Affibody-Attached Hyperbranched Conjugated Polyelectrolyte for

ARTICLE pubs.acs.org/Biomac

Affibody-Attached Hyperbranched Conjugated Polyelectrolyte for Targeted Fluorescence Imaging of HER2-Positive Cancer Cell Kan-Yi Pu,† Jianbing Shi,‡ Liping Cai,† Kai Li,† and Bin Liu*,† †

Department of Chemical and Biomolecular Engineering, 4 Engineering Drive 4, National University of Singapore, Singapore 117576, Singapore ‡ College of Materials Science and Engineering, Beijing Institute of Technology, Beijing 100081, China

bS Supporting Information ABSTRACT: A hyperbranched conjugated polyelectrolyte (HCPE) with a core
shell structure is designed and synthesized via alkyne polycyclotrimerization and click chemistry. The HCPE has an emission maximum at 565 nm with a quantum yield of 12% and a large Stokes shift of 143 nm in water. By virtue of its poly(ethylene glycol) shell, this polymer naturally forms spherical nanoparticles that minimize nonspecific interaction with biomolecules in aqueous solution, consequently allowing for efficient bioconjugation with anti-HER2 affibody via carbodiimide-activated coupling reaction. The resulting affibody-attached HCPE can be utilized as a reliable fluorescent probe for targeted cellular imaging of HER2-overexpressed cancer cells such as SKBR-3. Considering its low cytotoxicity and good photostability, the HCPE nanoprobe holds great promise in practical imaging tasks. This study also provides a molecular engineering strategy to overcome the intrinsic limitations of traditional fluorescent polymers (e.g., chromophore-tethered polymers and linear conjugated polyelectrolytes) for bioconjugation and applications.

’ INTRODUCTION Fluorescent cellular probes with high selectivity and sensitivity are of central importance not only for fundamental biology and pathophysiology but also for clinical diagnosis and therapy.1 Various materials including organic fluorophores, fluorescent proteins, and semiconductor quantum dots (QDs) have been intensively applied for cellular imaging.2 However, each of these materials has its own disadvantages, for example, low photobleaching thresholds for organic and genetic fluorophores and severe cytotoxicity for QDs under oxidative environment,3 which stimulates the unremitting pursuit of new fluorescent probes. In this regard, fluorescent conjugated polyelectrolytes (CPEs) with π-electron delocalized backbones and water-soluble side chains have provided a versatile platform for biological sensing.4
9 In particular, primary investigation using CPEs as simple nonspecific stains have been conducted, which reveals that they possess good photostability, sufficient brightness, and low cytotoxicity.10
14 Despite their great potential in bioimaging, CPEs with celltargeting capability have been rarely developed to date. A general method to construct cellular probes is to functionalize fluorescent materials with specific recognition elements, such as antibodies that have high affinity to specific receptors overexpressed in target cells.1 However, implementation of this strategy on CPEs appears to be difficult because the hydrophobic aromatic backbone and charged side chains of CPEs can inevitably induce strong nonspecific interactions with biomolecules,15,16 consequently r 2011 American Chemical Society

impeding bioconjugation reaction. In addition, these detrimental interactions would depress the selectivity of probes,17,18 potentially giving rise to false-positive staining. As a result of these structure-related limitations, it is of great challenge to develop fluorescent cellular probes based on CPE
antibody conjugates. Our recent study illuminated that attachment of poly(ethylene glycol) (PEG) as the side chain to a linear CPE can effectively minimize its nonspecific interactions with biomolecules.19 This improvement takes advantage of the polymer self-assembly to form core
shell nanoparticles in aqueous media, within which the PEG side chains serve as a protective shell to shield the CPE backbone from biomolecules. However, because self-assembly-induced nanoparticle formation is a thermodynamically controlled process, the dimension and morphology of the obtained nanoparticles are strongly influenced by solution conditions, such as solute concentration, ionic strength, and temperature.20 In contrast with linear polymers, hyperbranched polymers are intrinsic nano-objects,21 and their dimension is mainly determined by segmental flexibility and generation number.22 As a result, adoption of hyperbranched architecture would be an effective strategy to develop CPE-based fluorescent nanoparticles with well-controlled size and morphology for efficient bioconjugation and specific cellular imaging. Moreover, Received: April 24, 2011 Revised: June 4, 2011 Published: June 28, 2011 2966

dx.doi.org/10.1021/bm200563a | Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules superior to dye-incorporated dendrimer probes that share fast photobleaching of organic dyes,23 CPEs with hyperbranched architecture should inherit intrinsic fluorescence and good photostability from their linear counterparts. In this contribution, we report an affibody-attached hyperbranched CPE (HCPE) for targeted fluorescence imaging of human epidermal growth factor receptor 2 (HER2) positive cancer cells. Early stage detection of HER2 is of clinical significance in personalizing cancer treatment because the HER2 expression level is closely associated with tumor behavior and clinical outcome.24 Anti-HER2 affibody instead of commonly used HER2-specific antibody (herceptin) is chosen as the recognition element, considering its better targeting capability and smaller size (∼7 kDa) as compared with that for herceptin (∼150 KDa).25 The HCPE is designed not only to have a unique core
shell molecular architecture to reduce nonspecific interaction with biomolecules but also to emit fluorescence at the long wavelength region from 500 to 700 nm with a large Stokes shift of 143 nm to minimize interference from cellular autofluorescence. This Article is organized as follows. The design, synthesis, and characterization of the HCPE are first described, which is followed by the study of its size and optical properties in aqueous solution. After conjugation between the HCPE and anti-HER2 affibody, application of the obtained nanoprobe in targeted cellular imaging is demonstrated and discussed.

