Journal of Controlled Release 152 (2011) 21–29

Contents lists available at ScienceDirect

Journal of Controlled Release

j ourna l homepage: www.e lsev ie r.com/ locate / jconre l

NANOMEDICIN

E

Comparative study of photosensitizer loaded and conjugated glycol chitosannanoparticles for cancer therapy

So Jin Lee a,b,1, Heebeom Koo a,1, Hayoung Jeong a,c, Myung Sook Huh a,d, Yongseok Choi b, Seo Young Jeong c,Youngro Byun e, Kuiwon Choi a, Kwangmeyung Kim a,⁎, Ick Chan Kwon a,⁎⁎a Biomedical Research Center, Korea Institute of Science and Technology, 39-1 Haweolgog-Dong, Sungbook-Gu, Seoul 136-791, South Koreab School of Life Sciences and Biotechnology, Korea University, Seoul 136-713, South Koreac Department of Life and Nanopharmaceutical Science, Kyung Hee University, 1 Hoegi-dong, Dongdaemun-gu, Seoul, 130-701, South Koread Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul 151-742, South Koreae Department of Molecular Medicine and Biopharmaceutical Sciences, Graduate School of Convergence Science and Technology and College of Pharmacy, Seoul National University,599 Gwanak-ro, Gwanak-gu, Seoul 151-742, South Korea

⁎ Corresponding author. Tel.: +82 2 958 5916; fax:⁎⁎ Corresponding author. Tel.: +82 2 958 5912; fax:

E-mail addresses: kim@kist.re.kr (K. Kim), ikwon@k1 These authors contributed equally to this paper.

0168-3659/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.jconrel.2011.03.027

a b s t r a c t

a r t i c l e i n f o

Article history:Received 9 December 2010Accepted 24 March 2011Available online 30 March 2011

Keywords:Glycol chitosanPhotodynamic therapyPhotosensitizerDrug-loaded nanoparticleDrug-conjugated nanoparticle

This study reports that tumor-targeting glycol chitosan nanoparticles with physically loaded and chemicallyconjugated photosensitizers can be used in photodynamic therapy (PDT). First, the hydrophobicphotosensitizer, chlorin e6 (Ce6), was physically loaded onto the hydrophobically-modified glycol chitosannanoparticles (HGC), which were prepared by self-assembling amphiphilic glycol chitosan-5β-cholanic acidconjugates under aqueous conditions. Second, the Ce6s were chemically conjugated to the glycol chitosanpolymers, resulting in amphiphilic glycol chitosan-Ce6 conjugates that formed self-assembled nanoparticlesin aqueous condition. Both Ce6-loaded glycol chitosan nanoparticles (HGC-Ce6) and Ce6-conjugated chitosannanoparticles (GC-Ce6) had similar average diameters of 300 to 350 nm, a similar in vitro singlet oxygengeneration efficacy under buffer conditions, and a rapid cellular uptake profile in the cell culture system.However, compared to GC-Ce6, HGC-Ce6 showed a burst of drug release in vitro, whereby 65% of physicallyloaded drugs were rapidly released from the particles within 6.5 h in the buffer condition. When injectedthrough the tail vein into tumor bearing mice, HGC-Ce6 did not accumulate efficiently in tumor tissue,reflecting the burst in the release of the physically loaded drug, while GC-Ce6 showed a prolonged circulationprofile and a more efficient tumor accumulation, which resulted in high therapeutic efficacy. Thesecomparative studies with drug-loaded and drug-conjugated nanoparticles showed that the photosensitizer-conjugated glycol chitosan nanoparticles with excellent tumor targeting properties have potential for PDT incancer treatment.

+82 2 958 5909.+82 2 958 5909.ist.re.kr (I.C. Kwon).

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Chemical drugsaregenerally considered tobeoneof themostefficientforms of cancer therapy. To achieve maximum therapeutic efficacy withminimal side effects, specific delivery of anticancer drugs to the targettumor site is highly desirable in cancer treatment [1]. For this purpose,many research groups have developed various nano-size drug carriers,suchas liposomes, polymer conjugates, inorganic particles, andpolymericnanoparticles [2–5]. Among these, polymeric nanoparticles have receivedmuch attention from biomedical researchers because they possess manyuseful properties and can be easily modified for clinical purposes [6,7].Moreover, nanoparticles provide prolonged circulation in the blood and

accumulation to high levels in tumor tissue by avoiding rapid renalclearance when injected intravenously [8].

A large number of chemical drugs for cancer therapy arehydrophobic, and two methods are used to introduce hydrophobicdrugs into polymeric nanoparticles: 1) physically loading ontopolymeric nanoparticles and 2) chemical conjugation to polymericnanoparticles [9]. The physical loading technique has been widelyused with amphiphilic nanoparticles, but problems, such as theinstability of nanoparticles during blood circulation causing a burst ofrelease and loss of the loaded drugs, have been encountered [10]. Thechemical conjugation technique allows the conjugation of hydropho-bic drugs to hydrophilic polymers, which can then self-assemble toform spherical nanoparticles in aqueous conditions [11]. Importantly,chemically conjugated drugs in nanoparticles demonstrate increasedstability, and unintended release is less frequent than with loadeddrugs. Even with similar polymers and drugs, the in vitro and in vivocharacteristics of nanoparticles can be largely changed by the methodemployed to introduce the drugs into the nanoparticles. However, to

22 S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

the best of our knowledge, few studies have been performed foranalyzing these changes. Therefore, a comparative study of nanopar-ticles physically loaded with or chemically conjugated to similarpolymers and drugs is expected to be helpful to researchers whodevelop or use nanoparticles for drug delivery in cancer treatment.

