lable at ScienceDirect

Biomaterials 35 (2014) 3885e3894

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

Biodegradable polymeric vesicles containing magnetic nanoparticles,quantum dots and anticancer drugs for drug delivery and imaging

Fei Ye a, Åsa Barrefelt a,1, Heba Asem a,1, Manuchehr Abedi-Valugerdi a, Ibrahim El-Serafi a,Maryam Saghafian a, Khalid Abu-Salah b,**, Salman Alrokayan b, Mamoun Muhammed c,Moustapha Hassan a,d,*

aDivision of Experimental Cancer Medicine, Department of Laboratory Medicine (LABMED), Karolinska Institutet, SE-141 86 Stockholm, SwedenbKing Abdullah Institute for Nanotechnology, King Saud University, Riyadh 11451, P.O. Box 2455, Saudi ArabiacDivision of Functional Materials, School of Information and Communication Technology, Royal Institute of Technology (KTH), SE-164 40 Stockholm, SwedendClinical Research Center, Karolinska University Hospital - Huddinge, SE-141 86 Stockholm, Sweden

a r t i c l e i n f o

Article history:Received 15 December 2013Accepted 16 January 2014Available online 1 February 2014

Keywords:Biodegradable polymerMultifunctional nanoparticlesAnticancer drug deliveryBusulfanFluorescence imagingMagnetic resonance imaging

* Corresponding author. Division of Experimental Cof Laboratory Medicine (LABMED), Karolinska InstiSweden. Tel.: þ46 8 5858 3862; fax: þ46 8 5858 380** Corresponding author. Tel.: þ966 996 4675956.

E-mail addresses: abusalah@ksu.edu.sa (K. Abu-Salmoustaphah@gmail.com (M. Hassan).

1 These authors contributed equally.

0142-9612/$ e see front matter � 2014 Elsevier Ltd.http://dx.doi.org/10.1016/j.biomaterials.2014.01.041

a b s t r a c t

We have developed biodegradable polymeric vesicles as a nanocarrier system for multimodal bio-imaging and anticancer drug delivery. The poly(lactic-co-glycolic acid) (PLGA) vesicles were fabricatedby encapsulating inorganic imaging agents of superparamagnetic iron oxide nanoparticles (SPION),manganese-doped zinc sulfide (Mn:ZnS) quantum dots (QDs) and the anticancer drug busulfan into PLGAnanoparticles via an emulsion-evaporation method. T*2-weighted magnetic resonance imaging (MRI) ofPLGAeSPIONeMn:ZnS phantoms exhibited enhanced negative contrast with r*2 relaxivity of approxi-mately 523 s�1 mM�1 Fe. Murine macrophage (J774A) cellular uptake of PLGA vesicles started fluores-cence imaging at 2 h and reached maximum intensity at 24 h incubation. The drug delivery ability ofPLGA vesicles was demonstrated in vitro by release of busulfan. PLGA vesicle degradation was studiedin vitro, showing that approximately 32% was degraded into lactic and glycolic acid over a period of 5weeks. The biodistribution of PLGA vesicles was investigated in vivo by MRI in a rat model. Change ofcontrast in the liver could be visualized by MRI after 7 min and maximal signal loss detected after 4 hpost-injection of PLGA vesicles. Histological studies showed that the presence of PLGA vesicles in organswas shifted from the lungs to the liver and spleen over time.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Functional nanoparticles used in the treatment of cancer attractextensive attention due to their intrinsic physical properties, longblood circulation time, specific targeting capability, enhancedintracellular uptake andmanipulation of molecular behavior on thenanometer scale [1e4]. So far, studies on the treatment of cancerand other diseases using nanomaterials have mainly focused ontherapy by locally produced cytotoxic heat [5,6], targeted drugdelivery [7,8] or a combination of these two strategies [9e11]. Inparticular, several nano-structured particles based on polymers

ancer Medicine, Departmenttutet, SE-141 86 Stockholm,0.

ah), moustapha.hassan@ki.se,

All rights reserved.

[12], lipids [13], inorganic materials [14] or natural materials [15]have been investigated for therapeutic purposes, mainly as drugdelivery vehicles. With appropriate encapsulation, drugs are morestable in a physiological environment and the kinetics of the drugscan be more carefully controlled [16]. Furthermore, targeted drugdelivery can be developed to improve chemotherapy in cancertreatment, not only by reducing the adverse effects in non-targetorgans but also by enhancing the therapeutic efficacy in the tar-geted organ [17].

A colloidal system based on biodegradable polyester nano-particles (such as polylactic acid (PLA)), polylactic-co-glycolic acid(PLGA) and polycaprolactone (PCL) nanoparticles represents one ofthe most promising candidates for in vivo diagnosis and treatmentfor cancers, from preclinical development to clinical translation[1,18]. The use of these amphiphilic polymers results in the for-mation of nanoparticles with a hydrophobic core and a hydrophilicshell. The coreeshell structure allows them to encapsulate andcarry poorly water-soluble drugs [19] and to release these drugs at asustained rate in the optimal range of drug concentration [20]. They

F. Ye et al. / Biomaterials 35 (2014) 3885e38943886

can be further functionalized with polyethylene glycol (PEG) toavoid nonspecific absorption by proteins and fast clearance by theimmune system [21,22] as well as equipped with targeting ligandsfor delivery of drugs to specific pathological sites [12]. Anotherimportant aspect is the encapsulation of inorganic nanoparticlestogether with anticancer drugs into the core of polymeric nano-particles for localized contrast enhancement in different medicalvisualization techniques, which makes them superior in physicaland chemical properties to commercial imaging agents [23,24],thereby providing precise diagnosis and evaluation of therapeuticefficacy.

In the present investigation, we report the development of amultifunctional polymeric drug delivery system aiming to deliveranti-cancer drugs and to enable in vivo and in vitro imaging in orderto study the cell uptake as well as biodistribution of polymericnanoparticles by magnetic resonance imaging (MRI) and histopa-thology. This drug delivery system consisted of PLGA nanoparticlesencapsulating two hydrophobic inorganic nanocrystals, super-paramagnetic iron oxide nanoparticles (SPION) for in vivo MRI [25]and cadmium-free manganese-doped zinc sulfide (Mn:ZnS) quan-tum dots (QDs) for fluorescence in vitro imaging. The PLGA vesicles(i.e., PLGAeSPIONeMn:ZnS) were also loaded with the chemo-therapeutic drug busulfan [26], which is used in high doses as aconditioning agent prior to stem cell transplantation. Our aim wasto construct a drug delivery system able to efficiently entrap andrelease lipophilic anticancer drugs and track cellular uptake in vitroas well as the biodistribution in vivo via noninvasive MRI. Such avehicle is of significant value for diagnosis and therapy of cancer,where simultaneous drug delivery and therapeutic efficacy followup is needed.

