PEGylated and MMP-2 Specifically DePEGylated Quantum Dots

Dec 31, 2008 - Department of Biological Sciences, Korea Advanced Institute of Science and Technology, Daejeon, South Korea 305-701, and Department of ...
1 downloads 5 Views 1MB Size
Download PDF
Langmuir 2009, 25, 1645-1650

1645

PEGylated and MMP-2 Specifically DePEGylated Quantum Dots: Comparative Evaluation of Cellular Uptake Hyejung Mok,† Ki Hyun Bae,† Cheol-Hee Ahn,‡ and Tae Gwan Park*,† Department of Biological Sciences, Korea AdVanced Institute of Science and Technology, Daejeon, South Korea 305-701, and Department of Materials Science and Engineering, Seoul National UniVersity, Seoul, South Korea 151-744 ReceiVed October 24, 2008. ReVised Manuscript ReceiVed NoVember 20, 2008 Polyethylene glycol (PEG)-immobilized quantum dot (QD) nanoparticles, which could be specifically dePEGylated in response to the presence of the matrix metalloprotease-2 (MMP-2) enzyme, were prepared. The degree of PEGylation (MW 3400) on the surface of 12 nm streptavidin-coated QDs was stoichiometrically controlled by varying the feed amount of a biotin-substrate-PEG conjugate, where the substrate contained an MMP-2 cleavable peptide sequence. A biotin-cell penetrating peptide (CPP) conjugate was also immobilized onto the surface of the PEGylated QD surface to enhance the cellular uptake after dePEGylation. It was found that more than nine PEG chains per single QD were required to effectively inhibit the cellular uptake of modified QD particles down to around 20%, as compared with that of QD without PEG chains. However, the treatment of MMP-2 enzyme in the medium resulted in a substantial enhancement in the extent of QD cellular uptake by dePEGylation with concomitant resurfacing of sterically hidden CPP moieties. This study analyzed the effects of surface PEGylation density and MMP-2 specific dePEGylation on the cellular uptake of CPP-QD nanoparticles in a quantitative manner.

1. Introduction Quantum dot nanoparticles (QDs), which are semiconductor nanocrystals, have been intensively investigated as an imaging agent in vivo and in vitro because of their high quantum yield, high molar extinction coefficient, and high resistance to photobleaching.1-7 To facilitate targeted intracellular delivery of QDs for enhanced cellular imaging in vitro and in vivo, various cationic liposomes and polymers have been employed as QD carriers for facile cellular uptake, or surface-modified QDs with proteins, antibodies, and peptides were utilized for cell-specific delivery.8 However, QDs have inherent cytotoxic problems caused by leaching out of free heavy metal ions, free radical formation, and unfavorable interaction with intracellular components.9 The surfaces of QDs were routinely modified with polyethylene glycol (PEG) for improved biocompatibility as well as with various targeting ligands such as peptides and antibodies for enhanced delivery efficiency to specific cells and tissues.10-12 In particular, * Corresponding author. Tel: +82-42-350-2661. Fax: +82-42-350-2610. E-mail: [email protected]. † Korea Advanced Institute of Science and Technology. ‡ Seoul National University. (1) Akerman, M. E.; Chan, W. C.; Laakkonen, P.; Bhatia, S. N.; Ruoslahti, E. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 12617–12621. (2) Voura, E. B.; Jaiswal, J. K.; Mattoussi, H.; Simon, S. M. Nat. Med. 2004, 10, 993–998. (3) So, M. K.; Xu, C.; Loening, A. M.; Gambhir, S. S.; Rao, J. Nat. Biotechnol. 2006, 24, 339–343. (4) Maysinger, D.; Behrendt, M.; Lalancette-Hebert, M.; Kriz, J. Nano Lett. 2007, 7, 2513–2520. (5) Medintz, I. L.; Uyeda, H. T.; Goldman, E. R.; Mattoussi, H. Nat. Mater. 2005, 4, 435–446. (6) Wu, X.; Liu, H.; Liu, J.; Haley, K. N.; Treadway, J. A.; Larson, J. P.; Ge, N.; Peale, F.; Bruchez, M. P. Nat. Biotechnol. 2003, 21, 41–46. (7) Biju, V.; Muraleedharan, D.; Nakayama, K.; Shinohara, Y.; Itoh, T.; Baba, Y.; Ishikawa, M. Langmuir 2007, 23, 10254–10261. (8) Biju, V.; Itoh, T.; Anas, A.; Sujith, A.; Ishikawa, M. Anal. Bioanal. Chem. 2008, 391, 2469–2495. (9) Hardman, R. EnViron. Health Perspect. 2006, 114, 165–172. (10) Higuchi, Y.; Oka, M.; Kawakami, S.; Hashida, M. J. Controlled Release 2008, 125, 131–136. (11) Schroeder, J. E.; Shweky, I.; Shmeeda, H.; Banin, U.; Gabizon, A. J. Controlled Release 2007, 124, 28–34. (12) Duan, H.; Nie, S. J. Am. Chem. Soc. 2007, 129, 3333–3338.

