diluted with blood at dilutions 1:10 and 1:100 and processed for bloodcell counting. Values represent mean ± SEM (n = 3), (∗) marks significant differences between PEGA and control.

1290 J. Biomed. Nanotechnol. 9, 1281–1294, 2013

Ali et al. Hemostasis Disorders Due to Polymer Coated SPIONs

(A) (B)

(C) (D)

(E)

Figure 9. The effect of P4VP-g-PEGA at different dilutions 1:100 and 1:10 (0.13 and 1.33 g/L) on: (A) Erythrocytes, (B) Leukocytes,(C) Platelets, (D) Hemoglobin and (E) Hematocrit. P4VP-g-PEGA was diluted with blood at dilutions 1:10 and 1:100 and processedfor blood cell counting.Values represent mean±SEM (n = 3).

these two options is true is to mix the abnormal plasmawith normal plasma to see if it corrects the prolongationof aPTT; if that happens it indicates a deficiency of one ormore hemostasis factors (normal plasma corrects the fac-tor deficiency). The next step is to measure the activity

of individual coagulation factors involved in the aPTTtest.In the event that the normal plasma does not correct

the aPTT, the most likely is that an anticoagulant (whichinhibits the normal factors) is present in the plasma. Then,

J. Biomed. Nanotechnol. 9, 1281–1294, 2013 1291

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

Table V. Statistical analysis data for the effect of PEGA (PEG(200)A+PEG(1000)A-COOH) and P4VP-g-PEGA on CBC, the numberof samples was 3 for each concentration; the original PEGA and P4VP-g-PEGA concentrations were 13.36 g/L.

PEGA P4VP-g-PEGA

CBC Mean±SD CBC Mean±SD

Dilution factor Test Control P (Mann-Whitney) Test Control P (Mann-Whitney)

Erythrocytes (106/�L)1:10 4�33±0�12 4�33±0�10 0.82 5�16±0�41 5�11±0�41 0.511:100 5�04±0�28 4�38±0�16 0.04 5�19±0�44 4�96±0�65 0.51

Leukocytes (103/�L)1:10 8�25±2�63 8�27±2�54 > 0�99 5�50±0�90 5�57±0�83 0.511:100 8�76±2�52 8�67±1�90 0.82 7�61±1�99 7�43±2�72 0.82

Platelets (103/�L)1:10 311�30±98�58 316�00±121�52 0.82 229�24±40�71 228�80±47�34 0.831:100 324�50±107�72 319�33±117�86 0.82 230�48±46�71 228�80±47�34 0.92

Hemoglobin (g/dL)1:10 12�32±0�98 12�43±0�98 0.82 14�92±1�40 15�07±1�46 0.821:100 15�11±0�61 12�70±1�28 0.04 15�08±0�99 14�93±1�69 0.82

Hematocrit (%)1:10 38�28±3�34 37�77±3�62 0.51 45�72±4�65 44�30±4�48 0.511:100 43�71±2�93 37�77±3�62 0.04 47�47±2�96 44�43±4�25 0.27

it is important to find out if it is a specific or nonspecificanticoagulant with new tests, including the thrombin timeand quantifying the activity of factors potentially affectedby the anticoagulant at various dilutions.As stated above, the mixture with normal plasma (1:1)

did not correct the prolongation of aPTT, the TT value wasnormal and the activities of all clotting factors (VIII, IX,XI, XII) in the intrinsic pathway were equally decreased,thus indicating that inhibitory activity is nonspecific. Wetherefore conclude that the bioferrofluid behaves as a non-specific anticoagulant. In one way this behaviour is similarto that of heparin from the fact that lengths the aPTT,although heparin has also a specific effect on anti-thrombinIII. In fact, the degrees of aPTT lengthening are compara-ble in both cases and therefore well within the therapeuticrange.In order to gain insight into the origin of this anti-

