Complexation and release of doxorubicin from its complexes with pluronic P85-b-poly(acrylic acid) block copolymers

Journal of Controlled Release 121 (2007) 137 – 145 www.elsevier.com/locate/jconrel

Complexation and release of doxorubicin from its complexes with pluronic P85-b-poly(acrylic acid) block copolymers Y. Tian a,b , L. Bromberg c , S.N. Lin b , T. Alan Hatton a,c , Kam C. Tam a,b,⁎,1 a

b

Singapore-MIT Alliance, Singapore School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798, Singapore c Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA Received 31 January 2007; accepted 8 May 2007 Available online 17 May 2007

Abstract Poly(acrylic acid) (PAA) was attached on both termini of Pluronic P85 copolymer (EO27PO39EO27) via atom transfer radical polymerization (ATRP) to produce a novel block copolymer, PAA-b-P85-b-PAA (P85PAA). The P85PAA–DOX complex formation and drug loading were strongly dependent on the PAA segment length and pH, where the protonation of carboxyl groups in the PAA segment at pH b 7.2 reduced the binding sites of DOX onto P85PAA chains, resulting in a diminished DOX uptake at low pH. The composition of copolymer–DOX complexes at pH 7.2 was close to the stoichiometric 1:1 DOX:carboxyl molar ratio, confirming the dominance of electrostatic interactions between cationic DOX molecules and carboxyl groups. The stability study of the copolymer–DOX complex suggested that non-polyelectrolyte interactions may also participate in the complexation of drug and P85PAA block copolymer. DOX loading at pH 5.0 decreased to 60% of the total binding capacity, indicating that protonation of carboxyl groups reduced the DOX binding to P85PAA block copolymer. DOX release from the complex is a pHresponsive process, where the protonation of carboxyl groups at mildly acidic condition resulted in a faster dissociation of copolymer-DOX complex, leading to an accelerated release of DOX at pH 5.0. Thus, complexation of DOX with P85PAA yielded a drug delivery system affording a pH-triggered release of DOX in an acidic environment of pH 5.0. © 2007 Elsevier B.V. All rights reserved. Keywords: Cancer drug; Doxorubicin; Drug delivery; Pluronics; Self-assembly; Polymer; Diffusion

1. Introduction Amphiphilic copolymers have been explored in the delivery of potent drugs in cancer treatment as they are capable of forming nano-particles loaded with anticancer drugs, which were shown to improve their stability and efficiency after incorporating with polymeric micelles [1–3]. Doxorubicin (DOX) is one of the most common chemotherapeutic drugs that possesses high anti-tumor activity. However, DOX gives rise to strong side effects [4,5]. To reduce the toxicity of DOX to normal tissue and to improve its therapeutic efficacy, DOX has

⁎ Corresponding author. School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798, Singapore. 1 Present address: Department of Chemical Engineering, University of Waterloo, Canada. Email: [email protected]. 0168-3659/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2007.05.010

been incorporated into the polymeric micelle by covalent attachment [6–10]. For instance, the primary amino group was used for conjugating DOX to poly(ethylene oxide)-b-poly (aspartic acid) through the amide bond [6]. DOX was also conjugated with poly(ethylene oxide)-b-poly(allyl glycidyl ether) block copolymer through pH-sensitive hydrazone bonds. Cleavage of hydrazone bonds at mildly acidic condition results in an accelerated release of DOX at pH 5.0. [7] Physical entrapment of DOX was developed mainly using hydrophobic interactions to entrap DOX into micellar cores [11–14], such as in poly(ethylene glycol)-poly(β-benzyl-L-aspartate) copolymer micelles [11], poly(ε-caprolactone)-poly(ethylene glycol) block copolymer [12], and poly(L-lactide)-b-poly(e-ethyl-s-oxazoline)-b-poly(L-lactide) block copolymer [13]. Compared to the chemical attachment, noncovalent entrapment of DOX is convenient and easy to prepare. Macromolecular surfactants such as Pluronic® copolymers have been studied extensively as a DOX delivery vehicle due to

