Synthesis and Grafting of Thioctic Acid− PEG− Folate Conjugates onto

Bioconjugate Chem. 2006, 17, 603−609

603

Synthesis and Grafting of Thioctic Acid-PEG-Folate Conjugates onto Au Nanoparticles for Selective Targeting of Folate Receptor-Positive Tumor Cells Vivechana Dixit,† Jeroen Van den Bossche,‡ Debra M. Sherman,# David H. Thompson,*,‡ and Ronald P. Andres*,† School of Chemical Engineering, Department of Chemistry, Life Science Microscopy Facility, Purdue University, West Lafayette, Indiana 47907. Received November 22, 2005; Revised Manuscript Received February 16, 2006

This paper reports the creation of Au nanoparticles (AuNP) that are soluble in aqueous solution over a broad range of pH and ionic strength values and that are capable of selective uptake by folate receptor positive (FR+) cancer cells. A novel poly(ethylene glycol) (PEG) construct with thioctic acid and folic acid coupled on opposite ends of the polymer chain was synthesized for targeting the AuNP to FR+ tumor cells via receptor-mediated endocytosis. These folic acid-PEG-thioctic acid conjugates were grafted onto 10-nm-diameter Au particles in aqueous solution. The resulting folate-PEG-coated nanoparticles do not aggregate over a pH range of from 2 to 12 and at electrolyte concentrations of up to 0.5 M NaCl with particle concentrations as high as 1.5 × 1013 particles/mL. Transmission electron microscopy was used to document the performance of these coated nanoparticles in cell culture. Selective uptake of folate-PEG grafted AuNPs by KB cells, a FR+ cell line that overexpress the folate receptor, was observed. AuNP uptake was minimal in cells that (1) do not overexpress the folate receptor, (2) were exposed to AuNP lacking the folate-PEG conjugate, or (3) were co-incubated with free folic acid in large excess relative to the folate-PEG grafted AuNP. Understanding this process is an important step in the development of methods that use targeted metal nanoparticles for tumor imaging and ablation.

INTRODUCTION Current anticancer agents suffer from several limitationss they are typically highly toxic and are often nonspecific in that they accumulate within normal as well as tumor cells. In many cases, they also have limited bioavailability due to poor solubility and limited stability in vivo, leading to short halflives and rapid clearance from the body. An ideal solution to these problems would be to deliver biologically effective concentrations of the therapeutic agent(s) to the cancer cells with high specificity. One approach to this ultimate goal could be achieved by attaching the agents to the surface of a gold nanoparticle (AuNP)1 that is small enough to be transported through the circulatory system, is not rapidly cleared from the body, and is selectively targeted to cancer cells. AuNPs are the most chemically stable of metal nanoparticles. They have been studied as “soluble” gold since the fifth or fourth century B.C. both for their aesthetic and for their curative properties. In ancient times, they were used in the manufacture of ruby glass and in the treatment of a wide range of diseases (1). The use of AuNP in modern biology dates to their application as immunological staining agents for electron microscopy (2), wherein the AuNP are functionalized by simple covalent attachment or physisorption of an antibody or enzyme to the particle surface prior to exposure to tissue sections. Recently, AuNPs have become the subject of increased interest in the context of the emerging fields of nanoscience * Corresponding authors. (R.P.A.) E-mail: [email protected]. Voice: 765-494-4047. Fax: 765-494-0805. (D.H.T.) E-mail: [email protected]. Fax: 765-496-2592. † School of Chemical Engineering. ‡ Department of Chemistry. # Life Science Microscopy Facility. 1 Abbreviations: Poly(ethylene glycol) (PEG), folate-PEG1500thioctamide-modified gold nanoparticles (F-PEG1500-T:AuNP), folate receptor (FR), folate receptor positive (FR+), gold nanoparticles (AuNP), mPEG2000-thioctamide-modified gold nanoparticles (mPEG2000-T:AuNP).

and nanotechnology due to the ease of synthesizing monolayerprotected AuNP. These inorganic-organic constructs typically have a core-shell morphology consisting of a gold core, as small as 1-2 nm in diameter, surrounded by a covalently attached monolayer of organic molecules. The organic shell passivates the metallic core and imparts steric stabilization to the nanoparticles, permitting their manipulation as stable organic compounds. Recent reviews provide a compilation of the methods used to fabricate monolayer-protected AuNPs and highlight some of their important applications (3, 4). AuNPs must be water-soluble over a wide range of pH values and electrolyte concentrations to successfully target tumor cells. They must also resist recognition and subsequent removal by macrophages associated with the reticuloendothelial system. We anticipated that water solubility and biocompatibility could be achieved by coating the Au core with a biomedically accepted, water-soluble polymer such as poly(ethylene glycol) (PEG), since PEGylation of proteins (5) and liposomes (6-9) is known to increase their bioavailability, improve their pharmacokinetics, and decrease their immunogenicity. One method of covalently attaching PEG chains to AuNP is through the use of a thiol-terminated PEG. Foos et al. (10) used this approach to synthesize water-soluble AuNP. Tshikhudo et al. (11) also synthesized water-soluble AuNP by terminating the PEG chain with an alkanethiol. The use of a short alkane chain between the Au surface and the PEG chain is thought to produce a dense, ordered “brush” morphology on the Au surface. Tokumitsu et al. (12) observed a conformational transition when alkanethiol-terminated PEG chains are grafted to a flat Au substrate, consistent with the theoretical predictions of Szleifer (13). The PEG chains exist in a so-called “mushroom” regime at low coverages; however, as the coverage increases, they undergo a transition to a more extended conformation (i.e., the “brush” regime) away from the surface. From the standpoint of producing stable AuNPs for in vivo use, chemisorption of PEG is preferred over physisorption. It is not clear, however, whether a more ordered PEG layer anchored by alkanethiol adsorption

