at SciVerse ScienceDirect

Biomaterials 33 (2012) 5574e5583

Contents lists available

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Doxorubicin-loaded highly porous large PLGA microparticles as a sustained-release inhalation system for the treatment of metastatic lung cancer

Insoo Kim a, Hyeong Jun Byeon a, Tae Hyung Kim a, Eun Seong Lee b, Kyung Taek Oh c, Beom Soo Shin d,Kang Choon Lee a, Yu Seok Youn a,*

a School of Pharmacy, Sungkyunkwan University, 300 Cheoncheon-dong, Jangan-gu, Suwon 440-746, Republic of KoreabDivision of Biotechnology, The Catholic University of Korea, 43-1 Yeokgok 2-dong, Wonmi-gu, Bucheon-si, Gyeonggi-do 420-743, Republic of KoreacCollege of Pharmacy, Chung-Ang University, 221 Heukseok dong, Dongjak-gu, Seoul 155-756, Republic of KoreadCollege of Pharmacy, Catholic University of Daegu, 330 Geumrak 1-ri, Hayang Eup, Gyeongsan si, Gyeongbuk 712-702, Republic of Korea

a r t i c l e i n f o

Article history:Received 17 February 2012Accepted 7 April 2012Available online 10 May 2012

Keywords:Porous microparticlesInhalationDoxorubicinLung cancerSustained-release

* Corresponding author. Tel.: þ82 31 290 7785; faxE-mail address: ysyoun@skku.edu (Y.S. Youn).

0142-9612/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.biomaterials.2012.04.018

a b s t r a c t

Doxorubicin-loaded highly porous large PLGA microparticles (Dox PLGA MPs) were prepared using a w/o/w double emulsification method using ammonium bicarbonate effervescent salt. The prepared DoxPLGA MPs were characterized by particle size analysis, scanning electron microscopy, and confocalmicroscopy. In vitro cytotoxicity to B16F10 melanoma cells and lung deposition in C57BL/6 mice wereexamined, and finally the anti-tumor efficacy of pulmonary administered Dox PLGA MPs was evaluatedin a mouse model of B16F10 melanoma metastasis. Results showed that Dox PLGA MPs were highlyporous, had high encapsulation efficiency, and good aerosolization characteristics. Doxorubicin wasgradually released from Dox PLGAMPs over 2 weeks, and after pulmonary administration, Dox PLGA MPswere deposited in lungs and remained in situ for up to 14 days. Furthermore, exposure to Dox PLGA MPskilled B16F10 cells in vitro within 24 h. In particular, tumors in B16F10-implanted mice treated with DoxPLGA MPs were remarkably smaller in terms of mass and number than those in non-treated B16F10-implanted mice. We believe that doxorubicin-loaded highly porous large PLGA microparticles havegreat potential as a long-term inhalation agent for the treatment of lung cancer.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Lung cancer is a leading cause of malignancy-related death inmost developed countries, and the incidence of lung cancer indeveloping countries is rapidly increasing [1]. The lungs area frequent site of metastasis, and >90% of deaths from lung cancerare attributed to the metastatic process [2,3]. Conventionalmodalities, such as, radiotherapy, chemotherapy, or their combi-nations, are considered to be the primary treatment choices fornon-small-cell lung cancer patients [4]. However, the systemicadministration of non-specific chemotherapeutic agents causessignificant toxicities and undesirable side effects because the anti-cancer agents used act on normal cells as well as tumor cells [5,6].Furthermore, the delivery efficiencies of intravenously adminis-tered agents to the lungs are low because they are diluted in thesystemic circulation, and thus, therapies are often unsatisfactoryand survival times are low [4,6].

: þ82 31 290 7724.

All rights reserved.

Direct local delivery of chemotherapeutic agents to lung cancersites offers an attractive alternative approach because it allows theconcentrated delivery of anti-cancer drugs to tumor sites [5e7].Moreover, high systemic levels of chemotherapeutic agents can beavoided when inhalatory delivery systems are used, because theyreduce adverse systemic side effects. In addition, this form ofdelivery is non-invasive, and thus improves patient compliance,versus intravenous injections [8].

Nevertheless, the need for frequent administration can be a realhurdle to the establishment of a drug requiring pulmonaryadministration, and can be bothersome to lung cancer patients[9e11], inwhom breathing is weak. Furthermore, frequent or failedinhalation of cytotoxic chemotherapeutics may induce higher riskto harm normal airway tissues frommouth to alveoli at the cellularlevel, when compared with fewer inhalations. Accordingly, therequirement for frequent inhalation should be avoided, whichdrives efforts to develop inhalatory systems that provide long-termsustained release type of active agents.