surface-engineered QD nanoprobes, emitting stimuli-sensitive fluorescent signals in response to tumor physiological conditions such as subtle change in pH and the presence of extracellular enzymes, have been fabricated for the optical visualization of specific cancer cells and tissues.13,14 PEG has been commonly conjugated to various drugs, liposomes, and polymeric micelles and nanoparticles to prolong their blood circulation time by reducing the nonspecific adsorption of proteins via a steric stabilization effect;15-20 however, the introduction of PEG chains onto the surface of nanoparticulates severely reduces the extent of cellular uptake.21,22 It becomes very desirable to design stimuliresponsive PEG-decorated delivery systems that can be dePEGylated in response to specific physiological conditions available only at the site of tumor tissues for enhanced cellular uptake.23-25 Recently, a PEG derivative containing a matrix metalloprotease (MMP) cleavable peptide sequence was used for the surface modification of nanosized liposomes to improve their intracellular delivery efficiency via an enzymatic dePEGylation effect.24,25 MMPs are known to be abundantly present in the vicinity of tumor tissues because they play a key role in facilitating the (13) Zhang, Y.; So, M. K.; Rao, J. Nano Lett. 2006, 6, 1988–1992. (14) Mok, H.; Park, J. W.; Park, T. G. Bioconjugate Chem. 2008, 19, 797–801. (15) Kim, D.; Park, S.; Lee, J. H.; Jeong, Y. Y.; Jon, S. J. Am. Chem. Soc. 2007, 129, 7661–7665. (16) Gref, R.; Minamitake, Y.; Peracchia, M. T.; Trubetskoy, V.; Torchilin, V.; Langer, R. Science 1994, 263, 1600–1603. (17) Kohler, N.; Fryxell, G. E.; Zhang, M. J. Am. Chem. Soc. 2004, 126, 7206–7211. (18) Mok, H.; Park, J. W.; Park, T. G. Pharm. Res. 2007, 24, 2263–2269. (19) Kim, S. H.; Jeong, J. H.; Chun, K. W.; Park, T. G. Langmuir 2005, 21, 8852–7. (20) Jeong, J. H.; Kim, S. W.; Park, T. G. J. Controlled Release 2003, 93, 183–191. (21) Ryman-Rasmussen, J. P.; Riviere, J. E.; Monteiro-Riviere, N. A. J. InVest. Dermatol. 2007, 127, 143–153. (22) Chang, E.; Thekkek, N.; Yu, W. W.; Colvin, V. L.; Drezek, R. Small 2006, 2, 1412–1417. (23) Sawant, R. M.; Hurley, J. P.; Salmaso, S.; Kale, A.; Tolcheva, E.; Levchenko, T. S.; Torchilin, V. P. Bioconjugate Chem. 2006, 17, 943–949. (24) Hatakeyama, H.; Akita, H.; Kogure, K.; Oishi, M.; Nagasaki, Y.; Kihira, Y.; Ueno, M.; Kobayashi, H.; Kikuchi, H.; Harashima, H. Gene Ther. 2007, 14, 68–77. (25) Terada, T.; Iwai, M.; Kawakami, S.; Yamashita, F.; Hashida, M. J. Controlled Release 2006, 111, 333–342.