138

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

their excellent biocompatibility and pronounced micellization behaviors [15–18]. Pluronic® copolymers have been found to promote active membrane transport of numerous anticancer drugs and thus help to overcome multiple drug resistance (MDR) phenomena in cancer therapy [15–18]. Physically encapsulated DOX in Pluronic micelles enhanced drug delivery to solid tumors and helped to overcome the MDR effects of the drug in vivo by inhibiting the membrane pump, P-glycoprotein (Pgp), responsible for the drug efflux causing MDR [15–17]. Efficacy of the Pluronic block copolymer composition in MDR cells have been evaluated, and it was shown that Pluronic P85 is one of the most efficacious agents in the Pgp inhibition [15–17], resulting in a stronger cytotoxic response of resistant cells to the DOX/P85 formulation compared to free DOX [19]. However, partition coefficient (LogP) of Doxorubicin between water and Pluronic micelles is only about 0.5, [20] which limits the loading capacity of Pluronic micelles for DOX. In order to improve the capability of Pluronic P85 with DOX, we grafted polyacrylic acid (PAA) at the ends of Pluronic P85 via atom transfer radical polymerization (ATRP), yielding a pH-responsive block copolymer, PAA- b-P85- b-PAA (P85PAA). The basic motivation for designing this polymer is to obtain high binding affinity of DOX within the polymeric carrier, and controlled DOX release can be achieved over the physiological pH range of 7.4 and 5.0. In previous studies, directly entrapping DOX via electrostatic interaction were mainly explored in high molecular weight crosslinked ionchange microgels [21,22]. Bromberg et al. have grafted Pluronic with poly(acrylic acid) (PAA) using a one-step radical polymerization with a chain transfer onto PAA, producing high molecular weight polymers, as well as bulk hydrogels and spherical microgel particles [23–25]. Pluronic-PAA-based microgels are suitable vehicles for oral chemotherapy, because of their benign nature, lack of penetration into the bloodstream, ability to be loaded with both ionic and hydrophobic therapeutic agents, mucoadhesion, and ease of processing into oral dosage forms. Polyether-modified poly(acrylic acid) microgels are capable of retaining loaded drugs in the collapsed state in acidic pH of the stomach and undergo volume transitions, which release drugs in the swollen state at intestinal pH. Kitaeva and coworkers had conducted fundamental studies on the direct complexation of DOX with anionic polymer to induce micellar formation. Directly complexation of DOX with poly(acrylic acid) (MW = 5000 Da) was achieved by the addition of 150 μM PAA solution to 50 μM DOX solution [35]. The size of complex particles was found to be between 600-900 nm. For our polymer PAA-b-P85-b-PAA, we hypothesized that the electrostatic interaction of DOX with carboxyl groups of PAA segments within the block copolymer could facilitate the incorporation water-soluble cationic drug such as DOX. In this study, DOX-induced assembly behavior and drug loading capacity were conducted using dynamic light scattering (DLS), fluorescence spectroscopic, transmission electron microscopic (TEM) and UV–vis spectroscopic techniques. The stability of the copolymer–DOX complex was investigated using isothermal titration calorimetry (ITC). The DOX dissociation from the complex was studied in buffer solution.

2. Experimental 2.1. Materials Pluronic P85 was obtained from BASF Corporation (Mount Olive, NJ). Trace amounts of water in P85 were removed by azeotropic distillation prior to use. Tert-Butyl acrylate (tBA) (Aldrich, 99%) was passed through a basic alumina column, dried over CaH2 and vacuum-distilled before polymerization. Triethylamine (TEA) and toluene were distilled prior to use. Copper (I) bromide (99.99%), N,N,N′,N″,N′″-pentamethyldiethylenetriamine (PMDETA), 2-bromoisobutyl bromide, tetrahydrofuran (THF), hexane, and methanol were purchased from Aldrich and used as received. Doxorubicin hydrochloride (99%) was purchased from Hande Tech (Houston, TX), and used without further treatment. 2.2. Synthesis of PAA-b-P85-b-PAA block copolymer and characterization All the synthetic steps were carried out under an argon atmosphere and the detailed synthesis procedure has been reported previously [26]. In a three-neck round-bottom flask, Pluronic P85 was dissolved in freshly distilled toluene at room temperature, then cooled to 0 °C. Deoxygenated triethylamine was added with stirring, and 2-bromoisobutryl bromide in dry toluene was added dropwise at 0 °C. After reacting for 24 h at r.t., macroinitiator, Br–P85–Br, was precipitated in excess n-hexane and dried under vacuum. Then, difunctional bromo-terminated P85 macroinitiator and CuBr were added to a pre-dried schlenk flask, and deoxygenated toluene and tBA were added via a syringe that had been purged with argon prior to use. The mixture was stirred until the macroinitiator was totally dissolved, and then evacuated with three freeze-thaw cycles to remove oxygen. The degassed PMDETA was added, and the reaction mixture was stirred at 80 °C for 6 h. After completion of the polymerization, the product was precipitated into excess of water:MeOH solution and dried under vacuum. Subsequently, hydrolysis was performed by adding an excess of trifluoroacetic acid (TFA) to the copolymer solution in methylene chloride and stirred at room temperature according to the procedure reported in the literature [27–29]. After 24 h reaction time, it was concentrated and precipitated in excess hexane, leading to the targeted P85PAA block copolymer. The content of the carboxylic groups was quantified by potentiometric titration. Titration of 0.1 wt.% polymer solution with 1 M standard NaOH solution was performed at 25 °C under a constant stirring. An ABU93 Triburet Titration system equipped with Radiometer pHG201 pH glass and Radiometer REF201 reference electrodes was used for pH measurement. 2.3. Fluorescence spectroscopy All fluorescence spectra were recorded on a AMINCOBowman series 2 spectrometer (Thermo Electron Corporation, USA) in a steady-state mode with 4 nm band-pass for excitation and 2 nm band-pass for emission, and equipped with a thermostated