Porous microparticles (MPs) are viewed as useful tool for thedelivery of drugs to the lungs because they are light and areinhaled deeply into the lungs [12e14]. Particles of low density

I. Kim et al. / Biomaterials 33 (2012) 5574e5583 5575

(<0.4 g/cm3) and as large as 5 mm or more can be aerosolized andpenetrate deeply into lungs and even reach alveoli. On the otherhand, heavy particles (density > 1.0 g/cm3) must be sized < 3 mmto achieve the same penetration [15,16]. However, 1e5 mm parti-cles tend to agglomerate because of increased van der Waalsattraction, which hampers particle mobility and aerosolization[7,17], and 1e3 mm particles are vigorously taken up by lungmacrophages, which reduces the localized effects of agents [15].For these reasons, particles as large as 5e30 mm are required [11],and large, light particles are considered optimal for the deliverysustained release preparations intended to penetrate deeply intolungs.

Here, we describe an inhalable sustained-release deliverysystem of doxorubicin for the treatment of lung cancer. For thispurpose, we developed a doxorubicin-loaded large and highlyporous poly(d,l-lactic-co-glycolic acid) (PLGA) microparticledelivery system with good aerosolization characteristics. In addi-tion, we examined the physicochemical properties, release profile,in vitro cytotoxicity (to B16F10 cells), lung deposition characteris-tics, and anti-tumor efficacy (in a B16F10 melanoma metastaticlung mouse model) of this delivery system.

2. Materials and methods

2.1. Materials

Poly(d,l-lactic-co-glycolic acid) (PLGA) (Mw: 10,000; lactic acid : glycolicacid ¼ 50:50) and poly(ethylene-alt-maleic anhydride) (PEMA, Mw: 400,000; 1:1)were purchased from Wako Pure Chemical (Tokyo) and Polysciences, Inc. (War-rington, PA), respectively. Doxorubicin hydrochloride (Dox) was obtained from theResearch Laboratories of Boryung Pharm. Co., Ltd. (Ansan, Korea). Ammoniumbicarbonate (ABC) was purchased from SigmaeAldrich. All other reagents wereobtained from SigmaeAldrich, unless otherwise specified.

2.2. Animals

C57BL/6 mice (males, 7 weeks old) were purchased from the Hyochang Exper-imental Animal Laboratory (Daegu, Korea). Animals were cared for in accordancewith the guidelines of the National Institute of Health (NIH) regarding the care anduse of laboratory animals (NIH publication 80-23, revised in 1996). Animals werehoused in groups of 6e8 under a 12-h light/dark cycle (lights on 6 am), allowed foodand water ad libitum, and acclimatized for 2 weeks. This study was approved by theEthical Committee on Animal Experimentation at Pusan National University.

2.3. Preparation of doxorubicin-loaded porous PLGA microparticles

Doxorubicin-loaded porous PLGA microparticles (Dox PLGA MPs) wereprepared by a w/o/w double emulsion-solvent evaporation technique using a gas-foaming porogen (ammonium bicarbonate), as previously described [14,18]. Briefly,20 mg of Dox was dissolved in 0.3 ml of deionized water (DW), poured into 3 ml ofmethylene chloride solution containing 150 mg of PLGA, and then sonicated in icebath using a Sonics Vibra-Cell Ultrasonic Processor (Sonics & Materials Inc. New-town, CT, USA) for 30 s at an amplitude of 50%. An aliquot (100 ml) of ammoniumbicarbonate solution (90 mg/ml) was then added to the mixture, which wassonicated again in an ice bath for 30 s. The primary emulsion obtained wasemulsified in 25 ml of an ice-cold 0.1 M sodium chloride solution (pH 7.0) con-taining 0.5% (w/v) of PEMA for 2 min at 4000 rpm using a Silverson LaboratoryMixer (model L4RT) with a 5/8-inch head (Silverson Machines, Inc. East Long-meadow, MA, USA). The resultant emulsion was poured into 50 ml of DW andallowed to evaporate under gentle stirring under an air current at 40 �C for 5 h. Thehardened MPs were harvested by centrifugation, washed three times with DW andlyophilized. Mean particle sizes of the Dox PLGA MPs produced were determined inwater using a laser diffraction particle size analyzer (Mastersizer, Malvern Instru-ments, USA) [11]. Mass median aerodynamic diameters (MMAD) were calculatedby the well-established method using density values obtained using a tappeddensity tester (Engelsmann AG, Ludwigshafen, Germany): MMAD ¼ d � (r/r0)1/2,where d is geometric mean diameter, r is the tapped density, and r0 is referencedensity (1 g/cm3) [19].

