pH-sensitive drug release of star-shaped micelles with OEG brush corona

Abstract

Star-shaped polymers are attractive as drug carriers due to the integration of relatively easy synthesis compared to the preparation of hyper-branched dendrimers and their potential ability to form unimolecular nanoparticles with enhanced stability to survive the high dilution-associated deformation and subsequently premature drug release in the body fluids and blood during circulation. Clearly, the star structure exerts a significant effect on their bioproperties and potential applications. In this study, three star-shaped amphiphilic copolymers with the same polymer compositions but different star structures were synthesized to investigate the effect of star architecture on their properties as well as potential applications as drug carriers. Three different branched alcohols, 2-(hydroxymethyl) propane-1,3-diol, pentaerythritol and dipentaerythritol with 3, 4, and 6 respective hydroxyl groups were chosen as the starting core to generate the target star-shaped amphiphilic block copolymers composed of hydrophobic poly(ε-caprolactone) (PCL) and hydrophilic poly(oligoethyleneglycol methacrylate) (POEGMA) with 3, 4, and 6 corresponding star arms by a combination of ring-opening polymerization (ROP) and atom transfer radical polymerization (ATRP). Dynamic light scattering (DLS) measurements and pyrene fluorescence probe technique confirmed the capability of the resultant star-shaped amphiphilic copolymers to form unimolecular micelles with average diameters smaller than 50 nm in an aqueous phase. A comparison of the drug loading capacity revealed that the micelles of 4-star-PCL–POEGMA (4s-PCL–POEGMA) exhibited the highest drug loading content (DLC) of all the three formulations. More importantly, in vitro doxorubicin (DOX) release study showed unique pH-mediated drug release behaviors, i.e., dramatically accelerated release at pH 6.0 but a much slower profile at pH 7.4, from these generally recognized “pH-insensitive” 4-arm star-shaped micelles. An acid-triggered degradation and an acid–base titration were performed to further reveal that such interesting pH-responsive drug release behaviors were attributed primarily to the hydrophilic corona of OEG brushes. This study is believed to provide a new insight into the structure–bioproperties relationship of star-shaped polymers, thus should be useful for the future design and development of novel star-shaped polymers for controlled drug delivery.

---

Introduction

In the past several decades, micelles formed by the self-assembly of amphiphilic block copolymers have drawn considerable attention in the field of nanomedicine as drug delivery systems due to their unique core–shell structures.^1–8 The micellar drug carriers can improve the solubility of water-insoluble drugs, stabilize and protect encapsulated therapeutic agents that are sensitive to the surrounding environment, reduce the nonspecific uptake by the reticulo-endothelial system (RES) and prolong the circulation time in the blood.^1,2 However, the in vivo instability of conventional micelles self-assembled from the linear amphiphilic block copolymers hampers their practical applications and clinical translations because of the largely diluted polymer concentration in the bloodstream and consequent disassembly of micelle structure once the polymer concentration drops below the critical micelle concentration (CMC).⁹ Such deformation further results in a premature release and loss of the therapeutic payloads in blood circulation, causing significant off-target and side effects. On the other hand, it is worth pointing out that permanently stable micelles are undesirable as well because the effective release of cargoes in the intracellular environment is also required to achieve an ideal therapeutic efficiency.

To improve the stability of traditional self-assembled micelles, two primary strategies have been developed by using cross-linked micelles¹⁰ or star/hyperbranched structure-based unimolecular formulations.^11–13 In the case of cross-linked micelles, cross-linking can be performed either in the micellar core, at the core/shell interface/junction, or in the shell. However, this strategy suffers mainly from additional chemistry design to incorporate cross-linking structures. Alternatively, fabrication of unimolecular micelles using star-shaped polymers or dendrimers is an elegant approach to address the stability issue due to the integration of enhanced stability irrespective of high dilution in the blood and facile modification of micellar corona with function groups available.^14–17 Compared to the complicated multi-step synthesis of hyperbranched dendrimers, the relative ease for the preparation of star-shaped polymers has made them attract great attention in past decade. A survey of literature indicates extensive and intensive studies regarding the star structure–properties relationships or targeted design of star polymers for biomedical applications.^18–27 Notably, Schubert's group reported the synthesis and structure-dependent solution behavior of star-shaped amphiphilic block copolymers based on hydrophobic poly(ε-caprolactone) (PCL) and hydrophilic poly(oligoethyleneglycol methacrylate) (POEGMA), and their potential biomedical applications as unimolecular drug carriers.^28–30 Specifically, a series of PPEGMA with M_n of 500, 950, and 1100 were used to fabricate the brush-like hydrophilic shell, and a comparison study revealed that the molar mass of the macromonomer used was a more crucial factor, compared to the chain length of the hydrophilic block, to form an efficiently stabilizing hydrophilic shell.³⁰ Since our group has a long-standing interest in fabricating nanocarriers with P(OEGMA300) brush-like shielding corona for drug and gene delivery,^31–34 we reported herein the construction of star-shaped amphiphilic block copolymer using OEGMA (M_n = 300), and further investigated the in vitro drug release behaviors of the resultant star-shaped micelles. This study is therefore believed to make a useful supplementary to Schubert's studies. More importantly, different from the traditional strategy leading to pH-responsive nanocarriers using pH-sensitive moiety, such as the reported pH-responsive star-shaped micelles based on a tri-block copolymer of PCL, poly(2-(diethylamino)ethyl methacrylate) (PDEADMA), and PPEGMA,^35–38 our study presented pH-mediated drug release behaviors from a generally recognized “pH-insensitive” star-shaped PCL–POEGMA micelles, thus providing new insights into the structure–bioproperties correlation of star-shaped polymers.