coagulant effect we have studied the bioferrofluid com-ponents separately. The surface of the nanoparticles inthe bioferrofluid is formed by 1/10 in weight of PEGAlong chains (MW = 1000) ending on carboxylic groupsand 9/10 of PEGA short chains (MW = 200) ending onhydroxyl groups (see Scheme 1). It seems that either com-ponent is not likely to be at the origin of aPTT lengthen-ing, because a solution of these compounds in the sameconcentration as in the bioferrofluid does not show anyeffect. The PEGA hydrophilic chains are linked to theP4VP core by pyridine-acrylate bonds that generate a pos-itive charge on the N atom of the pyridine, thus the result-ing P4VP-g-PEGA grafted copolymer is polycationic andhas a brush-like structure. This copolymer alone produceda strong aPTT lengthening (see Table III), larger than thewhole bioferrofluid, and must be the responsible for theanticoagulant effect. However, the mechanism of action is

not clear. In fact, although their effects are comparable tothose of heparin the molecular structure is very differentbecause heparin action is associated to the large presenceof negatively charged groups such as sulfate, sulfonamideand carboxylic groups whereas P4VP-g-PEGA copolymeris positively charged.We have only found a precedent of bioferrofluids show-

ing a similar anticoagulant effect.32 The particle structurein this case was very different from ours as they consistedof a mixture of iron and iron oxide nanoparticles encapsu-lated in a carbon matrix. Although they share a commonfeature that is the presence of carboxylic groups on thesurface, this cannot cause of aPTT lengthening as inferredfrom PEGA blank experiments.As previously remarked the information about the

effect of nanoparticles in the coagulation process isscarce. Studies on metal oxide nanoparticles includingiron oxide have shown that positive charged surfacesincrease coagulation time.22 However, these experimentsare not exactly comparable to ours since the particles wereuncoated. Actually, the coating can change the behaviourof nanoparticles in blood, i.e., it has been reportedthat PEGylated poly(N -isopropylacrylamide) nanoparti-cles are hemocompatible with insignificant toxicity butnon-PEGylated caused some hemolysis.27 Similar resultshave been reported for poly(lactide-co-glycolide) acid.37

Another factor concerning nanoparticles that could affecttheir behaviour in blood is size. It has been shown thatthe negatively charged polystyrene nanoparticles becomehematotoxic when their size decreases below 60 nm.26 Ournanoparticles exceed this size and therefore are within thesafe size range. Finally, protein adsorption is important inblood compatibility as it is an activation factor in the bloodcoagulation process. That is the main reason for using PEG

1292 J. Biomed. Nanotechnol. 9, 1281–1294, 2013

Ali et al. Hemostasis Disorders Due to Polymer Coated SPIONs

as a coating component in our particles, as PEG is the mostaccepted coating for minimising protein adsorption.1 Thus,it should be expected that protein adsorption is weak inthis case. On the other hand, an extensive protein adsorp-tion would probably promote coagulation, which is notobserved here.

CONCLUSIONSIt has been shown that these bioferrofluids do not haveany prejudicial effect on erythrocytes, leukocytes, platelets,hemoglobin and hematocrit even at the highest concen-tration tested (0.77 g of Fe2O3 per liter of blood). It isalso clear that the bioferrofluid does not cause any pro-coagulant effect that would prevent its use in intravenousapplications. On the contrary, it shows an anticoagulanteffect that is reflected on a lengthening of the aPTT.In addition, the anticoagulant effect of the bioferrofluid isnon-specific because it does not affect to the thrombin time(TT), and it reduces equally the activity of clotting factors(VIII, IX, XI, XII) in the intrinsic pathway. Therefore, thebioferrofluids formed by stable suspensions of P4VP-g-PEG-coated SPIONs in PBS act as non-specific circulatinganticoagulant agents in vitro. While PEG component doesnot seem to have any effect on the coagulation process,the coating copolymer P4VP-g-PEG shows strong antico-agulant behaviour indicating that P4VP is at the origin ofthe effect. Bioferrofluids have no effect on the CBC nei-ther they show hemolytic effect on blood in vitro. Furtherexperiments using small animals in vivo are in progress toassess the validity of these conclusions.

