2017 Volume 65 Issue 6 Pages 530-537
RSPP050 (AG50) is one of the semi-synthetic andrographolide that is isolated from Andrographis paniculata NEES (Acanthaceae). The anti-proliferation effects of AG50 against cholangiocarcinoma (HuCCT1) were displayed high cytotoxicity. Unfortunately, poor water solubility of AG50 limited its clinical applications. This study aimed to increase the concentration of AG50 in water and drug loading and release study in phosphate-buffered saline (PBS) in the absence/presence of pig liver esterase enzyme. Cytotoxicity of AG50-loaded polymeric micelles was evaluated against HuCCT1. AG50 loaded micelles were prepared by film sonication and encapsulated by polymers including poly(ethylene glycol)-b-poly(ε-caprolactone) (PEG-b-PCL) or poly(ethylene glycol)-b-poly(D,L-lactide) (PEG-b-PLA). Micelle properties were characterized such as solubility, drug loading, drug release and in vitro cytotoxicity against HuCTT1. AG50 was successfully loaded into both types of polymeric micelles. The best drug–polymer (D/P) ratio was 1 : 9. AG50/PCL and AG50/PLA-micelles had small particle size (36.4±5.1, 49.0±2.7 nm, respectively) and high yield (58.2±1.8, 58.8±2.9, respectively). AG50/PLA-micelles (IC50=2.42 µg/mL) showed higher cytotoxicity against HuCCT1 than AG50/PCL-micelles (IC50=4.40 µg/mL) due to the higher amount of AG50 released. Nanoencapsulation of AG50 could provide a promising development in clinical use for cholangiocarcinoma treatment.
Andrographolide (AG, Fig. 1a) is identified from diterpenoid lactone, which is isolated from Andrographis paniculata (Burm.f.) NEES (Acanthaceae). This herb is widely used in Southeast Asia. AG is commonly used for antibacterial, anti-inflammatory and anticancer; however, it exhibits low potency for cancer inhibition that can be improved by modification of its chemical structure. One of the many modified AGs is RSPP050 (AG50, Fig. 1b) has high toxicity against cancer cell lines.1–3) The worldwide incidence of cholangiocarcinoma steadily increase. Cholangiocarcinoma account for 3% of all gastrointestinal tumors and 10–15% of all hepatocellular carcinoma and it is second most common primary liver tumor.4) Unfortunately, andrographolide and its analogs exhibit poor water solubility (<1 µg/mL) which is an obstacle for AG50 development and medical applications; therefore, it is necessary to increase its water-solubility while maintaining its chemical structure5,6) to retain its toxic effect. Hydrophobic compound-loaded polymeric micellar nanocarriers (10–200 nm) have been reported to overcome these limitations: high toxicity, low water solubility, and short circulation time by encapsulating the drug and protecting it from harsh condition (i.e., immune system and elimination system), polymeric micelles’ favorable properties overcoming these limitations lead to prolong circulation time in blood.7) Additionally, it can be used to achieve high drug accumulation in tumor tissues by the enhanced permeability retention (EPR) effect.8) This allowed macromolecules or nanoparticles to accumulate in the tumor tissues due to the large porous blood vessel. In contrast, the small porous endothelial of normal blood vessel allowed only small molecules or free drugs. Furthermore, micelles can be transported and retained within cells to a higher extent than free drug probably due to the mechanism of endocytosis that might effectively escape the multi-drug resistance (MDR) effect that can be found in free drugs.9)
Andrographolide was substituted by 19-triphenylmethyl ether to be (b) RSPP050.
