Article pubs.acs.org/Biomac
Photoresponsive Poly(S-(o-nitrobenzyl)-L-cysteine)-b-PEO from a LCysteine N-Carboxyanhydride Monomer: Synthesis, Self-Assembly, and Phototriggered Drug Release Gang Liu and Chang-Ming Dong* Department of Polymer Science & Engineering, School of Chemistry and Chemical Engineering, Shanghai Jiao Tong University, Shanghai 200240, P. R. China S Supporting Information *
ABSTRACT: A photoresponsive S-(o-nitrobenzyl)-L-cysteine N-carboxyanhydride (NBC-NCA) monomer was for the first time designed, and the related poly(S-(o-nitrobenzyl)-Lcysteine)-b-poly(ethylene glycol) (PNBC-b-PEO) block copolymers were synthesized from the ring-opening polymerization (ROP) of NBC-NCA in DMF solution at 25 °C. Their molecular structures, physical properties, photoresponsive selfassembly, and drug release of PNBC-b-PEO were thoroughly investigated. The β-sheet conformational PNBC block within copolymers presented a thermotropic liquid crystal phase behavior, and the crystallinity of PEO block was progressively suppressed over the PNBC composition. The characteristic absorption peaks of these copolymers at about 310 and 350 nm increased over UV irradiation time and then leveled off, indicating that the o-nitrobenzyl groups were gradually photocleaved from copolymers until the completion of photocleavage. The PNBC-b-PEO copolymers self-assembled into spherical nanoparticles in aqueous solution, presenting a photoresponsive self-assembly behavior, together with a size reduction of nanoparticles after irradiation. The anticancer drug doxorubicin can be released in a controlled manner by changing the light irradiation time, which was induced by gradually photocleaving the PNBC core of nanoparticles. This work provides a facile strategy not only for the synthesis of photoresponsive polypeptide-based block copolymers but also for the fabrication of photoresponsive nanomedicine potential for anticancer therapy.
■
been approved for clinical human trials;25 whereas the NIR light can penetrate several millimeter up to centimeter depths of tissue, giving the NIR-triggered nanomedicine potential for clinical use. Therefore, discovering photoresponsive, especially NIR-responsive biodegradable polymers is still important for polymeric nanomedicine and tissue engineering.26−28 Because of their inherent ionizable groups (e.g., NH2 and COOH), hierarchical self-assembly and multiple secondary or tertiary structures, biodegradable polypeptides or proteins hybrids have been intensively designed and fabricated into stimuli-responsive (e.g., pH and temperature) nanomedicine for drug/gene delivery, and some entered clinical trials.29−49 For example, thermosensitive elastin-like polypeptides have been intensively studied for drug/gene delivery and tissue engineering scaffolds.50−53 Several kinds of pH-sensitive micelles based on PEO-b-polypeptides (e.g., poly (L-aspartic acid), poly(L-lysine), and poly(L-histidine)) have been thoroughly investigated for drug/gene/siRNA delivery.54−59 Our group synthesized linear−dendritic PEO-b-poly(L-glutamic
INTRODUCTION Owing to the precise remote, spatiotemporal control and easy use of light, photoresponsive polymers have been widely investigated for various applications in smart materials, actuators, phototriggered drug/gene delivery, and hydrogels for tissue engineering.1−12 For instance, several kinds of photoresponsive nanocarriers have been fabricated from onitrobenzyl-/coumarin-containing linear copolymers, presenting a triggered burst release of hydrophobic Nile Red dye.13−17 Fréchet et al. and we reported the near-infrared (NIR) photoresponsive micelles from diazonaphthoquinone-decorated dendritic copolymers, which showed a NIR-triggered release of Nile Red and anticancer drug doxorubicin (DOX), respectively.18−21 Anseth et al. discovered photodegradable poly(ethylene glycol) (PEO)-based hydrogels for remote and realtime manipulation of hydrogel properties.22,23 Utilizing independent photocleavage and photoclick conjugation reactions, they further created a kind of cytocompatible PEO/ oligopeptide hybrid hydrogels with programmable biomechanical and biochemical signals for 3D cell culture system.24 However, most photoresponsive copolymers lacked biodegradability, which to some extent limited their in vivo and clinical trials. In addition, some porphyrin-based photosensitizers have © 2012 American Chemical Society
Received: February 25, 2012 Revised: April 12, 2012 Published: April 20, 2012 1573
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
Scheme 1. Synthesis of Photoresponsive NBC-NCA Monomer (A) and the Related PNBC-b-PEO Block Copolymers and Their Photocleavage Reaction (B)
Scheme 2. Photoresponsive Self-Assembly and Phototriggered Drug-Release of Amphiphilic PNBC-b-PEO Block Copolymers in Aqueous Solution
cells, or both, minimizing the side effects of anticancer drugs.10,26,27 Motivated by the photoresponsive polypeptide nanomedicine, herein, a photoresponsive S-(o-nitrobenzyl)-L-cysteine Ncarboxyanhydride (NBC-NCA) monomer was for the first time designed, and the related PNBC-b-PEO block copolymers were synthesized from the ring-opening polymerization (ROP) of NBC-NCA in DMF solution at 25 °C (Scheme 1). Their molecular structures, physical properties, photoresponsive selfassembly, and phototriggered drug release were thoroughly