In this study, three star-shaped amphiphilic block copolymers with similar polymer compositions but different star structures were synthesized to investigate the effect of star architecture on their bioproperties as well as biomedical applications as drug carriers. Three different branched alcohols, 2-(hydroxymethyl) propane-1,3-diol, pentaerythritol, and dipentaerythritol with 3, 4, and 6 respective hydroxyl groups were chosen as the starting core to generate the target star-shaped amphiphilic copolymers composed of hydrophobic PCL and hydrophilic POEGMA with 3, 4, and 6 corresponding star arms by integrated ring-opening polymerization (ROP) and atom transfer radical polymerization (ATRP). The size and morphology of micelles formed were analyzed as a function of the star structure, and compared with each other. A further comparison of the in vitro drug loading capacity and drug release profile of each formulation was presented. Interestingly, a unique pH-mediated drug release profile, i.e., dramatically accelerated release at pH 6.0 while much slower profile at pH 7.4, was recorded for 4-star-PCL–POEGMA (4s-PCL–POEGMA) micelles. Such pH-responsiveness was therefore evaluated by an acid-triggered degradation and an acid–base titration study.

Experimental section

Materials

ε-Caprolactone (CL) (Sigma-Aldrich) was dried over CaH₂ and distilled under reduced pressure prior to use. Oligo(ethylene glycol) monomethylether methacrylate (OEGMA, M_n = 300 g mol⁻¹ and 4–5 pendent EO units, Sigma-Aldrich) was purified by passing through a column filled with basic alumina to remove the inhibitor. Pentaerythritol, 2-(hydroxymethyl)propane-1,3-diol, stannous(II) octanoate (Sn(Oct)₂), 2-bromoisobutyryl bromide, triethylamine (TEA), copper(I) bromide (CuBr) and bipyridine (bpy) were purchased from Sigma-Aldrich and used as received. Dipentaerythritol (analytical grade) was provided by Aladdin. Tetrahydrofuran (THF) and toluene were dried by refluxing over sodium and distilled prior to use. All other reagents and solvents were used without further purification.

Preparation of 4-arm star-shaped PCL–OH (4s-PCL–OH)

4s-PCL–OH was synthesized by Sn(Oct)₂-catalyzed ROP of ε-CL using pentaerythritol as the initiator. In a typical procedure, pentaerythritol (0.1375 g, 1 mmol) and ε-CL (4.6117 g, 40 mmol) were added into a thoroughly dried tube equipped with a magnetic stirring bar. The tube was connected to a standard Schlenk line and the system was degassed via three freeze–pump–thaw cycles, and then the tube was put into an oil bath preheated at 140 °C and stirred for ∼5 min to obtain a homogeneous solution. Thereafter, Sn(Oct)₂ (32.4 mg, 0.08 mmol) in dry toluene was added to the mixture under the protection of nitrogen flow and three freeze–pump–thaw cycles were applied again, followed by immersing the tube in the oil bath thermostated at 140 °C. After 6 h, the very viscous reaction mixtures were cooled down, diluted using THF, and precipitated in 10-fold excess of ice-cold methanol to yield crude product. The product was purified by re-dissolving/precipitating in THF/methanol for three times, and further dried under vacuum until constant weight (yield, 86%).

Synthesis of 4-bromoend-capped macroinitiators (4s-PCL–iBuBr)

4s-PCL–iBuBr was synthesized by esterification of 4s-PCL–OH with excess 2-bromoisobutyryl bromide in the presence of TEA. 4s-PCL–OH (1.0003 g, 0.25 mmol) and TEA (0.3986 g, 3.94 mmol) were dissolved in 8 ml of dry dichloromethane (DCM) and cooled to 0 °C in an ice bath. 2-Bromoisobutyryl bromide (0.7038 g, 3.03 mmol) was added and the reaction mixture was stirred at 0 °C for 30 min, followed by stirring at room temperature for another 48 h. After completion, the reaction mixture was concentrated by vacuum evaporation, and further dissolved in THF. After filtration of insoluble quaternary ammonium salts, the filtrate was added dropwise into excess ice-cold methanol to precipitate the crude product. The product was purified by re-dissolving/precipitating in THF/methanol for three times, and further dried under vacuum (yield, 92%).

