ARTICLE IN PRESS+ModelPDPDT-480; No. of Pages 10

Photodiagnosis and Photodynamic Therapy (2013) xxx, xxx—xxx

Available online at www.sciencedirect.com

jou rn al hom epage: www.elsev ier .com/ locate /pdpdt

Photosan-II loaded hollow silicananoparticles: Preparation and its effect inkilling for QBC939 cells

Xiaofeng Denga, Li Xionga,1, Liangwu Linb, Guangzhong Xionga,Xiongying Miao MDa,∗

a Department of General Surgery, Second Xiangya Hospital, Central South University, Changsha 410011,Hunan, PR Chinab State Key Laboratory for Powder Metallurgy, Central South University, Changsha 410083, Hunan, PR China

KEYWORDSPhotosan-II;Hollow silicananoparticles;Photodynamictherapy;Cholangiocarcinoma;QBC939 cell

SummaryBackground: Nanoparticles have been explored recently as an efficient means to deliver photo-sensitizers for photodynamic therapy. However, it is largely unknown if polyhematoporphyrin(C34H38N4NaO5, Photosan-II, PS) or other photosensitizers can be efficiently delivered by hollowsilica nanoparticles (HSNP).Methods: Polyhematoporphyrin (C34H38N4NaO5, Photosan-II, PS) was loaded into hollow silicananoparticles (HSNP) by one-step wet chemical-based synthetic route. Dynamic light scattering(DLS) and polydispersive index (PDI) were used for measurement of the particles size and sizedistribution. Transmission electron microscope and scanning electron microscopy were usedfor the microstructure, morphological and chemical composition analysis. Fourier transforminfrared spectrometry spectra and fluorescence emission spectrum were obtained. The photo-biological activity of the PS-loaded HSNP was evaluated on human cholangiocarcinoma QBC939cells. The cellular viability was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetra-zolium bromide (MTT) assay. Apoptotic and necrotic cells were measured by flow cytometry.Results: DLS measurements showed that the size of the particles is in the range of 25—90 nm.PDI of the PS-loaded HSNP is 0.121 ± 0.01, indicating that samples have excellent quality with

narrow size distribution to monomodal systems. In MTT assay, PS-loaded HSNP and free PS of thesame concentration killed about 95.3% ± 2.0% and 55.7% ± 1.9% of QBC939 cells, respectively.The flow cytometry demonstrated that the laser induced cell death with PS-loaded HSNP wasmuch more severe than that of free PS (P < 0.05).Conclusions: Photosan-II-loaded hollow silica nanoparticles not only can quickly deliver

Please cite this article in press as: Deng X, et al. Photosan-II loaded hollow silica nanoparticles: Prepa-ration and its effect in killing for QBC939 cells. Photodiagnosis and Photodynamic Therapy (2013),http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

Photosan-II into cells but also can reach a more high concentration than free Photosan-II. HSNP

∗ Corresponding author. Tel.: +86 731 84895199; fax: +86 731 84895199.E-mail address: 13787782059@163.com (X. Miao).

1 Co-first author.

1572-1000/$ — see front matter © 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

ARTICLE IN PRESS+ModelPDPDT-480; No. of Pages 10

2 X. Deng et al.

is a desirable vehicle and the release system that shows promises for photodynamic therapy use,which not only improve the aqueous solubility, stability and transport efficiency of PS, but also

ficacs res

I

Ptanmof

pptadlbaTlaedplmcvntepltttmct

pIbtpaesoefidsc[

odItrinntwsaTcntim[Dlntl

Pccmra

M

M

T(wAcIFsTb((Fabove-mentioned chemicals were used without any furtherpurification. A 630 nm semiconductor laser photodynamic

increase its photodynamic ef© 2013 Elsevier B.V. All right

ntroduction

hotodynamic therapy (PDT) is a promising novel therapeu-ic method for the treatment of various kinds of tumorsnd other diseases, such as ovarian cancer [1], choroidaleovascularization [2], and oral precancer [3]. Due to mini-al invasion and nontoxicity, PDT provides patients with the

pportunity to be treated painlessly and repeatedly whenailed in traditional therapy.

The conventional PDT utilizes photosensitizers, appro-riate wavelength laser for photoexcitation and oxygen toroduce singlet oxygen and other reactive oxygen species;o cause lipid peroxidation and photooxidation, and eventu-lly leads to cell death [4,5]. The photosensitizers are oftenelivered via vein injection, and the PDT is performed at 48 hater. However, the conventional PDT efficacy is restrictedy insufficient selectivity, low solubility of photosensitizers,nd the limited penetration depth of the 630 nm laser light.hese disadvantages made the PDT efficacy for the tumors

ocated at the deeper tissues are lower than that locatedt the body surface. Recently, with the development ofndoscope and laparoscopy, the 630 nm laser light can beelivered through cylinder diffusing tip quartz fibers androvided a new route to perform the PDT for the tumorsocated at deeper part inside the body. On the other hand, byeans of the endoscope or laparoscopy, the photosensitizers

an be directly planted to the tumor tissues instead of byein injection, and PDT can be performed in situ. In situ PDTot only shortens the time when photosensitizers reach tohe tumor tissues but also avoids the pain when the body wasxposed to the sunlight. In addition, the photosensitizserslanted directly to the tumor tissues via the endoscope oraparoscopy can make the concentration of the photosensi-izers reaching to the required level for an in situ PDT clinicalreatment in a short time. Therefore, in order to improvehe photodynamic efficacy, a photosensitizer with high per-eability and low side effects must be provided [5,6], which

an fastly reach the concentration of the photosensitizers tohe required level for PDT.