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

cuvette compartment. Pyrene was prepared at 0.6 μM in 0.1 wt.% P85PAA solution. All samples were excited at 335 nm. Fluorescence spectra were recorded between wavelength of 360–430 nm. The particle size distribution of 0.1 wt.% P85PAA solution was measured at room temperature using a Brookhaven BIS200 laser scattering system at a scattering angle of 90°. An argon ion laser with a wavelength of 488 nm was used as the light source. 2.4. Transmission electron microscope (TEM) TEM was performed on a JEOL JEM-2010 electron microscope at an acceleration voltage of 120 kV. Samples were prepared on a 400 meshes copper grid precoated with a carbon film and stained with osmium tetroxide (OsO4). 2.5. Isothermal titration calorimetry Calorimetric experiments were performed using a Microcal Isothermal Titration Calorimeter (ITC) (MicroCal, Northampton, MA). A sample cell with a volume of 1.35 mL was filled with a 0.01 wt.% polymer solution. ITC experiments were carried out at 25 °C by injecting 5 mM DOX solution from a 250 μL injection syringe into the sample cell, which was stirred at 400 rpm to ensure an optimum mixing efficiency. The observed experimental released heat data include the heat of dilution for the DOX in water and heat of DOX binding to the polymer chains. The heat of dilution can be determined by performing a blank titration in aqueous solution without polymer. In this study, the differential enthalpy curves of DOX binding onto P85PAA were subtracted from heat of dilution of DOX. Data analysis was performed using Microcal ORIGIN software. The uptake of DOX in the complex was quantified by a UV– vis spectrophotometer. P85PAA/DOX complexes were isolated by centrifugation (15,000 g, 30 min), and free DOX concentration in supernatant solutions was measured at λ = 481 nm, using an extinction coefficient, ε = 10,410 M− 1cm−1at 25 °C. 2.6. DOX release study The DOX release from the copolymer–DOX complex was studied using the dialysis procedure. DOX–polymer complexes were prepared in 10 mM phosphate buffer (pH 7.2) with a final DOX concentration of 0.09 mg/ml, and equilibrated in a shaking water bath at 37 °C for about 18 h in the dark. Then, the complexes in buffer (total volume is 3 ml, “donor” solution) were placed into a dialysis membrane with a molecular weight cut-off of 8000 Da and dialyzed against 100 ml PBS (10 mM at pH 7.2, “receiver” solution) in a shaking water bath at 37 °C kept in the dark for the DOX release study. The released DOX from “receiver” solution was assayed spectrophotometrically by measuring the absorbance of the solution at 481 nm. The samples taken for measurement were returned to the 100 ml receiver solution after measurement. The data were expressed as the mean value of three independent experiments with the reported standard deviation.

139

The cumulative fraction of released DOX versus time was expressed by Eq. (1). Cumulative released DOX% ¼

½DOXt  100k ½DOXcomplex

ð1Þ

where [DOX]t is the measured DOX concentration (mM) at time t, [DOX]complex is the total DOX complexed with copolymer in buffer solution. The cumulative amount of DOX released (μg) at sampling time t was expressed by Eq. (2): Cumulative released DOX ðμgÞ ¼ ½DOXt  ðVdonor þ Vreceiver Þ  MWDOX

ð2Þ

where [DOX]t is the measured DOX concentration (mM) at sampling time t, Vdonor and Vreceiver are the volume of “donor” solution and “receiver” solutions, respectively. MWDOX is the molecular weight of DOX. 3. Results and discussion 3.1. Polymer characterization Block copolymer PAA-b-P85-b-PAA with different PAA units (P85PAA60 and P85PAA28) were synthesized by the ATRP technique using protected group chemistry, followed by hydrolysis in acidic conditions [27–29]. The polymer composition and structure were characterized by nuclear magnetic resonance (NMR), gel permeation chromatographic (GPC) and potentiometric titration. The molecular weights of P85PAA60 and P85PAA28 were ∼10,000 and ∼7000 Da based on NMR measurements. The contents of carboxylic groups in the P85PAA60 and P85PAA28 were found to be ∼60 and ∼28 units, respectively. 3.2. Aggregation phenomena in aqueous solution of PAA-bP85-b-PAA block copolymer It is known that polyethylene oxide forms rather stable Hbonded complexes with polyacrylic acid in slightly acidic medium [30–33]. These complexes are water insoluble and their formation is enthalpy-driven [30]. We believe that the formation of such inter-or intra-molecular complexes may occur with P85PAA block copolymers. The I1/I3 ratio of the pyrene fluorescence emission intensity for 0.1 wt.% P85PAA60 aqueous solution is presented in Fig. 1. A sharp decrease in I1/I3 was observed at pH b 5.0, indicating the formation of complexes composed of a hydrophobic domain. At pH above 5.0, the increase of I1/I3 value indicated the dissociation of the complexes due to the ionization of PAA segments and the increase of electrostatic repulsion. For P85PAA60 solution at pH 7.2, the I1/I3 intensity ratio is 1.77, suggesting that the presence of unimeric polymer chains and the micellar structures may not exist in the polymer solution [34]. Dynamic light scattering and TEM studies were carried out to examine the aggregation behaviors of P85PAA60 block copolymer. It was found that P85PAA60 block copolymer formed large

140

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

produces physical cross-linking, resulting in the formation of P85PAA/DOX complex [26]. The proposed mechanism and microstructure of the complexes is illustrated in Fig. 4d [26]. 3.3. pH dependence of DOX binding capacity

Fig. 1. Variation of I1/I3 with the ionization of 0.1 wt.% P85PAA60 solution at 25oC, 0.6 μM pyrene.