Synthesis of 4-arm star-shaped amphiphilic block copolymers (4s-PCL–POEGMA) by ATRP of OEGMA using 4s-PCL–iBuBr as a macroinitiator

4s-PCL–POEGMA was prepared by ATRP of OEGMA using 4s-PCL–iBuBr as a macroinitiator. Typically, 4s-PCL–iBuBr (30 mg, 0.008 mmol), bpy (10.2 mg, 0.06 mmol) and OEGMA (0.9575 g, 3.19 mmol) were dissolved in anisole (3.192 ml). After three freeze–pump–thaw cycles, CuBr (4.6 mg, 0.03 mmol) was introduced under the protection of nitrogen flow. After another three freeze–pump–thaw cycles, the reaction mixture was sealed and placed in an oil bath thermostated at 60 °C to start the polymerization. After 2.5 h, the reaction was stopped by exposure to air and diluted using THF. The crude product was collected by precipitation in excess ice-cold n-hexane. To remove the copper catalyst and any unreacted monomer, the crude product was dissolved in 2 ml of DMF, placed in a dialysis tube (molecular weight cut-off (MWCO), 3.5 kDa) and then subjected to dialysis against distilled water for 24 h, during which the water was renewed every 8 h. The purified 4s-PCL–POEGMA was harvested by freeze-drying (yield, 83%).

3s-PCL–POEGMA and 6s-PCL–POEGMA were prepared following the same procedures except using 2-(hydroxymethyl) propane-1,3-diol and dipentaerythritol as the starting initiators, respectively. The detailed amount of each reagent used for the synthesis of 3s- and 6s-PCL–POEGMA was summarized in Tables S1–S3 as ESI.†

Polymer characterization

¹H NMR spectra were recorded on a JNM-ECS spectrometer at 400 MHz using CDCl₃ as the solvents. Molecular weight and polydispersity index (PDI) of the polymers synthesized were analyzed by gel permeation chromatography (GPC) equipped with a Waters 1515 pump, a Waters 2414 differential refractive index detector and a series of three linear Styragel columns (HT2, HT4, and HT5) at 35 °C. THF was used as the eluent at a flow rate of 1 ml min⁻¹ and poly(methyl methacrylate) (PMMA) standards were used for the calibration.

Preparation and characterization of star-shaped micelles

Take 4-arm star-shaped micelles (4s-micelles) as an example, 4s-PCL–POEGMA (1.5 mg) in 1 ml of DMF was placed in a dialysis tube and dialyzed against distilled water for 24 h to prepare the micelle solution with a concentration of 0.25 mg ml⁻¹.

The TEM images were recorded on a JNM-2010 instrument operating at an acceleration voltage of 200 keV. To prepare specimens for TEM observation, a drop of micelle solution was deposited onto a carbon-coated copper grid. After deposition, excess solution was removed using a strip of filter paper. The sample was further stained using phosphotungstic acid (2% w/w) and dried in air prior to visualization.

The average hydrodynamic size of star-shaped micelles was measured by dynamic light scattering (DLS) on a BI-200SM (Brookhaven, USA) instrument. The scattering angle was fixed at 90°. Polymer solutions with various concentrations of 1.0, 0.5, 0.25, and 0.125 mg ml⁻¹ were evaluated.

Fluorescence spectra were recorded on a LS55 luminescence spectrometer (Perkin-Elmer) using pyrene as a fluorescence probe. 1 ml of pyrene solution (3 × 10⁻⁶ M in acetone) was added to containers, and the acetone was allowed to evaporate. Then 5 ml of polymer aqueous solution at different concentrations were added to the containers containing the pyrene residue and the combined solution of pyrene and copolymers was equilibrated at room temperature in dark for 24 h prior to measurements. The final concentration of pyrene was 6 × 10⁻⁷ M in water. Excitation was carried out at 340 nm, and emission spectra were recorded ranging from 350 to 600 nm. Both excitation and emission bandwidths were 10 nm. From the pyrene emission spectra, the intensities (peak height) of the first (I₃₇₃) and the third bands (I₃₉₃) were recorded.

In vitro drug loading and drug release study

DOX·HCl (1 mg) and TEA (0.26 g) were dissolved in 2 ml of DMSO and stirred overnight in dark at room temperature to obtain DOX. Next, the star-PCL–POEGMA (10 mg) in 2 ml of DMSO was added to the above DOX solution and stirred at room temperature for 1 h. Thereafter, the above mixture was added dropwise into 4 ml of ultra-purified water under vigorous stirring. After stirring for another 1 h, the solution was put into a dialysis tube and dialyzed against 5 l of distilled water for 24 h, during which the water was renewed every 8 h. Finally, the drug-loaded star-shaped micelles were harvested by freeze-drying. To determine the drug loading content (DLC) and entrapment efficiency (EE), the freeze-dried drug-loaded micelles were re-dispersed in PBS (pH 7.4). The concentration of DOX was determined by measuring the absorbance at 485 nm using a Lambda 35 UV-Vis spectrometer (Perkin-Elmer).