DISCLOSURE OF CONFLICTS OF INTERESTThe authors state that they have no conflict of interest.

AcronymsADP Adenosine diphosphateaPTT Activated partial thromboplastin

timeCBC Complete blood countsDLS Dynamic light scattering

Maghemite–P4VP Nanocomposites formed bymaghemite and P4VP

MRI Magnetic resonance imageP4VP poly(4-vinyl pyridine)

P4VP-g-PEGA P4VP-grafted PEGAPBS Phosphate buffer solutionPEG Polyethylene glycol

PEG(1000)A Polyethylene glycol monoacrylateof MW= 1000 Da

PEG(200)A Polyethylene glycol monoacrylateof MW= 200 Da

PEG(1000)A-COOH Carboxyl end-capped polyethyleneglycol monoacrylate

PEGA Polyethylene glycol monoacrylate(generic)

PPP Platelets-poor plasmaPT Prothrombin time

SPION Superparamagnetic iron oxidenanoparticle

SPIONs Superparamagnetic iron oxidenanoparticles

TEM Transmission electron microscopyTT Thrombin time

Acknowledgments: This work has been funded bythe Spanish MINECO, MAT2011-25991, FEDER grantMAT2007-61621 and by Consolider Programme in Molec-ular Nanoscience Ref. CSD2007-00010. Lamiaa M. A. Aliacknowledges financial support from the Spanish Ministryof Science and Innovation FPI research grants. Techni-cal support from the University Hospital Lozano Blesa,Zaragoza, Spain, and from María Angeles Gracia, MaríaIsabel Ferrando, Concha Sanz, Ana Isabel Martínez deTernero, Carmen Morata, Jose Luis Polo.

REFERENCES1. M. De, P. S. Ghosh, and V. M. Rotello, Applications of nanoparticles

in biology. Adv. Mater. 20, 4225 (2008).2. R. Duncan and R. Gaspar, Nanomedicine(s) under the Microscope.

Molecular Pharmaceutics 8, 2101 (2011).3. Y.-W. Jun, J.-H. Lee, and J. Cheon, Chemical design of nanoparticle

probes for high-performance magnetic resonance imaging. Angew.Chem.-Int. Ed. 47, 5122 (2008).

4. A. K. Gupta and M. Gupta, Synthesis and surface engineering ofiron oxide nanoparticles for biomedical applications. Biomaterials26, 3995 (2005).

5. M. S. Fallon, M. Varshney, D. M. Dennis, and A. Chauhan,A physiologically-based pharmacokinetic model of drug detoxifica-tion by nanoparticles. Journal of Pharmacokinetics and Pharmaco-dynamics 31, 381 (2004).

6. Y. Jing, L. R. Moore, P. S. Williams, J. J. Chalmers, S. S.Farag, B. Bolwell, and M. Zborowski, Blood progenitor cell sep-aration from clinical leukapheresis product by magnetic nanopar-ticle binding and magnetophoresis. Biotechnol. Bioeng. 96, 1139(2007).

7. P. Elamanchili, M. Diwan, M. Cao, and J. Samuel, Characterizationof poly(D,L-lactic-co-glycolic acid) based nanoparticulate system forenhanced delivery of antigens to dendritic cells. Vaccine 22, 2406(2004).

8. L. Brannon-Peppas and J. O. Blanchette, Nanoparticle and tar-geted systems for cancer therapy. Advanced Drug Delivery Reviews56, 1649 (2004).

9. R. Weissleder, A. Moore, U. Mahmood, R. Bhorade, H. Benveniste,E. A. Chiocca, and J. P. Basilion, In vivo magnetic resonance imag-ing of transgene expression. Nat. Med. 6, 351 (2000).

10. H. Amiri, R. Bustamante, A. Millan, N. J. O. Silva, R. Pinol,L. Gabilondo, F. Palacio, P. Arosio, M. Corti, and A. Lascialfari,Magnetic and relaxation properties of multifunctional polymer-basednanostructured bioferrofluids as MRI contrast agents. Magnetic Res-onance in Medicine 66, 1715 (2011).