The electrostatic interaction between ionized PAA segments and DOX was believed to be the basis for the drug binding with PAA modified Pluronic P85, which sustained the stability of the DOX/polymer complex. Thus, the effects of the polymer ionization on the DOX-to-polymer binding capacity were studied. The degree of neutralization, αN, of the carboxylic groups is defined as: ½BASE þ ½Hþ   ½OH  C½COOH

spherical particles with hydrodynamic diameter of 400–700 nm in the absence of DOX, as shown in Fig. 2a and b. Hydrogenbonding association between protons of carboxylic acid and ether

aN ¼

oxygen of Pluronic P85 (

Here, [BASE], [H+] and [OH–] are the molar concentrations of added NaOH, free hydrogen ion, and hydroxide ion, respectively, and C[COOH] is the total concentration of carboxylic groups. The titration curve yielded the ionization degree (α) of carboxyl groups in P85PAA60 copolymer versus pH value, and the ionization of P85PAA60 responsive to small changes in pH was observed in Fig. 5a, where 86% (molar ratio) of carboxyl groups was ionized at pH 7.2, and only 35% (molar ratio) of carboxyl groups were ionized at pH 5.0. Fig. 5b shows pH dependence of DOX content in the complexes, as determined spectrophotometrically. At pH 7.2, the equilibrium uptake of DOX was close to the stoichiometric 1:1 ratio of DOX and concentration of carboxyl groups, indicating the high binding affinity of DOX within the complex at pH 7.2. At pH 5.0, DOX content in the complex reached the plateau at 0.30 mM due to the lower ionization degree (α = 0.35) of PAA and thus lower binding capacity of DOX, as shown in Fig. 5b. We hypothesized that the protonation of carboxyl groups reduced the uptake of DOX by the block copolymer, which could be utilized for the pH-responsive character of DOX, which will be discussed later.

) as well as

hydrophobic interaction between PPO blocks promote the selfassociation of P85PAA60 block copolymer. It was reported that mixtures of poly(acrylic acid) and ethoxylated nonionic surfactants or PEO aqueous solutions led to the formation of compact hydrogen-bonding intra- or inter-polymer complexes [31–33]. When DOX was added to the polymer solution at pH 3.87, the formation of larger spherical aggregates were observed upon the binding of DOX, in which the hydrogen bonding between Pluronic-b-PAA block copolymer and DOX contributed to the formation of drug–polymer complexes [26]. The hydrogenbonding as well as the hydrophobic interaction enhanced the hydrophobicity of the P85PAA60/DOX complex, resulting in the formation of flocculated complex with a hydrodynamic diameter of ∼1000 nm (Fig. 3a and b). The proposed mechanism of the complexation in the P85PAA60/DOX system at pH 3.87 is illustrated in Fig. 3c. At pH 7.2, polymer aggregates dissociated completely into unimeric polymer chains with an intensity-average hydrodynamic diameter of 9–10 nm, as shown in Fig. 4a. With the addition of DOX, the formation of P85PAA/DOX complex was induced by DOX binding via electrostatic interaction that can be expressed as: ∼ COO− þ Naþ þ DOXþ þ Cl− ⇔∼COO− DOXþ þ Naþ þ Cl−

The pKa of DOX is 8.6, and protonated amino nitrogen provides the positive charge at pH 7.2 [5,36]. P85PAA60/DOX complex was stabilized by the electrostatic repulsion of free [COO–] from PAA segments and steric effect of PEO block from P85 segment. At pH 7.2, the formation of spherical particles with size of 150–200 nm was observed in the TEM image shown in Fig. 4c. The smaller clusters were found to be about 20–30 nm, which were probably formed by several polymeric chains and DOX molecules, which was consistent with the dynamic light scattering measurement shown in Fig. 4b. The electrostatic interaction between ionized PAA segments and DOX is the basis for the drug binding with P85PAA block copolymer, and the stacking interaction of polymer-bound DOX

ð3Þ

3.4. Stability of P85PAA/DOX complex in the presence of salt For P85PAA60/DOX complex, the net positive charge of DOX molecules at pH 7.2 provided the driving force for

Fig. 2. Aggregation behavior of 0.1 wt.% P85PAA60 solution at pH 3.87, 25 °C (a) Hydrodynamic diameter of DOX-free polymer; (b) TEM image of DOX-free polymer.

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

141

Fig. 3. Complexation of P85PAA60/DOX at pH 3.87, [DOX] = 0.05 mM (a) Hydrodynamic diameter of DOX-loaded complex, (b) TEM image of DOX-loaded complex, (scale bar is equal to 1 μM); (c) Schematic representation of the complexation behavior at pH 3.87.

binding with negatively charged [COO − ] sites on PAA segments by electrostatic interaction. The dissociation of complexes may occur in the presence of high concentration of salt due to the shielding effect of salt between the charged

binding sites. The practical value of this work lies in the stability of the complex at physiological ionic strength. ITC experiments were performed to investigate DOX binding in the presence of 0.15 M sodium chloride. Fig. 5c shows the differential enthalpy

Fig. 4. Complexation of P85PAA60/DOX at pH 7.2, [DOX] = 0.05 mM (a) Hydrodynamic diameter of DOX-free polymer; (b) Hydrodynamic diameter of DOXloaded complex; (c) TEM image of DOX-loaded complex; (d) Schematic representation of the complexation behavior at pH 7.2 [refer to Ref. [26]].

142

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

uptake at higher salt concentration. However, within the range of concentrations studied, similar linear increase in DOX uptake was observed upon DOX addition to the polymer solutions, as shown in Fig. 5d. The binding capacity of P85PAA block copolymer was not significantly affected by the presence of salt, indicating that the dissociation of P85PAA60/DOX complex did not occur with the addition of 0.15 M NaCl. The result of DOX uptake suggested that other non-electrostatic interaction may also play a role in the stability of the copolymer–DOX complex. The DOX binding to PAA has been shown to be governed by electrostatic and stacking interactions resulting in the formation of PAA/DOX complexes [35]. The PAA-bound DOX molecules can induced strong binding with each other via stacking interaction of planar aromatic anthracycline rings, leading to the formation of stable complex. Hydroxyl group attached to the anthraquinone ring system is known to be a key participant in the hydrogen-bonding interaction between DOX molecules and negatively charged DNA [5]. In the DOX binding to P85PAA block copolymer, it may participate in the formation of hydrogen-bonds. Therefore, the presence of 0.15 M NaCl in solution can weaken the electrostatic interactions between oppositely charged DOX and polymer, but may not trigger the significant dissociation of the complex due to the contribution of the non-polyelectrolyte interaction, such as π−π stacking interaction and H-bonding. 3.5. DOX release from PAA-b-P85-b-PAA/DOX complexes The DOX molecule contains an amino group with a pKa of 8.6 [37], and the pK0 for PAA carboxylic groups is about 4.8.

Fig. 5. (a) Potentiometric titration curve of P85PAA60, 0.1 wt.%, 25 °C; (b) Uptake of DOX by P85PAA60 at pH 5.0 and pH 7.2, obtained from titrating 5 mM DOX into 0.01 wt.% P85PAA60 solution; (c) Binding isotherms obtained from titration 5 mM DOX in 0.01 wt.% P85PAA solution with or without 0.15 M sodium chloride; (d) Uptake of DOX by P85PAA60.

curves at pH 7.2, which were measured from the gradual injection of 5 mM DOX solutions into P85PAA60 aqueous solutions in the presence and absence of salt. The experimental results indicated that the interaction of copolymer and DOX was strongly salt dependent. Salt causes a shielding effect, as in the case of drug binding shown in Fig. 5c, the enthalpy of DOX binding decreased from ∼ − 25 to ∼ − 15 kJ/mol in the presence of 0.15 M NaCl, indicating that addition of salt shielded the electrostatic attraction between oppositely charged copolymer and DOX, and thus weakened the DOX binding with P85PAA block copolymer. Thus, we expected to observe lower DOX

Fig. 6. Release of DOX in 10 mM PBS, 37 °C. (a) pH-dependent release of DOX from P85PAA60-DOX complex at pH of 5.0, 6.0 and 7.2. (b) Cumulative DOX release (μg) at pH of 5.0 and 7.2. The loading capacity of DOX for each sample is 270 μg.

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

143

This provided the right range at physiological pH (∼ 7 to 7.2), where electrostatic interactions governed the complex formation. The complex particles dissociated into different aggregated states at different pHs. Hence, the “release” curves that we reported in Fig. 6 represented several dissociation processes. It was observed that the release of DOX from P85PAA60 is a pHdependent process, where the cumulative fraction of DOX release increased with decreasing pH in PBS buffer (Fig. 6a). The release profile was characterized by two stages: (I) a rapid release stage (the first 10 h), (II) a slow release stage (from 10– 30 h). Fig. 6b shows that the cumulative DOX released at pH 5.0 was about 200 μg in the first 10 h (stage I), but only another ∼ 50 μg DOX was released during the next 20 h (stage II). To clearly observe these dissociation processes at pH 5.0 and 7.2, DOX release rate profile was calculated and plotted against time, as shown in Fig. 7. DOX release rate (μg/h) was expressed as: d ½DOX ½DOXtn  ½DOXtn1 ¼ dt tn  tn1

ð4Þ

where d[DOX] / dt (μg/h) is the average DOX release rate during sampling time interval, [DOX]tn (μg) and [DOX]tn − 1 (μg) are the DOX concentration at sampling time tn (h) and tn − 1(h), respectively. tn − tn − 1 is the interval time between two samplings. Again, two release processes were clearly discernible from Fig. 7. At pH 5.0 (Fig. 7a), a gradual decrease in the release rate was observed in the first 10 h, and stage II appeared to be nearly constant, suggesting that different mechanisms may be in operation. In the initial stage (I), diffusion of DOX present in the donor solution in equilibrium with the fast dissociation of P85PAA/DOX complex, and the release rate depended on pH.