2. Materials and methods

2.1. Chemicals

Poly(lactic-co-glycolic acid) (PLGA) with the brand name PURASORB� PDLG5002A (molecular weight ca. 15 kDa), terminated with carboxylic acid and having aratio of 50/50 for DL-lactide/glycolide, was obtained from Purac Biomaterials, Gor-inchem, the Netherlands. Sodium oleate, ferric chloride hexahydrate (FeCl3$6H2O),n-hexane, octyl ether, dichloromethane, manganese chloride (MnCl2), stearic acid(SA), tetramethylammonium hydroxide (TMAOH), zinc acetate dihydrate (ZnAc2),sulfur, oleylamine (OLA), octadecene (ODE), 1-dodecanethiol, and PVA were pur-chased from Sigma Aldrich, Munich, Germany and used without any furtherpurification.

2.2. Synthesis of SPION, Mn:ZnS QDs, and PLGAeSPIONeMn:ZnS nanoparticles

Monodisperse SPION were synthesized by thermal decomposition of a Fe-oleatecomplex in octyl ether at approximately 297 �C in the presence of oleic acid ac-cording to a previously reported method [27]. The Fe3O4 nanocrystals were stabi-lized with oleic acid and dispersed in dichloromethane at a concentration of 9.1 mg/mL Fe. Mn:ZnS QDs were synthesized by a nucleation-doping strategy [28]. First,manganese stearate (MnSt2) was prepared by dropwise addition of methanolicMnCl2 solution into a mixture of SA and TMAOH in methanol [29]. A mixture ofMnSt2 and 1-dodecanethiol in ODEwas then degassed at 100 �C for 15min, followedby the addition of sulfur and ZnAc2 in sequence at 250 �C. The Mn:ZnS nanoparticlesthus obtained were washed against acetone and finally re-dispersed in dichloro-methane. Dichloromethane solutions of PLGA, SPION and Mn:ZnS were mixed withPVA aqueous solution (1:20 oil to water ratio) using a probe-type sonicator to forman emulsion, which was agitated overnight to evaporate the organic solvent andwashed against de-ionized (DI) water (15 MU cm) to collect PLGAeSPIONeMn:ZnSnanoparticles. The PBS suspension of these particles was deposited on a copper gridand positively stained for TEM examination using a 2% aqueous solution of phos-photungstic acid (H3PW12O40).

2.3. Characterization of nanoparticles

The morphology and elemental composition of SPION, Mn:ZnS, and PLGAeSPIONeMn:ZnS nanoparticles were characterized by JEM-2100F field emissiontransmission electron microscope (FE-TEM) operating at an accelerating voltage of200 kV. The hydrodynamic size of the particles was measured by dynamic lightscattering (DLS) (Delsa�Nano particle size analyzer, Beckman Coulter, Brea, CA,USA). The magnetization measurements were performed using a vibrating samplemagnetometer (VSM-NUOVO MOLSPIN, Newcastle-upon-Tyne, UK). The optical

absorbance and fluorescence intensity of Mn:ZnS and PLGAeSPIONeMn:ZnSnanoparticles were measured by Lambda 900 UVeViseNIR spectrometer (PerkinElmer, Waltham, MA, USA) and LS 55 Fluorescence spectrometer (Perkin Elmer,Waltham, MA, USA), respectively. Electron paramagnetic resonance (EPR) mea-surement of Mn:ZnS was done on a Bruker ELEXSYS EPR spectrometer (X-band, 9e10 GHz) at 113 K (Bruker, Billerica, MA, USA). Concentrations of iron, manganese andzinc in samples were measured by Thermo Scientific iCAP 6500 inductively coupledplasma atomic emission spectroscopy (ICP-AES) (Thermo Fisher Scientific, KungensKurva, Sweden).

2.4. In vitro phantom magnetic resonance imaging

Phantoms (10 mL) of PLGAeSPIONeMn:ZnS nanoparticles with 0.05 mM,0.1 mM, 0.2 mM, 0.3 mM, 0.4 mM, 0.5 mM, and 1 mM of iron were made by mixingthe nanoparticle suspensionwith agarose gel (3 wt%) in DI water at 85 �C and lettingit cool down naturally overnight in 50mL Eppendorf centrifuge tubes. The phantomswere placed in the extremity coil of a 3 T MRI scanner (Siemens Trio, Siemens,Erlangen, Germany). A gradient echo T*

2 sequencewith a fixed repetition time (TR) of2000 ms and 12 TEs of 2e22.9 ms was used for MR imaging to obtain T*

2-weightedimages. Circular ROIs (region of interest) were placed manually on the images andthe negative logarithmic values of the signal intensities at different TEs were plottedversus the respective TE values. The T*2 relaxation time was calculated as the slope ofa semi-log plot of the signal intensities versus the TEs. In phantoms with a highconcentration of iron oxide, the calculations were based on fewer TEs excludingthose TEs where full transaxial relaxation had already occurred.

2.5. In vitro cellular uptake and fluorescence imaging

To evaluate the effects of cellular uptake for PLGAeSPIONeMn:ZnS nano-particles, we used the murine J774A macrophage cell line (European Type TissueCulture Collection, CAMR, Salisbury, UK). These cells were obtained as a kind giftfrom Professor Carmen Fernandez, Department of Immunology, Wenner-GrenInstitute, Stockholm University, Stockholm, Sweden. First, the J774A cells werecultured in DMEM supplemented with 10% heat-inactivated fetal bovine serum,penicillin (100 U/mL) and streptomycin (100 mg/mL) (Invitrogen�, Life Technologies,Carlsbad, CA, USA) in a 50 cm2 tissue culture flask (Costar, Corning, NY, USA). Thecultures were maintained at 37 �C in a humidified atmosphere containing 5% carbondioxide. J774A cells were then cultured in 8-chamber polystyrene vessel tissueculture treated glasses at a density of 5 � 105 cells/chamber, at 37 �C, for 12 h in anatmosphere containing 5% carbon dioxide to allow cell attachment. Thereafter, thecell culture medium was aspirated from each chamber and substituted with themedium alone (negative control) or the same medium containing PLGAeSPIONeMn:ZnS nanoparticles at concentrations of 1000, 100, 50, 25, 12.5, 6.25 and/or 3 mg/mL. Chambers were then incubated at 37 �C for 1, 2, 4 and 24 h in an atmospherecontaining 5% carbon dioxide. The uptake experiment was terminated at each timepoint by aspirating the test samples, removing the chamber and washing the cellmonolayers with ice-cold PBS three times. Each slide was then fixed with meth-anoleacetone (1:1, v/v), followed by examination under a Nikon Eclipse i80 fluo-rescence microscope (Nikon, Tokyo, Japan) at a wavelength of 520 nm.