’ EXPERIMENTAL SECTION Characterization. Nuclear magnetic resonance (NMR) spectra were collected on Bruker Avance 500 (DRX 500, 500 MHz). Matrixassisted laser desorption/ionization time-of-flight (MALDI-TOF) was performed by using 2,5-dihydroxybenzoic acid (DHB) as the matrix under the reflector mode for data acquisition. Gel permeation chromatography (GPC) analysis was conducted with a Waters 2690 liquid chromatography system equipped with Waters 996 photodiode detector and Phenogel GPC columns, using polystyrenes as the standard and tetrahydrofuran (THF) as the eluent at a flow rate of 1.0 mL/min at 35 °C. UV
vis spectra were recorded on a Shimadzu UV-1700 spectrometer. Photoluminescence (PL) measurements were carried out on a Perkin-Elmer LS-55 equipped with a xenon lamp excitation source and a Hamamatsu (Japan) 928 PMT using 90° angle detection for solution samples. Photographs of the polymer solutions were taken using a Canon EOS 500D digital camera under a hand-held UV-lamp with λmax = 365 nm. Fisher brand regenerated cellulose dialysis tubes with 3.5 and 6.5 kDa molecular weight cutoff were used for polymer and polymer
affibody dialysis, respectively. Freeze-drying was performed using Martin Christ Model Alpha 1-2/LD. Dynamic light scattering (DLS) was performed on Malvern Zetasizer Nano Series at 25 °C, and the data were analyzed by Dispersion Technology Software 5.0. The polymer solution for DLS measurement was freshly prepared in Milli-Q water. High-resolution transmission electron microscopy (HR-TEM) images were obtained from a JEOL JEM-2010 transmission electron microscope with an accelerating voltage of 200 kV. The samples were prepared by drop-coating P2 aqueous solution (0.1 μg/mL) onto copper grid, followed by freeze-drying. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was carried out for the crude conjugation product at 100 V using 15% acrylamide gel prepared by the standard method. All UV and PL spectra were collected at 24 ( 1 °C. Milli-Q water (18.2 MΩ) was used for all experiments. Materials. Anti-HER2 affibody was purchased from Affibody AB. Fetal bovine serum (FBS) was purchased from Gibco (Lige Technologies, Ag, Switzerland). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) and penicillin
streptomycin solution were

ARTICLE

purchased from Sigma-Aldrich. O-(2-Aminoethyl)-O0 -(2-azidoethyl) nonaethylene glycol (NH2
PEG-N3) with the purity of 95% was also purchased from Sigama-Aldrich. Dulbecco’s modified essential medium (DMEM) was a commercial product of National University Medical Institutes (Singapore). Phosphate-buffered saline (PBS) buffer (10, pH 7.4) (ultrapure grade) and tris-borate buffer (10, pH 8.3) are a commercial product of first BASE Singapore. Milli-Q water (18.2 MΩ) was used to prepare diluted buffer solutions. NMR solvents, CDCl3 (99%), and CD3OD (99.5%) were purchased from Cambridge Isotope Laboratories. Other chemicals were ordered from SigmaAldrich unless otherwise stated. 2-(9,90 -Bis(6-bromohexyl)fluorenyl)4,4,5,5-tetramethyl-1,3,2-dioxaborolane (1) was synthesized according to our previous report.26,27 Cell Cultures. KBR-3 breast cancer cells, MCF-7 breast cancer cells, and NIH-3T3 fibroblast cells were cultured in DMEM containing 10% fetal bovine serum and 1% penicillin streptomycin at 37 °C in a humidified environment containing 5% CO2. Before experiment, the cells were precultured until confluence was reached. Confocal Imaging. For imaging after living cell incubation, cells were cultured in the chambers (LAB-TEK, Chambered Coverglass System) at 37 °C. After 80% confluence, the medium was removed and the adherent cells were washed twice with 1 PBS buffer. Affibodyattached P2 or P2 (90 μL, 0.5 μg/mL) solution was then added to the chamber. After incubation for 2 h, cells were washed three times with 1 PBS buffer and then fixed by 75% DMSO for 20 min and further washed twice with 1 PBS buffer. The nuclei were stained with PI for 40 min. The cell monolayer was washed twice with 1 PBS buffer and imaged by a confocal laser scanning microscope (CLSM, Zeiss LSM 410, Jena, Germany) with imaging software (Fluoview FV1000). For fixed cell imaging, cells were cultured in the chambers (LAB-TEK, Chambered Coverglass System) at 37 °C. After 80% confluence, the adherent cells were washed twice with 1 PBS buffer and then fixed by 75% DMSO for 20 min and were further washed twice with 1 PBS buffer and blocked for 30 min in 0.6 mL of 1 PBS containing BSA (2% BSA, 5 mM MgCl2). Affibody-attached P2 (90 μL, 0.5 μg/mL) was then added to each well, and the cells were further incubated for 20 min at 4 °C. The nuclei were then stained with propidium iodide (PI) for 20 min. After being washed with 1 PBS buffer (5 mM MgCl2), the cells were imaged. The average fluorescence intensities were calculated with Image-Pro. Cytotoxicity Test. MTT assays were performed to assess the metabolic activity of NIH-3T3 fibroblast. NIH-3T3 cells were seeded in 96-well plates (Costar, Chicago, IL) at an intensity of 2 104 cells/mL. After 48 h of incubation, the medium was replaced by P2 solution at the concentration of 0.01, 0.02, or 0.06 mg/mL, and the cells were then incubated for 8 and 24 h, respectively. After the designated time intervals, the wells were washed twice with 1 PBS buffer, and freshly prepared MTT (100 μL, 0.5 mg/mL) solution in culture medium was added to each well. The MTT medium solution was carefully removed after 3 h of incubation in the incubator. Isopropanol (100 μL) was then added to each well and gently shaken for 10 min at room temperature to dissolve all of the precipitate formed. The absorbance of MTT at 570 nm was monitored by the microplate reader (Genios Tecan). Cell viability was expressed by the ratio of the absorbance of the cells incubated with P2 solution to that of the cells incubated with culture medium only.

Synthesis of 4-(9,90 -Bis(6-bromohexyl)fluorenyl)-7-bromobenzothiadiazole (2). 2-(9,90 -Bis(6-bromohexyl)fluorenyl)-4,4,

5,5-tetramethyl-1,3,2-dioxaborolane (1) (2.84 g, 4.60 mmol), 4,7dibromobenzothiadiazole (2.16 g, 7.36 mmol), Pd(PPh3)4 (53 mg, 0.046 mmol), and potassium carbonate (4.43 g, 32.0 mmol) were placed in a 100 mL round-bottomed flask. A mixture of water (12 mL) and toluene (30 mL) was added to the flask, and the reaction vessel was degassed. The mixture was vigorously stirred at 90 °C for 2 days. After it was cooled to room temperature, dichloromethane was added to the 2967

