Invertible Micellar Polymer Assemblies for Delivery of Poorly Water

Jul 3, 2012 - The IMA capability to solubilize lipophilic drugs and deliver and release the cargo molecules by conformational inversion of polymer ...
0 downloads 0 Views 5MB Size
Download PDF
Article pubs.acs.org/Biomac

Invertible Micellar Polymer Assemblies for Delivery of Poorly WaterSoluble Drugs Ivan Hevus,† Amit Modgil,‡ Justin Daniels,§ Ananiy Kohut,† Chengwen Sun,‡ Shane Stafslien,§ and Andriy Voronov*,† †

Department of Coatings and Polymeric Materials, ‡Department of Pharmaceutical Sciences, and §Center for Nanoscale Science and Engineering, North Dakota State University, Fargo, North Dakota 58108, United States S Supporting Information *

ABSTRACT: Strategically designed amphiphilic invertible polymers (AIPs) are capable of (i) self-assembling into invertible micellar assemblies (IMAs) in response to changes in polarity of environment, polymer concentration, and structure, (ii) accommodating (solubilizing) substances that are otherwise insoluble in water, and (iii) inverting their molecular conformation in response to changes in the polarity of the local environment. The unique ability of AIPs to invert the molecular conformation depending on the polarity of the environment can be a decisive factor in establishing the novel stimuli-responsive mechanism of solubilized drug release that is induced just in response to a change in the polarity of the environment. The IMA capability to solubilize lipophilic drugs and deliver and release the cargo molecules by conformational inversion of polymer macromolecules in response to a change of the polarity of the environment was demonstrated by loading IMA with a phytochemical drug, curcumin. It was demonstrated that four sets of micellar vehicles based on different AIPs were capable of delivering the curcumin from water to an organic medium (1-octanol) by means of unique mechanism: AIP conformational inversion in response to changing polarity from polar to nonpolar. The IMAs are shown to be nontoxic against human cells up to a concentration of 10 mg/L. On the other hand, the curcumin-loaded IMAs are cytotoxic to breast carcinoma cells at this concentration, which confirms the potential of IMA-based vehicles in controlled delivery of poorly water-soluble drug candidates and release by means of this novel stimuli-responsive mechanism.

■

INTRODUCTION Transcellular transport involves drugs crossing the barrier thorough exploitation of one of the following mechanisms: passive diffusion, active transport, or endocytosis.1−3 Many nanoparticles used in therapeutic drug delivery are intended to deposit a drug at a specific intracellular location such as the cytosol.4 In many instances, nanoparticles may enter cells via multiple endocytic pathways,5 as well as they also enter cells using nonendocytic pathways.6 In the past two decades, over 90 chemotherapeutic drugs have been approved by the FDA for clinical applications, but the efficacy of these drugs is dramatically hindered due to ineffective delivery systems. Poor solubility of many drugs and drug candidates in water is one of the main problems for formulating clinically useful pharmaceuticals.7−9 In recent years, significant progress has been achieved in developing novel delivery systems based on liposomes, emulsions, hydrogels, © XXXX American Chemical Society

polymer micelles, and so on, to refine the critical parameter of solubility in the delivery of different hydrophobic drugs.10−21 Most of these potential new delivery systems are classified as nanoscale therapeutics, and considered to be emerging modalities in the treatment of many diseases.22 Polymer micelles are formed through spontaneously selfassembling amphiphilic macromolecules in an aqueous environment and possess characteristics that can fulfill the requirements of a versatile drug carrier for pharmaceutical applications: stability, small size, high payload capacity, prolonged blood circulation times, passive site targeting, reduced toxicity to healthy tissues, and improved efficacy as drug carriers.17−28 The inner part of the polymer micelles is Received: May 21, 2012 Revised: June 28, 2012

A

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

platforms have been established to trigger drug release from the delivery system through a stimuli-sensitivity mechanism. Among the ways to achieve the stimuli-responsive drug release, there are strategies including pH,35−38 temperature,39,40 and ultrasound-stimulated release.41−43 In this regard, a polymersomes system possessing a pHinduced “breathing” (reversible vesicle size change) that can be repeated many times, with a relaxation time of about 1 min, was recently reported.44 To the best of our knowledge, there is no previous report in the literature on invertible polymer micelles as potential drug delivery candidates, inducing release of the solubilized hydrophobic drug in response to rapidly changing polarity of the environment. Such a strategy of drug delivery and release was targeted and realized in our study. Loaded with a waterinsoluble drug, polymeric micellar assemblies can successfully transfer drug molecules from an aqueous medium to a polar/ nonpolar interface and release the payload upon inverting the macromolecular conformation by entering/contacting the less polar medium (Figure 1B). In this way, two critical drug performance parameters, namely, enhanced solubility and promoted release, are simultaneously achieved using the responsive polymeric micellar assemblies. In this study, micellar polymeric assemblies represent a new polymer class, amphiphilic invertible polymers (AIPs), which were synthesized by our group.45−50 The synthesis results in macromolecules containing a precisely controlled number of hydrophilic and hydrophobic short fragments with a welldefined length. The incompatibility of alternating small fragments along the AIP macromolecular backbone results in microphase separation at smaller-length scale (in comparison to similar block copolymers’ structure), which in turn enables a greater degree of tunability in the formation and response of the micelle.47,48 We also demonstrated that AIP macromolecules self-assemble into invertible micellar assemblies in response to changing polarity of the environment, polymer structure and concentration.45,46 Additional control over the self-assembly benefits from alternated distribution of hydrophilic and lipophilic fragments in the AIP macromolecular backbone. What varies in the invertible macromolecule is, thus, the hydrophilic−lipophilic balance (HLB), which considerably influences surface activity and self-assembly, thus, broadening the available variety of IMA properties.46−49 To demonstrate the capability of the IMAs to deliver lipophilic drugs and release the payload using stimuliresponsive inversion of macromolecules, the IMAs were loaded with poorly water-soluble curcumin. Curcumin is a phytochemical agent that occurs naturally in plants. It has low intrinsic toxicity but possesses a great potential in the treatment of diverse diseases including cancer, arthritis, Alzheimer’s disease, and so on51 due to broad biological activity such as antioxidant, anti-inflammatory, and antitumor activity at the molecular level, following oral or topical administration.52 Curcumin has been subjected to several clinical trials for the development of therapeutic agents for various diseases, which include cancer,53 but its clinical development has been hindered due to the insolubility of the curcumin in water, restricting the use of the drug.54