10.1021/bc050335b CCC: $33.50 © 2006 American Chemical Society Published on Web 05/03/2006

604 Bioconjugate Chem., Vol. 17, No. 3, 2006

Dixit et al.

Scheme 1. Schematic Representation of the PEG-Modified AuNP Used and Their Exposure to FR+ and FR- Cells.

on gold or a more amorphous monolayer provided by direct disulfide termination of the PEG chain is more desirable for imparting water solubility and biocompatibility to the Au nanoparticles. A second method for covalently attaching PEG chains to AuNP is by replacing the terminal thiol substituent with a disulfide moiety. Mangenay et al. (14) have shown that hydrophilic polymers that are end-capped with a disulfide can be efficiently grafted onto citrate-stabilized AuNPs in aqueous solution. This is the method adopted in the present study. Thioctic acid was chosen as the grafting agent due to its low cost, simple coupling chemistry, and ease of handling relative to thiol-based conjugates. Folic acid was chosen as the targeting agent. Folate is an essential precursor for the synthesis of nucleic acids and some amino acids, is not produced endogenously by mammalian cells, and requires internalization by cells via either receptor-mediated endocytosis or a carrier-based uptake mechanism (15-17). The folic acid receptor (FR)-based mechanism constitutes a particularly useful target for tumor-specific drug delivery because it is upregulated in many human cancers, including malignancies of the ovary, brain, kidney, breast, and lung. The FR density also appears to increase as the stage/grade of the cancer increases (18). An additional feature of FR that is particularly attractive for intravenous drug delivery strategies is the fact that FR generally becomes accessible to the vascular epithelium only after malignant transformation. This is because FR is selectively expressed on the apical membrane surface of normal polarized epithelial cells and is, therefore, protected from FR-directed therapeutics delivered in the plasma. Upon epithelial cell transformation, however, cell polarity is lost and FR becomes accessible to targeted drugs in blood circulation (19). The folate receptor has a high affinity for folic acid (Kd ∼ 0.1 nM), which results in high uptake by FR+ cells, even at low folate loadings on the therapeutic agent (17, 19). Because of these attractive characteristics, folate conjugation has become a widely used strategy for targeting liposomes (20-25), plasmid complexes (26-33), nanoparticles (34-36), polymer micelles (37), and other polymer constructs (38) for selective uptake by tumor cells. The 10 nm AuNPs chosen for this study were a compromise between providing a reasonable surface area for attachment of PEG chains while at the same time enabling rapid nanoparticle penetration into solid tumors based on the enhanced permeation and retention effect (e.g., EPR) (39, 40). The AuNPs were purchased as a citrate-stabilized aqueous colloid. Two PEG polymers were synthesized for grafting onto AuNPs, a mPEGthioctic acid conjugate (mPEG2000-T) for steric stabilization

and a folic acid-PEG-thioctic acid conjugate (F-PEG1500T) that would enable both steric stabilization and tumor cell targeting. After the AuNPs were modified with either of these conjugates, they were monitored for colloidal stability by absorption spectroscopy. KB cells were then treated with either the mPEG-thioctic acid-modified AuNP (mPEG2000-T: AuNP) or the folate-PEG1500-thioctic acid-modified AuNP (F-PEG1500-T:AuNP) constructs (Scheme 1) and their capacity for internalization observed using TEM.