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules reaction mixture. The organic portion was separated and washed with brine before drying over anhydrous MgSO4. The solvent was evaporated off, and the solid residues were purified by column chromatography on silica gel using dichloromethane/hexane (1:5) as eluent to afford 2 as grassy yellow liquid (2 g, 62%). 1H NMR (500 MHz, CDCl3, δ): 8.0
7.87 (m, 3 H), 7.85 (d, 1 H, J = 7.84 Hz), 7.77 (d, 1 H, J = 7.26 Hz), 7.66 (d, 1 H, J = 7.57 Hz), 7.45
7.30 (m, 3 H), 3.27 (t, 4 H, J = 6.84 Hz), 2.14
1.97 (m, 4 H), 1.74
1.62 (m, 4 H), 1.32
1.18 (m, 4 H), 1.17
1.04 (m, 4 H), 0.83
0.66 (m, 4 H). 13C NMR (125 MHz, CDCl3, δ): 154.00, 153.35, 152.83, 150.90, 141.76, 140.50, 135.37, 134.49, 132.31, 128.24, 128.05, 127.58, 127.08, 123.79, 122.91, 120.13, 119.89, 112.81, 55.16, 40.12, 33.92, 32.60, 29.04, 27.73, 23.61. MS (MALDI-TOF): m/z 707.37 [M]+.

Synthesis of 4-Bromo-7-(7-bromo-9,90 -bis(6-bromohexyl) fluorenyl)benzothiadiazole (3). Compound 2 (0.80 g, 1.14 mmol)

was dissolved in dichloromethane (20 mL) and cooled in an ice bath. Bromide liquid (0.45 g, 2.72 mmol) was then slowly added. After being stirred at 45 °C for 12 h, the reaction was quenched with sodium sulfite solution. Dichloromethane was added, and the organic portion was separated and washed with brine before drying over anhydrous MgSO4. The solvent was evaporated off, and the solid residues were purified by column chromatography on silica gel using dichloromethane/hexane (1:5) as eluent to afford 3 as yellow crystals (0.81 g, 90%). 1H NMR (500 MHz, CDCl3, δ): 7.95 (d, 1H, J = 7.75 Hz), 7.91 (dd, 1 H, J = 1.33, 7.89 Hz), 7.88 (s, 1 H), 7.81 (d, 1 H, J = 7.88 Hz), 7.64 (dd, 2 H, J = 8.12, 13.86 Hz), 7.50 (m, 2 H), 3.28 (t, 4 H, J = 6.70 Hz), 2.0 (m, 4 H), 1.67 (m, 4 H), 1.23 (m, 4 H), 1.11 (m, 4 H), 0.73 (td, 4 H, J = 7.74, 15.61 Hz). 13 C NMR (125 MHz, CDCl3, δ): 153.98, 153.14, 150.46, 140.60, 139.54, 135.86, 134.20, 132.29, 130.30, 128.46, 128.13, 126.23, 123.83, 121.50, 120.04, 113.04, 55.51, 40.05, 33.96, 32.61, 29.00, 27.74, 23.60. MS (MALDI-TOF): m/z 785.44 [M]+.

Synthesis of 4-(9,90 -Bis(6-bromohexyl)-7-((trimethylsilyl) ethynyl)fluorenyl)-7-((trimethylsilyl)ethynyl)benzothiadiazole (4). A solution of trimethylsilyl acetylene (1.08 g, 1.55 mL, 11.0 mmol,

d = 0.695 g/mL) in diisopropylamine ((iPr)2NH) (20.0 mL) was slowly added to a solution of 3 (3.9 g, 5.0 mmol), (Ph3P)2PdCl2 (0.175 g, 0.25 mmol), and CuI (0.047 g, 0.25 mmol) in (iPr)2NH (50.0 mL) under nitrogen at room temperature. The reaction mixture was then stirred at 70 °C for 8 h. The solvent was removed under reduced pressure, and the residue was chromatographed on silica gel using hexane as eluent to give 4 (2.8 g, 65%) as yellow crystals. 1H NMR (500 MHz, CDCl3, δ): 7.94 (m, 2 H), 7.87 (d, 1 H, J = 7.39 Hz), 7.81 (d, 1 H, J = 7.85 Hz), 7.73 (d, 1 H, J = 7.28 Hz), 7.69 (d, 1 H, J = 7.82 Hz,), 7.50 (d, 1 H, J = 7.86 Hz), 7.47 (s, 1 H), 3.26 (t, 4 H, J = 6.79 Hz), 2.00 (m, 4 H), 1.66 (m, 4 H), 1.21 (m, 4 H), 1.09 (m, 4 H), 0.70 (td, 4 H, J = 7.70, 15.16 Hz), 0.36 (s, 9 H), 0.30 (s, 9 H). 13C NMR (125 MHz, CDCl3, δ): 155.41, 153.20, 151.10, 150.87, 141.01, 140.91, 136.19, 135.16, 133.82, 131.43, 128.51, 127.27, 126.27, 123.86, 121.85, 120.23, 119.85, 115.58, 106.05, 101.84, 100.52, 94.46, 55.27, 40.09, 33.90, 32.64, 29.00, 27.76, 23.57, 0.10, 0.04. MS (MALDI-TOF): m/z 819.70 [M]+.

Synthesis of 4-(9,90 -Bis(6-bromohexyl)-7-ethynylfluorenyl) -7-ethynylbenzothiadiazole (5). A KOH aqueous solution

(3.0 mL, 20.0%) was diluted with methanol (15.0 mL) and added to a stirred solution of 4 (2.1 g, 2.5 mmol) in THF (20.0 mL). The mixture was stirred at room temperature for 6 h and extracted with hexane. The organic fraction was washed with water and dried over sodium sulfate. The crude product was chromatographed on silica gel using hexane as the eluent. Recrystallization of the product from methanol gave 5 (1.6 g, 92%) as yellow crystals. 1H NMR (500 MHz, CDCl3, δ): 7.98 (dd, 1 H, J = 1.47, 7.87 Hz), 7.94 (s, 1 H), 7.91 (d, 1 H, J = 7.34 Hz), 7.84 (d, 1 H, J = 7.90 Hz,), 7.76 (d, 1 H, J = 7.47 Hz,), 7.72 (d, 1 H, J = 7.80 Hz,), 7.53 (dd, 1 H, J = 1.10, 7.63 Hz,), 7.50 (s, 1 H), 3.64 (s, 1 H), 3.27 (t, 1 H, J = 6.74, 6.74 Hz,), 3.17 (s, 1 H), 2.03 (m, 4 H), 1.66 (m, 4 H), 1.22 (m, 4 H), 1.10 (m, 4 H), 0.71 (td, 4 H, J = 7.72, 15.20 Hz). 13C NMR

ARTICLE

(125 MHz, CDCl3, δ): 155.61, 153.16, 151.15, 150.97, 141.24, 140.97, 136.15, 135.69, 133.98, 131.46, 128.55, 127.25, 126.55, 123.91, 120.84, 120.31, 120.07, 114.48, 84.52, 83.70, 79.55, 77.47, 55.27, 40.06, 33.88, 32.61, 29.02, 27.75, 23.60. MS (MALDI-TOF): m/z 673.01 [M]+.

