Paclitaxel-Loaded Poly(γ-glutamic acid)-poly(lactide) - ACS Publications

Feb 18, 2006 - development of chemotherapeutic agents for the treatment of ... groups and pharmaceutical companies have endeavored to .... phosphortun...
1 downloads 0 Views 685KB Size
Download PDF
Bioconjugate Chem. 2006, 17, 291−299

291

Paclitaxel-Loaded Poly(γ-glutamic acid)-poly(lactide) Nanoparticles as a Targeted Drug Delivery System against Cultured HepG2 Cells Hsiang-Fa Liang,† Sung-Ching Chen,† Mei-Chin Chen,† Po-Wei Lee,† Chiung-Tong Chen,‡ and Hsing-Wen Sung*,† Department of Chemical Engineering, National Tsing Hua University, Hsinchu, Taiwan, R.O.C., and Division of Biotechnology and Pharmaceutical Research, National Health Research Institutes, Taipei, Taiwan, R.O.C.. Received July 19, 2005; Revised Manuscript Received December 20, 2005

The study was to develop paclitaxel-loaded formulations using a novel type of self-assembled nanoparticles that was composed of block copolymers synthesized from poly(γ-glutamic acid) and poly(lactide) via a simple coupling reaction. The nanoparticles (the NPs) were prepared with various feed weight ratios of paclitaxel to block copolymer (the P/BC ratio). The morphology of all prepared nanoparticles was spherical and the surfaces were smooth. Increasing the P/BC ratio significantly increased the drug loading content of the prepared nanoparticles, but remarkably reduced the drug loading efficiency. The release rate of paclitaxel from the NPs decreased significantly as the P/BC ratio increased. For the potential of targeting liver cancer cells, galactosamine was further conjugated on the prepared nanoparticles (the Gal-NPs) as a targeting moiety. It was found that the activity in inhibiting the growth of HepG2 cells (a liver cancer cell line) by the Gal-NPs was comparable to that of a clinically available paclitaxel formulation, while the NPs displayed a significantly less activity. This may be attributed to the fact that the Gal-NPs had a specific interaction with HepG2 cells via ligand-receptor recognition. Cells treated with distinct paclitaxel formulations resulted in arrest in the G2/M phase. The arrest of cells in the G2/M phase was highly suggestive of interference by paclitaxel with spindle formation and was consistent with the morphological findings presented herein. In conclusion, the active targeting nature of the Gal-NPs prepared in the study may be used as a potential drug delivery system for the targeted delivery to liver cancers.

INTRODUCTION The toxicity of anticancer drugs to normal tissues limits the development of chemotherapeutic agents for the treatment of cancers. Various drug carriers have been investigated to reduce the toxicity and increase the therapeutic efficacy of anticancer drugs (1-4). The pharmaceutical applications of self-assembled nanoparticles have recently attracted considerable attention (5). The self-assembled nanoparticles, composed of amphiphilic block copolymers, have a hydrophobic inner core and a hydrophilic outer shell. In our previous study, poly(γ-glutamic acid) (γ-PGA) and poly(lactide) (PLA) were used to synthesize amphiphilic block copolymers via a simple coupling reaction to prepare a novel type of self-assembled nanoparticles (6). γ-PGA, a natural compound, is produced as capsular substance or as slime by members of the genus Bacillus (7). It is uniquely composed of naturally occurring L-glutamic acids linked by amide bonds. It is known that γ-PGA is water-soluble, biodegradable, nontoxic (7), and relatively nonimmunogenic (8). In our previous stability study, no aggregation or precipitation of the nanoparticles was observed during storage for up to 1 month, since it was prevented by the electrostatic repulsion between the negatively charged nanoparticles (6). The aim of this study was to develop paclitaxel-loaded formulations using the aforementioned nanoparticles that were composed of amphiphilic γ-PGA-PLA block copolymers. Galactosamine was further conjugated on the prepared nanoparticles as a targeting moiety to develop the potential of targeting liver cancer cells. The unique structure of paclitaxel enables it

readily to enter mammalian cells and preferentially bind to tubulin in polymerized microtubules (9). This stabilizes microtubules and greatly interferes with microtubular reorganization necessary, among other processed, for spindle formation and cell division (10). Therefore, the exposure of susceptible cells to paclitaxel has been shown to arrest the G2/M phase initially and finally cause cell death through apoptosis (11). However, the hydrophobicity of paclitaxel raises great difficulties in the preparation of formulations. Many research groups and pharmaceutical companies have endeavored to develop better and improved paclitaxel formulations that are suitable for different applications (12). Even though paclitaxel is mainly prescribed to treat breast and ovarian cancers, it is known that various cancer cells including hepatoma cells can be killed effectively by this drug (13, 14). In this study, the effects of the feeding ratio of drug to copolymer on the distribution of particle sizes, the zeta potential, the drug loading content, and the drug loading efficiency of the prepared nanoparticles and their in vitro release profiles were evaluated. The morphology of the prepared nanoparticles was examined by the transmission electron microscopy (TEM) and the atomic force microscopy (AFM). Additionally, the cytotoxicity of the prepared nanoparticles with or without galactosamine conjugated on HepG2 cells (a liver cancer cell line) was compared to that of a clinically available paclitaxel formulation in vitro. The effect of the prepared nanoparticles on the cellular microtubules and the restriction of HepG2 cells in specific cell cycle stages was investigated using a confocal laser scanning microscope (CLSM) and a flow cytometer.