Fig. 8. Comparison of DOX release from P85PAA60 and P85PAA28 block copolymers in 10 mM PBS, at 37 °C. (a) pH 7.2; (b) pH 5.0. The loading capacity of DOX for each sample is ~ 270 μg.

Faster release rate was achieved at pH 5.0 due to the lower electrostatic binding compared to pH 7.2. This process can be described by the following dissociation equilibrium: ~ COO DOXþ þ Hþ þ Cl ⇔~ COO Hþ þ DOXþ þ Cl ð5Þ Once a substantial portion of DOX in the dissociated complex was released, the “release” of DOX proceeded to stage II, and the “release rate” maybe governed by the re-equilibration of the complex, which is probably associated with the particle conformational change, and ultimately, with the polymer chain relaxation and diffusion processes [38]. The dissociation of DOX from the polymer-drug complex in the second stage may be expressed by: ~ COO DOXþ þ Naþ þ Cl ⇔ ~ COO Naþ þ DOXþ þ Cl ð6Þ

Fig. 7. DOX release rate profile from DOX–P85PAA60 complex in 10 mM PBS, 37 °C. (a) DOX release rate at pH 5.0 (b) DOX release rate at pH 7.2. The loading capacity of DOX for each sample is 270 μg.

This stage is similar to the release behavior at pH 7.2, but the rate of dissociation of the DOX/polymer complex was limited by the high DOX concentration present in the buffer. At pH 7.2, fast transitions of DOX release rate in the first 2 h were observed, as shown in Fig. 7b. Initially, the release rate at pH 7.2 decreased rapidly in the earlier part of stage I (∼2 h), and then gradually decreased until 10 h. At pH 7.2, the overall release rate of DOX was much lower than at pH 5.0, due to the stronger electrostatic interaction between DOX molecules and PAA segments, which resulted in less DOX being released over the first 10 h (only ∼ 108 μg DOX was released at pH 7.2 compared to 202 μg at pH 5.0 over the first 10 h).

144

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145

Fig. 8 compares the cumulative release profiles for DOX from P85PAA containing different molecular weight PAA segments. Compared to P85PAA60, DOX loaded in P85PAA28 block copolymer (containing ∼28 carboxylic acid) was released at a faster rate at both pH 7.2 (Fig. 8a) and pH 5.0 (Fig. 8b). At pH 7.2, 50% of DOX was released from P85PAA28 block copolymer after 10 h, while 40% of DOX loaded in P85PAA60 was released in 10 h. At pH 5.0, over the first 6 h, 70% of DOX was released from P85PAA28 compared to 50% from P85PAA60 block copolymer. Ende and Peppas reported that the transport of ionizable drugs was influenced by the copolymer composition [39]. For instance, for the increased amounts of PAA from 30 to 40 mol%, the diffusion coefficient of oxprenolol HCl shifted to a lower value due to their interaction with ionized carboxylic acid groups [39]. In our PAA-modified Pluronic system, faster release rate was achieved from P85PAA28 block copolymer, which was probably due to the shorter PAA segments, resulting in the weaker electrostatic interaction and the fast dissociation equilibrium of P85PAA28/DOX complex in the buffer solutions. 4. Conclusions New pH-responsive block copolymer PAA-b-P85-b-PAA was prepared for the delivery of DOX. We examined the effect of pH on the aggregation behaviors of the block copolymer, DOX loading and release behaviors. It was found that complexation of DOX with the block copolymer was a pHdependent process. DOX binding at pH 3.87 induced the formation of large spherical complexes (radius of about 500 nm) probably due to the intra- or intermolecular hydrogen bonding between DOX and the copolymer. The complexes of P85PAA60 and DOX appeared to be spherical particles with sizes of about 150–200 nm at pH 7.2. The electrostatic interaction between ionized PAA segments and DOX as well as the stacking of polymer-bound DOX molecules produced the DOX/PAA-b-P85-b-PAA complex at pH of 7.2. DOX release from polymer-drug complexes was observed, depending on the release medium and polymer composition. The release of DOX was highly pH-responsive over a physiological pH range. Protonation of carboxyl groups at a mildly acidic condition resulted in an accelerated release of DOX at pH 5.0 compared to pH 7.2. The overall release rate of DOX was much lower at pH of 7.2, due to the strong electrostatic interaction between DOX molecules and PAA segments. For the block copolymer with shorter PAA segments, faster release of DOX was achieved at the initial stage due to the fast dissociation of polymer/DOX complex. Acknowledgements We are grateful for the financial support provided by the Singapore-MIT Alliance. References [1] A. Rősler, G.W.M. Vandermeulen, H. Klok, Advanced drug delivery devices via self-assembly of amphiphilic block copolymers, Adv. Drug Deliv. Rev. 53 (2001) 95–108.