11. N. Nitin, L. E. W. LaConte, O. Zurkiya, X. Hu, and G. Bao, Func-tionalization and peptide-based delivery of magnetic nanoparticlesas an intracellular MRI contrast agent. J. Biol. Inorg. Chem. 9, 706(2004).

J. Biomed. Nanotechnol. 9, 1281–1294, 2013 1293

Hemostasis Disorders Due to Polymer Coated SPIONs Ali et al.

12. H. Lee, E. Lee, D. K. Kim, N. K. Jang, Y. Y. Jeong, andS. Jon, Antibiofouling polymer-coated superparamagnetic iron oxidenanoparticles as potential magnetic resonance contrast agentsfor in vivo cancer imaging. J. Am. Chem. Soc. 128, 7383(2006).

13. H. Lee, M. K. Yu, S. Park, S. Moon, J. J. Min, Y. Y. Jeong,H.-W. Kang, and S. Jon, Thermally cross-linked superparamag-netic iron oxide nanoparticles: Synthesis and application as a dualImaging probe for cancer in vivo. J. Am. Chem. Soc. 129, 12739(2007).

14. M. Kumagai, Y. Imai, T. Nakamura, Y. Yamasaki, M. Sekino,S. Ueno, K. Hanaoka, K. Kikuchi, T. Nagano, E. Kaneko,K. Shimokado, and K. Kataoka, Iron hydroxide nanoparticles coatedwith poly(ethylene glycol)–poly(aspartic acid) block copolymer asnovel magnetic resonance contrast agents for in vivo cancer imaging.Colloids Surf., B: Biointerfaces 56, 174 (2007).

15. A. Jordan, R. Scholz, K. Maier-Hauff, F. K. H. Landeghem,N. Waldoefner, U. Teichgraeber, J. Pinkernelle, H. Bruhn,F. Neumann, B. Thiesen, A. Deimling, and R. Felix, The effectof thermotherapy using magnetic nanoparticles on rat malignantglioma. Journal of Neuro-Oncology 78, 7 (2006).

16. C. C. Berry and A. S. G. Curtis, Functionalisation of magneticnanoparticles for applications in biomedicine. J. Phys. D: Appl. Phys.36, R198 (2003).

17. C. Boyer, M. R. Whittaker, V. Bulmus, J. Liu, and T. P. Davis,The design and utility of polymer-stabilized iron-oxide nanopar-ticles for nanomedicine applications. NPG Asia Materials 2, 23(2010).

18. S. Laurent, D. Forge, M. Port, A. Roch, C. Robic, L. V. Elst,and R. N. Muller, Magnetic iron oxide nanoparticles: synthesis,stabilization, vectorization, physicochemical characterizations, andbiological applications. Chem. Rev. 108, 2064 (2008).

19. R. Weissleder, D. D. Stark, B. L. Engelstad, B. R. Bacon, C. C.Compton, D. L. White, P. Jacobs, and J. Lewis, Superparamag-netic iron oxide—pharmacokinetics and toxicity. American Journalof Roentgenology 152, 167 (1989).

20. P. Reimer, E. J. Rummeny, H. E. Daldrup, T. Balzer, B. Tombach,T. Berns, and P. E. Peters, Clinical results with resovist—A phase-2clinical trial. Radiology 195, 489 (1995).

21. M. Johannsen, A. Jordan, R. Scholz, M. Koch, M. Lein, S. Deger,J. Roigas, K. Jung, and S. Loening, Evaluation of magnetic fluidhyperthermia in a standard rat model of prostate cancer. Journal ofEndourology 18, 495 (2004).

22. T. A. Ostomel, Q. Shi, P. K. Stoimenov, and G. D. Stucky, Metaloxide surface charge mediated Hemostasis. Langmuir 23, 11233(2007).

23. S.-Q. Li, R.-R. Zhu, H. Zhu, M. Xue, X.-Y. Sun, S.-D. Yao, and S.-L.Wang, Nanotoxicity of TiO2 nanoparticles to erythrocyte in vitro.Food and Chemical Toxicology 46, 3626 (2008).