composed from hydrophobic fragments of the amphiphilic macromolecule and serves as a container for the solubilization of poorly soluble drug molecules. The micellar interior is surrounded by the outer part of the micelle, which is made of hydrophilic fragments of the amphiphilic macromolecule and provides micellar stability, longevity in the bloodstream, and drug administration.29−32 Solubility of hydrophobic drugs can be significantly increased by drug solubilization within the micellar hydrophobic interior by physical (noncovalent) interaction. The advantage of the incorporation strategy over the alternative approaches that include covalent attachment of the drug molecules to the carrier is that the drug molecules remain chemically intact, a feature that leads to higher drug activity and improved release profile.28 The unique attribute that differentiates polymer micelles from alternative potential nanoscale therapeutics is the chemical flexibility of micellar structure, which permits the design of custom-made carriers that can be developed individually with respect to drug properties, site of action, and administration pathway.33 Although the utility of polymer micelles in pharmaceutical formulations is now well recognized, their thermodynamic stability in biological media often complicates the release of the active agents.34 The release of physically incorporated drug molecules from the polymeric micellar containers usually occurs by diffusion (Figure 1A).33 However, it is desirable that, upon entering the action site, the solubilized drug be released in a controlled fashion to reach the appropriate therapeutic efficacy. Recently, several smart polymeric micellar

■

MATERIALS AND METHODS

Materials. Poly(ethylene glycol) (PEG, molecular weight 300, 600, and 1000 g/mol), polytetrahydrofuran (PTHF, molecular weight 250 and 650 g/mol), sebacic acid, pyrene, carbon tetrachloride, methanol,

Figure 1. Diffusion mechanism (A) and stimuli-responsive (inversion) mechanism (B) for delivery of poorly water-soluble drugs using polymeric micellar platforms. B

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

and deuterium oxide were purchased from Sigma-Aldrich. Dodecanedioic acid was obtained from TCI, while curcumin and 1-octanol were purchased from Alfa Aesar. Synthesis and Characterization of Amphiphilic Invertible Polymers (AIPs). Amphiphilic invertible polymers were synthesized using previously reported methods.45,50 from PEG-600 and PTHF-650 (PEG600PTHF650), PEG-300 and PTHF-250 (PEG300PTHF250), PEG1000 and sebacic acid (S10), and PEG-1000 and dodecanedioic acid (D10), respectively. Critical micelle concentration (CMC) of AIPs was measured as described elsewhere49 (see Supporting Information). Size distribution and zeta potential of AIP micelles in aqueous solution were measured using Malvern Zetasizer Nano-ZS90 at 25 °C. The final numbers represent an average of a minimum of 5 (size) or 10 (zeta potential) individual measurements. 1 H NMR spectra were recorded on a JEOL ECA 400 MHz NMR spectrometer using chloroform-d and deuterium oxide as solvents. Preparation of Curcumin-Loaded IMAs. Micellar curcumin was prepared by thin film method.55 Following this method, polymer (0.1 g) and methanol solution of curcumin taken in excess amount (0.5 mL, 1 mg/mL) were dissolved in carbon tetrachloride (10 mL). The solvent was removed by rotary evaporation at 60 °C for 1 h to obtain a solid drug/AIP matrix. Residual carbon tetrachloride remaining in the drug/AIP matrix was evaporated overnight in vacuo. The resultant thin film was hydrated with Millipore water (10 mL), and unincorporated excessive curcumin aggregates were removed by filtration through 0.45 and 0.2 μm filters. Study of IMA-Mediated Phase Transfer. For phase transfer, curcumin-loaded aqueous IMA solutions (3 mL) were mixed with an equal volume of 1-octanol. To reach equilibrium, the mixtures were stirred for 10 min. Subsequently, the aqueous phase was separated from the organic phase by centrifugation. Determination of the Curcumin Concentration in Aqueous and 1-Octanol Phase. The initial concentration of the curcumin in aqueous solution was estimated using UV−vis spectroscopy. UV−vis spectra were recorded on a Cary 5000 UV−vis-NIR spectrophotometer (Varian, Inc.). The absorbance values were measured in the range of 350−800 nm. The height of a curcumin adsorption peak at 425−430 nm was attributed to a particular drug concentration using the calibration method. If necessary, curcumin-loaded solution samples were diluted with the corresponding 1% aqueous AIP solution to maintain measurable absorbance levels. To build a calibration curve, sets of 1% micellar solutions for each amphiphilic invertible polymer containing known amounts of solubilized curcumin were prepared and their UV−vis spectra were recorded. The final concentration of the drug in the 1-octanol phase after the experiment was determined using UV−vis spectroscopy (samples were diluted with 1-octanol if necessary). The calibration curves were built after recording the UV−vis spectra of a set of curcumin solutions of known concentration in the 1-octanol. Determination of the AIP Concentration in 1-Octanol. To study whether the AIPs cross the interface between water and 1octanol during the phase transfer experiment, 1% aqueous solutions of the polymers (3 mL) were mixed with an equal volume of 1-octanol and stirred for 10 min. Subsequently, two phases were separated by centrifugation, both solvents were evaporated, and the amount of polymer in each phase was estimated gravimetrically. The fraction of polymer transferred to the organic phase was calculated as follows:

mL of Millipore water at room temperature with water exchange every 6 h. The released curcumin content was measured by UV−vis spectroscopy. Chemical stability of curcumin-loaded IMAs was evaluated using UV−vis spectroscopy. The decrease in absorbance of micellar solutions was monitored with time at room temperature. Cytotoxicity of IMAs against Human Embryonic Kidney 293 Cells. Cytotoxicity of IMAs against human living cells was tested using MTT assay. Hyman embryonic kidney cells (HEK 293) were grown and maintained in the DMEM medium supplemented with 10% fetal bovine serum, penicillin (100 U/mL), and streptomycin (100 μg/mL) at 37 °C in a humidified incubator in an atmosphere of 5% CO2. The cells were plated in a 96-well plate at density of 1 × 104 cells/well, 24 h prior to addition of polymers. Following this, IMAs made from aqueous solutions of S10, D10, PEG600-PTHF650, and PEG300PTHF250 at different concentrations above CMC were added to the cells and incubated for another 24 h. After 24 h, a 2 mg/mL solution of MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) in Hank’s balanced salt solution (HBSS) was added to all the wells (25 μL/well) and the plate was reincubated for 6 h in an incubator to allow for the formation of formazan crystals. After incubation, the media were discarded carefully from the wells and dimethyl sulfoxide (DMSO, 100 μL) was added to solubilize the formazan crystals that formed. The absorbance was measured in each well at 570 nm using microplate spectrophotometer. Untreated cells were used as controls and Triton-100 (at 0.025 and 0.015%) was used as positive control. Cytotoxicity of Curcumin-Loaded IMAs against Breast Carcinoma Cells. The curcumin-loaded IMAs were evaluated for cytotoxicity toward the breast carcinoma cell line T47D (CRL-2865) obtained from the American Type Culture Collection (Rockville, MD).56 The T47D cells were maintained in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS), penicillin (100 U/mL), streptomycin (100 μg/mL), and glutamine (2 mM). The cells were trypsinized, resuspended into fresh DMEM, and then seeded into 96-well poly-L-lysine plates at an inoculum density of 2 × 104 cells/well. The plates were incubated for 24 h at 37 °C (5% CO2) and then the DMEM was carefully removed from each well using an 8-channel pipet. Each curcumin-loaded IMA sample (0.150 mL) was transferred to five replicate wells containing attached T47D cells. The samples were prepared in DMEM by adding Na2CO3 solution to adjust pH to 7.2. In addition to the curcumin-loaded polymer micelle samples, a set of Triton X-100 cytotoxic controls (0.01 and 0.001% v/v) were also prepared in DMEM and added to five replicate wells. A growth positive control (i.e., noncytotoxic) was included by adding fresh DMEM to the wells only (i.e., without curcumin-loaded polymer micelles or Triton X-100). The plates were then transferred to a 37 °C growth chamber (5% CO2) and incubated for 3 h. Following the 3 h incubation period, the curcumin-loaded polymer micelle and control samples were discarded and the entire plate was rinsed three times with DMEM (0.150 mL). Subsequently, fresh DMEM (0.3 mL) was added to each well and the plates were placed back into the growth chamber and incubated at 37 °C (5% CO2). T47D cell viability was determined after 18, 42, and 66 h of incubation using an MTT colorimetric assay.57 In particular, a 0.5 g/L solution of MTT in HBSS (0.033 mL) was added to each well of the plate and then incubated for 4 h at 37 °C. Following the 4 h incubation period, the MTT solution was removed from each well and DMSO (0.150 mL) was added. The plates were then placed on an orbital shaker for 15 min (150 rpm; ambient laboratory temperature) to lyse the cells and solubilize the MTT dye. The plates were transferred to a multiwell plate spectrophotometer and the absorbance values were measured at 570 nm. The mean absorbance values reported (n = 5) were considered to be directly proportional to the number of viable cells that survived after the curcumin-loaded polymer micelle and control sample treatments. Statistical analysis was performed using a JMP 7.0 statistical software package (SAS Institute Inc.). A one-way ANOVA was used to evaluate the differences in cell viability of T47D breast carcinoma cells after 18, 42, and 66 h of exposure to curcumin-loaded AIPs. The p

x = [AIPorg ]/[AIPaq ]0 = m(AIPorg )/m0(AIPaq ) where [AIPorg] and [AIPaq]0 comprise the final AIP concentration in the organic phase and initial AIP concentration in the aqueous phase, mol/L, m(AIPorg) and m0(AIPaq) is the weight of the AIP in the organic phase after the experiment and the weight of the AIP in the initial aqueous solution, g, respectively. Release Behavior and Chemical Stability of CurcuminLoaded IMAs. Drug release from curcumin-loaded micelles was studied by dialyzing those using 15 mL Slide-A-Lyzer dialysis cassettes (Thermo Scientific, molecular-weight cutoff of 3500 Da) against 700 C

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Figure 2. Chemical structure and characteristics of the AIPs used in the study.