EXPERIMENTAL PROCEDURES Materials and Methods. Size selected (10 nm) gold colloid was obtained from Ted Pella Inc. (Redding, CA). mPEG-NH2 was purchased from Nektar Therapeutics (Huntsville, AL); all other chemicals were purchased from Sigma-Aldrich (St. Louis, MO) and used as received. All reactions were performed under anhydrous conditions and under an Ar atmosphere unless otherwise mentioned. All solvents were reagent grade and were distilled from CaH2 under Ar before use. NMR spectra were recorded on Varian 200 or 300 MHz spectrometers using 1H solvent peaks as internal reference. Column chromatography was typically performed on 230-400 mesh silica gel using HPLC grade eluents. Thin-layer chromatography was performed using Baker-flex IB-F plates (J. T. Baker) and visualized using UV, I2 adsorption, KMnO4/heat, and/or H2SO4/heat. Synthesis of Functionalized PEG Polymers. Synthesis of mPEG-NH-thioctamide (mPEG2000-T). mPEG2000-NH2 (1 g, 0.50 mmol) and NHS-activated thioctic acid (287 mg, 1 mmol; prepared as described by Ponpipom et al. (41) were dissolved in 20 mL anhydrous CH2Cl2. Triethylamine (1 mL, 6.9 mmol) was then added dropwise, and the resulting solution was stirred for 3 days at reflux. After cooling of the reaction to 25 °C, the mixture was evaporated, redissolved in 1 mL of CH2Cl2, and precipitated from a 1:1 Et2O/hexane mixture at -20 °C. After filtration, the product was isolated via silica gel chromatography using 9:1 CHCl3/MeOH as eluent. The white powder was determined to be pure by NMR analysis (>95% purity) Yield: 63%, RF: 0.48 (9:1 CHCl3/MeOH, I2 stain), 1H NMR (CDCl3, 300 MHz): 6.32 (br s, 1H), 3.89 (t, 2H), 3.743.55 (m, 200H), 3.19 (m, 3H), 2.51 (m,2H), 2.37 (t, 2H), 2.23 (t, 2H), 1.96 (m, 3H), 1.7 (m, 8 H), 1.50 (m, 4H). Synthesis of NH2-PEG1500-NHBoc. NH2-PEG1500-NH2 (5 g, 3 mmol) in 50 mL anhydrous MeOH was added to a solution of Boc2O (0.7 g, 3 mmol). Et3N (1 mL, 6.9 mmol) was then added in dropwise fashion, and the resulting solution was stirred overnight at reflux. After cooling of the reaction to 25 °C, the

Grafting of Thioctic Acid-PEG-Folate Conjugates

mixture was evaporated, dissolved in 5 mL of CH2Cl2, and precipitated from a 1:1 Et2O/hexane mixture at -20 °C. After filtration, the product was isolated via silica gel chromatography using 89:10:1 CHCl3/MeOH/NH4OH as eluent. The off-white powder was determined to be pure by NMR (>95%). Yield: 27%, RF: 0.3 (89:10:1 CHCl3/MeOH/NH4OH, I2 stain), 1H NMR (CDCl3, 300 MHz): 5.03 (s, 3H), 3.75-3.52 (m, 150H), 3.21 (q, 2H), 2.1 (s, 3H), 1.76 (m, 2H), 1.44 (s, 9H). Synthesis of tBocNH-PEG1500-NH-thioctamide. H2NPEG1500-NHBoc (1 g, 0.62 mmol) and NHS-thioctate (400 mg, 1.2 mmol) were dissolved in 10 mL CH2Cl2, and 1 mL of Et3N was added dropwise. This mixture was stirred at reflux for 3 days, after which it was evaporated, redissolved in 1 mL of CH2Cl2, and precipitated from 1:1 Et2O/hexane at -20 °C. After filtration, the product was purified by silica gel chromatography using 95:5 CHCl3/MeOH as eluent. The off-white powder was pure by NMR (>95%) Yield: 61%, RF: 0.25 (95:5 CHCl3/ MeOH, I2 stain), 1H NMR (CDCl3, 300 MHz): 4.28 (t, 2H), 3.92-3.45 (m, 150H), 3.28 (m, 3H), 2.78 (s, 2H), 2.52 (m, 1H), 2.41 (t, 1H), 2.0-1.65 (m, 14H), 1.49 (s, 9H). Synthesis of NH2-PEG1500-NH-thioctamide. Boc-NHPEG1500-NH-thioctamide (610 mg) was dissolved in 10 mL of a 1:1 (vol/vol) solution of TFA in CH2Cl2. This mixture was stirred at 25 °C for 3 h after which the solvent was evaporated. The recovered solid was redissolved in 1 mL of CH2Cl2 and precipitated from a 1:1 Et2O/hexane mixture at -20 °C. The isolated solid was pure by TLC analysis. Yield: 98%, RF: 0.14 (89:10:1 CHCl3/MeOH/NH4OH, I2 stain), 1H NMR (CDCl3, 300 MHz): 4.24 (t,2H), 3.88-3.41 (m, 150H), 3.2 (q, 4H), 2.48 (m, 2H), 2.36 (t, 2H), 2.0 (m, 3H), 1.91 (m,1H), 1.67 (m, 3H), 1.5 (m,2H), 0.80 (s,1H). Synthesis of Folate-NH-PEG-NH-thioctamide (F-PEG1500-T). Thioctic acid-NH-PEG1500-NH2 (200 mg, 0.11 mmol) and folic acid (100 mg, 0.20 mmol) were dissolved in 5 mL of anhydrous DMSO. Pyridine (500 µL) was added dropwise to this mixture, followed by DCC (100 mg, 0.6 mmol). The solution was stirred overnight at 70 °C, then cooled to 25 °C, and 50 mL of Et2O was added. The precipitated product was redissolved and reprecipitated from 1:1 Et2O/hexane at -20 °C, and the solid was further purified via silica gel chromatography using a 20:79:1 to 50:49:1 MeOH/CH2Cl2/NH4OH step gradient elution. After purification, the isolated product was redissolved in CH2Cl2 and filtered through a PFTE filter to remove dissolved Si gel. The filtered solution was then evaporated and dissolved in 60:30:10 benzene/CH2Cl2/MeOH and lyophilized to give a yellow powder (>95% pure by NMR). Yield: 59%, RF: 0.44 (50:49:1 MeOH/CH2Cl2/NH4OH, I2/UV visualization), 1H NMR (DMSO, 300 MHz): 8.67 (d, 1H), 8.32 (br s, 1H), 7.92-7.85 (m, 2H), 7.72 (d, 1H), 7.64 (d, 1H), 7.43 (s, 1H), 7.20 (br s, 2H), 6.95 (q, 2H), 6.70 (t, 2H), 4.5 (t,3H), 4.18 (t, 2H), 3.80-3.11 (m, 150h), 2.70 (s, 1H), 2.49 (m, 2H), 2.37 (t, 2H), 2.29 (t, 1H), 2.20-1.80 (m, 4H), 1.64 (m, 5H), 1.44 (m, 2H), 1.30 (s, 1H). Grafting of Functionalized PEG Polymers onto AuNP. All glassware was cleaned with aqua regia and rinsed with DI water. Citrate-stabilized gold colloid solution (3 mL, used as supplied at ∼5.7 × 1012 particles/mL) was adjusted to pH 9-11 using 0.5 M NaOH. The AuNP suspension was then combined with 3 mL of 0.25 mM mPEG2000-T or F-PEG1500-T, and the solution was stirred at 25 °C for 16-18 h. The resulting suspension was split into two equal volumes and centrifuged twice at 13,000 rpm for 60 min at 8 °C with decantation of the supernatant and redispersion of the particles in DI water. The two rinsed samples of AuNP were then recombined in 1 mL of DI water. The UV-Vis spectra of the AuNP in DI water were measured using a Jasco V-550 spectrophotometer operating at a resolution