However, the hydrophobic nature of many conventionalhotosensitizers, such as polyhematoporphyrin (Photosan-I, PS), results in two inevitable problems: the delivery inlood circulation and the low photo-physical properties dueo the aggregation of photosensitizers, which decreased thehoto-oxidation efficacy to achieve the photodynamic ther-py. These characterizations suggested that it is difficult tonsure the high permeability of the conventional photosen-itizers for the in situ PDT clinical treatment. In order tovercome this problem, nanotechnology-based drug deliv-ry system has become one of the most extensive studiedelds in nanomedicine in the recent decade [7]. A variety of

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

rug delivery systems have been developed to deliver photo-ensitizers, such as liposomes [8,9], polymer carriers [10],remophor emulsion [11], microspheres and nanoparticles12]. Most of the delivery systems improve the properties

scS

y compared to free PS.erved.

f photosensitizers, however most of the designed drugelivery nano-systems need releasing of the loaded drug.t will slow down the seepage velocity of photosensitizers toumors tissues and prolong the time of the photosensitizerseaching to the required concentration level. Therefore, its important that a designed drug delivery nano-system needot releasing of the loaded photosensitizers. Recently, silicaanoparticles have been attached more and more atten-ion due to the easily prepared with desired size and shape,ater-soluble, stable and biocompatible. More importantly,

ilica nanoparticles are permeable to small molecule suchs singlet oxygen [13,14], which is the key effecter of PDT.hese made the photosensitizers loaded silica nanoparti-les be different from conventional delivery systems whicheed releasing of the loaded drug [15]. Thus, silica nanopar-icle is an ideal photosensitizer delivery nano-system formproving the photodynamic efficacy. However, at present,ost of the studies focus on the solid silica nanoparticles

16—18], few focused on the hollow silica nanoparticles.ue to the cavum of the hollow silica nanoparticles (HSNP),

oading-capacity of HSNP is higher than that of solid silicaanoparticles, which will benefit for shortening the time ofhe photosensitizers reaching to the required concentrationevel.

Herein, Photosan-II as a classical photosensitizer forDT loaded with HSNP was prepared by one-step wethemical-based synthetic route at room temperature andharacterized by physical—chemical and photophysicalethods. In vitro photobiological activity assays were car-

ied out to evaluate the capability of the delivery systemgainst human cholangiocarcinoma QBC939 cells.

aterials and methods

aterials

etraethyl orthosilicate (TEOS, 99.99%), polyacrylic acidPAA, M.W = 3000), sodium polyacrylate (PAAS, M.W = 45,000)ere purchased from Aladdin Chemistry Co. Ltd (Shanghai).nhydrous ethanol (99.7%) and ammonia (25—28%) were pur-hased from Sinopharm Chemical Reagent Co. Ltd. Photosan-I (C34H38N4NaO5) was purchased from Seehof Laboratorium&E GmbH Wesselburenerkoog, Germany. The moleculartructure of the Photosan-II was shown in Scheme 1.he Dulbecco’s modified eagle’s medium (DMEM) and fetalovine serum (FBS) were purchased from Gibco Co. 3-4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromideMTT) were obtained from Sigma-Aldrich Co. Annexin-V-LUOS staining kit was purchased from Roche Co. All the

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

ystem was provided by LM Inc. (Shenzhen, China). QBC939ells were available in the cell store of Chinese Academy ofcience.

ARTICLE IN+ModelPDPDT-480; No. of Pages 10

Photosan-II loaded hollow silica nanoparticles

9i2bcatoddlTPstctpp(tvt

M1oPppwnwtpboAsVeulfw

SFdaumwc

R

S

Scheme 1 Molecule structure of Photosan-II.

Methods

Preparation of nanoparticlesNanoparticles containing Photosan-II were prepared byone-step wet chemical-based synthetic route at room tem-perature. Briefly, 0.615 mg of polyacrylic acid (or sodiumpolyacrylate) and 1 mg Photosan-II were dissolved in 4 mlammonia, and then 100 ml anhydrous ethanol was addedrapidly and the system was stirred in the dark until it becameclear (about 0.5 h). After that, 1.395 g TEOS was added intothe system slowly in 10 h at a time interval of 2 h undervigorous magnetic stirring. After the reaction, the productswere centrifuged and washed above three times in deion-ized water and anhydrous ethanol, and dried at 60 ◦C in avacuum drying oven.

CharacterizationParticle size analysis. Particle size and size distributionwere measured by laser light scattering using a particle sizeanalyzer (Zetasizer Nano-S90, Malvern, UK). Size distribu-tion was analyzed over the range of 1—10,000 nm and meandiameter was calculated for each sample. The equipmentalso gives a polydispersive index, which should be low (0.1)for a monodisperse system. The nanoparticles suspensionwas measured in a 1.0 cm quartz cell, at 25 ◦C. Nanoparticleswere analyzed three times (n = 3) with 30 readings taken foreach sample.Microstructure, morphological and chemical compositionanalysis. A JEOL JEM-2100F high-resolution transmissionelectron microscope (TEM) with 1.9 point-to-point resolu-tion operating with a 200 kV accelerating voltage, equippedwith energy dispersive X-ray spectroscopy (EDX INCA,OXFORD), and a FEI Nova NanoSEM 230 field emissionscanning electron microscopy (SEM) were used for themicrostructure, morphological and chemical compositionanalysis. Fourier transform infrared spectrometry (FT-IR)spectra of the PS-loaded HSNP were obtained from a WQF-410 spectrometer in the transmission mode and diffusereflection mode, respectively, at room temperature with aresolution of 4 and a scan frequency of 32.Photochemical characterization. Fluorescence emissionspectrum was measured with a Hitachi F-4500 fluorescencespectrophotometer with a Xe lamp at room temperature.The excitation wavelength was 410 nm.