2.6. In vitro drug release

For in vitro busulfan release experiments, 30 mg busulfan was dissolved indichloromethane solution containing PLGA, SPION and Mn:ZnS QDs, and thenemulsified with PVA at a total volume of 6 mL. After evaporation of organic solventand centrifugation to wash off unloaded drugs, PLGA vesicles containing drugs weretransferred into a cellulose permeable membrane bag with a molecular weight cut-off (MWCO) of 12e14 kDa to dialyze against PBS solution at 37 �C. Entrapment ef-ficiency of busulfan in PLGAeSPIONeMn:ZnS nanoparticles was calculated as [(massof the total drug � mass of free drug) � 100%/mass of total drug]. Three parallelrelease experiments were conducted and samples were taken at specific time points.Concentrations of busulfan released in dialysis media, left in dialysis bag or left incentrifuged supernatant were measured by gas chromatography (SCION 436-GC;Bruker, Billerica, MA, USA) with electron capture detector (ECD) according to amethod reported previously by Hassan et al. [30]. The release percentage of loadedbusulfan is averaged from the three parallel experiments with error bars repre-senting standard deviation.

2.7. In vitro degradation of PLGA vesicles

The synthesized PLGA vesicles were placed in a cellulose permeable membranebag (MWCO 12e14 kDa) and dialyzed against 1 L PBS (pH 7.4) at 37 �C. At pre-determined time intervals, a 5 mL aliquot of PBS solution was withdrawn and freshPBS was added into the dialysis solution. The concentration of lactic acid released inPBS was measured by high-performance liquid chromatography (HPLC). The HPLCsystem consisted of a Gilson autoinjector (100 mL loop), an LKB HPLC pump 2150(Pharmacia Inc., Sweden), an LDC analytical spectromonitor 3200 UV detector(Riviera Beach, FL, USA) and a CSW32 chromatography station integrator. Separationwas performed on a Zorbax SB-CN column (4.6 mm � 150 mm; 5 mm) from AgilentTechnologies (Santa Clara, CA, USA), and the column was maintained at roomtemperature during analysis. The mobile phase was composed of NH4H2PO4 (0.1 M,

Fig. 1. Schematic diagram showing the composition of PLGAeSPIONeMn:ZnS nano-particles and their multiple applications.

F. Ye et al. / Biomaterials 35 (2014) 3885e3894 3887

pH 2.5) and flow rate was set at 0.8 mL/min with a running time of 6 min. Injectionvolume was 50 mL. Detection of lactic acid was performed at 210 nm compared to acalibration curve for lactic and glycolic acids. Lactic acid eluted after 4.0 min.

2.8. In vivo magnetic resonance imaging

The animal study was approved by the Stockholm Southern Animal ResearchEthics Committee and was performed in accordance with Swedish Animal Welfarelaw. Two male Sprague Dawley rats (500 g, Charles River Laboratories, Germany)were anaesthetized using an intraperitoneal (I.P.) injection of 60 mg/kg sodiumpentobarbital (APL, Kungens Kurva, Sweden). The rats were put head first in anextremity coil and imaged in the coronal plane at 3 T in a Siemens Trio MRI scanner(Siemens, Erlangen, Germany), pre- and post-intravenous (I.V.) injections of 0.85 mLof PLGAeSPIONeMn:ZnS nanoparticles suspended in the aforementioned DMEM-fetal bovine serum-penicillin-streptomycin media ([Fe] ¼ 0.25 mg/mL). Imagingwas performed using a T*

2-weighted sequence before and after injection at 4, 8, 12and 30 min as well as at 1, 2, 3, 4 and 24 h. The images were evaluated using a PACSworkstation (Sectra, Linköping, Sweden) and R*2 ð1=T*2Þ was determined for liver,spleen, brain and kidneys by measuring the signal intensity in circular ROIs on theimages at different TEs and then calculating T*

2 using several TEs.

2.9. Histological examination and tissue distribution

Rats were injected intravenously with 0.85 mL of PLGAeSPIONeMn:ZnSnanoparticles suspended in the aforementioned media ([Fe] ¼ 0.25 mg/mL). Theanimals were sacrificed at 30 min, 4 h and 24 h post-injection by injecting anoverdose of pentobarbital. Various organs such as liver, spleen, and lungs weredissected and fixed in paraformaldehyde (PFA, 4%) for 48 h followed by ethanol

Fig. 2. Field emission transmission electron microscope (FE-TEM) images of PLGA nanopartiresolution image of a single (c) SPION and (d) Mn:ZnS QD; (e) TEM image of positively sta

(70%). Further dehydration and paraffinization of the selected tissuewere performedin a vacuum infiltration processor before they were embedded in paraffin accordingto RENI trimming guidelines [31]. Sections (4 mm) were mounted on superfrost glassslides and stained with Perls’ Prussian blue staining for iron. In addition, immuno-histochemistry was performed with a primary antibody against CD68 macrophagemarker (MCA 341, AbD Serotec, Kidlington, UK) in the lung, liver, and spleen sec-tions. Biotinylated rabbit anti-mouse IgG (E0354, Dako, Stockholm, Sweden) wasused as a secondary antibody. The sections were then treated by DAB peroxidasesubstrate (SK-4100, Vector, Burlingame, CA, USA). CD68 brown signal was developedusing VECTASTAIN� ABC kit (PK6100, Vector, Burlingame, CA, USA). To producecounterstaining, immunohistochemistry was performed followed by additionalPrussian blue staining. The slides were cover slipped and mounted with Pertexmounting medium (177070, Histolab, Spånga, Sweden).