acid) copolymers, establishing a platform for the fabrication of pH-sensitive micelles and hydrogels.60−63 Interestingly, Mezzenga et al. reported photochromic PEO-b-poly(L-glutamic acid) decorated with 50% spiropyran and the photoinduced reversible micellar transition.64 Despite this example, the studies on the synthesis and self-assembly of photoresponsive polypeptide-based copolymers remain scarce, although this would provide a platform for the generation of phototriggered biodegradable nanomedicine. Importantly, this might realize both quick and efficient drug release in diseased site within 1574
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
Kα radiation source (wavelength = 1.54 Å). The supplied voltage and current were set to 40 kV and 30 mA, respectively. Samples were exposed at a scan rate of 2θ = 4° min−1 between 2θ = 2 and 40°. Circular dichroism (CD) spectra were recorded on a JASCO J-815 spectrometer in deionized water at a concentration of 1 mg/mL at 25 °C. The critical aggregation concentration (cac) values of copolymers were determined employing the hydrophobic dye solubilization method using 1,6-diphenyl-1,3,5-hexatriene as a probe molecule.65−67 UV−vis spectra of samples were recorded in the range of 200−500 nm at room temperature. The mean size of nanoparticles was determined at 25 °C by DLS using a Malvern ZS90 instrument. All of the measurements were repeated three times, and the average values reported were the mean diameter ± standard deviation. TEM was performed without staining by using a JEM-2010 at 200 kV accelerating voltage. Samples were deposited onto the surface of 300 mesh Formvar-carbon film-coated copper grids, and excess solution was removed in vacuo. For the UV light irradiation, the sample solution (∼2 mL) was placed vertically under a high-pressure mercury lamp (wavelength: 365 nm; power: 150 W), and the distance was kept at 15 cm. Synthesis of S-(o-Nitrobenzyl)-L-cysteine (NBC). The NBC intermediate was synthesized according to the previous publication.68 L-Cysteine (0.6245 g, 5 mmol) was dissolved in 10 mL of deionized water and then neutralized by triethylamine (665 μL, 4.6 mmol). The solution was cooled to 0 °C; then, o-nitrobenzyl bromide solution (1.014 g, 4.6 mmol in 10 mL of methanol) was added dropwise over 15 min. The reaction solution was stirred vigorously at 0 °C for 30 min and then at room temperature for 60 min. The pale-yellow precipitate was washed sequentially with ethyl acetate (3 × 20 mL) and deionized water (3 × 20 mL). By recrystallizing the crude product in hot water (NBC/H2O, 1:36, w/w), the purified NBC was obtained as yellow long needles (0.8564 g, 72.7%). 1H NMR of NBC (CDCl3/ CF3COOD, v/v 1/1): δ = 3.16−3.22 (d, ArCH2SCHH, 1H), 3.36− 3.40 (d, ArCH2SCHH, 1H), 4.21−4.29 (q, ArCH2S, 2H), 4.55−4.58 (q, ArCH2SCH2CH, 1H), 7.52−8.15 (m, Ar, 4H). 13C NMR of NBC: δ = 169.42 (CHCOOH), 148.78, 134.42, 134.09, 133.03, 129.26, 125.93 (Ar), 53.96 (ArCH2SCH2CH), 33.06 (ArCH2SCH2CH), 32.67 (ArCH2SCH2CH). FT-IR (KBr, cm−1): 3066 (νNH3+), 1649 (νasNH3+), 1602 (νasC=O), 1527 (νsNH3+), 1500 (νasNO2), 1399 (νsC=O), 1354 (νsNO2), 863 (νC−NO2), 711 (νC−S). Synthesis of S-(o-Nitrobenzyl)-L-cysteine N-Carboxyanhydride (NBC-NCA). The photoresponsive NBC-NCA monomer was for the first time designed and synthesized from NBC with triphosgene according to the literature.69−71 NBC (1 g, 3.9 mmol) was suspended in THF (16 mL). The mixture was heated to 50 °C under a nitrogen atmosphere, and then triphosgene (0.6561 g, 2.2 mmol) in 5 mL of THF was added quickly via a syringe. The reaction solution was continued for 1 h after it became clear (about 30 min). The NBCNCA solution was dropwise poured into 80 mL of hexane, and the light-yellow viscous solid was dried under reduced pressure at 45 °C for 2 days (0.6892 g, 63% yield). 1H NMR of NBC-NCA (DMSO-d6): δ = 2.87−2.89 (dd, ArCH2SCH2, 2H), 4.07−4.16 (q, ArCH2S, 2H), 4.75−4.78 (dt, ArCH2SCH2CH, 1H), 7.55−8.05 (m, Ar, 4H), 9.16 (s, CHNH, 1H). 13C NMR of NBC-NCA (CDCl3/CF3COOD, v/v 1/1): δ = 168.63 (CHCOO),155.59 (NHCOO), 148.09, 134.35, 133.10, 132.38, 129.26, 126.00 (Ar), 58.54 (ArCH2SCH2CH), 34.48 (ArCH2SCH2CH), 32.79 (ArCH2SCH2CH). FT-IR (KBr, cm−1): 3284 (νN−H), 1851 (νasCH(CO)2O), 1826 (νNH(CO)2O), 1790 (νsCH(CO)2O), 1519 (νasNO2), 1342 (νsNO2), 1286 (νasO=C−O−CO), 938 (νsO=C−O−CO), 859 (νC−NO2), 706 (νC−S). TOF-MS: calcd for C11H10SN2O5, 282.0310; found, 281.0226 [M-1]+, 282.0308 M+, 283.0255 [M+1]+. Synthesis of Monofunctional Primary Amine-Terminated Poly(ethylene glycol) Macroinitiator (PEO-NH2). The PEO-NH2 macroinitiator was synthesized using PEO with N-(tertbutoxycarbonyl)glycine (N-Boc-glycine) catalyzed by DCC/DMAP, followed by deprotection of BOC group.72 PEO (2.0 g, 0.4 mmol), NBoc-glycine (141.0 mg, 0.8 mmol), DMAP (12.0 mg, 0.1 mmol), and DCC (165.1 mg, 0.8 mmol) were completely dissolved in THF (60
investigated by means of FT-IR, NMR, gel permeation chromatography, differential scanning calorimetry, polarized optical microscope (POM), wide-angle X-ray diffraction (WAXD), UV−vis, dynamic light scattering (DLS), and transmission electron microscope (TEM). The o-nitrobenzyl (NB) groups were gradually photocleaved from copolymers until the completion of photocleavage. The self-assembled nanoparticles became smaller over UV irradiation and then leveled off, presenting a phototriggered drug-release profile (Scheme 2).