24. N. Singh, G. J. S. Jenkins, R. Asadi, and S. H. Doak, Potential tox-icity of superparamagnetic iron oxide nanoparticles (SPION). NanoReviews 1, 5358 (2010), DOI: 10.3402/nano.v1i0.5358.

25. P. A. Gentry, Comparative aspects of blood coagulation. The Veteri-nary Journal 168, 238 (2004).

26. A. Mayer, M. Vadon, B. Rinner, A. Novak, R. Wintersteiger, andE. Fröhlich, The role of nanoparticle size in hemocompatibility. Tox-icology 258, 139 (2009).

27. N. Gulati, R. Rastogi, A. M. Dinda, R. Saxena, and V. Koul, Char-acterization and cell material interactions of PEGylated PNIPAAMnanoparticles. Colloids Surf., B-Biointerfaces 79, 164 (2010).

28. D. Simberg, W.-M. Zhang, S. Merkulov, K. McCrae, J.-H. Park,M. J. Sailor, and E. Ruoslahti, Contact activation of kallikrein–kininsystem by superparamagnetic iron oxide nanoparticles in vitro andin vivo. J. Controlled Release 140, 301 (2009).

29. A. A. Rugal, V. M. Gun’ko, V. N. Barvinchenko, V. V. Turov, T. V.Semeshkina, and V. I. Zarko, Interaction of fibrinogen with nanosil-ica. Central European Journal of Chemistry 5, 32 (2007).

30. B. W. Neun and M. A. Dobrovolskaia, Method for in vitro analysis ofnanoparticle thrombogenic properties, Characterization of Nanopar-ticles Intended for Drug Delivery, edited by S. E. McNeil, HumanaPress, New York (2011), pp. 225–235.

31. M.-T. Zhu, W.-Y. Feng, B. Wang, T.-C. Wang, Y.-Q. Gu, M. Wang,Y. Wang, H. Ouyang, Y.-L. Zhao, and Z.-F. Chai, Comparative studyof pulmonary responses to nano- and submicron-sized ferric oxidein rats. Toxicology 247, 102 (2008).

32. R. Fernández-Pacheco, C. Marquina, J. G. Valdivia, M. Gutiérrez,M. S. Romero, R. Cornudella, A. Laborda, A. Viloria, T. Higuera,A. García, J. A. G. de Jalón, and M. R. Ibarra, Magnetic nanoparti-cles for local drug delivery using magnetic implants. J. Magn. Magn.Mater. 311, 318 (2007).

33. Z. Yang, W. Zhang, J. Zou, and W. Shi, Synthesis and thermallyresponsive characteristics of dendritic poly(ether-amide) graftingwith PNIPAAm and PEG. Polymer 48, 931 (2007).

34. A. Millán, F. Palacio, G. Ibarz, and E. Natividad, Systems contain-ing magnetic nanoparticles and polymers, like nanocomposites andferrofluids, and their applications, Patent No. ES2308901B1, Spain(2007).

35. A. Millán, F. Palacio, A. Falqui, E. Snoeck, V. Serin, A. Bhattachar-jee, V. Ksenofontov, P. Gütlich, and I. Gilbert, Maghemite polymernanocomposites with modulated magnetic properties. Acta Mater.55, 2201 (2007).

36. O. Veiseh, C. Sun, J. Gunn, N. Kohler, P. Gabikian, D. Lee,N. Bhattarai, R. Ellenbogen, R. Sze, A. Hallahan, J. Olson, andM. Q. Zhang, Optical and MRI multifunctional nanoprobe for tar-geting gliomas. Nano Lett. 5, 1003 (2005).

37. D. Kim, H. El-Shall, D. Dennis, and T. Morey, Interaction of PLGAnanoparticles with human blood constituents. Colloids Surf., B:Biointerfaces 40, 83 (2005).

1294 J. Biomed. Nanotechnol. 9, 1281–1294, 2013