Synthesis of Neutral Hyperbranched Conjugated Polymer (P0). This precursor polymer was also synthesized in our recent report.28 A Schlenk tube charged with 5 (100 mg, 0.15 mmol) was degassed with three vacuum-nitrogen cycles. A solution of cyclopentadienylcobaltdicarbonyl (CpCo(CO)2) in anhydrous toluene (1.5 mL, 0.01 M) was then added to the tube, and the system was further frozen, evacuated, and thawed three times to remove oxygen. The mixture was vigorously stirred at 65 °C under irradiation with a 200 W Hg lamp (operating at 100 V) placed close to the tube for 8 h. After the mixture was cooled to room temperature, it was dropped in methanol (100 mL) through a cotton filter. The precipitate was collected and redissolved in tetrahydrofuran. The resultant solution was filtered through a 0.22 μm filter, and poured in hexane to precipitate further the product. After being dried in vacuum at 40 °C, P0 was obtained as a brown powder (65 mg, 65%). 1H NMR (500 MHz, CDCl3, δ): 8.50
7.30 (m, 8 H), 7.20 (br, 1 H), 3.67 (s, ∼0.56 H), 3.30 (br, 4 H), 3.20 (s, 0.20 H), 2.0 (br, 4 H), 1.70 (br, 4 H), 1.42
1.06 (m, 8 H), 0.77 (br, 4 H). 13C NMR (125 MHz, CDCl3, δ): 155.41, 154.34, 153.73, 153.06, 151.10, 150.97, 150.91, 150.08, 141.43, 140.50, 137.87, 134.02, 131.45, 129.04, 128.53, 128.23, 126.54, 125.30, 123.97, 120.68, 120.30, 119.98, 84.60, 83.30, 80.88, 77.92, 55.27, 40.10, 33.91, 32.64, 29.06, 27.77, 23.65. Mn = 6700, Mw/Mn = 1.8. Synthesis of Cationic HCPE (P1). Trimethylamine (2 mL) was added dropwise to a solution of P0 (50 mg) in THF (10 mL) at
78 °C. The mixture was stirred for 12 h, and then allowed to warm to room temperature. The precipitate was redissolved by the addition of methanol (8 mL). After the mixture was cooled to
78 °C, additional trimethylamine (2 mL) was added, and the mixture was stirred at room temperature for 24 h. After solvent removal, acetone was added to precipitate P1 as a brown powder (55 mg, 95%). 1H NMR (500 MHz, CD3OD, δ): 8.77
7.35 (m, 9 H), 3.63 (s, ∼0.56 H), 3.28 (br, 4 H), 3.05 (s, 18 H), 2.05 (br, 4 H), 1.58 (br, 4 H), 1.20 (br, 8 H), 0.77 (br, 4 H). 13 C NMR (125 MHz, CD3OD, δ): 155.49, 154.10, 150.97, 141.91, 141.37, 140.70, 138.15, 134.00, 133.43, 131.07, 130.22, 128.54, 128.32, 126.23, 125.97, 123.89, 121.27, 121.13, 119.92, 87.08, 84.05, 80.08, 77.82, 55.21, 52.20, 39.52, 28.73, 25.38, 23.29, 22.17. Synthesis of Core
Shell HCPE (P2). P1 (35 mg, 0.05 mmoL alkyne) and N3
PEG
NH2 (140 mg, 0.25 mmoL) were dissolved in DMF (5 mL). The mixture was degassed; then, PMDETA (12 mg, 0.0825 mmol) and CuBr (11.8 mg, 0.0825 mmoL) were added. After reaction at 65 °C under nitrogen for 24 h, the reaction mixture was cooled to room temperature and filtered through a 0.22 μm syringedriven filter. The filtrate was precipitated into diethyl ether to give red powders. The crude product was redissolved in water and further purified by dialysis against Mill-Q water using a 3.5 kDa molecular weight cutoff dialysis membrane for 3 days. After freeze-drying, P2 (45 mg, 78%) was obtained as brown fibers. 1H NMR (500 MHz, CD3OD, δ): 8.60
7.05 (m, ∼10.6 H), 4.56
3.40 (m, 145 H), 3.00
2.65 (m, 8 H), 2.47
1.70 (m, 22 H), 1.66
0.78 (m, 12 H), 0.56 (br, 4 H). Synthesis of Core
Shell HCPE (P3). P0 (30 mg, 0.05 mmoL alkyne) and N3
PEG
NH2 (140 mg, 0.25 mmoL) were dissolved in the mixture of THF (3 mL) and (2 mL). The mixture was degassed; then, PMDETA (12 mg, 0.0825 mmoL) and CuBr (11.8 mg, 0.0825 mmoL) were added. After reaction at 65 °C under nitrogen for 24 h, the reaction mixture was cooled to room temperature and filtered through a 0.22 μm syringe-driven filter. The filtrate was precipitated into diethyl ether to give red powders. The red powders were dissolved in dichloromethane and washed with water and brine. The solvent was removed in vacuum, and the residue was precipitated in diethyl ether twice. After drying 2968