In vitro studies of photobiological activity

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

Human cholangiocarcinoma QBC939 cell line was used inthese studies. Cells were cultured in DMEM containing 10%FBS in a 5% atmosphere CO2 at 37 ◦C. 200 �l of cell suspensionwith a concentration of 2 × 105 cells/ml were placed into

h

Te

PRESS3

6-well plates. A PS-loaded HSNP suspension was preparedn culture medium (DMEM) at 10 mg/l, 7.5 mg/l, 5 mg/l,.5 mg/l and 1 mg/l. The cells were washed with phosphateuffered saline (PBS) and incubated with 200 �l of variousoncentration of PS-loaded HSNP suspension at 37 ◦C in antmosphere of 5% CO2 for 1 h, 2 h and 4 h. After incubation,he cells were washed twice with PBS and added with 200 �lf fresh culture medium each well. Then, the cells were irra-iated with 630 nm laser for 112, 224 and 336 s, and the lightose was 5, 10 and 15 J/cm2 respectively. Finally, the cellu-ar viability was assessed by MTT assay after 24 h incubation.he photobiological activity of control group (PBS) and freeS group were also assessed with the same method. At theame time, we also assessed the dark toxicity of photosensi-izers and the photobiological activity of the light only. Theellular viability of each sample above was calculated fromhe data of 10 repeated wells, and the evaluation of thehotobiological activity of each sample came from the com-arison of the mean optical density between the untreated100%) and treated cells. The effect of different parame-ers on the in vitro photobiological activity was studied byarying the incubation time (1—4 h), concentration of pho-osensitizer (1—10 mg/l) and light dose (5—15 J/cm2).

easurement of apoptotic and necrotic cells ml of QBC939 cell suspension with a concentrationf 4 × 105 cells/ml were placed into 24-well plates.S-loaded HSNP and free PS suspension were both pre-ared at 7.5 mg/l. The tumor cells, treated with thesehotosensitizers and the light exposure of 10 J/cm2 dose,ere collected to quantitatively assess the apoptotic andecrotic cells by flow cytometry. Then, cells were stainedith Annexin V and propidium iodide (PI) according to

he manufacturer’s instructions. The quantity of cellsositive for Annexin V, for PI or for both was determinedy flow cytometry. In the fluorescence dot plot histogramf Annexin V/PI stained cells, the cells negative for bothnnexin V and PI are in the lower left quadrant, whichtands for normal viable cells; the cells positive for Annexin

only are in the lower right quadrant, which stands forarly-apoptotic cells; the cells positive for PI only are in thepper left quadrant, which stands for necrotic cells withoss of cytomembrane integrity; while the cells positiveor both Annexin V and PI are in the upper right quadrant,hich stands for late-apoptotic cells and necrotic cells.

tatistical analysisor MTT assay, the cellular viability was calculated from theata of 10 repeated wells (n = 10) and the photobiologicalctivity was expressed as a percentage, compared with thentreated cells (100%). The results were expressed as theeans ± standard deviation. Data of all the above studiesere analyzed by Student’s t-test. The differences wereonsidered significant at a level of P < 0.05.

esults

ize and size distributation of Photosan-II-loaded

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

ollow silica nanoparticles

he smallest capillaries in the body are 5—6 �m in diam-ter. The size of particles into the bloodstream must be

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

ARTICLE IN+ModelPDPDT-480; No. of Pages 10

4

Figure 1 Dynamic light scattering measurement of a nanopar-ticle size of Photosan-II-loaded hollow silica nanoparticles. Itindicates that over 95% of them are in the range between 25and 90 nm, with a mean value of 43.1 nm.

snsitHmr9htuHq[q

Mc

TTHst(lf

Hm(tCbvCCC

Figure 2 Typical scanning electron microscopy image of Photosobtained Photosan-II-loaded hollow silica nanoparticles have spheric

PRESSX. Deng et al.

ignificant smaller than 5 �m to ensure that the particles doot form an embolism. Thus, it is important to measure theize particles. Dynamic light scattering (DLS) was used fornitial characterization of the PS-loaded HSNP. Fig. 1 showshe DLS measurement of a nanoparticle size of PS-loadedSNP. Fig. 1 indicates that over 95% of the PS-loaded HSNPanufactured by one-step wet chemical-based synthetic

oute at room temperature are in the range between 25 and0 nm, with a mean value of 43.1 nm. The PS-loaded HSNPas nanometric size and this size is considered satisfactoryo intravenous administration. Polydispersive index (PDI) issed for judging the sample quality. PDI of the PS-loadedSNP is 0.121 ± 0.01, indicating that samples have excellentuality with narrow size distribution to monomodal systems19]. PDI in the range of 0.5—1.0 indicates samples with baduality, constituting polyimodal systems.

icrostructure, morphological and chemicalomposition analysis

he morphology of the particles was examined by SEM andEM. Fig. 2 shows the typical SEM image of the PS-loadedSNP, indicating that the obtained PS-loaded HSNP havepherical shape, smooth regular surface, which depend onhe Polyacrylic acid or Sodium polyacrylate. TEM imageFig. 3) showed that the obtained nanoparticles are hol-ow nanoparticles and the Photosan-II are loaded inside it,orming a nanocomposite photosensitizer.