EXPERIMENTAL SECTION * To whom correspondence should be addressed. Tel:886-3-5742504, Fax:886-3-572-6832, E-mail: [email protected]. † National Tsing Hua University. ‡ National Health Research Institutes.

Materials. Paclitaxel powder (purity > 99%) and clinical commercial paclitaxel [Phyxol, contained 6 mg paclitaxel, 527 mg Cremaphor EL and 47.7% (v/v) alcohol per milliliter] were

10.1021/bc0502107 CCC: $33.50 © 2006 American Chemical Society Published on Web 02/18/2006

292 Bioconjugate Chem., Vol. 17, No. 2, 2006

Liang et al.

Figure 1. Schematic illustrations of synthesis of γ-PGA-PLA block copolymers and formation of self-assembled nanoparticles with conjugated galactosamine.

obtained from Sinphar Pharmaceutical Co., Ltd. (Taipei, Taiwan). PLA [poly(L-lactide), Mn: 10 kDa, with a polydispersity of 1.1 determined by the GPC analysis] was kindly supplied by the Biomedical Engineering Center, Industrial Technology Research Institute (Hsinchu, Taiwan). Dimethyl sulfoxide (DMSO < 0.01% water), N,N′-carbonyldiimidazole (CDI, 98%), and dichloromethane were acquired from Fluka (Buchs, Switzerland). L-Glutamic acid (purity > 99%), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide (EDC), N-hydroxysuccinimide (NHS), galactosamine, and sodium cholate were purchased from Sigma (St. Louis, MO). 4-(Dimethylamino)pyridine (DMAP) and 1,4-dioxane were obtained from ACROS (Janssen Pharmaceuticalaan, Belgium). All other used chemicals were reagent grade. Production of γ-PGA. γ-PGA (Figure 1) was produced by Bacillus licheniformis (ATCC 9945, Bioresources Collection and Research Center, Hsinchu, Taiwan) and purified as per a method reported by Yoon et al. with slight modifications (15). Details of the methodology were reported in our previous publication (6). Hydrolysis of γ-PGA. The average molecular weight (Mn) of the purified γ-PGA obtained via the previous fermentation procedure was about 320 kDa. The purified γ-PGA was then hydrolyzed in a tightly sealed steel container at 150 °C for specified durations (16). The average molecular weight along with the polydispersity of the hydrolyzed γ-PGA were determined using a gel permeation chromatography (GPC) system equipped with a series of PL aquagel-OH columns (one Guard 8 µm, 50 × 7.5 mm and two MIXED 8 µm, 300 × 7.5 mm, PL Laboratories, UK) and a refractive index (RI) detector (RI2000F, SFD, Torrance, CA). Poly(ethylene glycol) (molecular weights of 106-22000 g/mol) and poly(ethylene oxide) (with molecular weights of 20000-1000000 g/mol) standards of narrow polydispersity (PL Laboratories, UK) were used to construct a calibration curve. The mobile phase contained 0.01 M NaH2PO4 and 0.2 M NaNO3 and was brought to a pH of 7.0. The flow rate of the mobile phase was 1.0 mL/min, and the columns and the RI detector cell were maintained at 30 °C. Synthesis of γ-PGA-PLA Block Copolymers. Block copolymers composed of γ-PGA and PLA were synthesized using CDI to activate the terminal hydroxyl group of PLA (17). CDI (82 mg) was dissolved in 1,4-dioxane (20 mL) in a nitrogen

atmosphere and PLA (0.1 g) was subsequently added to the solution. The clear solution was stirred at 37 °C for 2 h. Afterward, the solution was dialyzed extensively against deionized water at 4 °C. Finally, the activated PLA was obtained via centrifugation. The acidified form of the hydrolyzed γ-PGA (10 mg, Mn ∼4 kDa, PDI ) 1.3) was dissolved in DMSO (5 mL) in a dry, stoppered 20 mL round-bottomed flask in a nitrogen atmosphere. After DMAP (3 mg) was dissolved, a calculated amount of activated PLA (25 mg) was added. The solution was stirred at room temperature for 3 days, after which the reaction was stopped by adding 0.1 mL of concentrated HCl to neutralize DMAP and imidazole. The reaction mixture was transferred to a dialysis tube (MWCO: 6000-8000), followed by dialysis for 2 days against deionized water for several times at 4°C to remove the un-conjugated γ-PGA. Additionally, the unconjugated PLA was separated due to its self-precipitation. Finally, the product was lyophilized and stored at -20 °C until used. The molecular weight distribution of the synthesized block copolymers was determined using a GPC system equipped with a Jordi Gel DVB Mixed Bed column (250 × 10 mm, Jordi Associates, Inc., MA) and an RI detector. Tetrahydrofuran (THF) was used as an elution solvent (1 mL/min) and polystyrene standards for column calibration. Preparation of the Paclitaxel-loaded Nanoparticles. The paclitaxel-loaded nanoparticles were produced using an emulsion/solvent evaporation technique (18). Briefly, 10 mg of block copolymers were dissolved in 1 mL of methylene chloride, and paclitaxel was subsequently added with various feed weight ratios to block copolymer [paclitaxel/copolymer (P/BC) ) 0.5/ 10, 1/10, 2/10, and 3/10]. The solution was then stirred for 2 h at room temperature and emulsified in 50 mL of a 0.1 wt % sodium cholate solution using a sonicator (VCX-750, Sonics & Materials Inc., Newtown, CT; cycles were of 1 s sonication followed by a 1 s pause, over a total period of 20 min). Afterward, the solvent was evaporated in a vacuum oven at 37 °C for 1 h. The resulting suspension was filtered through a 0.8 µm membrane filter (Whatman) and then centrifuged for 60 min at 18000 rpm at 4 °C. The supernatant was subsequently discarded and the pellet was resuspended in 10 mL of phosphatebuffered saline (PBS, pH 7.4, Sigma). The size distribution and