■
EXPERIMENTAL SECTION
Materials. Acetonitrile (≥99%), chloroform (≥99%), dichloromethane (≥99.5%), ethyl acetate (≥99.5%), methanol (≥99.5%), and triethylamine (≥99%) were purchased from Shanghai Sinopharm and distilled before use. Dimethylformamide (DMF, 99.5%) was distilled from calcium hydride under reduced pressure and stored over 4 Å molecular sieves. Tetrahydrofuran (THF, ≥99%) and hexane (≥97%) were purchased from Shanghai Sinopharm and distilled from sodium and stored over 4 Å molecular sieves. Acetone (≥99.5%), ether (≥99.7%), and trifluoroacetic acid (≥99%) were purchased from Shanghai Sinopharm and used as received. L-Cysteine (97%, Aldrich), 1,6-diphenyl-1,3,5-hexatriene (DPH, Aldrich), o-nitrobenzyl bromide (98%, Aldrich), 2-propynylamine (98%, Aladdin Chemistry) and triphosgene (99%, Aladdin Chemistry) were used as received. Dicyclohexylcarbodiimide (DCC, 99.2%), N-(tert-butoxycarbonyl)glycine (99.7%), and 4-(dimethylamino)-pyridine (DMAP, 99.1%) were purchased from Shanghai GL Biochem and used as received. DOX hydrochloride was purchased from Beijing Huafeng United Technology and used as received. Poly(ethylene glycol) methyl ether (PEO, Mn,NMR = 5000, Aldrich) was dried overnight at 50 °C in vacuum before use. Methods. Fourier transform infrared (FT-IR) spectra were recorded on a Perkin-Elmer Paragon 1000 spectrometer at frequencies ranging from 400 to 4000 cm−1 at room temperature, and powder samples were thoroughly mixed with KBr crystal and pressed into pellet form. 1H NMR (400 MHz) and 13C NMR (100 MHz) spectra were recorded on a Varian Mercury-400 spectrometer at room temperature. Deuterated chloroform (CDCl3), dimethyl sulfoxide (DMSO-d6), and the mixed solvents of deuterated trifluoroacetic acid and deuterated chloroform (CF3COOD/CDCl3: 1/1 v/v) were used as solvent, respectively, and tetramethylsilane was used as an internal standard. Time-of-flight mass spectrometry (TOF-MS) was performed on a Waters Acquity UPLC/Premier QTOF MS Premier. Molecular weights (MWs) and polydispersities (Mw/Mn) of polymers were determined on a gel permeation chromatograph (GPC, HLC-8320, Tosoh Corporation, Japan) equipped with two HLC-8320 columns (TSKgel Super AWM-H, pore size: 9 μm; 6 × 150 mm, Tosoh) and a double-path, double-flow refractive index (RI) detector (Bryce) at 30 °C. The elution phase was DMF-LiBr (0.01 mol·L−1) with an elution rate of 0.6 mL·min−1, and a series of polymethyl methacrylate was used as the calibration standard. The differential scanning calorimetry (DSC) analysis was carried out using a TA Instruments Q2000 instrument under nitrogen flow (10 mL/min). All samples were first heated from 0 to 220 °C at 10 °C/min and held for 3 min to erase the thermal history, then cooled to 0 at 10 °C/min, and finally heated to 220 at 10 °C/min. The indium standard was used for temperature and enthalpy calibrations. The melting temperature (Tm) and the crystallization temperature (Tc) were taken as the maximal points of both endothermic and exothermic peaks, respectively. Thermogravimetric analysis (TGA) was obtained using a Perkin-Elmer TGA 7 instrument under nitrogen flow (10 mL/min) from 40 to 600 °C at 10 °C/min. The liquid crystal (LC) phase of polymer was observed using a Leica Microsystems Leica DMLP polarized optical microscope (POM). The power sample was sandwiched between two glass plates, heated to 210 at 10 °C/min and then held for 180 min for observing LC phase. WAXD patterns of powder samples were obtained at room temperature on a Shimadzu XRD-6000 X-ray diffractometer with a Cu 1575
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