3. Results

3.1. Synthesis and characterization of PLGAeSPIONeMn:ZnSnanoparticles

PLGAeSPIONeMn:ZnS nanoparticles were synthesized by anemulsioneevaporation process, and the composition of the parti-cles is shown in Fig. 1. First, hydrophobic SPION and Mn:ZnS QDs tofunction as imaging agents were fabricated separately. Next, PLGAvesicles entrapping the payloads of imaging agents and busulfanwere prepared by an oil-in-water (O/W) emulsionmethod followedby solvent evaporation of the volatile organic phase at room tem-perature. Specifically, the SPION, Mn:ZnS QDs and busulfan wereincorporated into the hydrophobic domain of PLGA molecules viahydrophobic interaction, and the PLGA vesicles were then formedin the presence of polyvinyl alcohol (PVA) emulsifier. After evapo-ration of the organic solvent in the emulsion, the PLGA vesiclesentrapping SPION and Mn:ZnS QDs with busulfan were washedusing de-ionized water and re-dispersed in phosphate buffer so-lution (PBS) or in Dulbecco’s modified Eagle medium (DMEM).

The morphology and crystal structure of SPION, Mn:ZnS, andPLGA vesicles were examined by transmission electron microscopy(TEM). According to the TEM images of nanoparticles (Fig. 2a, b),the average diameter was 10.7 nm (standard deviation s z 8%) forSPION, 3.1 nm (s z 10%) for Mn:ZnS QDs and 93 nm (s z 20%) forPLGAeSPIONeMn:ZnS vesicles. High resolution TEM images(Fig. 2c, d) show the single crystalline nature of SPION and Mn:ZnSQDs, respectively. The loading of these particles in PLGA vesiclescan be seen clearly in Fig. 2e. Images of unstained (Fig. S1a) andlarge area (Fig. S1b) PLGA vesicles as well as the EDS spectrum(Fig. S2) can be found in Supplementary Data.

cles. (a) SPION and (b) Mn:ZnS QDs with inset of electron diffraction patterns, and highined PLGAeSPIONeMn:ZnS nanoparticles.

Fig. 3. The magnetic property of SPION and the elemental analysis of PLGAeSPIONeMn:ZnS. (a) Field-dependent magnetization measurement of SPION at room temperature. (b)EPR spectra of Mn:ZnS QDs. (c) Line scan for elemental analysis and (d) UVeVis absorbance and fluorescence spectra of PLGAeSPIONeMn:ZnS nanoparticles.

F. Ye et al. / Biomaterials 35 (2014) 3885e38943888

The magnetic property of SPION was examined on field-dependent magnetization and no hysteresis appeared (Fig. 3a),demonstrating the superparamagnetic property desired for T2 andT*2 MRI application. The optical property of Mn:ZnS QDs was alsostudied. Fig. S3a shows theUVeVisible absorbance andfluorescencespectra. The characteristic peak of fluorescence emission at 594 nmis due to doping Mn2þ ions into a ZnS lattice (3 wt% of Mn to Znaccording to ICP results). In order to understand the location of Mnions in the ZnS host matrix, we recorded the electron paramagneticresonance (EPR) spectrum of Mn:ZnS nanocrystals (Fig. 3b). Theobserved six-line spectrum resulted from the hyperfine interactionof the unpaired electrons with 55Mn nuclear spin (I ¼ 5/2), whichprovided evidence for well-dispersed doping of Mn2þ ions withoutany clustering. The hyperfine coupling constant of Mn:ZnS,A¼68.8G, determined fromthe EPR spectrum, is in close agreementwith the literature data forMn2þ ions in the tetrahedral sites of cubiczinc blende lattice [32,33]. The elemental composition of PLGAeSPIONeMn:ZnS vesicles was analyzed by energy dispersive spec-troscopy (EDS) and inductively coupled plasma (ICP) techniques. Atypical EDS line scan on one PLGAvesicle (Fig. 3c) shows the spectralcounts corresponding to the elements of Fe, O, Zn, S, and Mn.Originated from the loaded Mn:ZnS QDs, the fluorescence emissionpeak of PLGAeSPIONeMn:ZnS is centered at 599 nm (Fig. 3d)whichis located in the observablewindowof fluorescencemicroscopy andfacilitates further imaging application.

Fig. 4. Magnetic resonance imaging of phantoms. (a) T*2-weighted MR phantom im-

ages of PLGAeSPIONeMn:ZnS at different TEs (TR ¼ 1200 ms; TE ¼ 2 ms, 3.9 ms,5.8 ms). (b) Proton transverse relaxation rate ðR*2 ¼ 1=T*

2Þ of phantom samples versusiron concentration.

3.2. In vitro phantom magnetic resonance imaging

An aqueous suspension of PLGAeSPIONeMn:ZnS vesicles wasused to prepare phantoms for MRI measurement. The hydrody-namic size distribution of these vesicles was studied and the results

Fig. 5. Superimposed light and fluorescence microscopy images of Quantum Dots (QD) labeled PLGAeSPIONeMn:ZnS nanoparticles in the J774A cells. (a, b) untreated control J774Aat 4 h and 24 h, (c, d) J774A incubated with PLGAeSPIONeMn:ZnS nanoparticles for 4 h and 24 h (e, f) The corresponding fluorescence imaging of J774A incubated with PLGAeSPIONeMn:ZnS-QD nanoparticles for 4 h and 24 h.

F. Ye et al. / Biomaterials 35 (2014) 3885e3894 3889

are shown in Fig. S4. The MRI contrasting effect of PLGAeSPIONeMn:ZnS was then evaluated by T*

2-weighted MR images using a 3 Tinstrument. With increasing concentrations of PLGA vesicles, thesignal intensity of MRI decreased owing to the increase in loadedSPION (Fig. 4a). For PLGA vesicle phantoms with the same con-centrations of iron oxide, the increase in echo time (TE; in Fig. 4aand Fig. S5a) and flip angle (Fig. S5b) also induce decreased signalintensity for MRI. It is noted that the signal intensity decreasesmore rapidly with increasing TE for phantoms containing higherconcentrations of iron oxide. These characteristics allow theapplication of PLGAeSPIONeMn:ZnS as a negative contrast agentfor MRI. In Fig. 4b, the transverse relaxivity (r*2) is obtained as theslope of linear fitting for the relaxation rate at different iron con-centrations, which is 523 s�1 mM�1 Fe for PLGAeSPIONeMn:ZnSvesicles.