Fig. 4 shows the typical FTIR spectra of the PS-loadedSNP obtained from a WQF-410 spectrometer in the trans-ission mode (curve (a)) and diffuse reflection mode (curve

b)), respectively, at room temperature. From the two spec-ra, the bands of Si O rocking vibration at 400—500 cm−1,

O in-plane bending vibration at 500—600 cm−1, Si Oending vibration at 802 cm−1, C H out-of-plane bending

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

ibration at 958 cm−1, C H in-plane bending vibration andO or C N or Si O stretching vibration at 1000—1350 cm−1,

N or C C or N H bending vibration at 1350—17,500 cm−1, O stretching vibration at 1750—2000 cm−1, C H or O H

an-II-loaded hollow silica nanoparticles. It indicates that theal shape and smooth regular surface.

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

ARTICLE IN PRESS+ModelPDPDT-480; No. of Pages 10

Photosan-II loaded hollow silica nanoparticles 5

Figure 3 Transmission electron microscopy image of Photosan-II-lnanoparticles are hollow nanoparticles and the Photosan-II are loade

or N H stretching vibration mode at 2900—3500 cm−1 areobserved [20,21], which indicated the encapsulation of thePhotosan-II in the hollow silica nanoparticles. It is interestedthat all the bands above of the spectrum (a) obtained in

Figure 4 Fourier transform infrared spectrometry spectra ofPhotosan-II-loaded hollow silica nanoparticles obtained in thetransmission mode (a) and diffuse reflection mode (b), respec-tively.

t(b2tha

SP

Ohtai[PswasFaef

oaded hollow silica nanoparticles. It showed that the obtainedd inside them, forming a nanocomposite photosensitizer.

he transmission mode are stronger than that of spectrumb) obtained in the diffuse reflection mode. Except for theands above, the bands C O or C N stretching vibration at000—2500 cm−1 were also observed. These facts suggestedhat the Photosan-II was successfully wrapped inside theollow silica nanoparticles, which is great accord with thenalysis of the TEM.

pectroscopic characterization ofhotosan-II-loaded hollow silica nanoparticles

ne of the best strategies to investigate the behavior ofeterogeneous or biological systems consists of the study ofheir photophysical properties using fluorescent compoundss probe. The fluorescence emission of the photosensitizerss sensitive in its environment yielding much information22]. Fig. 5 represents the fluorescence emission spectra ofhotosan-II standard in deionized water and aqueous disper-ion of the PS-loaded HSNP at same concentration, whichere measured from 520 to 750 nm under light excitationt 410 nm at room temperature. The fluorescence emissionpectra of Photosan-II standard shown by the curve (a) in

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

ig. 5 indicate the position of the maximum emission 635 nmnd shoulder peak at around 690 nm of classical Photosan-IImission maximum in deionized water. The emission signalrom the Photosan-II in the PS-loaded HSNP shown by curve

ARTICLE IN PRESS+ModelPDPDT-480; No. of Pages 10

6 X. Deng et al.

Figure 5 Fluorescence emission spectra of Photosan-IIstandard in deionized water and aqueous dispersion ofPhotosan-II-loaded hollow silica nanoparticles at the same con-centration. They were measured from 520 to 750 nm underlsn

(stttPcT

I

Twfiaowi4tglsPPa

cirttnao

Figure 6 Effect of the incubation time (h) in the photo-biological activity of free PS and PS-loaded HSNP. The cellswere treated with 7.5 mg/l of free PS and PS-loaded HSNPfor different incubation times, and with 10 J/cm2 of lightdose. Dark toxicity was studied at the same time. Eachdata point represents the mean of n = 10 determinations.Notes: *significant difference (P < 0.05), **no significant dif-ference (P > 0.05). Abbreviations: PS, Photosan-II; PS-loadedHSNP, Photosan-II-loaded hollow silica nanoparticles; MTT, 3-(4,5-dimethyl-thiazol-2-yl)-2,5-biphenyl tetrazolium bromide.

Figure 7 Effect of the photosensitizer concentration in thephotobiological activity of free PS and PS-loaded HSNP. Thecells were treated with various concentration of free PSand PS-loaded HSNP for 2 h incubation, and with 10 J/cm2

of light dose. Dark toxicity was studied at the same time.Each data point represents the mean of n = 10 determinations.

ight excitation at 410 nm at room temperature. (a) Photosan-IItandard in deionized water. (b) Photosan-II-loaded hollow silicaanoparticles.

b) was similar to that of Porphyrin standard, and the inten-ity was almost 60% of the Photosan-II standard. Compareo the Photosan-II standard, the fluorescence intensity ofhe PS-loaded HSNP decreased significantly, suggesting thathe encapsulation efficiency is low; notwithstanding, thehotosan-II is encapsulated in the hollow silica nanoparti-les which is great accord with the analysis of the FT-IR andEM.