Bioconjugate Chem., Vol. 17, No. 3, 2006 605

of 1 nm with a 400 nm/min scan rate. The concentration of nanoparticles in the final suspension was in the range 1.0 × 1013 to 1.5 × 1013 particles/mL as determined by absorption spectroscopy. TEM Characterization of AuNP Samples. AuNP samples were prepared for TEM analysis by placing a drop of the colloidal solutions on Formvar+C-coated Cu TEM grids and air-drying the grid at 25 °C. The samples were viewed with a Philips CM-10 TEM using 80 kV accelerating voltage. Cell Culture and AuNP Treatment. KB cells, a human cancer cell line of nasopharyngeal origin that is known to overexpress folate receptors, were cultured in folate-deficient Dulbecco’s modified Eagle medium (FD-DMEM) supplemented with 10% heat inactivated fetal calf serum, 2 mM L-glutamine, and penicillin/streptomycin for 3 weeks to establish a folate deficiency. The folate-deficient KB cells were then subcultured in 75 cm2 flasks at 37 °C in a 5% CO2 atmosphere. Cells were used during the log phase of growth and discarded after the sixth passage. WI-38 cells, a normal embryonic human diploid cell line, were used as a control. These cells were cultured in FD-DMEM supplemented with 10% heat inactivated fetal calf serum, 2 mM L-glutamine, and penicillin/streptomycin for 3 weeks. The WI38 cells were subcultured in 75 cm2 flasks at 37 °C in a 5% CO2 atmosphere. Cells were used during the log phase of growth and discarded after the sixth passage. Both the KB and the WI-38 cells were plated at ∼250,000 cells/well in six-well plates and incubated for 48 h. At this point, 100 µL of AuNP solution (∼1.5 × 1013 particles/mL; sterilized by filtration through a 0.22 µm filter) was added to three of the wells. After a 1 h incubation period, 100 µL of nanoparticle solution was added to the remaining three wells, and all were incubated for an additional 1 h before rinsing with fresh media to remove free AuNP before fixation. TEM Imaging of AuNP Uptake by Cells. TEM imaging was used to evaluate the cellular uptake of AuNP. Gold intensification was used to enhance the observability of the 10 nm AuNPs followed by osmium fixation to delineate the cell membranes. Gold enhancement, unlike silver intensification, is not degraded by exposure to osmium tetroxide. The samples were processed in a PELCO model 3450 microwave equipped with variable wattage, a Coldspot temperature regulator, and vacuum chamber (Ted Pella, Inc., Redding, CA). Processing times and wattage (P1 ) 233 W, P2 ) 345 W) are noted for each step. Cells were fixed in 2% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) containing 2 mM MgCl2, 1 mM CaCl2, 40 mM NaCl (P1: 1 min hold > 40 s on > 3 min hold). Samples were then washed with buffer three times followed by a water wash (P1: 40 s each). Cells were scraped from the wells and pelleted by centrifugation. Samples were incubated for 2 min at 25 °C in GoldEnhance kit #2113 (Nanoprobes, Inc. Yaphank, NY) formulated according to kit directions to increase the size of the gold particles. Following a water wash (P1: 40 s), the pellets were post-fixed in reduced osmium (1% OsO4 + 1.5% K4Fe(CN)6) for two 40 s cycles at P1). The samples were then enrobed with 1.5% agarose (Sigma Type VII) and processed further as tissue blocks. Sample blocks were dehydrated through a graded ethanol series (P1: 30, 50, 70, 90 and 100%, two cycles each, 40 s per change), followed by propylene oxide (P1 × 40 s). They were then infiltrated in propylene oxide (LX-112 resin mix, 3:1, 1:1 for 30 min each followed by overnight in 1:3 mix plus accelerator on a rotator at 25 °C). Blocks were embedded in LX-112 resin mixture (Ladd Research Industries, Williston, VT) and polymerized at 60°C for 48 h. Ultrathin sections (∼100 nm) were picked up on Formvar+C coated 100 mesh Cu grids. The samples were viewed with a