3.3. In vitro cellular uptake and fluorescence imaging

Fig. S6(aec) in Supplementary Data shows the fluorescenceimages of Mn:ZnS QDs with different emissionwavelengths using aset of filters. We then evaluated the cell uptake effect of PLGAe

SPIONeMn:ZnS nanoparticles in the J774Amurine macrophage cellline, in virtue of the intrinsic fluorescence property of loadedMn:ZnS QDs. Fig. 5a, b shows the superimposed optical and fluo-rescence imaging of non-treated J774A cells at 4 h and 24 h incu-bation, respectively. In comparison, the cells treated with PLGAeSPIONeMn:ZnS nanoparticles exhibited much stronger fluores-cence intensity in the cell plasma area at the same incubation timeof 4 h (Fig. 5c) and 24 h (Fig. 5d). The corresponding fluorescenceimaging of the treated cells seen in Fig. 5e, f shows that the uptakeof PLGA vesicles can greatly enhance visualization of the cellscompared with non-treated cells with very low fluorescence in-tensity (data not shown). Maximum fluorescence intensity fortreated cells was achieved after 24 h incubation due to continuouslocalization of PLGA vesicles in the cells, demonstrating their highuptake efficiency.

3.4. In vitro drug release and degradation of PLGA vesicles

The drug release kinetics of busulfan-loaded PLGAeSPIONeMn:ZnS vesicles was studied at pH 7.4 using a dialysis method; therelease profile versus time is demonstrated in Fig. 6. The initial

Fig. 6. Profile of busulfan release from PLGAeSPIONeMn:ZnS nanoparticles in PBSsolution at pH 7.4. Data represent average values for n ¼ 3, and the error bars indicatestandard deviation.

F. Ye et al. / Biomaterials 35 (2014) 3885e38943890

concentration of busulfan in a mixture of dichloromethane and PBSsolutionwas 5 mg/mL, and the entrapment efficiency of busulfan inPLGA vesicles was calculated to be 89 � 2%. The percentage ofrelease, represented on the vertical axis, was calculated by dividingthe amount of busulfan diffused into dialysis media by the totalamount of busulfan loaded into the PLGA vesicles. We found thataround 70e80% of busulfan was released after 5 h of dialysis. Thedegradation of the drug carrier was also tested and Fig. 7 shows thepercentage of lactic acid degraded from PLGA vesicles at pH 7.4 and37 �C for a period of 5 weeks. At 2 weeks, about 12% lactic acid wasdegraded and released. Thedegradationwas found to followaquasi-linear trend and around 32% lactic acid was degraded at 5 weeks.

3.5. In vivo magnetic resonance imaging

The in vivoMRI of PLGA vesicles was tested in a rat model. Fig. 8shows a coronal view of liver slices, where the liver appears whiteon T*

2-weighted images before injection of PLGA vesicles. There wasa rapid signal decrease in the liver and spleen post-injection. Justafter 7 min post-injection the liver imaging becomes darker, indi-cating the fast accumulation of iron-containing PLGA vesicles in the

Fig. 7. Release profile of lactic acid by degradation of PLGAeSPIONeMn:ZnS nano-particles in PBS solution during a period of 5 weeks (pH 7.4, T ¼ 37 �C). Data werecollected from three parallel experiment and error bars indicate standard deviation.

liver. The signal continues to decrease and the imaged liver be-comes very dark, especially on the images taken with longer TEs.Minimal signal intensity appears at 4 h post-injection and remainslow until 24 h post-injection. In Fig. 9, the relaxation rate of theliver is plotted against post-injection time. It quickly increases from64 s�1 pre-injection to 152 s�18 min post-injection, with its highestvalue being 161 s�1 4 h post-injection, after which it slowly de-creases to 137 s�1 24 h post-injection. Fig. 10 shows theT*2-weighted MR images of kidney slices. In these, a smaller signaldecrease can be observed over the 24 h post-injection periodcompared with liver slices. However, no change in signal intensitywas calculated in the brain and testes.

3.6. Histological examination and tissue distribution

The pattern of PLGA vesicle biodistribution and uptake werestudied histologically in the lung (Fig. 11aec) and liver (Fig. 11def).We employed Prussian blue staining todetect PLGAvesicles throughtheir iron content. As shown in Fig.11a andd, clusteredPLGAvesicleswere observed in the lungs and liver respectively as early as 30 minpost-injection. The number of clustered PLGAvesicles in pulmonarymacrophages reached theirmaximumat 4hpost-injection (Fig.11b)and then decreased until 24 h post-injection, when PLGA vesiclescould rarely be observed (Fig. 11c). PLGA vesicles were also found inliver tissue at 30 min post-injection (Fig. 11d), but the amount ofvesicles remained similar at 4 h and 24 h post-injection with only aminor decline (Fig. 11e and f, respectively). We confirmed thephagocytosis of PLGA vesicles by macrophages through applyingimmunohistological staining for macrophages followed by addi-tional Prussian blue staining in liver (Fig. 12aec) and spleen(Fig. 12def) sections. In the liver, clusters of PLGA vesicles werefound in the sinusoids at 30 min post-injection, mostly associatedwith macrophages (Fig. 12a). At 4 h post-injection (Fig. 12b), clusterfrequency was similar to that seen at 30 min post-injection. Mean-while, the PLGA vesicle clusters were clearly less frequent at 24 hpost-injection, as seen in Fig. 12c. In the spleen, fewer PLGA vesicleswere observed from 30 min post-injection (Fig. 12d) and increasedin number at 4 h to reach amaximumat 24 h post-injection (Fig.12eand f, respectively). The frequency of vesicles at 4 h post-injection inthe spleen was higher compared to that seen at 30 min post-injection; however, the number of clusters increased significantlyat 24 h post-injection, which indicates an increased amount of PLGAvesicles accumulated in the spleen over time.

4. Discussion

The aim of the present investigation was to construct a biode-gradable drug delivery system that is easy to image using MRI inorder to study its biodistribution in vivo. The nano-carrier was alsoconstructed to be utilized in cellular uptake as well as for histo-pathological studies, as illustrated in Fig. 1.