mL) under a nitrogen atmosphere. The reaction mixture was stirred vigorously at 25 °C for 48 h. After adding several drops of acetone, the mixture was filtered to remove dicyclohexylcarbodiurea. The solution was concentrated and then precipitated into a large excess of cold ether (0 °C, 200 mL). The white precipitate was filtered and dried in vacuo at 30 °C to give PEO-NHBoc (1.775 g, 86.1%). PEO-NHBoc (1.0 g, 0.2 mmol) was completely dissolved in the mixed solvents of trifluoroacetic acid (2.5 mL) and dichloromethane (7.5 mL) and then stirred at 25 °C for 3 h. The solution was concentrated and then precipitated into a large excess of cold ether (0 °C, 100 mL) to give white solid. The crude product was dissolved completely in chloroform (100 mL); then, the solution was washed sequentially with saturated sodium bicarbonate aqueous solution three times (3 × 20 mL) and distilled water twice (2 × 20 mL). The organic phase was collected, dried over sodium sulfate, and then dropwise poured into a large excess of cold ether. The powder was dried in vacuo to give 879.9 mg PEO-NH2 (88% yield). 1H NMR (CDCl3): δ = 1.78 (s, NH2), 3.38 (s, CH3O, 3H), 3.46−3.48 (t, CH3OCH2CH2O, 2H), 3.54−3.56 (t, CH3OCH2CH2O, 2H), 3.59−3.70 (m, O(CH2CH2O)n, 450H), 3.81−3.83 (t, OCH 2 CH 2 OOCCH 2 NH 2 , 2H), 4.28−4.30 (t, OCH2CH2OOCCH2NH2, 2H). FT-IR (KBr, cm−1): 2888 (νC−H), 1750 (νCO), 1113 (νC−O−C). Mn,GPC = 7830, Mw/Mn = 1.19. Preparation of Poly(S-(o-nitrobenzyl)-L-cysteine)-b-poly(ethylene glycol) Block Copolymers (PNBC-b-PEO). Linear PNBC-b-PEO block copolymers were synthesized by the ROP of NBC-NCA using monofunctional primary amine-terminated PEONH2 as macroinitiator in DMF solution at room temperature.60−63 As control experiment, PNBC homopolypeptide was synthesized using primary amine-terminated 2-propynylamine as initiator. In a representative polymerization reaction, PEO-NH2 (179.2 mg, 35.8 μmol) was dissolved completely in 2 mL of DMF under a nitrogen atmosphere; then, a degassed solution of NBC-NCA (100 mg, 0.35 mmol) in 1 mL of DMF was added via a syringe. The resulting solution was stirred vigorously at room temperature for 48 h, diluted with 1 mL of THF, and then precipitated dropwise into a large excess of ether (30 mL). The white precipitate was filtered and dried in vacuo at 40 °C to give 229.9 mg of PNBC9-b-PEO (87% yield). The copolymer yield varied within 75−90 wt %. 1H NMR of PNBC9-bPEO (CDCl3/CF3COOD, v/v 1/1): δ = 2.91−3.05 (m, ArCH2SCH2, 18H), 3.50 (s, CH3O, 3H), 3.59−3.74 (m, O(CH2CH2O)n, 450H), 4.09−4.12 (m, ArCH2SCH2, 18H), 4.69−4.77 (m, ArCH2SCH2CH, 9H), 7.44−7.95 (m, Ar, 36H). FT-IR (KBr, cm−1): 3281 (νN−H), 2887 (νC−H), 1629 (amide I), 1526 (amide II), 1344 (νsNO2), 1115 (νC−O−C), 858 (νC−NO2), 709 (νC−S). Mn,GPC = 6740, Mw/Mn = 1.50. Preparation of DOX-Loaded Nanoparticles in Aqueous Solution. Using a dialysis method,65−67 a typical procedure for the fabrication of DOX-loaded nanoparticles in aqueous solution is as follows. Both PNBC-b-PEO (10 mg) and DOX·HCl (5 mg, 8.6 μmoL) were dissolved in 6.5 mL of DMF, in which 1.5 fold of Et3N (12.9 μmoL) was added to neutralize HCl in solution. Distilled water (1.5 mL) was then added gradually at a speed of 30 μL/min using a microsyringe until the formation of nanoparticles. The resulting nanoparticles solution was then put into a dialysis bag (MWCO = 3500) and subjected to dialysis against 4 × 1 L of distilled water for 24 h. The blank nanoparticles were fabricated similarly. The drug-loaded nanoparticles solution was lyophilized, and the nanoparticles powder (1 mg) was dissolved in 5 mL of DMF and then analyzed at 500 nm by UV−vis. The drug-loading capacity of nanoparticles is calculated as the weight ratio of actual drug to drug-loaded nanoparticles, and the drugloading efficiency of nanoparticles is calculated as the weight ratio of actual and added drug content. In Vitro DOX Release. The DOX-loaded nanoparticles solution was diluted with PBS (pH 7.4) to a concentration of 0.4 mg/mL (this concentration is the same as that of blank nanoparticles used for photoresponsive study), then the solution (5 mL) was irradiated at 365 nm for the predetermined time (2, 5, and 10 min). The irradiated sample in dialysis bag (MWCO = 3500) was then put in 15 mL of the PBS buffer solution at 37 °C. The drug-released solution was changed periodically, and the amount of DOX released from nanoparticles was
determined at 500 nm by UV−vis. All release experiments were carried out in duplicate (three determinations in each), and all data were averages of six determinations used for drawing figures.