In vitro drug release study was carried out in phosphate buffer saline (PBS, pH 7.4, 12 mM) and saline sodium citrate (SSC, pH 6.0, 12 mM) at 37 °C. The freeze-dried drug-loaded nanoparticles was re-dispersed in buffer solution to prepare a drug-loaded micelle solution at a concentration of 1.0 mg ml⁻¹. 1 ml of the solution was loaded in a dialysis tube, and then immersed in a tube containing 25 ml of release medium of different pHs. The tube was kept in a horizontal laboratory shaker thermostated at a constant temperature of 37 °C and a stirring speed of 120 rpm. At predetermined time intervals, 3 ml of release medium was taken out and replenished with equal volume of fresh medium. The drug concentration was determined by measuring the absorbance of DOX at 485 nm using a calibration curve. The amount of DOX released in PBS (pH 7.4) or SSC (pH 6.0) was determined by UV-Vis spectrometer. The experiment was performed in triplicate for each sample.

Acid-triggered degradation study

4s-PCL–POEGMA (50 mg) was dissolved in 50 ml of SSC (pH 6.0). Hydrolysis was performed in a shaker thermostated at 37 °C for 96 h. At predetermined time intervals, 8 ml of the solution was taken out and lyophilized. The freeze-dried products were dissolved in 1 ml of THF or 600 μl of CDCl₃ and filtered to remove any insoluble salts prior to GPC and ¹H NMR analyses.

Acid–base titration

The buffering capacity of three star-PCL–POEGMA copolymers was determined by acid–base titration over a pH range from 3.0 to 11.0. Briefly, certain amount of each polymer was firstly dissolved in 0.15 M NaCl aqueous solution (0.2 mg ml⁻¹). After adjusting the starting pH to 11.0 with 0.1 M NaOH, the solution was titrated with 0.1 M HCl using a pH meter. The buffering capacity was determined as μmol of H⁺ per mg of polymer required to decrease the pH of 0.2 mg ml⁻¹ polymer solution from 7.4 to 5.0.^32,39

Results and discussion

Synthesis and characterization of the star-PCL–POEGMA copolymers

Well-defined star-shaped amphiphilic copolymers, star-PCL–POEGMA with the same polymer composition but different star architectures of 3, 4 and 6 arms were prepared in a consecutive three-step approach^28–30 including: (i) synthesis of star-PCL–OH by Sn(Oct)₂-catalyzed bulk ROP of ε-CL using three commercially available, branched primary alcohols, 2-(hydroxymethyl)propane-1,3-diol, pentaerythritol, and dipentaerythritol as the initiators,^40–42 (ii) esterification of star-PCL–OH with excess 2-bromoisobutyryl bromide to produce star-PCL–iBuBr macroinitiator,^43,44 and (iii) generation of target star-PCL–POEGMA block copolymers with 3, 4, and 6 arms by ATRP of OEGMA using star-PCL–iBuBr as the macroinitiator and CuBr/bpy as the catalyst.³² The synthesis of star-PCL–POEGMA block copolymer was schematically illustrated in Scheme 1.

Scheme 1 Synthesis of star-PCL–POEGMA block copolymers by integrated ROP and ATRP.

The molecular weights (MWs), polydispersity indexs (PDIs) and degree of polymerizations (DPs) of all the polymers synthesized were determined by ¹H NMR and GPC analysis. The results are summarized in Table 1. Taking 4-arm star-shaped polymer as an example, the typical ¹H NMR spectra of 4s-PCL–OH, 4s-PCL–iBuBr and 4s-PCL–POEGMA are presented in Fig. 1. The DP of PCL was determined to be ∼15 by comparing the integral ratio of peak e′ assigned to the methylene protons adjacent to the terminal hydroxyl group to peak e attributed to the methylene protons adjacent to carbonyl groups (Fig. 1a). Next, the successful conversion of terminal hydroxyl groups to ATRP initiating sites was confirmed by a clear shift of the resonance signal of methylene protons adjacent to the terminal hydroxyl groups in 4s-PCL–OH from 3.64 ppm (Fig. 1a) towards a position at 4.17 ppm in the lower field (Fig. 1b) and the appearance of a new peak f (methyl protons of 2-bromoisobutyryl moieties) at 1.93 ppm (Fig. 1b). In addition, the integral ratio of peak f to the characteristic peaks of PCL repeating units further indicates almost quantitative conversion (Fig. 1b). Finally, the presence of characteristic signals of both PCL and POEGMA blocks in the ¹H NMR spectrum of 4s-PCL–POEGMA (Fig. 1c) indicates successful preparation of target star-shaped block copolymers. The DP of POEGMA block was thus calculated to be ∼12. Successful synthesis of well-defined star-shaped polymers is also confirmed by GPC analysis (Fig. 2). All the star-shaped polymers show unimodal GPC elution peaks with narrow distributions, indicating well-controlled ROP and ATRP processes. The clear shift of GPC elution trace of star-PCL–POEGMAs towards higher MW without tailing at low-MW-side, compared to that of star-PCL–iBuBr macroinitiators, also supports successful chain extension by ATRP of OEGMA.