10.1021/la803542v CCC: $40.75  2009 American Chemical Society Published on Web 12/31/2008

1646 Langmuir, Vol. 25, No. 3, 2009

Mok et al.

Figure 1. Schematic presentation of MMP-2-enzyme-specific dePEGylation and intracellular QD delivery.

migration of tumor cells and inducing angiogenesis by degrading the extracellular matrix.26-28 Although a great number of studies have been reported on the basis of PEGylated nanoparticles for delivery and imaging applications, there are few studies quantitatively analyzing the effect of surface PEG density on the extent of cellular uptake. In this study, a 12 nm QD-streptavidin (QD-strep) was surface immobilized with biotin-cell penetrating peptide (CPP) and biotinsubstrate-PEG conjugates via biotin-streptavidin interactions, where the substrate shows a specific selectivity for MMP-2.29,30 The number of immobilized PEG chains was changed in a quantitative manner by controlling the feed molar ratio of biotinsubstrate-PEG conjugate to QD, and the amount of biotin-CPP conjugate was kept constant for all of the modified QDs. The dependence of QD cellular uptake on the number of surface PEG chains as well as the extent of MMP-2 specific dePEGylation with concomitant re-exposure of the CPP on the surface was thoroughly investigated at various MMP-2 concentrations.

2. Materials and Methods 2.1. Materials. Streptavidin-coated quantum dots (QD-strep, 12 nm) were purchased from Invitrogen (Carlsbad, CA). A C-terminal biotin-modified MMP-2 enzyme-specific peptide substrate (biotinsubstrate: NH2-GGGGPLGVRGGGGK-biotin) and an N-terminal biotin-modified cell-penetrating peptide (biotin-CPP: biotinYARVRRRGPRR) were synthesized from Peptron Inc. (Daejeon, South Korea). Recombinant human matrix metalloprotease-2 (MMP2) catalytic domains were obtained from Biomol (Butler Pike, PA). NH2-PEG-COOH (MW 3400) and 9-fluorenylmethoxycarbonylPEG-N-hydroxysuccinimide (Fmoc-PEG-NHS, MW 3400) were the products of Nektar (Huntsville, AL). Cy5-N-hydroxysuccinimide (26) Egeblad, M.; Werb, Z. Nat. ReV. Cancer 2002, 2, 161–174. (27) Turpeenniemi-Hujanen, T. Biochimie 2005, 87, 287–297. (28) Itoh, T.; Tanioka, M.; Yoshida, H.; Yoshioka, T.; Nishimoto, H.; Itohara, S. Cancer Res. 1998, 58, 1048–1051. (29) Choi, J. M.; Ahn, M. H.; Chae, W. J.; Jung, Y. G.; Park, J. C.; Song, H. M.; Kim, Y. E.; Shin, J. A.; Park, C. S.; Park, J. W.; Park, T. K.; Lee, J. H.; Seo, B. F.; Kim, K. D.; Kim, E. S.; Lee, D. H.; Lee, S. K. Nat. Med. 2006, 12, 574–579. (30) Lee, S.; Cha, E. J.; Park, K.; Lee, S. Y.; Hong, J. K.; Sun, I. C.; Kim, S. Y.; Choi, K.; Kwon, I. C.; Kim, K.; Ahn, C. H. Angew. Chem., Int. Ed. 2008, 47, 2804–2807.

Figure 2. Gel retardation assay of QDs modified with different numbers of biotin-substrate-PEG.

