Article pubs.acs.org/Biomac
Biarmed Poly(ethylene glycol)-(pheophorbide a)2 Conjugate as a Bioactivatable Delivery Carrier for Photodynamic Therapy Wool Lim Kim,†,‡ Hana Cho,†,§ Li Li,‡ Han Chang Kang,*,§ and Kang Moo Huh*,‡ ‡
Department of Polymer Science and Engineering, Chungnam National University, 99 Daehak-ro, Yuseong-gu, Daejeon 305-764, Republic of Korea § Department of Pharmacy and Integrated Research Institute of Pharmaceutical Sciences, College of Pharmacy, The Catholic University of Korea, 43 Jibong-ro, Wonmi-gu, Bucheon-si, Gyeonggi-do 420-743, Republic of Korea ABSTRACT: In the study presented here, we developed a bioreducible biarmed methoxy poly(ethylene glycol)-(pheophorbide a)2 (mPEG-(ss-PhA)2) conjugate for cancer-cellspecific photodynamic therapy (PDT). PhA molecules were chemically conjugated with biarmed linkages at one end of the mPEG molecule via disulfide bonds. Under aqueous conditions, the amphiphilic mPEG-(ss-PhA)2 conjugate selfassembled to form core−shell-structured nanoparticles (NPs) with good colloidal stability. The mPEG-(ss-PhA)2 NPs exhibited intramolecular and intermolecular self-quenching effects that enabled the NPs to remain photoinactive in a physiological buffer. However, the dissociation of the NP structure was effectively induced by the cleavage of the disulfide bonds in response to intracellular reductive conditions, triggering the rapid release of PhA molecules in a photoactive form. In cellculture systems, in addition to significant phototoxicity and intracellular uptake, we observed that the dequenching processes of PhA in the mPEG-(ss-PhA)2 NPs highly depended on the expression of intracellular thiols and that supplementation with glutathione monoethylester facilitated more rapid PhA release and enhanced the PhA phototoxicity. These findings suggest that the bioreducible activation mechanism of mPEG-(ss-PhA)2 NPs in cancer cells can maximize the cytosolic dose of active photosensitizers to achieve high cytotoxicity, thereby enhancing the treatment efficacy of photodynamic cancer treatment.
■
INTRODUCTION Photodynamic therapy (PDT) is a promising medical technology for the treatment of cancers and other nonmalignant diseases.1−6 PDT is based on a photochemical reaction that mediates interactions among chemical photosensitizers (PSs), light, and molecular oxygen. The lightinduced activation of PSs results in the in situ production of reactive singlet oxygen (1O2), which can damage cell viability.2 Nevertheless, clinical applications of PDT agents, such as porphyrin and phthalocyanine derivatives, have remained limited because of the poor water solubility and unfavorable pharmacokinetic properties of such agents in addition to the occurrence of post-treatment cutaneous photosensitivity.7,8 To reduce these limitations, one potential strategy is to chemically link hydrophobic PSs with various water-soluble, biofunctional molecules (e.g., tumor-targeting ligands or peptides,9−12 carrier proteins,13,14 carbohydrates,10,15 and hydrophilic biomacromolecules16−18) for the preparation of water-soluble or waterdispersible PS conjugates and nanoparticles (NPs). Poly(ethylene glycol) (PEG) is a nontoxic, nonimmunogenic, nonantigenic, and water-soluble polymer that is approved by the Food and Drug Administration as a safe excipient for drug delivery.19,20 PEGylation is one of the leading strategies for prolonging the in vivo circulation time of therapeutics by means of reduced phagocytosis and renal clearance.21,22 In © XXXX American Chemical Society
cancer therapy specifically, PEG conjugations with chemical drugs (e.g., doxorubicin,23,24 cis-platinum,25 taxol,26,27 and PDT agents28) are an attractive strategy for improving drug delivery by enhancing aqueous solubility, reducing renal clearance, decreasing enzymatic degradation, and limiting immunogenic and antigenic reactions. PEGylated platforms also exhibit a superior capacity for passive tumor targeting through the enhanced permeability and retention (EPR) effect because PEGylated formulations are capable of remaining in the circulatory system for an extended period of time and thus provide a sufficient level of accumulation in the pathological area (e.g., a solid tumor).29,30 On the other hand, the long circulation that is characteristic of PEG−PS conjugates must be carefully monitored because it carries a risk of causing longlasting skin photosensitivity, as the PEG−PS conjugates are photodynamically active throughout the postadministration period.31,32 In addition, in PEG−PS conjugates, the hydrophilic, repellant nature of PEG may negatively impact the intracellular distribution of PS at the target site and may also pose a diffusion barrier that negatively affects the release of 1O2 generated under light exposure33 because the irreversible Received: March 7, 2014 Revised: May 4, 2014
A
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 1. Conceptual illustration of the mechanism of bioreducible mPEG-(ss-PhA)2 NPs for the switchable photoactivity of PhA: mPEG-(ss-PhA)2 conjugates readily form self-assembled NPs under aqueous conditions. The NPs do not exhibit photoactivity during blood circulation because of the intramolecular and intermolecular self-quenching of PhA. When the NPs are internalized by cancer cells, the intracellular reduction-triggered cleavage of the disulfide bonds accelerates the dissociation of the NP structure and causes the rapid release of PhA, thereby restoring the photoactivity of the PS.
Second, the mPEG-(ss-PS)2 NPs should accumulate at tumor sites because of the EPR effect. Third, after the internalization of the mPEG-(ss-PS)2 NPs by the tumor cells, the thiolresponsive disulfide linkers in the NPs should be rapidly cleaved in the cytosol, as the concentrations of glutathione (GSH) in the cytosol (approximately 2−10 mM) are much higher than those in extracellular fluids (approximately 2−20 μM).36,37 The remarkable difference in GSH levels between the two environments should trigger the rapid dissociation of the mPEG-(ss-PS)2 NPs, followed by the rapid release of PS molecules into the intracellular environment, resulting in efficient photoactivity of the PS molecules. Recently, several GSH-responsive activatable photosensizing systems were reported, such as 2,4-dinitrobenzenesulfonate-substituted zinc(II) phthalocyanine,38 nitroethenyl-BODIPY conjugate,39 graphene oxide-ss-chlroin e6 conjugate,40 and hyaluronic acid-sschlroin e6 conjugates.41 Our group also has developed various activatable PS formulations regarding GSH-mediated activation of PS, including glycol chitosan-ss-pheophorbide a (PhA) conjugate,17 heparin-PhA-conjugated gold NP,18 and heparinPhA-conjugated iron oxide/gold core/shell NP.42 These GSHresponsive formulations showed superior capacity for selective release and efficient activation of the PSs in tumor cells, which might maximize the cytosolic dose of active PSs to achieve higher cytotoxicity, thereby enhancing the treatment efficacy. Therefore, we expect self-quenchable mPEG-(ss-PS)2 NPs with rapid intracellular-release kinetics and a bioreducible activation mechanism should be capable of engaging in highly selective cancer treatment using the PDT approach. Thus, the intent of this work was to synthesize such a bioreducible mPEG-(ss-PS)2
photodamage to pivotal biomacromolecules and other cellular components essentially depends on the amount and intracellular distribution of 1O2, which, in combination with the limited diffusion distance and short lifetime of 1O2,34 may significantly reduce the therapeutic efficacy of PDT. To develop an elegant PEG−PS conjugate with a highly potent PDT effect, we designed an activatable PDT agent known as a bioreducible mPEG-(ss-PS)2 conjugate, which has a methoxy PEG (mPEG) with two arms at one end, each of which is chemically linked to one PS molecule via a disulfide bond. It was expected that the designed mPEG-(ss-PS)2 conjugate and its self-assembled NPs should possess the aforementioned typical characteristics of PEGylated drugdelivery carriers. Specifically, the mPEG-(ss-PS)2 conjugate and its NPs were expected to display several characteristic features, including potent cancer-cell-specific PDT effects and minimal traditional PDT complications because of the design strategy of site-specific controllable photoactivity (Figure 1). First, in the mPEG-(ss-PS)2 conjugate, the proximity of two PS molecules should create a donor−acceptor pair that enables fluorescence resonance energy transfer (FRET)-mediated intramolecular photoquenching.35 In addition, the amphiphilic nature of the mPEG-(ss-PS)2 conjugate should allow it to readily self-assemble to form core−shell-structured NPs, resulting in π−π-stacked self-aggregation among hydrophobic PS molecules in the NPs to form FRET-based donor−acceptor pairs, which should lead to efficient intermolecular quenching.16 These desired characteristics should allow the PS molecules to remain photodynamically inactive in the mPEG-(ss-PS)2 NPs during systemic circulation to inhibit harmful 1O2 generation. B
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 2. (a) Overall synthesis scheme for the mPEG-(ss-PhA)2 conjugate, (b) the 1H NMR spectra of mPEG derivatives obtained during mPEG(ss-PhA)2 conjugation, and (c) FT-IR spectra of mPEG, mPEG-serinol, and mPEG-(ss-NH2)2.