Photodynamic therapy (PDT) with photosensitizers has emergedas an effective therapeutic option for various tumors and otherdiseases [12,13]. When the proper wavelength of light irradiatesphotosensitizers, highly reactive singlet oxygen is generated, causingdamage to tumor tissues [14]. Along with singlet oxygen, photosen-sitizers can emit fluorescence, which enables easy detection andtracking of the photosensitizers during both in vitro and in vivo studies[15]. Especially under in vivo conditions, this fluorescence is beneficialfor the imaging and quantification of photosensitizers in target tumortissues or other organs [16]. Moreover, even if they are conjugated toother molecules like polymers, photosensitizers remain therapeuticunlike most other drugs [17]. Because of these characteristics,photosensitizers are suitable as model drugs for a comparative studyof loaded and conjugated polymeric nanoparticles under both in vitroand in vivo conditions.

Chitosan is a natural polymer and a form of deacetylated chitin.Chitosan is both biodegradable and biocompatible. Consequently,chitosan has been widely used in various biomedical and pharmaceu-tical formulations [18]. Glycol chitosan (GC) is a chitosanderivativewithethylene glycol groups on its backbone, and itswater solubility is highlyenhanced by these glycol groups. In previous papers, we developedtumor-homing glycol chitosan nanoparticles and applied them as drugcarriers for cancer therapy [5]. When hydrophobic molecules, such as5β-cholanic acid or protophorphyrin IX, were conjugated to a GCpolymer, the resulting amphiphilic conjugates formed self-assembledhydrophobic GC nanoparticles (HGCs) with hydrophilic GC shells andhydrophobic cores under aqueous conditions [19,20]. These HGCsharbored various anticancer drugs in their hydrophobic inner cores, andshowedprolonged circulation in the blood and specific delivery of drugsto tumors for cancer therapy [21,22].

Herein, we synthesized two kinds of tumor targeting nanoparticlescontaining photosensitizers for PDT. We selected chlorin e6 (Ce6) asthe photosensitizer because of its hydrophobicity, its activation bynear infrared wavelengths, allowing it to act in deep tissue layers, andhigh singlet oxygen generation efficiency [23]. We obtained Ce6-loaded glycol chitosan nanoparticles (HGC-Ce6) and Ce6-conjugatedchitosan nanoparticles (GC-Ce6), and compared the in vitro and in vivocharacteristics of these two nanoparticles for PDT in cancer therapy.We developed new photosensitizer-containing glycol chitosan nano-particles and demonstrated their potential for efficient PDT of tumors.Furthermore, this comparative study provides valuable informationabout the in vivo behaviors of nanoparticles for drug delivery.

2. Materials and methods

2.1. Materials

Glycol chitosan (average molecular weight=250 kDa; degree ofdeacetylation=82.7%), chlorin e6(Ce6), N-hydroxysuccinimide(NHS), 5β-colanic acid, 1-ethyl-3-(3-dimethylaminopropyl)-carbo-diimide hydrochloride (EDC), fluorescein isothiocyanate (FITC), andp-nitroso-N,N′-dimethylaniline (RNO) were purchased from Sigma(St. Louis, MO). Anhydrous methanol and dimethyl sulfoxide (DMSO)were purchased from Merck (Darmstadt, Germany). All otherchemicals and solvents were analytical grade and were used withoutfurther purification.

2.2. Preparation of HGC-Ce6 and GC-Ce6

Firstly, hydrophilic glycol chitosan (0.25 g, 1 μmol) was chemicallyconjugated with hydrophobic 5β-cholanic acids (0.01 g, 28 μmol) in

1:1 (v/v) methanol/distilled water solution, and 42 μmol of NHS and42 μmol of EDC were added as catalysts to the solution as previouslyreported [5]. After stirring at room temperature for 1 day, the solutionwasdialyzed for 3 days againstmethanol andwater, and then lyophilizedto give awhite solid powder. The resultingHGCwas optimized to contain160 molecules of 5β-cholanic acids per glycol chitosan polymer andit formed a spherical nanoparticle (size=250 nm) in aqueous condi-tion [5,24]. To prepare Ce6-loaded glycol chitosan nanoparticles (HGC-Ce6), the photosensitizer, Ce6, was physically loaded onto HGCs by asimple dialysismethod [25]. In brief, 50 mgof HGC in 25 ml of DMSO andCe6 solution (2.5, 5 or 10 mg in6.25 mlofDMSO)weremixed, vigorouslystirred for 12 h at room temperature, and dialyzed for 3 days againstdistilled water. The reaction mixture was then filtered through a 0.8umsyringefilter and lyophilized togive a greenpowder containingHGC-Ce6.

Secondly, Ce6-conjugated glycol chitosan nanoparticles (GC-Ce6)were prepared by in 1:9 (v/v) DMSO/distilled water containing glycolchitosan and Ce6. Glycol chitosan polymer (50 mg, 200 nmol) wasdissolved in distilled water (10 ml) and Ce6 (2.5, 5 or 10 mg) wasdissolved in DMSO (2 ml). Then, 3.45 mg (30 μmol) of NHS and 5.73 mg(30 μmol) of EDC were added to the solution, it was gently stirred atroom temperature. After 1 day, the reaction mixture was dialyzed for3 days against methanol and water to remove free Ce6 molecules.Finally, the product was finally filtered through a 0.8 μm syringe filterand lyophilized to give a green powder containing GC-Ce6.

2.3. Characterization of HGC-Ce6 and GC-Ce6

The physical loading and chemical conjugation efficiency of Ce6 inthe nanoparticles were quantified using a Lambda UV–vis 7spectrophotometer (Perkin-Elmer, CT). One percent (v/v) DMSO/water solution was added to each nanoparticle (1 mg/ml) to make aclear solution. The Ce6 concentration was determined was measuringabsorbance at 405 nm and referring to a standard curve of free Ce6concentrations in the same DMSO/water solution. The size distribu-tions of HGC-Ce6 and Ce6-GC were determined in PBS (1 mg/ml,37 °C, pH 7.4) using dynamic light scattering (DLS) (Spectra Physics,Mountain View, CA) and a digital autocorrelator (BI-9000AT,Brookhaven, NY, USA) at 633 nm. The morphological images of thenanoparticles were confirmed by transmission electron microscopy(TEM) (CM30 electron microscope, Philips, CA). For this, each samplewas prepared on a 300-mesh copper grid coated with carbon at aconcentration of 1 mg/ml in distilled water. After drying the sample,negative staining was performed with 2 wt.% uranyl acetate solution.The chemical structures of HGC and GC-Ce6 were confirmed byperforming 1 H NMR (UnityPlus 300, Varian, CA, USA) on samplesprepared in 1:1 (v/v) D2O/CD3OD solution (data not shown).