(Cy5-NHS, MW 792) was obtained from GE Healthcare (Buckinghamshire, U.K). 1-Ethyl-3-(dimethylamino)propyl carbodiimide hydrochloride (EDC), 9-fluorenylmethyl-N-succinimidyl carbonate (Fmoc-NHS ester, MW 337), and 4-hydroxy-azobenzene-20carboxylic acid (HABA)/avidin reagent were the products of Sigma (St. Louis, MO). N-Hydroxysulfosuccinimide (sulfo-NHS) was from Pierce (Rockford, IL). MDA-MB-435-GFP (human breast cancer cell) cells stably expressing GFP were donated by Samyang Corp. (Daejeon, South Korea). Dulbecco’s phosphate-buffered saline (PBS), fetal bovine serum (FBS), and DMEM were purchased from Gibco BRL (Grand Island, NY). All other chemicals and reagents were of analytical grade. 2.2. Preparation of the Fmoc- or Cy5-Terminated PEG-biotinMMP-2 Substrate Peptide Conjugate (Biotin-substrate-PEGFmoc or Cy5). To synthesize the biotin-substrate-PEG-Fmoc conjugate, biotin-substrate (1 mg, 0.74 µmol) in 100 µL of PBS was added to 400 µL of PBS solution (pH 7.5) containing Fmoc-PEGNHS (50.3 mg, 14.8 µmol). After 4 h of reaction, glycine (22.2 mg, 296 µmol) was added to block unreacted NHS groups in FmocPEG-NHS. After reacting overnight, the reactant was dialyzed against deionized water using a dialysis membrane (MWCO 5000). To synthesize the biotin-substrate-PEG-Cy5 conjugate, NH2-PEGCOOH (5.0 mg, 1.5 µmol) was dissolved in 200 µL of PBS solution (pH 7.5), and Cy5-NHS (0.48 mg, 0.6 µmol) in 45 µL of DMSO was added to the solution. After reacting overnight, Fmoc-NHS ester (1.5 mg, 1.5 µmol) was added, and the reaction mixture was stirred for an additional 5 h to block the unreacted amine group in NH2-PEG-COOH. The solution was dialyzed against deionized water using a dialysis membrane (MWCO 3500). The carboxyl group of Cy5-PEG-COOH (0.2 µmol) was activated with EDC (0.4 mg, 2

ComparatiVe EValuation of Cellular Uptake

Langmuir, Vol. 25, No. 3, 2009 1647

Figure 3. (A) Confocal microscopy images of GFP-overexpressing MDA-MB-435 cells transfected with QD containing different numbers of biotinsubstrate-PEG per QD (GFP expressing cell, green; QD, red; and intracellular GFP + QD, yellow). (Top row) QD-only image; (bottom row) GFP + QD merged image. (B) Relative percentages of QD uptake within GFP-overexpressing MDA-MB-435 cells.

Figure 4. Synthesis scheme of biotin-substrate-PEG-Cy5 and recovery of fluorescence intensity as a function of MMP-2 concentration.

µmol) and sulfo-NHS (0.4 mg, 2 µmol), and biotin-modified (MMP-2 specific) substrate peptide (2 mg, 1.3 µmol) in 200 µL of PBS solution was added to the solution. After reacting overnight at 4 °C, the mixture was dialyzed against deionized water using a dialysis membrane (MWCO 3500). The conjugation of substrate peptide to the PEG chain was confirmed by gel permeation chromatography (GPC) and matrix-assisted laser desorption/ionization time-of-flight mass spectromety (MALDI-TOF, Bruker, Germany). For GPC analysis, PBS solution was used as an eluant, and the flow rate was 1.0 mL/min. The eluate was detected by UV at 280 nm. The

concentration of biotin in the conjugate was determined by HABA/ avidin assay according to the manufacturer’s protocol. 2.3. HABA/Avidin and Gel Retardation Assay of PEGModified QDs. The number of biotin binding sites on a single QDstrep was measured by the HABA/avidin assay. Biotin-CPP conjugate (0.82 nmol) was incubated with QD-strep (16.0 pmol) in 50 µL of PBS solution for 30 min at room temperature. The reaction mixture was diluted with deionized water and centrifuged using a Microcon centrifugal filter unit (Millipore, MWCO 10 000) to separate biotinCPP bound QDs. After concentration, the amount of free biotin in

1648 Langmuir, Vol. 25, No. 3, 2009

Mok et al.

