Hollow Silica Nanocontainers as Drug Delivery Vehicles

Langmuir 2008, 24, 3417-3421

3417

Hollow Silica Nanocontainers as Drug Delivery Vehicles Jaemoon Yang,† Jaewon Lee,† Jinyoung Kang,†,‡ Kwangyeol Lee,§ Jin-Suck Suh,| Ho-Geun Yoon,⊥ Yong-Min Huh,*,‡,| and Seungjoo Haam*,‡,| Department of Chemical Engineering, Yonsei UniVersity, Seoul 120-749, Korea, Graduate Program for Nanomedical Science, Yonsei UniVersity, Seoul 120-752, Korea, Department of Chemistry, Korea UniVersity, Seoul 136-701, Korea, Department of Radiology, College of Medicine, Yonsei UniVersity, Seoul 120-752, Korea, and Department of Biochemistry and Molecular Biology, Yonsei UniVersity, Seoul 120-752, Korea ReceiVed June 8, 2007. In Final Form: September 7, 2007 Novel hollow silica nanoparticles (HSNPs) for drug delivery vehicles were synthesized using silica-coated magnetic assemblies, which are composed of a number of Fe3O4 nanocrystals, as templates. The core cavity was obtained by removal of Fe3O4 phase with hydrochloric acid and subsequent calcination at a high temperature. HSNPs were modified by amine in order to introduce positive surface charge and further PEGylated for increased solubility in aqueous medium. Doxorubicin as a model drug was loaded into the HSNPs, and notable sustained drug release from HSNPs was demonstrated.

Introduction The synthesis of novel nanomaterials with well-tailored properties is a major challenge in biomedical applications for drug delivery carriers, diagnostic agents, sensing probes, and tracking labels.1 Due to the monodispersity, large surface area, high drug loading efficiency, and potential for hybridization with other organic/inorganic materials, hollow nanostructures for drug containers were prepared and tested as a drug delivery system.2 Hollow nanoparticles can be synthesized with various materials such as organic polymer, silicates, carbon, titania, and phosphates.3 To produce the cavity in the nanoparticle, various removable templates, such as polymeric micelles or surfactants, silicates, gold nanoparticles, and luminescence semiconductors, were used.4 In particular, biocompatible silica nanoparticles have been extensively used due to easy formation and a convenient surface modification procedure.5,6 However, silica nanoparticles without * To whom correspondence should be addressed. S.H.: phone, 82-22123-2751; fax, 82-2-312-6401; e-mail, [email protected]. Y.-M.H.: phone, 82-2-2228-2375; fax, 82-2-362-8647; e-mail, [email protected]. † Department of Chemical Engineering, Yonsei University. ‡ Graduate Program for Nanomedical Science, Yonsei University. § Department of Chemistry, Korea University. | Department of Radiology, College of Medicine, Yonsei University. ⊥ Department of Biochemistry and Molecular Biology, Yonsei University. (1) (a) Yang, J.; Lee, C.-H.; Park, J.; Seo, S.; Lim, E.-K.; Song, Y. J.; Suh, J.-S.; Yoon, H.-G.; Huh, Y.-M.; Haam, S. J. Mater. Chem. 2007, 17, 2695-2699. (b) Sengupta, S.; Eavarone, D.; Capila, I.; Zhao, G.; Watson, N.; Kiziltepe, T.; Sasisekharan, R. Nature 2005, 28, 568-572. (c) Koide, A.; Kishimura, A.; Osada, K.; Jang, W.-D.; Yamasaki, Y.; Kataoka, K. J. Am. Chem. Soc. 2006, 128, 59885989. (d) Lee, J.; Huh, Y.-M.; Jun, Y.; Seo, J.; Jang, J.; Song, H.; Kim, S.; Cho, E.; Yoon, H.; Suh, J.; Cheon, J. Nat. Med. 2007, 13, 95-99. (e) Quinti, L.; Weissleder, R.; Tung, C. H. Nano Lett. 2006, 6, 488-490. (f) Song, H.-T.; Choi, J.-S.; Huh, Y.-M.; Kim, S.; Jun, Y.-W.; Suh, J.-S.; Cheon, J. J. Am. Chem. Soc. 2005, 127, 5732-5733. (2) (a) Zhu, Y.; Shi, J.; Li, Y.; Chen, H.; Shen, W.; Dong, X. Microporous Mesoporous Mater. 2005, 85, 75-81. (b) Hao, L.; Gong, X.; Xuan, S.; Zhang, H.; Gong, X.; Jiang, W.; Chen, Z. Appl. Surf. Sci. Appl. Surf. Sci. 2006, 252, 8724-8733. (c) Li, Z.-Z.; Wen, L.-X.; Shao, L.; Chen, J.-F. J. Controlled Release 2004, 98, 245-254. (d) Arruebo, M.; Fernandez-Pacheco, R.; Irusta, S.; Arbiol, J.; Ibarra, M. R.; Santamaria, J. Nanotechnology 2006, 17, 4057-4064. (3) (a) Huo, Q.; Liu, J.; Wang, L. Q.; Jiang, Y.; Lambert, T. N.; Fang, E. J. Am. Chem. Soc. 2006, 126, 6447-6453. (b) Deng, Z.; Chen, M.; Zhou, S.; You, B.; Wu, L. Langmuir 2006, 22, 6403-6407. (c) Su, F.; Zhao, X. S.; Wang, Y.; Wang, L.; Lee, J. Y. J. Mater. Chem. 2006, 16, 4413-4419. (d) Zhang, K.; Zhang, X.; Chen, H.; Chen, X.; Zheng, L.; Zhang, J.; Yang, B. Langmuir 2004, 20, 11312-11314. (e) Tjandra, W.; Ravi, P.; Yao, J.; Tam, K. C. Nanotechnology 2006, 17, 5988-5994.