2.4. Singlet oxygen generation and release profile of HGC-Ce6 andGC-Ce6

The generation of singlet oxygen was determined using p-nitroso-N,N′-dimethylaniline (RNO) as an indicator for photo-induced singletoxygen with histidine as the singlet oxygen trap [26]. One-hundred μlof RNO (100 μl in distilled water, or 250 μM) and 300 μl of histidine indistilled water (0.03 M) were dissolved in a quartz cuvette. Free Ce6,HGC-Ce6 and GC-Ce6 (1 μg of Ce6) were dissolved in 1% (v/v) DMSO/distilled water. To quantify singlet oxygen generation, 1.0 ml of eachsolution was added to the RNO solution. Before measuring, eachsample solution was bubbled with oxygen for 10 min and thenirradiated with a 671 nm laser (220 mW/cm2). The optical density at440 nm (λmax of RNO) was monitored every 2 min using a spectro-photometer (ISS, Champaign, IL).

To obtain thedrug release profiles of HGC-Ce6 andGC-Ce6, 1.0 mgofeach sample was dissolved in 1.0 ml of PBS (pH 7.4) and sonicated at90W for 4 min. The solution was placed in a dialysis tube (molecularweight cutoff=12 to 14 kDa, Spectrum®), and the tube was fully

A

B

Hydrophobically modified glycol chitosan (HGC)

Hydrophobic photosensitizer conjugated glycol chitosan (GC-Ce6)

C

= 5 -cholanic acid

= Glycol chitosan

= chlorin e6

HGC-Ce6(Drug loaded nanoparticle)

GC-Ce6(Drug conjugated nanoparticle)

β

Fig. 1. Preparation of HGC-Ce6 and GC-Ce6. Chemical structure of (A) glycol chitosan-5β-cholanic acid conjugate (HGC) and (B) glycol chitosan-chlorin e6 conjugate (GC-Ce6). (C) Schematic illustration of HGC-Ce6 (drug-loaded nanoparticle) and GC-Ce6(drug-conjugated nanoparticle).

23S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

immersed in 20 ml of PBS and gently shaken at 100 rpm in awater bathat 37 °C. At the proper time interval, samples (1.0 ml) were collectedfrom the outer solution and the concentration of Ce6wasdeterminedbyUV–vis spectroscopy (Lambda Vis 7 Spectrometer, Perkin-Elmer, CT).

2.5. Cellular imaging of HGC-Ce6 and GC-Ce6

To determine the cellular uptake characteristic of nanoparticles,GC polymers of each nanoparticle were labeled with 0.1 wt.% FITC inDMSO. FITC-labeled nanoparticles emitted strong green fluorescence,as previously reported [15]. After FITC labeling, the physicochemicalproperties, such as average particle size and drug content, wereunchanged compared to unlabeled nanoparticles. Squamous cellcarcinoma (SCC-7) cells were originally obtained from the AmericanType Culture Collection (Rockville, MD) and cultured in RPMI 1640(Gibco, Grand Island, NY) containing 10% FBS (Gibco, Grand Island,NY). They were seeded onto 6-well plates (2×105 cells/well) andincubated for 1 day. Then, the medium was replaced with 2 ml ofserum-free RPMI medium containing free Ce6 (2.5 μg/ml), HGC-Ce6,or GC-Ce6 (2.5 μg/ml of Ce6) and then incubated for 1 h. The cellswere then washed twice with PBS (with Ca2+, Mg2+) and fixed with4 wt.% paraformaldehyde. The intracellular localization of FITC labelednanoparticles and Ce6 were determined using a fluorescencemicroscope (IX81, Olympus). The fluorescence images were obtainedwith the FITC filter set (Ex=488 nm, Em=500–550 nm) for FITC-labeled nanoparticles and with the Cy5.5 filter set (Ex=670 nm,Em=720 nm) for Ce6, respectively.

2.6. In vitro phototoxicity of HGC-Ce6 and GC-Ce6

SCC-7 cells were seeded onto 96-well cell culture plates (1×104

cells/well) and incubated for 1 day. Then the medium was replacedwith 200 μl of serum-free medium containing free Ce6, HGC-Ce6, orGC-Ce6 (0.5, 1, 2, 4, 8 μg of Ce6, respectively). After 1 h incubation, thecells werewashed three times with PBS and freshmediumwas added.After these processes, cells were illuminated with a LED lamp (670–690 nm, 100 mW/cm2) for 20 min. Cells were then incubated at 37 °Cfor 1 day, and the cell viability was measured by performing MTTassays. Cell viability was represented as the percentage of live cells pertotal none-treated cells. Using an Annexin V-FITC fluorescencemicroscope kits (BD bioscience, San Jose, CA, USA), apoptotic cellswere observed in green fluorescence images after 12 h incubation.