To optimize the particle size and morphology of PLGA vesiclesfor drug delivery application, we investigated the influence oforganic solvents (oil phase) and surfactants (emulsifier) on for-mation of these vesicles during an emulsion-evaporation process.We found that a water-immiscible solvent (e.g., dichloromethane)is more effective than water-miscible (e.g., tetrahydrofuran,ethanol, acetone) or partially miscible (e.g., ethyl acetate) solventsin forming separated and spherical polymeric vesicles. Non-ionicsurfactant (e.g., PVA) was found to be superior to ionic surfac-tants (e.g., cetyltrimethylammonium bromide, sodium dodecylsulfate) in forming high quality monodisperse PLGAeSPIONeMn:ZnS nanoparticles in a water-dichloromethane emulsion sys-tem; however, the previously reported method [34] using PluronicF127 was not reproduced in the current work.

Fig. 8. T*2-weighted in vivo MR images of rat before and after intravenous injection of PLGAeSPIONeMn:ZnS nanoparticles at a dose of 0.42 mg Fe/kg. Images were taken pre-

injection and at 7 min, 2 h 40 min, 4 h, and 24 h post-injection with different TEs (2 ms, 3.9 ms, and 5.8 ms). The regions of liver in the coronal planes are encircled by whitedashed lines.

F. Ye et al. / Biomaterials 35 (2014) 3885e3894 3891

For manganese-doped QDs, once Mn2þ ions are doped inside asemiconductor host the strong electronic interaction betweend states of Mn2þ and sep states of the host generates intermediateenergy states, through which the host generated exciton relaxes(4T1/6A1) resulting in emission centered at 595 nm [35,36]. Basedon the hyperfine coupling constant of Mn:ZnS QDs, if Mn2þ ions

Fig. 9. Variation of relaxation rate (R*2, s�1) of PLGAeSPIONeMn:ZnS nanoparticles in

the liver at different time points pre- and post-injection.

locate on the surface of a host particle a much higher value of hy-perfine splitting is expected [37], which is not valid here. However,additional peaks at the lower and higher magnetic field wereobserved (markedwith arrows in Fig. 3b) indicating the presence ofa small fraction of Mn2þ ions on the surface of ZnS nanoparticles.

The encapsulated SPION are responsible for MRI contrastenhancement. Theyarewell known for their capability to shorten thetransverse (spinespin) relaxation time, resulting in a decrease inMRsignal intensity [38]. The transverse relaxivity (r*2), i.e. the changes inrelaxation rate ðR*2 ¼ 1=T*

2Þperunit concentration, is theassessmentfor efficacy of MRI T*2 contrast enhancement. The value of transverserelaxivity for PLGAeSPIONeMn:ZnS vesicles, 523 s�1 mM�1 Fe, ismuch higher than that seen inprevious reports for single SPION [39],clusters [34] or assembly [40] of SPION, indicating the high efficiencyofPLGAeSPIONeMn:ZnS forMRT*

2 imaging. This canbeattributed tothe synergetic effect obtained when SPION are in intimate contact,thereby enhancing the local magnetic field.

To investigate the interaction between PLGA vesicles and cellsby fluorescence imaging, we used nontoxic zinc sulfide QDs. Bydoping a small amount of manganese into ZnS QDs, the fluores-cence emission peak was tuned from violet blue (approximately400 nm) to the red end of the visible spectrum (approximately595 nm), fitting with the observable range of emission filters forfluorescence microscopy. The increase in fluorescence intensity ofPLGA vesicle-treated cells until 24 h incubation provides the pos-sibility of long-term imaging for tracking and observation purposes.These results indicate the capacity of PLGA vesicles for delivery ofchemotherapeutic agents into the cells as well as their usefulness asan optical imaging agent.

Fig. 10. T*2-weighted in vivo MRI of coronal slices taken at pre-injection and at 7 min, 45 min, 1 h, 3 h, 4 h, and 24 h post-injection of PLGAeSPIONeMn:ZnS nanoparticles. The two

kidneys are encircled by white dashed lines.

F. Ye et al. / Biomaterials 35 (2014) 3885e38943892

Simple entrapment of payload in PLGA vesicles allows loadingand delivery of lipophilic drugs for specific cancer treatment. In thisstudy, we have demonstrated the validity of using PLGA vesicles forloading and release of the anticancer drug, busulfan. Currently, itsmain application is for conditioning prior to bone marrow trans-plantation [26,41]. Unlike other systems, where the drug releaseshows a burst model at the beginning of the process, our resultsshowed a relatively constant release rate during the first 2 h, afterwhich drug release continued at a lower rate for 5e6 h to release atotal of 70e80% of the loaded drugs. This behavior might be relatedto the amphiphilic nature of PLGA nanoparticles and coating effectof PVA layers. Preferably, a steady zero-order release profile needsto be built by employing surface coating or other precise controls ofthe drug release rate. This would prevent side effects due to a drug

Fig. 11. Tissue distribution of nanoparticles in rat (aec) lungs and (def) liver visualized by Pee) 4 h, and (c, f) 24 h after intravenous injection of PLGAeSPIONeMn:ZnS nanoparticles. T

release burst from polymers as well as maintain the drug plasmalevel during conditioning to ensure success of the followingengraftment. To investigate the stability of PLGA vesicles used as adrug carrier, we tested the degradation of PLGA vesicles bymeasuring the concentration of lactic acid, one of the degradationproducts. This relatively slow degradation behavior of PLGAvesiclessuggests their potential for use in long-term drug release.

We further investigated the effects of PLGAeSPIONeMn:ZnSvesicles on in vivo MR imaging using a rat model. T*

2-weighted MRimaging of the rat was performed at different time points afterintravenous tail injection of PLGA vesicles. In a comparison ofcontrast efficiency, our PLGA vesicles provided higher relaxationrate by using a dose of iron 20 times lower than the previouslyreported dextran-coated SPION [42]. The rapid decrease of signal

rls’ Prussian blue iron staining. Histological sections were extracted at (a, d) 30 min, (b,he scale bars represent 20 mm.

Fig. 12. Combination of Immunohistochemical (IHC) and Prussian blue staining on liver and spleen. (aec) liver and (def) spleen sections removed at (a, d) 30 min, (b, e) 4 h, and (c,f) 24 h post-injection of PLGAeSPIONeMn:ZnS nanoparticles. The scale bars represent 20 mm.

F. Ye et al. / Biomaterials 35 (2014) 3885e3894 3893

intensity or faster T*2 relaxation in the liver compared to other or-gans is due to the PLGAeSPIONeMn:ZnS nanoparticles being takenup by hepatic macrophages [43]. Other organs had fewer macro-phage cells able to take up the nanoparticles, and the signal in-tensity therefore exhibited less or no change.