any modification as drug carriers demonstrated a rapid drug release from the nanostructure.2c,d To accomplish sustained release with a hollow structure, a surface modification process seems to be indispensable.2a,6 In this paper, thus, hollow silica nanoparticles (HSNPs) with a large cavity were synthesized as nanocontainers for drug delivery vehicles. As templates for the embodiment of a large cavity in the silica nanoparticles, magnetic assemblies (MAs) were prepared by clustering Fe3O4 magnetic nanocrystals with a surfactant binder. Acting as seeds, MAs were then wrapped with a silica shell by the modified Sto¨ber method.7 The resulting magnetic silica nanoparticles (MSNPs) were treated with hydrochloric acid and calcinated at high temperature to form a large core cavity. Although these HSNPs could load a large amount of drug, release from the HSNPs was too fast to control. In order to control the drug release, we surface-modified HSNPs, as illustrated in Figure 1, and demonstrated the sustained drug release from the hollow structure prepared in this work. Experimental Methods Materials. Iron(III) acetylacetonate, 1,2-hexadecanediol, dodecanoic acid, dodecylamine, benzyl ether, 3-aminopropyltrimethoxy silane, anhydrous dichloromethane, succinic anhydride, 4-dimethylaminopyridine, triethylamine, polyethylene glycol (PEG; Mw, 6 kDa), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide, N-hydroxysuccinimide, tetraethylorthosilicate (TEOS), and poly(vinyl alcohol) (PVA; Mw, 15-20 kDa) were purchased from Sigma-Aldrich. Doxorubicin (DOX) was obtained from Fluka, and phosphatebuffered saline (PBS, 10 mM; pH 7.4) was purchased from Gibco. All other chemicals and reagents were of analytical grade. (4) (a) Khanal, A.; Inoue, Y.; Yada, M.; Nakashima, K. J. Am. Chem. Soc. 2007, 129, 1534-1535. (b) Xu, H.; Wang, W. Angew. Chem., Int. Ed. 2007, 46, 1489-1492. (c) Yoon, B. S.; Kang, S.; Yu, J.-S. Curr. Appl. Phys. 2006, 6, 1054-1058. (d) Marinakos, S. M.; Novak, J. P.; Brousseau, L. C.; Blaine House, A.; Edeki, E. M.; Feldhaus, J. C.; Feldheim, D. L. J. Am. Chem. Soc. 1999, 121, 8518-8522. (e) Darbandi, M.; Thomann, R.; Nann, T. Chem. Mater. 2007, 19, 1700-1703. (5) (a) Wu, W.; Caruntu, D.; Martina, A.; Yua, M. H.; O’Connora, C. J.; Zhoua, W. L.; Chenb, J.-F. J. Magn. Magn. Mater. 2007, 311, 578-582. (b) Wang, J.; Xiao, Q.; Zhou, H.; Sun, P.; Yuan, Z.; Li, B.; Ding, D.; Shi, A.-C.; Chen, T. AdV. Mater. 2006, 18, 3284-3288. (c) Wu, H.; Liu, Z.; Wang, X.; Zhao, B.; Zhang, J.; Li, C. J. Colloid Interface Sci. 2006, 302, 142-148. (6) Kim, J.; Lee, J. E.; Lee, J.; Yu, J. H; Kim, B.C.; An, K.; Hwang, Y.; Shin, C.-H.; Park, J.-G.; Kim, J.; Hyeon, T. J. Am. Chem. Soc. 2006, 128, 688-689. (7) Graf, C.; Blaaderen, A. Langmuir 2002, 18, 524-534.