Table 1 Summary of MW, PDI, and DP of star-PCL–POEGMA block copolymers

	n^a	m^a	Mn^b (kDa)	PDI^b	OEGMA Conv.^a (%)
a Determined by ^1H NMR.b Determined by GPC.
3s-PCLn–POEGMAm	15	11	18.5	1.15	11.03
4s-PCLn–POEGMAm	15	12	21.3	1.20	11.60
6s-PCLn–POEGMAm	14	11	25.0	1.16	11.33

---

Fig. 1 ¹H NMR spectra of (a) 4s-PCL–OH, (b) 4s-PCL–iBuBr, and (c) 4s-PCL–POEGMA in CDCl₃.

Fig. 2 GPC elution traces of 3, 4, and 6-arm star-PCL–OH, star-PCL–iBuBr and star-PCL–POEGMA using THF as the eluent.

Size and morphology of star-shaped micelles

Size is a critical factor for polymeric micelles due to its significant effect on the in vivo performance of micellar drug carriers.⁴⁵ Small size (<100 nm) can guarantee a lower level of nonspecific uptake by RES, minimal renal excretion, and promotion of the EPR effect to achieve passive tumor targeting.⁴⁶ The average size of micelles self-assembled from star-shaped copolymers was determined by DLS at various polymer concentrations in the range of 0.125–1.0 mg ml⁻¹ (Fig. 3). Interestingly, the size of micelles formed by 4s- and 6s-PCL–POEGMA copolymers remains almost constant at all the concentrations evaluated, probably confirming the formation of unimolecular micelles with superior stability irrespective of dilution. On the other hand, micelles self-assembled from 3s-PCL–POEGMA become much smaller in dimension with concentration decreasing from 1.0 to 0.125 mg ml⁻¹, implying formation of aggregates to some extent at high concentrations. It should be noticed that although the average sizes of 4s- and 6s-PCL–POEGMA micelles at 1.0 mg ml⁻¹ are slightly larger than the data recorded at lower concentrations, such tendency is statistical insignificant, therefore we don't think the results is supportive of forming multimolecular micelles at a concentration of 1.0 mg ml⁻¹ for 4s- and 6s-PCL–POEGMA. In addition, we monitored micellar formation of 4s- and 6s-PCL–POEGMA using pyrene as a fluorescence probe. Interestingly, the intensity ratios (I₃₉₃/I₃₇₃) of pyrene in the emission spectra remain almost constant irrespective of polymer concentration (Fig. 4), probably also confirming the formation of unimolecular micelles at the concentrations examined. The overall results demonstrate that 4s- and 6s-PCL–POEGMA are more inclined to form unimolecular micelles than 3s-PCL–POEGMA, we therefore conclude a higher likelihood for star-shaped polymers with more star arms to form unimolecular micelles.

Fig. 3 Average size of micelles self-assembled from (a) 3s-, (b) 4s-, and (c) 6s-PCL–POEGMA block copolymers at various polymer concentrations (*p < 0.001).

Fig. 4 Plots of intensity ratios (I₃₉₃/I₃₇₃) as function of logarithm of 4s- and 6s-PCL–POEGMA concentrations.

The mean hydrodynamic diameter for micelles of 3s-, 4s-, and 6s-PCL–POEGMA is 81.1 ± 2.2, 41.8 ± 3.9 and 51.9 ± 4.4 nm at a polymer concentration of 0.25 mg ml⁻¹, respectively (Fig. 5a–c). The morphology of 3s-, 4s-, and 6s-micelles was visualized by TEM (Fig. 5d–f). Micelles with regularly spherical shape and uniform size are clearly observed for all the three star-shaped micelles. Base on the TEM images, the mean size of 3s-, 4s-, and 6s-micelles is 30.4, 29.8, and 27.9 nm, respectively. The size observed by TEM is smaller than that determined by DLS. Such discrepancy is reasonable given that the latter is the hydrodynamic diameter of micelles in solution, whereas the former reflects the morphology size of micelles in a dry/dehydrated state.⁴⁷ The dimension of micelles is generally determined by the size of both hydrophobic core and hydrophilic shell. On one hand, a comparison of TEM images showed that compared to 6s-PCL–POEGMA micelles, 4s-PCL–POEGMA micelles showed slightly larger size (29.8 nm vs. 27.9 nm) under dehydrated conditions, probably implying a larger hydrophobic core domain with more loading capacity, thus leading to a higher drug loading content (DLC) for 4s-PCL–POEGMA micelles mentioned below. On the other hand, it is reasonable to postulate that the size of hydrophilic POEGMA coronas exerts a more significant effect on the hydrodynamic diameters of micelles in a hydrated medium. Therefore the larger dimension of hydrophilic shell consisting of more POEGMA arms may account for the bigger size of 6s-PCL–POEGMA micelles compared to 4s-PCL–POEGMA analogues in an aqueous phase. Larger hydrodynamic size was observed for 6s-PLGA–PEG micelles compared to 4-PLGA–PEG systems as well.⁴⁸