n vitro studies of photobiological activity

he photobiological activity of different incubation timesith PS-loaded HSNP was investigated and compared with

ree PS, as demonstrated in Fig. 6. In the assays of 0 hncubation, the cells were not treated with photosensitizernd receive the irradiation. The photobiological activityf PS-loaded HSNP was time dependent up to 2 h. Thereas no significant difference (P > 0.05) in the photobiolog-

cal activity of the PS-loaded HSNP between the 2 h and h incubation. Therefore, we chose 2 h as the incubationime for the following phototobiological activity investi-ations (effects of the photosensitizer’s concentration andight dose). There was little phototoxicity of the photosen-itizers in the dark. The photobiological activity of the freeS was incubation time dependent in the range of 1—4 h.S-loaded HSNP showed significantly higher photobiologicalctivity than free PS (P < 0.05) (Fig. 6).

The photobiological activity of various photosensitizeroncentrations of free PS and PS-loaded HSNP is exhibitedn Fig. 7. In the group of 0 mg/l, the photosensitizers wereeplaced with PBS. For PS-loaded HSNP group, increasinghe concentration from 1 to 7.5 mg/l was responsible for

Please cite this article in press as: Deng X, et al. Photosan-II loaded hollow silica nanoparticles: Prepa-ration and its effect in killing for QBC939 cells. Photodiagnosis and Photodynamic Therapy (2013),http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

he improvement of the photobiological activity. There waso significant difference (P > 0.05) in the photobiologicalctivity of the PS-loaded HSNP between the concentrationsf 7.5 and 10 mg/l. There was little phototoxicity of the

Notes: *significant difference (P < 0.05), **no significant dif-ference (P > 0.05). Abbreviations: PS, Photosan-II; PS-loadedHSNP, Photosan-II-loaded hollow silica nanoparticles; MTT, 3-(4,5-dimethyl-thiazol-2-yl)-2,5-biphenyl tetrazolium bromide.

ARTICLE IN+ModelPDPDT-480; No. of Pages 10

Photosan-II loaded hollow silica nanoparticles

Figure 8 Effect of the light dose in the photobiologi-cal activity. The cells were treated with 7.5 mg/l of freePS and PS-loaded HSNP for 2 h incubation, and with differ-ent light dose irradiation. Each data point represents themean of n = 10 determinations. Notes: *significant difference(P < 0.05), **no significant difference (P > 0.05). Abbreviations:

aPii

s

D

Pctbpcvrfssosba[

sppspaessaaptSaspaddmccetelThtu

PS, Photosan-II; PS-loaded HSNP, Photosan-II-loaded hollowsilica nanoparticles; MTT, 3-(4,5-dimethyl-thiazol-2-yl)-2,5-biphenyl tetrazolium bromide.

photosensitizers in the dark. Similarly, no phototoxicitywas observed in the PBS group. The cellular viability offree PS gradually decreased for concentration ranging from1 to 10 mg/l. PS-loaded HSNP showed significantly higherphotobiological activity than free PS (P < 0.05) (Fig. 7).

The influence of the light dose on the photobiologicalactivity was investigated with 2 h incubation and 7.5 mg/lphotosensitizer concentration, according to the studiesabove. The PS-loaded HSNP showed photobiological activityat the light dose of 5 J/cm2, and more intense at 10 J/cm2,but a light dose of 15 J/cm2 did not show a significantincrease (P > 0.05) in the photobiological activity comparedto 10 J/cm2. There was little phototoxicity of the photo-sensitizers without laser irradiation. The free PS showedphotobiological activity at the light dose of 10 J/cm2, andmore intense at 15 J/cm2. PS-loaded HSNP showed signifi-cantly higher photobiological activity than free PS (P < 0.05)(Fig. 8).

Analysis of free PS and PS-loaded HSNP inducedapoptosis and necrosis

Percentage of apoptotic and necrotic QBC939 cells inducedby free PS/PS-loaded HSNP and light irradiation was deter-mined by flow cytometry. Early apoptosis exhibit positivestaining for Annexin V while necrosis and later stage apo-

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

ptosis are detected by positive staining for both Annexin Vand PI, and necrosis with loss of cytomembrane integrityshows positive for PI only. Generally, there are less cellsin the upper left quadrant even the vast majority of cells

pi

s

PRESS7

re necrosis, so are our studies. Cells incubated with freeS for 2 h and 4 h demonstrated a time-dependent increasen necrosis and later stage apoptosis (Fig. 9). The photo-nduced cell death by PS-loaded

HSNP was much more severe than that of free PS in theame incubation time and light dose (P < 0.05).

iscussion

atients with cholangiocarcinoma have a poor prognosis andurative surgery can only be performed in a small propor-ion of early diagnosed patients. Palliative biliary drainagey either percutaneous or endoscopic insertion of endo-rostheses improves quality of life by reducing pruritis,holangitis, and pain, but has been reported to improve sur-ival time only slightly [23—25]. Photodynamic therapy is aelatively new local, minimally invasive palliative strategyor unresectable cholangiocarcinoma. PDT uses a photo-ensitive molecule that accumulates in proliferating tissueuch as tumors. Activation of the photosensitizer by usef light of a specific wavelength generates reactive oxygenpecies leading to selective tumor cell death. So far, PDT haseen shown to improve the quality of life, biliary drainage,nd survival of patients with advanced cholangiocarcinoma26—28].