Bioconjugate Chem., Vol. 17, No. 2, 2006 293

Drug Delivery System against Cultured HepG2 Cells

Table 1. Particle Size, Zeta Potential, and Drug Loading Content (LC) and Loading Efficiency (LE) of Nanoparticles Prepared with Various Feed Weight Ratios of Paclitaxel to Block Copolymer (the P/BC ratio)

Figure 2. Chromatograms of purified γ-PGA obtained by fermentation (γ-PGA before hydrolysis), obtained γ-PGA after 5 h of hydrolysis at 150 °C (γ-PGA hydrolysis for 5h), and hydrolyzed γ-PGA after dialysis twice against deionized water (γ-PGA hydrolysis for 5 h then dialysis).

P/BC ratio (n ) 4)

particle size (nm)

zeta potential (mV)

LC (%)

LE (%)

0/10 0.5/10 1/10 2/10 3/10

115.4 ( 4.2 125.9 ( 5.5 128.8 ( 3.4 144.4 ( 2.6 263.2 ( 6.8

-21.4 ( 2.3 -22.5 ( 3.2 -19.6 ( 1.8 -20.3 ( 2.7 -19.2 ( 2.2

3.7 ( 0.1 5.1 ( 0.2 5.8 ( 0.2 6.1 ( 0.2

76.5 ( 2.4 53.7 ( 1.7 30.8 ( 2.3 21.7 ( 4.2

Loading Content and Loading Efficiency of the PaclitaxelLoaded Nanoparticles. The drug loading content and the loading efficiency of the nanoparticles were determined using a high-performance liquid chromatography (HPLC) system equipped with a C18 analytic column (4.6 × 250 mm, particle size 5 µm, ThermoQuest, BDS, Runcorn, UK) (19). Two milligrams of the freeze-dried paclitaxel-loaded nanoparticles were dissolved in 1 mL dichloromethane with vigorous vortexing. This solution was dried by evaporating dichloromethane in a vacuum. It then was dissolved in a mixture of 50/50 (v/v) ethanol and deionized water to perform the HPLC analysis. The flow rate of the mobile phase (60% acetonitrile and 40% deionized water by v/v), delivered by an HPLC pump (TCP, P-100, Riviera Beach, FL), was 1 mL/min at 30 °C. The injection volume was 40 µL, and paclitaxel eluted from the column was monitored with an UV detector (Jasco 875-UV, Tokyo, Japan) at 227 nm. The drug loading content and the loading efficiency of the nanoparticles were calculated using the following equations (20).

loading content (%) ) weight of paclitaxel in the nanoparticles × 100% weight of the nanoparticles loading efficiency (%) ) weight of paclitaxel in the nanoparticles × 100% weight of the feeding paclitaxel

Figure 3. (a) 1H NMR spectrum of synthesized γ-PGA-PLA block copolymer. (b) GPC chromatograms of PLA (solid curve) and the synthesized γ-PGA-PLA block copolymer (dotted curve).

zeta potential of the prepared nanoparticles were measured using a Zetasizer (3000HS, Malvern Instruments Ltd., Worcestershire, UK). TEM and AFM were used to observe the morphology of the paclitaxel-loaded nanoparticles. The TEM sample was prepared by placing a drop of the paclitaxel-loaded nanoparticle solution onto a 400 mesh copper grid coated with carbon. About 2 min after deposition, a filter paper was tapped to the grid to remove surface water and negatively stained using a 2% (by w/v) phosphortungsten acid (PTA) solution. The AFM sample was prepared by placing a drop of the paclitaxel-loaded nanoparticle solution on a glass slide and then drying it in a vacuum.

Release of Paclitaxel from the Loaded Nanoparticles. The release profiles of paclitaxel from the prepared nanoparticles in PBS at 37 °C were investigated (19). The freeze-dried paclitaxel-loaded nanoparticles were weighed and resuspended in a centrifuge tube containing 20 mL of PBS. The tube was placed in a shaker water bath at 37 °C. At particular time intervals, the tube was removed and centrifuged. The supernatant was poured out, freeze-dried, and then dissolved in a mixture of 50/50 (v/v) ethanol and deionized water to conduct HPLC analysis. The pellet was resuspended in 20 mL of fresh PBS to make continuous release measurements. The paclitaxel released at each time point was calculated using a calibration curve (19). Conjugation of Galactosamine to the Paclitaxel-Loaded Nanoparticles. Galactosamine was conjugated to the paclitaxelloaded nanoparticles via an amide link by EDC in the presence of NHS (21, 22). The conditions in our previous study, under which galactosamine was conjugated on nanoparticles that had the most nanoparticles internalized in HepG2 cells, were used herein (6). The obtained galactosylated nanoparticles were separated from unreacted molecules via ultrafiltration and then lyophilized. The content of galactosamine conjugated on the nanoparticles was determined by the Morgan Elson assay (23). Viability of HepG2 Cells Treated with Distinct Paclitaxel Formulations. The cytotoxicity of the paclitaxel-loaded nanoparticles with or without galactosamine conjugated was evaluated in vitro by the MTT assay, using a clinically available paclitaxel formulation (Phyxol, Sinphar Pharmaceutical) as a control (24). The assay is based on mitochondrial dehydrogenase

294 Bioconjugate Chem., Vol. 17, No. 2, 2006

Liang et al.