Fig. 5 TEM images and size distributions of (a and d) 3s-, (b and e) 4s-, and (c and f) 6s-PCL–POEGMA micelles at a polymer concentration of 0.25 mg ml⁻¹.

In vitro drug loading and drug release

In vitro drug loading and drug release study was performed to investigate the effect of star structure on the potential applications of star-PCL–POEGMA as drug carriers. The anti-cancer drug, doxorubicin (DOX) was chosen as the model drug, and encapsulated within the micellar core following the classical dialysis method. The DLC and EE of 3s-, 4s-, and 6s-micelles are 4.70% and 46.3%, 4.92% and 52.9%, 4.51% and 44.3%, respectively. Of the three formulations, 4s-micelles show the highest DLC and EE values.

In vitro DOX release behaviors were evaluated in the physiological condition (PBS, pH 7.4) and in an acidic medium (SSC, pH 6.0) at 37 °C (Fig. 6). These pH values represent the typical extracellular pH and endosomal/lysosomal pH, respectively. Incubation at pH 7.4 results in ∼28%, 16% and 51% DOX release for 3s-, 4s- and 6s-micelles within 72 h. Besides the star structure, the effect of drug loading amount on the drug release rate from micelles should be taken into account. It has been reported that lower drug loading amount resulted in faster drug release.^48,49 Consequently, the drug release profiles of three different star-shaped micelles at pH 7.4 showed accelerated tendency following the order of 6s-\3s-\4s-PCL–POEGMA with decreased DLC of these formulations. The minimal 16% DOX release from 4s-micelles demonstrates the superior stability resulting from the unimolecular structure as well as the hydrophilic corona of POEGMA brushes stabilizing the micelles. Surprisingly, incubation at pH 6.0 results in significantly accelerated drug release with over 60% DOX release within 72 h for both 3s- and 4s-micelles. Such unexpected pH-sensitivity could be used to prompt intracellular drug release for an enhanced therapeutic efficacy. The interesting pH-mediated drug release behaviors of these generally recognized “pH-insensitive” star-shaped micelles have inspired us to figure out the essential reason leading to the pH responsiveness. Although it has been reported that the diffusion of DOX from polymeric micelles becomes easier at low pH due to the increased solubility of DOX in water upon protonation of the glycosidic amine,^50–52 there should be more fundamental reasons accounting for such great difference. To clarify this issue, an acid-triggered degradation study and an acid–base titration were carried out further. The degradation experiment was performed in SSC (pH 6.0) at 37 °C, and MW change of 4s-PCL–POEGMA was monitored by GPC analysis at preselected time intervals similar to the time points of sampling for in vitro drug release study (Fig. S3†). The GPC elution traces of 4s-PCL–POEGMA remain the same irrespective of various incubation time, indicating the nonoccurrence of PCL degradation within such a short duration of 96 h, which was also confirmed by the ¹H NMR characterization of products showing the same spectra (Fig. S4). Note that 4s-PCL–POEGMA of different batches with slightly different chain lengths of POEGMA block were used for GPC analysis in Fig. 2 and S3,† which is believed to account for the slight inconsistency of the GPC elution traces recorded for the polymer.

Fig. 6 In vitro drug release of various DOX-loaded star micelles.

Next, the buffering capacity (BC) of three star-shaped block copolymers was assessed by acid–base titration since BC reflects the ability for endosomal escape, a crucial factor to achieve high delivery efficacy. To make a comparison, RAFT-synthesized POEGMA₁₇ ^31,39 was included as a control. The pH of endosomal/lysosome are in the range of 5.0–6.5, and the buffering capacity of polymer is defined as the μmol H⁺ per mg polymer required to decrease the pH value of a 0.2 mg ml⁻¹ polymer solution from 7.4 to 5.0.^32,39 Interestingly, all the star-PCL–POEGMAs show some BC albeit not that high compared to that of typical polycations^32,39 (Fig. 7). The BC of 3s-, 4s-, and 6s-PCL–POEGMA and POEGMA₁₇ is 2.17, 1.79, 1.68 and 2.20 μmol mg⁻¹, respectively. The results show the BC of star-PCL–POEGMA is affected by hydrophilic OEG brush corona as well as the star structures, but with primary contribution from the former. We therefore conclude that it is the OEG brush corona that endows the star-PCL–POEGMA with pH-sensitive drug release behaviors.