■
RESULTS AND DISCUSSION Synthesis of Photoresponsive NBC-NCA Monomer and PNBC-b-PEO Block Copolymers. From a synthetic viewpoint, the major strategies for the synthesis of photoresponsive polymers can be reduced into the following three methods: (a) the postmodification of traditional functional polymers (e.g., poly(methacrylic acid)) by utilizing photochromic molecules such as azobenzene, coumarin, NB derivatives, and spiropyran; (b) the homo/copolymerization of photoresponsive monomer; and (c) the (co)polymers decorated by multiple photochromic molecules or bearing a photocleavable junction.1−24,73−83 Notably, the atom transfer radical polymerization of NB-containing (meth)acrylates was polymerized to low conversion because of the radicalterminating reactions of NB groups.84,85 To overcome the disadvantage, Thomas et al. put forward the ring-opening metathesis polymerization of NB-containing norbornenes for the well-defined synthesis of photoresponsive (co)polymers.85 In this Article, a photoresponsive S-(o-nitrobenzyl)-L-cysteine N-carboxyanhydride (NBC-NCA) monomer was for the first time designed, which was polymerized to produce photoresponsive polypeptide copolymers via ROP. The monomer was consumed quantitatively, and NB groups had no obvious interference with ROP. The NBC-NCA monomer was synthesized by two steps (Scheme 1A), according to the previous publications.68−71 The overall yield for NBC-NCA monomer was ∼45%, and its chemical structure was fully characterized by FT-IR, 1H NMR, 13C NMR, and TOF-MS spectroscopy (Figure 1 and Supporting Information Figure S1). FT-IR spectroscopy showed the characteristic N-carboxyanhydride peaks at 1851 cm −1 (ν asCH(CO) 2 O ), 1826 cm −1 (νNH(CO)2O), and 1790 cm−1 (νsCH(CO)2O) and the nitro (NO2) peaks at 1519 cm−1 (νasNO2) and 1342 cm−1 (νsNO2). 1H NMR spectrum of NBC-NCA (DMSO-d6) clearly shows a single proton peak at 9.16 ppm for the cyclic amide (1H, NH− CO), the aromatic protons of NB at 7.55−8.05 ppm (4H, Ar), the cyclic methine proton at 4.75−4.78 ppm (1H, OCCHNH), the methylene protons of NB at 4.07−4.16 ppm (2H, CH2), and the methylene protons linking NCA ring at 2.87−2.89 ppm (2H, CH2). In addition, both 13C NMR and TOF-MS also confirmed the molecular structure of NBC-NCA. In all, these results demonstrate that the NBC-NCA monomer is pure and can be used for the following ROP. The homopolymerization of NBC-NCA monomer was first tested in DMF solution at room temperature (25 °C). It was observed that the homopolypeptide PNBC would precipitate from solution during the polymerization process, and the MW or the chain length of PNBC did not increase over the molar ratio of NCA monomer to primary amine initiator (see Supporting Information Table S1). As a note, the more the molar ratio of NBC-NCA to initiator, the more precipitate. This result is probably due to the fact that poly(L-cysteine) homopolypeptide exclusively formed a β-sheet conformation and had lower solubility in DMF, as reported in literature (also see the following FT-IR analysis).41,87 Thus, using monofunctional primary amine-terminated poly(ethylene glycol) (PEONH2) as macroinitiator, the ROP of NBC-NCA monomer was carried out in DMF solution at 25 °C to produce the 1576
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
block was obtained with reasonable polydispersity (Mw/Mn) of ∼1.5. Compared with the GPC trace of PNBC9 homopolypeptide (Figure 2), the unimodal elution peak of PNBC9-b-PEO
Figure 2. GPC traces of PEO-NH2, PNBC, and PNBC-b-PEO.
moved to a lower elution time region, verifying the successful synthesis of block copolymer. However, as for the higher MW block copolymers with longer chain length of PNBC (i.e., PNBC20-b-PEO, PNBC56-b-PEO, and PNBC82-b-PEO), their unimodal elution peaks reversely shifted toward the higher elution time region. This abnormal result can be attributed to the β-sheet structure of PNBC block,41,87 where the strong intermolecular interactions among PNBC chains induced the copolymers’ hydrodynamic volume smaller over the MW of PNBC block. To test this hypothesis, we did the multi-angle laser light-scattering (MALLS) coupled GPC experiment for PNBC-b-PEO (Supporting Information Figure S2). PNBC9-bPEO gave an unreasonable MW of 122 900 g/mol, which was 18 times more than the actual MW (Mn,NMR) of 7140. This result suggests that the strong intermolecular hydrogenbondings among PNBC-b-PEO chains indeed existed, resulting in the larger aggregates with higher MW. Similar phenomenon was observed for other polypeptide-based block copolymers.88−90 The representative 1H NMR spectrum of PNBC-b-PEO block copolymers clearly shows the typical proton signals at 7.44−7.95 ppm (f, g, h, i) and 4.09−4.12 ppm (e) assignable to the NB group, those at 2.91−2.99 ppm (d) assignable to the methylene group (−CHCH2−S-NB), those at 4.60−4.80 ppm
Figure 1. 1H NMR spectra (DMSO-d6) of NBC-NCA (A) and the 13 C NMR spectra (CDCl3/CF3COOD) of NBC-NCA (B).
photoresponsive PNBC-b-PEO block copolymers, as shown in Scheme 1B and Table 1. No precipitate was apparently observed during the polymerization process, and a series of PNBC-b-PEO block copolymers with different MW PNBC Table 1. Synthesis of Photoresponsive PNBC-b-PEO Block Copolymers entry
a
PEO-NH2 PNBC9e PNBC9-bPEO PNBC20b-PEO PNBC56b-PEO PNBC82b-PEO
[NCA]/ [initiator] (mol/mol)
PNBC/ PEO% (wt %)d
Mn, GPCb
Mw/ Mnb
Mn,NMR
10/1 10/1
7830 1660 6740
1.19 1.45 1.50
5060 2140 7140
2140 2140
32/68
25/1
5580
1.40
9760
4760
49/51
70/1
4610
1.46
18340
13330
73/27
110/1
4470
1.46
24530
19520
80/20
c
Mn,PNBC
c
a
Subscript number represents the degree of polymerization of PNBC, which was determined by 1H NMR. bNumber-average molecular weight (Mn,GPC) and the polydispersity (Mw/Mn) of polymer was determined from gel permeation chromatography (GPC) using DMFLiBr (0.01 mmol·L−1) as eluent. cBoth Mn,NMR of copolymer and Mn,PNBC of PNBC block were determined by 1H NMR. dPNBC/PEO % denotes the weight fraction of PNBC and PEO within copolymers. e As a control, PNBC homopolypeptide was synthesized from the ROP of NBC-NCA using 2-propynylamine as initiator in DMF solution at room temperature.