conjugate using PhA as a model PS and to develop selfquenchable and bioactivatable mPEG-(ss-PhA)2 NPs. The conjugate and the NPs were investigated with regard to their
physicochemical characteristics, self-assembling behaviors, photoactivity, and in vitro PDT efficacy in tumor cell culture systems. C
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
■
Article
low-molecular-weight impurities. The mPEG-(ss-PhA)2 conjugate was obtained in powder form via lyophilization (yield: 69%). 1H NMR: PEG segment, δ = 2.6 (2 H, α-CH2-), 2.8 (2 H, β-CH2-), 3.4 (3 H, CH3-), and 3.5−3.7 (196 H, -CH2-CH2-O-); PhA moiety, 1.2 (3 H, -C11H3), 1.6 (3 H, -C4H3), 1.8 (2 H, -C1H2-), 2.1 (2 H, -C2H2-), 3.1 (3 H, -C6H3), 3.2 3.2 (3 H, -C13H3), 3.8 (3 H, -C10H3), 4.0 (3 H, -C15H3), 4.2 (1 H, -C3H-), 6.1 (2 H, =C8H2), 6.3 (1 H, -C14H-), 7.9 (1 H, -C7H), 8.9 (1 H, -C12H), 9.2 (1 H, -C9H), and 9.7 (1 H, -C5H) (Figure 3a). FT-IR (KBr) v/cm−1: PEG segment, 1098 (C−
MATERIALS AND METHODS
Materials. Methoxy poly(ethylene glycol) (mPEG, MW 2000 Da), cystamine dihydrochloride, p-nitrophenyl chloroformate (pNC), dithiothreitol (DTT), N-(3-(dimethylamino)propyl)-N′-ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), triethylamine (TEA), acetonitrile, dimethyl sulfoxide (DMSO), dichloromethane (DCM), 9,10-dimethylanthracene (DMA), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), RPMI 1640 medium, Ca2+-free and Mg2+-free Dulbecco’s phosphate-buffered saline (DPBS), 4-(2-hydroxy-ethyl)-1-piperazine (HEPES), D-glucose, sodium bicarbonate, glutathione monoethylester (GSHOEt), recombinant human insulin, fetal bovine serum (FBS), penicillin-streptomycin antibiotics, bovine serum albumin (BSA), and trypsin-EDTA solution were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). 2-Amino1,3-propanediol (serinol) and pheophorbide a (PhA) were obtained from Tokyo Chemical Industry (Tokyo, Japan) and Frontier Scientific, Inc. (Logan, Utah, U.S.A.), respectively. Spectra/Por membranes were purchased from Spectrum Laboratories, Inc. (Rancho Dominguez, CA, U.S.A.). All chemicals that were used for polymer synthesis were at least analytical grade and were used without further purification. Synthesis of Bioreducible mPEG-(ss-PhA)2 Conjugate. As shown in Figure 2a, a bioreducible mPEG-(ss-PhA)2 conjugate was synthesized via a multistep synthetic route. First, pNC-activated mPEG was synthesized from mPEG and pNC. In brief, mPEG (10 g, 5 mmol) was dried in a vacuum at 70 °C for 12 h and was then dissolved in 20 mL of acetonitrile and activated by pNC (12.60 g, 62.5 mmol) in the presence of TEA at room temperature (RT) under nitrogen atmosphere for 24 h. The pNC-activated mPEG solution was filtered to remove the triethylammonium chloride that had precipitated out, and the synthesized pNC-activated mPEG was then precipitated in diethyl ether and dried in a vacuum. The synthesis of pNC-activated mPEG was confirmed via 1H NMR (JNM-AL 400 spectrometer, JEOL Ltd., Akishima, Japan; 400 MHz, CDCl3): δ = 3.4 (3 H, CH3-), 3.5− 3.7 (196 H, -CH2-CH2-O-), 7.4 (2 H, Ar-H), and 8.2 (2 H, Ar-H) (Figure 2b). Second, for the synthesis of mPEG-serinol, pNC-activated mPEG (8.50 g, 3.93 mmol) was dissolved in DMSO and reacted with serinol (2.86 g, 31.41 mmol) at RT under nitrogen atmosphere for 36 h, followed by dialysis using a dialysis membrane (molecular-weight cutoff (MWCO) 1000 Da) against deionized water for 2 days and then lyophilization. The crude product was dissolved in DCM and filtered to remove unreacted serinol, followed by precipitation in cold diethyl ether and drying in a vacuum. 1H NMR (400 MHz, CDCl3): δ = 3.4 (3 H, CH3-), 3.5−3.7 (196 H, -CH2-CH2-O-), and 5.5 (1 H, -NH-) (Figure 2b); FT-IR (Magna 560, Nicolet, U.S.A.; KBr) v/cm−1: 1098 (C−O, ether), 1735 (CO, ester), and 2930 (C−H, methylene) (Figure 2c). Third, the activation of the hydroxyl end groups of mPEG-serinol using pNC was conducted under conditions similar to those described above for the synthesis of the mPEG-serinol. Then, mPEG-(pNC)2 was reacted with cystamine dihydrochloride in DMSO in the presence of TEA at RT for 24 h. The resultant solution was filtered and precipitated in cold diethyl ether, followed by drying in a vacuum. The product was dissolved in DCM to remove unreacted cystamine and extracted three times with 60 mL of saturated NaCl solution. The final product, mPEG-(ss-NH2)2, was collected after precipitation in cold diethyl ether and dried in a vacuum. 1H NMR (400 MHz, CDCl3): δ = 2.6 (2 H, α-CH2-), 2.8 (2 H, β-CH2-), 3.4 (3 H, CH3-), 3.5−3.7 (196 H, -CH2-CH2-O−), and 5.3 (1 H, -NH-) (Figure 2b); FT-IR (KBr) v/ cm−1: 1098 (C−O, ether), 1400 (C−N, amide), 1735 (CO, ester), and 2930 (C−H, methylene) (Figure 2c). Fourth, to synthesize the mPEG-(ss-PhA)2 conjugate, PhA (60 mg, 0.1 mmol) was dissolved in 12 mL of DMSO and activated using EDC (97 mg, 0.506 mmol) and NHS (58 mg, 0.506 mmol). After the NHSactivated PhA solution had reacted for 12 h, mPEG-(ss-NH2)2 (80 mg, 0.033 mmol) was added to the solution. The reaction mixture was stirred for an additional 12 h under nitrogen atmosphere and then dialyzed (MWCO 1000 Da), first against DMSO for 2 days and then against deionized water for 1 day, to remove unreacted PhA and other
Figure 3. 1H NMR and FT-IR spectra of the mPEG-(ss-PhA)2 conjugate. O, ether), 1400 (C−N, amide), 1735 (CO, ester), and 2930 (C−H, methylene); PhA moiety, 990, 1235, 1370, 1470, 1580 (chlorin skeleton), 1680 (CC), and 1706 (CO, keto group in exocycle) (Figure 3b). Characterization of the mPEG-(ss-PhA)2 Conjugate. The GPC traces of main PEG derivatives and mPEG-(ss-PhA)2 conjugate were recored using by a Agilent 1100 gel permeation chromatography system (Agilent Technologies, CA, U.S.A.) equipped with quaternary pump and refractive index detector using Agilent PL gel 5 μm mixed-C and mixed-D (300 mm × 7.5 mm) columns. The eluent was THF, and the flow rate was 1.0 mL/min. The temperature of columns was set and maintained at 40 °C. The number of PhA molecules conjugated on the mPEG backbone was determined using a colorimetric method. In brief, the lyophilized mPEG-(ss-PhA)2 conjugate was dissolved in 10 mL of DMF at various concentrations. The absorbance of the mPEG-(ss-PhA)2 conjugate at 667 nm was determined using a UV/ visible spectrometer (SINCO PDA, UV/visible spectrometer operation, Korea). The PhA content in the mPEG-(ss-PhA)2 conjugate was calculated based on a calibration curve for PhA (667 nm) in DMF. D
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
changed, and the cells were irradiated using a 670 nm light source. MCF7 cells were irradiated at 8.67 mW/cm2 during different irradiation times (16, 32, 63, 126, and 252 s), and HeLa cells was irradiated with light at 1.7 J/cm2 (8.67 mW, 63 s) to evaluate the lightinduced toxicity. At 36 h post-treatment (i.e., 24 h after light exposure), a MTT solution (0.5 mg/mL) was added to the culture medium, and the cells were incubated for an additional 4 h. After the MTT-containing medium was removed, the formazan crystals produced by the live cells were dissolved in DMSO (100 μL/well), and the absorbance at 570 nm was measured. In addition, to evaluate the effect of GSHOEt on the phototoxicity of mPEG-(ss-PhA)2 NPs, GSHOEt (2 mM) was added to the NP culture medium at 2 h posttreatment. To assess the cellular uptake of free PhA and mPEG-(ss-PhA)2 NPs, the cells (5 × 105 cells/well) were seeded in a 6-well plate and incubated for 24 h. After a 4 h treatment with PhA and mPEG-(ssPhA)2 ([PhA] = 0.5 μg/mL), the cells were rinsed twice with DPBS and detached. The cellular uptake of PhA was measured using a flow cytometer (FACScanto II, Becton−Dickinson, Franklin Lakes, NJ, U.S.A.) with a primary HeNe laser (633 nm) and a fluorescence detector (660 ± 10 nm). The PhA uptake of the cells was analyzed based on a gated viable population of at least 1 × 104 cells. To measure the intracellular localization and intracellular intensity of PhA delivered by the mPEG-(ss-PhA)2 NPs, MCF7 cells (1.5 × 104 cells/well) were seeded in an eight-chamber slide and incubated for 24 h. Before treatment with the NPs, the culture medium was replaced with fresh medium (0.2 mL). The cells were exposed to the NPs for 6 h. To assess the effect of GSHOEt, the cells were exposed to the NPs for 2 h and subsequently incubated in the presence of GSHOEt for an additional 4 h. After the 6 h incubation, the cells were rinsed with DPBS for further evaluation using a laser scanning confocal microscope (LSM710; Carl Zeiss, Oberkochen, Germany) equipped with excitation lasers (633 nm for HeNe) and variable band-pass emission filters.
In addition, the critical self-quenching concentration (CQC) was determined via fluorescence spectrometry using pyrene as a fluorescence probe. mPEG-(ss-PhA)2 solutions at a variety of concentrations ranging from 1 × 10−5 to 0.2 mg/mL were prepared and mixed with pyrene solutions that contained 6.0 × 10−7 M pyrene in each vial. Then, the fluorescence excitation spectra of pyrene were recorded using a fluorescence spectrophotometer (Cary Eclipse, Varian, U.S.A.) with an emission wavelength of 390 nm. Preparation and Characterization of Self-Assembled mPEG(ss-PhA)2 NPs. The mPEG-(ss-PhA)2 NPs were prepared from the mPEG-(ss-PhA)2 conjugate using a dialysis method. In brief, the conjugate was dissolved in DMF, followed by sonication for 10 min to obtain a clear solution. The solution was transferred into a dialysis membrane (MWCO 1000 Da) and dialyzed against deionized water for 24 h. The size and size distribution of the mPEG-(ss-PhA)2 NPs were determined using a particle size analyzer (ELS-Z, Photal, Japan). The morphology of the NPs was characterized using a field-emission scanning electron microscope (FE-SEM, JSM-6700, JEOL Ltd., Japan) at 10 keV. To assess the colloidal stability of the mPEG-(ss-PhA)2 NPs, NPs in PBS (0.5 mg/mL) and PBS containing 1 mg/mL of BSA were prepared, respectively, and the solutions were incubated in a shaking bath (200 rpm) at 37 °C. The change in particle size was monitored over time. Reduction-triggered destabilization of the mPEG-(ss-PhA)2 NPs was observed by detecting the change in size in the presence of DTT in an aqueous medium. The NPs were incubated with 10-mM DTT while being subjected to shaking, and the change in their size was measured over time. Photoactivity of mPEG-(ss-PhA)2 NPs. To evaluate the intramolecular quenching effect, the fluorescence emission spectra of mPEG-(ss-PhA)2 NPs (5 μg/mL) in DMSO, representing the soluble conjugates, were recorded using a fluorescence spectrophotometer (Cary Eclipse, Varian, USA) with an excitation wavelength of 405 nm and compared to those of NPs in DMSO in the presence of 10-mM DTT and in PBS. To assess the intermolecular quenching caused by self-assembly, the fluorescence emission spectrum of NPs in PBS was also measured. The generation of 1O2 by the mPEG-(ss-PhA)2 NPs was monitored using DMA as an 1O2 trap. The NPs were dissolved in DMSO in the absence or presence of 10 mM DTT or in PBS (pH 7.4) and then added to a DMA stock solution to reach a final concentration of 1.184 × 10−2 mM DMA. The solutions containing DMA were irradiated using a 670 nm laser source (LWRL-100 K, Laserwave, China) at a light intensity of 4.2 mW/cm2. The decrease in the fluorescence intensity of DMA (emission from 380 to 550 nm, with excitation at 360 nm) caused by 1O2 generation was monitored using a fluorescence spectrometer. The 1O2 quantum yield (SOQ) was calculated as follows: ϕΔ =
■
RESULTS AND DISCUSSION Synthesis and Physicochemical Characteristics of mPEG-(ss-PhA)2 Conjugates. As shown in Figure 2a, the mPEG-(ss-PhA)2 conjugate was synthesized via a five-step synthesis route. First, the hydroxyl end group of the mPEG was activated by pNC; second, pNC-activated mPEG was coupled with serinol to create a dihydroxylated mPEG; third, the hydroxyl groups of mPEG-serinol was further activated by pNC molecules; forth, disulfide-bond-containing biarmed mPEG-(ssNH2)2 was synthesized by conjugation of cystamine to the chain ends of mPEG-serinol; and finally, the PhA molecules were coupled to the mPEG-(ss-NH2)2 by a conventional carbodiimide reaction. The characteristic peaks of mPEG, pNCactivated mPEG, mPEG-serinol, mPEG-(pNC)2, mPEG-(ssNH2)2, and mPEG-(ss-PhA)2 were observed in the 1H NMR spectra at each step, indicating their successful syntheses (Figure 2b and Figure 3a). The FT-IR spectra of major mPEG derivatives and mPEG-(ss-PhA)2 conjugate were also measured as shown in Figure 2c and 3b. The GPC traces of mPEGserinol, mPEG-(ss-NH2)2, and mPEG-(ss-PhA)2 conjugate revealed a symmetrical elution peak with reasonably narrow polydispersity (1.03−1.05) and the shifts of the chromatograms toward higher molecular weight according to the stage of mPEG derivatization. The number-average MW for mPEG, mPEG-serinol, mPEG-(ss-NH2)2, and mPEG-(ss-PhA)2 conjugates were measured to be 1836, 1954, 2203, and 3054 Da, respectively. Using a concentration-absorbance standard curve for free PhA based on a colorimetric method, the existence of PhA in the mPEG-(ss-PhA)2 conjugate was reconfirmed, and the number of PhA molecules on one mPEG-(ss-PhA)2 molecule
KS A × R × ref · ϕΔ KR AS
where A is the absorbance at 670 nm (the excitation wavelength), K is the gradient of the absorbance curve, and the subscripts S and R refer to the sample and reference, respectively. The SOQ (ΦΔ) of each sample was calculated based on the reference ΦΔ value (free PhA, 0.52) Biological Characterizations of mPEG-(ss-PhA)2 NPs. HeLa cells (a human cervical adenocarcinoma cell line) and MCF7 cells (a human breast adenocarcinoma cell line) were used in this study. HeLa and MCF7 cells were cultured in RPMI 1640 supplemented with 10% FBS and D-glucose (2 g/L). The cells were maintained and grown under humidified air that contained 5% CO2 at 37 °C. Insulin (4 mg/ L) was added to the culture medium for the MCF7 cells. To evaluate the dark toxicity and light-induced toxicity (phototoxicity), the two cell lines were exposed to free PhA and mPEG-(ssPhA)2 NPs. The cells (5000 cells/well) were seeded in a clear 96-well plate with a clear bottom and incubated for 24 h. Next, the cells were treated with free PhA and mPEG-(ss-PhA)2 NPs at various concentrations. At 12 h post-treatment, the culture medium was E
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 4. Physicochemical characteristics of mPEG-(ss-PhA)2 conjugates. (a) UV/visible spectra of free PhA in DMSO and in H2O and of the mPEG-(ss-PhA)2 conjugate in DMSO with and without DTT and in H2O. (b) CQC measurement of mPEG-(ss-PhA)2 conjugates: the fluorescence intensity of pyrene in the mPEG-(ss-PhA)2 conjugate in an aqueous solution with respect to the conjugate concentration. (c) Size distribution of mPEG-(ss-PhA)2 NPs measured using a particle analyzer based on dynamic light scattering. (d) Morphology of mPEG-(ss-PhA)2 NPs observed via FE-SEM.