Fig. 7 Buffering capacity of star-PCL–POEGMA copolymers obtained by titrating polymer aqueous solution (0.2 mg ml⁻¹) in 0.15 M NaCl aqueous solution (pH 11, adjusted with 0.1 M NaOH) with 0.1 M HCl. As a reference, the titration curve of RAFT-synthesized POEGMA₁₇ is also included.

Taken together, 4s-PCL–POEGMA is the optimal formulation in terms of its micellar size, DLC, stability and unique pH-sensitive drug release behaviors.

Conclusion

Three star-shaped amphiphilic block copolymers, star-PCL–POEGMA with different star structures were synthesized successfully by integrated ROP and ATRP using branched alcohols as the starting core to investigate the effect of star architecture on their properties as well as potential biomedical applications as drug carriers. DLS measurements, TEM observation, and pyrene fluorescence probe technique confirmed formation of unimolecular spherical micelles with average hydrodynamic diameters less than 50 nm by 4s- and 6s-PCL–POEGMA copolymers. In vitro drug loading and drug release study showed that the DOX-loaded 4s-micelles possessed the highest DLC and EE of all the three formulations. Surprisingly, these “pH-insensitive” 4s-micelles showed unique pH-mediated drug release, i.e., dramatically accelerated release at pH 6.0 while much slower profile at pH 7.4. Such interesting pH-responsiveness was attributed primarily to the hydrophilic OEG brush corona revealed by an acid–base titration study. This study therefore provides new insights into the structure–bioproperties correlation of star-shaped polymers, and should be useful for the future design and development of novel star-shaped polymers with better performance for controlled drug delivery.

Acknowledgements

The authors acknowledge the financial support from the National Natural Science Foundation of China (51473072 and 21504035) and the Fundamental Research Funds for the Central Universities (lzujbky-2015-k05 and lzujbky-2016-ct05).