However, the hydrophobic nature of many photosen-itizers such as Photosan-II results in two inevitableroblems: (1) the delivery in blood circulation; (2) the lowhoto-physical properties due to the aggregation of photo-ensitizers. These two problems have greatly decreased thehoto-oxidation efficacy to achieve the photodynamic ther-py. These characterizations suggested that it is difficult tonsure the high permeability of the conventional photosen-itizers for the in situ PDT clinical treatment. In order toolve these problems, we used hollow silica nanoparticless a drug delivery system for photosensitizers. Compared to

variety of drug delivery nano-systems such as liposomes,olymer carriers, cremophor emulsion and microspheres,he silica nanoparticles have quite a lot of advantages.ilica nanoparticles have been obtained more and morettention due to the easily prepared with desired size andhape, water-soluble, stable, biocompatible, especially theorosity which let it be permeable to small molecular suchs singlet oxygen [13,14]. Here, we provide convincingata to show that hollow silica nanoparticles is a perfectelivery system to efficiently delivery PS into cells. Theean size of Photosan-II-loaded hollow silica nanoparti-

les is 43.1 nm which is much smaller than the smallestapillaries in the body. Besides the perfect size, transmissionlectron microscope, scanning electron microscopy, Fourierransform infrared spectrometry spectra and fluorescencemission spectrum showed that our prepared Photosan-II-oaded hollow silica nanoparticles are with excellent quality.aken the two features in consideration, we expected thatollow silica nanoparticles should be a perfect deliver sys-em to deliver Photosan-II in vivo. Therefore, in vivo studiessing different animal models are our interesting aims to be

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

ursued in further. However, the current in vitro character-stics are still important for our future in vivo studies.

Another interesting and important finding in our currenttudy is that Photosan-II-loaded hollow silica nanoparticles

ARTICLE IN PRESS+ModelPDPDT-480; No. of Pages 10

8 X. Deng et al.

Figure 9 Free PS and PS-loaded HSNP induced apoptotic and necrotic cell death in QBC939 cells after irradiation. (a) Cells weretreated with PBS instead of photosensitizer for 2 h. (b) Cells were treated with PS-loaded HSNP of 7.5 mg/l for 2 h incubation. (c)Cells were treated with free PS of 7.5 mg/l for 2 h incubation. (d) Cells were treated with free PS of 7.5 mg/l for 4 h incubation.A Abbh

iLPlPstImslPebPqaaP

hptcsahp

afbihgPisond

C

Iagpeoafi

ll the above cells were then exposed to light dose of 10 J/cm2.ollow silica nanoparticles; PBS, phosphate buffered saline.

s significantly more potent than Photosan-II to kill cells.ight can kill more than 70% cells at 1 h incubation withhotosan-II-loaded hollow silica nanoparticles; howeveright only kill less than 70% cells at 4 h incubation with freehotosan-II. This is suggested that Photosan-II-loaded hollowilica nanoparticles can more efficiently enter into cellshan free Photosan-II. More importantly, 2.5 mg Photosan-I-loaded hollow silica nanoparticles/L can efficiently killore than 50% cells and 10 mg free Photosan-II/L has the

imilar killing ability. This is further supported that 10 J/cm2

ight dose can kill more than 90% cells by incubation withhotosan-II-loaded hollow silica nanoparticles, howeverven using 15 J/cm2 light dose only kill less than 60% cellsy incubation with free Photosan-II. Taken these together,hotosan-II-loaded hollow silica nanoparticles not only canuickly deliver Photosan-II into cells but also can reach

more high concentration than free Photosan-II. Thisdvantage is quite important for the targeted therapy usinghotodynamic therapy (PDT) in clinic.

The incorporation of photosensitizer into nanoparticlesas been shown to reduce toxicity, provide solubility inlasma [29], enhance therapeutic activity [30,31], prolonghe delivery and, in some cases, provide targeting to spe-ific tissues [32]. Nanoparticles can accumulate at the tumor

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

ite due to enhanced endocytotic activity and leaky vascul-ture in the tumor [33]. Recently silica-based nanoparticlesave been widely developed as an efficient means for drug,rotein and gene delivery platforms due to their unique

espt

reviations: PS, Photosan-II; PS-loaded HSNP, Photosan-II-loaded

dvantages such as small and uniform pore size, large sur-ace area and pore volume, as well as nontoxicity andiocompatibility [12,34—37]. The porous structure of sil-ca nanoparticles can not only act as a suitable carrier forydrophobic photosensitizers, but also allow the oxygen andenerated singlet oxygen permeability that is essential forDT therapy [13,38]. Therefore, photosensitizer loaded sil-ca nanoparticles are different from conventional deliveryystems which need releasing of the loaded drug [15]. Basedn our data obtained here, we showed that hollow silicaanoparticles is a high potential vehicle can be used toeliver Photosan-II probably other photosensitizers.

onclusion

n summary, QBC939 cells can be photodynamically dam-ged with PS-loaded HSNP or free PS. The photo-stability,eneration of singlet oxygen, and therapeutic efficiency ofhotosensitizer Photosan-II were significantly improved byncapsulation into porous hollow silica nanoparticles. More-ver, Photosan-II transports into cells much more efficientlynd reaches a much higher concentration. We provide therst evidence to show that hollow silica nanoparticles can

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

fficiently deliver Photosan-II. Because most of the photo-ensitizers have similar features. This may represent aromising delivery system for photosensitizers used for PDTherapy of tumors.