In general, a typical mono-PS-conjugated PEG has a low content of hydrophobic PS, resulting in limited lipophilicity. For example, a folate-PEG3.4k-chlorin conjugate has been found to exhibit good water solubility (40.1 mg/mL in PBS), causing no obvious aggregation of chlorine molecules.28 Unlike a mono-PS-conjugated PEG, for the unique biarmed structure of the mPEG-(ss-PhA)2 conjugate that was prepared in this study, the self-assembly of NPs of the mPEG-(ss-PhA)2 conjugate was expected because the conjugate had a relatively high PhA loading. Thus, to evaluate whether the mPEG-(ss-PhA)2 conjugate would form self-assembled NPs, the CQC of the mPEG-(ss-PhA)2 conjugate and the size and morphology of the formed mPEG-(ss-PhA)2 NPs were investigated. First, using pyrene as a fluorescence probe, the CQC of the mPEG-(ssPhA)2 conjugate was determined to be 3.78 mg/L, which was the threshold concentration of the conjugates at which selfassembly and the simultaneous fluorescence quenching of pyrene would occur (Figure 4b). It appears that a strong hydrophobic interaction of the mPEG-(ss-PhA)2 conjugates is responsible for the low CQC value, further suggesting that the mPEG-(ss-PhA)2 conjugate could maintain stable NPs under
was calculated to be 1.57. In the UV/visible spectra, the mPEG(ss-PhA)2 conjugate in DMSO had a very similar Q-band absorption at 667 nm to that of free PhA in DMSO, and the Qband absorption of the mPEG-(ss-PhA)2 conjugate was not influenced by the presence of DTT (Figure 4a), implying that the chemical PEGylation of PhA or the presence/absence of the disulfide bonds did not affect the spectrometric characteristics of PhA itself. The Q-band absorption of free PhA in DMSO was shifted toward longer wavelengths (e.g., 683 nm) compared with that of free PhA in water and had a much lower intensity, indicating that free PhA in water exists in its dimeric form.43 The Q-band absorption of the mPEG-(ssPhA)2 NPs in water exhibited a clear bathochromic shift toward 699 nm, which is related to the self-aggregation of PhA molecules in the NPs. However, the absorbance intensity of the mPEG-(ss-PhA)2 NPs in water was much higher than that of free PhA in DMSO and was similar to that of the mPEG-(ssPhA)2 conjugate in DMSO, indicating that the mPEG-(ssPhA)2 NPs effectively enhanced the aqueous solubilization of PhA. F
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 5. (a) Colloidal stability of mPEG-(ss-PhA)2 NPs in PBS and PBS containing 1 mg/mL BSA. (b) Change in the size distribution of mPEG(ss-PhA)2 NPs in PBS before and after incubation with 10 mM DTT.
Figure 6. (a) Fluorescence emission spectra of mPEG-(ss-PhA)2 NPs in DMSO with and without DTT and in PBS. (b) 1O2 generation of free mPEG-(ss-PhA)2 NPs in DMSO with and without DTT and in PBS.
were exposed to harsh reductive conditions (e.g., 10-mM DTT), the NPs exhibited a remarkable change in colloidal stability. For example, when mPEG-(ss-PhA)2 NPs with an initial unimodal size distribution that corresponded to a diameter of approximately 190 nm were incubated in the presence of DTT (10 mM), the size distribution broadened, and the average size was increased to 240 nm after 1 h of incubation (Figure 5b). The NPs continued to grow, and their size distribution became trimodal after 2 h of incubation; the particle size could not be detected after 3 h of incubation. The variations in the particle size and the pattern of the size distribution resulted from the presence of disulfide bonds between the mPEG and PhA molecules, which cause the NPs to be reductively cleavable upon exposure to a reductive environment. These results suggest that mPEG-(ss-PhA)2 NPs may remain stable during blood circulation but then rapidly dissociate and release PhA in a reductive intracellular environment because of the cleavage of the disulfide linkers. Photoactivity of mPEG-(ss-PhA)2 NPs. The photoactivity of the PS, including photoquenching and 1O2 generation, is very significant to the estimation of PDT effects. The photoquenching properties of the mPEG-(ss-PhA)2 NPs were assessed by monitoring their fluorescence emission spectra in DMSO with and without DTT and in PBS. As shown in Figure 6a, no obvious fluorescence emission of the NPs in PBS was observed, implying that the PhA molecules existed in a photoinactive state because of the FRET-based photoquench-
diluted conditions, such as in blood circulation. Second, above the CQC of the amphiphilic mPEG-(ss-PhA)2 conjugate in an aqueous solution, the conjugate readily self-assembled to form NPs with a mean diameter of 189 nm (Figure 4c). The dried mPEG-(ss-PhA)2 NPs were observed using an FE-SEM and were nearly spherical with a diameter of approximately 190 nm (Figure 4d). The size and morphology results suggest that hydrophobic PhA molecules in mPEG-(ss-PhA)2 conjugates self-assemble into spherical NPs with a core−shell structure that consists of an inner hydrophobic PhA core surrounded by a hydrophilic PEG outer shell. The mPEG-(ss-PhA)2 NPs were expected to persist in blood circulation because their “stealth” PEG surface and nanoscale size could avoid phagocytosis and renal clearance. The NPs were also expected to selectively accumulate at tumor sites via an EPR-based passive targeting mechanism because of their nanoscale size. However, for the mPEG-(ss-PhA)2 NPs to achieve their expected extended circulation time and targeted tumor accumulation, the NPs must maintain their colloidal stability in vivo. Thus, the colloidal stability of the mPEG-(ssPhA)2 NPs was investigated in PBS and PBS containing 1 mg/ mL of BSA at 37 °C, respectively. As shown in Figure 5a, the average particle size of the NPs remained unchanged for 5 days in PBS. In addition, by mixing with BSA, the particle size was slightly increased, and then remained unchanged for 5 days, indicating that such NPs have long-term colloidal stability in a physiological condition (e.g., blood). However, when the NPs G
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
ing that was induced by the self-aggregation of PhA molecules via hydrophobic interactions and π−π stacking. However, the NPs in DMSO, which presented as soluble conjugates, exhibited a strong fluorescence emission, indicating that the disassembly of the NPs inhibited the intermolecular quenching, leading to the recovery of the fluorescence emission. Interestingly, when NPs in DMSO were treated with 10 mM DTT, their fluorescence intensity was increased 1.5-fold. This result can be attributed to the cleavage of the disulfide bonds under reductive conditions, which caused the release of free PhA and, consequently, the reduction of the intramolecular quenching between the two PhA molecules in the mPEG-(ssPhA)2 conjugate. In our design concept (Figure 1), it is proposed that the proximity of two PhA molecules at one end of a hydrophilic polymer could create a donor−acceptor pair that could facilitate efficient FRET-based intramolecular photoquenching. However, the coupling ratio of PhA in the mPEG-(ss-PhA)2 conjugate was found to be approximately 1.6, indicating the coexistence of the mPEG-(ss-PhA)2 conjugate and a mono-PhA-conjugated polymer (i.e., an mPEG-ss-PhA conjugate). Thus, although the fluorescence of the two PhA molecules in the mPEG-(ss-PhA)2 conjugate could be quenched, the coexisting mPEG-ss-PhA conjugate could not generate an energy transfer pair and would thus emit strong fluorescence when the excitation beam was applied. This insufficient intramolecular quenching of the conjugate mixture could suppress the photoactivity of PhA. Nevertheless, it was observed that additional intermolecular quenching, which arose from the spontaneous self-assembly of the conjugate mixture, could be effective in controlling the photoactivity of the mPEG(ss-PhA)2 NPs. Furthermore, the rapid cleavage of the disulfide bonds between the mPEG and PhA molecules could accelerate the dissociation of the NPs, which may lead to the efficient recovery of the photoactivity in a reductive environment. The generation of 1O2, which governs the phototoxicity, by mPEG-(ss-PhA)2 NPs was observed in DMSO with and without DTT and in PBS based on the calculation of the SOQ using DMA as a 1O2 trap (Figure 6b). The mPEG-(ss-PhA)2 NPs exhibited a sharp decline in DMA fluorescence intensity in DMSO because of the rapid 1O2 generation that occurred upon light exposure. The SOQ of the NPs in the presence of DTT was calculated to be 0.51, which was slightly higher than that of the NPs under nonreductive conditions (SOQ, 0.49) because of the reduced intramolecular quenching. On the other hand, no obvious decrease in the fluorescence intensity of DMA was observed for the NPs in PBS (SOQ, 0.03), indicating that the generation of 1O2 was dramatically suppressed because of the efficient intramolecular and intermolecular quenching. As a result, it can be concluded that the mPEG-(ss-PhA)2 NPs could not be activated into the triplet state via mutual energy transfer because of the efficient self-quenching and could inhibit the ability of PhA to generate 1O2. The self-quenching effect of the mPEG-(ss-PhA)2 NPs can improve the therapeutic efficiency of PDT by suppressing the phototoxicity of the PDT agent in the bloodstream. Cytotoxicity and Cellular Characteristics of mPEG-(ssPhA)2 NPs. The phototoxicity of PhA delivered by mPEG-(ssPhA)2 NPs was evaluated using MCF7 and HeLa cells and compared with that of free PhA. At the PhA concentrations used in this study, the dark toxicities of free PhA and the mPEG-(ss-PhA)2 NPs did not represent the IC50 (the drug concentration that causes 50% growth inhibition) values for MCF7 cells. However, when free PhA- and mPEG-(ss-PhA)2
NP-treated MCF7 cells were irradiated, cell viability dramatically decreased according to the irradiation time (Figure 7a,b).
Figure 7. Cytotoxicity of (a) free PhA, (b) mPEG-(ss-PhA)2 NPs according to light irradiation time in MCF7 cells, and (c) dark toxicity and phototoxicity of free PhA and mPEG-(ss-PhA)2 NPs in HeLa cells with 63 s light irradiation (n ≥ 6, mean ± standard error).
The IC50 of free PhA and mPEG-(ss-PhA)2 NPs decreased from 0.32 to 0.033 μg/mL and 4.3 to 0.27 μg/mL, respectively, when the irradiation time increased from 16 to 252 s. In addition, we also observed no obvious cytotoxicity in free PhAand mPEG-(ss-PhA)2 NP-treated HeLa cells that maintained in darkness, whereas the phototoxicities of the mPEG-(ss-PhA)2 NPs were 0.16 μg/mL based on the IC50 values and represented 3.9-fold decreases compared with those of free PhA (0.041 μg/mL) when the cells were exposed to light for 63 s. The significant phototoxicity of the mPEG-(ss-PhA)2 NPs implies that the photoactivity of the NPs that was suppressed by photoquenching was restored in living cells, which resulted from a series of dequenching processes caused by the cleavage of the disulfide bonds and dissociation of the self-assembled nanostructure. It is worthy to note that phototoxicity of mPEG(ss-PhA)2 NPs was lower than free PhA for all irradiation times H
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 8. Representative cellular-uptake histogram of free PhA and mPEG-(ss-PhA)2 NPs in (a) MCF7 and (b) HeLa cells at 4 h postincubation. The concentration of free PhA or PhA in the mPEG-(ss-PhA)2 NPs was 0.5 μg/mL.