Figure 5. (A) Confocal microscopy images of GFP overexpressing MDA-MB-435 cells transfected with QD containing nine PEG chains in the absence and presence of 10 µg/mL MMP-2 (GFP expressing cell, green; QD, red; intracellular GFP + QD, yellow). (Top row) QD-only image; (bottom row) GFP + QD merged image. (B) Quantitative analysis of recovery of QD uptake by MMP-2 incubation.

the eluant solution was quantified using the HABA/avidin assay according to the manufacturer’s protocol. PEG immobilization on the QDs was confirmed by gel retardation assay. Different amounts of biotin-substrate-PEG conjugate (0, 1, 3, 9, 18, and 36 pmol) were added to the PBS solution containing QD-strep (1.0 pmol) and incubated for 30 min. The solution was loaded onto 0.5% agarose gel, and electrophoresis was run at 100 V for 40 min. The QDs were visualized by a UV illuminator. 2.4. Fluorescence Resonance Energy Transfer (FRET)-Based QD Quenching Assay. Cleavage of the substrate by the MMP-2 enzyme was confirmed by the FRET-based QD quenching assay. Biotin-substrate-PEG-Cy5 (0.34 nmol) was incubated with QDs (1.0 pmol) for 30 min, and various amounts of MMP-2 enzyme (0, 0.5, 1, and 5 µg/mL) were added to the reaction buffer (50 mM HEPES, 200 mM NaCl, 10 mM CaCl2, 1 mM ZnCl2, pH 7.4) and incubated for 2 h at 37 °C. The incubation time of PEGylated QDs with MMP-2 for 2 h was sufficient for the full cleavage of PEG chains immobilized onto the QD surface, as previously reported for MMP-2 substrate peptide-coated gold nanoparticles.30 After dilution of the resulting solutions, the solution was analyzed using a spectrofluorophotometer (SLM-AMINCO 8100, SLM Instrument Inc., Rochester, NY) with excitation and emission wavelengths of 350 and 400-800 nm, respectively. There was no significant effect of MMP-2 on the inherent QD emission up to a concentration of 5 µg/mL, when biotin-substratePEG-Cy5 was not bound on the surface. 2.5. In Vitro Cellular Uptake Experiment in MDA-MB-435 Cells. GFP-overexpressing MDA-MB-435 cells were maintained in DMEM supplemented with 10% FBS, 100 units/mL penicillin, and 100 µg/mL streptomycin at 37 °C in a humidified atmosphere of 5% CO2. Cells were plated on a four-well chamber slide at a density of 1 × 105 cells/well 24 h prior to transfection with biotin-substratePEG and biotin-CPP-coated QDs. QDs (2.0 pmol) were incubated with biotin-CPP (20.0 pmol) for 15 min to immobilize the CPP peptide onto the QD surface. Then, different amounts of biotinsubstrate-PEG (0, 2, 6, 18, and 36 pmol) were incubated with CPPimmobilized QDs (2.0 pmol) for 30 min. The QDs coated with