Figure 3. Excitation spectra of pyrene in the aqueous solutions of a polyester PEG600PTHF650 at different concentrations (A). The intensity ratio I336.5/I333 of the excitation spectra of pyrene in AIP solutions vs AIP concentration (B). value was reported and a Tukey-Kramer HSD Posthoc multiple comparison test was conducted (α = 0.05). Data points that share the same letter (i.e., A or B) are statistically the same, whereas data points that do not share the same letter are statistically different.

solubilization of pyrene within the micellar hydrophobic environment and the transfer of pyrene molecules from water to the polymer micelles. The sharp increase in the intensity ratio corresponds to the critical micelle concentration for each polymer. The CMC values correlate with the chemical structure of the AIPs (Figure 2). Changing both length of hydrophobic fragments and length ratio of hydrophilic to hydrophobic chains in the AIP macromolecules result in a significant difference in recorded CMC values. The CMC increases with an increasing HLB of polymers and varies between 5.5 × 10−7 mol/L (3.5 mg/L) for the most hydrophobic PEG600PTHF650 and 8.2 × 10−5 mol/L (490 mg/L) for the most hydrophilic S10. In addition, very low CMC for the most hydrophobic polymer indicates that micelles from PEG600PTHF650 would provide a very good stability in solution even after strong dilution by the large volume of systemic circulation in the body. Micellar self-assembly was studied using 1H NMR spectroscopy by increasing the concentration of amphiphilic invertible polymers in water. Increasing the concentration of each amphiphilic invertible polymer leads to the formation of micellar assemblies due to the aggregation of single unimolecular micelles and the formation of hydrophilic and lipophilic domains. The characteristic 1H NMR spectra of PEG600PTHF650 and D10 taken in deuterium oxide over the wide polymer concentration range indicate the formation of invertible micellar assemblies (Figures 1S and 2S in Supporting Information). Characterization of the Blank and Curcumin-Loaded IMAs. Curcumin-loaded micelles were prepared using 1% polymer solutions. At this concentration, the micelles were selfassembled into the IMAs, and the hydrophobic curcumin was solubilized through physical interactions with the polymer hydrophobic fragments of the IMA interior. To determine loading, UV−vis spectroscopy measurements were carried out.

■

RESULTS AND DISCUSSION AIP Synthesis, Characterization, and Formation of IMAs in Water. Figure 2 shows the chemical structures of the amphiphilic invertible polymers that were used in this study for the formation of IMAs, as well as their molecular weight, polydispersity index, hydrophilic lipophilic balance (HLB), and critical micelle concentration (CMC). The chemical structures of the PEG300PTHF250, PEG600PTHF650, D10, and S10 were confirmed by 1H NMR and FT-IR spectroscopies.48,49 Broad surface activity for synthesized AIPs was targeted to achieve the varying capacity of IMAs in the solubilization of curcumin in water. To this end, the HLB of the invertible macromolecules was varied by changing length and length ratio of hydrophilic and hydrophobic fragments in macromolecules between 13.8 and 15.4, as calculated.58 To confirm the formation of micelles from AIPs in aqueous solution, CMC values were measured using solubilization of pyrene, a well-known fluorescent probe for studying the association behavior of amphiphilic polymers.59,60 Depending on the environment of the pyrene, a red shift of the absorption band with enhanced excitation intensity was observed due to the migration of the probe from the hydrophilic to the hydrophobic region of the polymer micelles (Figure 3A). In our experiments, pyrene excitation spectra were monitored in the wavelength range of 300−360 nm. From the pyrene excitation spectra, the intensity ratios I336.5/I333 were plotted as a function of amphiphilic invertible polymer concentration (Figure 3B). A red shift of the fluorescence excitation spectra from 333 to 336.5 nm with an increasing polymer concentration in aqueous solution indicates the D

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Table 1. Physical Properties of Blank and Curcumin-Loaded IMAs at 1% Concentration AIP

drug loading, wt %

size, nm (blank)

size, nm (loaded)

ζ-potential, mV (blank)

ζ-potential, mV (loaded)

PEG600PTHF650 PEG300PTHF250 D10 S10

10.3 ± 0.4 3.6 ± 0.3 1.9 ± 0.3 0.14 ± 0.03

12.0 ± 0.2 12.4 ± 0.2 6.3 ± 0.3 3.3 ± 0.5

17.5 ± 2.6 18.4 ± 1.1 7.8 ± 0.6 3.6 ± 0.2

−24.2 ± 3.8 −43.0 ± 2.3 −10.3 ± 0.9 −7.1 ± 0.3

−18.2 ± 0.6 −42.6 ± 1.5 −7.4 ± 0.3 −6.7 ± 0.9

Table 1 shows the curcumin loading content for each micellar formulation in this work. Hydrodynamic diameters of the IMAs were compared for amphiphilic invertible polymers with different HLB, before and after the curcumin loading. Figure 4 shows the characteristic

Figure 5. Chemical stability of curcumin in IMAs with time. Inset: chemical structure of curcumin.

hydroxy-3′-methoxyphenyl)-2,4-dioxo-5-hexanal, ferulic acid, and feruloyl methane.62 More than 90% of the curcumin rapidly decomposes within 30 min of placement in phosphatebuffered saline at pH 7.2. In this work, the stability of curcumin loaded in micellar assemblies was monitored by UV−vis spectroscopy. The data show that the change in absorbance of curcumin-loaded IMAs was negligibly small for micellar assemblies based on AIPs with lower HLB, confirming the very good stability of micellar curcumin (Figure 5). In particular, extremely low curcumin decomposition was observed for micellar formulation based on PEG600PTHF650 and PEG300PTHF250. More curcumin decompose when micelles are formed from more hydrophilic polymers, but even for the most hydrophilic S10, about 50% of the loaded drug remains intact, while it is entrapped in the micellar assemblies, after 10 days of the experiment. Therefore, curcumin-loaded IMAs from four polymers were employed for the phase transfer study. IMA-Mediated Delivery of Curcumin from Aqueous Polymer Solution to 1-Octanol. To probe the IMAmediated drug transfer through a polar/nonpolar interface, 1octanol (solvent in which curcumin is soluble, and that is immiscible with water) was chosen as a nonpolar phase. 1Octanol is considered to be a valid hydrophobic scale-model system to study partitioning in biomembrane.57 The 1-octanol was added to the top of an aqueous solution containing curcumin-loaded micellar assemblies. Mixtures were shaken for 1 min and then allowed to separate. All chosen polymeric micellar formulations were capable of transferring sequestered curcumin molecules from water to an organic medium of opposite polarity that was detected using UV−vis spectroscopy measurements (Figure 6). This observation indicates that micellar assemblies were efficient in transferring cargo molecules. Analysis of the curcumin concentration in aqueous (before the transfer) and 1-octanol (after the transfer) phases revealed that the vast majority of the drug molecules was transferred to the 1-octanol (Table 2). The latter demonstrates that the major determinant of the extent of IMA-mediated drug transfer through the polar/

