EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb
2
Research paper
6 4 7
Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin
5 8
Q1
a
9 10 11
Tianjin Key Laboratory for Modern Drug Delivery & High Efficiency, School of Pharmaceutical Science & Technology, Tianjin University, Tianjin, China Institute of Biomedical Engineering, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China c State Key Laboratory of Medicinal Chemical Biology, Key Laboratory of Bioactive Materials, Ministry of Education, College of Life Sciences, Nankai University, Tianjin, China b
13 12 14
a r t i c l e
1 2 6 8 17 18 19 20 21 22 23 24 25 26 27
Zheng Wang a, Chao Chen a, Qi Zhang a, Min Gao a, Ju Zhang b, Deling Kong c, Yanjun Zhao a,⇑
i n f o
Article history: Received 9 September 2014 Accepted in revised form 7 November 2014 Available online xxxx
Q2
Keywords: Drug delivery Micelle Self-assembly Polymeric conjugate Curcumin
a b s t r a c t To precisely manipulate the intracellular delivery of pleiotropic curcumin, this work reports two types of acid-responsive polymer–curcumin conjugates with different structures. One or two amphiphilic poly(ethylene glycol)-co-poly(lactic acid) (PEG-PLA) copolymer(s) were linked to curcumin via pH-liable hydrazone, producing a linear or phospholipid-like conjugate, respectively. Both conjugates efficiently self-assembled into micellar nanocarriers with enhanced stability in contrast to the paralleling PEG-PLA micelles. In comparison to the micelles assembled by phospholipid-like conjugates, the linear conjugate micelles exhibited similar size, doubled loading dose, higher release rate, and hence enhanced cellular uptake and cytotoxicity at the cost of increased critical micelle concentration. This work highlighted the importance of precise tuning of polymer–drug conjugate architecture in determining the pharmaceutical properties and thus delivery efficiency of the corresponding conjugate micelles. Ó 2014 Published by Elsevier B.V.
29 30 31 32 33 34 35 36 37 38 39 40 41
42 43
1. Introduction
44
Taking the advantage of enhanced permeability and retention (EPR) effect, nanocarriers have been prevalent for delivering active pharmaceutical ingredient (API) to tumours since the mid-1980s [1–3]. Prior to reaching the tumour site, the API and nanocarriers have to travel over a long distance in the systemic circulation. In many cases the API is loaded in the nanocarrier via simply physical encapsulation or adsorption, e.g. commercially available DoxilÒ and AbraxaneÒ. Since the blood circulation maintains a perfect sink condition, the premature drug release during the journey to the tumour target is inevitable no matter how strong the avidity between the API and the nanocarrier is [4,5]. This significantly contributes to both physical side-effects and psychological harm to the patients. Attaching the API to nanocarrier covalently is an effective means to address the issue of premature release [6–8]. However, the rapid breakdown of chemical bonds often requires certain conditions, i.e. specific enzymes, and catalyst. This induces another problem of poor drug release and delayed onset of action upon reaching the target [9,10]. Polymer–drug conjugate micelle offers a versatile platform for addressing the dilemma of premature release and
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
⇑ Corresponding author. School of Pharmaceutical Science & Technology, Tianjin University, 92 Weijin Road, Nankai District, Tianjin 300072, China. Tel.: +86 22 2740 7882; fax: +86 22 2740 4018. E-mail address: [email protected] (Y. Zhao).
on-demand release in the target site [11–13]. The abundance of polymer variety and specialty makes possible the tailored design and manufacture of polymer–drug conjugates with biodegradability, biocompatibility, self-assembly, long circulation, and stimuliresponsive drug release and other desired characteristics for efficient delivery [14–17]. In addition, the tumour tissue is well known to exhibit a low pH (ca. 6.8). This earmark together with the low pH milieu (4.5–6.5) of endosome and lysosome provides a superb sitespecific trigger for on-demand anti-cancer drug delivery. A rich collection of pH-liable cleavable linkers has been reported including hydrazone, hydrazide, oxime, imine, acetal and ketal, which makes ease of fabricating acid-responsive polymer–drug conjugate [18]. The diversity of polymer and API as well as their tailored combination can generate conjugates with multifarious architecture that is believed to make a significant impact on the pharmaceutical properties and ultimately the therapeutic outcome [19–21]. Curcumin (Cur) has been recognized as a pleiotropic agent with anti-inflammatory, antioxidant, anticancer, and many other medical properties as a consequence of its ability to regulate ca. 100 cell signalling pathways [22]. In terms of chemotherapy, it also shows the capability to reverse tumour multi-drug resistance [23]. Efficient curcumin delivery has been suffering from the poor aqueous solubility (ca. 1 lg/mL) and chemical instability, leading to poor bioavailability [24]. The presence of two phenol groups in one curcumin molecule offers the opportunity of drug conjugation with either one or two chain(s) of polymer simultaneously, bringing
http://dx.doi.org/10.1016/j.ejpb.2014.11.002 0939-6411/Ó 2014 Published by Elsevier B.V.
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 2 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115
Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
about different polymer–curcumin conjugate architectures or topologies. Utilizing the hydrophobicity of curcumin (log P = 3.2), Tang et al. connected two segments of hydrophilic poly(ethylene glycol) (PEG) to one molecule of curcumin via beta-thioester bond, producing a structure similar to phospholipid with two hydrophilic tail and one hydrophobic head [25]. Yang et al. employed ester bond to link curcumin and hydrophobic poly(lactic acid) (PLA) that was further tied with PEG, generating linear PEG-PLA-Cur conjugate [10]. All the above reported conjugates showed amphiphilicity and could capably self-assemble into micellar nanocarriers. It was postulated that such topology divergence of polymer–curcumin conjugate would make a significant difference in the conjugate self-assembly behaviour (e.g. critical micelle concentration, aggregation number, and the critical packing parameter), pharmaceutical properties of micellar nano-assemblies (e.g. particle size, drug loading, release rate, stimuli-responsiveness, and cellular uptake), and hence the eventual therapeutic efficacy (e.g. cytotoxicity). Thus the aim of the current study was to correlate the architecture of acid-responsive polymer–curcumin conjugate and the delivery efficiency of corresponding assembled micelles. To achieve this, biodegradable amphiphilic PEG-PLA copolymer was selected as the building block. pH-liable hydrazone was used as the linker between the polymer and curcumin to realize on-demand rapid drug release inside the target cells, but minimized premature release in the systemic circulation. Both linear (PEG-PLA-Cur) and phospholipid-like (Bi(PEG-PLA)-Cur) conjugates were created and compared (Scheme 1). Their architectures were illustrated below (Fig. 1).
Fig. 1. Schematic illustration of the topology of amphiphilic acid-liable PEG-PLACur (single-tail) and Bi(PEG-PLA)-Cur (double-tail, phospholipid-like) conjugate and the factors determining their critical packing parameter (CPP): equilibrium area per conjugate molecule at the aggregate interface (ae), hydrophobic tail volume (v0), and hydrophobic tail length (l0). (For the interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Scheme 1. The synthetic route of two types of acid-responsive polymer–drug conjugates: PEG-PLA-Cur and Bi(PEG-PLA)-Cur. PEG, PLA, and Cur represents poly(ethylene glycol), poly(lactic acid), and curcumin, respectively. Hydrazone is the pH-liable linker between polymer and curcumin.