2.4. Characterization of Dox PLGA MPs

The surface morphologies of Dox PLGA MPs were investigated by scanningelectron microscopy (SEM, Hitachi S3500N, Japan). Briefly, dry MP specimens wereattached to specimen stubs using double-side tape and then sputter-coated with

gold-palladium in an argon atmosphere using a Hummer I sputter coater (AnatechLtd. St. Alexandria, VA, USA). Separately, the fluorescence images of individual DoxPLGA MPs were visualized by confocal laser scanning microscopy (CLSM, Carl ZeissMeta LSM510, Germany).

2.5. Drug loading efficiency and drug release in Dox PLGA MPs

In order to determine Dox loading efficiency, 5 mg of Dox PLGA MPs were dis-solved in 1 ml dimethylsulfoxide (DMSO), and then diluted ten times with a 7:3solution of sodium acetate buffer (pH 3.0) and methanol. The mixture was shakenthoroughly and centrifuged at 12,000 rpm for 5 min. The supernatant solutioncollected was then subjected to RP-HPLC (reversed-phase high-performance liquidchromatography) on a LiChrospher 100 RP-18 column (250 � 4.0 mm, 5 mm, Merck,Germany) at ambient temperature. Isocratic elutionwas carried out at a flow-rate of1.0 ml/min using solution A (sodium acetate buffer, pH 3.0) and solution B (meth-anol) using the following elution profile: 70% A and 30% B for 20 min. Eluates weremonitored at 480 nm. Separately, to examine drug release, 4 mg of Dox PLGA MPswere suspended in 4mL of 10mMPBS (pH 7.4) containing 0.02% (v/v) Tween 20, andgently shaken at 37 �C. At predetermined times, MP samples were centrifuged at12,000 rpm for 5 min, supernatants were discarded, and pellets were lyophilized.Amounts of Dox in samples were determined by HPLC using the extraction methodmentioned above. In addition, the surface morphologies of Dox PLGA MPs wereinvestigated by SEM and CLSM.

2.6. Cytotoxicity of Dox PLGA MPs to B16F10 cells

Murine melanoma cells, B16F10 (Korea Cell Line Bank, Seoul) were cultivated inDulbecco’s Modified Eagle’s Medium (DMEM, Gibco, Carlsbad, CA) supplementedwith 10% (v/v) fetal bovine serum (FBS, Gibco) and 1% penicillin/streptomycin(Gibco), in a 5% CO2, 95% RH incubator at 37 �C. Cells were then seeded in 12-wellplates at 2 � 104 cells/well, pre-incubated for 24 h, and further incubated withsuspended MPs (final Dox concentration w3 mg/ml) that were previously sterilizedby the UV light emission overnight,for 0, 6, or 24 h. Dox internalization into B16F10cells was visualized by CLSM, and the cytotoxic effect of Dox PLGA MPs on B16F10cells was determined by flow cytometry and FACS (BD Bioscience Mountain View,CA, USA), as described previously [20,21].

2.7. Aerosolization of Dox PLGA MPs

Images of the aerosolization of Dox PLGA MPs were captured at 0.04 s intervalsafter actuation using a digital video-camera, as described previously [12]. The airvolume used by dry powder insufflators during a single actuation was set at 1.0 ml.

2.8. Pulmonary administration and lung deposition of Dox PLGA MPs

Pulmonary administration was accomplished using a modification of a previ-ously described procedure [11,12]. In brief, male C57BL/6 mice were anesthetizedwith a single intraperitoneal (i.p.) injection of tiletamine (20 mg/kg). Freeze-driedDox PLGA MPs (w1 mg) were then directly insufflated into lungs using an insuf-flator device (DP-4M) and an air pump (AP-1) (Penn-Century, Philadelphia, PA). Anotoscope (Heine Mini 3000, Germany) was used to visualize tracheal openings. Toobserve lung deposition, mice administered Dox PLGA MPs (1 mg containingw100 mg Dox) were sacrificed at 3 and 12 h and at 1, 2, 3, 7, 10, and 14 days post-administration and entire lung lobes were excised. After brief rinsing withisotonic saline, excised lungs were visualized using a Maestro 2 in vivo imagingsystem at excitation and emission wavelengths of 523 and 560 nm, respectively(Cambridge Research & Instrumentation, Inc., Darmstadt, Germany).

2.9. Implantation of B16F10 melanoma cells into the lungs of mice and Dox PLGAMPs treatment

To obtain a metastatic lung cancer-bearing animal model, male C57BL/6 micewere injected with 100 ml of 5 � 105 B16F10 cells using a 27-gauge needle througha tail vein (0 day). Dox PLGA MPs (1 mg containing w100 mg Dox) were insufflatedinto the trachea of mice twice at 5 and 14 days after B16F10 cell implantation.Animals were sacrificed on day 21 (n ¼ 3) or day 28 (n ¼ 6) and lungs were har-vested. The mice B16F10 cell-implanted but administered Dox PLGA MPs were alsosacrificed on day 21 (n ¼ 3) or day 28 (n ¼ 6), and lungs were harvested. In addition,the lungs of mice administered Dox PLGAMPs but not B16F10 cells (n¼ 3) and thoseof age-matched untreated mice (n ¼ 3) were also harvested. In addition, the lungweight of eachmouse treated with B16F10 cells or Dox PLGAMPs was measured at 3and 4 week.

2.10. Histologies of lungs in each study group

The histologies of harvested lung tissues were evaluated using a modification ofa previously described procedure [11,12]. Lung specimens were treated withformalin, paraffin embedded, sectioned, hematoxylin and eosin (H&E) stained, and

Fig. 1. Morphologies of Dox PLGA MPs by scanning electron microscopy (SEM) (A) and confocal microscopy (CLSM) (B).

I. Kim et al. / Biomaterials 33 (2012) 5574e55835576

examined under a light microscope. Histopathologies of lung tissues in each groupwere compared.

2.11. Data analysis

Data are presented as means � SDs. Statistical significances were determinedusing the Student’s t-test, and P-values of <0.05 were considered statisticallysignificant.

Fig. 2. Doxorubicin release and the morphological characteristics of Dox

3. Results

3.1. Preparation and characterization of Dox PLGA MPs

Dox PLGA MPs were prepared using a double emulsification/solvent evaporation method; using ammonium bicarbonate asthe porogen. As shown in Fig. 1A, prepared microparticles

PLGA MPs. (A) Release profile; (B) SEM images; (C) CLSM images.

Fig. 3. CLSM images of B16F10 melanoma cells at 0, 6, and 24 h after incubation with Dox PLGA MPs (A). Flow cytometry histograms of Dox PLGA MPs (w3 mg/ml) in B16F10 cells asdetermine by FACS.

I. Kim et al. / Biomaterials 33 (2012) 5574e5583 5577

I. Kim et al. / Biomaterials 33 (2012) 5574e55835578

were found to have considerable surface porosity. CLSM images ofDox PLGA MPs showed strong red fluorescence (Fig. 1B), indi-cating a high loading of doxorubicin into the PLGA matrix. Themean particle geometric and mass median aerodynamic diame-ters (density-based) were 14.1 � 2.1 and 3.6 � 0.4 mm, respec-tively. The preparation yield of Dox PLGA MPs was found to be80.4 � 7.4%, and the encapsulation efficiency of doxorubicin intoDox PLGA MPs at a Dox loading of 20 mg per 150 mg PLGA was74.9 � 3.7%.

3.2. Doxorubicin release from Dox PLGA MPs

As shown in Fig. 2A, doxorubicin was released gradually fromDox PLGAMPs, and almost all was released at day 14 day. Dox PLGA

Fig. 4. (A) Aerosolization of Dox PLGA MPs using a dry powder insufflator at 0.02 s after acmice using a dry powder insufflator and an otoscope.

MPs were degraded gradually with incubation time in PBS, andsurface pores were found to aggregate (probably due to thedecomposition of PLGA) (Fig. 2B). CLSM images showed that the redfluorescence intensity of doxorubicin in Dox PLGA MPs decreasedwith time (Fig. 2C). In addition, the apparent particle sizes of DoxPLGA MPs gradually decreased with increasing the incubation timedue to PLGA degradation.

3.3. Cytotoxicity of Dox PLGA MPs by FACS

As shown in Fig. 3A, doxorubicin released from Dox PLGA MPswas increasingly incorporated into B16F10 cells, as indicated byincreasing red fluorescence. Thus, almost all B16F10 cells seemed tobe influenced by the cytotoxic effect by doxorubicin. FACS showed

tuation. (B) Photograph of the pulmonary administration of Dox PLGA MPs to C57BL/6

I. Kim et al. / Biomaterials 33 (2012) 5574e5583 5579

that 81.6 and 88.8% of B16F10 cells were killed at 6 and 24 h,respectively (Fig. 3B).

3.4. Aerosolization and pulmonary administration of Dox PLGA MPs

Fig. 4A shows that Dox PLGA MPs aerosolization was accom-plished w0.06 s after actuation. Aerosolized particles were deeppink due to the incorporated doxorubicin and displayed goodmobility. Dox PLGA MPs were directly administered into thetrachea of C57BL/6 mice using a dry powder insufflator (Fig. 4B).

3.5. Lung deposition of Dox PLGA MPs

Lung deposition images were obtained over 14 days afteradministering Dox PLGA MPs to C57BL/6 mice. RGB images showthe distribution of Dox PLGA MPs in mouse lungs. At 3 h afteradministration, Dox PLGAMPs appeared to be located the centers oflungs, and then spread throughout the lungs. At one week afteradministration, considerable number of Dox PLGAMPs remained inlungs, but the intensity of coloration diminished. At 14 days after

Fig. 5. Monitoring of the lung deposition of Dox PLGA MPs in C57BL/

administration only residual levels of doxorubicinwere observed inlung tissues (Fig. 5).

3.6. Anti-tumor efficacy of Dox PLGA MPs on B16F10 melanomametastatic cancers

Macroscopic lung features were investigated after B16F10implantation and/or pulmonary Dox PLGA MPs treatment. Asshown in Fig. 6, lungs excised from treatment naive exhibitednormal physiological features, and the lungs of mice treated withDox PLGA MPs resembled those of naive controls at 21 and 28 daysafter implantation. However, the lungs of B16F10 melanoma-bearing mice showed obvious tumors. Interestingly, the lungs ofmice treated both performed (B16F10 implantation and pulmonaryDox PLGA MPs treatment) clearly displayed much reduced tumortissues in terms of mass and number versus B16F10 melanoma-bearing mice. Specifically, the average lung weight of micetreated with both B16F10 cells and Dox PLGA MPs was determinedto be significantly lower than that of mice treated with only B16F10cells at 4 week (384.3 � 76.4 vs. 723.0 � 215.4 mg; P ¼ 0.0046)(Fig. 7).

6 mice using a RGB spectrum until 14 days post-administration.

Fig. 6. Photographs of lungs obtained from C57BL/6 mice according to B16F10 cell implantation or pulmonary Dox PLGA MPs administration.

I. Kim et al. / Biomaterials 33 (2012) 5574e55835580

3.7. Histological evaluation of lung tissues in B16F10 melanomametastasis mice

At 3 and 4 weeks post-implantation, the histologies of the lungsin each group were investigated microscopically by H&E staining.

Fig. 7. Lung weights of C57BL/6 mice according to B16F10 cell implantation orpulmonary Dox PLGA MPs administration.

As shown in Fig. 8, the lung tissue specimens of treatment naiveand Dox PLGA MPs treated mice were not significantly different.Actually, numerous pores, a typical physiological feature of cross-sectional lung specimen, were seen in both mice group, and theircolors (dark blue/purple) by H&E staining demonstrated that thelung cells from each mouse group are robust or well tolerable. Onthe other hand, the lung tissue sections of melanoma-bearing miceshowed many melanomas, and at 4 weeks post-implantation,a great number of pulmonary parenchyma seemed to be replacedbymelanomas. Especially, the lung specimen of melanoma-bearingmice without Dox PLGA MPs treatment seldom showed typicalpores, unlike in the lung specimen of non-treated negative controlgroup mice. In contrast, the lungs of mice treated with B16F10 cellsand Dox PLGA MPs showed obvious reductions in melanoma masssizes and numbers and many normal pores.

4. Discussion

The lungs are an essential organ where human breathe, and the5-year survival rate of lung cancer is less than 10% [22]. Further-more, the lungs are vulnerable to metastatic processes initiated bytumor cells from other tissues or organs, and lung cancer is itselfhighly metastatic in nature [2,7]. The injection-based chemother-apies used to treat lung cancer are associated with many clinicalproblems, such as, low delivery efficiency, low anti-tumor efficacy,systemic toxicity, and side effects [7,23]. Here, we described a newinhalation-based sustained-release system for the treatment ofmetastatic lung cancer. This pulmonary system can deliver anti-

Fig. 8. Histologies of lung tissues in the four study groups at 3 or 4 weeks after B16F10 cell implantation (5 � 105 ea, 0 day) and/or pulmonary administration of Dox PLGAMPs (5, 14day). (A) age-matched treatment naive control mouse; (B) Dox PLGA MP administered mouse; (C) B16F10-implanted mouse; and (D) B16F10-implanted and Dox PLGA MPsadministered mouse.

I. Kim et al. / Biomaterials 33 (2012) 5574e5583 5581

cancer drugs to lung tumors with high efficiency, and is likely tominimize damage to other fast-growing tissues. Doxorubicin wasconsidered a good model anti-cancer drug for an inhalationdelivery because its administration by injection is limited by itssevere cardio-toxicity [20].

Large porous polymeric microparticles provide optimal deliverysystems for long-term treatment via inhalation. Such particles havetwo advantages, namely; (i) high aerosolization efficiency and (ii)macrophage bypass [15,16]. It has been well established thatporosities of microparticles decreases aerodynamic densitiessufficiently for aerosolization [16]. More specifically, porousmicroparticles with a MMAD of 1e5 mm have been shown toaerosolize well and penetrate deeply into lungs. Furthermore, it hasbeen estimated that wsix billion alveolar macrophages (14e22 mmin size) are present in the deep lungs, which are capable of rapidlyphagocytosing small inhaled microparticles (1e2 mm). Hence, largeporous particles of >10 mm in diameter are a prerequisite for aninhalatory system. In this respect, the prepared Dox PLGA MPs(diameter and MMAD 14.1 � 2.1 and 3.6 � 0.4 mm, respectively)appear to be suitable, and have good aerosolization and in vivoparticle mobility characteristics (Fig. 4A).

Many methods have been devised to make porous microparti-cles, such as, methods that use; (i) extractable porogens (e.g.,pluronics) [24], (ii) osmogens (e.g., cyclodextrins) [12,13], (iii)effervescent salts (ammonium bicarbonate) [15], and (iv) gasbubbles (hydrogen peroxide þ catalase) [25]. Extractable porogensand osmogens produce many water channels in PLGA matrices toform surface pores, and thus, considerable amounts of doxorubicin(as hydrochloride) could leach from these matrices and reduceencapsulation efficiencies. On the other hand, ammonium bicar-bonate is a carbon dioxide-producing foaming agent, and does notappear to accelerate the leach-out of doxorubicin [18]. Thus,a doxorubicin encapsulation efficiency of 74.9 � 3.7% was achieveddespite high microparticles surface porosities. However, nonethe-less such porosity, significant initial burst release from Dox PLGAMPs was not found, and such character may decrease the

probability of harming normal respiratory cells in the context ofsafety issue (Fig. 2A). Moreover, the sustained doxorubicin releasehas potential of continual killing malignant cells in lung tumortissues, which was responsible for improved anti-tumor efficacywithout frequent inhalation.

The inability of mice to spontaneously inhale Dox PLGAMPs dueto an anesthetized state is a shortcoming of our in vivomodel. Eventhough it is conscious, the spontaneous inhalation accompanied bydeep breath in mice seems probably impossible, otherwise per-formed in an inhalation chamber instrument. Furthermore,delivery efficiency to the lungs is known to be low for aerosoladministration, and thus, we chose to use a dry powder insufflationmethod, which guarantees high local delivery efficiency to thelungs. Unlike intrapulmonary delivery for systemic absorption, oursystem did not require non-stop particle movement to deep lungs(e.g. alveoli) because the tumors were located throughout the lungsof mice (Fig. 6). Nevertheless, immediately after insufflation, DoxPLGA MPs were found only distributed into central lungs (i.e.,bronchi to bronchioles), and then to gradually spread throughoutthe lungs for 2 days, mainly by sedimentation and Browniandiffusion, as reported previously [9e12].

Metastatic lung cancer preclinical models based on B16F10melanoma cell implantation via a tail vein are well-established andconsidered efficient, because they produce massive metastatictumors in only w2e3 weeks [7]. Furthermore, B16F10 cells arehighly sensitive to doxorubicin (IC50, 0.3 mg/ml, data not shown)[26], and thus doxorubicin readily internalizes into the cells. Also,>90% B16F10 cells are killed at a doxorubicin concentration ofw3 mg/ml [26], and thus this concentration was chosen for thecytotoxicity test. From the flow cytometry analysis (Fig. 3B), B16F10cells were killed mainly by necrosis due to DNA damage, but suchcells also seemed to be killed via apoptosis induced in response tosuch DNA damage [27e29]. Accordingly, doxorubicin released fromDox PLGA MPs might have the ability to kill melanoma cells in thelungs of mice within a day, as shown in Fig. 3. As shown in Fig. 5,Dox PLGAMPs appeared to stay in lungs forw2 weeks, presumably

Fig. 9. An illustrated cartoon for the understanding of anti-tumor effects of pulmonary administered Dox PLGA MPs in lung cancer.

I. Kim et al. / Biomaterials 33 (2012) 5574e55835582

due to their particle sizes (14 mm) large enough for macrophagebypass [15,18]. Therefore, doxorubicin could be gradually releasedfrom Dox PLGAMPs over 2 weeks. In the present study, melanomasin Dox PLGA MPs administered mice were clearly reduced versusnon-Dox PLGA MP treated mice. This anti-tumor efficacy was alsodemonstrated by microscopic analysis of lung sections, that is,tumor tissues decreased remarkably in Dox PLGA MPs-administered mice vs. non-administered mice. Furthermore, DoxPLGA MPs seemed not to have significant toxic effects on normallungs at the doses administered macroscopically or microscopi-cally, when compared with the lungs from negative control groupmice. This fact might be partially due to much better physiologicaldefense systems, i.e., respiratory mucus lining or mucocilliaryclearance, of the lungs of normal mice than those of damaged lungsby tumor infiltration.

PLGA is considered safe due its biodegradability and biocom-patibility because many PLGA-based injection depots have beenapproved by the FDA. Nevertheless, it has been reported to invokeinflammation or necrosis at injection sites [30]. Therefore, the long-term clinical administration of inhalatory PLGA should beapproached more cautiously because the marginal duration forinhaled particle deposition is controversial [11,31]. Nevertheless,chemotherapeutics-loaded sustained-release inhalation systemsseem to be relatively free of such concerns in lung cancer patients,given the seriousness of the condition. Hence, we believe if theapplication system is well optimized in terms of selective targetinglung tumors that it has clinical potential.

5. Conclusion

In the present study, we prepared inhalable and highly porouslarge doxorubicin-loaded PLGA microparticles for treating lungcancer. These particles were found to have desirable aerosolization,high encapsulation, and phagocytosis escapement characteristics.These doxorubicin-loaded microparticles were also found to bedeposited in the lungs of mice and to remain in situ for up to 14days. Remarkably, these doxorubicin-loaded microparticlesadministered by inhalation greatly reduced the masses andnumbers of metastatic lung tumors caused by B16F10 cellimplantation in C57BL/6 mice. Furthermore, doxorubicin-loadedmicroparticles were not found to have significant toxic effect onthe lung tissues of healthy mice. To date, no inhalation-type anti-tumor agents for lung cancer are developed, although manyinjection-based agents show serious side-effects to cancer patients.In this respect, this doxorubicin-loaded PLGAmicroparticles systemwould be a promising potential inhalation agent with long-termsustained-release characteristics for the treatment of lung tumors(Fig. 9).

Acknowledgments

This work was supported by the National Research Foundationof Korea (NRF), supported by the Korea government (MEST)(#20110012711).

I. Kim et al. / Biomaterials 33 (2012) 5574e5583 5583

References

[1] Johnson DH. Treatment strategies for metastatic non small-cell lung cancer.Clin Lung Cancer 1999;1:34e41.

[2] Carballo M, Maish MS, Jaroszewski DE, Holmes CE. Video-assisted thoracicsurgery (VATS) as a safe alternative for the resection of pulmonary metas-tases: a retrospective cohort study. J Cardiothorac Surg 2009;4:13.

[3] Goto H, Yano S, Zhang H, Matsumori Y, Ogawa H, Blakey DC, et al. Activity ofa new vascular targeting agent, ZD6126, in pulmonary metastases by humanlung adenocarcinoma in nude mice. Cancer Res 2002;62:3711e5.

[4] Tseng CL, Su WY, Yen KC, Yang KC, Lin FH. The use of biotinylated-EGF-modified gelatin nanoparticle carrier to enhance cisplatin accumulation incancerous lungs via inhalation. Biomaterials 2009;30:3476e85.

[5] Burris 3rd HA. Shortcomings of current therapies for non-small-cell lungcancer: unmet medical needs. Oncogene 2009;28:S4e13.

[6] Wauthoz N, Deleuze P, Hecq J, Roland I, Saussez S, Adanja I, et al. In vivoassessment of temozolomide local delivery for lung cancer inhalation therapy.Eur J Pharm Sci 2010;39:402e11.

[7] Garbuzenko OB, Saad M, Pozharov VP, Reuhl KR, Mainelis G, Minko T. Inhi-bition of lung tumor growth by complex pulmonary delivery of drugs witholigonucleotides as suppressors of cellular resistance. Proc Natl Acad Sci USA2010;107:10737e42.

[8] Tseng CL, Wu SY, Wang WH, Peng CL, Lin FH, Lin CC, et al. Targeting efficiencyand biodistribution of biotinylated-EGF-conjugated gelatin nanoparticlesadministered via aerosol delivery in nude mice with lung cancer. Biomaterials2008;29:3014e22.

[9] Patton JS, Byron PR. Inhaling medicines: delivering drugs to the body throughthe lungs. Nat Rev Drug Discov 2007;6:67e74.

[10] Youn YS, Lee KC, Bae YH, Na K, Lee ES. Advanced pulmonary delivery ofpeptides or proteins using polymeric particles. In: Jorgensen L, Nielsen HM,editors. Delivery technologies for biopharmaceuticals: peptides, proteins,nucleic acids, and vaccines. WILEY; 2009. p. 228e44.

[11] Kim H, Park H, Lee J, Kim TH, Lee ES, Oh KT, et al. Highly porous largepoly(lactic-co-glycolic acid) microspheres adsorbed with palmityl-acylatedexendin-4 as a long-acting inhalation system for treating diabetes. Biomate-rials 2011;32:1685e93.