2.7. In vivo imaging of HGC-Ce6 and GC-Ce6

Athymic nude mice (5 weeks old, Institute of Medical Science,Tokyo) were used for comparing in vivo biodistribution and tumortargeting efficacy of free Ce6, HGC-Ce6 and GC-Ce6. For tumor tissuegeneration, HT-29 human colon adenocalcinoma cells (5.0×106)wereinjected into the left flanks of mice. When the tumor volume grew toapproximately 150±30 mm3, a saline solution of free Ce6, HGC-Ce6 orGC-Ce6 containing 2.5 mg/kg of Ce6 were injected into the tumor-bearing mice via tail vein. Then, the in vivo biodistribution and tumortargeting efficacy of each samplewere evaluated by puttingmice on ananimal plate at 37 °C inside the eXplore Optix system (AdvancedResearch Technologies Inc., Montreal, Canada) [27,28]. During non-invasive whole-body imaging, laser power and detection time wereadjusted to20 μW and 0.3 s per point, respectively. The excitation andemission spots were evaluated by raster-scanning in 1 mm steps overthe region of interest (ROI). To excite Ce6 molecules, a 670 nm pulseddiode laser was used. Near infrared fluorescence emission (650 to700 nm) was detected with a time-correlated single photon countingsystem (Becker and Hickl GmbH, Berlin, Germany) and a fastphotomultiplier tube (Hamamatsu, Japan). To compare ex vivo organbiodistributions of Ce6 in the nanoparticles, mice were sacrificed at2 day after i.v. injection, and the excised organs (livers, lungs, spleens,

kidneys, hearts, and the tumors) were analyzed with a 12-bit CCDcamera (Image Station 4000 mm; Kodak, New Haven, CT) equippedwith a special C-mount lens and a near infrared emission filter (600–700 nm; Omega Optical). The quantification data of fluorescenceintensity in all images was calculated as total photons per centimetersquared per steradian (p/s/cm2/sr) (n=5) [16].

2.8. Therapeutic efficacy of HGC-Ce6 and GC-Ce6 in HT-29tumor-bearing mice

The HT-29 tumor-bearing mice models for PDT were made by thesame methods as described above for the imaging studies. Whentumor volume grew to approximately 150±30 mm3, saline, free Ce6,HGC-Ce6, and GC-Ce6 containing 2.5 mg/kg of Ce6 were injected intothe tail vein (n=5). At 4 h and 12 h post-injection, mice wereirradiated twice with a red laser (671 nm, 220 mW/cm2) for 30 mineach at the tumor site. Then, the therapeutic efficacy of treated micewas evaluated by monitoring the tumor volumes calculated aswidth×length×height×1/2 for 25 days. Differences between exper-imental and control groups were determined using one-way ANOVAand deemed statistically significant (indicated by an asterisk (*) infigure) if pb0.01.

3. Results and discussion

3.1. Physicochemical properties of HGC-Ce6 and GC-Ce6

Hydrophobically modified glycol chitosan nanoparticles (HGC)were prepared as nano-size drug carriers for cancer treatment

Table 1Characteristics of HGC-Ce6 and GC-Ce6 (n=3).

Sample Ce6 amount (%) Loading & conjugationefficiency (%)

5 wt.% a HGC-Ce6 4.79±0.8 96.23±4.810 wt.% HGC-Ce6 9.02±0.9 90.18±9.020 wt.% HGC-Ce6 16.29±1.2 81.45±6.05 wt.% GC-Ce6 4.63±0.8 92.57±4.110 wt.% GC-Ce6 8.73±0.6 87.34±6.020 wt.% GC-Ce6 10.1±0.9 50.27±4.9

a The weight feed ratio of Ce6 to HGC and GC nanoparticles.

24 S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

(Fig. 1A) [5]. HGC containing no drugs formed stable self-assemblednanoparticle with an average diameter of 250 nm in aqueouscondition. Many hydrophobic anticancer drugs can be successfullyloaded into the hydrophobic inner-cores of HGC by a simple dialysismethod [15]. Tomake photosensitizer-loadednanoparticles, bothHGCand Ce6 were dissolved in DMSO and then dialyzed against distilledwater to produce Ce6-loaded HGC (HGC-Ce6) (Fig. 1A and C). When5 wt.%, 10 wt.% and 20 wt.% of Ce6 were loaded into the HGCnanoparticles, the physical loading efficiencies were approximately96%, 90% and 81%, respectively (Table 1). On the other hand, thephotosensitizer-glycol chitosan conjugates were produced by the directchemical conjugation of Ce6 to glycol chitosan polymer in the presenceof NHS and EDC (Fig. 1B and C).When the feed ratio of Ce6was less than10 wt.% in the reactionmixture, the chemical conjugation efficiencywasabove 85%, but it dramatically decreased to about 50% when the feedratio of Ce6 was above 20 wt.% compared with the glycol chitosan

Fig. 2. Particle sizes and morphological shapes of (A) HGC-Ce6 and (B) GC-Ce6. Sizedistributions were determined by dynamic light scattering, and inserted images wereTEM images of each type of nanoparticle.

polymer in the reaction (Table 1). To compare the in vitro and in vivocharacteristics of HGC-Ce6 and GC-Ce6 for cancer treatment, weoptimized the feed ratio of Ce6 to 10 wt.%, wherein the content of Ce6was about 8.7 to 9.0 wt.% in both types of nanoparticles. Therefore, weused these two nanoparticles (10 wt.%HGC-Ce6 and 10 wt.%GC-Ce6) inthe following studies unless otherwise noted.

In aqueous condition, both HGC-Ce6 and GC-Ce6 formed stableself-assembled nanoparticle structures and their average size wasabout 300 to 350 nm, asmeasured by dynamic light scattering (Fig. 2).TEM images also showed a spherical morphology, because thehydrophilic glycol chitosan molecules are located outside and thephysically loaded or chemically conjugated Ce6s are located insidedue to their hydrophobic intermolecular interactions.

3.2. In vitro singlet oxygen generation and release profile of HGC-Ce6and GC-Ce6

We next measured the singlet oxygen generation of free Ce6 andboth types of nanoparticles using p-nitroso-N,N′-dimethylaniline(RNO) as an indicator (Fig. 3A) [26]. Free Ce6 showed a very sharpdownward curve at OD 440 nm, indicating rapid generation of singletoxygen upon light exposure. The optical density of both HGC-Ce6 andGC-Ce6 also decreased when irradiated, but the rate was slower thanthat of free Ce6, due to the self-quenching properties of thenanoparticles [11]. In particular, the amounts of singlet oxygengeneration of HGC-Ce6 and GC-Ce6 were very similar despite their

5 10 15 20 25

10

20

30

40

50

60

70

80

90

100

Cu

mu

lati

ve r

elea

se (

%)

Time (hr)

HGC-Ce6

GC-Ce6

B

0 5 10 15 20 25 300

20

40

60

80

100

RN

O C

on

cen

trat

ion

(%

)

Time (min)

Free Ce6

GC-Ce6

HGC-Ce6

00

A

Fig. 3. Singlet oxygen generation and release profile of HGC-Ce6 and GC-Ce6. (A) Singletoxygen generation determined using RNO as a sensor according to irradiation time.Results represent the means±s.e. (n=3). (B) In vitro release profile of HGC-Ce6 andGC-Ce6measured with UV–vis spectroscopy in PBS at 37 °C. Results represent means±s.e. (n=3).