We used the histological method to study the biodistribution ofPLGA vesicles in the rat in order to understand the biological fate ofthese vesicles in lung, liver and spleen. The result of histologicalstudies on lung and liver is consistent with that of in vivo MRI forliver slices, where the signal decreased rapidly at the beginning andreached its lowest level at 4 h post-injection due to a large amount ofPLGA accumulated in the liver. Alternatively, the biodistribution anduptake of PLGA vesicles can also be studied in macrophages treatedwith immunohistological staining. The distribution of vesicles wasfound to gradually shift from liver to spleen, which is also seen inprevious reports on nanoparticle biodistribution. Panagi et al. havereported that the dose affects the biodistribution of PLGA particlesbetween blood and the mononuclear phagocyte system, while nodose effect on the biodistribution was observed when stealth pol-y(Lactide-co-glycolide)-monomethoxypoly(ethyleneglycol) (PLGA-mPEG) was used [44]. In the present investigation, the PLGA parti-cles were rapidly cleared from the lungs to the liver and within 24 hwere found in spleen. No sign of higher uptake into the lungs ordistribution to the brain tissues was observed as was reported withPLGAePEGePLGA [22]. These results show that the present con-struction of PLGA differs from other nanoparticles and suggest thatthe present vesiclesmight be used as drug carriers in therapy aswellas for monitoring the cellular uptake in order to follow treatmentefficacy.

5. Conclusions

In summary, we have synthesized PLGAeSPIONeMn:ZnS vesi-cles as a multifunctional drug delivery system consisting of abiodegradable polymeric shell containing a payload of multipleimaging agents and an anti-cancer drug. The MRI phantom of PLGAvesicles exhibits high r*2 relaxivity and greatly enhancedT*2-weighted MR imaging contrast. J774A macrophage cells showed

high uptake of PLGA vesicles labeled with quantum dots, and theuptake was improved over time. The quantum dots enhanced the

fluorescence visualization of cell uptake. PLGA vesicles have alsobeen demonstrated to have high entrapment efficiency for thelipophilic drug busulfan as well as sustained drug release. Thein vivo MRI and histological studies in a rat model elucidated thebiodistribution of PLGA vesicles. The biodistribution to differenttissues was confirmed using immunohistochemistry and Prussianblue staining. Altogether, these results strongly suggest furtherdevelopment of PLGA as a multifunctional diagnostic and thera-peutic tool for cancer treatment. Further investigation of a PLGAdrug delivery system for cancer treatment is ongoing, includingorgan targeting, controlling of drug release and in vivo evaluation ofthe treatment efficacy of chemotherapy in tumor models.

Acknowledgments

The authors thank Dr. T. Astlind (Dept. of Biophysics, Arrheniuslaboratories, StockholmUniversity) for EPRmeasurement. Financialsupport from The Swedish Cancer Society, The Swedish ChildhoodCancer Foundation and the NPST program of King Saud University(Project No: 11-NAN1462-02) is also acknowledged.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.biomaterials.2014.01.041

References

[1] Hrkach J, Von Hoff D, Ali MM, Andrianova E, Auer J, Campbell T, et al. Pre-clinical development and clinical translation of a PSMA-targeted docetaxelnanoparticle with a differentiated pharmacological profile. Sci Transl Med2012;4:128ra39.

[2] Kam NWS, O’Connell M, Wisdom JA, Dai H. Carbon nanotubes as multifunc-tional biological transporters and near-infrared agents for selective cancer celldestruction. Proc Natl Acad Sci U S A 2005;102:11600e5.

[3] Medina SH, El-Sayed MEH. Dendrimers as carriers for delivery of chemo-therapeutic agents. Chem Rev 2009;109:3141e57.

[4] Yang F, Jin C, Subedi S, Lee CL, Wang Q, Jiang Y, et al. Emerging inorganicnanomaterials for pancreatic cancer diagnosis and treatment. Cancer TreatRev 2012;38:566e79.

[5] Huang X, El-Sayed IH, Qian W, El-Sayed MA. Cancer cell imaging and photo-thermal therapy in the near-infrared region by using gold nanorods. J AmChem Soc 2006;128:2115e20.

F. Ye et al. / Biomaterials 35 (2014) 3885e38943894

[6] Kawano T, Niidome Y, Mori T, Katayama Y, Niidome T. PNIPAM gel-coatedgold nanorods for targeted delivery responding to a near-infrared laser. Bio-conjug Chem 2009;20:209e12.

[7] Scott RC, Crabbe D, Krynska B, Ansari R, Kiani MF. Aiming for the heart: tar-geted delivery of drugs to diseased cardiac tissue. Expert Opin Drug Deliv2008;5:459e70.

[8] Shi M, Ho K, Keating A, Shoichet MS. Doxorubicin-conjugated immuno-nanoparticles for intracellular anticancer drug delivery. Adv Funct Mater2009;19:1689e96.

[9] Chen R, Zheng X, Qian H, Wang X, Wang J, Jiang X. Combined near-IR pho-tothermal therapy and chemotherapy using gold-nanorod/chitosan hybridnanospheres to enhance the antitumor effect. Biomater Sci 2013;1:285e93.

[10] Kumar CSSR, Mohammad F. Magnetic nanomaterials for hyperthermia-basedtherapy and controlled drug delivery. Adv Drug Deliv Rev 2011;63:789e808.

[11] You J, Zhang R, Zhang G, Zhong M, Liu Y, Van Pelt CS, et al. Photothermal-chemotherapy with doxorubicin-loaded hollow gold nanospheres: a plat-form for near-infrared light-trigged drug release. J Control Release2012;158:319e28.

[12] Nicolas J, Mura S, Brambilla D, Mackiewicz N, Couvreur P. Design, function-alization strategies and biomedical applications of targeted biodegradable/biocompatible polymer-based nanocarriers for drug delivery. Chem Soc Rev2013;42:1147e235.

[13] Arias JL, Clares B, Morales ME, Gallardo V, Ruiz MA. Lipid-based drug deliverysystems for cancer treatment. Curr Drug Targets 2011;12:1151e65.

[14] Vallet-Regí M, Balas F, Arcos D. Mesoporous materials for drug delivery.Angew Chem Int Ed 2007;46:7548e58.

[15] Huang S, Fu X. Naturally derived materials-based cell and drug delivery sys-tems in skin regeneration. J Control Release 2010;142:149e59.

[16] Lasic DD. Doxorubicin in sterically stabilized liposomes. Nature 1996;380:561e2.