Figure 4. Size distributions of nanoparticles prepared with various feed weight ratios of paclitaxel to block copolymer (P/BC ratios). PI: polydispersity index of size distribution of prepared nanoparticles.

Figure 5. Morphology of nanoparticles prepared with various feed weight ratios of paclitaxel to block copolymer (P/BC ratios) obtained by AFM and TEM.

Drug Delivery System against Cultured HepG2 Cells

Figure 6. Release profiles of paclitaxel from nanoparticles prepared with various feed weight ratios of paclitaxel to block copolymer (P/ BC ratios).

Figure 7. Viability of HepG2 cells treated with distinct paclitaxel formulations at various paclitaxel concentrations for 3 days. Phyxol: cells treated with a clinically available paclitaxel formulation (Sinphar Pharmaceutical); NPs: cells treated with paclitaxel-loaded nanoparticles without conjugated galactosamine; and Gal-NPs: cells treated with paclitaxel-loaded nanoparticles with conjugated galactosamine.

cell activity as an indicator of cell viability. Briefly, MTT [3-(4,5-dimethyl-thiazol-yl)-2,5-diphenyltetrazolium bromide, Sigma] was dissolved in PBS to a concentration of 5 mg/mL as a stock MTT solution and filtered for sterilization. An amount of 5 × 104 HepG2 cells/well was seeded in 24-well plates and allowed to adhere overnight. The growth medium was replaced with a fresh medium that contained various concentrations (0.25-8 µg/mL) of the following paclitaxel formulations: Phyxol, nanoparticles without conjugated galactosamine (NPs), and nanoparticles with conjugated galactosamine (Gal-NPs). The cells were then incubated for 3 days and washed twice in 1 mL of PBS. Subsequently, the cells were incubated in a growth medium that contained 1 mg/mL MTT agent for an additional 4 h at 37 °C; 500 µL of DMSO was added to each well to ensure that the formazan crystals dissolved. Finally, the optical density readings were taken using a multiwell scanning spectrophotometer (MRX Microplate Reader, Dynatech Laboratories Inc., Chantilly, VA) at a wavelength of 570 nm. Immunofluorescence Analysis of HepG2 Cells Treated with Distinct Paclitaxel Formulations. HepG2 cells were grown on glass coverslips and then treated with distinct paclitaxel formulations with a final paclitaxel concentration of 8 µg/mL. After incubation for 3 days, the cells were fixed with

Bioconjugate Chem., Vol. 17, No. 2, 2006 295

3.7% formaldehyde in PBS for 10 min at room temperature and then permeabilized in 0.1% Triton X-100 in PBS that contained 1% bovine serum albumin (PBS-BSA) and RNase 100 µg/mL. After they had been washed three times with PBSBSA, the cells were treated with Oregon Green 514 palloidin (1:100 v/v, Molecular Probes) in PBS-BSA for 20 min. The cells were then incubated for 60 min with anti-bovine R-tubulin mouse mAb (1 µg/mL, Molecular Probes) in PBS-BSA. The Alexa Fluor 633-conjugated goat anti-mouse IgG antibody (2 µg/mL, Molecular Probes) was added, and the cells were incubated for another 60 min. They were then rinsed three times with PBS-BSA and treated with 100 nM propidium iodide (PI, Sigma) for 5 min. Before the samples were mounted for CLSM examinations, they were again washed with PBS and deionized water. Oregon Green 514 palloidin, PI, or Alexa Fluor 633 stainings were visualized with excitations at 488, 543, and 633 nm, respectively, using an inverted CLSM (TCS SL, Leica, Germany). Images were imposed using LCS Lite software (version 2.0). Altered Cycling States of HepG2 Cells Treated with Distinct Paclitaxel Formulations. To demonstrate whether paclitaxel released from the prepared NPs or the Gal-NPs could restrict HepG2 cells in specific cell cycle stages, flow cytrometric studies were performed. HepG2 cells treated with distinct paclitaxel formulations with a final paclitaxel concentration of 1 µg/mL for 3 days were pelleted at 1500 rpm for 5 min and then resuspended in PBS (25). Methanol (100%) precooled to -20 °C for 15 min was then added to the cell suspension, which was then centrifuged at 1500 rpm for 5 min. The supernatant was discarded, and the cell pellet was rehydrated with PBS. The pellet was stained with DNA staining solution (10 µg/mL PI and 1 mg/mL RNase A) for 45 min. The DNA content of each cell was measured using a Becton Dickinson FACSCalibur flow cytometer (San Jose, CA). Statistical Analysis. Statistical analysis was performed to determine differences between the measured properties of groups. One-way analysis of variance was performed, and confidence intervals were determined using a statistical program (Statistical Analysis System, Version 6.08, SAS Institute Inc., Cary, NC). All data are presented as a mean value with its standard deviation indicated (mean ( SD).