biotin-substrate-PEG and biotin-CPP were treated with MDA-MB435 cells. After 4 h of incubation, cells were washed three times with cold PBS solution and fixed with 1% (w/v) paraformaldehyde solution for 30 min at 4 °C. The uptake of QD within GFP overexpressing MDA-MB-435 cells was visualized by confocal microscopy (Carl Zeiss, Germany). To determine the number of QDs within cells, fluorophotometric analysis was performed. CPPmodified QDs (4.0 pmol, the molar ratio of feed biotin-CPP to QDstrep was 10) and PEGylated CPP-modified QDs (the molar ratios of PEG to QD-strep were 0, 1, 3, 9, and 18) were transfected to cells in a 12 well plate (2 × 105 cells/well). After 4 h of incubation, QD-treated cells were washed with PBS and treated with a cell lysis solution (1% Triton X-100 in PBS) and centrifuged at 5000 rpm for 5 min to remove the cell debris. The relative amounts of QD and GFP proteins within cells were determined by analyzing the supernatant using a spectrofluorophotometer (SLM-AMINCO 8100, SLM Instrument Inc., Rochester, NY) with an excitation/emission wavelengths of 350/605 nm for QD particles and 488/509 nm for GFP protein, respectively. To determine the cellular uptake of dePEGylated QDs after MMP-2 enzyme treatment, QDs coated with biotin-substrate-PEG and biotinCPP were added to HEPES buffer solutions (50 mM HEPES, 200 mM NaCl, 10 mM CaCl2, 1 mM ZnCl2, pH 7.4) containing various concentrations of MMP-2 enzyme (0, 5, 10 µg/mL) and incubated for 2 h at 37 °C. MMP-2-enzyme-treated QDs were transfected to GFP-overexpressing MDA-MB-435 cells for 4 h.

3. Results and Discussion PEGylated and CPP-immobilized QDs, which could be dePEGylated by the MMP-2 enzyme, were prepared via streptavidin-biotin interactions between QD-strep and biotinsubstrate-PEG conjugates and biotin-CPP peptide. The immobilization of multiple PEG (MW 3400) chains on the surfaces of QDs was likely to sterically hide CPP moieties, suppressing the cellular uptake of QD. Upon exposure to the MMP-2 enzyme,

ComparatiVe EValuation of Cellular Uptake

however, the surface-covering PEG chains were cleaved with resurfacing CPP outside, leading to enhanced intracellular delivery of dePEGylated QDs via CPP-cell membrane interactions. Schematic illustrations of MMP-2 activated QDs modified with biotin-CPP and biotin-substrate-PEG and their cellular uptake behaviors are shown in Figure 1. The synthesis of the biotin-substrate-PEG conjugate was confirmed by GPC and MALDI-TOF analysis as shown in Figure S1. The GPC result suggests that almost all of the biotinderivatized substrate peptide was conjugated to PEG. Although unreacted and amine-group-blocked PEG species remained in solution, they would have no effect on the cellular uptake of PEGylated and CPP-immobilized QDs. The number of biotin binding sites per single QD-strep was determined to prepare modified QD coated with a precisely controlled number of biotinsubstrate-PEG conjugates along with a fixed number of cellpenetrating peptides on the surface. The HABA/avidin assay revealed that as-received QD-strep with a size of 12 nm had 28.6 ( 4.0 biotin binding sites (considering 4 biotin binding sites/ streptavidin, ∼7 streptavidin molecules on the surface), coinciding with the previous result that a QD-strep particle ranging from 10 to 30 nm had 20-40 biotin binding sites.13 It was assumed that all biotin-substrate-PEG conjugates in the feed were stoichiometrically bound to QD-strep, considering a strong binding affinity between biotin and streptavidin with a dissociation constant, KD, of 4 × 10-14 M.31 Thus, it was postulated that the numbers of PEG chains and CPP moieties immobilized on the QD surface were directly proportional to the feed molar ratios of the two biotin-derivatized conjugates. Assuming 28 biotin binding sites available per QD-strep, the number of immobilized biotin-CPP conjugates per QD-strep were prefixed at 10, and that of immobilized biotin-substrate-PEG conjugates was varied from 0 to 18 by controlling its feed molar ratio. As shown in Figure 2, the gel electrophoresis assay displayed retarded migration of QD modified with an increasing feed ratio of biotinsubstrate-PEG of up to 3, indirectly supporting that different feed ratios of biotin-derivatized conjugates successfully produced PEGylated QD with a controlled number of PEG chains seeded on the surface. However, the gel migration profiles of QDs modified with feed ratios of biotin-substrate-PEG from 9 to 36 were similar, probably because of the fact that excessively PEGylated QD beyond a certain surface density does not proportionally increase their hydrodynamic sizes. Cellular uptake behaviors of QD immobilized with a fixed number of biotin-CPP and different numbers of biotin-substratePEG were investigated using GFP-overexpressing MDA-MB435 cells. It was previously reported that unmodified QDs showed severely limited cellular uptake without conjugating cell penetrating peptides on the surface.29 The confocal microscopy images in Figure 3A show that QD-strep did not show any measurable red fluorescent signals, whereas CPP-QD-strep alone exhibited strong, well-dispersed red fluorescent signals over the whole intracellular region. As shown in Figure 3A, the red fluorescence intensity for MDA-MB-435 cells gradually decreased with increasing number of biotin-substrate-PEG per CPPQD-strep, suggesting that the surface immobilization of PEG chains on the QD surface reduced the extent of intracellular uptake by sterically preventing the exposure of cell-penetrating peptides to the cell membrane. The extent of QD uptake within cells was quantitatively evaluated by measuring the intracellular red fluorescence intensity after cell lysis as shown in Figure 3B. Extents of cellular uptake of CPP-QD with 1, 3, 9, and 18 biotin(31) Avrantinis, S. K.; Stafford, R. L.; Tian, X.; Weiss, G. A. ChemBioChem 2002, 3, 1229–1234.