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx 116
2. Materials and methods
117
2.1. Materials
118
138
Curcumin was purchased from Guangfu Fine Chemical Research Institute (Tianjin, China). 40 ,6-diamidino-2-phenylindole (DAPI) and methoxy poly(ethylene glycol) (PEG, 2000 Da) were obtained from Sigma–Aldrich (Beijing, China). Levulinic acid, stannous octoate, hydrazine monohydrate, pyrene, and 4-dimethylaminopyridine (DMAP) were from Aladdin (Shanghai, China). N-(3Dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDCHCl) was from Medpep Co., Ltd. (Shanghai, China). D,L-lactide was from GLACO Ltd. (Beijing, China). 4-Nitrophenyl chloroformate (NPC) was from Energy Chemical (Shanghai, China). Methanol and tetrahydrofuran (THF) were from Concord Technology Co., Ltd. (Tianjin, China). Dulbecco’s modification of eagle’s medium (DMEM), foetal bovine serum, penicillin–streptomycin, and phosphate buffer solution (PBS) were from HyClone, Inc. (Logan City, Utah, USA). Trypsin was from Life Technologies (Beijing, China). Cell Counting Kit-8 (CCK-8) was from Dojindo Laboratories (Shanghai, China). The human epithelial carcinoma (HeLa) and human hepatocellular liver carcinoma (HepG2) cell lines were provided by Institute of Biomedical Engineering (Chinese Academy of Medical Sciences & Peking Union Medical College). All other chemicals were sourced from Jiangtian Chemicals (Tianjin, China).
139
2.2. Polymeric conjugate synthesis
140
2.2.1. Cur-1 and Cur-2 Carboxyl-terminated curcumin derivatives (Cur-1 and Cur-2) (Scheme 1) were prepared as follows based on a previous report [26]. Levulinic acid (0.0631 g, 0.543 mmol), DMAP (0.0662 g, 0.543 mmol), and EDCHCl (0.1041 g, 0.543 mmol) were mixed in a round-bottom flask containing 20 mL dichloromethane (DCM) for 1 h. Curcumin (0.2 g, 0.543 mmol) was dissolved in 30 mL DCM. The former solution was transferred slowly to the curcumin solution and the reaction was maintained at 25 °C for 24 h with nitrogen protection. Then the crude product was washed in triplicate with water post DCM evaporation and filtration. The collected solid was dissolved in ethyl acetate (EA) following the extraction with hydrochloride acid (HCl) solution; the water residue in the organic phase was removed with sodium sulphate (anhydrous), followed by EA removal to get unpurified Cur-1. The purification was carried out using the silica gel column chromatography with a gradient elution. The starting eluent was a mixture of petroleum ether and EA (2:1, v/v) containing 0.2% (v/v) glacial acetic acid for unreacted curcumin that came out first. Then a mixture of petroleum ether and EA (1:1–1:5, v/v) was used to elute Cur-1 (yield: 50%). The final product was obtained via solvent evaporation and vacuum-drying. 1H NMR (600 MHz, CDCl3), d [ppm]: 2.16 (s, 3H); 2.81 (s, 4H); 3.80 (s, 3H); 3.88 (s, 3H); 5.76–5.81 (d, 2H); 6.41– 6.49 (q, 2H); 6.84–7.10 (m, 6H); 7.50–7.56 (q, 2H) (Fig. S1, Supporting Information/SI). Likewise, the synthesis of Cur-2 was maintained at the same condition except an altered reactant feeding: levulinic acid (0.630 g, 5.43 mmol), DMAP (0.662 g, 5.43 mmol), EDCHCl (1.041 g, 5.43 mmol), and curcumin (0.2 g, 0.543 mmol). The isocratic eluent for Cur-2 purification was 1, 2-dichloroethane (yield: 40%). 1H NMR (600 MHz, CDCl3), d [ppm]: 2.23 (s, 6H); 2.88 (s, 8H); 3.87 (s, 6H); 5.85 (s, 2H); 6.55–6.57 (d, 2H); 7.06–7.17 (m, 6H); 7.58–7.64 (d, 2H) (Fig. S2, SI).
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
173 174 175
2.2.2. PEG-PLA PEG (5 g, 2.5 mmol) was melted in a 100 mL flask at 80 °C and then lactide (5 g, 34.7 mmol) was added. The mixture was heated
3
until melting, and then stannous octoate (1 mL, 3.1 mmol) was supplemented. The reaction was kept at 125 °C for 24 h under nitrogen atmosphere. Afterwards, the crude product was cooled down to ambient temperature and dissolved in tetrahydrofuran (THF), which was followed by precipitation in ice-cooled diethyl ether. This process was repeated in triplicate and the resultant solid was vacuum-dried at ambient temperature to get PEG-PLA (yield: 93%). 1H NMR (600 MHz, CDCl3), d [ppm]: 1.56 (m, –CH3 PLA repeating unit); 3.38 (s, –CH3 PEG end group); 3.64 (m, – CH2 PEG repeating unit); 5.17 (m, –CH PLA repeating unit) (Fig. S3, SI).
176
2.2.3. PEG-PLA-NPC PEG-PLA (3 g, 0.857 mmol) and pyridine (0.345 mL, 4.286 mmol) were mixed in 50 mL DCM at 0 °C with nitrogen protection. Then NPC (0.691 g, 3.428 mmol) solution in DCM (10 mL) was gently added to the above solution. The mixture was maintained at 0 °C for 1 h and then reacted for 24 h at ambient temperature (25 °C). The crude product was filtered and the filtrate was transferred to appropriate amount of ice-cooled diethyl ether. The obtained white solid was precipitated in triplicate and then vacuum-dried to get PEG-PLA-NPC (yield: 95%). 1H NMR (600 MHz, CDCl3), d [ppm]: 1.56 (m, –CH3 PLA repeating unit); 3.38 (s, –CH3 PEG end group); 3.64 (m, –CH2 PEG repeating unit); 5.17 (m, –CH PLA repeating unit); 7.4 (t, –CH aromatic); 8.28 (d, –CH aromatic) (Fig. S4, SI).
187
2.2.4. PEG-PLA-Cur Activated PEG-PLA-NPC (0.5 g, 0.143 mmol) and zinc powder (0.5 mg, 7.65 lmol) were blended in 40 mL dimethylformamide (DMF) under nitrogen atmosphere. Then hydrazine monohydrate (26 lL, 0.429 mmol) was added to the mixture and the reaction was maintained at 25 °C for 6 h. The crude product was purified by filtration, repeated precipitation in diethyl ether, and then vacuum-dried to get the amino-terminated polymer, PEG-PLANHNH2. Then PEG-PLA-NHNH2 (1.0 g, 0.286 mmol) and Cur-1 (0.2 g, 0.429 mmol) were dissolved in 15 mL DMF, and reacted at ambient temperature for 48 h. The crude product was dialysed against 1000 mL water with light protection using a dialysis tube (molecular weight cut-off/MWCO: 3500 Da). The dialysis medium was maintained at ca. pH 7.0 with the diluted ammonium, and replaced regularly. After 24 h, the dialysed solution was centrifuged for 10 min (1000g), filtered using a microfilter (0.45 lm), and then freeze-dried to get single-tailed polymer–drug conjugate PEG-PLA-Cur (yield: 70%). 1H NMR (600 MHz, CDCl3), d [ppm]: 1.56 (m, –CH3 PLA repeating unit); 2.23 (s, –CH3 spacer); 2.88 (s, –CH2 spacer); 3.38 (s, –CH3 PEG end group); 3.64 (m, –CH2 PEG repeating unit); 5.17 (m, –CH PLA repeating unit); 5.80 (d, 2H); 6.45–6.51 (q, 2H); 6.91–7.16 (m, 6H); 7.57–7.62 (q, 2H); 8.09 (s, NH) (Fig. 2).