25S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

different nanoparticle structures, suggesting that the singlet oxygengeneration of Ce6 was not significantly changed by the type ofnanoparticle under these conditions.

To evaluate the in vitro drug release profile, HGC-Ce6 and GC-Ce6were immersed at 37 °C in PBS (pH 7.4) to mimic in vivo conditionsand time-dependent changes in drug concentration weremeasured inthe buffer system [15]. Ce6 was released rapidly from HGC-Ce6(Fig. 3B), demonstrating that the hydrophobic interactions of Ce6s inHGC were relatively weaker than those of covalent conjugation. Overthe 5 h incubation period, 58% of the loaded Ce6 was rapidly releasedfrom the HGC. Hence HGC-Ce6 may lose a large amount of physicallyloaded drugs into the blood stream under in vivo conditions. However,as expected, GC-Ce6 did not release the drug even up to 1 day after thestart of incubation because of the robust nature of the covalent bondbetweenGC and Ce6. Thus, chemically conjugated drugs are expected toshow the same biodistribution of nano-sized drug carriers without anyloss of conjugated drugs in the blood stream. However, based on thedrug release study, the drug release profiles of HGC-Ce6 and GC-Ce6showed substantial differences under aqueous conditions.

3.3. Cellular uptake and in vitro phototoxicity of HGC-Ce6 and GC-Ce6

To assay cellular uptake of nanoparticles in the SCC7 tumor cellculture system, both nanoparticles were labeled with FITC andobserved through an FITC filter set (green color). Also, the preciselocation of Ce6s after cellular uptake was observed by using a Cy5.5filter set (red color). After 1 h of incubation, free Ce6 (2.5 μg/ml)-treated cells presented a strong red color in the cytoplasm, indicatingthat a large amount of Ce6 had been transported to and diffusedinto the cytoplasm (Fig. 4). Tumor cells incubated with HGC-Ce6 orGC-Ce6 nanoparticles (2.5 μg/ml of Ce6 in each case) showed greenFITC spots in the cytoplasm, indicating efficient cellular uptake of bothnanoparticles. In the case of HGC-Ce6 treated cells, most of the FITC

Fig. 4. Fluorescence microscopy images of SCC-7 cells incubated w

fluorescence was localized to the cytoplasm, and physically loadedCe6s were diffused throughout the cytoplasm, indicating that the Ce6swere rapidly released from HGC after cellular uptake. These resultsconfirmed that HGC could deliver Ce6s into the cytoplasm of tumorcells almost as efficiently as free Ce6 within one hour. These resultsare in accordance with a previous study demonstrating that HGCshows fast cellular uptake by several nondestructive endocyticpathways, such as caveolae-mediated endocytosis, clathrin-mediatedendocytosis, and macropinocytosis [25,29]. However, in the case ofGC-Ce6-treated cells, the red fluorescent spots congregatedmore thanthose of HGC-Ce6. Furthermore, these large red spots mainly co-localized with the green spots in the fluorescence images, indicatingthat Ce6 of GC-Ce6was chemically conjugated to the GC polymers andcould not be released from the nanoparticles. The different intracel-lular behaviors of HGC-Ce6 and GC-Ce6 might substantially affecttheir therapeutic efficacy against cancer.

Next, we studied the in vitro phototoxicity of HGC-Ce6 and GC-Ce6in a tumor cell culture system. As shown in Fig. 5A, tumor cells wereincubated with free Ce6, HGC-Ce6, or GC-Ce6 (0.5, 1, 2, 4, 8 μg of Ce6,respectively) for 2 h and then irradiated using an LED lamp at 671 nmfor 20 min. At 1 day post-irradiation, the phototoxicity of each samplewas determined by the MTT assay. The cell viability decreased as afunction of the increase in the concentration of Ce6, reflecting Ce6phototoxicity. When the Ce6 concentration of free Ce6, HGC-Ce6, orGC-Ce6 was less than 4 μg/ml, tumor cells treated with free Ce6showed the greatest loss of cell viability. Importantly, HGC-Ce6-treatedcells showed higher phototoxicity than GC-Ce6-treated cells, whereasboth cell treatments showed similar in vitro singlet oxygen generationefficiencies. This is consistent with the idea that rapidly released Ce6s inthe HGC-Ce6 can generate singlet oxygen at target cellular organellessuch as mitochondria, compared with congregated photosensitizers ofGC-Ce6 [30]. At a higher Ce6 concentration of 8 μg/ml, all tumor cellstreatedwith free Ce6, HGC-Ce6, or GC-Ce6 showed severe phototoxicity

ith free Ce6, FITC labeled-HGC-Ce6 or FITC-labeled GC-Ce6.

0 2 4 6 80

20

40

60

80

100

Cel

l Via

bili

ty (

%)

Concentration of Ce6 (μg/ml)

Free Ce6

HGC-Ce6

GC-Ce6

A

B

DIC DAPI Annexin V

No treat

HGC-Ce6

GC-Ce6

Free Ce6

50μm

Fig. 5. In vitro phototoxicity of HGC-Ce6 and GC-Ce6. (A) MTT assay data and(B) Annexin V treated images of SCC-7 cells treated with free Ce6, HGC-Ce6, or GC-Ce6and laser irradiation.