between the phototoxicities of free PhA and the NPs in the cells. In addition, the results further suggest that mPEG-(ssPhA)2 NPs are a potential delivery system for the poorly watersoluble PhA. To understand how the mPEG-(ss-PhA)2 NPs produce their phototoxic effects in cells, the cell-culture media were supplemented with GSHOEt during the phototoxicity studies because GSHOEt increases the concentration of intracellular thiols. When free-PhA-uptaken MCF7 cells were treated with 2 mM GSHOEt, the GSHOEt did not influence the phototoxicity of the free PhA (Figure 9). This result indicates that the phototoxicity of PhA is not affected by the presence of intracellular thiols. However, in light-exposed MCF7 cells and HeLa cells, mPEG-(ss-PhA)2 NPs in the presence of GSHOEt (IC50 values of 0.20 μg/mL and 0.11 μg/ mL for MCF7 cells and HeLa cells, respectively) produced an approximately 2.7-fold and 1.3-fold higher phototoxicity than that observed in the absence of GSHOEt (IC50 values of 0.53 μg/mL and 0.14 μg/mL for MCF7 cells and HeLa cells, respectively; Figure 9). This finding indicates that the increased level of thiols could more quickly break the disulfide linkages in the mPEG-(ss-PhA)2 NPs, thereby facilitating the release of PhA from the NPs and increasing their phototoxicity. The intracellular localization and intracellular degradation of the mPEG-(ss-PhA)2 NPs were further investigated using confocal microscopy (Figure 10). The fluorescence intensity of free PhA originated mostly from the cytoplasm but also somewhat from
and in both MCF7 and HeLa cells. That is likely related to the different cellular uptake efficiency between free PhA and mPEG-(ss-PhA)2 NPs, because the cell-killing activity of an antitumor drug is correlated with its cellular uptake. Therefore, for further analysis of the therapeutic efficacy of the NPs, the cellular uptake of the mPEG-(ss-PhA)2 NPs was compared with that of free PhA in the two cell lines. After the cells were incubated with either free PhA or mPEG-(ss-PhA)2 NPs for 4 h, the intracellular fluorescence intensities of PhA were monitored via flow cytometry. The NPs presented approximately 4.7-fold and 3.4-fold lower fluorescence intensities (i.e., cellular uptakes) than those of free PhA in MCF7 and HeLa cells, respectively (Figure 8). The lower cellular uptake of the NPs might be attributable to a weak interaction between the cellular membrane and the PEG-decorated NPs, unlike the moderate interaction of hydrophobic PhA with the cellular membrane, and such a difference in interaction strength would further lead to different endocytosis mechanisms (i.e., fluidphase endocytosis vs adsorptive endocytosis). Nevertheless, if the same amount of PhA were internalized, the phototoxicity of the mPEG-(ss-PhA)2 NPs would be very similar to that of free PhA because the former would be only approximately 1.5-fold or 1.2-fold lower than the latter in MCF7 or HeLa cells, respectively. This estimation indicates that the difference between the amounts of free PhA and mPEG-(ss-PhA)2 NPs internalized in the cells may be responsible for the difference I
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
fluorescence intensity, and increased phototoxicity of the PhA delivered by the NPs. In the current stage, the mPEG-(ss-PhA)2 NPs have shown its perspective for controllable PDT with cancer cells in the cell culture system. Further studies will be focused to evaluate the tumor specificity and PDT efficacy of the designed mPEG-(ssPhA)2 NPs in vivo and to further explore their practical applications in clinical PDT.
■
CONCLUSION
In this study, a novel bioreducible mPEG-(ss-PhA)2 conjugate was synthesized through multiple reaction steps and readily self-assembled to construct stable core−shell-structured NPs under aqueous conditions. The thiol-sensitive disulfide linkers introduced between the mPEG and PhA molecules accelerated the intracellular release of PhA. The self-assembled NPs could suppress the photoactivity of the PhA molecules through FRET-based self-quenching and modulated 1O2 generation. However, when the NPs were exposed to a reductive environment, the cleavage of the disulfide bonds facilitated the dissociation of the NP structure and caused rapid release of PhA, and as a result, the photoactivity of the PhA was effectively restored. In vitro cell studies revealed that the mPEG-(ss-PhA)2 NPs exhibited significant phototoxicity in cancer cells, and the cytotoxicity was lower than that of free PhA because of the relatively low cellular uptake. However, enhanced phototoxicity of the NPs was observed when the expression of intracellular thiols in the cytoplasm was increased. These findings suggest that the bioreducible mPEG-(ss-PhA)2 NPs could provide a potential implementation of controllable 1 O2 generation within cancer cells, which could lead to more effective PDT with minimal phototoxicity in normal tissue and enhanced efficacy for photodynamic cancer treatment.
Figure 9. Effect of GSHOEt on the phototoxicity of free PhA and mPEG-(ss-PhA)2 NPs in (a) MCF7 cells and (b) HeLa cells with 63 s light irradiation (n ≥ 6, mean ± standard error).
the nucleus, and it remained similar, regardless of the presence of GSHOEt. Although the fluorescence intensity of the PhA delivered by the mPEG-(ss-PhA)2 NPs also originated from the cytoplasm and nucleus, the fluorescence intensity in this case was increased by the addition of GSHOEt (Figure 10). These results indicate that the presence of intracellular thiols in the cytoplasm breaks the disulfide linkages in the mPEG-(ss-PhA)2 NPs, resulting in the increased cytoplasmic release, increased
Figure 10. Intracellular localization and fluorescence intensity of PhA in free-PhA-uptaken and mPEG-(ss-PhA)2 NP-uptaken MCF7 cells in the absence or presence of GSHOEt (2 mM). The concentration of free PhA or PhA in the mPEG-(ss-PhA)2 NPs was 0.5 or 4 μg/mL, respectively. J
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
■
Article
(25) Ohya, Y.; Shirakawa, S.; Matsumoto, M.; Ouchi, T. Polym. Adv. Technol. 2000, 11, 635−641. (26) Ding, Y.; Zhou, Y. Y.; Chen, H.; Geng, D. D.; Wu, D. Y.; Hong, J.; Shen, W. B.; Hang, T. J.; Zhang, C. Biomaterials 2013, 34, 10217− 10227. (27) Wermeckes, B.; Hess, M.; Dehne, S.; Jo, B. W.; Sohn, J. S. Mater. Res. Innov. 2006, 10, 88−90. (28) Li, D. H.; Li, P. X.; Lin, H. Y.; Jiang, Z. L.; Guo, L. F.; Li, B. H. J. Photochem. Photobiol. B 2013, 127, 28−37. (29) Howard, M. D.; Jay, M.; Dziublal, T. D.; Lu, X. L. J. Biomed. Nanotechnol. 2008, 4, 133−148. (30) Veronese, F. M.; Pasut, G. Drug Discovery Today 2005, 10, 1451−1458. (31) Verhille, M.; Couleaud, P.; Vanderesse, R.; Brault, D.; BarberiHeyob, M.; Frochot, C. Curr. Med. Chem. 2010, 17, 3925−3943. (32) Lovell, J. F.; Liu, T. W. B.; Chen, J.; Zheng, G. Chem. Rev. 2010, 110, 2839−2857. (33) Allison, R. R.; Mota, H. C.; Bagnato, V. S.; Sibata, C. H. Photodiagn. Photodyn. 2008, 5, 19−28. (34) Fernandez, J. M.; Bilgin, M. D.; Grossweiner, L. I. J. Photochem. Photobiol. B 1997, 37, 131−140. (35) Campo, M. A.; Gabriel, D.; Kucera, P.; Gurny, R.; Lange, N. Photochem. Photobiol. 2007, 83, 958−965. (36) Gamcsik, M. P.; Kasibhatla, M. S.; Teeter, S. D.; Colvin, O. M. Biomarkers 2012, 17, 671−691. (37) Cheng, R.; Feng, F.; Meng, F. H.; Deng, C.; Feijen, J.; Zhong, Z. Y. J. Controlled Release 2011, 152, 2−12. (38) He, H.; Lo, P.-C.; Ng, D. K. P. Chem.Eur. J. 2014, 20, 1−6. (39) Isik, M.; Ozdemir, T.; Turan, I. S.; Kolemen, S.; Akkaya, E. U. Org. Lett. 2013, 15, 216−219. (40) Park, S. Y.; Oh, K. T.; Oh, Y. T.; Oh, N. M.; Youn, Y. S.; Lee, E. S. Chem. Commun. 2013, 48, 2522−2524. (41) Kim, H.; Mun, S.; Choi, Y. J. Mater. Chem. B 2013, 1, 429−431. (42) Li, L.; Nurunnabi, M.; Nafiujjaman, M.; Jeong, Y. Y.; Lee, Y. K.; Huh, K. M. J. Mater. Chem. B 2014, 2, 2929−2937. (43) Eichwurzel, I.; Stiel, H.; Roder, B. J. Photochem. Photobiol. B 2000, 54, 194−200.
AUTHOR INFORMATION
Corresponding Authors
*Tel.: +82-42-821-6663. Fax: +82-42-821-8910. E-mail: khuh@ cnu.ac.kr. *Tel.: +82-2-2164-6533. Fax: +82-2-2164-4059. E-mail: [email protected]. Author Contributions †
These authors contributed equally to this work (W.L.K. and H.C.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF-2012R1A1A2005029 and NRF2013R1A2A2A04015914). H.C.K. appreciates Research Fund of The Catholic University of Korea (2012).
■
REFERENCES
(1) Juarranz, A.; Jaen, P.; Sanz-Rodriguez, F.; Cuevas, J.; Gonzalez, S. Clin. Transl. Oncol. 2008, 10, 148−154. (2) MacDonald, I. J.; Dougherty, T. J. J. Porphyrins Phthalocyanines 2001, 5, 105−129. (3) Mitton, D.; Ackroyd, R. Photodiagn. Photodyn. 2008, 5, 103−111. (4) Ibbotson, S. H. Photodiagn. Photodyn. 2010, 7, 16−23. (5) Ortner, M. A. Curr. Opin. Gastroen. 2009, 25, 472−476. (6) Kohl, E. A.; Karrer, S. G. Ital. Dermatol. Venereol. 2011, 146, 473− 485. (7) Vrouenraets, M. B.; Visser, G. W.; Snow, G. B.; van Dongen, G. A. Anticancer Res. 2003, 23, 505−22. (8) Detty, M. R.; Gibson, S. L.; Wagner, S. J. J. Med. Chem. 2004, 47, 3897−3915. (9) Gravier, J.; Schneider, R.; Frochot, C.; Bastogne, T.; Schmitt, F.; Didelon, J.; Guillemin, F.; Barberi-Heyob, M. J. Med. Chem. 2008, 51, 3867−77. (10) Huang, P.; Xu, C.; Lin, J.; Wang, C.; Wang, X. S.; Zhang, C. L.; Zhou, X. J.; Guo, S. W.; Cui, D. X. Theranostics 2011, 1, 240−250. (11) Ranyuk, E.; Cauchon, N.; Klarskov, K.; Guerin, B.; van Lier, J. E. J. Med. Chem. 2013, 56, 1520−1534. (12) Mukai, H.; Wada, Y.; Watanabe, Y. Ann. Nucl. Med. 2013, 27, 625−639. (13) Toneatto, J.; Garcia, P. F.; Arguello, G. A. J. Inorg. Biochem. 2011, 105, 1299−1305. (14) Allen, C. M.; Sharman, W. M.; La Madeleine, C.; Weber, J. M.; Langlois, R.; Ouellet, R.; van Lier, J. E. Photochem. Photobiol. 1999, 70, 512−523. (15) Lee, D. J.; Park, S. Y.; Oh, Y. T.; Oh, N. M.; Oh, K. T.; Youn, Y. S.; Lee, E. S. Macromol. Res. 2011, 19, 848−852. (16) Li, L.; Bae, B. C.; Tran, T. H.; Yoon, K. H.; Na, K.; Huh, K. M. Carbohydr. Polym. 2011, 86, 708−715. (17) Oh, I. H.; Min, H. S.; Li, L.; Tran, T. H.; Lee, Y. K.; Kwon, I. C.; Choi, K.; Kim, K.; Huh, K. M. Biomaterials 2013, 34, 6454−6463. (18) Li, L.; Nurunnabi, M.; Nafiujjaman, M.; Lee, Y. K.; Huh, K. M. J. Controlled Release 2013, 171, 241−250. (19) Greenwald, R. B.; Choe, Y. H.; McGuire, J.; Conover, C. D. Adv. Drug Delivery Rev. 2003, 55, 217−250. (20) Knop, K.; Hoogenboom, R.; Fischer, D.; Schubert, U. S. Angew. Chem., Int. Ed. 2010, 49, 6288−6308. (21) Joralemon, M. J.; Mcrae, S.; Emrick, T. Chem. Commun. 2010, 46, 1377−1393. (22) Bruckdorfer, T. Biopolymers 2009, 92, 337−337. (23) Yoo, H. S.; Park, T. G. J. Controlled Release 2004, 100, 247−256. (24) Ye, W. L.; Zhou, S. Y. Acta Pharmacol. Sin. 2013, 34, 18−18. K
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biarmed Poly(ethylene glycol)-(pheophorbide a)2 Conjugate as a Bioactivatable Delivery Carrier for Photodynamic Therapy Wool Lim Kim,†,‡ Hana Cho,†,§ Li Li,‡ Han Chang Kang,*,§ and Kang Moo Huh*,‡ ‡
Department of Polymer Science and Engineering, Chungnam National University, 99 Daehak-ro, Yuseong-gu, Daejeon 305-764, Republic of Korea § Department of Pharmacy and Integrated Research Institute of Pharmaceutical Sciences, College of Pharmacy, The Catholic University of Korea, 43 Jibong-ro, Wonmi-gu, Bucheon-si, Gyeonggi-do 420-743, Republic of Korea ABSTRACT: In the study presented here, we developed a bioreducible biarmed methoxy poly(ethylene glycol)-(pheophorbide a)2 (mPEG-(ss-PhA)2) conjugate for cancer-cellspecific photodynamic therapy (PDT). PhA molecules were chemically conjugated with biarmed linkages at one end of the mPEG molecule via disulfide bonds. Under aqueous conditions, the amphiphilic mPEG-(ss-PhA)2 conjugate selfassembled to form core−shell-structured nanoparticles (NPs) with good colloidal stability. The mPEG-(ss-PhA)2 NPs exhibited intramolecular and intermolecular self-quenching effects that enabled the NPs to remain photoinactive in a physiological buffer. However, the dissociation of the NP structure was effectively induced by the cleavage of the disulfide bonds in response to intracellular reductive conditions, triggering the rapid release of PhA molecules in a photoactive form. In cellculture systems, in addition to significant phototoxicity and intracellular uptake, we observed that the dequenching processes of PhA in the mPEG-(ss-PhA)2 NPs highly depended on the expression of intracellular thiols and that supplementation with glutathione monoethylester facilitated more rapid PhA release and enhanced the PhA phototoxicity. These findings suggest that the bioreducible activation mechanism of mPEG-(ss-PhA)2 NPs in cancer cells can maximize the cytosolic dose of active photosensitizers to achieve high cytotoxicity, thereby enhancing the treatment efficacy of photodynamic cancer treatment.