201
2.2.5. Bi(PEG-PLA)-Cur The synthesis and purification of double-tailed conjugate Bi(PEG-PLA)-Cur were carried out using a similar approach to that PEG-PLA-Cur described in the above section (yield: 60%). The reactants were PEG-PLA-NHNH2 (1.0 g, 0.286 mmol) and Cur-2 (0.054 g, 0.095 mmol). The dialysis tube had a MWCO of 7,000 Da. 1H NMR (600 MHz, CDCl3), d [ppm]: 1.49 (m, –CH3 PLA repeating unit); 3.31 (s, –CH3 PEG end group); 3.57 (m, –CH2 PEG repeating unit); 3.88 (s, 6H); 5.09 (m, –CH PLA repeating unit); 5.74 (s, 2H); 6.38–6.44 (d, 2H); 6.82–7.10 (m, 6H); 7.49–7.55 (d, 2H); 8.02 (s, –NH) (Fig. 3).
224
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
177 178 179 180 181 182 183 184 185 186
188 189 190 191 192 193 194 195 196 197 198 199 200
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
225 226 227 228 229 230 231 232 233 234
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 4
Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
Fig. 2. 1H NMR spectrum of PEG-PLA-Cur conjugate in CDCl3. PEG, PLA, and Cur indicated poly(ethylene glycol), poly(lactic acid), and curcumin, respectively. (For the interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
1
Fig. 3. H NMR spectrum of Bi(PEG-PLA)-Cur conjugate in CDCl3. PEG, PLA, and Cur indicated poly(ethylene glycol), poly(lactic acid), and curcumin, respectively. The proton resonance peaks at position 7, 8, and 9 were not seen due to the influence of water peak in the solvent (ca. 1.7–2.5 ppm). 235
2.3. Molecular weight analysis
236
The molecular weight and its distribution of both types of polymer–curcumin conjugates were analysed by a gel permeation chromatography (GPC) system (Malvern Viscotek TDA 305). Both samples were dissolved in THF to get a concentration of 5 mg/
237 238 239
mL. Polystyrene was used as the calibration standard with THF as the eluent at a speed of 1 mL/min at ambient temperature.
240
2.4. Determination of critical micelle concentration and mean aggregation number
242
The conjugate micelles were prepared using the dialysis method (MWCO: 2000 Da) with THF as the organic solvent. The critical micelle concentration (CMC) of micelles was investigated using a FluorologÒ3–21 spectrofluorometer (HORIBA JobinYvon). Micellar aqueous solution (0.4–400 lg/mL) was supplemented with pyrene probe to reach a constant probe concentration of 0.5 lM. The emission spectra were recorded from 350 nm to 450 nm with an excitation wavelength of 333 nm and spectral slit width of 5 nm. The intensity ratio of pyrene probe band at 384 nm and 373 nm was plotted against the logarithm of the conjugate micelles’ concentration and the CMC was determined by getting the flexion point of the sigmoidal curve [10]. The mean aggregation number (nagg) of both polymer–curcumin conjugate micelles was examined by a dynamic light scatter (DLS) method [27,28]. The analysis was carried out in a concentration range of 0.1–12 mg/ mL for both samples. Prior to DLS measurement, the specific refractive index increment of the conjugate aqueous solution was determined by a digital Abbe refractometer (WAY-2S) at ambient temperature. The DLS analysis employed toluene as the standard with the scattering angle fixed at 90°. The Debye plot was generated to get intercept that was reciprocal of the mean molecular weight of one single nanocarrier. The mean aggregation number was obtained simply via dividing the molecular weight of the micelle by that of the corresponding conjugate. All measurements were taken in triplicate.
244
2.5. Particle size and zeta potential analysis
269
The hydrodynamic diameter and zeta potential of the conjugate micelles were analysed in triplicate by a Zetasizer Nano ZS (Malvern Instruments). The measurement was taken at 25 °C with a sample concentration of 5 mg/mL in water. The size and morphology of both types of micelles were determined by a JEM-100 CXII transmission electron microscope (TEM). One drop of aqueous micelle solution (1 mg/mL) was transferred to a collodion support on copper grids followed by air-drying and then the sample is ready for imagerecording. Atomic force microscope (AFM) (Veeco, Multimode) was also utilized to characterize the micelles. The aqueous sample (50 lg/mL) was placed on the surface of microscope glass slide and then freeze-dried prior to AFM analysis via a tapping mode.
270
2.6. Drug loading and release analysis
282
The curcumin content in the conjugate micelles was determined using a Cary 60 UV–vis spectrophotometer. Prior to analysis, the UV–vis spectra of both conjugates, PEG-PLA-NHNH2, and curcumin in methanol were obtained to verify no shift of the wavelength, at which the drug/conjugate showed the maximum absorption. The quantification employed curcumin as the calibration standard with methanol as the solvent. The absorption was recorded at 424 nm (n = 3). The drug release experiment utilized the same method in our previously published work [10]. Briefly, PEG-PLA-Cur (4 mg) or Bi(PEG-PLA)-Cur (9.4 mg) micelle dispersed in 2 mL water was added to the donor compartment of a Franz cell. The receiver compartment (ca. 17 mL) was filled with citric acid/ disodium hydrogen phosphate buffer at one of three designed pH values (5.0, 6.0, and 7.4). The receiver fluid contains 5% (w/v) sodium dodecyl sulphate (SDS) to keep the sink condition. A porous regenerated cellulose membrane (MWCO: 3500 Da) was used to separate the donor and receiver compartment. The release study
283
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
241
243
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
271 272 273 274 275 276 277 278 279 280 281
284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
5
306
was carried out at 37 °C in triplicate and the drug content in the receiver fluid at pre-determined time points was analysed by an Agilent 1100 high performance liquid chromatography (HPLC) system coupled with a UV detector. The cumulative amount of released drug was plotted against time. Due to the chemical instability of curcumin, a mass balance study was also performed to calculate the total drug recovery at the end of experiments.