606 Bioconjugate Chem., Vol. 17, No. 3, 2006

Dixit et al.

Figure 1. Synthetic schemes for mPEG2000-T and F-PEG1500-T.

Figure 2. Absorption spectra of colloidal suspensions of 10 nm diameter AuNPs: (1) citrate capped AuNPs; (2) T-PEG-F coated AuNPs.

Philips CM-10 TEM using 80 kV accelerating voltage. TEM negatives were scanned at 600 dpi and image processed using Adobe Photoshop.

RESULTS AND DISCUSSION The methods developed for both mPEG-thioctamide (mPEG2000-T) and folate-PEG-thioctamide (F-PEG1500-T) enable the synthesis of these constructs on the gram scale (Figure 1). Boc-protection of NH2-PEG1500-NH2 was found to be necessary since initial attempts at the direct synthesis of monosubstituted PEG-thioctamide resulted in inseparable mixtures. Boc protection results in a mixture of products that is easier to separate and produces acceptable yields of mono-Boc protected PEG. Coupling of this intermediate with NHSthioctate, followed by deprotection and coupling with folic acid gives the desired F-PEG1500-T in reasonable yield. Initial attempts at grafting the thioctic acid-functionalized polymers onto citrate-capped AuNP under a wide variety of conditions and concentrations in aqueous solution proved unsuccessful. However, it was found that efficient displacement of citrate by the thioctic acid functionalized polymers could be achieved by adjusting the solution pH to 9-11 with NaOH. UV-Vis absorption spectra for the 10-nm-diameter, citratestabilized AuNP, with and without a chemisorbed F-PEG1500-T coating in deionized H2O, are presented in Figure 2. These spectra have a strong absorption feature at ∼520 nm due to the characteristic surface plasmon resonance of AuNPs (3).

The height of this feature is proportional to the concentration of AuNPs in solution, while peak position and width are indicative of the size, size distribution, aggregation, and surface coating of the particles. Figure 2 shows that there is minimal broadening of the peak and only a small red-shift in the position of the maximum when the citrate-capped particles are coated with F-PEG1500-T. This is an indication that the AuNPs are not aggregated in solution. Adding HCl, NaOH, or NaCl to an aqueous suspension of the AuNP (∼1.5 × 1013 particles/mL) does not markedly affect the UV-Vis spectra (i.e., cause particle aggregation) over a pH range of 2-12 and an electrolyte concentration range of 0-0.5 M NaCl. Aqueous suspensions of AuNP within these ranges of pH and electrolyte concentration are observed to be stable for up to a year as determined by their absorption spectra. Figure 3 shows TEM micrographs illustrating the diameter and size distribution of the citrate-capped AuNPs and F-PEG1500-T:AuNPs. Although the nanoparticles aggregate due to surface tension as the water evaporates during sample preparation, the Au cores of the particles do not coalesce and fuse. Analysis of the mean diameter and size distribution of these two AuNP preparations shows that the Au cores of the F-PEG1500-T coated particles are similar to those of the citrate-capped particles. These data indicate that AuNP integrity is unaffected by F-PEG1500-T grafting. Of the two known mechanisms of folic acid transport into cells, only the receptor-mediated endocytosis pathway enables the uptake of folate conjugates by mammalian cells (42) (Scheme 2). A series of experiments were undertaken to determine whether the 10-nm-diameter F-PEG1500-T:AuNPs are selectively internalized by FR+ tumor cells. The nanoparticles were added to the cell cultures at a particle concentration of ∼5 × 1011 particles/mL or a loading of ∼6 × 106 AuNPs/ cell, and the cells were incubated for periods of 1 and 2 h. After the incubation period, the cells were rinsed with FD-DMEM, chemically fixed, washed, and processed as described in the Experimental Section prior to TEM analysis. KB cells were incubated with mPEG-T:AuNP as a control to establish that (1) nontargeted PEG-coated AuNPs are not internalized and that (2) AuNPs not taken up by the cells are eliminated in the washing process. Figure 4A shows a typical micrograph of a KB cell exposed to mPEG-T:AuNP. Only an occasional Au nanoparticle was observed inside the cells, which

Grafting of Thioctic Acid-PEG-Folate Conjugates

Bioconjugate Chem., Vol. 17, No. 3, 2006 607

Figure 3. TEM micrographs of 10 nm diameter Au nanoparticles: left micrograph citrate-capped AuNPs; right micrograph T-PEG-F-coated AuNPs.