Nanocarrier is a promising strategy to increase efficacy of chemotherapeutic agents. Several kinds of nanocarriers including liposome, micelle, dendrimer, and nanocrystal have been used for drug delivery systems.5,10–12) Polymeric micelles were selected as a nanocarrier in this study. Polymeric micelles are self-assembling nanostructure comprising of core and shell which are made up by amphiphilic copolymers. The molecular weight of these copolymers can be varied which in turn controls the particle size, drug encapsulation, and drug release of the polymeric micelles.13,14) Several methods for micelle preparations have been examined to encapsulate hydrophobic drug in polymeric micelles. Solvent evaporation method, film sonication, and dialysis method are widely used for micelle preparation.15) Previous works suggested that film sonication is the most suitable for high semi-synthetic hydrophobic compound loading due to its short-time organic solvent exchange.15) On the other hand, solvent evaporation method and dialysis method are time consuming, which lead to lesser drug loading in the same amount of time. Film sonication method was employed for micelles preparation because this method provided high drug loading content as a result of short preparation time.14,15)
Block copolymers with amphiphilic property were used to prepare micelles. These polymers composed of two different segments16) including the outer shell segment and inner core segment. For the outer segment, polyethylene glycol (PEG) is most commonly used as the hydrophilic segment in order to prevent protein absorption, immune recognition, and to exhibit non-toxicity and biocompatibility.17–20) Poly(D,L-lactide) (PLA) and poly(ε-caprolactone) (PCL) can be used as the inner segment which requires the hydrophobic property and holds various advantages such as non-toxicity, biocompatibility, and biodegradable property20–22) or enzyme23) in cytoplasm of the cells.24) Therefore, PEG-b-PCL and PEG-b-PLA are excellent block copolymers for encapsulation of hydrophobic compounds, prolonged circulation time in blood and large accumulation amounts of particles in the tumor tissues.
This study aimed to enhance water solubility of andrographolide using polymeric micelles, i.e., PEG-b-PCL and PEG-b-PLA. Polymeric micelles were prepared by film sonication using different ratio of drug–polymer (D/P) (1 : 9, 2 : 8, 3 : 7). Drug release profile was measured in phosphate-buffered saline (PBS) at pH 7.4 in the absence or in the presence of an enzyme (porcine esterase) to mimic the physiological pH and ion strength and cytoplasm, respectively. Effects of the enzyme on degradation of polymers were evaluated using porcine liver esterase (PLE). Finally, cytotoxicity against HuCCT1 with the best D/P loading ratio was determined in order to confirm potential efficacy of AG50-loaded polymeric micelles.
Methoxy-terminated PEG with the Mn of 5000 Da (MeO–PEG–OH) was purchased from Fluka (Buchs, Switzerland) and was recrystallized in ethyl acetate. D,L-Lactide was purchased from Shenzhen Brightchina Industrial (Brightchina, China) and was purified by recrystallization in ethyl acetate. ε-Caprolactone was purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.) and was purified by distillation with calcium hydride (CaH2). Stannous (ІІ) octoate (Sn(Oct)2) was obtained from Aldrich. All of the organic solvents including tetrahydrofuran (THF), dimethyl sulfoxide (DMSO), and chloroform (CHCl3) were purchased from RCI Lab-Scan Ltd. (Wilwaukee, WI, U.S.A.). Toluene was refluxed with sodium–benzophenone under argon. PBS 0.01 M at pH 7.4 was prepared by mixing 8.00 g sodium chloride (NaCl), 0.2 g potassium chloride (KCl), 1.44 g disodium hydrogenphosphate (Na2HPO4) and 0.24 g potassium phosphate monobasic (KH2PO4) in 1000 mL distilled water. AG50 was kindly provided by Prof. Rungnapha Saeeng (Department of Chemistry, Faculty of Science, Burapha University, Chon Buri, Thailand). The purity of AG50 was assessed by HPLC and was approximately 99%. Esterase solution from porcine liver (10 units/mL, 1 mL, E-code: 3.1.1.1, Sigma-Aldrich). Cholangiocarcinoma cells (HuCCT1) was purchased from Japanese Collection of Research Bioresources Cell Bank (Tokyo, Japan). It was cultured by Roswell Park Memorial Institute (RPMI) that was originally obtained from Gibco (Grand Island, New York, U.S.A.). Added supplemental compounds were 100 U/mL of penicillin, 100 µg/mL of streptomycin and 10% fetal bovine, which were obtained from JR Scientific Inc. (Woodland, CA, U.S.A.). These cell lines were cultured in an incubator and controlled in physiological pH, which is 5% CO2 in humidified atmosphere at 37°C.
Synthesis of PEG-b-PLA and PEG-b-PCL Block CopolymersPEG-b-PLA and PEG-b-PCL were synthesized by ring-opening polymerization. First, the humidified MeO–PEG–OH (2 g, 0.4 mmol) was dried by applying heat at 65°C and subsequently vacuum. The exact amount of D,L-lactide or ε-caprolactone added into the flask. Then, anhydrous toluene was added into MeO–PEG–OH under argon atmosphere. Lastly, the catalyst Sn(Oct)2 was added. The whole process was treated under 140°C.13,25)
Water Solubility StudySolubility of AG50 in water was detected by UV-Vis spectroscopy. First, 1 mg of AG50 was dissolved in 2.5 mL of THF. The mixture was underwent orbital shaking in the fume hood for 3 d to allow THF to evaporate. It is the conventional protocol to evaporate THF. It was centrifuged at 3000 rpm for 10 min at 4°C and 0.45 µm syringe filter was used for eliminating any water-insoluble AG50. Supernatants, which are the water-soluble AG50 was lyophilized. DMSO (5 mL) was added to the lyophilized sample, and the AG50 content was measured at the wavelength of 260 nm.26) Moreover, AG50 solubility in PBS containing 10 units of porcine esterase (PLE/PBS) was determined by incubating excess AG50 (1 mg) in 1 mL (n=3) of each esterase solution at 37°C for 24 h with agitation. Samples were centrifuged at 10000 rpm for 20 min to separate a supernatant. AG50 dissolved in PLE/PBS was filtered with 0.45 µm nylon syringe filters, lyophilized, extracted with ethyl acetate, and analyzed with 270 nm of UV-Vis.27) For solubility of AG50-encapsulated micelles, micelles from different formulations were centrifuged at 3000 rpm for 120 min by centrifugal filter. The volume of water that was filtered off was controlled to be the same for all micelle formulations. Their micelles were centrifuged until micelle solution started to be significantly viscous.
Preparation of AG50-Loaded Polymeric MicellesFilm sonication method was applied to prepare AG50-loaded polymeric micelles. AG50 and block copolymer were dissolved in tetrahydrofuran (THF); THF was evaporated from the sample in a glass vial by rotary vacuum evaporator (IK, RV10) to produce dry film. Distilled water was added and the solution was sonicated for 1 min by Sonic-VibraCell™ (Model CV.18).15,28)
Particle Size and Determination of AG50 Loading ContentThe particle sizes were measured by laser light scattering (Zetasizer Nano ZS, Malvern). The measurement was carried out at the scattering angle of 90° and at 25°C. The final concentration of solution was determined at 2 mg/mL. Drug loading content was determined by UV-Vis spectrophotometer. Free drugs and solutions of AG50-loaded micelle were lyophilized and the amount of AG50 was measured by dissolving it in chloroform after that the solution was measured at the wavelength of 242 nm. Drug loading content, drug loading efficiency, and micelle yield were obtained by the following set of equations.29–32)
 | (1) |
 | (2) |
 | (3) |
First, free AG50 was removed from micelles by millipore centrifugal filter (Vivaspin from G&E Healthcare) with 50 kDa molecular weight cut-off from purified AG50-loaded micelles. The purified micelle solution was placed in dialysis bags (Spectra/por dialysis from spectrumlab) and incubated in 20 mL of PBS (pH 7.4). For the study of an enzymatic effect, porcine liver esterase (PLE) solution (10 units/mL) was added into dialysis bag with the molecular weight cut off at 50 kDa and subsequently placed into glass bottles containing 20 mL of PBS (pH 7.4). Release study was carried out in an incubator shaker at 37°C and 90 rpm. Sink conditions were performed 2 mL of micelle solution dialysis against large volume of PBS, pH 7.2 (20 mL) of the same medium. At desired time intervals, 20 mL release media was taken out and replenished with an equal volume of fresh media and added 2 mL of ethanol to measure the amount of AG50 by UV-Vis spectrophotometer.26) This experiment was carried out in triplicate and the release curve was fitted with different mathematical models including Zero order, First order, Higuchi, Korsmeyer–Peppas, Hixson–Crowell, Hopfenberg, Baker–Lonsale, Makoid–Banakar, Peppas–Sahlin, Quadratic, Weibull and Logistic Model.33)
Cytotoxicity Study of AG50-Loaded Copolymeric MicellesCytotoxicity test of micelles was carried out against cholangiocarcinoma cells (HuCCT1) and was tested by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay following the standard procedure (Molecular Probes, V-13154).34) HuCCT1 (5×103 cells) were seeded into each well of 96-well plates in 100 µL medium. After 48 h of co-incubation, cells were washed with medium (RPMI1640) and treated by various drug concentrations in the medium. The cell viability was determined after 72 h by MTT assay. The medium was removed and cells were incubated with fresh medium containing 0.5 mg/mL of MTT for 4 h. The formazan crystal was dissolved by adding 100 mL of DMSO. The intensity was determined at 570 nm using M965 Micro-plate reader (Metertech Inc.). The cell viability were measured as percentage of cell death with AG50 concentrations compared to untreated cells (the controlled group). Cell viability was calculated from the following equation.