Figure 3. 1H NMR spectrum (CDCl3/CF3COOD) of PNBC9-b-PEO. 1577
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
intramolecular and intermolecular hydrogen-bondings, respectively. It is well known that both the amide I and II bands at ∼1650 and ∼1550 cm−1, together with the amide bands at ∼1630 and ∼1530 cm−1, are characteristic of both α-helix and β-sheet conformations of polypeptides.60−63,89−92 All PNBC-bPEO block copolymers showed the strong amide I and II bands at 1629 and 1526 cm−1, demonstrating that the PNBC block within copolymers mainly presented a β-sheet conformation. This is consistent with the fact that poly(L-cysteine) exclusively assumed a β-sheet conformation.41,87 In addition, the temperature-dependent FT-IR spectra of PNBC-b-PEO were studied after the samples were annealed from 220 to 25 °C (Supporting Information Figure S3). These results confirmed that PNBC exclusively adopted a β-sheet conformation, which did not change within 220 °C. Physical Properties of PNBC-b-PEO Block Copolymers in Solid State. DSC was performed to characterize the LC phase of PNBC block and the crystalline property of PEO block within copolymers in solid state, as shown in Figure 5A and Table 2. Compared with PEO and PNBC homopolymers, all PNBC-b-PEO block copolymers showed the maximal melting temperature (Tm = 53−60 °C) for PEO block and the LC phase transition temperature (TLC = 184−206 °C) for PNBC block in the first heating runs. The LC phase transition gave a small endothermic enthalpy of about 0.1 to 0.6 J/g, whereas it disappeared in the cooling and the second heating runs during DSC experiment (Supporting Information Figure S4). The DSC was repeated to measure after the PNBC-b-PEO copolymers were annealed at 200 °C for 1 h. This also confirmed that the PNBC block presented the thermotropic LC phase transition (Supporting Information Figure S5). As determined by TGA, the thermal decomposition temperature of PNBC-b-PEO copolymers was ∼265 °C (Supporting Information Figure S6), which indicates that the small enthalpy determined by DSC was not induced by the experimental error, such as the thermal degradation of copolymers. These results demonstrate that the LC phase transition of the polypeptide PNBC block indeed existed and was irreversible, which was observed in other polypeptide-b-PEO copolymers.60−63,89,90 Moreover, POM was used to verify further the LC phase of PNBC within copolymers (Figure 5B). After both PNBC20-bPEO and PNBC56-b-PEO samples were annealed at 210 °C, the LC phase of PNBC block appeared for about 30 min and then
(c) assignable to the backbone (−CH−) of PNBC, and both PEO backbone signals (b) at 3.55−4.00 ppm and its methoxyl group (a) at 3.50 ppm. By taking into account the integral ratios of protons “b” to “c” (or d, f), the actual MWs and the fraction (PNBC/PEO wt %) of these block copolymers can also be calculated, as shown in Table 1. Moreover, FT-IR spectroscopy verified the presence of both PNBC and PEO components in these purified block copolymers. As shown in Figure 4, these block copolymers exhibit strong NH stretching
Figure 4. FT-IR spectra of PEO-NH2, PNBC, and PNBC-b-PEO.
band at 3281 cm−1, both amide I and II bands at 1629 and 1526 cm−1, and nitro (NO2) band at 1344 cm−1 for PNBC block, and the distinct bands at 2887 (CH2) and 1115 cm−1 (C−O− C) for PEO block. Meanwhile, it was observed that the relative intensity of amide I and II bands, nitro band, or both increased with increasing PNBC fraction. Taken together, the above results convincingly demonstrated that the photoresponsive PNBC-b-PEO block copolymers with different PNBC fraction (32−80 wt %) have been successfully synthesized by the PEONH2 initiated ROP of NBC-NCA monomer in DMF solution at room temperature (Scheme 1). FT-IR spectroscopy also provides a simple method to characterize the secondary conformations such as α-helix and βsheet of polypeptides in solid state, which are stabilized by
Figure 5. DSC curves (left) of PEO-NH2, PNBC, and PNBC-b-PEO in the first heating runs (solid lines; the letters A, B, C, D, and E denote TLC) and in the cooling runs (dot lines); POM photographs (right) of PNBC-b-PEO annealed for 1 h. 1578
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
compared with the crystalline PEO homopolymer.60−63 Meanwhile, the diffraction intensity of PEO block within copolymers was attenuated gradually with increasing the PNBC fraction. This indicates that the crystallinity of PEO block was progressively prohibited, which is in agreement with that determined by DSC. Taken together, the aforementioned analyses indicate that the PNBC block within PNBC-b-PEO copolymers exclusively assumed a β-sheet conformation and presented a thermotropic LC phase behavior; the crystallinity of PEO block was progressively suppressed over the PNBC fraction. Photoresponsive Self-Assembly and Phototriggered Drug-Release of Amphiphilic PNBC-b-PEO in Aqueous Solution. Under UV irradiation at 365 nm, the photoresponsive property of PNBC9-b-PEO in acetonitrile solution (0.4 mg/mL) was monitored by UV−vis spectroscopy (Figure 7A). The characteristic absorption peaks of PNBC9-b-PEO at
Table 2. Solid Physical Properties of PEO-NH2, PNBC, and the PNBC-b-PEO Block Copolymers entry PEO-NH2 PNBC9 PNBC9-bPEO PNBC20-bPEO PNBC56-bPEO PNBC82-bPEO
Tm,PEO (°C)a
ΔHm,PEO (J/g)b
Xc, PEO (%)c
TLC,PNBC (°C)d
ΔHLC,PNBC (J/g)e
58.7
185.4
95.1
57.0
105.1
78.5
195.7 206.0
3.44 0.34
52.8
58.8
58.5
196.2
0.53
49.6
25.9
48.7
183.7
0.56
52.9
17.4
44.2
197.4
0.14
a Tm,PEO denotes the maximal melting temperature of PEO in the second heating run. bΔHm,PEO denotes the fusion enthalpy of PEO crystal in the second heating run. cXC,PEO denotes the degree of crystallinity of PEO block within copolymers, where XC = ΔHm/ (WPEO × ΔH0m,PEO), ΔH0m,PEO = 197.0 J/g. dTLC,PNBC denotes the liquid crystal (LC) phase transition temperature of PNBC in the first heating run. eΔHLC,PNBC denotes the fusion enthalpy of PNBC LC phase in the first heating run, which has been corrected by taking into account the mass percentage of the PNBC block in each copolymer sample.