26 S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

at 1 day post-irradiation, which was confirmed by FITC-labeledannexin-V (green apoptotic cells) (Fig. 5B). This result indicated thatall Ce6 molecules in HGC-Ce6 and GC-Ce6 would be therapeuticallyefficient for PDT in cancer treatment.

3.4. In vivo biodistribution of HGC-Ce6 and GC-Ce6

To compare the in vivo biodistribution of physically loaded andchemically conjugated Ce6s in nanoparticles, saline, free Ce6, HGC-Ce6,or GC-Ce6 (2.5 mg/kg of Ce6) was injected intravenously into HT-29human colon adenocarcinoma tumor-bearing mice. The in vivobiodistribution of Ce6 molecules in each sample could be directlymonitored by non-invasive and real-time fluorescence imaging of thewhole body, because Ce6 can emit strong near infrared (NIR)fluorescence for efficient tracking [31]. Therefore, it is possible toimage the in vivo localization of physically loaded or chemicallyconjugated Ce6 molecules in in vivo systems without using additionalfluorescence dyes.

At 3 h post-injection, free Ce6-treated mice showed whole-bodyfluorescence that was not seen in saline-injected mice. A strong signalwas detected in the liver tissue (Fig. 6A). Thus, we did not observetumor tissue specificity for free Ce6 in these mice. In addition, whole-body fluorescence decreased rapidly, due to the rapid excretion of Ce6[32,33]. In the case of HGC-Ce6, intense whole-body fluorescence wasobserved at 3 h post-injection, indicating that physically loaded Ce6circulates for a longer time than free Ce6. Moreover, there was a slightincrease in Ce6 tumor target specificity. However, the fluorescence ofthe whole body and the tumor tissue decreased similarly and rapidly,as observed for free Ce6 , indicating that physically loaded Ce6 wasrapidly released from HGC, as predicted by the in vitro release data inFig. 3B. However, GC-Ce6-treated mice displayed much higher tumortarget specificity at 1 h post-injection than either free Ce6 or HGC-Ce6,and the subcutaneous tumor tissues could be easily distinguishedfrom the surrounding tissue. Furthermore, the fluorescence of GC-Ce6in tumor tissue was maintained for up to 2 days, indicating that thechemically conjugated Ce6 molecules were not subject to rapidexcretion from the body. The total photon counts of GC-Ce6 in tumortissue were about 6–7 fold higher than that of free Ce6 and HGC-Ce6(Fig. 6B).

Ex vivo fluorescence images also confirmed higher fluorescence inthe tumor tissue of GC-Ce6-treated mice, indicating superior in vivotumor targeting of GC-Ce6 compared to that of free Ce6 or HGC-Ce6(Fig. 6C). These in vivoand exvivo results demonstrated thatdrug loadednanoparticles had greater tumor targeting specificity than the free drug,although drug accumulation in the tumor tissue from HGC-Ce6 wasinefficient because of the burst in the release of the drug into the bloodstream. In contrast, the chemically conjugated Ce6 in GC-Ce6 nano-particles was unaffected by the burst in drug release, and consequentlythe GC-Ce6 nanoparticle displayed a more prolonged blood circulationtime and more efficient accumulation in the tumor tissue.

3.5. In vivo therapeutic efficacy of HGC-Ce6 and GC-Ce6

The in vivo therapeutic efficiencies of HGC-Ce6 and GC-Ce6 wereevaluated by measuring tumor growth rates in HT-29 tumor-bearingmice with laser irradiation for PDT. When the inoculated tumorsize grew to 150±30 mm3, saline, free Ce6, HGC-Ce6 or GC-Ce6(2.5 mg/kg of Ce6) was intravenously injected into the mice. Afterpost-injection periods of 4 h and 12 h, the tumor tissues wereirradiated with a red laser (670 nm, 250 mW/cm2) for 30 min andthe tumor sizewasmonitored up to 20 days. After 7 daypost-injection,only tumor tissue containing GC-Ce6 and irradiated with the lasershowed severe necrosis, indicating excellent phototoxicity in cancertreatment, whereas free Ce6- and HGC-Ce6-treated mice failed toshow noticeable phototoxicity in tumor tissue, indicating their lowertherapeutic efficacy in cancer treatment (Fig. 7A).

The tumor growth rate of individual mice clearly showed theexcellent therapeutic results of GC-Ce6 in cancer therapy. As shown inFig. 7B, free Ce6-treated mice did not show an effective decrease intumor size compared to the saline-treatedmice. This is perhaps due tolower tumor target specificity. HGC-Ce6-treated mice had slowertumor growth rates than free Ce6-treated mice, but tumor suppres-sion was incomplete in these mice. However, GC-Ce6-treated miceshowed a dramatic decrease in tumor volume. The final tumorvolumes in these mice were approximately 160 mm3 at 20 day post-injection, which was significantly smaller than that of free Ce6 orHGC-Ce6-treated mice (about 760 mm3 and 560 mm3, respectively).This suggested that GC-Ce6 has potential as an effective Ce6-deliverysystem for cancer treatment, most likely because of the robust natureof the chemically conjugated Ce6 molecules in the glycol chitosannanoparticles. Taken together, the results demonstrate that there aresignificant differences between drug-loaded and conjugated nano-particles with respect in vivo biodistribution, tumor target specificity,and cancer treatment.

Fig. 6. In vivo biodistribution of HGC-Ce6 and GC-Ce6. (A) In vivo time-dependant whole body imaging of athymic nude mice bearing HT-29 tumors after i.v. injection of saline, freeCe6, HGC-Ce6, or GC-Ce6 (2.5 mg/kg of Ce6). (B) Quantification of in vivo tumor target specificity recorded as total photon counts per centimeter squared per steradian (p/s/cm²/sr)of each tumor. (C) Ex vivo images of normal organs (liver, lung, spleen, kidney, and heart) and tumors excised at 2 days post-injection of saline, free Ce6, HGC-Ce6, or GC-Ce6.