 |
All experiments were carried out in triplicate and represented as mean±standard deviation (SD). The independent samples t-test was performed using SPSS Statistics 17.0. Statistical significance differences were considered when the p-value is less than 0.05.
PEG-b-PLA and PEG-b-PCL copolymers were synthesized by ring opening polymerization with the number-average molecular weight of 9.8 and 9.7 kDa respectively. PEG-b-PCL and PEG-b-PLA were characterized by integral peak ratio of 1H-NMR spectroscopy with intensity of the strong characteristic peaks of methoxy-PEG (mPEG) at δ=3.66 ppm and the signal of methylene proton of PCL at δ=2.31 ppm and the strong signal of PLA at δ=1.46 ppm.
Water Solubility StudyThe solubility of AG50-loaded PEG-b-PCL micelles at different drug to polymer ratios (1 : 9, 2 : 8, 3 : 7) were compared to the solubility of free AG50 (0.94 µg/mL) and results revealed an increase in the solubility values approximately 300-, 600-, and 900-fold, respectively. In the same manner, AG50-loaded PEG-b-PLA micelles also increased the water solubility of AG50 approximately 300-, 600-, and 200-fold, respectively. Results showed that the large amount of compound was loaded into the core of micelles due to the higher drug to polymer ratio resulting in higher AG50 solubility. However, the water solubility of AG50-loaded PEG-b-PLA micelles was limited at the ratio of 2 : 8 due to the loss of micelle stability leading to precipitation at the ratio of 3 : 735) as shown in Fig. 2. AG50 solubility in Esterase/PBS is 55.75±4.54 µg/mL.
Nanoscale drug carriers are preferentially accumulated at tumor tissues due to many pathophysiological factors such as extravasation of nanocarriers by leaky tumor vasculatures that promotes passive accumulation of drugs in the tumors, a phenomenon called enhanced permeability and retention (EPR) effect; therefore, selection of a proper particle size (10–200 nm) is vital. Results show that all ratios of drug-to-polymer provided AG50-loaded micelles with the proper size (30–60 nm). The type of polymers used produced noticeable effect on the size of micelles where the size of PEG-b-PCL micelles was smaller than that of PEG-b-PLA; this was due to the more hydrophobic property of PEG-b-PCL that lead to less swollen structure and more compact micelle core.14) Micelle size, drug loading contents (% DLC), encapsulation efficiency (% EE) and yields (% yields) of AG50-loaded PEG-b-PCL and PEG-b-PLA micelles were shown in Table 1.
| Block copolymer | Initial ratio of AG50 : polymer | Micelle size (nm) | DLC (%) | EE (%) | Yield (%) |
|---|---|---|---|---|---|
| PEG(5k)-b-PCL(5k) | 1 : 9 | 36.4±5.1 | 9.7±0.3 | 55.6±0.3 | 58.2±1.8 |
| 2 : 8 | 44.4±8.1 | 18.3±0.2 | 59.6±0.4 | 33.2±1.2 | |
| 3 : 7 | 53.5±2.4 | 29.1±0.6 | 60.5±1.0 | 34.6±1.3 | |
| 4 : 6 | 61.9±0.9 | 35.4±1.6 | 19.1±0.4 | 11.7±1.2 | |
| PEG(5k)-b-PLA(5k) | 1 : 9 | 49.0±2.7 | 8.2±0.2 | 54.2±2.4 | 58.8±2.9 |
| 2 : 8 | 54.2±2.3 | 17.8±0.6 | 56.2±1.6 | 30.7±1.2 | |
| 3 : 7 | 64.9±0.2 | 27.4±1.8 | 14.1±0.2 | 5.1±0.4 | |
| 4 : 6 | 82.7±3.3 | 31.4±1.1 | 13.0±0.5 | 6.6±1.2 |
Drug-to-polymer ratio influenced particle size and drug loading content. It was observed that an increase in the drug concentration led to an increase in the mean particle size and drug loading content (% DLC). This was due to the high hydrophobic interaction between drug molecules and polymer; it was reported that % DLC of a hydrophobic drug could go up to 70% while % DLC of a hydrophilic drug was approximately 10%.36) The decrease in the production yield with increasing drug–polymer ratios for both types of polymers was probably due to an increase in the diffusion rate of solvent from the core as polymer ratio decreased. Therefore, the best ratio of AG50 to polymer for further studies including drug release study and cytotoxicity test was 1 : 9 due to its high micelle yields and acceptable encapsulation efficiency.
In Vitro Release StudyAG50 was encapsulated in two different polymeric micelles including PEG-b-PCL and PEG-b-PLA by film sonication method; in vitro release of AG50-loaded PEG-b-PCL and PEG-b-PLA micelles was carried out in buffer (PBS, pH 7.4) at 37°C. The amount of AG50 release was measured at the wavelength of 203 nm. Moreover, porcine liver esterase (PLE) was also used to study enzymatic degradation’s effect on the drug release. PLE, one type of carboxylesterase, was chosen for in vitro studies. It should be noted that this enzyme overexpresses in the cytoplasm of various cells, especially lung, colon and liver tumors and plays important roles in the metabolism of lipid and the splitting of ester bond into acid and alcohol.37–39)
The release profiles of AG50 from PEG-b-PCL and PEG-b-PLA micelles throughout 1 month are shown in Fig. 3. In Fig. 3b, the amount of AG50 released from PEG-b-PLA micelles was 0.4 and 1.17 mg in PBS and PLE, respectively. This was around 2 times greater than that of PEG-b-PCL micelle, which was 0.2 and 0.7 mg in PBS and PLE, respectively. In Figs. 3a and b, the cumulative release of AG50 from PEG-b-PLA micelles was 25% in PBS and 78% in enzyme, which was faster than PEG-b-PCL micelles at 10% in PBS and 46% in enzyme with noticeable cumulative release improvement as the environment changes from PBS to PLE. It was found that the cumulative AG50 release was increased by 53 and 36% for PEG-b-PLA and PEG-b-PCL micelles, respectively. The release enhancement of polymeric micelles by PLE environment, which was 1.03 mg or 63% for AG50-loaded PEG-b-PLA micelles, and 0.78 mg or 40% for AG50-loaded PEG-b-PCL micelles were shown in Fig. 3c. The release rate of AG50 from PEG-b-PLA micelles was faster than PEG-b-PCL micelles. For example, 10% of AG50 released from PEG-b-PLA micelles at 72 h compared with 5% for PEG-b-PCL micelles. This was due to the higher hydrophobicity and crystalline structure of PCL core that exhibited slower release behavior in comparison to lower hydrophobicity and amorphous PLA core in PBS (pH 7.4) and PLE (enzymatic solution).40) The burst release of AG50 was attributed to the hydration of AG50 located near the surface, and the subsequent slow rate of release can be related to AG50 that located in the core of micelles.
Experimental data was fitted by Korsmeyer–Peppas model (solid lines). The symbol (**) indicates significant statistical difference of AG50 release from polymer in enzyme compared to PBS (** p<0.01). c) AG50 release enhancement from PEG-b-PLA (square) and PEG-b-PCL (triangle) micelles by enzyme for mg (solid symbol) and percent (dash symbol) and d) release rate of AG50 from PEG-b-PLA (triangle) and PEG-b-PCL (square) micelles in PBS (filled symbols) and enzyme (open symbols).
AG50 release relied mainly on diffusion mechanism where the release rate of AG50 in PLE was higher than PBS. On the third day, the release rate of AG50-loaded PEG-b-PCL and PEG-b-PLA micelles in PLE were 3.8 and 4.7 µg/h, respectively while they were 1.4 and 2.1 µg/h in PBS as shown in Fig. 3d. Based on the structure of polymeric micelles, it is possible that the release rate of AG50 is controlled by both the rate of hydrolysis of ester bond between polymers and its rate of diffusion. Therefore, enzymes that play important role for hydrolysis of ester bond can enhance the AG50 release from AG50-loaded micelles.
The release study in esterase enzyme which mimics the cytoplasm of cancer cells showed an increase in AG50 release due to enzymatic cleavage of polymer. The release of AG50 from PEG-b-PCL and PEG-b-PLA micelles were enhanced by 218 and 225% at 6 h and 472 and 330% on day 27, respectively compared to the release in PBS. Many research reported that enzymatic biodegradation occurred mainly on the surface of micelles. This was due to the difficulty of hydrophilic enzyme to diffuse through a hydrophobic core. Based on previous reports, PEG-b-PLA micelles could take over 4 d to degrade41,42) whereas PEG-b-PCL micelles could easily be degraded within 10 d.40) This result suggested that the influence of enzyme on polymer cleavage is based on hydrophobicity due to the easier release of AG50-loaded PEG-b-PLA micelles, which is less hydrophobic than that of PEG-b-PCL micelles. As such, this suggested that hydrophobic structure impede enzyme from accessing the micelle’s core.
In conclusion, such controlled release of AG50 can be related to the hydrophobic interaction of AG50 to the polymer, which causes the release process to be predominantly controlled by ester bond cleavage of polymers and simple diffusion. Release study of AG50-loaded polymeric micelles was done in PBS at pH 7.4 to simulate blood circulation environment where only small amount of AG50 was released. On the other hand, more AG50 were released from micelles in PLE at the faster rate and greater amount than PBS. It should be noted that AG50 release from PEG-b-PLA micelles was faster than PEG-b-PCL micelles. This result suggested that enzymes can help regulating the release of AG50 from different types of polymeric micelles.
In order to describe the behavior of drug release from polymer matrix, different mathematical models were chosen to explain the mechanism of drug release and find the best fitting model for the release profiles. Twelve release models were selected including Zero order, First order, Higuchi, Korsmeyer–Peppas, Hixson–Crowell, Hopfenberg, Baker–Lonsale, Makoid–Banakar, Peppas–Sahlin, Quadratic, Weibull and Logistic Model. Results showed that mathematical models that provided good fitting for drug release in PBS were Korsmeyer-Peppas, Makoid-Banaker, Peppas-Sahlin, and Weibull (R2adjusted>0.9) as shown in Table 2. The best model for AG50-loaded PEG-b-PCL and PEG-b-PLA micelles in both enzyme and PBS was Makoid-Banakar with R2adjusted>0.99. Drug release in PBS occurred by diffusion mechanism as confirmed by both n values of Korsmeyer–Peppas model that were less than 0.43. In contrast, drug release in enzyme occurred from a combination between diffusion and polymer relaxation as confirmed by n values of Korsmeyer–Peppas model that were between 0.43 and 0.85. Furthermore, Peppas–Sahlin model also suggested that AG50 mainly diffused through polymer matrix in the micelle core as indicated by the high value of K1 (diffusion) and small value of K2 (polymer relaxation). Comparison of in vitro AG50 release from two types of block copolymers revealed that K1 value of AG50 release from PEG-b-PLA micelles was higher than that of PEG-b-PCL micelles.
| Formulation | Korsmeyer–Peppas | Makoid–Banakar | Peppas–Sahlin | Weibull | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Kkp | n | KMB | n | c | K1 | K2 | m | α | β | Ti | |
| PEG-b-PCL (water) | 4.157 | 0.251 | 3.861 | 0.338 | 0.007 | 4.279 | −0.444 | 0.391 | 23.06 | 0.253 | 0.053 |
| PEG-b-PLA (water) | 6.666 | 0.376 | 5.622 | 0.516 | 0.009 | 5.854 | −0.322 | 0.553 | 14.52 | 0.406 | 0.062 |
| PEG-b-PCL (enzyme) | 8.929 | 0.459 | 8.361 | 0.509 | 0.003 | 8.675 | −0.220 | 0.516 | 12.50 | 0.572 | −0.12 |
| PEG-b-PLA (enzyme) | 12.32 | 0.517 | 9.713 | 0.676 | 0.009 | 9.411 | −0.256 | 0.718 | 15.31 | 0.886 | −0.67 |
In-depth analysis of the release mechanism of AG50 from two types of micelles was carried out by the Korsmeyer–Peppas equation as shown below.