Langmuir, Vol. 25, No. 3, 2009 1649

substrate-PEG conjugates per QD were 80.2 ( 14.8, 75.2 ( 18.9, 25.1 ( 7.4, and 20.3 ( 9.8%, respectively, when CPP-QD alone was used as a control (100% cellular uptake). It is noticeable that >9 PEG chains/QD was required to effectively inhibit the cellular uptake of CPP-QD down to around 25%. Considering that the radius of gyration of PEG (MW 3500) is 2.3 nm,32 it can be roughly estimated that the covered surface areas of 12 nm QDs tethered by 1, 3, 9, and 18 PEG chains were about 3.7, 11.0, 33.1, and 66.1%, respectively. Even though the number of PEG chains was increased from 9 to 18 with concomitant doubling of the covering surface area, there was no significant reduction in cellular uptake. This implies that approximately one-third coverage of the surface area of a spherical QD by highly mobile and hydrated PEG chains might be sufficient to suppress the extent of cellular uptake down to 25%, primarily because of sterically limiting the surface exposure of CPP moieties to the cellular membrane. It appears that extensively PEGylated QDs could be taken up by cells through a nonspecific fluid-phase pinocytic pathway, resulting in unavoidable ∼20% cellular uptake. On the basis of the above relationship between the PEG surfacecovering area and cellular uptake for 12 nm QDs, it can be roughly deduced that any nanoparticle ∼100 nm in size would require about 1250 PEG (MW 3400) chains on the basis of the assumption of 66% surface coverage for the critical minimization of cellular recognition and internalization. This is in good agreement with the previously reported theoretical value: ∼100 nm liposomes would need 1560 PEG (MW 2000) chains on the surface for full surface coverage and long circulation. 33 To confirm the cleavage of PEG chains immobilized onto the surface of QDs by the MMP-2 enzyme, the biotin-substratePEG-Cy5 conjugate was immobilized up to a saturated level of 28 PEG chains/QD-strep. The cleavage of Cy-5-labeled PEG chains was monitored on the basis of the fluorescence resonance energy transfer (FRET)-based QD quenching effect at various enzyme concentrations as shown in Figure 4.34-36 The QDstrep/biotin-substrate-PEG-Cy5 particles were incubated with the MMP-2 enzyme for 2 h before fluorescence measurement to allow a sufficient reaction time to cleave accessible peptide substrate sequences located on the bottom surface of each PEGylated QD. The relative fluorescence intensity of QD emission at 605 nm was effectively quenched from 8570 to 1070 when biotin-substrate-PEG-Cy5 was anchored on the QD-strep surface. After enzyme treatment, no significant enhancement of Cy5 emission at 664 nm was detected from cleaved Cy5-PEG fragments, although a markedly increased intensity of QD emission was observed. This was probably due to the fact that the amount of released Cy5 dyes in the solution was not sufficient for detection, whereas they efficiently quenched QD emission when bound to the surface. As the concentration of the MMP-2 enzyme increased by 0.5, 1, and 5 µg/mL, the fluorescence recovery intensity was gradually enhanced to 2360, 4720, and 7620, respectively. A recovery percentage of 74.9 ( 19.8% was attained at an MMP-2 concentration of 5 µg/mL. In contrast, no detectable increase in the QD emission signal was observed upon exposure of luciferase enzyme, confirming that the recovery of QD emission signals was caused by the cleavage of the MMP(32) Tanaka, S.; Ataka, M.; Onuma, K.; Kubota, T. Biophys. J. 2003, 84, 3299–3306. (33) Allen, C.; Dos Santos, N.; Gallagher, R.; Chiu, G. N.; Shu, Y.; Li, W. M.; Johnstone, S. A.; Janoff, A. S.; Mayer, L. D.; Webb, M. S.; Bally, M. B. Biosci. Rep. 2002, 22, 225–250. (34) Oh, E.; Lee, D.; Kim, Y. P.; Cha, S. Y.; Oh, D. B.; Kang, H. A.; Kim, J.; Kim, H. S. Angew. Chem., Int. Ed. 2006, 45, 7959–7963. (35) Bagalkot, V.; Zhang, L.; Levy-Nissenbaum, E.; Jon, S.; Kantoff, P. W.; Langer, R.; Farokhzad, O. C. Nano Lett. 2007, 7, 3065–3070. (36) Zhou, D.; Ying, L.; Hong, X.; Hall, E. A.; Abell, C.; Klenerman, D. Langmuir 2008, 24, 1659–1664.