Figure 4. Size of blank and curcumin-loaded IMAs as determined by dynamic light scattering: (1) D10, (2) curcumin−D10, (3) PEG600PTHF650, and (4) curcumin−PEG600PTHF650.

average diameter and size distribution of blank and drug-loaded micellar assemblies from D10 and PEG600PTHF650. The diameters of micellar assemblies vary between 4.7 ± 0.2 and 12 ± 0.6 nm for chosen polymers and become larger for macromolecules with higher HLB. The loaded micellar assemblies showed a larger diameter size in comparison to blank IMAs, which confirms the incorporation of curcumin into the IMA interior. Small size and narrow unimodal size distribution indicate that IMAs possess good physical properties for being considered as nanocarriers for poorly watersoluble drugs. Analysis of the zeta potential for blank and drug-loaded micellar assemblies showed that all polymer formulations have negative surface charge at room temperature (Table 1). The zeta potentials of loaded micelles decrease but still remain negative after the solubilization of curcumin. The negative zeta potentials indicate that IMAs potentially can provide an enhanced adhesion and interactions with gastrointestinal mucus and cellular linings and, thus, facilitate the bioadhesion between the micellar carriers and intestinal epithelial cells.61 The obtained results demonstrate that amount of solubilized curcumin in the IMAs is the highest for the most hydrophobic PEG600PTHF650 (HLB = 13.8). The loading capacity of micellar assemblies is essentially smaller for PEG300PTHF250, a polymer with shorter hydrophobic fragments. Much less curcumin was loaded in IMAs developed from the more hydrophilic D10 and S10 (HLB = 14.4 and 15.4, respectively). However, all four polymers were chosen for further experiments on the chemical stability of micellar curcumin and phase transfer study. Chemical Stability of Curcumin in IMAs. Curcumin (Figure 5, inset) is stable at acidic pH but not stable at neutral and basic pH. Under these conditions it decomposes to (4′E

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

disassemble (lose their container properties), and release curcumin into the 1-octanol, where the drug is soluble (Figure 7B). Finally, the loaded micellar assemblies can undergo thermodynamically driven distribution between the aqueous and the 1-octanol phases. This can happen when drug-loaded micelles cross the interface, invert their conformation, and form new micelles in the 1-octanol (Figure 7C). If this mechanism works, then curcumin-loaded assemblies can act as responsive carriers for solubilized drug molecules and release the drug upon inversion within a nonpolar phase. To distinguish between three possible mechanisms and their contributions to the established IMA-mediated curcumin transfer, two additional experiments were carried out. First, the presence of AIP macromolecules in 1-octanol was analyzed to confirm crossing the interface by polymer and, thus, the possibility for the IMAs to deliver curcumin to 1-octanol using the mechanism, as is shown in Figure 7C. Second, release of curcumin from micellar assemblies was studied using conventional dialysis membrane to evaluate the potential contribution of anticipated partitioning of curcumin between the micellar interior and aqueous phase due to limited solubility of the drug in water (Figure 7A). Table 2 shows that AIP macromolecules were detected in 1octanol for each polymer employed in the IMA-mediated curcumin transfer experiment, indicating that all the polymers were able to cross the interface between water and 1-octanol during the experiment. Interestingly, the hydrophilic-lipophilic balance of the polymers does not show a great influence on the amount of the transferred AIP. In turn, curcumin release in dialysis differs greatly for AIPs with different HLB values. To this end, almost no release was detected for drug molecules loaded in PEG600PTHF650 and PEG300PTHF250. In turn, the release of curcumin from more hydrophilic polymers (especially, S10) is much higher, which demonstrates differences in the thermodynamic stability of loaded micelles and the significant influence of HLB on partitioning mechanism contribution during the transfer. Quantitative analysis of data in Table 2 indicates that most of the micellar curcumin was delivered to 1-octanol using the conformational change of macromolecules that is induced in response to the changing polarity of the environment (Figure 7B). The total amount of transferred curcumin suggests that IMAs are superior in curcumin delivery from water to 1octanol, and the drug was efficiently transferred through the polar/nonpolar interface, predominantly using the inversion mechanism. To further investigate the potential of IMA formulations in cellular uptake, the cytotoxicity of curcuminloaded micellar assemblies was evaluated against human embryonic kidney cells and breast carcinoma cells.

Figure 6. Spectra of curcumin in (1) aqueous phase before the transfer, as well as in (2) 1-octanol and (3) aqueous phase after the transfer from water to 1-octanol. Inset: appearance of water−octanol mixture before (left-side tube) and after (right-side tube) the IMAmediated transfer of curcumin from water (bottom phase) to 1-octanol (top phase).