 | (4) |
AG50 release profile from both types of micelles was divided into two stages, i.e., day 0 to day 1 and day 1 to day 40. Results in Fig. 4 show good curve fitting for these two stages for both types of polymeric micelles in both in PBS and enzyme. In PBS, all of n values of AG50-loaded PEG-b-PCL and PEG-b-PLA micelles were less than 0.43 indicating combination mechanisms between diffusion and erosion control. On the other hand, the n values for both stages of AG50-loaded PEG-b-PCL and PEG-b-PLA micelles in the presence of enzyme were between 0.43 and 0.85 demonstrating diffusion and polymer relaxation. It was found that the k constant was increased when the environment was changed from PBS to enzyme (Table 3) suggesting that AG50 release rate was increased in the presence of enzyme. This was due primarily to the role of enzyme in causing the polymer relaxation that facilitated the drug diffusion process. AG50 release from PEG-b-PLA micelles showed k values at 0.077 and 0.078 for the first and second stage in PBS, and 0.135 and 0.102 for the first and second stage in enzyme. These values were higher than that of PEG-b-PCL micelles at 0.04 and 0.039 for the first and second stage in PBS, and 0.104 and 0.064 for first and second stage in enzyme, respectively. This demonstrated that the release rate of AG50 from PEG-b-PLA micelles was faster than PEG-b-PCL micelles. The reason might be that the high hydrophobicity of PCL interacted with AG50 and resulted in slow AG50 release. In conclusion, the release of AG50 was enhanced by polymer relaxation as a result of polymer hydrolysis by the enzyme.
| Polymer | Conditions | n (0–1 d) | k (0–1 d) | n (1–40 d) | k (1–40 d) |
|---|---|---|---|---|---|
| PEG-b-PCL | PBS | 0.267 | 0.037 | 0.398 | 0.039 |
| Enzyme | 0.464 | 0.104 | 0.534 | 0.064 | |
| PEG-b-PLA | PBS | 0.195 | 0.077 | 0.335 | 0.078 |
| Enzyme | 0.474 | 0.135 | 0.577 | 0.102 |
Cytotoxicity of AG50 and AG50-loaded micelles were evaluated against cholangiocarcinoma cells (HuCCT1) using MTT assay. Results showed morphology change of HuCCT1 cells after being treated by AG50 and AG50-loaded polymeric micelles. Due to the cytotoxicity of AG50, cell debris can be observed as circle spot on Fig. 5, which were obvious sign of cell death; cytotoxicity was observed as the dose of AG50 was elevated. Therefore, using high concentration of AG50 caused extensive morphology changes and a lot of cell debris. As shown in Fig. 6, results showed that cell viability of free AG50 (IC50=0.98 µg/mL) was slightly lower than that of AG50-loaded polymeric micelles where the IC50 was 2.42 and 4.40 µg/mL for PEG-b-PLA and PEG-b-PCL, respectively. This was due to the gradual and prolonged release of AG50 from micelles; therefore, its cytotoxicity is slightly lower than free drug. The higher release rate of AG50 released from PEG-b-PLA micelles caused approximately two times more toxicity than that of PEG-b-PCL micelles. It should be noted that blank micelles displayed no cytotoxic effects. Since cytotoxicity of AG50-loaded polymeric micelles was less than free AG50, polymeric micelles can provide advantages in in vivo with high accumulation in tumor tissues via EPR effect and targeting delivery.9)
The error bars represent mean and S.D. of triplicated experiments.
AG50 exhibited high cytotoxicity against cholangiocarcinoma cells (HuCCT1); however, it has poor water solubility and it is cytotoxic to normal cells. This study successfully presented the preparation of AG50 in two types of polymeric micelles (PEG-b-PCL and PEG-b-PLA) by film sonication method. The size of these micelles was in the range of 10 to 100 nm and the solubility of AG50 was raised as high as 900-fold compared to free AG50. Release studies revealed slower release of AG50 from polymeric micelles in PBS (pH 7.4) compared to the release study in the presence of enzyme (porcine esterase). Encapsulation of AG50 in polymeric micelles maintained cytotoxic effect against cancer cells. Therefore, AG50-loaded polymeric micelles offered a possible therapeutic solution for cholangiocarcinoma cancer chemotherapy as an intravenously injectable aqueous formulation for AG50.
This research project is supported by Mahidol University, Thailand. The financial support for Komgrit Eawsakul from the Center of Excellence for Innovation in Chemistry (PERCH-CIC) is gratefully acknowledged.
The authors declare no conflict of interest.