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules

ARTICLE

Scheme 1. Synthesis of the Monomer 5a

Reagents and conditions: (i) Pd(PPh3)4/K2CO3, toluene/H2O, 90 °C, 48 h; (ii) Br2, CH2Cl2, 45 °C, 12 h; (iii) (Ph3P)2PdCl2/CuI, (iPr)2NH, 70 °C, 8 h; and (iv) KOH, THF/CH3OH/H2O, 24 °C, 6 h. a

under vacuum at 40 °C for 24 h, P3 is obtained as a red powder (51 mg, 73%). 1H NMR (500 MHz, CDCl3, δ): 8.80
7.30 (m, ∼10.7 H), 4.35
3.87 (m, ∼4 H), 3.80
3.55 (∼143 H), 3.26 (br, 4 H), 2.88 (br, 4 H), 2.10 (br, 4 H), 2.16 (br, 4 H), 1.36
0.98 (m, 8 H), 0.77 (br, 4 H). Synthesis of P2-Affibody. 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) aqueous solution (5 μL, 0.1 M) and P2 (2 μL, 1 mM, 2.8 M) were added to borate buffer (150 μL, 10 mM). The molar concentration of P2 is estimated according to the molar molecular weight of ∼26 000 based on each P2 containing 25 fluorene-BT units for the core and 14 PEG-NH2 for the shell. N-Hydroxysulfosuccinimide sodium salt (sulfo-NHS) (10 μL, 0.1 M) was subsequently added to the solution. An aqueous solution of anti-HER2 affibody (20 μL, 0.14 mM, 2.8 M) was then added under gentle stirring, and the solution was incubated for 1 h at room temperature. The reaction solution was passed through a NapTM-5 column (Sephadex G-25, GE Healthcare) with water as an eluent to remove unreacted EDC and sulfo-NHS. The crude product was used for SDS-PAGE. Before cellular imaging experiments, the product was purified against 10 mM PBS using a 6.5 kDa molecular weight cutoff dialysis membrane for 2 h.

’ RESULTS AND DISCUSSION Synthesis and Characterization. The HCPE with a hyperbranched conjugated polymer core and a linear PEG shell was synthesized via alkyne polycyclotrimerization,29 followed by click chemistry.30 Alkyne polycyclotrimerization is an efficient method to produce hyperbranched conjugated polymers with good solubility and active alkyne groups for postfunctionalization. In addition, click chemistry between the polycyclotrimerization product and azide-modified PEG is a facile way toward core
shell molecular architecture. To achieve an absorption maximum near the commercial excitation lines at 408 or 457 nm and emission at a relatively longwavelength region, 4-(9,90 -bis(6-bromohexyl)-7-ethynylfluorenyl)7-ethynylbenzothiadiazole (5) was designed and synthesized as the monomer for polycyclotrimerization (Scheme 1). In the first step, the palladium-mediated Suzuki cross-coupling reaction between 2-[9,90 -bis(6-bromohexyl)fluorenyl]-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (1) and 4,7-dibromobenzothiadiazole gave 4-[9,90 -bis(6bromohexyl)fluorenyl]-7-bromobenzothiadiazole (2) in 62% yield. Bromination of 2 in dichloromethane at 45 °C led to 4-bromo-7[7-bromo-9,90 -bis(6-bromohexyl)fluorenyl]benzothiadiazole (3) in 90% yield. It should be noted that the temperature is crucial for the precise control of bromination extent. Subsequently, 3 was

reacted with trimethylsilyl acetylene in a (Ph3P)2PdCl2/CuI catalyzed Sonagashira coupling reaction to afford 4-[9,90 -bis(6bromohexyl)-7-((trimethylsilyl)ethynyl)fluorenyl]-7-[(trimethylsilyl) ethynyl]benzothiadiazole (4) in 65% yield. At last, a trimethylsilyl deprotection process was carried out for 4 in a basic solution to yield 5 in 92% yield. The correct structure of 5 was confirmed by NMR and mass spectroscopies. The synthetic route toward the HCPE (P2) is shown in Scheme 2. Homopolycyclotrimerization of 5 in anhydrous toluene using 10 mol % CpCo(CO)2 as the catalyst gave the neutral hyperbranched conjugated polymer (P0), which has alkyne groups as the end-capper for click reaction. Because GPC measurement significantly underestimates the molecular weights of hyperbranched conjugated polymers,31,32 the 1H NMR spectrum (Figure S1 in the Supporting Information) of P0 is used to evaluate the molecular weight.33 As compared with the 1H NMR spectrum of 5, the appearance of the single peak at 7.2 ppm for P0 proves the formation of phenyl units as the branch points in the polymer. The ratio of the integrated area of the peak at 3.67 ppm (corresponding to the proton resonance of endcapping alkynes) to that of the peak at 2.00 ppm (corresponding to the proton resonance of methylene groups next to the nineposition of fluorene) is ∼0.14, which indicates that the degree of polymerization (DP) for P0 is ∼25.33 The Mn of P0 is thus estimated to be ∼16 000. Subsequent quaternization of P0 with trimethylamine in THF/ methanol mixture yielded the cationic precursor (P1). According to the ratio of the integrated area for the aromatic peaks ranging from 8.50 to 7.30 ppm to that for the peak at 3.05 ppm (the proton resonance of (CH3)3N) in the 1H NMR spectrum of P1, the quaternization degree of P1 is ∼95%. Click chemistry was carried out in dimethyl formamide (DMF) between P1 and azide-functionalized monodispersed PEG-NH2 (N3
PEG-NH2) at 65 °C using N,N,N0 ,N00 ,N000 -pentametyldiethylenetriamine (PMDETA) and CuBr as the catalyst, affording the bioamenable core
shell HCPE (P2). The appearance of the resonance peak of the triazole proton at 8.54 ppm (Figure S1 of the Supporting Information) reveals successful click reaction.34 Additionally, the ratio of the integrated area for the peak at 8.54 ppm to that at 2.28 ppm (corresponding to the proton resonance of methylene groups next to the nine-position of fluorene) is ∼0.14, which reflects that almost all of the alkynes of P1 are converted to triazole groups. As such, each P2 molecule has ∼14 PEG chains as the shell. 2969

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules

ARTICLE

Scheme 2. Synthesis of the HCPE (P2) and Its Neutral Counterpart (P3)a

Reagents and conditions: (i) CpCo(CO)2, toluene, 65 °C, UV radiation, 8 h; (ii) THF/methanol, NMe3, 24 h; (iii) NH2
PEG-N3, THF/DMF, PMDETA/CuBr, 65 °C, 24 h; and (iv) NH2
PEG-N3, DMF, PMDETA/CuBr, 65 °C, 24 h. a