Table 2. Phase Transfer Characteristics of Curcumin-Loaded IMAs at 1% Concentration AIP PEG600PTHF650 PEG300PTHF250 D10 S10

curcumin transferred, %

curcumin released in dialysis, %

± ± ± ±

1.4 ± 0.7 0.3 ± 0.5 16 ± 2 49 ± 2

89 91 99 96

3 3 2 3

AIP transferred to 1-octanol,% 14 26 21 13

± ± ± ±

2 2 3 1

nonpolar interface is the ability of the IMAs to bind curcumin molecules in water (loading capacity). With the IMA-mediated drug transfer established, a mechanism of release of hydrophobic substance from micellar assemblies was considered next. Obviously, three different possibilities exist to explain how curcumin molecules could cross a polar/nonpolar interface. One of these assumes that the molecules that are poorly soluble in water can be partitioned between the micellar interior and bulk water due to their limited aqueous solubility.63 At the beginning of the experiment, curcumin molecules are distributed between the bulk water and the IMAs, which act as containers of poorly soluble material. The limited solubility of drug in water, and thus, its low concentration in the aqueous phase, is obviously sufficient to transfer the drug from the micellar interior, first to water and then from water to the nonpolar phase (Figure 7A).50,64 At the same time, the curcumin-loaded micellar containers can migrate from the aqueous phase to the polar/nonpolar interface, change their conformation due to the macromolecular inversion upon changing polarity of local environment,

Figure 7. Possible mechanisms of IMAs-mediated curcumin delivery from water to 1-octanol. F

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

In Vitro Cytotoxicity of Blank and Curcumin-Loaded IMAs. Prior to studying the cytotoxicity of curcumin-loaded micellar assemblies against carcinoma cells, the cell viability of human embryonic kidney cells (HEK 293) in the presence of blank IMAs was examined using standard MTT assay. As a potential drug-delivering carrier, the material has to be nontoxic to human cells and highly biocompatible. Various concentrations of IMAs exhibit no cytotoxicity against living human cells at concentrations of 2 × 10−3−10 mg/L (Figure 8). Three different polymers are comparable on the cell viability of HEK 293 cells.

Figure 9. Cytotoxicity of curcumin-loaded IMAs on breast carcinoma cells after 18, 42, and 66 h incubation.

This difference could be related to the fact that the polymerbased formulations have different profiles for releasing curcumin. The results from the phase transfer study indicate that a large portion of S10-loaded curcumin could be released by diffusion (Figure 7A), while this mechanism is almost negligible in the case of highly thermodynamically stable PEG600PTHF650 formulation (Table 2). However, although the IMAs from PEG600PTHF650 remain stable in a homogeneous environment, they can rapidly invert conformation due to subtle change in polarity of the environment. Thus, once the curcumin-loaded micellar assemblies developed from PEG600PTHF650 reach the cell sites, the drug release can be induced by rapid inversion of the macromolecular conformation. As a result, the cytotoxicity of curcumin-loaded PEG600PTHF650 assemblies becomes comparable to the cytotoxicity of the formulation made from S10 already after 42 h and further remains constant. This behavior of PEG600PTHF650 formulation demonstrates the promising ability of the IMAs to remain stable in a homogeneous environment, but then rapidly undergo inverse conformation due to changes in environmental polarity. To summarize, the ability of IMAs-based micellar formulations to solubilize a poorly water-soluble drug and improve drug bioavailability was demonstrated. Drug-loaded IMAs delivered poorly water-soluble curcumin to carcinoma cells and showed an anticancer efficiency. For the moment, we can speculate that there are two possible different mechanisms of IMA-mediated drug uptake by the cells: diffusion and conformational inversion of AIP macromolecules in response to change of the polarity of the environment. The obtained results demonstrate the potential of IMA-based responsive vehicles as a promising platform for controlled delivery of poorly water-soluble drug candidates by means of new stimuli-responsive release mechanism. Further exploration is needed to reveal the nature and insights of this unique phenomenon in drug delivery and release.

Figure 8. Cell viability of HEK 293 cells treated with various concentrations of blank IMAs.

The efficiency of curcumin-loaded IMAs on the viability of T47D breast carcinoma cells was assessed using MTT assay for S10 and PEG600PTHF650 formulations at concentration of 10 μg/mL and maximum curcumin loading. Both formulations showed high efficiency in the phase transfer experiment, while they are, obviously, significantly different in terms of releasing mechanism. We observed that the very small amount of drug was released through dialysis from PEG600PTHF650-based formulation, while almost half of the drug amount was released in the same experiment from the S10-based formulation (Table 2). Figure 9 shows the in vitro viability of T47D breast carcinoma cells after 18, 42, and 66 h treatment with curcumin formulated in IMAs at maximum loading (Table 2). A few general conclusions can be drawn from Figure 9. First, the IMA-based vehicles are cytotoxic against the carcinoma cells, which confirms their successful administration or/and curcumin delivery to the cells. Second, the micellar curcumin remains stable for at least 66 h under conditions that (without solubilization in micelles) facilitate its rapid chemical decomposition with water, which indicates the improved bioavailability of micellar curcumin for the cellular uptake. Third, the cytotoxicity results demonstrate that two IMA formulations obviously exhibit different profiles of carcinoma cell viability in time, which we believe may be due to variations in the drug-releasing mechanism. For the formulation based on PEG600PTHF650, the cytotoxic effect is small after 18 h, but it improves with an increasing time of incubation. In case of curcumin-loaded S10 assemblies, the cell viability significantly decreases after 18 h and remains almost constant during further incubation.