Figure 4. TEM Micrographs of thin sections of cells that were incubated for 2 h with various AuNP constructs. Magnification of primary images is 6000×; the inset is shown at 28000×. (A) KB cells incubated with mPEG2000-T:AuNP. Occasionally, AuNP are found adjacent to the cell plasma membrane (arrow), but most are removed by washes during sample preparation. (B) KB cells incubated with F-PEG1500-T:AuNP. Region corresponding to the higher magnification inset is denoted by the arrow which shows significant uptake of nanoparticles throughout the cytoplasm. (C) KB cells incubated with F-PEG-T:AuNP in the presence of excess folic acid. (D) WI-38 cells incubated with F-PEG-T:AuNP.

was presumably internalized via nonspecific adsorption to the cell surface. The number of nanoparticles observed outside the cell membrane was also extremely small as indicated by the single AuNP imaged in Figure 4A. These data support the expectation that mPEG2000-T-coated particles are not internalized to a significant extent by KB cells. The absence of a specific binding event between the mPEG2000-T:AuNPs and the folate receptor on the surface of these cells prevents their uptake, enabling their facile removal from the cells by washing. Figure 4B shows a typical micrograph of a KB cell that was

incubated with F-PEG-T:AuNPs. After both 1 and 2 h of exposure, about 20% of the KB cells were found to contain a large number of the folate functionalized nanoparticles. The intracellular distribution of the nanoparticles was similar to that observed with folate-targeted liposomes (21), i.e., localization within lysosomes near the nucleus and within endosomes distributed throughout the endosomal pathway. It should be noted that gold intensification of AuNPs enclosed inside the cells resulted in a nonuniform particle size distribution and caused particles adjacent to each other to coalesce as seen in

608 Bioconjugate Chem., Vol. 17, No. 3, 2006 Scheme 2. Conceptual Diagram of Folate Receptor-Mediated Binding, Internalization, Endosomal Acidification, Intracellular Trafficking, and Endosomal Escape of F-PEG1500-T:AuNP by Folate Receptor-Positive Cells (43).

Dixit et al.

the particles. However, these cells took them up in large numbers. It is not clear whether this is related to the metabolic state of the cells during the initial exposure to the particles or to other physiological reasons unique to this type of cell. Additional experiments will be done to increase the uptake efficiency and further develop the responsiveness of the system.

ACKNOWLEDGMENT We would like to acknowledge the use of the Life Science Microscopy Facility and the expertise of Chia-Ping Huang in preparation of samples for TEM. Dr. Vivechana Dixit performed the AuNP coating, spectroscopy, and stability studies. Mr. Jeroen Van den Bossche performed the PEG syntheses and the cell culture experiments. This study was supported by the Showalter Trust.

LITERATURE CITED

the high magnification inset in Figure 4B. This phenomenon and the random nature of the sectioning process made accurate estimation of the AuNP density within the cells difficult. However, by taking multiple high magnification images, it was possible to estimate the particle density in the thin sections. Assuming these thin sections to be 100 nm thick, this analysis yields an estimated density of 0.25-1.0 × 1015 nanoparticles/ mL of cell volume, which is 2-3 orders of magnitude greater than the particle density in the growth medium. Variations in the incubation times (e.g., 1, 2, or 12 h) of the cells with F-PEG1500-T:AuNP did not appear to affect the number of particles internalized. Two additional control experiments lend further support for the uptake of F-PEG1500-T:AuNP via receptor-mediated endocytosis. When F-PEG1500-T/AuNP were incubated with KB cells in a medium that was augmented with free folate (i.e., 2 µM folic acid added to the FD-MEM during nanoparticle incubation with the cells), the KB cells did not internalize the nanoparticles (see Figure 4C). Furthermore, it was observed that WI-38 cells rarely internalized any of the particles when incubated with F-PEG1500-T:AuNP (see Figure 4D). Since the WI-38 cells lack the folate receptor, they are incapable of internalizing significant amounts of the folate-functionalized nanoparticles.