RESULTS AND DISCUSSION Production of γ-PGA. In the 1H NMR spectrum of the purified γ-PGA produced by fermentation, the five main signals at 1.73, 1.94, 2.19, 4.14, and 8.15 ppm represented the protons of β-CH2, γ-CH2, R-CH, and amide, respectively. Additionally, the fermented product after purification showed no macromolecular impurity, according to the 1H NMR analysis, suggesting that the obtained white γ-PGA powder was highly pure. Hydrolysis of γ-PGA. Low-molecular-weight γ-PGA was produced by hydrolyzing purified γ-PGA obtained from fermentation at 150 °C for distinct durations. Solutions of purified γ-PGA obtained by fermentation and hydrolyzed γ-PGA were analyzed by a GPC system. Figure 2 shows that the purified γ-PGA obtained by fermentation had a high average molecular weight (Mn ∼320 kDa) with a polydispersity of about 1.8. When γ-PGA was hydrolyzed at 150 °C for 5 h, its average molecular weight of γ-PGA was reduced to about 5 kDa. To reduce the polydispersity of the hydrolyzed γ-PGA, the hydrolyzed γ-PGA (∼5 kDa) was further dialyzed twice (using a membrane with MWCO: 3500 and a membrane with MWCO: 6000-8000) against deionized water. Thus, the obtained γ-PGA had an average molecular weight of about 4 kDa with a polydispersity of 1.3 (Figure 2). This specific γ-PGA was used subsequently, together with PLA, to synthesize block copolymers to prepare the nanoparticles.

296 Bioconjugate Chem., Vol. 17, No. 2, 2006

Liang et al.

Figure 8. CLSM images of HepG2 cells treated with distinct paclitaxel formulations for 3 days. Control: untreated cells; Phyxol: cells treated with a clinically available paclitaxel formulation (Sinphar Pharmaceutical); NPs: cells treated with paclitaxel-loaded nanoparticles without conjugated galactosamine; and Gal-NPs: cells treated with paclitaxel-loaded nanoparticles with conjugated galactosamine. N: normal nuclei; n: micronuclear formation; bold arrow right: centrosomes; and f: condensation of cytoplasmic microtubules.

Synthesis of γ-PGA-PLA Block Copolymers. Figure 3a and 3b show the 1H NMR spectrum and the GPC chromatogram of the synthesized γ-PGA-PLA block copolymers, respectively. In the 1H NMR spectrum (Figure 3a), the chemical shifts at 1.42 and 5.20 ppm corresponded to the protons of the methyl group (CH3) and methine (CH) on PLA, respectively, while the main chemical shifts exhibited by γ-PGA, described above, were also detected. The GPC chromatograms (Figure 3b) indicated that the molecular weight of γ-PGA-PLA slightly exceeded that of PLA. The aforementioned results indicated that γ-PGA was successfully conjugated to PLA by CDI.

Particle Size and Morphology of Paclitaxel-Loaded Nanoparticles. The distribution of sizes and the zeta potential of the prepared nanoparticles may importantly determine their fates after administration (19). As shown in Table 1, the particle sizes of the prepared nanoparticles increased significantly with the P/BC ratio. Dynamic light scattering measurements further demonstrated that all of the prepared nanoparticles had a narrow size distribution, with the exception of those prepared with a P/BC ratio of 3/10 (Figure 4). AFM and TEM showed that the morphologies of all of the prepared nanoparticles were spherical, and their surfaces were smooth (Figure 5).

Drug Delivery System against Cultured HepG2 Cells

Bioconjugate Chem., Vol. 17, No. 2, 2006 297

Figure 9. Analysis, using a flow cytometer, of DNA content in cell cycle of HepG2 cells treated with distinct paclitaxel formulations for 3 days. Control: untreated cells; Phyxol: cells treated with a clinically available paclitaxel formulation (Sinphar Pharmaceutical); NPs: cells treated with paclitaxel-loaded nanoparticles without conjugated galactosamine; and Gal-NPs: cells treated with paclitaxel-loaded nanoparticles with conjugated galactosamine.

Hashida et al. reported that the majority of the diameter of the fenestrate of the liver sinusoid is usually smaller than 200 nm (26). Thus, large particles hardly reach the parenchymal cells of the liver. Additionally, drug carriers with diameters of larger than 200 nm are readily scavenged nonspecifically by monocytes and the reticuloendothelial system (27). It was reported that smaller particles tended to accumulate at the tumor sites because of the EPR (enhanced permeability and retention) effect (28); greater internalization was also observed (1). It was found that the prepared paclitaxel-loaded nanoparticles had a negative surface charge with a zeta potential of about -20 mV (Table 1), because of the carboxyl (COO) groups on the hydrophilic γ-PGA shell. This may affect the cellular uptake of the prepared nanoparticles due to electrostatic repulsion forces between the nanoparticles and the rather negatively charged surface of cells (1). However, Wakebayashi et al. suggested that the introduction of a specific ligand onto the nanoparticles may enhance their cellular uptake by receptor-mediated endocytosis (29). Additionally, positively charged carriers are reported to be able to induce a nonspecific interaction with unintended target tissues, particularly under in vivo conditions after administration (29, 30). Loading Content and Loading Efficiency of PaclitaxelLoaded Nanoparticles. Paclitaxel is highly hydrophobic and has a solubility of approximately 1 µg/mL in aqueous solution at pH 7.4 (31). Thus, in the drug loading process, the incorporation of paclitaxel in the nanoparticles competed with the precipitation of paclitaxel in aqueous solution (31). As the P/BC ratio increased, the incorporation of paclitaxel in the nanoparticles (the drug loading content) appeared to be increased, while the precipitation of paclitaxel in aqueous solution was more pronounced and thus significantly reduced the drug loading efficiency (Table 1, p < 0.05). Release of Paclitaxel from Loaded Nanoparticles. As shown in Figure 6, paclitaxel was continuously released from

nanoparticles prepared with distinct P/BC ratios. All samples released a burst of paclitaxel in the initial stage. An amount of 10-25% of the encapsulated drug was released in the first hour. This may be due to some of the drugs were deposited near the γ-PGA shell of the prepared nanoparticles. As the P/BC ratio increased, the rate of release of paclitaxel from the prepared nanoparticles decreased significantly. It was reported that a hydrophobic drug encapsulated within nanoparticles partially crystallized at a higher drug loading content, while it formed a molecular dispersion at a lower drug loading content (32). The crystallized drug in the hydrophobic core of the nanoparticles is expected to dissolve more gradually and to diffuse to their outer aqueous phase more slowly than the drug in the form of a molecular dispersion (2). Additionally, the encapsulated drug took longer to diffuse through the polymer matrix to the aqueous medium when the nanoparticles were larger (i.e., as the P/BC ratio increased, as in Table 1) (32). Conjugation of Galactosamine to Paclitaxel-Loaded Nanoparticles. As discussed above, the drug loading content of the prepared nanoparticles increased significantly with the P/BC ratio, while their drug loading efficiency decreased remarkably (Table 1). To obtain a comparatively high drug loading content with a high loading efficiency (Table 1), nanoparticles prepared with a P/BC ratio of 1/10 (the NPs) were used in the rest of this study. For the potential of targeting liver cancer cells, galactosamine was conjugated to the paclitaxel-loaded nanoparticles (the Gal-NPs). As determined by the Morgan Elson assay, the amount of galactosamine conjugated on the Gal-NPs was 66.2 ( 2.4 nmol/mg nanoparticles (or 4.0 ( 0.1 × 1016 galactosamine moieties per mg of the targeted nanoparticles, n ) 4). The particle size of the Gal-NPs (127.5 ( 2.5 nm) was comparable to that of the NPs (128.8 ( 3.4 nm, p > 0.05). However, the zeta potential of the former (-10.6 ( 2.0 mV) was significantly lower than that of the latter (-19.6 ( 1.8 mV, p < 0.05), because galactosamine was conjugated to the

298 Bioconjugate Chem., Vol. 17, No. 2, 2006

carboxyl (COO) groups on γ-PGA, reducing the negative surface charge of the Gal-NPs. Viability of HepG2 Cells Treated with Distinct Paclitaxel Formulations. Figure 7 shows the viability of HepG2 cells treated with the paclitaxel formulations investigated herein. The inhibition of the growth of cells by the Gal-NPs was comparable to that of a clinically available paclitaxel formulation (Phyxol, p > 0.05); the NPs exhibited significantly less inhibitory activity (p < 0.05). Hepatoma cells are known to recognize galactose- and N-acetylgalactosamine-terminated glycoproteins via the asialoglycoprotein (ASGP) receptors located on their surfaces (33). It was found in our previous study that during incubation with rhodamine-123-containing nanoparticles without conjugated galactosamine, little fluorescence was observed on the images of HepG2 cells taken by the CLSM (6). This indicated that without galactosamine only a small proportion of the nanoparticles could be internalized in the cells, because of the electrostatic repulsion forces between the nanoparticles and the cells as mentioned earlier. Hence, the NPs prepared herein released paclitaxel mainly outside the cells (in the culture medium). The released paclitaxel then diffused into the HepG2 cells, inhibiting the growth of the cells. Accordingly, under in vivo conditions after administration, the normal tissues may be nonselectively exposed to paclitaxel released from the NPs in the blood stream, potentially causing unwanted toxic side effects. In contrast, with increasing the galactosamine content conjugated on the rhodamine-123-containing nanoparticles, the intensity of fluorescence observed in HepG2 cells increased significantly at 30 min after incubation (6). This indicated that the galactosylated nanoparticles interacted in a specified manner with HepG2 cells via ligand-receptor (ASGP) recognition. Therefore, the Gal-NPs prepared in this study were first internalized into HepG2 cells via the ASGP receptors, and then the encapsulated paclitaxel was released inside the cytoplasm, inhibiting the growth of the cells. Thus, the active targeting nature of the Gal-NPs may lead to high selectivity to the hepatic tumor and enhance their cellular uptake, with a consequent decrease in systemic toxicity. Figure 8 shows CLSM images of HepG2 cells treated with Phyxol, the NPs, or the Gal-NPs. The untreated HepG2 cells (control) demonstrated normal nuclei, centrosomes, and microtubule networks. In contrast, HepG2 cells that had been exposed to Phyxol, NPs, or Gal-NPs exhibited significantly disrupted polar spindles of cells and the characteristic condensation of cytoplasmic microtubules. Additionally, improper mitotic spindle assembly caused micronuclear formation. Figure 9 shows that HepG2 cells that had been treated with distinct paclitaxel formulations arrested the G2/M phase. The arrest of the growth of cells in the G2/M phase strongly suggested interference by paclitaxel with spindle formation and was consistent with the morphological findings presented in Figure 8. As compared to the untreated cells (control), the sharp peak observed in the G0/G1 phase was markedly attenuated and was instead mostly in the G2/M phase for the cells treated with Phyxol or the Gal-NPs. In contrast, for those cells treated with the NPs, this observation was less significant (p < 0.05). This finding again may be owing to that without galactosamine, only a small proportion of the nanoparticles (the NPs) could be internalized in the cells. Therefore, the NPs released paclitaxel mainly outside the HepG2 cells.

CONCLUSIONS In conclusion, the aforementioned results indicated that the paclitaxel-loaded nanoparticles with conjugated galactosamine prepared herein may be internalized into HepG2 cells via receptor-mediated endocytosis, inhibiting the growth of cells.

Liang et al.

Therefore, the prepared nanoparticles may be used as a potential targeted drug delivery system to fight liver cancers.

ACKNOWLEDGMENT The authors would like to thank the National Science Council of the Republic of China, Taiwan (Contract No. NSC 93-2213B-007-051), and the National Health Research Institutes of the Republic of China, Taiwan (Contract No. NHRI-EX93-9221EI), for financially supporting this research.

LITERATURE CITED (1) Li, C. (2002) Poly(L-glutamic acid)-anticancer drug conjugates. AdV. Drug DeliVery ReV. 54, 695-713. (2) Na, K., Park, K. H., Kim, S. W., and Bae, Y. H. (2000) Self-assembled hydrogel nanoparticles from curdlan derivatives: Characterization, anti-cancer drug release and interaction with a hepatoma cell line (HepG2). J. Controlled Release 69, 225-236. (3) Saul, J. M., Annapragada, A., Natarajan, J. V., and Bellamkonda, R. V. (2003) Controlled targeting of liposomal doxorubicin via the folate receptor in vitro. J. Controlled Release 92, 49-67. (4) Liang, H. C., Chang, W. H., Lin, K. J., and Sung, H. W. (2003) Genipin-crosslinked gelatin microspheres as a drug carrier for intramuscular administration: In vitro and in vivo studies. J. Biomed. Mater. Res. 65A, 271-282. (5) Na, K., Lee, T. B., Park, K. G., Shin, E. K., Lee, Y. B., and Choi, H. K. (2003) Self-assembled nanoparticles of hydrophobically modified polysaccharide bearing vitamin H as a targeted anti-cancer drug delivery system. Eur. J. Pharm. Sci. 18, 165-173. (6) Liang, H. F., Yang, T. F., Huang, C. T., Chen, M. C., and Sung, H. W. (2005) Preparation of nanoparticles composed of poly(γglutamic acid)-poly(lactide) block copolymers and evaluation of their uptake by HepG2 cells. J. Controlled Release 105, 213-225. (7) Richard, A., and Margaritis, A. (2001) Poly(glutamic acid) for biomedical applications. Crit. ReV. Biotechnol. 21, 219-232. (8) Prodhomme, E. J. F., Tutt, A. L., Glennie, M. J., and Bugg, T. D. H. (2003) Multivalent conjugates of poly-γ-D-glutamic acid from Bacillus Licheniformis with antibody F(ab′) and glycopeptide ligands. Bioconjugate Chem. 14, 1148-1155. (9) Rao, S., Orr, G. A., Chaudhary, A. G., Kingston, D. G., and Horwitz, S. B. (1995) Characterization of Taxol binding site on the microtubule, 2-(M-Azidobenzoyl) Taxol photolabels a peptide (amino acids 217-231) of beta-tubulin. J. Biol. Chem. 270, 20235-20238. (10) Gagandeep, S., Novikoff, P. M., Ott, M., and Gupta, S. (1999) Paclitaxel shows cytotoxic activity in human Hepatocellular carcinoma cell lines. Cancer Lett. 136, 109-118. (11) Jordan, M. A., Wendell, K., Gardiner, S., Derry, W. B., Copp, H., and Wilson, L. (1996) Mitotic block induced in HeLa cells by low concentrations of paclitaxel results in abnormal mitotic exit and apoptotic cell death. Cancer Res. 56, 816-825. (12) Lee, I. H., Park, Y. T., Roh, K., Chung, H., Kwon, I. C., and Jeong, S. Y. (2005) Stable paclitaxel formulations in oily contrast medium. J. Controlled Release 102, 415-425. (13) Lin, H. L., Liu, T. Y., Chau, G. Y., Lui, W. Y., and Chi, C. W. (2000) Comparison of 2-methoxyestradiol-induced, docetaxelinduced, and paclitaxel-induced apoptosis in hepatoma cells and its correlation with reactive oxygen species. Cancer 89, 983-994. (14) Kang, T. H., Pae, H. O., Yoo, J. C., Kim, N. Y., Kim, Y. C., Ko, G. I., and Chung, H. T. (2000) Antiproliferative effect of alkaloids from Sedum sarmentosum on murine and human hepatoma cell line. J. Ethnopharmacol. 70, 177-182. (15) Yoon, S. H., Do, J. H., Lee, S. Y., and Chang, H. N. (2000) Production of poly-γ-glutamic acid by fed-batch culture of Bacillus licheniformis. Biotechnol. Lett. 22, 585-588. (16) Troy, F. A. (1973) Chemistry and biosynthesis of the poly(γ-Dglutamyl) capsule in Bacillus licheniformis II. Characterization and structural properties of the enzymatically synthesized polymer. J. Biol. Chem. 248, 316-324. (17) Van Dijk-Wolthuis, W. N. E., Tsang, S. K. Y., Kettenes-van den Bosch, J. J., and Hennink, W. E. (1997) A new class of polymerizable dextrans with hydrolysable groups: Hydroxyethyl methacrylated

Bioconjugate Chem., Vol. 17, No. 2, 2006 299

Drug Delivery System against Cultured HepG2 Cells dextran with and without oligolactate spacer. Polymer 38, 62356242. (18) Gref, R., Luck, M., Quellec, P., Marchand, M., Dellacherie, E., Harnisch, S., Blunk, T., and Muller, R. H. (2000) “Stealth” coronacore nanoparticles surface modified by polyethylene glycol (PEG): Influences of the corona (PEG chain length and surface density) and of the core composition on phagocytic uptake and plasma protein adsorption. Colloid. Surface. B18, 301-313. (19) Dong, Y., and Feng, S. S. (2004) Methoxy poly(ethylene glycol)poly(lactide) (MPEG-PLA) nanoparticles for controlled delivery of anticancer drugs. Biomaterials 25, 2843-2849. (20) Zhang, L., Hu, Y., Jiang, X., Yang, C., Lu, W., and Yang, Y. H. (2004) Camptothecin derivative-loaded poly(caprolactone-co-lactide)-b-PEG-b- poly(caprolactone-co-lactide) nanoparticles and their biodistribution in mice. J. Controlled Release 96, 135-148. (21) Sung, H. W., Chang, W. H., Ma, C. Y., and Lee, M. H. (2003) Crosslinking of biological tissues using genipin and/or carbodiimide. J. Biomed. Mater. Res. 64A, 427-438. (22) Liang, H. C., Chang, W. H., Liang, H. F., Lee, M. H., and Sung, H. W. (2004) Crosslinking structures of gelatin hydrogels crosslinked with genipin or a water-soluble carbodiimide. J. Appl. Polym. Sci. 91, 4017-4026. (23) David, A., Kopeckova, P., Rubinstein, A., and Kopeccek, J. (2001) Enhanced biorecognition and internalization of HPMA copolymers containing multi- or multivalent carbohydrate side-chains by human hepatocarcinoma cells. Bioconjugate Chem. 12, 890-899. (24) Chen, C. N., Sung, H. W., Liang, H. F., and Chang, W. H. (2002) Feasibility study using a natural compound (reuterin) produced by Lactobacillus reuterin in sterilizing and crosslinking biological tissues. J. Biomed. Mater. Res. 61, 360-369. (25) Saito, T., Dojima, T., Toriyama, M., Park, E. Y. (2002) The effect of cell cycle on GFPuv gene expression in the baculovirus expression system. J. Biotechnol. 93, 121-129.

(26) Hashida, M., Takemura, S., Nishikawa, M., and Takakura, Y. (1998) Targeted delivery of plasmid DNA complexed with galactosylated poly(L-lysine). J. Controlled Release 53, 301-310. (27) Litzinger, D. C., Buiting, A. M. J., Rooijen, N., and Huang, L. (1994) Effect of liposome size on the circulation time and intraorgan distribution of amphipathic poly(ethylene glycol)-containing liposomes. Biochim. Biophys. Acta 1190, 99-107. (28) Maeda, H., and Matsumura, Y. (1989) Tumoritropic and lymphotropic principles of macromolecular drugs. Crit. ReV. Ther. Drug Carrier Syst. 6, 193-210. (29) Wakebayashi, D., Nishiyama, N., Yamasaki, Y., Itaka, K., Kanayama, N., Harada, A., Nagasaki, Y., and Kataoka, K. (2004) Lactose-conjugated polyion complex micelles incorporating plasmid DNA as a targetable gene vector system: their preparation and gene transfecting efficiency against cultured HepG2 cells. J. Controlled Release 95, 653-664. (30) Ward, C. M., Read, M. L., and Seymour, L. W. (2001) Systemic circulation of poly(L-lysine)/DNA vectors is influenced by polycation molecular weight and type of DNA: differential circulation in mice and rats and the implications for human gene delivery. Blood 97, 2221-2229. (31) Lee, S. C., Kim, C., Kwon, I. C., Chung, H., and Jeong, S. Y. (2003) Polymeric micelles of poly(2-ethyl-2-oxazoline)-block-poly(-caprolactone) copolymer as a carrier for paclitaxel. J. Controlled Release 89, 437-446. (32) Gref, R., Minamitake, Y., Peracchia, M. T., Trubetskoy, V., Torchilin, V., and Langer, R. (1994) Biodegradable long-circulating polymeric nanospheres. Science 263, 1600-1603. (33) Han, J. H., Oh, Y. K., Kim, D. S., and Kim, C. K. (1999) Enhanced hepatocyte uptake and liver targeting of methotrexate using galactosylated albumin as a carrier. Int. J. Pharm. 188, 39-47. BC0502107