■

CONCLUSIONS Strategically designed amphiphilic invertible polymers (AIPs) are capable of the following: (i) self-assembling into invertible micellar assemblies (IMAs) in response to changing polarity of the environment, polymer concentration, and structure; (ii) accommodating (solubilizing) drugs that are otherwise G

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

(5) Hyung Park, J.; Kwon, S.; Lee, M.; Chung, H.; Kim, J. H.; Kim, Y. S.; Park, R. W.; Kim, I. S.; Bong Seo, S.; Kwon, I. C. Biomaterials 2006, 27 (1), 119−126. (6) Taylor, U.; Klein, S.; Petersen, S.; Kues, W.; Barcikowski, S.; Rath, D. Cytometry, Part A 2010, 77 (5), 439−446. (7) Kennedy, T. Drug Discovery Today 1997, 2, 436−444. (8) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Delivery Rev. 2001, 46, 3−26. (9) Serajuddin, A. T. Adv. Drug Delivery Rev. 2007, 59, 603−616. (10) Duncan, R. Nat. Rev. Drug Discovery 2003, 2, 347−360. (11) Ferrari, M. Nat. Rev. Cancer 2005, 5, 161−171. (12) Peer, D.; Karp, J. M.; Hong, S.; Farokhzad, O. C.; Margalit, R.; Langer, R. Nat. Nanotechnol. 2007, 2, 751−760. (13) Zhang, J. A.; Anyarambhatla, G.; Ma, L.; Uqwu, S.; Xuan, T.; Sardone, T.; Ahmad, I. Eur. J. Pharm. Biopharm. 2005, 59, 177−187. (14) Lundberg, B. B.; Risovic, V.; Ramaswamy, M.; Wasan, K. M. J. Controlled Release 2003, 86, 93−100. (15) Maranhão, R. C.; Tavares, E. R.; Padoveze, A. F.; Valduga, C. J.; Rodrigues, D. G.; Pereira, M. D. Atherosclerosis 2008, 197, 959−966. (16) Obara, K.; Ishihara, M.; Ozeki, Y.; Ishizuka, T.; Hayashi, T.; Nakamura, S.; Saito, Y.; Yura, H.; Matsui, T.; Hattori, H.; Takase, B.; Ishihara, M.; Kikuchi, M.; Maehara, T. J. Controlled Release 2005, 110, 79−89. (17) Torchilin, V. P. J. Controlled Release 2001, 73, 137−172. (18) Batrakova, E. V.; Kabanov, A. V. J. Controlled Release 2008, 130, 98−106. (19) Torchilin, V. P. Expert Opin. Ther. Pat. 2005, 15, 63−75. (20) Kabanov, A. V.; Lemieux, P.; Vinogradov, S.; Alakhov, V. Adv. Drug Delivery Rev. 2002, 54, 223−233. (21) Torchilin, V. P. Cell. Mol. Life Sci. 2004, 61, 2549−2559. (22) Blanco, E.; Kessinger, C. W.; Sumer, B. D.; Gao., J. Exp. Biol. Med. 2009, 234, 123−131. (23) Croy, S. R.; Kwon, G. S. Curr. Pharm. Des. 2006, 12, 4669− 4684. (24) Nishiyama, N.; Kataoka, K. Pharmacol. Ther. 2006, 112, 630− 648. (25) Sutton, D.; Nasongkla, N.; Blanco, E.; Gao, J. Pharm. Res. 2007, 24, 1029−1046. (26) Torchilin, V. P. Pharm. Res. 2007, 24, 1−16. (27) Yokoyama, M. Expert Opin. Drug Delivery 2010, 7, 145−158. (28) Kwon, G. Crit. Rev. Ther. Drug Carrier Syst. 2003, 20, 357−403. (29) Kwon, G. S.; Suwa, S.; Yokoyama, M.; Okano, T.; Sakurai, Y.; Kataoka, K. J. Controlled Release 1994, 29, 17−23. (30) Yokoyama, M.; Okano, T.; Sakurai, Y.; Fukushima, S.; Okamoto, K.; Kataoka, K. J. Drug Target. 1999, 7, 171−186. (31) Nishiyama, N.; Okazaki, S.; Cabral, H.; Miyamoto, M.; Kato, Y.; Sugiyama, Y.; Nishio, K.; Matsumura, Y.; Kataoka, K. Cancer Res. 2003, 63, 8977−8983. (32) Bae, Y.; Jang, W.-D.; Nishiyama, N.; Fukushima, S.; Kataoka, K. Mol. Biosyst. 2005, 1, 242−250. (33) Aliabadi, H. M.; Lavasanifar, A. Expert Opin. Drug Delivery 2006, 3, 139−162. (34) Bawarski, W. E.; Chidlowsky, E.; Bharali, D. J.; Mousa, S. Nanomedicine 2008, 4, 273−282. (35) Bae, Y.; Nishiyama, N.; Fukushima, S.; Koyama, H.; Yasuhiro, M.; Kataoka, K. Bioconjugate Chem. 2005, 16, 122−130. (36) Potineni, A.; Lynn, D. M.; Langer, R.; Amiji, M. M. J. Controlled Release 2003, 86, 223−234. (37) Stapert, H. R.; Nishiyama, N.; Jiang, D.-L.; Aida, T.; Kataoka, K. Langmuir 2000, 16, 8182−8188. (38) Tang, Y.; Liu, S. Y.; Armes, S. P.; Billingham, N. C. Biomacromolecules 2003, 4, 1636−1645. (39) Liu, S. Q.; Tong, Y. W.; Yang, Y. Y. Mol. Biosyst. 2005, 1, 158− 165. (40) Chung, J. E.; Yokoyama, M.; Yamato, M.; Aoyagi, T.; Sakurai, Y.; Okano, T. J. Controlled Release 1999, 62, 115−127. (41) Mitragotri, S. Nat. Rev. Drug Discovery 2005, 4, 255−260. (42) Gao, Z.-G.; Fain, H. D.; Rapoport, N. J. Controlled Release 2005, 102, 203−222.

insoluble in water; and (iii) inverting their molecular conformation in response to changes in the polarity of the local environment. The unique ability of AIPs to invert the molecular conformation depending on the polarity of the environment can be a decisive factor in establishing the novel stimuli-responsive mechanism of solubilized drug release that is induced just in response to a change of the environment polarity. The IMA capability to solubilize lipophilic drugs, deliver, and release the cargo molecules by conformational inversion of polymer macromolecules in response to changing polarity of the environment was demonstrated by loading IMAs with curcumin, a phytochemical drug poorly soluble in water. Four sets of micellar vehicles based on different amphiphilic invertible polymers were capable of delivering the curcumin from water to an organic medium (1-octanol). Analyses before and after the drug transfer reveal that the vast majority of the curcumin was delivered from water to 1-octanol, and the release of the drug was triggered upon entering the nonpolar environment. The release of curcumin from IMAs is primarily caused by macromolecular inversion due to changing local environmental polarity and can be controlled by AIP structure and concentration. The IMAs are shown to be nontoxic against human cells up to a concentration of 10 mg/L. On the other hand, the curcumin-loaded IMAs are cytotoxic to breast carcinoma cells at this polymer concentration. An additional advantage of IMA vehicles is their ability to stabilize curcumin molecules against chemical decomposition. The unique ability of the IMAs to remain stable in homogeneous environments and rapidly invert their conformation based on subtle changes in polarity of the environment can be considered as a new and promising concept in developing polymer micelles-based nanopharmaceuticals.

■

ASSOCIATED CONTENT

S Supporting Information *

1

H NMR spectra of AIPs and AIP assemblies and information on measurements of AIP critical micellar concentration. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel.: +1 701 231 9563. Fax: +1 701 231 8439. E-mail: andriy. [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The work was partially supported by National Science Foundation (CBET 0966574).

■

REFERENCES

(1) Stenberg, P.; Luthman, K.; Artursson, P. J. Controlled Release 2000, 65 (1−2), 231−243. (2) Siccardi, D.; Turner, J. R.; Mrsny, R. J. Nat. Protoc. 2006, 1 (2), 633−636. (3) Gad, S. C. Preclinical Development Handbook: ADME and Biopharmaceutical Properties; Wiley-Interscience: New York, 2008. (4) Prokop, A. Intracellular Delivery: Fundamentals and Applications; Springer: New York, 2011. H

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

(43) Priutt, J. D.; Pitt, W. G. Drug Delivery 2002, 9, 253−258. (44) Yu, S.; Azzam, T.; Rouiller, I.; Eisenberg, A. J. Am. Chem. Soc. 2009, 131, 10557−10566. (45) Voronov, A.; Kohut, A.; Peukert, W.; Voronov, S.; Gevus, O.; Tokarev, V. Langmuir 2006, 22, 1946−1948. (46) Voronov, A.; Kohut, A.; Vasylyev, S.; Peukert, W. Langmuir 2008, 24, 12587−12594. (47) Martinez Tomalino, L.; Voronov, A.; Kohut, A.; Peukert, W. J. Phys. Chem. B 2008, 112, 6338−6343. (48) Kohut, A.; Voronov, A. Langmuir 2009, 25, 4356−4360. (49) Kohut, A.; Sieburg, L.; Vasylyev, S.; Kudina, O.; Hevus, I.; Stafslien, S.; Daniels, J.; Kislenko, V.; Voronov, A. In Amphiphiles: Molecular Assembly and Applications; Nagarajan, R., Ed.; ACS Symposium Series 1070; American Chemical Society: Washington, DC, 2011; pp 205−224. (50) Hevus, I.; Kohut, A.; Voronov, A. Polym. Chem. 2011, 2, 2767− 2770. (51) Aggarwal, B. B.; Harikumar, K. B. Int. J. Biochem. Cell Biol. 2009, 41, 40−59. (52) Maheshwari, R. K.; Singh, A. K.; Gaddipati, J.; Srimal, R. C. Life Sci. 2006, 78, 2081−2087. (53) Goel, A.; Kunnumakkara, A. B.; Aggarwal, B. B. Biochem. Pharmacol. 2008, 75, 787−809. (54) Sharma, R. A.; Gescher, A. J.; Steward, W. P. Eur. J. Cancer 2005, 41, 1955−1968. (55) Zhang, X.; Jackson, J. K.; Burt, H. M. Int. J. Pharm. 1996, 132, 195−206. (56) Stumm, S.; Meyer, A.; Lindne, M.; Baster, G.; Wallwiener, D.; Guckel, B. Oncology 2004, 66, 101−111. (57) Wang, Y.; Yu, L.; Han, L.; Sha, X.; Fang, X. Int. J. Pharm. 2007, 337, 63−73. (58) Davies, J. T.; Rideal, E. K. Interfacial Phenomena; Academic Press: New York, 1961. (59) Schmitz, C.; Mourran, A.; Keul, H.; Möller, M. Macromol. Chem. Phys. 2008, 209, 1859−1871. (60) Wilhelm, M.; Zhao, C. L.; Wang, Y.; Xu, R.; Winnik, M. A.; Mura, J. L.; Riess, G.; Croucher, M. D. Macromolecules 1991, 24, 1033−1040. (61) Lo, C. L.; Lin, K. M.; Huang, C. K.; Hsiue, G. H. Adv. Funct. Mater. 2006, 16, 2309−2316. (62) Kumar, A.; Ahuja, A.; Ali, J.; Baboota, S. Crit. Rev. Ther. Drug Carrier Syst. 2010, 27, 279−312. (63) Berthod, A.; Garcia-Alvarez-Coque, C. Micellar Liquid Chromatography; Marcel Dekker Inc.: New York, 2000. (64) Basu, S.; Vutukuri, D. R.; Thayumanavan, S. J. Am. Chem. Soc. 2005, 127, 16794−16795.

I

dx.doi.org/10.1021/bm3007924 | Biomacromolecules XXXX, XXX, XXX−XXX