References

1. V. P. Torchilin, J. Controlled Release, 2001, 73, 137–172.
2. K. Kataoka, A. Harada and Y. Nagasaki, Adv. Drug Delivery Rev., 2001, 47, 113–131.
3. A. Halperin, Macromolecules, 1991, 24, 1418–1419.
4. Y. H. Ma, T. Cao and S. E. Webber, Macromolecules, 1998, 31, 1773–1778.
5. L. F. Zhang and A. Eisenberg, J. Am. Chem. Soc., 1996, 118, 3168–3181.
6. S. Pispas, Y. Poulos and N. Hadjichristidis, Macromolecules, 1998, 31, 4177–4181.
7. G. Gaucher, M. H. Dufresne, V. P. Sant, N. Kang, D. Maysinger and J. C. Leroux, J. Controlled Release, 2005, 109, 169–188.
8. W. Wu, W. G. Wang and J. S. Li, Prog. Polym. Sci., 2015, 46, 55–85.
9. F. Wang, T. K. Bronich, A. V. Kabanov, R. D. Rauh and J. Roovers, Bioconjugate Chem., 2005, 16, 397–405.
10. H. Wei, C. Cheng, C. Chang, W. Q. Chen, S. X. Cheng, X. Z. Zhang and R. X. Zhuo, Langmuir, 2008, 24, 4564–4570.
11. S. Chen, X. Z. Zhang, S. X. Cheng, R. X. Zhuo and Z. W. Gu, Biomacromolecules, 2008, 9, 2578–2585.
12. H. Wei, C. E. Wang, N. Tan, A. J. Boydston and S. H. Pun, ACS Macro Lett., 2015, 4, 938–941.
13. S. Aryal, M. Prabaharan, S. Pilla and S. Q. Gong, Int. J. Biol. Macromol., 2009, 44, 346–352.
14. G. Kreutzer, C. Ternat, T. Q. Nguyen, C. J. G. Plummer, J. A. E. Månson, V. Castelletto, I. W. Hamley, F. Sun, S. S. Sheiko, A. Herrmann, L. Ouali, H. Sommer, W. Fieber, M. I. Velazco and H. A. Klok, Macromolecules, 2006, 39, 4507–4516.
15. I. J. Majoros, A. Myc, T. Thomas, C. B. Mehta and J. R. Baker, Biomacromolecules, 2006, 7, 572–579.
16. J. H. Zou, Y. B. Zhao and W. F. Shi, J. Phys. Chem. B, 2006, 110, 2638–2642.
17. J. H. Zou, W. F. Shi, J. Wang and J. Bo, Macromol. Biosci., 2005, 5, 662–668.
18. H. F. Gao, Macromol. Rapid Commun., 2012, 33, 722–734.
19. G. Lapienis, Prog. Polym. Sci., 2009, 34, 852–892.
20. J. M. Ren, T. G. McKenzie, Q. Fu, E. H. H. Wong, J. T. Xu, Z. S. An, S. Shanmugam, T. P. Davis, C. Boyer and G. G. Qiao, Chem. Rev., 2016, 116, 6743–6836.
21. C. L. Shi, X. Guo, Q. Q. Qu, Z. M. Tang, Y. Wang and S. B. Zhou, Biomaterials, 2014, 35, 8711–8722.
22. M. Byrne, R. Murphy, A. Kapetanakis, J. Ramsey and S. A. Cryan, Macromol. Rapid Commun., 2015, 36, 1862–1876.
23. S. Corneillie and M. Smet, Polym. Chem., 2015, 6, 850–867.
24. J. Park, S. Jang and J. K. Kim, J. Polym. Sci., Part B: Polym. Phys., 2015, 53, 1–21.
25. B. V. K. J. Schmidt, M. Hetzer, H. Ritter and C. Barner-Kowollik, Prog. Polym. Sci., 2014, 39, 235–249.
26. P. G. Kadam and S. Mhaske, Des. Monomers Polym., 2011, 14, 515–540.
27. M. E. Fox, F. C. Szoka and J. M. J. Frechet, Acc. Chem. Res., 2009, 42, 1141–1151.
28. O. G. Schramm, G. M. Pavlov, H. P. van Erp, M. A. R. Meier, R. Hoogenboom and U. S. Schubert, Macromolecules, 2009, 42, 1808–1816.
29. K. Knop, G. M. Pavlov, T. Rudolph, K. Martin, D. Pretzel, B. O. Jahn, D. H. Scharf, A. A. Brakhage, V. Makarov, U. Möllmann, F. H. Schacherab and U. S. Schubert, Soft Matter, 2013, 9, 715–726.
30. K. Knop, D. Pretzel, A. Urbanek, T. Rudolph, D. H. Scharf, A. Schallon, M. Wagner, S. Schubert, M. Kiehntopf, A. A. Brakhage, F. H. Schacher and U. S. Schubert, Biomacromolecules, 2013, 14, 2536–2548.
31. H. Wei, J. G. Schellinger, D. S. H. Chu and S. H. Pun, J. Am. Chem. Soc., 2012, 134, 16554–16557.
32. H. Wei, L. R. Volpatti, D. L. Sellers, D. O. Maris, I. W. Andrews, A. S. Hemphill, L. W. Chan, D. S. H. Chu, P. J. Horner and S. H. Pun, Angew. Chem., Int. Ed., 2013, 52, 5377–5381.
33. C. E. Wang, H. Wei, N. Tan, A. J. Boydston and S. H. Pun, Biomacromolecules, 2016, 17, 69–75.
34. C. Zuo, X. Y. Dai, S. J. Zhao, X. N. Liu, S. L. Ding, L. W. Ma, M. Z. Liu and H. Wei, ACS Macro Lett., 2016, 5, 873–878.
35. X. J. Huang, Y. Xiao and M. D. Lang, Macromol. Res., 2011, 19, 113–121.
36. Y. Q. Yang, W. J. Lin, L. J. Zhang, C. Z. Cai, W. Jiang, X. D. Guo and Y. Qian, Macromol. Res., 2013, 21, 1011–1020.
37. Y. Q. Yang, B. Zhao, Z. D. Li, W. J. Lin, C. Y. Zhang, X. D. Guo, J. F. Wang and L. J. Zhang, Acta Biomater., 2013, 9, 7679–7690.
38. W. J. Lin, S. Y. Nie, Q. Zhong, Y. Q. Yang, C. Z. Cai, J. F. Wang and L. J. Zhang, J. Mater. Chem. B, 2014, 2, 4008–4020.
39. H. Wei, J. A. Pahang and S. H. Pun, Biomacromolecules, 2013, 14, 275–284.
40. C. M. Dong, K. Y. Qiu, Z. W. Gu and X. D. Feng, Macromolecules, 2001, 34, 4691–4696.
41. G. D. García-Olaiz, K. A. Montoya-Villegas, A. Licea-Claverie and N. A. Cortez-Lemus, React. Funct. Polym., 2015, 88, 16–23.
42. F. L. Meng, Z. G. Xu and S. X. Zheng, Macromolecules, 2008, 41, 1411–1420.
43. K. H. Kim, G. H. Cui, H. J. Lim, J. Huh, C. H. Ahn and W. H. Jo, Macromol. Chem. and Phys., 2004, 205, 1684–1692.
44. X. J. Li, Y. F. Qian, T. Liu, X. L. Hu, G. Y. Zhang, Y. Z. You and S. Y. Liu, Biomaterials, 2011, 32, 6595–6605.
45. S. D. Perrault, C. Walkey, T. Jennings, H. C. Fischer and W. C. W. Chan, Nano Lett., 2009, 9, 1909–1915.
46. C. Chang, H. Wei, C. Y. Quan, