Folic Acid-Conjugated Chitosan Nanoparticles Enhanced

Mar 11, 2010 - Assay was performed by using a TNBS kit (Pierce Chemical Company; Rockford, IL) according to the manufacturer's instructions. ... The s...
9 downloads 13 Views 3MB Size
Bioconjugate Chem. 2010, 21, 679–689

679

Folic Acid-Conjugated Chitosan Nanoparticles Enhanced Protoporphyrin IX Accumulation in Colorectal Cancer Cells Shu-Jyuan Yang,§ Feng-Huei Lin,§ Kun-Che Tsai,§ Ming-Feng Wei,§ Han-Min Tsai,§ Jau-Min Wong,§ and Ming-Jium Shieh*,§,‡ Institute of Biomedical Engineering, College of Medicine and College of Engineering, National Taiwan University, Taipei 10051, Taiwan, and Department of Oncology, National Taiwan University Hospital and College of Medicine, Taipei 10002, Taiwan. Received November 2, 2009; Revised Manuscript Received February 11, 2010

Folic acid can be covalently conjugated to chitosan molecules via its γ-carboxyl moiety and thus retain a high affinity for colorectal cancer cells bearing folate receptor overexpression. Colorectal cancer is one of the leading causes of malignant death and often goes undetected with current colonoscopy practices. Improved methods of detecting dysplasia and tumors during colonoscopy will improve mortality. A folic acid conjugated chitosan nanoparticle as a suitable vehicle for carrying 5-aminolaevulinic acid (5-ALA) is developed to enhance the detection of colorectal cancer cells in vivo after a short-term uptake period. Chitosan can be successfully conjugated with folic acid to produce folic acid-chitosan conjugate, which is then loaded with 5-ALA to create nanoparticles (fCNA). The loading efficiency of 5-ALA in fCNA particles and the z-average diameter were in the range 35-40% and 100 nm, respectively. The zeta-potential for fCNA was 20 mV, enough to keep the nanoparticle stable without aggregation. The fCNA is then incubated with HT29 and Caco-2 colorectal cancer cell lines overexpressing folate receptor on the surface of the cell membrane to determine the rate of accumulation of protoporphyrin IX (PpIX). The results show that fCNA can be taken up more easily by HT29 and Caco-2 cell lines after short-term uptake period, most likely via receptor-mediated endocytosis, and the PpIX accumulates in cancer cells as a function of the folate receptor expression and the folic acid modification. Therefore, the folic acid-chitosan conjugate appears to be an ideal vector for colorectal-specific delivery of 5-ALA for fluorescent endoscopic detection.

INTRODUCTION Folic acid is a ligand that is useful for targeting cell membrane and enhancing nanoparticle endocytosis via the folate receptor. It is a stable, inexpensive, and generally poorly immunogenic chemical with a high affinity for the folate receptor (1, 2). Because the folate receptor overexpresses on many human epithelial cancer cell surfaces, including cancers of the ovary, kidney, uterus, colon, and lung, conjugation of drugs and macromolecules with folic acid can enhance their uptake and targeting ability (3-9). Folic acid conjugates, which are covalently derivative via folate’s γ-carboxyl moiety, can maintain a high affinity to the folate receptor, and the mechanism of cellular uptake of folic acid conjugates by folate receptors is as effective as that of folic acid chemical (2). They can enter cells by folate receptor-mediated endocytosis and move through many organelles by vesicular trafficking, which can supply materials to release into cell cytoplasm. Later, the unligated folate receptor may recycle to the cell surface to transport more folic acid conjugates (2, 3). This process has more advantages for the efficient intracellular delivery of anticancer agents or macromolecules than the one that relies on cell membrane markers. The incidence of colorectal cancer is increasing worldwide and its prognosis remains poor (10). In general, survival is inversely related to extent of tumor spread at time of detection. Relative rates of survival of 50% at three years and 40% at five years have remained unchanged since the 1960s (11). Prognosis * Corresponding author. Shieh, M.-J., Tel: 886-2-23123456 ext 67512; Fax: 886-2-23940049; E-mail: [email protected]; [email protected]. § National Taiwan University. ‡ National Taiwan University Hospital and College of Medicine.

is excellent with detection at an early stage. Unfortunately, small colorectal neoplasia and early cancer are frequently overlooked during endoscopy. Therefore, a powerful and highly sensitive tool for the detection of precancerous lesions would be of great value. Photodiagnosis is one of the most promising and noninvasive methods for detecting malignant or premalignant tissue (12). Currently, detection of abnormal tissue usually involves the use of an exogenous chromophore, such as protoporphyrin IX (PpIX), excited by optima light to generate fluorescence in cancer lesions (13). The 5-aminolevulinic acid (5-ALA) used in the study, a precursor in heme group synthesis, is totally degraded intracellularly and converted to PpIX. Because the decomposition rate of PpIX in cancer cells differs from that in normal cells, the photosensitive fluorophore, PpIX, can be used to detect cancer lesions (14, 15). Protoporphyrin IX in normal cells is usually maintained at a constant level and quickly degraded if it exceeds the equilibrium level. However, this feedback mechanism does not function well in cancer cells, where PpIX degrades at a slow pace and accumulates at very high concentrations. In addition, ferrochelatase, one of the major enzymes involved in converting PpIX into heme, remains at low activity in cancer cells and results in the limited availability of iron that partially contributes to the high concentration of PpIX in those abnormal cells (16). Generally, PpIX can be totally degraded within 2 to 4 h in normal cells and 12-24 h in cancer cells (17). Because unprotected 5-ALA is easily engulfed by bacteria in the gastrointestinal tract (GI tract), its presence is open to misinterpretation and it has not been suitable as a marker during endoscopy (18). We have developed a chitosan nanoparticle vehicle for carrying 5-ALA that prevents bacterial uptake (19). Chitosan is the polymer of 2-amino-2-deoxy-β-D-glucan by

10.1021/bc9004798  2010 American Chemical Society Published on Web 03/11/2010

680 Bioconjugate Chem., Vol. 21, No. 4, 2010

glycosidic linkages. The primary amino groups on the molecular chain of chitosan have special properties that make chitosan more useful in pharmaceutical applications. Compared with other biological polymers, chitosan is more cationic, which allows it to approach cell membranes more easily. In addition, it has muco-adhesive properties that prolong its retention to targeted substrates (20-24). Additionally, chitosan induces no allergic reactions or immuno-rejections. Also, its bacteriostatic properties discourage bacterial uptake in the GI tract (25). Furthermore, chitosan is a linear polyamine with an abundance of free amino groups that promote cross-linkage. Its cationic nature also promotes ionic cross-linkage with multivalent anions, such as sodium tripolyphosphate (STPP) (26). Our previous study using chitosan as a vehicle for 5-ALA has shown that the accumulation of PpIX in colorectal cancer cells could be detected after 24 h incubation (19). In order to enhance the uptake and targeting ability of nanoparticles into targeted cells, target-specific ligands are usually used to conjugate to the surface of the polymer carrier. The main aim of this study was to determine the suitability of the synthesized folic acid-chitosan conjugates for carrying 5-ALA and to determine its targeting and uptake efficiency in different human colorectal cancer cell lines (HT29 and Caco-2) by folate receptor-mediated endocytosis after a short-term uptake period for the clinical practice.

EXPERIMENTAL PROCEDURES Materials. Chitosan (99% deacetylation and 150 kDa M.W.) and 5-aminolevulinic acid (5-ALA) were purchased from Fluka (Germany). Sodium tripolyphosphate (STPP) was purchased from Wako Pure Chemical Industries (Osaka, Japan). Folic acid was purchased from TCI (Tokyo, Japan). 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) was purchased from Acros Organics (New Jersey, USA); anhydrous dimethyl sulfoxide (DMSO) was purchased from Riedel-de Hae¨n (Germany). They were all reagent grade and used without further purification. Conjugation of Folic Acid with Chitosan. The process of conjugation of folic acid to chitosan molecules is described as follows: A solution of folic acid and EDC in 20 mL anhydrous DMSO (molar ratio was 1:1) was prepared and stirred at room temperature until EDC and folic acid were well-dissolved and mixed. The mixture was then added slowly to 0.5% (w/v) chitosan in acetic acid aqueous solution (0.1 M, pH 4.7) and stirred at room temperature in the dark for 16 h to let folic acid conjugate onto chitosan molecules. The solution was brought to pH 9.0 by dripping with NaOH aqueous solution (1.0 M) and centrifuged at 2500 rpm to spin down the folic acid-chitosan conjugate. The precipitate was dialyzed first against phosphate buffered saline (PBS, pH 7.4) for 3 days and then against water for 4 days. Finally, the folic acid conjugated chitosan was isolated as a sponge by lyophilization and kept for follow-up study (6, 7). Folic acid-chitosan conjugates were formed at selected amino groups of chitosan to folic acid molar ratios from 1:0.02 to 1:0.2. Abbreviations for different folic acid-chitosan conjugates are shown in Table 1. Characterization of the Folic Acid-Chitosan Conjugates. The chemical compositions of chitosan, folic acid-chitosan conjugates, and folic acid were characterized with an ATR-FTIR spectrometer (VARIAN UMA600; Palo Alto, CA) to determine the characteristics of the folic acid-chitosan conjugates. The 1H NMR spectra of folic acid-chitosan conjugates were recorded on a Bruker Avance-400 MHz NMR spectrometer (Bruker BioSpin, MA, USA) using deuterium acetic acid-d4/ deuterium oxide solution (1/4 v/v) as a solvent. The folic acid to amino groups molar ratios in folic acid-chitosan conjugates were estimated by determining the

Yang et al. Table 1. Molar Ratio of Folic Acid/Amino Groups in Folic Acid-Chitosan Conjugates and Yield of the Synthesis Conjugatesa folic acid-chitosan conjugates abbreviation

formulated molar ratio of folic acid/NH2

molar ratio of folic acid/NH2 in folic acid-chitosan conjugate mean ( SD

yield (%)

f02Chi f04Chi f06Chi f08Chi f10Chi f20Chi

0.02 0.04 0.06 0.08 0.10 0.2

0.008 ( 0.0003 0.015 ( 0.0008 0.018 ( 0.0007 0.028 ( 0.0011 0.033 ( 0.0018 0.059 ( 0.0039

81.6 ( 2.02 77.0 ( 0.27 71.2 ( 2.73 68.7 ( 3.13 67.5 ( 0.39 53.2 ( 1.15

a

(n ) 5).

intensity of absorption maximum (285 nm) of folate derivative in 0.02% folic acid-chitosan conjugate solution with an ultraviolet-visible (UV-vis) spectrophotometry (Cary 50 Conc, Varian; Australia). In this study, folic acid powder was dissolved in 0.1 M NaOH solution and diluted to a series of gradient folic acid standard solutions that were used to prepare the calibration curve in the range 0.0025-0.05 mg/mL for further analysis. Synthesis of Nanoparticles. The preparation of chitosan nanoparticles loaded with 5-ALA was based on the ionic gelation interaction between positively charged chitosan and negatively charged STPP at room temperature (27). The negative charge of ionized 5-ALA also partially contributed to the gelation interactions in this study. The preparation process is briefly described as follows: 0.05% chitosan and folic acidchitosan conjugate solutions were prepared by dissolving 0.5 mg chitosan powder and folic acid-chitosan conjugate sponge in 1 mL 0.01 M acetic acid solution at pH 4.0, respectively. 0.15% 5-ALA solution was prepared by dissolving 1.5 mg 5-ALA powder in 1 mL 0.05% STPP solution, where STPP were dissolved in PBS beforehand. Either 2 mL of STPP solution or 2 mL of 5-ALA solution was added to 5 mL of chitosan or folic acid-chitosan conjugate solution with a peristaltic pump at a flow rate of 0.5 mL/min to prepare the CN, CNA, fCN, and fCNA nanoparticle suspended solutions, where CN and fCN were represented as chitosan and folic acid conjugated chitosan nanoparticles without 5-ALA loading, and chitosan and folic acid conjugated chitosan nanoparticles loaded with 5-ALA were termed CNA and fCNA. The preparation of fCNA was schematically illustrated in Figure 1. The prepared nanoparticle suspended solutions were later used directly without further treatment. Particle Size and the Zeta-Potential of CN, CNA, fCN, and fCNA. Measurements of particle size and zeta-potential of various nanoparticles were performed on the Zetasizer-3000 (Malvern Instruments; Westborough, MA) by dynamic light scattering measurements and laser Doppler electrophoresis, respectively. Particle size was measured at 25 °C with a 90° scattering angle based on the Zetasizer-3000 internal setting. The cumulative curve was used to present the mean hydrodynamic diameter. Measurements of zeta-potential were made using the aqueous flow cell in the automatic mode at 25 °C (28). Loading Efficiency of 5-ALA in CNA and fCNA. Loading efficiency of 5-ALA in nanoparticles was assessed by centrifuging the particle solution at a 20 000 g force to spin down the particles. The suspension was collected and then reacted with 2,4,6-trinitrobenzene sulfonic acid (TNBS) as an assay reagent to detect 5-ALA in the suspension. Assay was performed by using a TNBS kit (Pierce Chemical Company; Rockford, IL) according to the manufacturer’s instructions. In the study, 5-ALA powder was dissolved in reaction buffer (0.1 M sodium bicarbonate, pH 8.5) and diluted to a serial of gradient 5-ALA standard solutions that were used to prepare the calibration curve in the range of 2-20 µg/mL for further analysis. The loading

Folic Acid-Conjugated Nanoparticles for Cancer Image

efficiency of 5-ALA in the nanoparticles was calculated with the following equation: loading efficiency (%) )

Ct - Cf × 100% Ct

where Ct and Cf were total amount of 5-ALA and free amount of 5-ALA in the suspension, respectively. Transmission Electron Microscope Examination. Carboncoated 200 mesh copper grids were immersed in CN, CNA, fCN, and fCNA solutions. Grids were placed on delicate-taskwipers to absorb excess liquid and then dried in a desiccator overnight. The dried copper grids with CN, CNA, fCN, and fCNA were examined under the Hitachi TEM H-7500 (Tokyo, Japan). Determining Expression of Folate Receptor among Colorectal Cancer Cell Lines. Expression content of folate receptor on cell membrane was determined by Western blot method. Two colorectal cancer cell lines (HT29 and Caco-2) were pretreated by scraping the cell cultures into cold lysis buffer (RIPA buffer) containing protease inhibitors. Whole cell lysate was assayed for the protein concentration by using the DC protein assay (Bio-Rad; Hercules, CA) with bovine serum albumin as standard. Total protein (30 µg) was separated on 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) gels, and then transferred to polyvinylidene fluoride (PVDF) membranes (Amersham Biosciences; Sweden). The PVDF membranes were incubated in blocking buffer (5% nonfat dry milk, 0.1% Tween20 in TBS) at room temperature for 60 min, and subsequently treated with primary antibody, rabbit antihuman folate receptor (1:1000 dilution, Santa Cruz Biotechnology;

Bioconjugate Chem., Vol. 21, No. 4, 2010 681

CA, USA), in blocking buffer overnight at 4 °C. Membranes were washed with Tris-buffered saline (TBS) containing 0.1% Tween20 (USB Corporation; Cleveland, OH), and further incubated with the horseradish peroxidase-conjugated antirabbit IgG (1:5000 dilution, Sigma) as secondary antibody at room temperature for 60 min. Finally, membranes were incubated in blocking buffer at room temperature for 60 min, and then washed with PBS containing 0.1% Tween20. The bands of proteins were visualized by using chemiluminescence enhancing (Amersham Parmacia Biotech; Piscataway, NJ) and exposed to Hyperfilm MP. The signals of antihuman folate receptor in the Western blotting were quantified with the software ImageQuant 5.1, and the data were normalized to β-actin expression. Measurement of Accumulated PpIX. Cell lines HT29 and Caco-2 were cultured in Dulbecco’s Modified Eagle’s Medium (GIBCO; Grand Island, NY) supplement with 10% (v/v) fetal bovine serum, 2 mM glutamine, 10 mM nonessential amino acids, 50 IU/mL penicillin, and 50 Ag/mL streptomycin, at 37 °C and in an atmosphere of 5% CO2, respectively. Cultured medium was changed on alternate days until confluent. The following quantities of cells were seeded onto 24-well culture plates: 1.2 × 105/well HT29 and 3 × 104/well Caco-2. After cell lines were cultured for 24 h, the medium was replaced with fresh medium containing CNA and fCNA. The CNA and fCNA solutions were concentrated to 10× by Amicon Ultra (Centrifugal Filter Devices, Millipore) to remove the unloaded 5-ALA beforehand. Moreover, because of no blood supplement in the GI tract, the fresh medium without serum supplement was used. Cells were further cultured for 3, 6, and 12 h to allow uptake of nanoparticles and to convert 5-ALA to PpIX. In order to determine the effect of folic acid-folate receptor mediated

Figure 1. Schematic illustration of the preparation folic acid-conjugated chitosan nanoparticles for PpIX accumulation in colorectal cancer cells.

682 Bioconjugate Chem., Vol. 21, No. 4, 2010

endocytosis on the endogenously synthesized PpIX, fresh culture media, both with or without free folic acid, were tested at the same time. The concentration of free folic acid (5 mM) used was expected to result in complete saturation of folate receptors (5). After removal of the cultured medium, plates were washed with PBS three times. The PpIX were extracted with DMSO, and the fluorescent intensity of PpIX was determined by the spectrofluorometer (Molecular Device Spectra Max Gemine XS; Sunnyvale, CA) with the excitation wavelength of 405 nm and the emission wavelength at 635 nm. The fluorescent intensity was expressed as accumulative percentage. In order to evaluate the feasibility of the prepared fCNA on colorectal cancer cells detection after short-term uptake period, the glass plate was placed in a Petri dish and seeded with HT29, Caco-2, or a normal cell line (human colon fibroblast (CCD112)). After cells were cultured for 24 h, medium was replaced with fresh medium containing CAN, fCNA, and 5-ALA. To confirm the nanoparticle uptake content as a function of the folate receptor expression and the folic acid modification, a fluorescent compound, calcein, was incorporated simultaneously into CNA and fCNA as a fluorescent marker. Cells were further cultured for 12 h to allow for uptake of nanoparticles and conversion of 5-ALA to PpIX. After the cultured medium was removed, the glass plate was washed three times with PBS and the cells were fixed with 10% formalin and then examined under a spectral confocal and multiphoton system (Leica TCS SP5, Wetzlar, Germany). The average fluorescent intensity of calcein in cells was quantified with the MetaMorph imaging software (v 7.5, Molecular Devices, USA) to demonstrate the content of nanoparticle uptake. The data represented an average of five regions obtained from randomly chosen areas of each sample, and the brightness of the green emission light of the CNA group was designated as 100% for the individual colorectal cancer cell lines. Statistical Analysis. Mean plus/minus standard deviation (SD) and graphs were used to describe the data. One-way analysis of variance (ANOVA) with post hoc muti-comparison methods, such as Fisher’s LSD test and Tamhane’s T2 test, was used to assess the differences in the fluorescent intensity of PpIX and calcein in colorectal cancer cells fed with CNA and fCNA. All p values were two-sided, and their significance level was 0.05. The software of statistic package for social science 11.0 (SPSS 11.0) was used to conduct all statistical analysis.

Yang et al.

Figure 2. FT-IR spectra of folic acid-chitosan conjugates (A) chitosan, (B) f06Chi, (C) f20Chi, and (D) pure folic acid. Table 2. Relative Intensity Ratio of the NsH Bending over CdO Stretching in Chitosan and Folic Acid-Chitosan Conjugates

sample

chitosan

f06Chi

Characterization of Folic Acid-Chitosan Conjugates. Folate acid-chitosan conjugates could be synthesized by chemically linking folic acid to chitosan via EDC as a carboxyl activating agent. They were formed at the selected amino groups of chitosan to folic acid molar ratios from 1:0.02 to 1:0.2. Abbreviations for different folic acid-chitosan conjugates are shown in Table 1. The FT-IR spectra of the chitosan, folic acid-chitosan conjugates (f06Chi, f20Chi), and pure folic acid were shown in Figure 2. Because the folic acid modification in the folic acid-chitosan conjugate, f02Chi, was too low to easily determine the difference between chitosan and folic acid-chitosan conjugate in FT-IR spectra, it was not shown in Figure 2. Characteristic FT-IR absorption peaks of folic acid at 1450, 1520, and 1605 cm-1 were observed in the spectrum of folic acid-chitosan conjugates. In the pure folic acid and folic acid-chitosan conjugates, the absorption bands at 1450 and 1605 cm-1 were assigned to the stretching vibrations of CdC in the backbone of the aromatic ring (29). Furthermore, chitosan and folic acid-chitosan conjugates all contained the -CONHgroup. The carbonyl absorption band, CdO, is known as the amide I band, and N-H is termed the amide II band. Their position depends on the degree of hydrogen bonding and, thus,

Amide II

free N-H

Amide I

bonded CdO free CdO free N-H

Amide II Amide I

RESULTS f20Chi

wavenumber (cm-1)

band

Amide II Amide I

bonded N-H bonded CdO free CdO free N-H bonded N-H bonded CdO free CdO

1541 1558 1635 1716 1541 1558 1565 1635 1716 1541 1558 1565 1635 1716

intensity ratio (Amide II/ Amide I)

1.813 1 2.267 1 2.276 1

the physical state of the compound. The most significant IR frequencies of the chitosan and folic acid-chitosan conjugates are the amide I band (carbonyl stretching) and the amide II band Table 3. z-Average Diameter, Zeta-Potential, and Loading Efficiency of Chitosan Nanoparticles Prepared in Different Conditions

sample

z-average diametera mean ( SD (nm)

CN CNA f02CN f02CNA f06CN f06CNA

102.4 ( 7.74 96.0 ( 4.82 100.1 ( 10.75 93.6 ( 5.80 99.0 ( 7.00 112.8 ( 13.19

PDI

zetapotential (mV)

loading efficiency mean ( SD (%)

0.252 0.392 0.434 0.381 0.444 0.457

25.0 ( 1.84 24.4 ( 1.78 22.6 ( 1.54 21.9 ( 2.09 21.1 ( 2.19 20.1 ( 1.76

43.1 ( 5.71 38.9 ( 3.52 35.5 ( 1.40

b

a z-average diameter is the mean diameter based on the intensity of scattered light. b PDI ) polydispersity index (ratio of weight- to number-average particle diameter).

Folic Acid-Conjugated Nanoparticles for Cancer Image

Bioconjugate Chem., Vol. 21, No. 4, 2010 683

Figure 3. 1H NMR spectrum of f02Chi (A), f06Chi (B), and f20Chi (C) folic acid-chitosan conjugates in an acetic acid-d4/D2O solution (1/4 v/v).

(N-H deformation mode) vibration. The former is typically around 1720-1740 cm-1 and is shifted to lower frequencies by hydrogen bonding (weakening of the force constant due to partial single-bond character). The latter is typically between 1500 and 1550 cm-1 and is believed to shift to the higher frequencies upon hydrogen-bond formation (strengthening of the force constant for deformation vibrations) (30). Figure 2 shows that there were two absorption peaks for the CdO group. The peak at about 1635 cm-1 represented the bonded CdO stretching and the peak at about 1716 cm-1 represented the free

CdO stretching (31). The amide II band was observed at a relatively high frequency (1565 cm-1), most probably due to the hydrogen-bond formation. The peaks at about 1541 and 1558 cm-1 were related to free N-H bending. In Table 2, it can be seen that the amide II/I intensity ratio increased with increasing folic acid conjugation in folic acid-chitosan conjugates. Successful synthesis of folic acid-chitosan conjugates was also confirmed by a 1H NMR spectroscopy, and the 1H NMR spectra of folic acid-chitosan conjugates are shown in Figure 3. The signals at δ 1.65, 2.88, and 3.08-3.64 ppm were assigned

684 Bioconjugate Chem., Vol. 21, No. 4, 2010

Yang et al.

Figure 4. Photos from TEM for CN and CNA modified with folic acid or without. The scale bar in the picture is 1 µm.

to the resonance of the monosaccharide residue protons, -COCH3, -CH-NH-, and -CH2-O-, respectively (2, 7). The signals appearing at δ 6.3-8.5 ppm in the 1H NMR spectra of folic acid-chitosan conjugates attributed to the resonance of the folate aromatic protons, and they revealed that the couple of the folate residue to the chitosan could be achieved via EDC-mediated reaction (7, 32). Table 1 shows the molar ratio of folic acid to amino groups in folic acid-chitosan conjugates evaluated by an absorbance at λ ) 285 nm in the UV-vis spectra of the conjugates. The lowest molar ratio was 0.008 in f02Chi, which was two-fifths of the formulated molar ratio; however, the highest molar ratio was up to 0.059 (f20Chi), three-tenths of the formulated molar ratio only. The incremental trend of the folic acid conjugation in the folic acid-chitosan conjugates matched that of folic acid addition, but it was still lower than the formulated molar ratio. Moreover, the yield of f02Chi was 81.6%, and decreased to 53.2% as the folic acid addition in the reaction increased (f20Chi) (Table 1). Particle Size, Zeta-Potential, and Loading Efficiency of 5-ALA in Nanoparticles. Table 3 shows the z-average diameter, zeta-potential, and loading efficiency of chitosan nanoparticles, which were prepared in different conditions. When chitosan was conjugated with an abundance of folic acid molecules (f08Chi, f10Chi, and f20Chi), it was difficult to dissolve it into 0.01 M acetic acid solution completely, most likely due to the unsolvable property of folic acid in acetic acid (33). Accordingly, folic acid-chitosan conjugates with large folic acid modification were not suitable for use as a carrier for 5-ALA in the present prepared nanoparticle condition. As shown in Table 3, the z-average diameter of the prepared nanoparticles did not vary significantly with change of folic acid on chitosan polymer (CN and fCN) or 5-ALA loading (CNA and fCNA). Also, the z-average diameter of prepared nanoparticles was about 100 nm. The zeta-potential analysis revealed that the nanoparticles

prepared in different conditions all bore a positive charge (in the range 20-25 mV). However, when the CN was modified with folic acid, as was so-called fCN, the zeta-potential of particles decreased with increase in the percentage with folic acid conjugation. If the CN and fCN were loaded with 5-ALA, the zeta-potential of the CNA and fCNA decreased slightly as compared with CN and fCN, respectively. The loading efficiency of 5-ALA in CNA was 43% and decreased to 35% with increasing percentage of folic acid conjugation. Figure 4 shows the particle sizes of CN, CNA, fCN, and fCNA as determined by transmission electron microscopic examination (TEM). The particle size distribution of CN and CNA was major in the range 60-200 nm, but fell into the range 100-600 nm for f06CN and f06CNA. Therefore, the particle size distribution strongly depended on and increased with the increase of percentage of conjugated folic acid. These results are in agreement with the polydispersity index (PDI) values of the prepared nanoparticles measured by dynamic light scattering measurements using the Zetasizer-3000.

Figure 5. (A) Western blot analysis of folate receptor expression and (B) the ratio of folate receptor to β-actin quantified with the software ImageQuant 5.1 on HT29 and Caco-2 colorectal cancer cell membrane (n ) 3).

Folic Acid-Conjugated Nanoparticles for Cancer Image

Bioconjugate Chem., Vol. 21, No. 4, 2010 685

Figure 6. (A) Relative fluorescent intensity of PpIX in HT29 and Caco-2 cells when fed with the prepared CNA, f02CNA, and f06CNA for 3, 6, and 12 h. (B) In order to determine the effect of folic acid-folate receptor mediated endocytosis on the endogenously synthesized PpIX, the relative fluorescent intensity of PpIX in HT29 and Caco-2 cells was also be estimated when cultured in a fresh medium with the nanoparticle suspension and 5 mM free folic acid addition for 3, 6, and 12 h. *: p < 0.05.

Expression of Folate Receptor among Colorectal Cancer Cell Lines. Because folate receptors are known to be overexpressed and actively internalized through folate receptormediated endocytosis in various types of colorectal cancer cells, folic acid has been considered to be a suitable ligand for improving the cellular uptake of drugs and macromolecules into cells (3). Therefore, the expression of folate receptor on HT29 and Caco-2 colorectal cancer cells was examined by the Western blot method and shown in Figure 5. It was apparent that expression of folate receptor depended on the cell line: expression of folate receptor was of a low level for Caco-2 cells, whereas HT29 cells exhibited 1.5-fold higher expression of folate receptors than did Caco-2 cells. Measurement of PpIX in Colorectal Cancer Cells. Figure 6 shows the relative fluorescent intensity of PpIX in HT29 and Caco-2 cells cultured in a medium with or without free folic acid addition. The brightness of the emission light of the CNA group was designated as 100% for the individual colorectal cancer cell lines. Figure 6A shows that increase of bright red emission light of PpIX in HT29 and Caco-2 cells was only 31.3% and 31.2%, respectively, when the cells were cultured with f02CNA for 12 h. However, when f06CNA was fed to HT29 and Caco-2 cells, increases in intensity of brightness were 42.5% and 28.8% after 6 h of incubation and significantly up to 74.8% and 57.2% after incubating for 12 h, respectively. In order to evaluate the role of folic acid in the cellular uptake of folic acid-conjugated chitosan nanoparticles, CNA, f02CNA, and f06CNA coincubated with free folic acid were fed to the HT29 and Caco-2 cells. Figure 6B shows that there were no significant brightness intensity differences among CNA, f02CNA, and f06CNA groups and that free folic acid molecules could compete with f02CNA and f06CNA for folate receptors and suppress the receptor-mediated endocytosis of the prepared particles. These results showed that folic acid-chitosan con-

jugates could significantly enhance the cell uptake of 5-ALA and the accumulation of PpIX in HT29 and Caco-2 cells via the folate receptor-mediated endocytosis. Figure 7 shows the red fluorescence of PpIX excited by a green laser and observed under a spectral confocal and multiphoton system in HT29, Caco-2, and CCD112 cells, which were fed with CNA, f02CNA, f06CNA, or 5-ALA for 12 h. Pure 5-ALA or 5-ALA in the prepared nanoparticles has been released in colorectal cancer cell lysosomes and then converted into PpIX, which accumulated in cells and then was excited by a green laser and emitted red light. Figure 7 shows CNA, fCNA, and 5-ALA taken up by CCD112 cells. Because of the quick degradation rate of PpIX in the normal cell, no red light was emitted and interference with malignant cell identification was unlikely. A compound, calcein, is a membrane-impermeant fluorescent dye and can be incorporated into the prepared CNA and fCNA simultaneously by the electrostatic interaction between negatively charged calcein and positively charged chitosan. When HT29 and Caco-2 cells were fed with the calcein-incorporated CNA and fCNA nanoparticles for 12 h, the green brightness of calcein excited by a UV light increased with the increasing folic acid modification in the prepared nanoparticles (Figure 8A). As presented in Figure 8B, HT29 cells fed with the highest folic acid conjugated nanoparticles (f06CNA) exhibited a substantial calcein fluorescent intensity, suggesting the high activity for folate receptor-mediated endocytosis in HT29 cells with higher folate receptor expression. These results also demonstrated that enhancement of cellular association and uptake was induced by the specific interaction of folic acid conjugated nanoparticle with its receptor on the cellular surface and by

686 Bioconjugate Chem., Vol. 21, No. 4, 2010

Yang et al.

Figure 7. Spectral confocal and multiphoton system images of PpIX in HT29, Caco-2, and CCD112 cells when incubated with the prepared CNA, f02CNA, f06CNA, and pure drug 5-ALA for 12 h. The scale bar in the image is 25 µm.

subsequent internalization through the receptor-mediated endocytosis.

DISCUSSION Folic acid is a stable, inexpensive, and generally poorly immunogenic chemical. It can be covalently conjugated to chitosan molecules via its γ-carboxyl moiety and thus retain a high affinity for folate receptors. Moreover, the ability of cellular uptake of folic acid-chitosan conjugate via folate receptormediated endocytosis is still as good as that of folic acid (6). In this study, the folic acid-chitosan conjugates bearing different content of folate derivatives could be fabricated by varying the folic acid to amino groups of chitosan molar ratios in the synthesis procedure (Table 1). The FT-IR and 1H NMR spectra of the folic acid-chitosan conjugates confirmed that folic acid could be conjugated onto chitosan molecules successfully, but the relevant signals in 1H NMR spectra, which were recommended by Wang et al. (32) and could be used to ascertain the regiochemistry of the linkage of folate derivatives and chitosan, were masked by the strong and broad signals owing to the resonances of the protons associated to chitosan monosaccharide residues (Figures 2 and 3). Thus, we were unable to determine the regiochemistry of the folate linkage in folic acid-chitosan conjugates from 1H NMR spectra, but it could be inferred that folic acid was conjugated to an amino group via its γ-carboxyl moiety mostly due to the reaction conditions chosen and the steric constraints. Chitosan was chosen as the cationic polymer because it provides many free amino groups that can be quickly gelled to form micro/nanoparticles once it contacts polyanions (34). In

this study, we found a weak association between the particle size of the prepared nanoparticles and the folic acid conjugation or 5-ALA load. However, the zeta-potential and loading efficiency of 5-ALA depended on the folic acid conjugation. Zeta-potential can greatly influence particle stability through electrostatic repulsion between particles. Greater zeta-potential resulted in improved stability of nanoparticles in the solution and a narrower size distribution, presumably because the charged particles repelled one another and therefore overcame the natural tendency to aggregate (35). As shown in Table 3, the zeta-potential of the CN and CNA was about 25 mV. This provided enough surface charge to stabilize the particles from aggregation, which was decreased when CN and CNA were conjugated with folic acid. When the more electroneutral folate derivatives displaced the protoned amino groups (f06CN and f06CNA), the zeta-potential was even down to 20 mV, and therefore it might not provide enough repulsive force to stabilize the particles from aggregation for long-term storage. These results are consistent with the Zetasizer-3000 measurements and TEM observations. When the prepared particles conjugated with folic acid, they got a large PDI value (Table 3) and a wide particle size distribution in the range 100-600 nm (Figure 4). When CNA was conjugated with folic acid (fCNA), the loading efficiency of 5-ALA decreased to 38% and 35% for f02CNA and f06CNA particles, respectively (Table 3). In f02CNA and f06CNA particles, a number of amino groups were replaced by folic acid and reduced the positive charges on chitosan molecules to attract 5-ALA molecules. Moreover, the spatial hinder of folate derivatives on chitosan backbone might

Folic Acid-Conjugated Nanoparticles for Cancer Image

Bioconjugate Chem., Vol. 21, No. 4, 2010 687

Figure 8. (A) Green fluorescent images of calcein in HT29 and Caco-2 cells when fed with the calcein-incorporated CNA, f02CNA, and f06CNA nanoparticles for 12 h. The scale bar in the image is 25 µm. (B) The relative fluorescent intensity of calcein in HT29 and Caco-2 cells. *: p < 0.05.

resist the interaction between chitosan and 5-ALA molecules. On the basis of these reasons, the loading efficiency appeared strongly dependent on the content of folic acid conjugation. In the previous studies (19), no red fluorescence of PpIX could be observed in E. coli when fed with the prepared CNA nanoparticles, and it pointed out that the uptake of 5-ALA in E. coli could be precluded by encapsulating 5-ALA in an antibacterial material, such as chitosan. Furthermore, the incorporation of 5-ALA in the CNA would be stable because of the lack of 5-ALA release from the particle and uptake by E. coli. Therefore, it was feasible to load 5-ALA in the CNA and fCNA nanoparticles and then deliver them into colorectal cancer cells via folate receptor-mediated endocytosis. Folate receptor is frequently overexpressed on the surface of many human cancer cell types, such as carcinomas of the kidney, ovary, uterus, testis, and adenocarcinomas of the colon (4). For this reason, many studies have used folic acid as a ligand with cationic liposomes and other polymers to target cells with folate receptor expression and promote its entry into malignant cells via folate receptor-mediated endocytosis. Figure 6A shows that, for f06CNA group, much more PpIX accumulated in HT29 and Caco-2 cells than that for CNA group. This enhancement of PpIX accumulation might be due to the high folic acid conjugation on f06CNA, which could result in stronger affinity to the folate receptor on HT29 and Caco-2 cell membrane and higher efficacy of internalization via folate receptor-mediated endocytosis. Moreover, variation in the degree of folate receptor among various colorectal cancer cell lines will result in variation

in degree of target uptake efficiency. The folate receptor expression on HT29 cell membrane was remarkable higher than that on Caco-2 cell membrane (Figure 5), and this resulted in a lower f06CNA engulfment and a lower difference in the enhanced PpIX accumulation in Caco-2 cells than in HT29 cells (Figure 6A). In addition, the improvement of cell uptake via electrostatic attraction between particles and cell membrane was not conspicuous for HT29 and Caco-2 cells due to no difference of PpIX accumulation in HT29 and Caco-2 cells when cultured in a fresh medium suspending with the prepared nanoparticles bearing different surface charge and free folic acid (Figure 6B). In previous studies, 5-ALA was encapsulated in chitosan nanoparticles and taken up by cells but not bacteria, which could interfere with the efficacy of detection of cancer cells. However, culturing the 5-ALA loading chitosan nanoparticle with cancer cells for 24 h would result in an observable fluorescent intensity (19). The conjugation of folic acid onto a chitosan molecular backbone via amide bond could promote the uptake of particles into a cell after only 12 h of incubation time. The HT29 and Caco-2 cells showed a stronger uptake of f06CNA nanoparticles than those of pure 5-ALA alone (Figure 7). When HT29 and Caco-2 cells were fed with f06CNA, the f06CNA was easily engulfed via folate receptor-mediated endocytosis, and then 5-ALA could be converted into to PpIX intracellularly to reach the visual fluorescent intensity noted after 12 h of incubation. Since PpIX in normal cells is always maintained at a constant level (16), no red fluorescence suggesting interference with the diagnostic results could be observed inside CCD112 cells

688 Bioconjugate Chem., Vol. 21, No. 4, 2010

(Figure 7). The results shown in Figure 8 confirmed that the f06CNA was easily taken up either in HT29 cells or in Caco-2 cells as against the CNA. However, the difference in the enhanced particle uptake evaluated from the calcein fluorescence intensity was much higher than that determined by PpIX accumulation, suggesting that the release of 5-ALA from f02CNA and f06CNA was slow and sustained and resulted in a lower improvement of PpIX accumulation. These findings suggest that 5-ALA would be a perfect precursor exogenous chromophore for detecting cancer lesions and that chitosan conjugated with folic acid could be a suitable vehicle for carrying 5-ALA. In clinical applications, because employing folic acid conjugated with chitosan nanoparticles only in oral form might result in dissolution in the stomach at a pH that is lower than that of the pKa of chitosan (34), coating the conjugate with a pH-sensitive polymer, such as poly(acrylic acid) (35), would be necessary. Furthermore, the usage of STPP and acetic acid solution in the particle preparation process will not cause any distress to the blood circulation or other part of human systems when the folic acid-conjugated chitosan nanoparticles are applied as an oral form for the colon-specific drug delivery system.

CONCLUSIONS In this study, a chitosan nanoparticle that was successfully conjugated with folic acid and loaded with 5-ALA was taken up by HT29 and Caco-2 cell lines, most likely via receptormediated endocytosis. Chitosan conjugated with folic acid appears to be an ideal vector for colorectal-specific drug delivery of 5-ALA for fluorescent endoscopic detection of colorectal cancer.

ACKNOWLEDGMENT This research work was funded by National Science Council, ROC (NSC98-2120-M-002-008).

LITERATURE CITED (1) Park, E. K., Lee, S. B., and Lee, Y. M. (2005) Preparation and characterization of methoxy poly(ethylene glycol)/poly(e-caprolactone) amphiphilic block copolymeric nanospheres for tumorspecific folate-mediated targeting of anticancer drugs. Biomaterials 26, 1053–1061. (2) Chan, P., Kurisawa, M., Chung, J. E., and Yang, Y. Y. (2007) Synthesis and characterization of chitosan-g-poly(ethylene glycol)folate as a non-viral carrier for tumor-targeted gene delivery. Biomaterials 28, 540–549. (3) Wang, S., and Low, P. S. (1998) Folate-mediated targeting of antineoplastic drugs, imaging agents, and nucleic acids to cancer cells. J. Controlled Release 53, 39–48. (4) Quintana, A., Raczka, E., Piehler, L., Lee, I., Myc, A., Majoros, I., Patri, A. K., Thomas, T., Mule´, J., Jr, and Baker, J. R. (2002) Design and function of a dendrimer-based therapeutic nanodevice targeted to tumor cells through the folate receptor. Pharm. Res. 19, 1310–1316. (5) Oyewumi, M. O., Yokel, R. A., Jay, M., Coakley, T., and Mumper, R. J. (2004) Comparison of cell uptake, biodistribution and tumor retention of folate-coated and PEG-coated gadolinium nanoparticles in tumor-bearing mice. J. Controlled Release 95, 613–626. (6) Mansouri, S., Cuie, Y., Winnik, F., Shi, Q., Lavigne, P., Benderdour, M., Beaumont, E., and Fernandes, J. C. (2006) Characterization of folate-chitosan-DNA nanoparticles for gene therapy. Biomaterials 27, 2060–2065. (7) Dube´, D., Francis, M., Leroux, J. C., and Winnik, F. M. (2002) Preparation and tumor cell uptake of poly(N-isopropylacrylamide) folate conjugates. Bioconjugate Chem. 13, 6856–6892.

Yang et al. (8) Liu, S. Q., Wiradharma, N., Gao, S. J., Tong, Y. W., and Yang, Y. Y. (2007) Bio-functional micelles self-assembled from a folate-conjugated block copolymer for targeted intracellular delivery of anticancer drugs. Biomaterials 28, 1423–1433. (9) Zhang, Y., Kohler, N., and Zhang, M. (2002) Surface modification of superparamagnetic magnetite nanoparticles and their intracellular uptake. Biomaterials 23, 1553–1561. (10) Rembacken, B., Fujii, T., and Kondo, H. (2001) The recognition and endoscopic treatment of early gastric and colonic cancer. Best Pract. Res. Clin. Gastroenterol. 15, 317–336. (11) Mountney, L., Sanderson, H., and Harris, J. (1994) Health Care Needs Assessment (Stevens, A., and Raftery, J., Eds.) Chapter 7, Vol 1, Radcliffe Medical Press, Oxford. (12) Allison, R. R., Mota, H. C., and Sibata, C. H. (2004) Clinical PD/PDT in North America: An historica review. Photodiagnosis Photodyn. Ther. 1, 263–277. (13) Allison, R. R., Cuenca, R., Downie, G. H., Randall, M. E., Bagnato, V. S., and Sibata, C. H. (2005) PD/PDT for gynecological disease: A clinical review. Photodiagnosis Photodyn. Ther. 2, 51–63. (14) Pottier, R. H., Chow, Y. F., LaPlante, J. P., Truscott, T. G., Kennedy, J. C., and Beiner, L. A. (1986) Non-invasivetech technique for obtaining fluorescence excitation and emission spectra in vivo. Photochem. Photobiol. 44, 679–687. (15) Kennedy, J. C., Pottier, R. H., and Ross, D. C. (1990) Photodynamic therapy with endogenous protoporphyrin IX: basic principles and present clinical experience. J. Photochem. Photobiol. B 6, 143–148. (16) Grimbergen, M. C., van Swol, C. F., Jonges, T. G., Boon, T. A., and van Moorselaar, R. J. (2003) Reduced specificity of 5-ALA induced fluorescence in photodynamic diagnosis of transitional cell carcinoma after previous intravesical therapy. Eur. Urol. 44, 51–56. (17) Steluti, R., De Rosa, F. S., Collett, J., Tedesco, A. C., and Bentley, M. V. (2005) Topical glycerol monooleate/propylene glycol formulations enhance 5-aminolevulinic acid in vitro skin delivery and in vivo protophorphyrin IX accumulation in hairless mouse skin. Eur. J. Pharm. Biopharm. 60, 439–444. (18) Gederaas, O. A., Rasch, M. H., Berg, K., Lagerberg, J. W., and Dubbelman, T. M. (1999) Photodynamically induced effects in colon carcinoma cells (WiDr) by endogenous photosensitizers generated by incubation with 5-aminolaevulinic acid. J. Photochem. Photobiol. B 49, 162–170. (19) Yang, S. J., Shieh, M. J., Lin, F. H., Lou, P. J., Peng, C. L., Wei, M. F., Yao, C. J., Lai, P. S., and Young, T. H. (2009) Colorectal cancer cell detection by 5-aminolaevulinic acid-loaded chitosan nano-particles. Cancer Lett. 273, 210–220. (20) Berscht, P. C., Nies, B., Liebendorfer, A., and Kreuter, J. (1994) Incorporation of basic fibroblast growth factor into methylpyrrolidinone chitosan fleeces and determination of the in vitro release characteristics. Biomaterials 15, 593–600. (21) Thanou, M., Verhoef, J. C., and Junginger, H. E. (2001) Chitosan and its derivatives as intestinal absorption enhancers. AdV. Drug. DeliVery ReV. 50, S91–S101. (22) Thanou, M., Verhoef, J. C., and Junginger, H. E. (2001) Oral drug absorption enhancement by chitosan and its derivatives. AdV. Drug. DeliVery ReV. 52, 117–126. (23) Agnihotri, S. A., Mallikarjuna, N. N., and Aminabhavi, T. M. (2004) Recent advances on chitosan-based micro- and nanoparticles in drug delivery. J. Controlled Release 100, 5–28. (24) Qi, L., Xu, Z., Jiang, X., Hu, C., and Zou, X. (2004) Preparation and antibacterial activity of chitosan nanoparticles. Carbohydr. Res. 339, 2693–2700. (25) Du, J., Zhang, S., Sun, R., Zhang, L. F., Xiong, C. D., and Peng, Y. X. (2005) Novel polyelectrolyte carboxymethyl konjac glucomannan-chitosan nanoparticles for drug delivery. II. Release of albumin in vitro. J. Biomed. Mater. Res. Part B 72B, 299– 304.

Folic Acid-Conjugated Nanoparticles for Cancer Image (26) Xu, Y., Du, Y., Huang, R., and Gao, L. (2003) Preparation and modification of N-(2-hydroxyl) propyl-3-trimethyl ammonium chitosan chloride nanoparticle as a protein carrier. Biomaterials 24, 5015–5022. (27) Kiang, T., Wen, J., Lim, H. W., and Leong, K. W. (2004) The effect of the degree of chitosan deacetylation on the efficiency of gene transfection. Biomaterials 25, 5293–5301. (28) Gan, Q., Wang, T., Cochrane, C., and McCarron, P. (2005) Modulation of surface charge, particle size and morphological properties of chitosan-TPP nanoparticles intended for gene delivery. Colloid Surf., B 44, 65–73. (29) Choy, J. H., Jung, J. S., Oh, J. M., Park, M., Jeong, J., Kang, Y. K., and Han, O. J. (2004) Layered double hydroxide as an efficient drug reservoir for folate derivatives. Biomaterials 25, 3059–3064. (30) Akasaka, K., Gyimesi-Forra´s, K., La¨mmerhofer, M., Fujita, T., Watanabe, M., Harada, N., and Lindner, W. (2005) Investigations of molecular recognition aspects related to the enantiomer separation of 2-methoxy-2-(1-naphthyl)propionic acid using quinine carbamate as chiral selector: An NMR and FT-IR

Bioconjugate Chem., Vol. 21, No. 4, 2010 689 spectroscopic as well as X-ray crystallographic study. Chirality 17, 544–555. (31) Yang, J. M., Lin, H. T., and Yang, S. J. (2005) Evaluation of poly(N-isopropylacrylamide) modified hydroxyl-terminated polybutadiene based polyurethane membrane. J. Membr. Sci. 258, 97–105. (32) Wang, S., Luo, J., Lantrip, D. A., Waters, D. J., Mathias, C. J., Green, M. A., Fuchs, P. L., and Low, P. S. (1997) Design and synthesis of [111In]DTPA-folate for use as a tumor-targeted radiopharmaceutical. Bioconjugate Chem. 8, 673–679. (33) Dawson, R. M. C., Elliott, D. C., Elliott, W. H., and Jones, K. M. (1986) Data for Biochemical Research, pp 134-135, Oxford University Press, New York. (34) Qi, L., Xu, Z., Jiang, X., Li, Y., and Wang, M. (2005) Cytotoxic activities of chitosan nanoparticles and copper-loaded nanoparticles. Bioorg. Med. Chem. Lett. 15, 1397–1399. (35) Hu, Y., Jiang, X., Ding, Y., Ge, H., Yuan, Y., and Yang, C. (2002) Synthesis and characterization of chitosan-poly(acrylic acid) nanoparticles. Biomaterials 23, 3193–3201. BC9004798