IN+Model

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

ARTICLEPDPDT-480; No. of Pages 10

Photosan-II loaded hollow silica nanoparticles

Conflicts of interests

The authors declare that they have no conflicts of interests.

Acknowledgements

This work was supported by the National Natural ScienceFoundation of China (Grant No. 51021063), and the ChinaPostdoctoral Science Foundation funded project (Grant No.2012M521539), and the Hunan Provincial Natural ScienceFoundation of China (Grant No. 12JJ5048), and the Spe-cialized Research Fund for the Doctoral Program of HigherEducation (Grant No. 20090162120008), and the Open-EndFund for the Valuable and Precision Instruments of CentralSouth University.

References

[1] Goff BA, Blake J, Bamberg MP, Hasan T. Treatment of ovar-ian cancer with photodynamic therapy and immunoconjugatesin a murine ovarian cancer model. British Journal of Cancer1996;74(8):1194—8.

[2] Ideta R, Tasaka F, Jang WD, et al. Nanotechnology-basedphotodynamic therapy for neovascular disease using asupramolecular nanocarrier loaded with a dendritic photosen-sitizer. Nano Letters 2005;5(12):2426—31.

[3] Soukos NS, Hamblin MR, Keel S, Fabian RL, Deutsch TF, HasanT. Epidermal growth factor receptor-targeted immunophoto-diagnosis and photoimmunotherapy of oral precancer in vivo.Cancer Research 2001;61(11):4490—6.

[4] Moan J, Berg K. Photochemotherapy of cancer: exper-imental research. Photochemistry and Photobiology1992;55(6):931—48.

[5] Konan YN, Gurny R, Allemann E. State of the art in the deliv-ery of photosensitizers for photodynamic therapy. Journal ofPhotochemistry and Photobiology B 2002;66(2):89—106.

[6] Soncin M, Polo L, Reddi E, et al. Effect of the delivery systemon the biodistribution of Ge(IV) octabutoxy-phthalocyanines intumour-bearing mice. Cancer Letters 1995;89(1):101—6.

[7] Hatakeyama H, Akita H, Harashima H. A multifunctional enve-lope type nano device (MEND) for gene delivery to tumoursbased on the EPR effect: a strategy for overcoming the PEGdilemma. Advanced Drug Delivery Reviews 2011;63:152—60.

[8] Oh EK, Jin SE, Kim JK, Park JS, Park Y, Kim CK. Retainedtopical delivery of 5-aminolevulinic acid using cationic ultra-deformable liposomes for photodynamic therapy. EuropeanJournal of Pharmaceutical Sciences 2001;44(1—2):149—57.

[9] Peterson CM, Lu JM, Sun Y, et al. Combination chemother-apy and photodynamic therapy with N-(2-hydroxypropyl)methacrylamide copolymer-bound anticancer drugs inhibithuman ovarian carcinoma heterotransplanted in nude mice.Cancer Research 1996;56(17):3980—5.

[10] Soncin M, Polo L, Reddi E, et al. Unusually high affinityof Zn(II)-tetradibenzobarrelenooctabutoxy-phthalocyanine forlow-density lipoproteins in a tumor-bearing mouse. Photo-chemistry and Photobiology 1995;61(3):310—2.

[11] Wieder ME, Hone DC, Cook MJ, Handsley MM, GavrilovicJ, Russell DA. Intracellular photodynamic therapy withphotosensitizer-nanoparticle conjugates: cancer therapy usinga ‘Trojan horse’. Photochemical & Photobiological Sciences

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

2006;5(8):727—34.[12] Tu HL, Lin YS, Lin HY, et al. In vitro studies of functionalized

mesoporous silica nanoparticles for photodynamic therapy.Advanced Materials 2009;21(2):172—7.

[

PRESS9

13] Roy I, Ohulchanskyy TY, Pudavar HE, et al. Ceramic-basednanoparticles entrapping water-insoluble photosensitizinganticancer drugs: a novel drug-carrier system for photody-namic therapy. Journal of the American Chemical Society2003;125(26):7860—5.

14] Wang SZ, Gao RM, Zhou FM, Selke M. Nanomaterialsand signlet oxygen photosensitizers: potential applicationsin photodynamic therapy. Journal of Materials Chemistry2004;14(4):487—93.

15] Snyder JW, Skovsen E, Lambert JD, Ogiby PR. Subcellular, time-resolved studies of singlet oxygen in single cells. Journal of theAmerican Chemical Society 2005;127(42):14558—9.

16] Cho Y, Shi R, Borgens RB, Ivanisevic A. Functionalizedmesoporous silica nanopartical-based drug delivery sys-tem to rescue acrolein-mediated cell death. Nanomedicine2008;3(4):507—19.

17] Zhao Y, Trewyn BG, Slowing II, Lin SY. Mesoporous silicananoparticle-based double drug delivery system for glucose-responsive controlled release of insulin and cyclic AMP. Journalof the American Chemical Society 2009;131:8398—400.

18] Slowing II, Trewyn BG, Giri S, Lin SY. Mesoporous silica nanopar-ticles for drug delivery and biosensing applications. AdvancedFunctional Materials 2007;17:1225—36.