[2] A. Lavasanifar, J. Samuel, G.S. Kwon, Poly(ethylene oxide)-block-poly (L-amino acid) micelles for drug delivery, Adv. Drug Deliv. Revs. 54 (2002) 169–190. [3] G.S. Kwon, K. Kataoka, Block copolymer micelles as long circulating drug vehicles, Adv. Drug Deliv. Rev. 16 (1995) 295–309. [4] J.G. Hardman, L.E. Limbird (Eds.), Goodman and Gilman's The Pharmacological Basis of Therapeutics, 9th edn., McGraw-Hill, New York, 1996, p. 1265. [5] J.B. Chaires, S. Satyanarayana, D. Suh, I. Fokt, T. Przewloka, W. Priebe, Parsing the Free Energy of Anthracycline Antibiotic Binding to DNA, Biochemistry 35 (1996) 2047–2053. [6] T. Nakanishi, S. Fukushima, K. Okamoto, M. Suzuki, Y. Matsumura, M. Yokoyama, T. Okano, Y. Sakurai, K. Kataoka, Development of the polymer micelle carrier system for doxorubicin, J. Control. Release 74 (2001) 295–302. [7] M. Hrubý, Č. Koňák, K. Ulbrich, Polymeric micellar pH-sensitive drug delivery system for doxorubicin, J. Control. Release 103 (2005) 137–148. [8] T. Mrkvan, M. Sirova, T. Etrych, P. Chytil, J. Strohalm, D. Plocova, K. Ulbrich, B. Rihova, Chemotherapy based on HPMA copolymer conjugates with pH-controlled release of doxorubicin triggers anti-tumor immunity, J. Control. Release 110 (2005) 119–129. [9] S.R. Yang, H.J. Lee, J.D. Kim, Histidine-conjugated poly(amino acid) derivatives for the noval endosomolytic delivery carrier of doxorubicin, J. Control. Release 114 (2006) 60–68. [10] E.R. Gillies, J.M.J. Frechet, pH-Responsive copolymer assemblies for controlled release of Doxorubicin, Bioconjug. Chem. 16 (2005) 361–368. [11] K. Kataoka, T. Matsumoto, M. Yokoyama, T. Okano, Y. Sakurai, S. Fukushima, K. Okamoto, G.S. Kwon, Doxorubicin-loaded poly(ethylene glycol)-poly(b-benzyl-L-aspartate) copolymer micelles: their pharmaceutical characteristics and biological significance, J. Control. Release 64 (2000) 143–153. [12] X. Shuai, H. Ai, N. Nasongkla, S. Kim, J. Gao, Micellar carriers based on block copolymers of poly(e-caprolactone) and poly(ethylene glycol) for doxorubicin delivery, J. Control. Release 98 (2004) 415–426. [13] C.H. Wang, C.H. Wang, G.H. Hsiue, Polymeric micelles with a Phresponsive structure as intracellular drug carriers, J. Control. Release 108 (2005) 140–149. [14] K. Greish, T. Sawa, J. Fang, T. Akaike, H. Maeda, SMA-doxorubicin, a new polymeric micellar drug for effective targeting to solid tumours, J. Control. Release 97 (2004) 219–230. [15] E.V. Batrakova, T.Y. Dorodnych, E.Y. Klinskii, E.N. Kliushnenkova, O.B. Shemchukova, O.N. Goncharova, S.A. Arjakov, V.Y. Alakhov, A.V. Kabanov, Anthracycline antibiotics non-covalently incorporated into the block copolymer micelles: in vivo evaluation of anti-cancer activity, Br. J. Cancer 74 (1996) 1545–1552. [16] A.V. Kabanov, E.V. Batrakova, D.W. Miller, Pluronic block copolymers as modulators of drug efflux transporter activity in the blood–brain barrier, Adv. Drug Deliv. Rev. 55 (2003) 151–164. [17] A.V. Kabanov, E.V. Batrakova, V.Yu. Alakhov, Pluronic block copolymers for overcoming drug resistance in cancer, Adv. Drug Deliv. Rev. 54 (2002) 759–779. [18] E.V. Batrakova, S. Li, V.Yu. Alakhov, W.F. Elmquist, D.W. Miller, A.V. Kabanov, Sensitization of cells overexpressing multidrug-resistant proteins by Pluronic P85, Pharm. Res. 20 (2003) 1581–1590. [19] T. Minko, E.V. Batrakova, S. Li, Y. Li, R.I. Pakunlu, V.Yu. Alakhov, A.V. Kabanov, Pluronic block copolymers alter apoptotic signal transduction of doxorubicin in drug-resistant cancer cells, J. Control. Release 105 (2005) 269–278. [20] N.S. Melik-Nubarov, M.Yu. Kozlov, Evaluation of partition coefficients of low molecules weight solutes between water and micelles of block copolymer of ethylene oxide based on dialysis kinetics and fluorescence spectroscopy, Colloid Polym. Sci. 276 (1998) 381–387. [21] Z. Liu, R. Cheung, X.Y. Wu, J.R. Ballinger, R. Bendayan, A.M. Rauth, A study of doxorubicin loading onto and release from sulfopropyl dextran ion-exchange microspheres, J. Control. Release 77 (2001) 213–224. [22] G.M. Eichenbaum, P.F. Kiser, A.V. Dobrynin, S.A. Simon, D. Needham, Investigation of the swelling response and loading of ionic microgels with drugs and proteins: the dependence on cross-link density, Macromolecules 32 (1999) 4867–4878.