307
2.7. In vitro cytotoxicity assay
308
323
HeLa or HepG2 cells were cultured in DMEM medium supplemented with 10% foetal bovine serum and 100 kU/L penicillin– streptomycin. The cells were seeded in 96-well plates at a density of 8 103 per well in 100 lL medium and incubated at 37 °C and 5% CO2. After 24 h, the medium was removed and the cells were washed with 100 lL PBS. Then 200 lL micelle-containing fresh medium was added to the plate; wherein the drug concentration ranged from 5 lg/mL to 100 lg/mL. After another 24 h’s incubation, the drug-containing medium was discarded and the cells were washed with PBS three times followed by the addition of 100 lL medium together with 10 lL CCK-8 solution, and further incubation for 30 min. Afterwards, the spectrophotometric measurement at 459 nm was taken and fresh cell medium was used as the control. The half maximal inhibitory concentration (IC50) of different micelles was calculated using a previously reported method (n = 4) [10]. 2.8. Cellular uptake study
3.1. Characterization of polymer–curcumin conjugate and assembled micelles
355
324 325
HeLa or HepG2 cells were seeded to a CLSM plate at a density of 1 104 cells per well in NEST-glass bottom cell culture dish (15 mm) containing 1 mL DMEM medium. After 24 h’ incubation, the medium was removed and the cells were washed with 0.5 mL PBS twice, followed by the addition of 200 lL fresh medium and then 200 lL micelle solution (100 lg/mL). At predesigned time point (2 h, 6 h, and 12 h), the drug-containing medium was withdrawn and 1 mL PBS was added to wash the cells in triplicate. Then
Two types of polymer–curcumin conjugates were synthesized, i.e. PEG-PLA-Cur and Bi(PEG-PLA)-Cur (Figs. 2 and 3). The GPC analysis revealed that the two-tail Bi(PEG-PLA)-Cur exhibited a roughly doubled molecular weight (MW) compared to the single-chain PEG-PLA-Cur (Fig. 4a). However, the MWs determined by GPC were higher compared to that obtained by 1H NMR (Table S1, SI). Although GPC is the most routinely used method for MW analysis of polymers, it is a relative approach with strong dependence of its
357
300 301 302 303 304 305
309 310 311 312 313 314 315 316 317 318 319 320 321 322
326 327 328 329 330 331 332
the cells were fixed with 1 mL paraformaldehyde (4%) for 20 min at 4 °C with light protection. Subsequently, the paraformaldehyde was removed following the cell rinse by 1 mL PBS twice. Then the cells were stained with 300 lL DAPI solution (1 lg/mL) for 5 min and washed with 1 mL PBS ready for CLSM analysis with the excitation wavelength at 488 nm.
333
2.9. Flow cytometry study
339
Flow cytometry was also employed to analyse cellular the uptake of micelles. HepG2 cells were seeded to a petri dish (60 15 mm) at a density of 1 104 cells in 1 mL medium and cultured for 24 h. Afterwards, the cells were washed with 1 mL PBS and further incubated in micelle-containing DMEM medium with a drug concentration fixed at 50 lg/mL. After 24 h, culture medium was removed and cells were washed with 1 mL PBS followed by the treatment with 1 mL trypsin for 3 min at 37 °C. Then 3 mL medium was added to each culture well and the solutions were centrifuged using an Eppendorf 5810 R centrifuge at 1000 rpm for 5 min. The supernatant was removed and the cells were suspended in 1 mL paraformaldehyde (4%) at 4 °C with light protection. The fluorescence was analysed via the BD FACSCalibur flow cytometer and the cells without micelle treatment were used as the control.
340
3. Results and discussion
354
Fig. 4. The characterization of acid-responsive polymer–curcumin conjugates: (a) molecular weight determined by GPC; (b) mean aggregation number analysed by DLS; (c and d) CMC of PEG-PLA-Cur and Bi(PEG-PLA)-Cur obtained by fluorescence techniques (n = 3). (For the interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
334 335 336 337 338
341 342 343 344 345 346 347 348 349 350 351 352 353
356
358 359 360 361 362 363 364
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 6 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
data on the calibrant and polymer architecture. In spite of the simplicity of 1H NMR, its utility in determining polymer MW has been somewhat underappreciated since it cannot get the information of weight-average molecular weight. Therefore, the combination of both methods would produce more reliable MW information of polymers. Although the polymer polydispersity index (PDI) obtained by GPC was slightly higher, it was still within the acceptable level since the choices of both calibrant standard and elution solvent would affect the PDI results. Both conjugates could successfully self-assemble to micelles. For two-tail conjugate, both CMC and nagg were lower compared to that of single-tail conjugate and the calculation utilized the MW data determined by GPC (Fig. 4b–d). The CMC of a typical nonionic single-chain amphiphilic copolymer usually decreases with the growth of hydrophobic moiety; meanwhile the nagg would move towards an opposite direction, which was due to the elevated driving force of micelle formation, i.e. to minimize the interfacial free energy [29]. In the current study, with the aid of a fluorescent probe, the two-tail conjugate’s CMC (0.6 ± 0.1 lM) was four times lower in contrast to that of single-tail conjugate (2.6 ± 0.3 lM), which manifested a decreased free energy in the former system. This was presumed as a consequence of the elevated hydrophobicity of the two-tail conjugate. The cohesion of hydrophobic core protected by the hydrophilic PEG moiety was thus enhanced, giving rise to a lower CMC and improved stability. The stability enhancement was crucial and advantageous to maintain the integrity of the micellar nanocarriers upon large-ratio blood dilution post intravenous dose administration. The nagg value of both conjugate micelles was 101 ± 3 (single-tail) and 70 ± 5 (double-tail), correspondingly. The reduced nagg of two-tail conjugate micelle contradicted to the traditional stability theory for micelles generated by linear amphiphilic copolymers [30]. According to the free energy model by Tanford, the presence of two PEG in one two-tail conjugate molecule would facilitate the repulsive interactions between hydrophilic heads, hence leading to the increased equilibrium cross-sectional area (ae) and hence decreased nagg compared to its single-tail counterpart (Fig. 1) [31,32]. In addition, the double-tail topology has a direct influence over ae since the relatively rigid structure forms a packing constraint inside the nano-aggregates, resulting in a lower nagg for Bi(PEG-PLA)-Cur [33]. The hydrodynamic diameter of both micelles was similar at 82 ± 1 nm (PEG-PLA-Cur) and 85 ± 3 nm (Bi(PEG-PLA)-Cur) in turn (Fig. 5). TEM analysis showed spherical micelles with the core diameter at 42 ± 9 nm and 40 ± 5 nm, for single-tail and phospholipid-like conjugate micelle, respectively. Further AFM observation also verified the comparable size of both types of micelles. The zeta potential evaluation showed a negative particle surface with the corresponding value of 15 ± 2 mV (PEG-PLA-Cur) and 11 ± 3 mV (Bi(PEG-PLA)-Cur). The spherical geometry of both types of conjugate micelles can be explained by the concept of critical packing parameter (CPP). For single-tail polymer–drug conjugate, the formation of spherical micellar nanocarriers has been reported previously [10], which is also the case in current study. This was because the actual CPP value located in the range of the theoretical window (
No. of Pages 10, Model 5G
18 November 2014 European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
European Journal of Pharmaceutics and Biopharmaceutics journal homepage: www.elsevier.com/locate/ejpb
2
Research paper
6 4 7
Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin
5 8
Q1
a
9 10 11
Tianjin Key Laboratory for Modern Drug Delivery & High Efficiency, School of Pharmaceutical Science & Technology, Tianjin University, Tianjin, China Institute of Biomedical Engineering, Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China c State Key Laboratory of Medicinal Chemical Biology, Key Laboratory of Bioactive Materials, Ministry of Education, College of Life Sciences, Nankai University, Tianjin, China b
13 12 14
a r t i c l e
1 2 6 8 17 18 19 20 21 22 23 24 25 26 27
Zheng Wang a, Chao Chen a, Qi Zhang a, Min Gao a, Ju Zhang b, Deling Kong c, Yanjun Zhao a,⇑
i n f o
Article history: Received 9 September 2014 Accepted in revised form 7 November 2014 Available online xxxx
Q2
Keywords: Drug delivery Micelle Self-assembly Polymeric conjugate Curcumin
a b s t r a c t To precisely manipulate the intracellular delivery of pleiotropic curcumin, this work reports two types of acid-responsive polymer–curcumin conjugates with different structures. One or two amphiphilic poly(ethylene glycol)-co-poly(lactic acid) (PEG-PLA) copolymer(s) were linked to curcumin via pH-liable hydrazone, producing a linear or phospholipid-like conjugate, respectively. Both conjugates efficiently self-assembled into micellar nanocarriers with enhanced stability in contrast to the paralleling PEG-PLA micelles. In comparison to the micelles assembled by phospholipid-like conjugates, the linear conjugate micelles exhibited similar size, doubled loading dose, higher release rate, and hence enhanced cellular uptake and cytotoxicity at the cost of increased critical micelle concentration. This work highlighted the importance of precise tuning of polymer–drug conjugate architecture in determining the pharmaceutical properties and thus delivery efficiency of the corresponding conjugate micelles. Ó 2014 Published by Elsevier B.V.