grew slowly with annealing time. Meanwhile, the phase morphology retained after the temperature decreased to 25 °C. This weak LC phase was possibly induced by the nematic orientation of the rod-like PNBC block,91 which needs further investigation. Collectively, these results demonstrate that PNBC block within copolymers indeed presented a thermotropic LC phase behavior. Furthermore, the degree of crystallinity (Xc) of PEO block within copolymers decreased from 78.5 to 44.2% with increasing the PNBC fraction, which demonstrates that the crystallinity of PEO block was progressively suppressed within copolymers. WAXD is another useful method to demonstrate both the secondary conformation of PNBC block and the crystallinity of PEO block, as shown in Figure 6. Similar to the PNBC homopolypeptide, all of these copolymers presented a diffraction peak at ∼0.39 Å−1, confirming that the PNBC segment within copolymers mainly assumed a β-sheet conformation.60−63,89−92 This result is consistent with the above FT-IR analysis. These block copolymers showed the similar Bragg diffraction peaks at about 1.35 and 1.64 Å−1
Figure 7. UV−vis spectra of PNBC9-b-PEO (0.4 mg/mL) in acetonitrile solution (A) and the self-assembled PNBC9-b-PEO nanoparticles (0.4 mg/mL) in aqueous solution (B) under UV irradiation at 365 nm for different time.
about 310 and 350 nm increased over irradiation time and then leveled off after 20 min, indicating that NB groups were gradually photocleaved from copolymers until the completion of photocleavage reaction.76−85 Note that the photoresponsive NB-containing (co)polymers and the photocleavage mechanism have been widely investigated.2,3,22−24,76−86 Moreover, the UV−vis spectra of self-assembled PNBC9-b-PEO nanoparticles
Figure 6. WAXD patterns of PEO-NH2, PNBC, and PNBC-b-PEO. 1579
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
in aqueous solution (0.4 mg/mL) demonstrated a similar trend over UV irradiation time (Figure 7B). The degree of photocleavage of NB groups within PNBC9-b-PEO can be calculated from the maximal absorbance change at 310 nm, and the photocleavage reaction finished within ∼20 min (Supporting Information Figure S7). Meanwhile, the self-assembled PNBC20-b-PEO nanoparticles in water gave a similar photocleavage trend under 365 nm irradiation, but the photocleavage reaction finished within ∼40 min (Supporting Information Figure S8). In addition, the nanoparticles solution turned from colorless to slightly yellow, which is typical of the nitrosobenzaldehyde produced by the photocleavage. These results indicate that the NB groups were gradually photocleaved from the PNBC core of nanoparticles, and these copolymers presented a photoresponsive self-assembly behavior in aqueous solution (Scheme 2). This will be thoroughly investigated as follows. The cac of amphiphilic block copolymers is an important parameter for characterizing their self-assembly in aqueous solution, which can be measured by the dye solubilization method.65−67 As shown in Figure 8, the absorbance intensity of
Figure 9. TEM photographs of the self-assembled PNBC-b-PEO nanoparticles cast from their aqueous solutions: PNBC9-b-PEO (A), PNBC9-b-PEO after irradiation for 30 min (B), drug-loaded nanoparticles of PNBC9-b-PEO (C), and PNBC82-b-PEO (D).
Figure 8. Relationship of the absorbance intensity of DPH as a function of the PNBC-b-PEO copolymer concentration in aqueous solution at room temperature.