CONCLUSIONS Poly(ethylene glycol) conjugates bearing either methoxy and thioctic acid or folic acid and thioctic acid substituents at opposing ends of the polymer have been synthesized. These conjugates have been grafted onto 10-nm-diameter Au nanoparticles, via the thioctic acid moiety. The PEG-folate-coated nanoparticles are soluble in water over a pH range of 2-12 and electrolyte concentration range of 0-0.5 M and exhibit selective uptake by a folate receptor positive (FR+) cell line when the AuNP construct contains folic acid. These sterically stabilized nanoparticles may have potential uses in targeted drug delivery to FR+ cancer cells. It is interesting to note that only about 20% of the KB cancer cells exposed to the F-PEG1500-T/AuNP actively took up

(1) Antonii, F. Panacea Aurea-Auro Potabile, Bibiopolio Frobeniano, Hamburg, 1618. (2) Faulk, W. P., and Taylor, G. M. (1971) Immunocolloid method for electron microscope. Immunochemistry 8, 1081. (3) Daniel, M.-C., and Astruc, D. (2004) Gold nanoparticles: Assembly, supramolecular chemistry, quantum-size-related properties, and applications toward biology, catalysis, and nanotechnology. Chem. ReV. 104, 293-346. (4) Goodman, C. M., and Rotello, V. M. (2004) Biomacromolecule surface recognition using nanoparticles. Mini-ReV. Org. Chem. 1, 103-114. (5) Abuchowski, A., Vanes, T., Palczuk, N., and Davis, F. (1977) Alteration of immunological properties of bovine serum-albumin by covalent attachment of polyethylene-glycol. J. Biol. Chem. 252, 3578-3581. (6) Klibanov, A. L., Maruyama, K., Torchilin, V. P., Huang, L (1990) Amphipathic poly(ethylene glycol)s effectively prolong the circulation time of liposomes. FEBS Lett 268, 235-7. (7) Allen, T. M., Hansen, C., Martin, F., Redemann, C.; Yau-Young A. (1991) Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo. Biochim. Biophys. Acta 1066, 29-36. (8) Stealth Liposomes (Lasic, D., Martin, F., Eds.) CRC Press, Boca Raton, FL, 1995. (9) Gabizon, A., Shmeeda H., and Barenholz, Y. (2003) Pharmacokinetics of pegylated liposomal doxorubicin - Review of animal and human studies. Clin. Pharmacokin. 42, 419-436. (10) Foos, E. E., Snow, A. W., Twigg, M. E., and Ancona, M. G. (2002) Thiol-terminated di-, tri-, and tetraethylene oxide functionalized gold nanoparticles: A water-soluble, charge-neutral cluster. Chem. Mater. 14, 2401-2408. (11) Tshikhudo, T. R., Wang, Z., and Brust, M. (2004) Biocompatible gold nanoparticles. Mater. Sci. Technol. 20, 980-984. (12) Tokumitsu, S., Liebich, A., Herrwerth, S., Eck, W., Himmelhaus, M., and Grunze, M. (2002) Grafting of alkanethiol-terminated poly(ethylene glycol) on gold. Langmuir 18, 8862-8870. (13) Carignano, M. A., and Szleifer, I. (2003) Controlling surface interactions with grafted polymers. Interface Sci. 11, 187-197 and references therein. (14) Mangeney, C., Ferrage, F., Aujard, I., Marchi-Artzner, V., Jullien, L., Ouari, O., Rekai, E. D., Laschewsky, A., Vikholm, I., and Sadowski, J. W. (2002) Synthesis and properties of water-soluble gold colloids covalently derivatized with neutral polymer monolayers. J. Am. Chem. Soc. 124, 5811-5821. (15) Folate and Human DeVelopment (Massaro, E. J., Rogers, J. M., Eds.) p 376, Humana Press, NJ, 2002. (16) Kamen, B. A., and Smith, A. K. (2004) A review of folate receptor alpha cycling and 5-methyltetrahydrofolate accumulation with an emphasis on cell models in vivo. AdV. Drug DeliVery ReV. 56, 10851097. (17) Sabharanjak, S., and Mayor, S. (2004) Folate receptor endocytosis and trafficking. AdV. Drug DeliVery ReV. 56, 1099-1109. (18) Parker, N., Turk, M. J., Westrick, E., Lewis, J. D., Low, P. S., and Leamon, C. P. (2005) Folate receptor expression in carcinomas