29 30 31 32 33 34 35 36 37 38 39 40 41
42 43
1. Introduction
44
Taking the advantage of enhanced permeability and retention (EPR) effect, nanocarriers have been prevalent for delivering active pharmaceutical ingredient (API) to tumours since the mid-1980s [1–3]. Prior to reaching the tumour site, the API and nanocarriers have to travel over a long distance in the systemic circulation. In many cases the API is loaded in the nanocarrier via simply physical encapsulation or adsorption, e.g. commercially available DoxilÒ and AbraxaneÒ. Since the blood circulation maintains a perfect sink condition, the premature drug release during the journey to the tumour target is inevitable no matter how strong the avidity between the API and the nanocarrier is [4,5]. This significantly contributes to both physical side-effects and psychological harm to the patients. Attaching the API to nanocarrier covalently is an effective means to address the issue of premature release [6–8]. However, the rapid breakdown of chemical bonds often requires certain conditions, i.e. specific enzymes, and catalyst. This induces another problem of poor drug release and delayed onset of action upon reaching the target [9,10]. Polymer–drug conjugate micelle offers a versatile platform for addressing the dilemma of premature release and
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
⇑ Corresponding author. School of Pharmaceutical Science & Technology, Tianjin University, 92 Weijin Road, Nankai District, Tianjin 300072, China. Tel.: +86 22 2740 7882; fax: +86 22 2740 4018. E-mail address: [email protected] (Y. Zhao).
on-demand release in the target site [11–13]. The abundance of polymer variety and specialty makes possible the tailored design and manufacture of polymer–drug conjugates with biodegradability, biocompatibility, self-assembly, long circulation, and stimuliresponsive drug release and other desired characteristics for efficient delivery [14–17]. In addition, the tumour tissue is well known to exhibit a low pH (ca. 6.8). This earmark together with the low pH milieu (4.5–6.5) of endosome and lysosome provides a superb sitespecific trigger for on-demand anti-cancer drug delivery. A rich collection of pH-liable cleavable linkers has been reported including hydrazone, hydrazide, oxime, imine, acetal and ketal, which makes ease of fabricating acid-responsive polymer–drug conjugate [18]. The diversity of polymer and API as well as their tailored combination can generate conjugates with multifarious architecture that is believed to make a significant impact on the pharmaceutical properties and ultimately the therapeutic outcome [19–21]. Curcumin (Cur) has been recognized as a pleiotropic agent with anti-inflammatory, antioxidant, anticancer, and many other medical properties as a consequence of its ability to regulate ca. 100 cell signalling pathways [22]. In terms of chemotherapy, it also shows the capability to reverse tumour multi-drug resistance [23]. Efficient curcumin delivery has been suffering from the poor aqueous solubility (ca. 1 lg/mL) and chemical instability, leading to poor bioavailability [24]. The presence of two phenol groups in one curcumin molecule offers the opportunity of drug conjugation with either one or two chain(s) of polymer simultaneously, bringing
http://dx.doi.org/10.1016/j.ejpb.2014.11.002 0939-6411/Ó 2014 Published by Elsevier B.V.
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 2 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115
Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
about different polymer–curcumin conjugate architectures or topologies. Utilizing the hydrophobicity of curcumin (log P = 3.2), Tang et al. connected two segments of hydrophilic poly(ethylene glycol) (PEG) to one molecule of curcumin via beta-thioester bond, producing a structure similar to phospholipid with two hydrophilic tail and one hydrophobic head [25]. Yang et al. employed ester bond to link curcumin and hydrophobic poly(lactic acid) (PLA) that was further tied with PEG, generating linear PEG-PLA-Cur conjugate [10]. All the above reported conjugates showed amphiphilicity and could capably self-assemble into micellar nanocarriers. It was postulated that such topology divergence of polymer–curcumin conjugate would make a significant difference in the conjugate self-assembly behaviour (e.g. critical micelle concentration, aggregation number, and the critical packing parameter), pharmaceutical properties of micellar nano-assemblies (e.g. particle size, drug loading, release rate, stimuli-responsiveness, and cellular uptake), and hence the eventual therapeutic efficacy (e.g. cytotoxicity). Thus the aim of the current study was to correlate the architecture of acid-responsive polymer–curcumin conjugate and the delivery efficiency of corresponding assembled micelles. To achieve this, biodegradable amphiphilic PEG-PLA copolymer was selected as the building block. pH-liable hydrazone was used as the linker between the polymer and curcumin to realize on-demand rapid drug release inside the target cells, but minimized premature release in the systemic circulation. Both linear (PEG-PLA-Cur) and phospholipid-like (Bi(PEG-PLA)-Cur) conjugates were created and compared (Scheme 1). Their architectures were illustrated below (Fig. 1).
Fig. 1. Schematic illustration of the topology of amphiphilic acid-liable PEG-PLACur (single-tail) and Bi(PEG-PLA)-Cur (double-tail, phospholipid-like) conjugate and the factors determining their critical packing parameter (CPP): equilibrium area per conjugate molecule at the aggregate interface (ae), hydrophobic tail volume (v0), and hydrophobic tail length (l0). (For the interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Scheme 1. The synthetic route of two types of acid-responsive polymer–drug conjugates: PEG-PLA-Cur and Bi(PEG-PLA)-Cur. PEG, PLA, and Cur represents poly(ethylene glycol), poly(lactic acid), and curcumin, respectively. Hydrazone is the pH-liable linker between polymer and curcumin.
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx 116
2. Materials and methods
117
2.1. Materials
118
138
Curcumin was purchased from Guangfu Fine Chemical Research Institute (Tianjin, China). 40 ,6-diamidino-2-phenylindole (DAPI) and methoxy poly(ethylene glycol) (PEG, 2000 Da) were obtained from Sigma–Aldrich (Beijing, China). Levulinic acid, stannous octoate, hydrazine monohydrate, pyrene, and 4-dimethylaminopyridine (DMAP) were from Aladdin (Shanghai, China). N-(3Dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDCHCl) was from Medpep Co., Ltd. (Shanghai, China). D,L-lactide was from GLACO Ltd. (Beijing, China). 4-Nitrophenyl chloroformate (NPC) was from Energy Chemical (Shanghai, China). Methanol and tetrahydrofuran (THF) were from Concord Technology Co., Ltd. (Tianjin, China). Dulbecco’s modification of eagle’s medium (DMEM), foetal bovine serum, penicillin–streptomycin, and phosphate buffer solution (PBS) were from HyClone, Inc. (Logan City, Utah, USA). Trypsin was from Life Technologies (Beijing, China). Cell Counting Kit-8 (CCK-8) was from Dojindo Laboratories (Shanghai, China). The human epithelial carcinoma (HeLa) and human hepatocellular liver carcinoma (HepG2) cell lines were provided by Institute of Biomedical Engineering (Chinese Academy of Medical Sciences & Peking Union Medical College). All other chemicals were sourced from Jiangtian Chemicals (Tianjin, China).
139
2.2. Polymeric conjugate synthesis
140
2.2.1. Cur-1 and Cur-2 Carboxyl-terminated curcumin derivatives (Cur-1 and Cur-2) (Scheme 1) were prepared as follows based on a previous report [26]. Levulinic acid (0.0631 g, 0.543 mmol), DMAP (0.0662 g, 0.543 mmol), and EDCHCl (0.1041 g, 0.543 mmol) were mixed in a round-bottom flask containing 20 mL dichloromethane (DCM) for 1 h. Curcumin (0.2 g, 0.543 mmol) was dissolved in 30 mL DCM. The former solution was transferred slowly to the curcumin solution and the reaction was maintained at 25 °C for 24 h with nitrogen protection. Then the crude product was washed in triplicate with water post DCM evaporation and filtration. The collected solid was dissolved in ethyl acetate (EA) following the extraction with hydrochloride acid (HCl) solution; the water residue in the organic phase was removed with sodium sulphate (anhydrous), followed by EA removal to get unpurified Cur-1. The purification was carried out using the silica gel column chromatography with a gradient elution. The starting eluent was a mixture of petroleum ether and EA (2:1, v/v) containing 0.2% (v/v) glacial acetic acid for unreacted curcumin that came out first. Then a mixture of petroleum ether and EA (1:1–1:5, v/v) was used to elute Cur-1 (yield: 50%). The final product was obtained via solvent evaporation and vacuum-drying. 1H NMR (600 MHz, CDCl3), d [ppm]: 2.16 (s, 3H); 2.81 (s, 4H); 3.80 (s, 3H); 3.88 (s, 3H); 5.76–5.81 (d, 2H); 6.41– 6.49 (q, 2H); 6.84–7.10 (m, 6H); 7.50–7.56 (q, 2H) (Fig. S1, Supporting Information/SI). Likewise, the synthesis of Cur-2 was maintained at the same condition except an altered reactant feeding: levulinic acid (0.630 g, 5.43 mmol), DMAP (0.662 g, 5.43 mmol), EDCHCl (1.041 g, 5.43 mmol), and curcumin (0.2 g, 0.543 mmol). The isocratic eluent for Cur-2 purification was 1, 2-dichloroethane (yield: 40%). 1H NMR (600 MHz, CDCl3), d [ppm]: 2.23 (s, 6H); 2.88 (s, 8H); 3.87 (s, 6H); 5.85 (s, 2H); 6.55–6.57 (d, 2H); 7.06–7.17 (m, 6H); 7.58–7.64 (d, 2H) (Fig. S2, SI).
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172
173 174 175
2.2.2. PEG-PLA PEG (5 g, 2.5 mmol) was melted in a 100 mL flask at 80 °C and then lactide (5 g, 34.7 mmol) was added. The mixture was heated
3
until melting, and then stannous octoate (1 mL, 3.1 mmol) was supplemented. The reaction was kept at 125 °C for 24 h under nitrogen atmosphere. Afterwards, the crude product was cooled down to ambient temperature and dissolved in tetrahydrofuran (THF), which was followed by precipitation in ice-cooled diethyl ether. This process was repeated in triplicate and the resultant solid was vacuum-dried at ambient temperature to get PEG-PLA (yield: 93%). 1H NMR (600 MHz, CDCl3), d [ppm]: 1.56 (m, –CH3 PLA repeating unit); 3.38 (s, –CH3 PEG end group); 3.64 (m, – CH2 PEG repeating unit); 5.17 (m, –CH PLA repeating unit) (Fig. S3, SI).
176
2.2.3. PEG-PLA-NPC PEG-PLA (3 g, 0.857 mmol) and pyridine (0.345 mL, 4.286 mmol) were mixed in 50 mL DCM at 0 °C with nitrogen protection. Then NPC (0.691 g, 3.428 mmol) solution in DCM (10 mL) was gently added to the above solution. The mixture was maintained at 0 °C for 1 h and then reacted for 24 h at ambient temperature (25 °C). The crude product was filtered and the filtrate was transferred to appropriate amount of ice-cooled diethyl ether. The obtained white solid was precipitated in triplicate and then vacuum-dried to get PEG-PLA-NPC (yield: 95%). 1H NMR (600 MHz, CDCl3), d [ppm]: 1.56 (m, –CH3 PLA repeating unit); 3.38 (s, –CH3 PEG end group); 3.64 (m, –CH2 PEG repeating unit); 5.17 (m, –CH PLA repeating unit); 7.4 (t, –CH aromatic); 8.28 (d, –CH aromatic) (Fig. S4, SI).
187
2.2.4. PEG-PLA-Cur Activated PEG-PLA-NPC (0.5 g, 0.143 mmol) and zinc powder (0.5 mg, 7.65 lmol) were blended in 40 mL dimethylformamide (DMF) under nitrogen atmosphere. Then hydrazine monohydrate (26 lL, 0.429 mmol) was added to the mixture and the reaction was maintained at 25 °C for 6 h. The crude product was purified by filtration, repeated precipitation in diethyl ether, and then vacuum-dried to get the amino-terminated polymer, PEG-PLANHNH2. Then PEG-PLA-NHNH2 (1.0 g, 0.286 mmol) and Cur-1 (0.2 g, 0.429 mmol) were dissolved in 15 mL DMF, and reacted at ambient temperature for 48 h. The crude product was dialysed against 1000 mL water with light protection using a dialysis tube (molecular weight cut-off/MWCO: 3500 Da). The dialysis medium was maintained at ca. pH 7.0 with the diluted ammonium, and replaced regularly. After 24 h, the dialysed solution was centrifuged for 10 min (1000g), filtered using a microfilter (0.45 lm), and then freeze-dried to get single-tailed polymer–drug conjugate PEG-PLA-Cur (yield: 70%). 1H NMR (600 MHz, CDCl3), d [ppm]: 1.56 (m, –CH3 PLA repeating unit); 2.23 (s, –CH3 spacer); 2.88 (s, –CH2 spacer); 3.38 (s, –CH3 PEG end group); 3.64 (m, –CH2 PEG repeating unit); 5.17 (m, –CH PLA repeating unit); 5.80 (d, 2H); 6.45–6.51 (q, 2H); 6.91–7.16 (m, 6H); 7.57–7.62 (q, 2H); 8.09 (s, NH) (Fig. 2).
201
2.2.5. Bi(PEG-PLA)-Cur The synthesis and purification of double-tailed conjugate Bi(PEG-PLA)-Cur were carried out using a similar approach to that PEG-PLA-Cur described in the above section (yield: 60%). The reactants were PEG-PLA-NHNH2 (1.0 g, 0.286 mmol) and Cur-2 (0.054 g, 0.095 mmol). The dialysis tube had a MWCO of 7,000 Da. 1H NMR (600 MHz, CDCl3), d [ppm]: 1.49 (m, –CH3 PLA repeating unit); 3.31 (s, –CH3 PEG end group); 3.57 (m, –CH2 PEG repeating unit); 3.88 (s, 6H); 5.09 (m, –CH PLA repeating unit); 5.74 (s, 2H); 6.38–6.44 (d, 2H); 6.82–7.10 (m, 6H); 7.49–7.55 (d, 2H); 8.02 (s, –NH) (Fig. 3).
224
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
177 178 179 180 181 182 183 184 185 186
188 189 190 191 192 193 194 195 196 197 198 199 200
202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
225 226 227 228 229 230 231 232 233 234
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 4
Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
Fig. 2. 1H NMR spectrum of PEG-PLA-Cur conjugate in CDCl3. PEG, PLA, and Cur indicated poly(ethylene glycol), poly(lactic acid), and curcumin, respectively. (For the interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
1
Fig. 3. H NMR spectrum of Bi(PEG-PLA)-Cur conjugate in CDCl3. PEG, PLA, and Cur indicated poly(ethylene glycol), poly(lactic acid), and curcumin, respectively. The proton resonance peaks at position 7, 8, and 9 were not seen due to the influence of water peak in the solvent (ca. 1.7–2.5 ppm). 235
2.3. Molecular weight analysis
236
The molecular weight and its distribution of both types of polymer–curcumin conjugates were analysed by a gel permeation chromatography (GPC) system (Malvern Viscotek TDA 305). Both samples were dissolved in THF to get a concentration of 5 mg/
237 238 239
mL. Polystyrene was used as the calibration standard with THF as the eluent at a speed of 1 mL/min at ambient temperature.
240
2.4. Determination of critical micelle concentration and mean aggregation number
242
The conjugate micelles were prepared using the dialysis method (MWCO: 2000 Da) with THF as the organic solvent. The critical micelle concentration (CMC) of micelles was investigated using a FluorologÒ3–21 spectrofluorometer (HORIBA JobinYvon). Micellar aqueous solution (0.4–400 lg/mL) was supplemented with pyrene probe to reach a constant probe concentration of 0.5 lM. The emission spectra were recorded from 350 nm to 450 nm with an excitation wavelength of 333 nm and spectral slit width of 5 nm. The intensity ratio of pyrene probe band at 384 nm and 373 nm was plotted against the logarithm of the conjugate micelles’ concentration and the CMC was determined by getting the flexion point of the sigmoidal curve [10]. The mean aggregation number (nagg) of both polymer–curcumin conjugate micelles was examined by a dynamic light scatter (DLS) method [27,28]. The analysis was carried out in a concentration range of 0.1–12 mg/ mL for both samples. Prior to DLS measurement, the specific refractive index increment of the conjugate aqueous solution was determined by a digital Abbe refractometer (WAY-2S) at ambient temperature. The DLS analysis employed toluene as the standard with the scattering angle fixed at 90°. The Debye plot was generated to get intercept that was reciprocal of the mean molecular weight of one single nanocarrier. The mean aggregation number was obtained simply via dividing the molecular weight of the micelle by that of the corresponding conjugate. All measurements were taken in triplicate.
244
2.5. Particle size and zeta potential analysis
269
The hydrodynamic diameter and zeta potential of the conjugate micelles were analysed in triplicate by a Zetasizer Nano ZS (Malvern Instruments). The measurement was taken at 25 °C with a sample concentration of 5 mg/mL in water. The size and morphology of both types of micelles were determined by a JEM-100 CXII transmission electron microscope (TEM). One drop of aqueous micelle solution (1 mg/mL) was transferred to a collodion support on copper grids followed by air-drying and then the sample is ready for imagerecording. Atomic force microscope (AFM) (Veeco, Multimode) was also utilized to characterize the micelles. The aqueous sample (50 lg/mL) was placed on the surface of microscope glass slide and then freeze-dried prior to AFM analysis via a tapping mode.
270
2.6. Drug loading and release analysis
282
The curcumin content in the conjugate micelles was determined using a Cary 60 UV–vis spectrophotometer. Prior to analysis, the UV–vis spectra of both conjugates, PEG-PLA-NHNH2, and curcumin in methanol were obtained to verify no shift of the wavelength, at which the drug/conjugate showed the maximum absorption. The quantification employed curcumin as the calibration standard with methanol as the solvent. The absorption was recorded at 424 nm (n = 3). The drug release experiment utilized the same method in our previously published work [10]. Briefly, PEG-PLA-Cur (4 mg) or Bi(PEG-PLA)-Cur (9.4 mg) micelle dispersed in 2 mL water was added to the donor compartment of a Franz cell. The receiver compartment (ca. 17 mL) was filled with citric acid/ disodium hydrogen phosphate buffer at one of three designed pH values (5.0, 6.0, and 7.4). The receiver fluid contains 5% (w/v) sodium dodecyl sulphate (SDS) to keep the sink condition. A porous regenerated cellulose membrane (MWCO: 3500 Da) was used to separate the donor and receiver compartment. The release study
283
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
241
243
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
271 272 273 274 275 276 277 278 279 280 281
284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
5
306
was carried out at 37 °C in triplicate and the drug content in the receiver fluid at pre-determined time points was analysed by an Agilent 1100 high performance liquid chromatography (HPLC) system coupled with a UV detector. The cumulative amount of released drug was plotted against time. Due to the chemical instability of curcumin, a mass balance study was also performed to calculate the total drug recovery at the end of experiments.
307
2.7. In vitro cytotoxicity assay
308
323
HeLa or HepG2 cells were cultured in DMEM medium supplemented with 10% foetal bovine serum and 100 kU/L penicillin– streptomycin. The cells were seeded in 96-well plates at a density of 8 103 per well in 100 lL medium and incubated at 37 °C and 5% CO2. After 24 h, the medium was removed and the cells were washed with 100 lL PBS. Then 200 lL micelle-containing fresh medium was added to the plate; wherein the drug concentration ranged from 5 lg/mL to 100 lg/mL. After another 24 h’s incubation, the drug-containing medium was discarded and the cells were washed with PBS three times followed by the addition of 100 lL medium together with 10 lL CCK-8 solution, and further incubation for 30 min. Afterwards, the spectrophotometric measurement at 459 nm was taken and fresh cell medium was used as the control. The half maximal inhibitory concentration (IC50) of different micelles was calculated using a previously reported method (n = 4) [10]. 2.8. Cellular uptake study
3.1. Characterization of polymer–curcumin conjugate and assembled micelles
355
324 325
HeLa or HepG2 cells were seeded to a CLSM plate at a density of 1 104 cells per well in NEST-glass bottom cell culture dish (15 mm) containing 1 mL DMEM medium. After 24 h’ incubation, the medium was removed and the cells were washed with 0.5 mL PBS twice, followed by the addition of 200 lL fresh medium and then 200 lL micelle solution (100 lg/mL). At predesigned time point (2 h, 6 h, and 12 h), the drug-containing medium was withdrawn and 1 mL PBS was added to wash the cells in triplicate. Then
Two types of polymer–curcumin conjugates were synthesized, i.e. PEG-PLA-Cur and Bi(PEG-PLA)-Cur (Figs. 2 and 3). The GPC analysis revealed that the two-tail Bi(PEG-PLA)-Cur exhibited a roughly doubled molecular weight (MW) compared to the single-chain PEG-PLA-Cur (Fig. 4a). However, the MWs determined by GPC were higher compared to that obtained by 1H NMR (Table S1, SI). Although GPC is the most routinely used method for MW analysis of polymers, it is a relative approach with strong dependence of its
357
300 301 302 303 304 305
309 310 311 312 313 314 315 316 317 318 319 320 321 322
326 327 328 329 330 331 332
the cells were fixed with 1 mL paraformaldehyde (4%) for 20 min at 4 °C with light protection. Subsequently, the paraformaldehyde was removed following the cell rinse by 1 mL PBS twice. Then the cells were stained with 300 lL DAPI solution (1 lg/mL) for 5 min and washed with 1 mL PBS ready for CLSM analysis with the excitation wavelength at 488 nm.
333
2.9. Flow cytometry study
339
Flow cytometry was also employed to analyse cellular the uptake of micelles. HepG2 cells were seeded to a petri dish (60 15 mm) at a density of 1 104 cells in 1 mL medium and cultured for 24 h. Afterwards, the cells were washed with 1 mL PBS and further incubated in micelle-containing DMEM medium with a drug concentration fixed at 50 lg/mL. After 24 h, culture medium was removed and cells were washed with 1 mL PBS followed by the treatment with 1 mL trypsin for 3 min at 37 °C. Then 3 mL medium was added to each culture well and the solutions were centrifuged using an Eppendorf 5810 R centrifuge at 1000 rpm for 5 min. The supernatant was removed and the cells were suspended in 1 mL paraformaldehyde (4%) at 4 °C with light protection. The fluorescence was analysed via the BD FACSCalibur flow cytometer and the cells without micelle treatment were used as the control.
340
3. Results and discussion
354
Fig. 4. The characterization of acid-responsive polymer–curcumin conjugates: (a) molecular weight determined by GPC; (b) mean aggregation number analysed by DLS; (c and d) CMC of PEG-PLA-Cur and Bi(PEG-PLA)-Cur obtained by fluorescence techniques (n = 3). (For the interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: Z. Wang et al., Tuning the architecture of polymeric conjugate to mediate intracellular delivery of pleiotropic curcumin, Eur. J. Pharm. Biopharm. (2014), http://dx.doi.org/10.1016/j.ejpb.2014.11.002
334 335 336 337 338
341 342 343 344 345 346 347 348 349 350 351 352 353
356
358 359 360 361 362 363 364
EJPB 11739
No. of Pages 10, Model 5G
18 November 2014 6 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430
Z. Wang et al. / European Journal of Pharmaceutics and Biopharmaceutics xxx (2014) xxx–xxx
data on the calibrant and polymer architecture. In spite of the simplicity of 1H NMR, its utility in determining polymer MW has been somewhat underappreciated since it cannot get the information of weight-average molecular weight. Therefore, the combination of both methods would produce more reliable MW information of polymers. Although the polymer polydispersity index (PDI) obtained by GPC was slightly higher, it was still within the acceptable level since the choices of both calibrant standard and elution solvent would affect the PDI results. Both conjugates could successfully self-assemble to micelles. For two-tail conjugate, both CMC and nagg were lower compared to that of single-tail conjugate and the calculation utilized the MW data determined by GPC (Fig. 4b–d). The CMC of a typical nonionic single-chain amphiphilic copolymer usually decreases with the growth of hydrophobic moiety; meanwhile the nagg would move towards an opposite direction, which was due to the elevated driving force of micelle formation, i.e. to minimize the interfacial free energy [29]. In the current study, with the aid of a fluorescent probe, the two-tail conjugate’s CMC (0.6 ± 0.1 lM) was four times lower in contrast to that of single-tail conjugate (2.6 ± 0.3 lM), which manifested a decreased free energy in the former system. This was presumed as a consequence of the elevated hydrophobicity of the two-tail conjugate. The cohesion of hydrophobic core protected by the hydrophilic PEG moiety was thus enhanced, giving rise to a lower CMC and improved stability. The stability enhancement was crucial and advantageous to maintain the integrity of the micellar nanocarriers upon large-ratio blood dilution post intravenous dose administration. The nagg value of both conjugate micelles was 101 ± 3 (single-tail) and 70 ± 5 (double-tail), correspondingly. The reduced nagg of two-tail conjugate micelle contradicted to the traditional stability theory for micelles generated by linear amphiphilic copolymers [30]. According to the free energy model by Tanford, the presence of two PEG in one two-tail conjugate molecule would facilitate the repulsive interactions between hydrophilic heads, hence leading to the increased equilibrium cross-sectional area (ae) and hence decreased nagg compared to its single-tail counterpart (Fig. 1) [31,32]. In addition, the double-tail topology has a direct influence over ae since the relatively rigid structure forms a packing constraint inside the nano-aggregates, resulting in a lower nagg for Bi(PEG-PLA)-Cur [33]. The hydrodynamic diameter of both micelles was similar at 82 ± 1 nm (PEG-PLA-Cur) and 85 ± 3 nm (Bi(PEG-PLA)-Cur) in turn (Fig. 5). TEM analysis showed spherical micelles with the core diameter at 42 ± 9 nm and 40 ± 5 nm, for single-tail and phospholipid-like conjugate micelle, respectively. Further AFM observation also verified the comparable size of both types of micelles. The zeta potential evaluation showed a negative particle surface with the corresponding value of 15 ± 2 mV (PEG-PLA-Cur) and 11 ± 3 mV (Bi(PEG-PLA)-Cur). The spherical geometry of both types of conjugate micelles can be explained by the concept of critical packing parameter (CPP). For single-tail polymer–drug conjugate, the formation of spherical micellar nanocarriers has been reported previously [10], which is also the case in current study. This was because the actual CPP value located in the range of the theoretical window (