For comparison, the counterpart polymer (P3) with the neutral conjugated core was also synthesized via click chemistry between P0 and N3
PEG
NH2 under similar reaction conditions. Solubility and Morphology. The solubility of P2 in water is measured to be ∼36 mg/mL at 25 °C, whereas its neutral counterpart P3 cannot be directly dissolved in water. Such a solubility difference highlights the importance of molecular design for P2, where the charged conjugated core plays an indispensable role in making the polymer soluble in aqueous media for further bioconjugation. To study the size of P2 in aqueous solution, we performed DLS for the polymer solution at a concentration of 1 μg/mL in Milli-Q water. Unimodal peak is observed for P2 solution, as shown in Figure 1a, revealing the formation of nanoparticles with an average diameter of ∼36 nm. HR-TEM was used to visualize further the morphology of P2 nanoparticles in dry state. As depicted in Figure 1b, P2 forms spherical nanoparticles with an average diameter of ∼30 nm. The slightly smaller particle size for TEM relative to that for DLS should be caused by the shrinkage of P2 during drying process. Moreover,

these nanoparticles possess a core
shell nanostructure, wherein the dark interior and the gray exterior correspond to the domains enriched with electron-rich conjugated segments and saturated PEG chains, respectively. Such a spherical core
shell nanostructure is beneficial to bioconjugation because the PEG shell could serve as a protective layer to minimize its nonspecific interaction with biomolecules. Optical Properties. The UV
vis absorption and PL spectra of P1 and P2 in water are shown in Figure 2a. The absorption peaks of P1 are at 309 and 410 nm, corresponding to fluorene and benzothiadiazole (BT) units, respectively. As compared with P1, the BT absorption peak of P2 is red-shifted by 12 nm from 410 to 422 nm. The PL maximum of P2 is at 565 nm, which is blue-shifted by 33 nm relative to that of P1. The PL quantum yields of P1 and P2 in water are ∼0.03 and 0.12, respectively, measured using quinine sulfate in 0.1 M H2SO4 as the standard. Previous reports have revealed that BT emission is sensitive to environmental polarity because of its charge-transfer electronic states, and the BT emission enhances as the environment 2970

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules

ARTICLE

Figure 1. (a) DLS of P2 in Milli-Q water at the polymer concentration of 1 μg/mL. (b) HR-TEM of P2.

Figure 2. (a) Normalized UV
vis absorption and PL spectra of P1 and P2 in water. Inset shows the photographs of polymer solutions under UV radiation at 365 nm. PL spectra of P1 (b) and P2 (c) (2 μg/mL) in 25 mM PBS (pH 7.4) upon addition of anti-HER2 affibody with the concentration ranging from 0 to 0.6 μM at intervals of 0.1 μM. Excitation at 420 nm.

becomes more hydrophobic.35
37 Accordingly, the higher quantum yield of P2 as compared with that of P1 reflects that the conjugated components of P2 are localized in a less polar microenvironment owing to the existence of PEG protective layer. With its bright fluorescence (inset of Figure 2a), large Stokes shift (143 nm), and obvious emission tail in the longwavelength region (>650 nm), P2 could serve as a good fluorescent molecule for optical imaging. Changes in the PL spectra for both P1 and P2 solutions upon the addition of anti-HER2 affibody were monitored to evaluate their nonspecific interactions with proteins. As shown in Figure 2b, the fluorescence of P1 significantly enhances with increased affibody concentration. This is consistent with our previous finding that BT-containing CPEs show fluorescence turnon responses toward charged biomolecules due to nonspecific electrostatic and hydrophobic interactions that induce the formation of complexes and in turn increase the local hydrophobicity of CPEs.35
39 In contrast, the fluorescence of P2 remains nearly the same upon addition of affibody, which indicates that the shielding effect of PEG shell significantly reduces nonspecific interactions.35
40 The difference between P1 and P2 in interaction with affibody is also examined by DLS. The effective diameter of P1 (2 μg/mL) significantly increases from ∼28 to 160 nm upon addition of affibody (0.6 μM), whereas it slightly changes from ∼36 to 45 nm for P2. The data further verify that the designed core
shell molecular architecture leads to very limited interaction between P2 and affibody, which is ideal for bioconjugation. Affibody Conjugation. To develop an HER2-specific probe, P2 was conjugated to anti-HER2 affibody via carbodiimideactivated reaction. The polymer was reacted with anti-HER2 affibody (1: 1.4 in molar ratio) in the presence of EDC and

N-sulfo-NHS. After conjugation, SDS-PAGE analysis was conducted to check the coupling reaction. The results show that the as-synthesized product migrates less and appears at the top of the gel (Figure S2 in the Supporting Information). This phenomenon is similar to the protein/polymer conjugates containing both PEG and charged components, indicating that the retention should be caused by the strong interaction with the gel.41 Moreover, the affibody band completely disappears for the product, whereas it remains for the P2/affibody mixture. This difference proves the successful conjugation of affibody to P2. Because there is almost no residual affibody after the reaction, the average number of affibody molecules conjugated to each P2 molecule is estimated to be ∼1.4 according to the feed ratio. Furthermore, the absorption and emission spectra of P2-affibody are similar to those of P2, which ensures a rational condition to compare their fluorescence imaging capabilities. Cellular Imaging. Fluorescence imaging of HER2-positive cancer cells based on P2-affibody was investigated and compared with P2. SKBR-3 breast cancer cell with high HER2 expression was chosen as the target, whereas MCF-7 breast cancer cell and NIH-3T3 fibroblast normal cell lacking HER2 expression were used as the negative controls. The CLSM images of these cells after treatment with P2 or P2-affibody are shown in Figure 3. The cellular nuclei are stained with PI. As shown in Figure 3a
c, green fluorescence from the cellular cytoplasm is visible for all P2-stained cells, indicating that P2 can be nonspecifically internalized by these cells and accumulated in the cytoplasm. The nonspecific cellular uptake was widely observed for polymeric nanoparticles with the size smaller than 50 nm.42 On the contrary, for P2-affibody stained cells (Figures 3d
f), green fluorescence in the cytoplasm is ∼3.5 times stronger in average 2971

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules

ARTICLE

Figure 3. CLSM images of (a) SKBR-3, (b) MCF-7, and (c) NIH-3T3 cells treated with P2 (0.5 μg/mL) for 2 h at 37 °C. CLSM images of (d) SKBR-3, (e) MCF-7, and (f) NIH-3T3 cells treated with P2-affibody conjugate (0.5 μg/mL) for 2 h at 37 °C. The cellular nuclei are stained by PI. The images are obtained from merging two fluorescence channels: 520
550 nm upon excitation at 408 nm for P2 and 575
635 nm upon excitation at 543 nm for PI.