■
INTRODUCTION Photodynamic therapy (PDT) is a promising medical technology for the treatment of cancers and other nonmalignant diseases.1−6 PDT is based on a photochemical reaction that mediates interactions among chemical photosensitizers (PSs), light, and molecular oxygen. The lightinduced activation of PSs results in the in situ production of reactive singlet oxygen (1O2), which can damage cell viability.2 Nevertheless, clinical applications of PDT agents, such as porphyrin and phthalocyanine derivatives, have remained limited because of the poor water solubility and unfavorable pharmacokinetic properties of such agents in addition to the occurrence of post-treatment cutaneous photosensitivity.7,8 To reduce these limitations, one potential strategy is to chemically link hydrophobic PSs with various water-soluble, biofunctional molecules (e.g., tumor-targeting ligands or peptides,9−12 carrier proteins,13,14 carbohydrates,10,15 and hydrophilic biomacromolecules16−18) for the preparation of water-soluble or waterdispersible PS conjugates and nanoparticles (NPs). Poly(ethylene glycol) (PEG) is a nontoxic, nonimmunogenic, nonantigenic, and water-soluble polymer that is approved by the Food and Drug Administration as a safe excipient for drug delivery.19,20 PEGylation is one of the leading strategies for prolonging the in vivo circulation time of therapeutics by means of reduced phagocytosis and renal clearance.21,22 In © XXXX American Chemical Society
cancer therapy specifically, PEG conjugations with chemical drugs (e.g., doxorubicin,23,24 cis-platinum,25 taxol,26,27 and PDT agents28) are an attractive strategy for improving drug delivery by enhancing aqueous solubility, reducing renal clearance, decreasing enzymatic degradation, and limiting immunogenic and antigenic reactions. PEGylated platforms also exhibit a superior capacity for passive tumor targeting through the enhanced permeability and retention (EPR) effect because PEGylated formulations are capable of remaining in the circulatory system for an extended period of time and thus provide a sufficient level of accumulation in the pathological area (e.g., a solid tumor).29,30 On the other hand, the long circulation that is characteristic of PEG−PS conjugates must be carefully monitored because it carries a risk of causing longlasting skin photosensitivity, as the PEG−PS conjugates are photodynamically active throughout the postadministration period.31,32 In addition, in PEG−PS conjugates, the hydrophilic, repellant nature of PEG may negatively impact the intracellular distribution of PS at the target site and may also pose a diffusion barrier that negatively affects the release of 1O2 generated under light exposure33 because the irreversible Received: March 7, 2014 Revised: May 4, 2014
A
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 1. Conceptual illustration of the mechanism of bioreducible mPEG-(ss-PhA)2 NPs for the switchable photoactivity of PhA: mPEG-(ss-PhA)2 conjugates readily form self-assembled NPs under aqueous conditions. The NPs do not exhibit photoactivity during blood circulation because of the intramolecular and intermolecular self-quenching of PhA. When the NPs are internalized by cancer cells, the intracellular reduction-triggered cleavage of the disulfide bonds accelerates the dissociation of the NP structure and causes the rapid release of PhA, thereby restoring the photoactivity of the PS.
Second, the mPEG-(ss-PS)2 NPs should accumulate at tumor sites because of the EPR effect. Third, after the internalization of the mPEG-(ss-PS)2 NPs by the tumor cells, the thiolresponsive disulfide linkers in the NPs should be rapidly cleaved in the cytosol, as the concentrations of glutathione (GSH) in the cytosol (approximately 2−10 mM) are much higher than those in extracellular fluids (approximately 2−20 μM).36,37 The remarkable difference in GSH levels between the two environments should trigger the rapid dissociation of the mPEG-(ss-PS)2 NPs, followed by the rapid release of PS molecules into the intracellular environment, resulting in efficient photoactivity of the PS molecules. Recently, several GSH-responsive activatable photosensizing systems were reported, such as 2,4-dinitrobenzenesulfonate-substituted zinc(II) phthalocyanine,38 nitroethenyl-BODIPY conjugate,39 graphene oxide-ss-chlroin e6 conjugate,40 and hyaluronic acid-sschlroin e6 conjugates.41 Our group also has developed various activatable PS formulations regarding GSH-mediated activation of PS, including glycol chitosan-ss-pheophorbide a (PhA) conjugate,17 heparin-PhA-conjugated gold NP,18 and heparinPhA-conjugated iron oxide/gold core/shell NP.42 These GSHresponsive formulations showed superior capacity for selective release and efficient activation of the PSs in tumor cells, which might maximize the cytosolic dose of active PSs to achieve higher cytotoxicity, thereby enhancing the treatment efficacy. Therefore, we expect self-quenchable mPEG-(ss-PS)2 NPs with rapid intracellular-release kinetics and a bioreducible activation mechanism should be capable of engaging in highly selective cancer treatment using the PDT approach. Thus, the intent of this work was to synthesize such a bioreducible mPEG-(ss-PS)2
photodamage to pivotal biomacromolecules and other cellular components essentially depends on the amount and intracellular distribution of 1O2, which, in combination with the limited diffusion distance and short lifetime of 1O2,34 may significantly reduce the therapeutic efficacy of PDT. To develop an elegant PEG−PS conjugate with a highly potent PDT effect, we designed an activatable PDT agent known as a bioreducible mPEG-(ss-PS)2 conjugate, which has a methoxy PEG (mPEG) with two arms at one end, each of which is chemically linked to one PS molecule via a disulfide bond. It was expected that the designed mPEG-(ss-PS)2 conjugate and its self-assembled NPs should possess the aforementioned typical characteristics of PEGylated drugdelivery carriers. Specifically, the mPEG-(ss-PS)2 conjugate and its NPs were expected to display several characteristic features, including potent cancer-cell-specific PDT effects and minimal traditional PDT complications because of the design strategy of site-specific controllable photoactivity (Figure 1). First, in the mPEG-(ss-PS)2 conjugate, the proximity of two PS molecules should create a donor−acceptor pair that enables fluorescence resonance energy transfer (FRET)-mediated intramolecular photoquenching.35 In addition, the amphiphilic nature of the mPEG-(ss-PS)2 conjugate should allow it to readily self-assemble to form core−shell-structured NPs, resulting in π−π-stacked self-aggregation among hydrophobic PS molecules in the NPs to form FRET-based donor−acceptor pairs, which should lead to efficient intermolecular quenching.16 These desired characteristics should allow the PS molecules to remain photodynamically inactive in the mPEG-(ss-PS)2 NPs during systemic circulation to inhibit harmful 1O2 generation. B
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 2. (a) Overall synthesis scheme for the mPEG-(ss-PhA)2 conjugate, (b) the 1H NMR spectra of mPEG derivatives obtained during mPEG(ss-PhA)2 conjugation, and (c) FT-IR spectra of mPEG, mPEG-serinol, and mPEG-(ss-NH2)2.
conjugate using PhA as a model PS and to develop selfquenchable and bioactivatable mPEG-(ss-PhA)2 NPs. The conjugate and the NPs were investigated with regard to their
physicochemical characteristics, self-assembling behaviors, photoactivity, and in vitro PDT efficacy in tumor cell culture systems. C
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
■
Article
low-molecular-weight impurities. The mPEG-(ss-PhA)2 conjugate was obtained in powder form via lyophilization (yield: 69%). 1H NMR: PEG segment, δ = 2.6 (2 H, α-CH2-), 2.8 (2 H, β-CH2-), 3.4 (3 H, CH3-), and 3.5−3.7 (196 H, -CH2-CH2-O-); PhA moiety, 1.2 (3 H, -C11H3), 1.6 (3 H, -C4H3), 1.8 (2 H, -C1H2-), 2.1 (2 H, -C2H2-), 3.1 (3 H, -C6H3), 3.2 3.2 (3 H, -C13H3), 3.8 (3 H, -C10H3), 4.0 (3 H, -C15H3), 4.2 (1 H, -C3H-), 6.1 (2 H, =C8H2), 6.3 (1 H, -C14H-), 7.9 (1 H, -C7H), 8.9 (1 H, -C12H), 9.2 (1 H, -C9H), and 9.7 (1 H, -C5H) (Figure 3a). FT-IR (KBr) v/cm−1: PEG segment, 1098 (C−
MATERIALS AND METHODS
Materials. Methoxy poly(ethylene glycol) (mPEG, MW 2000 Da), cystamine dihydrochloride, p-nitrophenyl chloroformate (pNC), dithiothreitol (DTT), N-(3-(dimethylamino)propyl)-N′-ethylcarbodiimide hydrochloride (EDC), N-hydroxysuccinimide (NHS), triethylamine (TEA), acetonitrile, dimethyl sulfoxide (DMSO), dichloromethane (DCM), 9,10-dimethylanthracene (DMA), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), RPMI 1640 medium, Ca2+-free and Mg2+-free Dulbecco’s phosphate-buffered saline (DPBS), 4-(2-hydroxy-ethyl)-1-piperazine (HEPES), D-glucose, sodium bicarbonate, glutathione monoethylester (GSHOEt), recombinant human insulin, fetal bovine serum (FBS), penicillin-streptomycin antibiotics, bovine serum albumin (BSA), and trypsin-EDTA solution were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). 2-Amino1,3-propanediol (serinol) and pheophorbide a (PhA) were obtained from Tokyo Chemical Industry (Tokyo, Japan) and Frontier Scientific, Inc. (Logan, Utah, U.S.A.), respectively. Spectra/Por membranes were purchased from Spectrum Laboratories, Inc. (Rancho Dominguez, CA, U.S.A.). All chemicals that were used for polymer synthesis were at least analytical grade and were used without further purification. Synthesis of Bioreducible mPEG-(ss-PhA)2 Conjugate. As shown in Figure 2a, a bioreducible mPEG-(ss-PhA)2 conjugate was synthesized via a multistep synthetic route. First, pNC-activated mPEG was synthesized from mPEG and pNC. In brief, mPEG (10 g, 5 mmol) was dried in a vacuum at 70 °C for 12 h and was then dissolved in 20 mL of acetonitrile and activated by pNC (12.60 g, 62.5 mmol) in the presence of TEA at room temperature (RT) under nitrogen atmosphere for 24 h. The pNC-activated mPEG solution was filtered to remove the triethylammonium chloride that had precipitated out, and the synthesized pNC-activated mPEG was then precipitated in diethyl ether and dried in a vacuum. The synthesis of pNC-activated mPEG was confirmed via 1H NMR (JNM-AL 400 spectrometer, JEOL Ltd., Akishima, Japan; 400 MHz, CDCl3): δ = 3.4 (3 H, CH3-), 3.5− 3.7 (196 H, -CH2-CH2-O-), 7.4 (2 H, Ar-H), and 8.2 (2 H, Ar-H) (Figure 2b). Second, for the synthesis of mPEG-serinol, pNC-activated mPEG (8.50 g, 3.93 mmol) was dissolved in DMSO and reacted with serinol (2.86 g, 31.41 mmol) at RT under nitrogen atmosphere for 36 h, followed by dialysis using a dialysis membrane (molecular-weight cutoff (MWCO) 1000 Da) against deionized water for 2 days and then lyophilization. The crude product was dissolved in DCM and filtered to remove unreacted serinol, followed by precipitation in cold diethyl ether and drying in a vacuum. 1H NMR (400 MHz, CDCl3): δ = 3.4 (3 H, CH3-), 3.5−3.7 (196 H, -CH2-CH2-O-), and 5.5 (1 H, -NH-) (Figure 2b); FT-IR (Magna 560, Nicolet, U.S.A.; KBr) v/cm−1: 1098 (C−O, ether), 1735 (CO, ester), and 2930 (C−H, methylene) (Figure 2c). Third, the activation of the hydroxyl end groups of mPEG-serinol using pNC was conducted under conditions similar to those described above for the synthesis of the mPEG-serinol. Then, mPEG-(pNC)2 was reacted with cystamine dihydrochloride in DMSO in the presence of TEA at RT for 24 h. The resultant solution was filtered and precipitated in cold diethyl ether, followed by drying in a vacuum. The product was dissolved in DCM to remove unreacted cystamine and extracted three times with 60 mL of saturated NaCl solution. The final product, mPEG-(ss-NH2)2, was collected after precipitation in cold diethyl ether and dried in a vacuum. 1H NMR (400 MHz, CDCl3): δ = 2.6 (2 H, α-CH2-), 2.8 (2 H, β-CH2-), 3.4 (3 H, CH3-), 3.5−3.7 (196 H, -CH2-CH2-O−), and 5.3 (1 H, -NH-) (Figure 2b); FT-IR (KBr) v/ cm−1: 1098 (C−O, ether), 1400 (C−N, amide), 1735 (CO, ester), and 2930 (C−H, methylene) (Figure 2c). Fourth, to synthesize the mPEG-(ss-PhA)2 conjugate, PhA (60 mg, 0.1 mmol) was dissolved in 12 mL of DMSO and activated using EDC (97 mg, 0.506 mmol) and NHS (58 mg, 0.506 mmol). After the NHSactivated PhA solution had reacted for 12 h, mPEG-(ss-NH2)2 (80 mg, 0.033 mmol) was added to the solution. The reaction mixture was stirred for an additional 12 h under nitrogen atmosphere and then dialyzed (MWCO 1000 Da), first against DMSO for 2 days and then against deionized water for 1 day, to remove unreacted PhA and other
Figure 3. 1H NMR and FT-IR spectra of the mPEG-(ss-PhA)2 conjugate. O, ether), 1400 (C−N, amide), 1735 (CO, ester), and 2930 (C−H, methylene); PhA moiety, 990, 1235, 1370, 1470, 1580 (chlorin skeleton), 1680 (CC), and 1706 (CO, keto group in exocycle) (Figure 3b). Characterization of the mPEG-(ss-PhA)2 Conjugate. The GPC traces of main PEG derivatives and mPEG-(ss-PhA)2 conjugate were recored using by a Agilent 1100 gel permeation chromatography system (Agilent Technologies, CA, U.S.A.) equipped with quaternary pump and refractive index detector using Agilent PL gel 5 μm mixed-C and mixed-D (300 mm × 7.5 mm) columns. The eluent was THF, and the flow rate was 1.0 mL/min. The temperature of columns was set and maintained at 40 °C. The number of PhA molecules conjugated on the mPEG backbone was determined using a colorimetric method. In brief, the lyophilized mPEG-(ss-PhA)2 conjugate was dissolved in 10 mL of DMF at various concentrations. The absorbance of the mPEG-(ss-PhA)2 conjugate at 667 nm was determined using a UV/ visible spectrometer (SINCO PDA, UV/visible spectrometer operation, Korea). The PhA content in the mPEG-(ss-PhA)2 conjugate was calculated based on a calibration curve for PhA (667 nm) in DMF. D
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
changed, and the cells were irradiated using a 670 nm light source. MCF7 cells were irradiated at 8.67 mW/cm2 during different irradiation times (16, 32, 63, 126, and 252 s), and HeLa cells was irradiated with light at 1.7 J/cm2 (8.67 mW, 63 s) to evaluate the lightinduced toxicity. At 36 h post-treatment (i.e., 24 h after light exposure), a MTT solution (0.5 mg/mL) was added to the culture medium, and the cells were incubated for an additional 4 h. After the MTT-containing medium was removed, the formazan crystals produced by the live cells were dissolved in DMSO (100 μL/well), and the absorbance at 570 nm was measured. In addition, to evaluate the effect of GSHOEt on the phototoxicity of mPEG-(ss-PhA)2 NPs, GSHOEt (2 mM) was added to the NP culture medium at 2 h posttreatment. To assess the cellular uptake of free PhA and mPEG-(ss-PhA)2 NPs, the cells (5 × 105 cells/well) were seeded in a 6-well plate and incubated for 24 h. After a 4 h treatment with PhA and mPEG-(ssPhA)2 ([PhA] = 0.5 μg/mL), the cells were rinsed twice with DPBS and detached. The cellular uptake of PhA was measured using a flow cytometer (FACScanto II, Becton−Dickinson, Franklin Lakes, NJ, U.S.A.) with a primary HeNe laser (633 nm) and a fluorescence detector (660 ± 10 nm). The PhA uptake of the cells was analyzed based on a gated viable population of at least 1 × 104 cells. To measure the intracellular localization and intracellular intensity of PhA delivered by the mPEG-(ss-PhA)2 NPs, MCF7 cells (1.5 × 104 cells/well) were seeded in an eight-chamber slide and incubated for 24 h. Before treatment with the NPs, the culture medium was replaced with fresh medium (0.2 mL). The cells were exposed to the NPs for 6 h. To assess the effect of GSHOEt, the cells were exposed to the NPs for 2 h and subsequently incubated in the presence of GSHOEt for an additional 4 h. After the 6 h incubation, the cells were rinsed with DPBS for further evaluation using a laser scanning confocal microscope (LSM710; Carl Zeiss, Oberkochen, Germany) equipped with excitation lasers (633 nm for HeNe) and variable band-pass emission filters.
In addition, the critical self-quenching concentration (CQC) was determined via fluorescence spectrometry using pyrene as a fluorescence probe. mPEG-(ss-PhA)2 solutions at a variety of concentrations ranging from 1 × 10−5 to 0.2 mg/mL were prepared and mixed with pyrene solutions that contained 6.0 × 10−7 M pyrene in each vial. Then, the fluorescence excitation spectra of pyrene were recorded using a fluorescence spectrophotometer (Cary Eclipse, Varian, U.S.A.) with an emission wavelength of 390 nm. Preparation and Characterization of Self-Assembled mPEG(ss-PhA)2 NPs. The mPEG-(ss-PhA)2 NPs were prepared from the mPEG-(ss-PhA)2 conjugate using a dialysis method. In brief, the conjugate was dissolved in DMF, followed by sonication for 10 min to obtain a clear solution. The solution was transferred into a dialysis membrane (MWCO 1000 Da) and dialyzed against deionized water for 24 h. The size and size distribution of the mPEG-(ss-PhA)2 NPs were determined using a particle size analyzer (ELS-Z, Photal, Japan). The morphology of the NPs was characterized using a field-emission scanning electron microscope (FE-SEM, JSM-6700, JEOL Ltd., Japan) at 10 keV. To assess the colloidal stability of the mPEG-(ss-PhA)2 NPs, NPs in PBS (0.5 mg/mL) and PBS containing 1 mg/mL of BSA were prepared, respectively, and the solutions were incubated in a shaking bath (200 rpm) at 37 °C. The change in particle size was monitored over time. Reduction-triggered destabilization of the mPEG-(ss-PhA)2 NPs was observed by detecting the change in size in the presence of DTT in an aqueous medium. The NPs were incubated with 10-mM DTT while being subjected to shaking, and the change in their size was measured over time. Photoactivity of mPEG-(ss-PhA)2 NPs. To evaluate the intramolecular quenching effect, the fluorescence emission spectra of mPEG-(ss-PhA)2 NPs (5 μg/mL) in DMSO, representing the soluble conjugates, were recorded using a fluorescence spectrophotometer (Cary Eclipse, Varian, USA) with an excitation wavelength of 405 nm and compared to those of NPs in DMSO in the presence of 10-mM DTT and in PBS. To assess the intermolecular quenching caused by self-assembly, the fluorescence emission spectrum of NPs in PBS was also measured. The generation of 1O2 by the mPEG-(ss-PhA)2 NPs was monitored using DMA as an 1O2 trap. The NPs were dissolved in DMSO in the absence or presence of 10 mM DTT or in PBS (pH 7.4) and then added to a DMA stock solution to reach a final concentration of 1.184 × 10−2 mM DMA. The solutions containing DMA were irradiated using a 670 nm laser source (LWRL-100 K, Laserwave, China) at a light intensity of 4.2 mW/cm2. The decrease in the fluorescence intensity of DMA (emission from 380 to 550 nm, with excitation at 360 nm) caused by 1O2 generation was monitored using a fluorescence spectrometer. The 1O2 quantum yield (SOQ) was calculated as follows: ϕΔ =
■
RESULTS AND DISCUSSION Synthesis and Physicochemical Characteristics of mPEG-(ss-PhA)2 Conjugates. As shown in Figure 2a, the mPEG-(ss-PhA)2 conjugate was synthesized via a five-step synthesis route. First, the hydroxyl end group of the mPEG was activated by pNC; second, pNC-activated mPEG was coupled with serinol to create a dihydroxylated mPEG; third, the hydroxyl groups of mPEG-serinol was further activated by pNC molecules; forth, disulfide-bond-containing biarmed mPEG-(ssNH2)2 was synthesized by conjugation of cystamine to the chain ends of mPEG-serinol; and finally, the PhA molecules were coupled to the mPEG-(ss-NH2)2 by a conventional carbodiimide reaction. The characteristic peaks of mPEG, pNCactivated mPEG, mPEG-serinol, mPEG-(pNC)2, mPEG-(ssNH2)2, and mPEG-(ss-PhA)2 were observed in the 1H NMR spectra at each step, indicating their successful syntheses (Figure 2b and Figure 3a). The FT-IR spectra of major mPEG derivatives and mPEG-(ss-PhA)2 conjugate were also measured as shown in Figure 2c and 3b. The GPC traces of mPEGserinol, mPEG-(ss-NH2)2, and mPEG-(ss-PhA)2 conjugate revealed a symmetrical elution peak with reasonably narrow polydispersity (1.03−1.05) and the shifts of the chromatograms toward higher molecular weight according to the stage of mPEG derivatization. The number-average MW for mPEG, mPEG-serinol, mPEG-(ss-NH2)2, and mPEG-(ss-PhA)2 conjugates were measured to be 1836, 1954, 2203, and 3054 Da, respectively. Using a concentration-absorbance standard curve for free PhA based on a colorimetric method, the existence of PhA in the mPEG-(ss-PhA)2 conjugate was reconfirmed, and the number of PhA molecules on one mPEG-(ss-PhA)2 molecule
KS A × R × ref · ϕΔ KR AS
where A is the absorbance at 670 nm (the excitation wavelength), K is the gradient of the absorbance curve, and the subscripts S and R refer to the sample and reference, respectively. The SOQ (ΦΔ) of each sample was calculated based on the reference ΦΔ value (free PhA, 0.52) Biological Characterizations of mPEG-(ss-PhA)2 NPs. HeLa cells (a human cervical adenocarcinoma cell line) and MCF7 cells (a human breast adenocarcinoma cell line) were used in this study. HeLa and MCF7 cells were cultured in RPMI 1640 supplemented with 10% FBS and D-glucose (2 g/L). The cells were maintained and grown under humidified air that contained 5% CO2 at 37 °C. Insulin (4 mg/ L) was added to the culture medium for the MCF7 cells. To evaluate the dark toxicity and light-induced toxicity (phototoxicity), the two cell lines were exposed to free PhA and mPEG-(ssPhA)2 NPs. The cells (5000 cells/well) were seeded in a clear 96-well plate with a clear bottom and incubated for 24 h. Next, the cells were treated with free PhA and mPEG-(ss-PhA)2 NPs at various concentrations. At 12 h post-treatment, the culture medium was E
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 4. Physicochemical characteristics of mPEG-(ss-PhA)2 conjugates. (a) UV/visible spectra of free PhA in DMSO and in H2O and of the mPEG-(ss-PhA)2 conjugate in DMSO with and without DTT and in H2O. (b) CQC measurement of mPEG-(ss-PhA)2 conjugates: the fluorescence intensity of pyrene in the mPEG-(ss-PhA)2 conjugate in an aqueous solution with respect to the conjugate concentration. (c) Size distribution of mPEG-(ss-PhA)2 NPs measured using a particle analyzer based on dynamic light scattering. (d) Morphology of mPEG-(ss-PhA)2 NPs observed via FE-SEM.
In general, a typical mono-PS-conjugated PEG has a low content of hydrophobic PS, resulting in limited lipophilicity. For example, a folate-PEG3.4k-chlorin conjugate has been found to exhibit good water solubility (40.1 mg/mL in PBS), causing no obvious aggregation of chlorine molecules.28 Unlike a mono-PS-conjugated PEG, for the unique biarmed structure of the mPEG-(ss-PhA)2 conjugate that was prepared in this study, the self-assembly of NPs of the mPEG-(ss-PhA)2 conjugate was expected because the conjugate had a relatively high PhA loading. Thus, to evaluate whether the mPEG-(ss-PhA)2 conjugate would form self-assembled NPs, the CQC of the mPEG-(ss-PhA)2 conjugate and the size and morphology of the formed mPEG-(ss-PhA)2 NPs were investigated. First, using pyrene as a fluorescence probe, the CQC of the mPEG-(ssPhA)2 conjugate was determined to be 3.78 mg/L, which was the threshold concentration of the conjugates at which selfassembly and the simultaneous fluorescence quenching of pyrene would occur (Figure 4b). It appears that a strong hydrophobic interaction of the mPEG-(ss-PhA)2 conjugates is responsible for the low CQC value, further suggesting that the mPEG-(ss-PhA)2 conjugate could maintain stable NPs under
was calculated to be 1.57. In the UV/visible spectra, the mPEG(ss-PhA)2 conjugate in DMSO had a very similar Q-band absorption at 667 nm to that of free PhA in DMSO, and the Qband absorption of the mPEG-(ss-PhA)2 conjugate was not influenced by the presence of DTT (Figure 4a), implying that the chemical PEGylation of PhA or the presence/absence of the disulfide bonds did not affect the spectrometric characteristics of PhA itself. The Q-band absorption of free PhA in DMSO was shifted toward longer wavelengths (e.g., 683 nm) compared with that of free PhA in water and had a much lower intensity, indicating that free PhA in water exists in its dimeric form.43 The Q-band absorption of the mPEG-(ssPhA)2 NPs in water exhibited a clear bathochromic shift toward 699 nm, which is related to the self-aggregation of PhA molecules in the NPs. However, the absorbance intensity of the mPEG-(ss-PhA)2 NPs in water was much higher than that of free PhA in DMSO and was similar to that of the mPEG-(ssPhA)2 conjugate in DMSO, indicating that the mPEG-(ssPhA)2 NPs effectively enhanced the aqueous solubilization of PhA. F
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 5. (a) Colloidal stability of mPEG-(ss-PhA)2 NPs in PBS and PBS containing 1 mg/mL BSA. (b) Change in the size distribution of mPEG(ss-PhA)2 NPs in PBS before and after incubation with 10 mM DTT.
Figure 6. (a) Fluorescence emission spectra of mPEG-(ss-PhA)2 NPs in DMSO with and without DTT and in PBS. (b) 1O2 generation of free mPEG-(ss-PhA)2 NPs in DMSO with and without DTT and in PBS.
were exposed to harsh reductive conditions (e.g., 10-mM DTT), the NPs exhibited a remarkable change in colloidal stability. For example, when mPEG-(ss-PhA)2 NPs with an initial unimodal size distribution that corresponded to a diameter of approximately 190 nm were incubated in the presence of DTT (10 mM), the size distribution broadened, and the average size was increased to 240 nm after 1 h of incubation (Figure 5b). The NPs continued to grow, and their size distribution became trimodal after 2 h of incubation; the particle size could not be detected after 3 h of incubation. The variations in the particle size and the pattern of the size distribution resulted from the presence of disulfide bonds between the mPEG and PhA molecules, which cause the NPs to be reductively cleavable upon exposure to a reductive environment. These results suggest that mPEG-(ss-PhA)2 NPs may remain stable during blood circulation but then rapidly dissociate and release PhA in a reductive intracellular environment because of the cleavage of the disulfide linkers. Photoactivity of mPEG-(ss-PhA)2 NPs. The photoactivity of the PS, including photoquenching and 1O2 generation, is very significant to the estimation of PDT effects. The photoquenching properties of the mPEG-(ss-PhA)2 NPs were assessed by monitoring their fluorescence emission spectra in DMSO with and without DTT and in PBS. As shown in Figure 6a, no obvious fluorescence emission of the NPs in PBS was observed, implying that the PhA molecules existed in a photoinactive state because of the FRET-based photoquench-
diluted conditions, such as in blood circulation. Second, above the CQC of the amphiphilic mPEG-(ss-PhA)2 conjugate in an aqueous solution, the conjugate readily self-assembled to form NPs with a mean diameter of 189 nm (Figure 4c). The dried mPEG-(ss-PhA)2 NPs were observed using an FE-SEM and were nearly spherical with a diameter of approximately 190 nm (Figure 4d). The size and morphology results suggest that hydrophobic PhA molecules in mPEG-(ss-PhA)2 conjugates self-assemble into spherical NPs with a core−shell structure that consists of an inner hydrophobic PhA core surrounded by a hydrophilic PEG outer shell. The mPEG-(ss-PhA)2 NPs were expected to persist in blood circulation because their “stealth” PEG surface and nanoscale size could avoid phagocytosis and renal clearance. The NPs were also expected to selectively accumulate at tumor sites via an EPR-based passive targeting mechanism because of their nanoscale size. However, for the mPEG-(ss-PhA)2 NPs to achieve their expected extended circulation time and targeted tumor accumulation, the NPs must maintain their colloidal stability in vivo. Thus, the colloidal stability of the mPEG-(ssPhA)2 NPs was investigated in PBS and PBS containing 1 mg/ mL of BSA at 37 °C, respectively. As shown in Figure 5a, the average particle size of the NPs remained unchanged for 5 days in PBS. In addition, by mixing with BSA, the particle size was slightly increased, and then remained unchanged for 5 days, indicating that such NPs have long-term colloidal stability in a physiological condition (e.g., blood). However, when the NPs G
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
ing that was induced by the self-aggregation of PhA molecules via hydrophobic interactions and π−π stacking. However, the NPs in DMSO, which presented as soluble conjugates, exhibited a strong fluorescence emission, indicating that the disassembly of the NPs inhibited the intermolecular quenching, leading to the recovery of the fluorescence emission. Interestingly, when NPs in DMSO were treated with 10 mM DTT, their fluorescence intensity was increased 1.5-fold. This result can be attributed to the cleavage of the disulfide bonds under reductive conditions, which caused the release of free PhA and, consequently, the reduction of the intramolecular quenching between the two PhA molecules in the mPEG-(ssPhA)2 conjugate. In our design concept (Figure 1), it is proposed that the proximity of two PhA molecules at one end of a hydrophilic polymer could create a donor−acceptor pair that could facilitate efficient FRET-based intramolecular photoquenching. However, the coupling ratio of PhA in the mPEG-(ss-PhA)2 conjugate was found to be approximately 1.6, indicating the coexistence of the mPEG-(ss-PhA)2 conjugate and a mono-PhA-conjugated polymer (i.e., an mPEG-ss-PhA conjugate). Thus, although the fluorescence of the two PhA molecules in the mPEG-(ss-PhA)2 conjugate could be quenched, the coexisting mPEG-ss-PhA conjugate could not generate an energy transfer pair and would thus emit strong fluorescence when the excitation beam was applied. This insufficient intramolecular quenching of the conjugate mixture could suppress the photoactivity of PhA. Nevertheless, it was observed that additional intermolecular quenching, which arose from the spontaneous self-assembly of the conjugate mixture, could be effective in controlling the photoactivity of the mPEG(ss-PhA)2 NPs. Furthermore, the rapid cleavage of the disulfide bonds between the mPEG and PhA molecules could accelerate the dissociation of the NPs, which may lead to the efficient recovery of the photoactivity in a reductive environment. The generation of 1O2, which governs the phototoxicity, by mPEG-(ss-PhA)2 NPs was observed in DMSO with and without DTT and in PBS based on the calculation of the SOQ using DMA as a 1O2 trap (Figure 6b). The mPEG-(ss-PhA)2 NPs exhibited a sharp decline in DMA fluorescence intensity in DMSO because of the rapid 1O2 generation that occurred upon light exposure. The SOQ of the NPs in the presence of DTT was calculated to be 0.51, which was slightly higher than that of the NPs under nonreductive conditions (SOQ, 0.49) because of the reduced intramolecular quenching. On the other hand, no obvious decrease in the fluorescence intensity of DMA was observed for the NPs in PBS (SOQ, 0.03), indicating that the generation of 1O2 was dramatically suppressed because of the efficient intramolecular and intermolecular quenching. As a result, it can be concluded that the mPEG-(ss-PhA)2 NPs could not be activated into the triplet state via mutual energy transfer because of the efficient self-quenching and could inhibit the ability of PhA to generate 1O2. The self-quenching effect of the mPEG-(ss-PhA)2 NPs can improve the therapeutic efficiency of PDT by suppressing the phototoxicity of the PDT agent in the bloodstream. Cytotoxicity and Cellular Characteristics of mPEG-(ssPhA)2 NPs. The phototoxicity of PhA delivered by mPEG-(ssPhA)2 NPs was evaluated using MCF7 and HeLa cells and compared with that of free PhA. At the PhA concentrations used in this study, the dark toxicities of free PhA and the mPEG-(ss-PhA)2 NPs did not represent the IC50 (the drug concentration that causes 50% growth inhibition) values for MCF7 cells. However, when free PhA- and mPEG-(ss-PhA)2
NP-treated MCF7 cells were irradiated, cell viability dramatically decreased according to the irradiation time (Figure 7a,b).
Figure 7. Cytotoxicity of (a) free PhA, (b) mPEG-(ss-PhA)2 NPs according to light irradiation time in MCF7 cells, and (c) dark toxicity and phototoxicity of free PhA and mPEG-(ss-PhA)2 NPs in HeLa cells with 63 s light irradiation (n ≥ 6, mean ± standard error).
The IC50 of free PhA and mPEG-(ss-PhA)2 NPs decreased from 0.32 to 0.033 μg/mL and 4.3 to 0.27 μg/mL, respectively, when the irradiation time increased from 16 to 252 s. In addition, we also observed no obvious cytotoxicity in free PhAand mPEG-(ss-PhA)2 NP-treated HeLa cells that maintained in darkness, whereas the phototoxicities of the mPEG-(ss-PhA)2 NPs were 0.16 μg/mL based on the IC50 values and represented 3.9-fold decreases compared with those of free PhA (0.041 μg/mL) when the cells were exposed to light for 63 s. The significant phototoxicity of the mPEG-(ss-PhA)2 NPs implies that the photoactivity of the NPs that was suppressed by photoquenching was restored in living cells, which resulted from a series of dequenching processes caused by the cleavage of the disulfide bonds and dissociation of the self-assembled nanostructure. It is worthy to note that phototoxicity of mPEG(ss-PhA)2 NPs was lower than free PhA for all irradiation times H
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 8. Representative cellular-uptake histogram of free PhA and mPEG-(ss-PhA)2 NPs in (a) MCF7 and (b) HeLa cells at 4 h postincubation. The concentration of free PhA or PhA in the mPEG-(ss-PhA)2 NPs was 0.5 μg/mL.
between the phototoxicities of free PhA and the NPs in the cells. In addition, the results further suggest that mPEG-(ssPhA)2 NPs are a potential delivery system for the poorly watersoluble PhA. To understand how the mPEG-(ss-PhA)2 NPs produce their phototoxic effects in cells, the cell-culture media were supplemented with GSHOEt during the phototoxicity studies because GSHOEt increases the concentration of intracellular thiols. When free-PhA-uptaken MCF7 cells were treated with 2 mM GSHOEt, the GSHOEt did not influence the phototoxicity of the free PhA (Figure 9). This result indicates that the phototoxicity of PhA is not affected by the presence of intracellular thiols. However, in light-exposed MCF7 cells and HeLa cells, mPEG-(ss-PhA)2 NPs in the presence of GSHOEt (IC50 values of 0.20 μg/mL and 0.11 μg/ mL for MCF7 cells and HeLa cells, respectively) produced an approximately 2.7-fold and 1.3-fold higher phototoxicity than that observed in the absence of GSHOEt (IC50 values of 0.53 μg/mL and 0.14 μg/mL for MCF7 cells and HeLa cells, respectively; Figure 9). This finding indicates that the increased level of thiols could more quickly break the disulfide linkages in the mPEG-(ss-PhA)2 NPs, thereby facilitating the release of PhA from the NPs and increasing their phototoxicity. The intracellular localization and intracellular degradation of the mPEG-(ss-PhA)2 NPs were further investigated using confocal microscopy (Figure 10). The fluorescence intensity of free PhA originated mostly from the cytoplasm but also somewhat from
and in both MCF7 and HeLa cells. That is likely related to the different cellular uptake efficiency between free PhA and mPEG-(ss-PhA)2 NPs, because the cell-killing activity of an antitumor drug is correlated with its cellular uptake. Therefore, for further analysis of the therapeutic efficacy of the NPs, the cellular uptake of the mPEG-(ss-PhA)2 NPs was compared with that of free PhA in the two cell lines. After the cells were incubated with either free PhA or mPEG-(ss-PhA)2 NPs for 4 h, the intracellular fluorescence intensities of PhA were monitored via flow cytometry. The NPs presented approximately 4.7-fold and 3.4-fold lower fluorescence intensities (i.e., cellular uptakes) than those of free PhA in MCF7 and HeLa cells, respectively (Figure 8). The lower cellular uptake of the NPs might be attributable to a weak interaction between the cellular membrane and the PEG-decorated NPs, unlike the moderate interaction of hydrophobic PhA with the cellular membrane, and such a difference in interaction strength would further lead to different endocytosis mechanisms (i.e., fluidphase endocytosis vs adsorptive endocytosis). Nevertheless, if the same amount of PhA were internalized, the phototoxicity of the mPEG-(ss-PhA)2 NPs would be very similar to that of free PhA because the former would be only approximately 1.5-fold or 1.2-fold lower than the latter in MCF7 or HeLa cells, respectively. This estimation indicates that the difference between the amounts of free PhA and mPEG-(ss-PhA)2 NPs internalized in the cells may be responsible for the difference I
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
fluorescence intensity, and increased phototoxicity of the PhA delivered by the NPs. In the current stage, the mPEG-(ss-PhA)2 NPs have shown its perspective for controllable PDT with cancer cells in the cell culture system. Further studies will be focused to evaluate the tumor specificity and PDT efficacy of the designed mPEG-(ssPhA)2 NPs in vivo and to further explore their practical applications in clinical PDT.
■
CONCLUSION
In this study, a novel bioreducible mPEG-(ss-PhA)2 conjugate was synthesized through multiple reaction steps and readily self-assembled to construct stable core−shell-structured NPs under aqueous conditions. The thiol-sensitive disulfide linkers introduced between the mPEG and PhA molecules accelerated the intracellular release of PhA. The self-assembled NPs could suppress the photoactivity of the PhA molecules through FRET-based self-quenching and modulated 1O2 generation. However, when the NPs were exposed to a reductive environment, the cleavage of the disulfide bonds facilitated the dissociation of the NP structure and caused rapid release of PhA, and as a result, the photoactivity of the PhA was effectively restored. In vitro cell studies revealed that the mPEG-(ss-PhA)2 NPs exhibited significant phototoxicity in cancer cells, and the cytotoxicity was lower than that of free PhA because of the relatively low cellular uptake. However, enhanced phototoxicity of the NPs was observed when the expression of intracellular thiols in the cytoplasm was increased. These findings suggest that the bioreducible mPEG-(ss-PhA)2 NPs could provide a potential implementation of controllable 1 O2 generation within cancer cells, which could lead to more effective PDT with minimal phototoxicity in normal tissue and enhanced efficacy for photodynamic cancer treatment.
Figure 9. Effect of GSHOEt on the phototoxicity of free PhA and mPEG-(ss-PhA)2 NPs in (a) MCF7 cells and (b) HeLa cells with 63 s light irradiation (n ≥ 6, mean ± standard error).
the nucleus, and it remained similar, regardless of the presence of GSHOEt. Although the fluorescence intensity of the PhA delivered by the mPEG-(ss-PhA)2 NPs also originated from the cytoplasm and nucleus, the fluorescence intensity in this case was increased by the addition of GSHOEt (Figure 10). These results indicate that the presence of intracellular thiols in the cytoplasm breaks the disulfide linkages in the mPEG-(ss-PhA)2 NPs, resulting in the increased cytoplasmic release, increased
Figure 10. Intracellular localization and fluorescence intensity of PhA in free-PhA-uptaken and mPEG-(ss-PhA)2 NP-uptaken MCF7 cells in the absence or presence of GSHOEt (2 mM). The concentration of free PhA or PhA in the mPEG-(ss-PhA)2 NPs was 0.5 or 4 μg/mL, respectively. J
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
■
Article
(25) Ohya, Y.; Shirakawa, S.; Matsumoto, M.; Ouchi, T. Polym. Adv. Technol. 2000, 11, 635−641. (26) Ding, Y.; Zhou, Y. Y.; Chen, H.; Geng, D. D.; Wu, D. Y.; Hong, J.; Shen, W. B.; Hang, T. J.; Zhang, C. Biomaterials 2013, 34, 10217− 10227. (27) Wermeckes, B.; Hess, M.; Dehne, S.; Jo, B. W.; Sohn, J. S. Mater. Res. Innov. 2006, 10, 88−90. (28) Li, D. H.; Li, P. X.; Lin, H. Y.; Jiang, Z. L.; Guo, L. F.; Li, B. H. J. Photochem. Photobiol. B 2013, 127, 28−37. (29) Howard, M. D.; Jay, M.; Dziublal, T. D.; Lu, X. L. J. Biomed. Nanotechnol. 2008, 4, 133−148. (30) Veronese, F. M.; Pasut, G. Drug Discovery Today 2005, 10, 1451−1458. (31) Verhille, M.; Couleaud, P.; Vanderesse, R.; Brault, D.; BarberiHeyob, M.; Frochot, C. Curr. Med. Chem. 2010, 17, 3925−3943. (32) Lovell, J. F.; Liu, T. W. B.; Chen, J.; Zheng, G. Chem. Rev. 2010, 110, 2839−2857. (33) Allison, R. R.; Mota, H. C.; Bagnato, V. S.; Sibata, C. H. Photodiagn. Photodyn. 2008, 5, 19−28. (34) Fernandez, J. M.; Bilgin, M. D.; Grossweiner, L. I. J. Photochem. Photobiol. B 1997, 37, 131−140. (35) Campo, M. A.; Gabriel, D.; Kucera, P.; Gurny, R.; Lange, N. Photochem. Photobiol. 2007, 83, 958−965. (36) Gamcsik, M. P.; Kasibhatla, M. S.; Teeter, S. D.; Colvin, O. M. Biomarkers 2012, 17, 671−691. (37) Cheng, R.; Feng, F.; Meng, F. H.; Deng, C.; Feijen, J.; Zhong, Z. Y. J. Controlled Release 2011, 152, 2−12. (38) He, H.; Lo, P.-C.; Ng, D. K. P. Chem.Eur. J. 2014, 20, 1−6. (39) Isik, M.; Ozdemir, T.; Turan, I. S.; Kolemen, S.; Akkaya, E. U. Org. Lett. 2013, 15, 216−219. (40) Park, S. Y.; Oh, K. T.; Oh, Y. T.; Oh, N. M.; Youn, Y. S.; Lee, E. S. Chem. Commun. 2013, 48, 2522−2524. (41) Kim, H.; Mun, S.; Choi, Y. J. Mater. Chem. B 2013, 1, 429−431. (42) Li, L.; Nurunnabi, M.; Nafiujjaman, M.; Jeong, Y. Y.; Lee, Y. K.; Huh, K. M. J. Mater. Chem. B 2014, 2, 2929−2937. (43) Eichwurzel, I.; Stiel, H.; Roder, B. J. Photochem. Photobiol. B 2000, 54, 194−200.
AUTHOR INFORMATION
Corresponding Authors
*Tel.: +82-42-821-6663. Fax: +82-42-821-8910. E-mail: khuh@ cnu.ac.kr. *Tel.: +82-2-2164-6533. Fax: +82-2-2164-4059. E-mail: [email protected]. Author Contributions †
These authors contributed equally to this work (W.L.K. and H.C.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (NRF-2012R1A1A2005029 and NRF2013R1A2A2A04015914). H.C.K. appreciates Research Fund of The Catholic University of Korea (2012).
■
REFERENCES
(1) Juarranz, A.; Jaen, P.; Sanz-Rodriguez, F.; Cuevas, J.; Gonzalez, S. Clin. Transl. Oncol. 2008, 10, 148−154. (2) MacDonald, I. J.; Dougherty, T. J. J. Porphyrins Phthalocyanines 2001, 5, 105−129. (3) Mitton, D.; Ackroyd, R. Photodiagn. Photodyn. 2008, 5, 103−111. (4) Ibbotson, S. H. Photodiagn. Photodyn. 2010, 7, 16−23. (5) Ortner, M. A. Curr. Opin. Gastroen. 2009, 25, 472−476. (6) Kohl, E. A.; Karrer, S. G. Ital. Dermatol. Venereol. 2011, 146, 473− 485. (7) Vrouenraets, M. B.; Visser, G. W.; Snow, G. B.; van Dongen, G. A. Anticancer Res. 2003, 23, 505−22. (8) Detty, M. R.; Gibson, S. L.; Wagner, S. J. J. Med. Chem. 2004, 47, 3897−3915. (9) Gravier, J.; Schneider, R.; Frochot, C.; Bastogne, T.; Schmitt, F.; Didelon, J.; Guillemin, F.; Barberi-Heyob, M. J. Med. Chem. 2008, 51, 3867−77. (10) Huang, P.; Xu, C.; Lin, J.; Wang, C.; Wang, X. S.; Zhang, C. L.; Zhou, X. J.; Guo, S. W.; Cui, D. X. Theranostics 2011, 1, 240−250. (11) Ranyuk, E.; Cauchon, N.; Klarskov, K.; Guerin, B.; van Lier, J. E. J. Med. Chem. 2013, 56, 1520−1534. (12) Mukai, H.; Wada, Y.; Watanabe, Y. Ann. Nucl. Med. 2013, 27, 625−639. (13) Toneatto, J.; Garcia, P. F.; Arguello, G. A. J. Inorg. Biochem. 2011, 105, 1299−1305. (14) Allen, C. M.; Sharman, W. M.; La Madeleine, C.; Weber, J. M.; Langlois, R.; Ouellet, R.; van Lier, J. E. Photochem. Photobiol. 1999, 70, 512−523. (15) Lee, D. J.; Park, S. Y.; Oh, Y. T.; Oh, N. M.; Oh, K. T.; Youn, Y. S.; Lee, E. S. Macromol. Res. 2011, 19, 848−852. (16) Li, L.; Bae, B. C.; Tran, T. H.; Yoon, K. H.; Na, K.; Huh, K. M. Carbohydr. Polym. 2011, 86, 708−715. (17) Oh, I. H.; Min, H. S.; Li, L.; Tran, T. H.; Lee, Y. K.; Kwon, I. C.; Choi, K.; Kim, K.; Huh, K. M. Biomaterials 2013, 34, 6454−6463. (18) Li, L.; Nurunnabi, M.; Nafiujjaman, M.; Lee, Y. K.; Huh, K. M. J. Controlled Release 2013, 171, 241−250. (19) Greenwald, R. B.; Choe, Y. H.; McGuire, J.; Conover, C. D. Adv. Drug Delivery Rev. 2003, 55, 217−250. (20) Knop, K.; Hoogenboom, R.; Fischer, D.; Schubert, U. S. Angew. Chem., Int. Ed. 2010, 49, 6288−6308. (21) Joralemon, M. J.; Mcrae, S.; Emrick, T. Chem. Commun. 2010, 46, 1377−1393. (22) Bruckdorfer, T. Biopolymers 2009, 92, 337−337. (23) Yoo, H. S.; Park, T. G. J. Controlled Release 2004, 100, 247−256. (24) Ye, W. L.; Zhou, S. Y. Acta Pharmacol. Sin. 2013, 34, 18−18. K
dx.doi.org/10.1021/bm5003619 | Biomacromolecules XXXX, XXX, XXX−XXX