Y. Tian et al. / Journal of Controlled Release 121 (2007) 137–145 [23] V. Alakhov, G. Pietrzynski, K. Patel, A. Kabanov, L. Bromberg, T.A. Hatton, Pluronic block copolymers and Pluronic poly(acrylic acid) microgels in oral delivery of megestrol acetate, J. Pharm. Pharmacol. 56 (2004) 1233–1241. [24] R. Barreiro-Iglesias, L. Bromberg, M. Temchenko, T.A. Hatton, A. Concheiro, C. Alvarez-Lorenzo, Solubilization and stabilization of camptothecin in micellar solutions of pluronic-g-poly(acrylic acid) copolymers, J Control. Release 97 (2004) 537–549. [25] L. Bromberg, M. Temchenko, T.A. Hatton, Dually responsive microgels from polyether-modified poly(acrylic acid): swelling and drug loading, Langmuir 18 (2002) 4944–4952. [26] Y. Tian, P. Ravi, L. Bromberg, T. Alan Hatton, K.C. Tam, Synthesis and aggregation behavior of Pluronic F87/Poly(acrylic acid) block copolymer in the presence of doxorubicin, Langmuir 23 (2007) 2638–2646. [27] Q. Ma, K.L. Wooley, The preparation of t-butyl acrylate, methyl acrylate, and styrene block copolylers by atom transfer radical polymerization: precursors to amphiphilic and hydrophilic block copolymers and conversion to complex nanostructured materials, J. Polym. Sci.: Part A: Polym. Chem. 38 (2002) 4805–4820. [28] S.J. Hou, E.L. Chaikof, D. Taton, Y. Gnanou, Synthesis of water-soluble star-block and dendrimer-like copolymers based on poly(ethylene oxide) and poly(acrylic acid), Macromolecules 36 (2003) 3874–3881. [29] Z.H. Lu, G.J. Liu, Polysulfone-graft-poly(tert-butyl acrylate): synthesis, nanophase separation, poly(tert-butyl acrylate) hydrolysis, and pHdependent iridescence, Macromolecules 37 (2004) 174–180. [30] V.Yu. Baranovsky, A.A. Litmanovich, I.M. Papisov, V.A. Kabanov, Quantitative studies of interaction between complementary polymers and oligomers in solutions, Eur. Polym. J. 17 (1981) 969–979.

145

[31] I.D. Robb, P. Stevenson, Interaction between poly(acrylic acid) and ethoxylated nonionic surfactant, Langmuir 16 (2000) 7168–7172. [32] S. Nishi, T. Kotaka, Complex-formation poly(oxyethylene):poly(acrylic acid) interpenetrating polymer networks. 1. Preparation, structure, and viscoelastic properties, Macromolecules 18 (1985) 1519–1524. [33] I. Iliopoulos, R. Audebert, Complexation of acrylic acid copolymers with polybases: importance of cooperative effects, Macromolecules 24 (1991) 2566–2575. [34] P. Alexandridis, T. Nivaggioli, T.A. Hatton, Temperature effects on structural properties of Pluronic P104 and F108 PEO-PPO-PEO block copolymer solutions, Langmuir 11 (1995) 1468–1476. [35] M.V. Kitaeva, N.S. Melik-Nubarov, F.M. Menger, A.A. Yaroslavov, Doxorubicin-poly(acrylic acid) complexes: interaction with liposome, Langmuir 20 (2004) 6575–6579. [36] M.V. Kitaeva, N.S. Melik-Nubarov, F.M. Menger, A.A. Yaroslavov, Membrane transport of a polyacid-tied Doxorubicin, Langmuir 20 (2004) 6796–6799. [37] P.R. Harrigan, K.F. Wong, T.E. Redelmeier, J.J. Wheeler, P.R. Cullis, Biochim. Biophys. Acta 1149 (1993) 329–338. [38] J. Wang, B.M. Wang, S.P. Schwendeman, Characterization of the initial burst release of a model peptide from poly(D,L-lactide-co-glucolide) microspheres, J. Control. Release 82 (2002) 289–307. [39] M.T. Ende, N.A. Peppas, Transport of ionizable drugs and proteins in crosslinked poly(acrylic acid) and poly(acrylic acid-co-2-hydroxyethyl methacrylate) hydrogels. II. Diffusion and release studies, J. Control. Release 48 (1997) 47–56.