Figure 4. CLSM fluorescence and fluorescence/transmission overlapped images of (a,b) SKBR-3, (c,d) MCF-7, and (e,f) NIH-3T3 cells treated with P2-affibody conjugate (0.5 μg/mL) for 20 min at 4 °C. The cellular nuclei are stained by PI. The fluorescence images are obtained from merging two fluorescent channels: 520
550 nm upon excitation at 408 nm for P2 and 575
635 nm upon excitation at 543 nm for PI.

for SKBR-3 cells than that for both MCF-7 and NIH-3T3 cells, which is also ∼4 times stronger on average than that for P2stained cells. The difference in these images clearly reflects that P2-affibody can be used for targeted imaging of SKBR-3 cells. To understand the origin of the targeting capability for P2affibody, we conducted two sets of cellular imaging experiments. In the first experiment, SKBR-3, MCF-7, and NIH-3T3 cells were incubated with P2-affibody conjugate at 4 °C. Such a low incubation temperature inhibits energy-dependent endocytosis,43 ensuring a simplified situation to evaluate the affinity of cellular probes to the specific extracellular receptors. Figure 4 shows the

corresponding CLSM fluorescence and fluorescence/transmission overlapped images of SKBR-3, MCF-7, and NIH-3T3 cells after treatment with P2-affibody. There is strong fluorescence localized on the cellular membrane of SKBR-3 cells, whereas nearly no visible fluorescence is observed for other cells. These images verify that the specific binding between P2-affibody and HER2 receptors in the extracellular domains of SKBR-3 does exist. In the second experiment, SKBR-3 cells were incubated with P2-affibody conjugate in the presence of 100-fold excess amounts of anti-HER2 affibody. As shown in Figure S3 in the Supporting Information, the fluorescence is very weak for cellular cytoplasm. This indicates that 2972

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules

Figure 5. CLSM fluorescence images of SKBR-3 breast cancer cells stained by P2-affibody under laser scanning for (a) 0 min and (b) 15 min upon excitation at 408 and collection of fluorescence from 520
550 nm. The scale bar is the same for both images.

ARTICLE

’ CONCLUSIONS In conclusion, we have demonstrated the first example of affibody-attached CPE and applied it for targeted fluorescence imaging of HER2 positive cancer cells. The HCPE (P2) synthesized via the combination of alkyne polycyclotrimerization and click chemistry exists as core
shell nanoparticles with an average size of ∼36 nm in aqueous solution. Such a unique nanoarchitecture of P2 minimizes nonspecific interactions with charged biomolecules and thus facilitates its bioconjugation with antiHER2 affibody. Comparison between P2-affibody and P2 in cellular imaging clearly illustrates that the specificity of antiHER2 affibody is preserved after bioconjugation. With the demonstrated good photostability and low cytotoxicity of P2, P2-affibody can be used as an effective organic nanoprobe for targeted fluorescence imaging of HER2-positive cancer cells in a high contrast and nonviral manner. Considering the variability of core
shell components and biorecognition elements, surfaceamenable HCPEs are appropriate for fulfilling various sensing and imaging tasks. This study thus provides a novel molecular design concept for CPEs to meet the structural and biochemical criteria for complicated biological applications. ’ ASSOCIATED CONTENT

bS

1

H NMR spectra of the polymers, SDS-PAGE, and CLSM image. This material is available free of charge via the Internet at http://pubs.acs.org. Supporting Information.

’ AUTHOR INFORMATION Figure 6. In vitro viability of NIH-3T3 cells treated with P2 solutions at concentrations of 10 (green), 20 (red) or 60 μg/mL (blue) for 8 and 24 h, respectively. The percentage cell viability of treated cells is calculated relative to that of the untreated cells with a viability arbitrarily defined as 100%.

free affibody molecules can efficiently block the binding between P2-affibody and HER2 receptors and thus significantly minimize the internalizaton of P2-affibody to SKBR-3 cells via a receptor mediated process. These data clearly illustrate that the specificity of anti-HER2 affibody is preserved in P2-affibody, allowing for effective discrimination of HER2-positive cancer cells from others. Photostability and Cytotoxicity. Changes in the CLSM images of SKBR-3 cells treated with P2-affibody conjugates before and after continuous laser scanning for 15 min were monitored to evaluate photostability of the nanoprobe. As shown in Figure 5a,b, the intensity decrease in the fluorescence image is not obvious, which indicates a relatively high photostability of the probe in cellular environment. This should be mainly attributed to the intrinsic good photostability of hyperbranched conjugated polymers,44,45 and also partially due to the existence of PEG shell that is able to shield the fluorescent conjugated core from an oxygen-rich environment.19 The cytotoxicity of P2 is evaluated for NIH-3T3 using a MTT cell-viability assay. Figure 6 summarizes the in vitro NIH-3T3 cell viability after being cultured with P2 solutions at the concentration of 10, 20, or 60 μg/mL for 8 or 24 h, respectively. It is noteworthy that these concentrations are much higher than that used for cellular imaging (0.5 μg/mL). The low cytotoxicity of P2 is witnessed by the nearly 100% cell viabilities within the tested period, which indicate that HPCEs are indeed ideal fluorescent molecules for long-term clinical applications.

Corresponding Author

*Fax: (+65) 6779-1936. E-mail: [email protected].