1650 Langmuir, Vol. 25, No. 3, 2009

2-enzyme-specific substrate. The results suggest that MMP-2specific dePEGylation could indeed occur by MMP-2 enzymes accessible to its peptide substrate sequence positioned near the surface of PEGylated QDs. Even though PEG chains were fully immobilized on the surface of QD-strep at all 28 biotin binding sites, the MMP-2 catalytic domain (MW 40 kDa) could penetrate into the bottom surface of highly PEGyated QDs with a prolonged incubation period of 2 h. The effect of MMP-2-induced dePEGylation on the cellular uptake of QD was investigated using MDA-MB-435 cells. Figure 5 shows that PEGylated and CPP-modified QDs with nine PEG chains on the surface significantly recovered intracellular red fluorescence intensity upon adding the MMP-2 enzyme at a concentration of 10 µg/mL. The confocal images revealed that dePEGylated QDs greatly enhanced the extent of intracellular uptake by exposing CPP moieties to the cell membrane. The quantitative increase in the uptake for QD with 18 PEG chains on the surface was substantial from 21.9 ( 13.3 to 79.1 ( 6.5% with increasing MMP-2 concentration from 0 to 10 µg/mL. These results suggest that the modified QD with biotin-CPP and biotin-

Mok et al.

substrate-PEG can be selectively delivered to cancer cells by specific substrate recognition by MMP-2 because MMP-2 is an overexpressed endopeptidase secreted from cancer cells. In conclusion, the introduction of nine or more PEG chains on the surfaces of 12 nm QDs effectively inhibited cellular uptake. MMP-2-enzyme-specific dePEGylation stimulated the cellular uptake of QD by exposing cell-penetrating peptides to the cell membrane after the cleavage of the MMP-2-specific substrate in the immobilized PEG chains. The modified QD with biotinCPP and biotin-substrate-PEG could be applied as a cancercell-specific imaging agent. Acknowledgment. This study was supported by grants from the National Research Laboratory project from the Korea Science and Engineering Foundation, Republic of Korea. Supporting Information Available: Gel permeation chromatography (GPC) data and MALDI-TOF analysis of biotin-substratePEG conjugate.This material is available free of charge via the Internet at http://pubs.acs.org. LA803542V