27S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

4. Conclusions

In this study, we have shown that drug-loaded and conjugatednanoparticles perform differently during both in vitro and in vivoexperiments. We selected Ce6 and GC as a photosensitizer and abiocompatible polymer, respectively, and developed drug-loadedand drug-conjugated nanoparticles as model drug delivery systemsin cancer treatment. Both nanoparticles were well dispersed in theaqueous system and formed stable nano-structures. HGC-Ce6showed time dependant release of Ce6 and more efficient photo-

dynamic therapy than GC-Ce6 in the cell culture system. HoweverHGC-Ce6 did not accumulate efficiently in tumor tissue because ofthe burst in drug release, even though it showed enhanced tumortargeting compared with free drug. On the other hand, GC-Ce6 had aprolonged circulation time and efficiently accumulated in thetumor, which resulted in excellent therapeutic efficacy in tumor-bearing mice. The results suggest that GC-Ce6 has great potentialin in vivo PDT for tumors, and provide valuable information aboutin vivo burst release of drugs and nanoparticle stabilities for cancertherapy.

Fig. 7. In vivo therapeutic efficacy of HGC-Ce6 and GC-Ce6. (A) Tumor images and(B) tumor growth data after photodynamic therapy with free Ce6, HGC-Ce6, or GC-Ce6(5 mg/kg of Ce6) in HT-29 tumor-bearing mice (*=pb0.01).

28 S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

NANOMEDICIN

E

Acknowledgements

This work was financially supported by the Real-Time MolecularImaging Project, the Global Research Laboratory Project, FusionTechnology Project (2009-0081876) of MEST, National R&D Programfor Cancer Control of Ministry for Health andWelfare from Republic ofKorea (1020260), and the Intramural Research Program of KIST.

References

[1] M.A. Phillips, M.L. Gran, N.A. Peppas, Targeted nanodelivery of drugs anddiagnostics, Nano Today 5 (2010) 143–159.

[2] X. Li, L. Ding, Y. Xu, Y. Wang, Q. Ping, Targeted delivery of doxorubicin usingstealth liposomes modified with transferrin, Int. J. Pharm. 373 (2009) 116–123.

[3] E. Lee, J. Lee, I.-H. Lee, M. Yu, H. Kim, S.Y. Chae, S. Jon, Conjugated chitosan as anovel platform for oral delivery of paclitaxel, J. Med. Chem. 51 (2008) 6442–6449.

[4] T.Y. Ohulchanskyy, I. Roy, L.N. Goswami, Y. Chen, E.J. Bergey, R.K. Pandey, A.R. Oseroff,P.N. Prasad, Organically modified silica nanoparticles with covalently incorporatedphotosensitizer forphotodynamic therapy of cancer, Nano Lett. 7 (2007)2835–2842.

[5] K. Kim, J.H. Kim, H. Park, Y.-S. Kim, K. Park, H. Nam, S. Lee, J.H. Park, R.-W. Park, I.-S.Kim, K. Choi, S.Y. Kim, K. Park, I.C. Kwon, Tumor-homing multifunctionalnanoparticles for cancer theragnosis: simultaneous diagnosis, drug delivery, andtherapeutic monitoring, J. Control. Release 146 (2010) 219–227.

[6] M.E. Fox, F.C. Szoka, J.M.J. Fréchet, Soluble polymer carriers for the treatment ofcancer: the importance of molecular architecture, Acc. Chem. Res. 42 (2009)1141–1151.

[7] Z. Liu, Y. Jiao, Y. Wang, C. Zhou, Z. Zhang, Polysaccharides-based nanoparticles asdrug delivery systems, Adv. Drug Deliv. Rev. 60 (2008) 1650–1662.

[8] H.Maeda, J.Wu, T. Sawa, Y.Matsumura, K.Hori, Tumorvascularpermeability and theEPR effect in macromolecular therapeutics: a review, J. Control. Release 65 (2000)271–284.

[9] K.Y. Choi, H. Chung, K.H.Min, H.Y. Yoon, K. Kim, J.H. Park, I.C. Kwon, S.Y. Jeong, Self-assembled hyaluronic acid nanoparticles for active tumor targeting, Biomaterials31 (2010) 106–114.

[10] A.S. Hasan, M. Socha, A. Lamprecht, F.E. Ghazouani, A. Sapin, M. Hoffman, P.Maincent, N. Ubrich, Effect of the microencapsulation of nanoparticles on thereduction of burst release, Int. J. Pharm. 344 (2007) 53–61.

[11] B.-c. Bae, K. Na, Self-quenching polysaccharide-based nanogels of pullulan/folate-photosensitizer conjugates for photodynamic therapy, Biomaterials 31 (2010)6325–6335.

[12] D.E.J.G.J. Dolmans, D. Fukumura, R.K. Jain, Photodynamic therapy for cancer,Nat. Rev. Cancer 3 (2003) 380–387.

[13] C.M. Moore, D. Pendse, M. Emberton, Photodynamic therapy for prostate cancer-areview of current status and future promise, Nat. Clin. Pract. Urol. 6 (2009) 18–30.

[14] L.K. Folkes, P. Wardman, Enhancing the efficacy of photodynamic cancer therapy byradicals from plant auxin (indole-3-acetic acid), Cancer Res. 63 (2003) 776–779.

[15] S.J. Lee, K. Park, Y.-K. Oh, S.-H. Kwon, S. Her, I.-S. Kim, K. Choi, S.J. Lee, H. Kim, S.G.Lee, K. Kim, I.C. Kwon, Tumor specificity and therapeutic efficacy of photosen-sitizer-encapsulated glycol chitosan-based nanoparticles in tumor-bearing mice,Biomaterials 30 (2009) 2929–2939.

[16] H. Koo, H. Lee, S. Lee, K.H. Min, M.S. Kim, D.S. Lee, Y. Choi, I.C. Kwon, K. Kim, S.Y.Jeong, In vivo tumor diagnosis and photodynamic therapy via tumoral pH-responsive polymeric micelles, Chem. Commun. 46 (2010) 5668–5670.

[17] N. Nishiyama, Y. Morimoto, W.-D. Jang, K. Kataoka, Design and development ofdendrimer photosensitizer-incorporated polymeric micelles for enhanced photo-dynamic therapy, Adv. Drug Deliv. Rev. 61 (2009) 327–338.

[18] J.H. Park, G. Saravanakumar, K. Kim, I.C. Kwon, Targeted delivery of low moleculardrugs using chitosan and its derivatives, Adv. Drug Deliv. Rev. 62 (2010) 28–41.

[19] B.S. Lee, K. Park, S. Park, G.C. Kim, H.J. Kim, S. Lee, H. Kil, S.J. Oh, D. Chi, K. Kim, K.Choi, I.C. Kwon, S.Y. Kim, Tumor targeting efficiency of bare nanoparticles does notmean the efficacy of loaded anticancer drugs: Importance of radionuclide imagingfor optimization of highly selective tumor targeting polymeric nanoparticles withor without drug, J. Control. Release 147 (2010) 253–260.

[20] S.J. Lee, H. Koo, D.-E. Lee, S. Min, S. Lee, X. Chen, Y. Choi, J.F. Leary, K. Park, S.Y.Jeong, I.C. Kwon, K. Kim, K. Choi, Tumor-homing photosensitizer-conjugatedglycol chitosan nanoparticles for synchronous photodynamic imaging andtherapy based on cellular on/off system, Biomaterials 32 (2011) 4021–4029.

[21] K.H. Min, K. Park, Y.-S. Kim, S.M. Bae, S. Lee, H.G. Jo, R.-W. Park, I.-S. Kim, S.Y. Jeong,K. Kim, I.C. Kwon, Hydrophobically modified glycol chitosan nanoparticles-encapsulated camptothecin enhance the drug stability and tumor targeting incancer therapy, J. Control. Release 127 (2008) 208–218.

[22] J.-H. Kim, Y.-S. Kim, K. Park, S. Lee, H.Y. Nam, K.H. Min, H.G. Jo, J.H. Park, K. Choi, S.Y.Jeong, R.-W. Park, I.-S. Kim, K. Kim, I.C. Kwon, Antitumor efficacy of cisplatin-loadedglycol chitosan nanoparticles in tumor-bearing mice, J. Control. Release 127 (2008)41–49.

[23] W.W. Chin, P.W. Heng, P.S. Thong, R. Bhuvaneswari, W. Hirt, S. Kuenzel, K.C. Soo,M. Olivo, Improved formulation of photosensitizer chlorin e6 polyvinylpyrroli-done for fluorescence diagnostic imaging and photodynamic therapy of humancancer, Eur. J. Pharm. Biopharm. 69 (2008) 1083–1093.

[24] H.-Y. Hwang, I.-S. Kim, I.C. Kwon, Y.-H. Kim, Tumor targetability and antitumoreffect of docetaxel-loaded hydrophobically modified glycol chitosan nanoparti-cles, J. Control. Release 128 (2008) 23–31.

[25] S. Park, S.J. Lee, H. Chung, S. Her, Y. Choi, K. Kim, K. Choi, I.C. Kwon, Cellular uptakepathway and drug release characteristics of drug-encapsulated glycol chitosannanoparticles in live cells, Microsc. Res. Tech. 73 (2010) 857–865.

[26] R. Bachor, C.R. Shea, R. Gillies, T. Hasan, Photosensitized destruction of humanbladder carcinoma cells treated with chlorin e6-conjugated microspheres, Proc.Natl. Acad. Sci. U. S. A. 88 (1991) 1580–1584.

[27] X.L. Wu, J.H. Kim, H. Koo, S.M. Bae, H. Shin, M.S. Kim, B.-H. Lee, R.-W. Park, I.-S.Kim, K. Choi, I.C. Kwon, K. Kim, D.S. Lee, Tumor-targeting peptide conjugated pH-responsive micelles as a potential drug carrier for cancer therapy, Bioconjug.Chem. 21 (2010) 208–213.

[28] S. Lee, K.Y. Choi, H. Chung, J.H. Ryu, A. Lee, H. Koo, I.-C. Youn, J.H. Park, I.-S. Kim, S.Y.Kim, X. Chen, S.Y. Jeong, I.C. Kwon, K. Kim, K. Choi, Real time, high resolution videoimaging of apoptosis in single cells with a polymeric nanoprobe, Bioconjug. Chem.22 (2011) 125–131.

[29] H.Y. Nam, S.M. Kwon, H. Chung, S.-Y. Lee, S.-H. Kwon, H. Jeon, Y. Kim, J.H. Park, J.Kim, S. Her, Y.-K. Oh, I.C. Kwon, K. Kim, S.Y. Jeong, Cellular uptake mechanism andintracellular fate of hydrophobically modified glycol chitosan nanoparticles,J. Control. Release 135 (2009) 259–267.

[30] J. Morgan, A.R. Oseroff, Mitochondria-based photodynamic anti-cancer therapy,Adv. Drug Deliv. Rev. 49 (2001) 71–86.

29S.J. Lee et al. / Journal of Controlled Release 152 (2011) 21–29

EDICIN

E

[31] S.Y. Park, H.J. Baik, Y.T. Oh, K.T. Oh, Y.S. Youn, E.S. Lee, A smart polysaccharide/drug conjugate for photodynamic therapy, Angew. Chem. Int. Ed. 50 (2011)1644–1647.

[32] G.P. Gurinovich, T.E. Zorina, S.B. Melnov, N.I. Melnova, I.F. Gurinovich, L.A.Grubina, M.V. Sarzhevskaya, S.N. Cherenkevich, Photodynamic activity of chlorin

e6 and chlorin e6 ethylenediamide in vitro and in vivo, J. Photochem. Photobiol., B13 (1992) 51–57.

[33] G.A.Kostenich, I.N. Zhuravkin, E.A. Zhavrid,Experimental grounds forusingchlorine6 inthe photodynamic therapy of malignant tumors, J. Photochem. Photobiol., B 22 (1994)211–217.

M

NANO