[17] Brannon-Peppas L, Blanchette JO. Nanoparticle and targeted systems forcancer therapy. Adv Drug Deliv Rev 2004;56:1649e59.

[18] Danhier F, Ansorena E, Silva JM, Coco R, Le Breton A, Préat V. PLGA-basednanoparticles: an overview of biomedical applications. J Control Release2012;161:505e22.

[19] Hevus I, Modgil A, Daniels J, Kohut A, Sun C, Stafslien S, et al. Invertiblemicellar polymer assemblies for delivery of poorly water-soluble drugs. Bio-macromolecules 2012;13:2537e45.

[20] Seremeta KP, Chiappetta DA, Sosnik A. Poly(ε-caprolactone), Eudragit� RS 100and poly(ε-caprolactone)/Eudragit� RS 100 blend submicron particles for thesustained release of the antiretroviral efavirenz. Colloids Surf B Biointerfaces2013;102:441e9.

[21] Gu B, Xie C, Zhu J, He W, Lu W. Folate-PEG-CKK2-DTPA, a potential carrier forlymph-metastasized tumor targeting. Pharm Res 2010;27:933e42.

[22] Song Z, Feng R, Sun M, Guo C, Gao Y, Li L, et al. Curcumin-loaded PLGA-PEG-PLGA triblock copolymeric micelles: preparation, pharmacokinetics and dis-tribution in vivo. J Colloid Interface Sci 2011;354:116e23.

[23] Jun YW, Lee JH, Cheon J. Chemical design of nanoparticle probes for high-performance magnetic resonance imaging. Angew Chem Int Ed 2008;47:5122e35.

[24] Resch-Genger U, Grabolle M, Cavaliere-Jaricot S, Nitschke R, Nann T. Quantumdots versus organic dyes as fluorescent labels. Nat Meth 2008;5:763e75.

[25] Lee D. Highly effective T2 MR contrast agent based on heparinized super-paramagnetic iron oxide nanoparticles. Macromol Res 2011;19:843e7.

[26] Hassan M. The role of busulfan in bone marrow transplantation. Med Oncol1999;16:166e76.

[27] Park J, An K, Hwang Y, Park JG, Noh HJ, Kim JY, et al. Ultra-large-scale syn-theses of monodisperse nanocrystals. Nat Mater 2004;3:891e5.

[28] Zhang W, Li Y, Zhang H, Zhou X, Zhong X. Facile synthesis of highly lumi-nescent Mn-doped ZnS nanocrystals. Inorg Chem 2011;50:10432e8.

[29] Pradhan N, Peng X. Efficient and color-tunable Mn-doped ZnSe nanocrystalemitters: control of optical performance via greener synthetic chemistry.J Am Chem Soc 2007;129:3339e47.

[30] Hassan M, Ehrsson H. Gas chromatographic determination of busulfan inplasma with electron-capture detection. J Chromatogr B Biomed Sci Appl1983;277:374e80.

[31] Morawietz G, Ruehl-Fehlert C, Kittel B, Bube A, Keane K, Halm S, et al. Revisedguides for organ sampling and trimming in rats and mice e Part 3. A jointpublication of the RITA and NACAD groups. Exp Toxicol Pathol 2004;55:433e49.

[32] Chin PTK, Stouwdam JW, Janssen RAJ. Highly luminescent ultranarrow Mndoped ZnSe nanowires. Nano Lett 2009;9:745e50.

[33] Srivastava BB, Jana S, Karan NS, Paria S, Jana NR, Sarma DD, et al. Highlyluminescent Mn-doped ZnS nanocrystals: gram-scale synthesis. J Phys ChemLett 2010;1:1454e8.

[34] Kim J, Lee JE, Lee SH, Yu JH, Lee JH, Park TG, et al. Designed fabrication of amultifunctional polymer nanomedical platform for simultaneous cancer-targeted imaging and magnetically guided drug delivery. Adv Mater2008;20:478e83.

[35] Bhargava RN, Gallagher D, Hong X, Nurmikko A. Optical properties ofmanganese-doped nanocrystals of ZnS. Phys Rev Lett 1994;72:416e9.

[36] Erwin SC, Zu L, Haftel MI, Efros AL, Kennedy TA, Norris DJ. Doping semi-conductor nanocrystals. Nature 2005;436:91e4.

[37] Nag A, Sapra S, Nagamani C, Sharma A, Pradhan N, Bhat SV, et al. A study ofMn2þ doping in CdS nanocrystals. Chem Mater 2007;19:3252e9.

[38] Weissleder R, Moore A, Mahmood U, Bhorade R, Benveniste H, Chiocca EA,et al. In vivo magnetic resonance imaging of transgene expression. Nat Med2000;6:351e4.

[39] Ye F, Laurent S, Fornara A, Astolfi L, Qin J, Roch A, et al. Uniform mesoporoussilica coated iron oxide nanoparticles as a highly efficient, nontoxic MRI T2contrast agent with tunable proton relaxivities. Contrast Media Mol Imaging2012;7:460e8.

[40] Lee JE, Lee N, Kim H, Kim J, Choi SH, Kim JH, et al. Uniform mesoporous dye-doped silica nanoparticles decorated with multiple magnetite nanocrystals forsimultaneous enhanced magnetic resonance imaging, fluorescence imaging,and drug delivery. J Am Chem Soc 2009;132:552e7.

[41] Hamerschlak N, Lima M, Kerbauy F. Hematopoietic stem cell transplantationin elderly patients with myelodysplastic syndrome and acute myelogenousleukemia: use of busulfan/fludarabine for conditioning. In: Hayat MA, editor.Stem cells and cancer stem cells. Netherlands: Springer; 2013. pp. 263e9.

[42] Kumar M, Yigit M, Dai G, Moore A, Medarova Z. Image-guided breast tumortherapy using a small interfering RNA nanodrug. Cancer Res 2010;70:7553e61.

[43] Muthiah M, Park I-K, Cho C-S. Surface modification of iron oxide nanoparticlesby biocompatible polymers for tissue imaging and targeting. Biotechnol Adv2013;31:1224e36.

[44] Panagi Z, Beletsi A, Evangelatos G, Livaniou E, Ithakissios DS, Avgoustakis K.Effect of dose on the biodistribution and pharmacokinetics of PLGA and PLGA-mPEG nanoparticles. Int J Pharm 2001;221:143e52.