1,6-diphenyl-1,3,5-hexatriene dye remained nearly constant below a certain concentration and then increased substantially, reflecting the incorporation of dye in the hydrophobic region of nanoparticles.65−67 The cac value gradually decreased from 4.58 × 10−2 to 4.30 × 10−3 mg/mL with increasing the hydrophobic PNBC fraction (32−80 wt %). This cac value decreases with increasing the hydrophobicity−hydrophilicity balance, which is a normal result similar to that reported in other amphiphilic block copolymers.65−67,93−96 After the PNBC-b-PEO nanoparticles were irradiated at 365 nm for 5 min, the cac value increased from original (4.58 to 7.57) × 10−2 mg/mL, and then to 8.75 × 10−2 mg/mL after 30 min of irradiation. This can be attributed to the gradual photocleavage of hydrophobic NB groups from PNBC block, which resulted in the progressively increased hydrophilicity, that is, the photocleaved product PNBCm−x-co-poly(L-cysteine)x-b-PEO (x ≤ m) compared with original PNBC-b-PEO (Scheme 1B). Both the morphology and the average size of the selfassembled PNBC-b-PEO nanoparticles were investigated by means of TEM and DLS, as shown in Figures 9 and 10 and
Figure 10. Nanoparticle size distribution of the self-assembled PNBC9-b-PEO nanoparticles in aqueous solution determined by DLS (A). The DLS-averaged diameter of nanoparticles dependence on the UV irradiation time (B).
Supporting Information Table S2. As for PNBC9-b-PEO with a lower PNBC fraction of 32 wt %, it spontaneously selfassembled into spherical micelles with a diameter of 79.0 ± 1580
dx.doi.org/10.1021/bm300304t | Biomacromolecules 2012, 13, 1573−1583
Biomacromolecules
Article
11.0 nm. This is similar to the conventional polymeric micelles (10−100 nm), suggesting that they had a hydrophobic PNBC core surrounded by a hydrophilic PEO corona.93−96 Note that the average size determined by DLS is bigger than that determined from TEM analysis. Specifically, DLS gives a hydrodynamic radius that corresponds to the core and the swollen corona of the micelles, whereas TEM often gives a size of the core for micelles in a dried state as the corona with low electronic density is not always visible. As the PNBC fraction increased to 80 wt % within PNBC82-b-PEO, similar spherical nanoparticles were obtained with a diameter of 191.4 ± 18.6 nm. This result is induced by both the increased chain length of β-sheet conformational PNBC block and the increased intermolecular aggregation among PNBC-b-PEO chains. In addition, PNBC82-b-PEO had the lowest cac value, also suggesting the strongest intermolecular aggregation trend.93−96 The photoresponsive self-assembly of PNBC-b-PEO was further monitored by means of DLS (Figure 10B) and TEM. The average size of PNBC9-b-PEO nanoparticles (i.e., original size of ∼79 nm) reduced sharply over irradiation time and then leveled off (i.e., ∼44 nm). TEM also confirmed the size reduction of PNBC9-b-PEO nanoparticles after 365 nm irradiation (Figure 9A,B), and the spherical morphology did not obviously change. This result can be explained as follows. After NB groups were gradually photocleaved from copolymers until completion, the resulting PNBCm−x-co-poly(L-cysteine)x block still aggregated in aqueous solution to form a relatively hydrophobic core, which was stabilized by a hydrophilic PEO corona (Scheme 2). As confirmed by IR and CD (Supporting Information Figures S9 and S10), the polypeptide segments within the irradiated nanoparticles of PNBCm−x-co-poly(Lcysteine)x-b-PEO still assumed a β-sheet conformation in aqueous solution. This demonstrated that the strong intermolecular hydrogen-bondings existed among the irradiated PNBCm−x-co-poly(L-cysteine)x chains, resulting in the aggregated polypeptide core in aqueous solution. Taken together, these results indicate that PNBC-b-PEO block copolymers selfassembled into spherical nanoparticles with a photocleavable PNBC core surrounded by a hydrophilic PEO corona in aqueous solution, and the nanoparticles were reduced and then leveled off after UV irradiation. As a widely used clinical anticancer drug,96−98 DOX was used to demonstrate the concept of phototriggered drug release. The hydrophobic DOX drug was encapsulated into the PNBC9-bPEO nanoparticles by a dialysis method, and the DOX-loaded nanoparticles were characterized by means of UV−vis, DLS, and TEM. Determined by UV−vis spectroscopy, the DOXloading efficiency and capacity was ∼47 and 19 wt %, respectively. The DOX-loaded nanoparticles mainly showed nearly spherical morphology with a DLS-determined diameter of 38.5 ± 7.7 nm. Note that a few drug-loaded micelles presented worm-like morphology, which could be attributed to the fact that the hydrophobic DOX changed the hydrophobicity−hydrophilicity balance of the drug/copolymer nanomedicine.94−96 After irradiating nanoparticles for different time (i.e., 2, 5, and 10 min), in vitro drug-release of DOX-loaded nanoparticles was performed in PBS at 37 °C (Figure 11). The DOX release became faster with increasing the UV irradiation time, demonstrating that DOX can be released in a controllable manner. The cumulative drug release amount nearly doubled, and ∼42% DOX was released within 108 h after irradiating the nanoparticles for 10 min. This result can be attributed to the fact that the irradiated nanoparticles of PNBCm−x-co-poly(L-
Figure 11. In vitro drug-release profiles (10 mM PBS, pH 7.4, and 37 °C) of DOX-loaded PNBC9-b-PEO nanoparticles after UV irradiation at 365 nm for different time.
cysteine)x-b-PEO had progressively decreased hydrophobic π−π interactions between anthracene-like DOX and NB groups, which resulted in the acceleration of the DOX diffusion from nanoparticles at 37 °C.21,96,99 As a note, the initial drugrelease was not obviously accelerated within 12 h when the DOX-loaded nanoparticles were irradiated within 5 min. This is due to the fact that the quick diffusion of DOX from the surface layer of nanoparticles played a dominant role during the initial drug-release period.21,67,96 In addition, using the longer UV irradiation to achieve the quicker drug release needs further investigation. Collectively, the self-assembled DOX-loaded nanomedicine with a diameter of 38.5 ± 7.7 nm (