19] Scholes PD, Coombes AGA, Illum L, Daviz SS, Vert M, DaviesMC. The preparation of sub-200 nm poly(lactide-co-glycolide)microspheres for site-specific drug delivery. Journal of Con-trolled Release 1993;25(1—2):145—53.

20] Yadav BS, Tyagi SK, Seema. Study of vibrational spectra of 4-methyl-3-nitrobenzaldehyde. Indian Journal of Pure & AppliedPhysics 2006;44:644—8.

21] Sundaraganesan N, Meganathan C, Anand B, Dominic Joshua B,Lapouge C. Vibrational spectra and assignments of 2-amino-5-iodopyridine by ab initio Hartree—Fock and density functionalmethods. Spectrochimica Acta Part A 2007;67:830—6.

22] Cuccovia IM, Quina FH, Chaimovich H. A remarkable enhance-ment of the rate of ester thiolysis by synthetic amphiphilevesicles. Tetrahedron 1982;38(7):917—20.

23] Gerhards MF, den Hartog D, Rauws EA, et al. Palliative treat-ment in patients with unresectable hilar cholangiocarcinoma:results of endoscopic drainage in patients with type III andIV hilar cholangiocarcinoma. European Journal of Surgery2001;167(4):274—80.

24] Paik WH, Park YS, Hwang JH, et al. Palliative treat-ment with self-expandable metallic stents in patients withadvanced type III or IV hilar cholangiocarcinoma: a percuta-neous versus endoscopic approach. Gastrointestinal Endoscopy2009;69(1):55—62.

25] De Palma GD, Pezzullo A, Rega M, et al. Unilateral placement ofmetallic stents for malignant hilar obstruction: a prospectivestudy. Gastrointestinal Endoscopy 2003;58(1):50—3.

26] Ortner ME, Caca K, Berr F, et al. Successful photodynamictherapy for nonresectable cholangiocarcinoma: a randomizedprospective study. Gastroenterology 2003;125(5):1355—63.

27] Zoepf T, Jakobs R, Arnold JC, Apel D, Riemann JF. Palliationof nonresectable bile duct cancer: improved survival afterphotodynamic therapy. American Journal of Gastroenterology2005;100(11):2426—30.

28] Witzigmann H, Berr F, Ringel U, et al. Surgical and pallia-tive management and outcome in 184 patients with hilarcholangiocarcinoma: palliative photodynamic therapy plusstenting is comparable to r1/r2 resection. Annals of Surgery2006;244(2):230—9.

29] Low K, Knobloch T, Wagner S, et al. Comparison of intracellularaccumulation and cytotoxicity of free mTHPC and mTHPC-

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

loaded PLGA nanoparticles in human colon carcinoma cells.Nanotechnology 2011;22(24):245102.

30] Zeisser-Labouebe M, Lange N, Gurny R, Delie F. Hypericin-loaded nanoparticles for the photodynamic treatment of

IN+ModelP

1

[

[

[

[

[

[

[

ARTICLEDPDT-480; No. of Pages 10

0

ovarian cancer. International Journal of Pharmaceutics2006;326(1—2):174—81.

31] Konan YN, Berton M, Gurny R, Allemann E. Enhanced photo-dynamic activity of meso-tetra(4-hydroxyphenyl)porphyrin byincorporation into sub-200 nm nanoparticles. European Journalof Pharmaceutical Sciences 2003;18(3—4):241—9.

32] Ricci-Junior E, Marchetti JM. Zinc(II) phthalocyanine loadedPLGA nanoparticles for photodynamic therapy use. Interna-tional Journal of Pharmaceutics 2006;310(1—2):187—95.

33] Saxena V, Sadoqi M, Shao J. Enhanced photo-stability,thermal-stability and aqueous-stability of indocyanine green inpolymeric nanoparticulate systems. Journal of Photochemistry

Please cite this article in press as: Deng X, et al. Pration and its effect in killing for QBC939 cells.

http://dx.doi.org/10.1016/j.pdpdt.2013.04.002

and Photobiology B 2004;74(1):29—38.34] Brevet D, Gary-Bobo M, Raehm L, et al. Mannose-targeted

mesoporous silica nanoparticles for photodynamic therapy.Chemical Communications (Cambridge) 2009;12:1475—7.

[

PRESSX. Deng et al.

35] Cheng SH, Lee CH, Yang CS, Tseng FG, Mou CY, LoLW. Mesoporous silica nanoparticles functionalized with anoxygen-sensing probe for cell photodynamic therapy: poten-tial cancer theranostics. Journal of Materials Chemistry2009;19(9):1252—7.

36] Guo HC, Feng XM, Sun SQ, et al. Immunization of mice byhollow mesoporous silica nanoparticles as carriers of porcinecircovirus type 2 ORF2 protein. Virology Journal 2012;9(1):108.

37] Gao Y, Chen Y, Ji X, et al. Controlled intracellular release ofdoxorubicin in multidrug-resistant cancer cells by tuning theshell-pore sizes of mesoporous silica nanoparticles. ACS Nano2011;5(12):9788—98.

hotosan-II loaded hollow silica nanoparticles: Prepa-Photodiagnosis and Photodynamic Therapy (2013),

38] Bechet D, Couleaud P, Frochot C, Viriot ML, GuilleminF, Barberi-Heyob M. Nanoparticles as vehicles for deliveryof photodynamic therapy agents. Trends in Biotechnology2008;26(11):612—21.