’ ACKNOWLEDGMENT We are grateful to the National University of Singapore (R279-000-301-646), National Research Foundation (R279000-323-281), Ministry of Education (R279-000-255-112), and Ministry of Defense (R279-000-301-232) for financial support. ’ REFERENCES (1) Medintz, I. L.; Uyede, H. T.; Goldman, E. R.; Mattoussi, H. Nat. Mater. 2005, 4, 435–446. (2) Michalet, X.; Pinaud, F. F.; Bentolila, L. A.; Tsay, J. M.; Doose, S.; Li, J. J.; Sundaresan, G.; Wu, A. M.; Gambhir, S. S.; Weiss, S. Science 2005, 307, 538–544. (3) Resch-Genger, U.; Grabolle, M.; Cavaliere-Jaricot, S.; Nitschke, R.; Nann, T. Nat. Methods 2008, 5, 763. (4) Thomas, S. W., III.; Joly, G. D.; Swager, T. M. Chem. Rev. 2007, 107, 1339–1380. (5) Ho, H. A.; Najari, A.; Leclerc, M. Acc. Chem. Res. 2008, 41, 168–178. (6) Jiang, H.; Taranekar, P.; Reynolds, J. R.; Schanze, K. S. Angew. Chem., Int. Ed. 2009, 48, 4300–4316. (7) Pu, K. Y.; Liu, B. Biosens. Bioelectron. 2009, 24, 1067–1073. (8) Duan, X.; Liu, L.; Feng, F.; Wang, S. Acc. Chem. Res. 2010, 43, 260–270. (9) Duarte, A.; Pu, K. Y.; Liu, B.; Bazan, G. C. Chem. Mater. 2011, 23, 501–515. (10) Sigurdson, C. J.; Nilsson, K. P. R.; Hornemann, S.; Manco, G.; Polymenidou, M.; Schwarz, P.; Leclerc, M.; Hammarstr€om, P.; W€uthrich, K.; Aguzzi, A. Nat. Methods 2007, 4, 1023–1030. 2973

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974

Biomacromolecules

ARTICLE

(11) McRae, R. L.; Phillips, R. L.; Kim, I. B.; Bunz, U. H. F.; Fahmi, C. J. J. Am. Chem. Soc. 2008, 130, 7851–7853. (12) Feng, X.; Tang, Y.; Duan, X.; Liu, L.; Wang, S. J. Mater. Chem. 2010, 20, 1312–1316. (13) Pu, K. Y.; Li, K.; Shi, J. B.; Liu, B. Chem. Mater. 2009, 21, 3816–3822. (14) Pu, K. Y.; Li, K.; Liu, B. Chem. Mater. 2010, 22, 6736–6741. (15) Pu, K. Y.; Liu, B. J. Phys. Chem. B 2010, 114, 3077–3084. (16) Wosnick, J. H.; Mello, C. M.; Swager, T. M. J. Am. Chem. Soc. 2005, 127, 3400–3405. (17) Fan, C.; Plaxco, K. W.; Heeger, A. J. J. Am. Chem. Soc. 2002, 124, 5642–5643. (18) Satrijo, A.; Swager, T. M. J. Am. Chem. Soc. 2007, 129, 16020–16028. (19) Pu, K. Y.; Li, K.; Liu, B. Adv. Funct. Mater. 2010, 20, 2770–2777. (20) Sheiko, S. S.; Sumerlin, B. S.; Matyjaszewskic, K. Prog. Polym. Sci. 2008, 33, 759–785. (21) Zeng, F.; Zimmerman, S. C. Chem. Rev. 1997, 97, 1681–1712. (22) Bosman, A. W.; Janssen, H. M.; Meijer, E. W. Chem. Rev. 1999, 99, 1665–1688. (23) Svenson, S.; Tomalia, D. A. Adv. Drug Delivery Rev. 2005, 57, 2106–2129. (24) Rabindran, S. K. Cancer Lett. 2005, 227, 9–23. (25) Nilsson, F. Y.; Tolmachev, V. Curr. Opin. Drug Discovery Dev. 2007, 10, 167–175. (26) Pu, K. Y.; Li, K.; Liu, B. Adv. Mater. 2010, 22, 643–646. (27) Pu, K. Y.; Liu, B. Adv. Funct. Mater. 2009, 19, 1371–1378. (28) Li, K.; Pu, K. Y.; Cai, L. P.; Liu, B. Chem. Mater. 2011, 23, 2113–2119. (29) Liu, J. Z.; Lam, J. W. Y.; Tang, B. Z. Chem. Rev. 2009, 109, 5799–5867. (30) Qin, A. J.; Lam, J. W. Y.; Tang, B. Z. Chem. Soc. Rev. 2010, 39, 2522–2544. (31) Grayson, S. M.; Frechet, J. M. J. Macromolecules 2001, 34, 6542–6544. (32) Shi, J. B.; Tong, B.; Li, Z.; Shen, J. B.; Zhao, W.; Fu, H. H.; Zhi, J. G.; Dong, Y. P.; H€aussler, M.; Lam, J. W. Y.; Tang, B. Z. Macromolecules 2007, 40, 8195–8204. (33) Liu, J. Z.; Zheng, R. H.; Tang, Y. H.; H€aussler, M.; Lam, J. W. Y.; Qin, A. J.; Ye, M. X.; Hong, Y. N.; Gao, P.; Tang, B. Z. Macromolecules 2007, 40, 7473–7486. (34) Gao, H.; Matyjaszewski, K. J. Am. Chem. Soc. 2007, 129, 6633–6639. (35) Wang, F.; Bazan, G. C. J. Am. Chem. Soc. 2006, 128, 15786–15792. (36) Pu, K. Y.; Liu, B. Adv. Funct. Mater. 2009, 19, 277–284. (37) Pu, K. Y.; Cai, L. P.; Liu, B. Macromolecules 2009, 42, 5933–5940. (38) Pu, K. Y.; Liu, B. J. Phys. Chem. B 2010, 114, 3077–3084. (39) Pu, K. Y.; Li, K.; Zhang, X. H.; Liu, B. Adv. Mater. 2010, 22, 4186–418. (40) Pu, K. Y.; Zhan, R. Y.; Liang, J.; Liu, B. Sci. China, Ser. B: Chem. 2011, 54, 567–574. (41) Magnusson, J. P.; Bersani, S.; Salmaso, S.; Alexander, C.; Caliceti, P. Bioconjugate Chem. 2010, 21, 671–678. (42) Lai, S. K.; Hida, K.; Man, S. T.; Chen, C.; Machamer, C.; Schroer, T. A.; Hanes, J. Biomaterials 2007, 28, 2876–2884. (43) Watson, P.; Jones, A. T.; Stephens, D. J. Adv. Drug Delivery Rev. 2005, 57, 43–61. (44) Tao, X. T.; Zhang, Y. D.; Wada, T.; Sasabe, H.; Suzuki, H.; Watanabe, T.; Miyata, S. Adv. Mater. 1998, 10, 226–230. (45) Paul, G. K.; Mwaura, J.; Argun, A. A.; Taranekar, P.; Reynolds, J. R. Macromolecules 2006, 39, 7789–7792.

2974

dx.doi.org/10.1021/bm200563a |Biomacromolecules 2011, 12, 2966–2974