10.1021/la701688t CCC: $40.75 © 2008 American Chemical Society Published on Web 03/07/2008

3418 Langmuir, Vol. 24, No. 7, 2008

Yang et al.

Figure 2. Three types of HSNPs as drug delivery vehicles: (a) HSNPs-OH, (b) HSNPs-NH2, and (c) HSNPs-PEG.

Figure 1. Schematic of drug-loaded hollow silica nanoparticles for drug delivery vehicles. Synthesis of Magnetic Manocrystals (MNCs). Benzyl ether (20 mL) was mixed with 2 mmol of iron(III) acetylacetonate, 10 mmol of 1,2-hexadecanediol, 6 mmol of dodecanoic acid, and 6 mmol of dodecylamine under ambient nitrogen conditions. The mixture was preheated to 150 °C for 30 min and heated to reflux at 300 °C for 30 min. The reactants were cooled to room temperature, and the products were purified with excess pure ethanol. Magnetic nanocrystals with approximate diameters of 12 nm were synthesized using the seed-mediated growth method.1d,8 Synthesis of Magnetic Silica Nanoparticles (MSNPs). To prepare the MAs, 5 mg of magnetic nanocrystals was dissolved in 4 mL of chloroform. The organic phase was added to 10 mL of an aqueous phase containing 200 mg of poly(vinyl alcohol) (PVA).1a After mutual saturation of the organic and continuous phase, the mixture was emulsified for 10 min with an ultrasonicator (ULH700S, Ulssohitech, Korea) at 200 W. After evaporation of the organic solvent, the products were purified by filtration and centrifuged at 21k rpm. The precipitates were redispersed in water, and the MA diameter was 45 ( 2.3 nm. The MSNPs containing MAs were synthesized by the modified Sto¨ber method. The particles were prepared in a mixture of alcohol and water at ambient temperature using MA seeds. A 1 mL amount of MA (5 mg/mL) was diluted with 4 mL of ethyl alcohol and 0.1 mL of ammonia solution (28% in water). TEOS (60 µL) was slowly added to the suspension and stirred for 12 h such that after hydrolysis and condensation a silica outer shell formed on the MA surface.7 Synthesis of Hollow Silica Nanoparticles (HSNPs). For preparation of hydroxyl-terminated HSNPs (HSNPs-OH), 50 mg of MSNPs was dispersed in D.D.I. water (5 mL) and 4 mL of HCl (37%) was added to the reaction mixture. After a while, the dark brown solution changed to a bright yellow color due to dissolution of the magnetic nanocrystals.9 After the MAs were completely eliminated by HCl etching, the product was precipitated with centrifugation at 10k rpm. The precipitates were washed several times using D.D.I. water, and the products were frozen and vacuum dried. The dried nanoparticles were calcinated at 300 °C to eliminate the organic compounds. To inhibit fast release of DOX from HSNPs-OH, amine-terminated HSNPs (HSNPs-NH2) were synthesized using 3-aminopropyl trimethoxy silane (APTMS, 97%). A 10 mg amount of HSNPs-OH was added to 0.05 mL of APTMS in deionized water. The mixture was vigorously stirred at 70 °C for 12 h. The products were washed several times aided with centrifugation. Carboxyl-terminated PEG (PEG-diCOOH) was synthesized for preparation of PEGylated HSNPs (HSNPs-PEG). PEG (0.5 mmol) was initially dissolved in dioxane (10 mL) and activated by adding (8) Sun, S.; Zeng, H.; Robinson, D. B.; Raoux, S.; Rice, P. M.; Wang, S. X.; Li, G. J. Am. Chem. Soc. 2004, 126, 273-279. (9) Rochelle, M.; Cornell, R. M.; Schwertmann, U. The Iron Oxides: Structure, Properties, Reactions, Occurrences and Uses, 2nd ed.; Wiley-VCH: New York, 2003.

2.0 mmol of succinic anhydride, 4-dimethylaminopyridine, and triethylamine.10 The reaction was performed for 24 h at room temperature under a nitrogen atmosphere. The products were filtered and purified with excess carbon tetrachloride. The precipitates by ethyl ether were dried under vacuum and stored until further use. A 20 mg amount of HSNPs-NH2 and 0.1 mmol of PEG-diCOOH were dissolved in PBS (500 µL). A 2 mmol amount of 1-ethyl-3(3-dimethylaminopropyl)-carbodiimide and N-hydroxysuccinimide were added to the above solution and incubated for 4 h. The unreacted materials were filtered, purified, and centrifuged (rpm, 21k). The three HSNPs (HSNPs-OH, HSNPs-NH2, and HSNPs-PEG) are illustrated in Figure 2. Preparation of Drug Delivery Vehicles. A 5 mg amount of DOX and 10 mg of HSNPs (HSNPs-OH, HSNPs-NH2, or HSNPsPEG) were mixed in 4 mL of PBS for 24 h. The drug-loaded HSNPs were precipitated with 10k of centrifugation and stored under subzero temperatures. For the drug release experiment, 2 mL of drug-loaded HSNPs suspension (5 mg/mL) in a dialysis tube was immersed in 10 mL of PBS at 37 °C. Released drug was monitored by a UV spectrophotometer (Optizen 2120UV, MECASYS Co.) at 480 nm. For induction of drug release from the prepared HSNPs, drug carriers were treated under ultrasonication (350 W) at room temperature. Released drugs were collected by centrifugation (5k rpm). The drug loading content and entrapment efficiency were calculated by the following equations drug loading contents (%) )

weight of drug in nanoparticles × weight of prepared nanoparticles 100 (1)

entrapment efficiency (%) )

weight of drug in nanoparticles × weight of injected drug 100 (2)

Characterization. MNCs, MAs, MSNPs, and HSNPs structures were confirmed by transmission electron microscopy (TEM, JEM2100, JEOL Ltd.). The size distribution of the nanoparticles was analyzed by laser scattering (ELS-Z, Otsuka Electronics). FT-IR spectra (ExcaliburTM series, Varian Inc.) analysis was used to confirm the characteristic bands of the synthesized nanoparticles. X-ray diffraction measurements were performed by a Rigaku D/maxRB (Tokyo, Japan) powder diffractometer and image-plate photography using graphite-monochromatized Cu KR radiation (λ ) 1.542 Å) to determine the crystallinity of the MNCs in the MSNPs. Data were collected from 20° to 80° with a step size of 0.05° and step time of 5 s. Nitrogen adsorption/desorption measurements were performed at liquid nitrogen temperature (77 K) with a Micromertics ASAT2020 apparatus. Prior to the measurements the samples were degassed for 2 h at 120 °C. The surface area was determined from the adsorption branch of the isotherm according to the BrunauerEmmett-Teller (BET) method in the relative pressure range 0.02 < P/P0 < 0.2. The total pore volume was calculated from the volume of adsorbed nitrogen at a relative pressure of P/P0 ) 0.97.

Results and Discussion For preparation of drug delivery vehicles, hollow silica nanoparticles (HSNPs) were synthesized using silica-coated (10) Lee, S.-H.; Kim, S. H.; Han, Y.-K.; Kim, Y. H. J. Polym. Sci.: Part A 2002, 40, 2545-2552.

Silica Nanocontainers as Drug DeliVery Vehicles

Figure 3. Transmission electron microscopy images of (a) MNCs, (b) MAs, (c) MSNP, and (d) HSNPs.

magnetic assemblies, which are composed of a number of Fe3O4 nanocrystals, as templates. Magnetic nanocrystals approximately 10 nm in diameter were synthesized in an organic phase (benzyl ether) (Figure 3a). Magnetic assemblies (MAs) of Fe3O4 nanoparticles were prepared with the nanoemulsion method using poly(vinyl alcohol) as a surfactant. The hydrophobic magnetic nanocrystals were well encapsulated by amphiphilic poly(vinyl alcohol), and the spherical morphology is shown in Figure 3b. Laser scattering indicated that the MA diameter was 45.3 ( 5.9 nm, and the surface charge was nearly zero due to the nonionic poly(vinyl alcohol) coating.11 The silica layer on MAs was formed by hydrolysis and condensation of the TEOS. Due to the presence of seed MAs, MSNPs with core-shell structures were successfully synthesized (Figure 3c). The MSNP diameter was 80.9 ( 9.3 nm, and compared with the MAs, a negative surface charge was observed due to the hydroxyl group of the silica layer.12 The MSNP zeta potential was -35.6 ( 7.8 mV. To prepare HSNPs-OH, inorganic (magnetic nanocrystals) and organic (poly(vinyl alcohol)) components were eliminated using acid etching and calcinations, respectively. First, the large core cavity was obtained by completely removing Fe3O4 with hydrochloric acid (37 wt %).9 The strong acid induced the dissolution of magnetic compounds of the MSNPs. To remove the poly(vinyl alcohol) polymeric compounds, the magnetic components were calcinated at 300 °C. The TEM image of HSNPs-OH is presented in Figure 3d, clearly showing the nanoparticle cavity. The size of the HSNPs-OH was 83.1 ( 8.9 nm, and the zeta potential was -43.0 ( 3.2 mV (Figure 4a). Figure 4b shows the suspension of MSNPs (i) and HSNPs-OH (ii) in the aqueous phase. Due to the dissolution and elimination of magnetic components, the suspension color changed from dark brown to white gray. After hydrochloride treatment and calcination of the MSNPs, the removed magnetic MSNPs components were evaluated using power XRD (Figure 5). The XRD pattern of the MSNPs in Figure 5a represents the inverse spinel structure of Fe3O4 and silicate. After the magnetic and organic components were removed from the MSNPs process, however, the Fe3O4 phase could not be detected in the XRD pattern (Figure 5b). The FT-IR spectra were consistent with the XRD data (Figure 6); the Fe-O band (∼580 cm-1) was not observed for HSNPs-OH due to elimination of magnetic components. (11) DeMerlis, C. C.; Schoneker, D. R. Food Chem. Toxicol. 2003, 41, 319326. (12) Jovanovic, A. V.; Flint, J. A.; Varshney, M.; Morey, T. E.; Dennis, D. M.; Duran, R. S. Biomacromolecules 2006, 7, 945-949.

Langmuir, Vol. 24, No. 7, 2008 3419

We conducted the thermogravimetric analysis for MSNPs and HSNPs. While MSNPs exhibited abrupt weight variation (∼20 wt %) near 260 °C due to decomposition of organic compounds such as dodecanoic acid and poly(vinyl alcohol), little weight variation was observed for HSNPs for temperatures over 300 °C (Figure 7). This confirms that HSNPs-OH contains little organic components after calcination at 300 °C. To determine the iron (Fe) content in HSNPs-OH after acid etching, we conducted the elemental analysis for MSNPs and HSNPs using energydispersive X-ray spectrometers (Figure 8). In Figure 8a and b, Fe contents were detected for MKs (96.92 ( 1.13%) and MSNPs (38.70 ( 0.77%). After hydrochloric acid treatments, however, Fe components were nearly completely eliminated (Figure 8c) to give the remnant Fe content of 2.86 ( 0.46% (Figure 8d). To characterize the cavity volume and size of pores in the shell, N2 adsorption/desorption measurements for HSNPs were performed using Brunauer-Emmett-Teller (BET). In Figure 9, HSNPs-PEG adsorbed nitrogen up to 198.7 cm3/g, much larger than MSNPs (92.7 cm3/g). However, the adsorbed volume of N2 for HSNPs-PEG was smaller than that of unmodified HSNPs because PEG molecules on the surface of HSNPs decrease the effective cavity size. The N2 adsorption/desorption study shows that HSNPs contain very small pores, which are not readily observed by TEM study. The average pore size of HSNPs-PEG was 1.64 nm, which is smaller than that of MSNPs (2.30 nm), because PEG molecules again reduce the effective radius of small pores. In summary, the PEG molecules reduce the size of the large cavity in HSNPs as well as the sizes of small pores on the silica wall, which endow the HSNPs the required passages, maybe in the form of connected small pores, to reach the large inner cavity. To use HSNPs as drug delivery vehicles, DOX was loaded into HSNPs-OH (DOX-HSNPs-OH, Figure 2a). The drug loading content and entrapment efficiency of DOX-HSNPs-OH were 23.5 and 47.7 wt %, respectively. The drug loading content was higher than previous works due to the increased nanoparticle cavity.6,13 Figure 10 shows the DOX release profile of the three HSNPs. HSNPs-OH released most of the drug within 1 day (filled circle; b), which was faster than those of the conventional polymeric drug carriers, resulting in limited ability to control the sustained drug release.7 To decrease the release rate, the HSNPs-OH surface was modified with an amine group; the hydroxyl group (-OH) of HSNPs-OH was exchanged with the amine group (-NH2) using APTMS (Figure 2b). After surface modification, the size was 85.9 ( 7.1 nm and the surface charge was -4.2 ( 2.2 mV due to the amine groups on the surface of HSNPs-NH2. The amine-modified HSNPs-OH (DOXHSNPs-NH2) showed a drug loading content and an entrapment efficiency of 21.6 and 42.3 wt %, respectively. The DOX release from DOX-HSNPs-NH2 was slightly slower than DOXHSNPs-OH (Figure 10, filled triangle; 2) due to ionic interactions of the DOX carboxyl group with the amine group of HSNPNH2.6 The PEGylated HSNPs (HSNPs-PEG, Figure 2c) was prepared by conjugating the diCOOH-PEG to HSNPs-NH2. The amide bond (OdC-N-H) was formed after the reaction, as confirmed in the FT-IR spectrum (Figure 6a). The size of the PEGylated HSNPs was 91.3 ( 8.1 nm, and the surface charge was -1.3 ( 3.2 mV due to the attached neutral PEG molecules. The drug loading content and entrapment efficiency of DOX-HSNPs-PEG were 19.5 and 38.7 wt %, respectively. The decrease in drug loading content and entrapment efficiency was due to the decrease (13) Zhao, W.; Gu, J.; Zhang, L.; Chen, H.; Shi, J. J. Am. Chem. Soc. 2005, 127, 8916-8917.

3420 Langmuir, Vol. 24, No. 7, 2008

Yang et al.

Figure 4. (a) Size distributions and zeta potential of (a) MAs, MSNP, and HSNPs and (b) photographs of (i) MSNPs and (ii) HSNPs in an aqueous phase.

Figure 5. X-ray diffraction patterns of (a) MSNPs and (b) HSNPs.

Figure 8. Energy-dispersive X-ray spectrometers for (a) MKs, (b) MSNPs, and (c) HSNPs. (d) Composition ratio of Si and Fe for MKs, MSNPs, and HSNPs.

Figure 6. FT-IR spectra of (a) HSNPs, (b) MSNPs, and (c) MAs.

Figure 9. Nitrogen adsorption-desorption isotherms for MSNPs, HSNPs, and HSNPs-PEG.

Figure 7. HSNPs.

Thermogravity analysis for (a) MSNPs and (b)

in surface area from the presence of PEG molecules. Notably, the release rate of absorbed DOX from the DOX-HSNPs-PEG was significantly lowered due to the reduced pore size by the PEG molecules on the small pore walls (Figure 10, filled square; 9). In addition, burst release from HSNPs-PEG at an early period

was insignificant. Specifically, on the first day, the release rate of MSNPs (2.2956 Ln%/day) was almost 10 times larger than that of HSNPs-PEG (0.2270 Ln%/day), with accompanying burst release in the early state (Table 1). Furthermore, HSNPs-PEG exhibited near zero-order kinetics, and the release rate was 0.0636 Ln%/day, whichwas smaller than HSNPs (0.1040 Ln%/day) and HSNPs-NH2 (0.1051 Ln%/day). Thus, HSNPs-PEG demonstrated a remarkable sustained release profile by surface modification process. Therefore, retardation of drug release from HSNPsPEG was successfully accomplished by PEG attachment to the surface of HSNPs.

Silica Nanocontainers as Drug DeliVery Vehicles

Langmuir, Vol. 24, No. 7, 2008 3421

Figure 10. (a) DOX release profiles of HSNPs (b), HSNPs-NH2 (2), and HSNPs-PEG (9). (b) Semilogarithmic plot of DOX release from HSNPs (b), HSNPs-NH2 (2), and HSNPs-PEG (9) as time went by. Table 1. Kinetics of Drug Release from HSNPs, HSNPs-NH2, and HSNPs-PEG rate ka (Ln%/day) period (days) HSNPs HSNPs-NH2 HSNPs-PEG

R2

0-1 2.2956 0.7108 0.2270

rate ka (Ln%/day)

Conclusion R2

2-15 0.9781 0.9734 0.9401

0.1040 0.1051 0.0636

Figure 11. Fluorescence microscopic image of (a) HSNPs-PEG and photograph of (b) precipitated HSNPs-PEG by centrifugation and (c) HSNPs-PEG dispersion in an aqueous solution.

0.8607 0.9958 0.9765

a The drug release data was analyzed using the equation14 X ) X (1 t inf - e-kt), where Xt and Xinf denote the absolute cumulative amounts of drug released at time t and infinite time, t is the release time, and k is a rate constant.

Furthermore, due to the fluorescence characteristics of DOX, the DOX-loaded HSNPs-PEG could be observed using a fluorescence microscope (Figure 11a).14,15 The DOX-loaded HSNPs-PEG were well dispersed in the aqueous phase (Figure 11c), although the DOX-loaded HSNPs-PEG could be precipitated in the microtube by centrifugation (Figure 11b). (14) Hitoshi, S. Drug deliVery; Hanlimwon Publishing Co.: Seoul, 1998. (15) Yoo, H. S.; Park, T. G. J. Controlled Release 2004, 96, 273-283.

Novel hollow silica nanoparticles (HSNPs) for drug delivery vehicles were synthesized using silica-coated magnetic kernels, which are composed of a number of Fe3O4 nanocrystals, as templates. The core cavity was obtained by removal of Fe3O4 phase with hydrochloric acid and subsequent calcination at a high temperature. HSNPs decorated by PEG on the surface exhibited superb drug release behavior due to the pore narrowing effect of PEG molecules, which leads to the potential utility of HSNPs-PEG as drug delivery vehicles and nanoreactors. Acknowledgment. This work was supported by Korea Science and Engineering Foundation (KOSEF) (R15-2004-02400000-0, M10755020001-07N5502-00110 and R01-2006-00010023-0) and a grant of the National R&D Program for Cancer Control, Korea Ministry of Health & Welfare (0620190-1). LA701688T