[12] Kim H, Lee J, Kim TH, Lee ES, Oh KT, Lee DH, et al. Albumin-coated poroushollow poly(lactic-co-glycolic acid) microparticles bound with palmityl-acylated exendin-4 as a long-acting inhalation delivery system for the treat-ment of diabetes. Pharm Res 2011;28:2008e19.

[13] Kwon MJ, Bae JH, Kim JJ, Na K, Lee ES. Long acting porous microparticles forpulmonary drug delivery. Int J Pharm 2007;333:5e9.

[14] Kim TK, Yoon JJ, Lee DS, Park TG. Gas foamed open porous biodegradablepolymeric microspheres. Biomaterials 2006;27:152e9.

[15] Edwards DA, Hanes J, Caponetti G, Hrkach J, Ben-Jebria A, Eskew ML, et al.Large porous particles for pulmonary drug delivery. Science 1997;276:1868e71.

[16] Edwards DA, Ben-Jebria A, Langer R. Recent advances in pulmonary drugdelivery using large, porous inhaled particles. J Appl Physiol 1998;85:379e85.

[17] Vanbever R, Mintzes JD, Wang J, Nice J, Chen D, Batycky R, et al. Formulationand physical characterization of large porous particles for inhalation. PharmRes 1999;16:1735e42.

[18] Yang Y, Bajaj N, Xu P, Ohn K, Tsifansky MD, Yeo Y. Development of highlyporous large PLGA microparticles for pulmonary drug delivery. Biomaterials2009;30:1947e53.

[19] Ungaro F, De Rosa G, Miro A, Quaglia F, La Rotonda MI. Cyclodextrins in theproduction of large porous particles: development of dry powders for thesustained release of insulin to the lungs. Eur J Pharm Sci 2006;28:423e32.

[20] Bae S, Ma K, Kim TH, Lee ES, Oh KT, Park ES, et al. Doxorubicin-loaded humanserum albumin nanoparticles surface-modified with TNF-related apoptosis-inducing ligand and transferrin for targeting multiple tumor types. Biomate-rials 2012;33:1536e46.

[21] Jiang HH, Kim TH, Lee S, Chen X, Youn YS, Lee KC. PEGylated TNF-relatedapoptosis-inducing ligand (TRAIL) for effective tumor combination therapy.Biomaterials 2011;32:8529e37.

[22] Welch HG, Schwartz LM, Woloshin S. Are increasing 5-year survival ratesevidence of success against cancer? JAMA 2000;283:2975e8.

[23] Roa WH, Azarmi S, Al-Hallak MH, Finlay WH, Magliocco AM, Löbenberg R.Inhalable nanoparticles, a non-invasive approach to treat lung cancer ina mouse model. J Control Release 2011;150:49e55.

[24] Kim HK, Chung HJ, Park TG. Biodegradable polymeric microspheres with“open/closed” pores for sustained release of human growth hormone.J Control Release 2006;112:167e74.

[25] Bae SE, Son JS, Park K, Han DK. Fabrication of covered porous PLGA micro-spheres using hydrogen peroxide for controlled drug delivery and regenera-tive medicine. J Control Release 2009;133:37e43.

[26] Veronese FM, Schiavon O, Pasut G, Mendichi R, Andersson L, Tsirk A, et al.PEG-doxorubicin conjugates: influence of polymer structure on drug release,in vitro cytotoxicity, biodistribution, and antitumor activity. Bioconjug Chem2005;16:775e84.

[27] Co NN, Tsang WP, Wong TW, Cheung HH, Tsang TY, Kong SK, et al. OncogeneAF1q enhances doxorubicin-induced apoptosis through BAD-mediated mito-chondrial apoptotic pathway. Mol Cancer Ther 2008;7:3160e8.

[28] Li L, Lu Q, Shen Y, Hu X. Schisandrin B enhances doxorubicin-inducedapoptosis of cancer cells but not normal cells. Biochem Pharmacol 2006;71:584e95.

[29] Xia M, Knezevic D, Vassilev LT. p21 does not protect cancer cells fromapoptosis induced by nongenotoxic p53 activation. Oncogene 2011;30:346e55.

[30] Shive MS, Anderson JM. Biodegradation and biocompatibility of PLA and PLGAmicrospheres. Adv Drug Deliv Rev 1997;28:5e24.

[31] Baker GL, Gupta A, Clark ML, Valenzuela BR, Staska LM, Harbo SJ, et al.Inhalation toxicity and lung toxicokinetics of C60 fullerene nanoparticles andmicroparticles. Toxicol Sci 2008;101:122e31.