Bioconjugate Chem., Vol. 17, No. 3, 2006 609

Grafting of Thioctic Acid-PEG-Folate Conjugates and normal tissues determined by a quantitative radioligand binding assay. Anal. Biochem. 338, 284-293. (19) Elnakat, H., and Ratnam, M. (2004) Distribution, functionality, and gene regulation of folate receptor isoforms: implications in targeted therapy. AdV. Drug DeliVery ReV. 56, 1067-1084. (20) Rui, Y., Wang, S., Low, P. S., and Thompson, D. H. (1998) Diplasmenylcholine-folate liposomes: an efficient vehicle for intracellular drug delivery. J. Am. Chem. Soc. 120, 11213-11218. (21) Qualls, M. M., and Thompson, D. H. (2001) Synergistic phototoxicity of chloroaluminum phthalocyanine tetrasulfonate delivered via acid-labile diplasmenylcholine-folate liposomes. Intl J. Cancer 93, 384-392. (22) Pan, X. Q., Wang, H., and Lee, R. J. (2002) Boron delivery to a murine lung carcinoma using folate receptor targeted liposomes. Anticancer Res. 22, 1629-1633. (23) Gabizon, A., Horowitz, A. T., Goren, D., Tzemach, D., Shmeeda, H. and Zalipsky, S. (2003) In vivo fate of folate-targeted poly(ethylene glycol) liposomes in tumor-bearing mice. Clin. Cancer Res. 9, 6551-6559. (24) Leamon, C. P., Cooper, S. R. and Hardee, G. E. (2003) Folateliposome-mediated antisense oligodeoxynucleotide targeting to cancer cells: evaluation in vitro and in vivo. Bioconjugate Chem. 14, 738-747. (25) Boomer, J. A., Qualls, M. M., Inerowicz, H. D., Haynes, R. H., Patri, G. V. S., Kim, J.-M. and Thompson, D. H., unpublished results. (26) Gottschalk, S., Cristiano, R. J., Smith, L. C., and Woo, S. L. C. (1994) Folate receptor mediated DNA delivery into tumor cells: potosomal disruption results in enhanced gene expression. Gene Ther. 1, 185-191. (27) Mislick, K. A., Baldeschwieler, J. D., Kayyem, J. F., and Meade, T. J. (1995) Transfection of folate -polylysine DNA complexes: evidence for lysosomal delivery. Bioconjugate Chem. 6, 512-515. (28) Li, S., and Huang, L. (1997) Targeted delivery of antisense oligodeoxynucleotides by LPDII. J. Liposome Res. 7, 63-75. (29) Benns, J. M., Maheshwari, A., Furgeson, D. Y., Mahato, R. I., and Kim, S. W. (2001) Folate-PEG-folate-graft-polyethyleniminebased gene delivery. J. Drug Targeting 9, 123-139. (30) Dauty, E., Remy, J. S., Zuber, G., and Behr, J.-P. (2002) Intracellular delivery of nanometric DNA particles via the folate receptor. Bioconjugate Chem. 13, 831-839. (31) Zhou, W., Yuan, X., Wilson, A., Yang, L., Mokotoff, M., Pitt, B., and Song, L. (2002) Efficient intracellular delivery of oligonucleotides formulated in folate receptor-targeted lipid vesicles. Bioconjate Chem. 13, 1220-1225.

(32) Quintana, A., Raczka, E., Piehler, L., Lee, I., Myc, A., Majoros, I., Patri, A. K., Thomas, T., Mule, J., 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, 13101316. (33) Ward, C. M., Pechar, M., Oupicky, D., Ulbrich, K., and Seymour, L. W. (2002) Modification of pLL/DNA complexes with a multivalent hydrophilic polymer permits folate-mediated targeting in vitro and prolonged plasma circulation in vivo. J. Gene Med. 4, 536547. (34) Turek, J. J., Leamon, C. P., and Low, P. S. (1991) Endocytosis of folate-protein conjugates: ultrastructural localization in KB cells. J. Cell. Sci. 106, 423-430. (35) Zhang, Y., Kohler, N., and Zhang, M. (2002) Surface modification of superparamagnetic magnetite nanoparticles and their intracellular uptake. Biomaterials 23, 1553-1561. (36) Stevens, P. J., Sekido, M., and Lee, R. J. (2004) A folate receptortargeted lipid nanoparticle formulation for a lipophilic paclitaxel prodrug. Pharm. Res. 21, 2153-2157. (37) Lee, E. S., Na, K., and Bae, Y. H. (2003) Polymeric micelle for tumor pH and folate-mediated targeting. J. Controlled Release 91, 103-113. (38) Aronov, O., Horowitz, A. T., Gabizon, A., and Gibson, D. (2003) Folate-targeted PEG as a potential carrier for Carboplatin analogues. Synthesis and in vitro studies. Bioconjugate Chem. 14, 563-574. (39) Jain, R. K. (2001) Delivery of molecular medicine to solid tumors: lessons from in vivo imaging of gene expression and function. J. Controlled Release 74, 7-25. (40) Maeda, H. (2001) The enhanced permeability and retention (EPR) effect in tumor vasculature: the key role of tumor-selective macromolecular drug targeting. AdV. Enzyme Regul. 41, 189-207. (41) Ponpipom, M. M., Bugianesi, R. L., and Blake, T. J. Inhibition of delayed hypersensitivity reactions by cinnamyl 1-thioglycosides (1987) J. Med. Chem. 30, 705-710. (42) Low, P. S., and Antony, A. C. (2004) Folate receptor-targeted drugs for cancer and inflammatory diseases. AdV. Drug DeliVery ReV. 56, 1055-1058. (43) Leamon, C. P., and Reddy, J. A. (2004) Folate-targeted chemotherapy. AdV. Drug DeliVery ReV. 56, 1127-1141. BC050335B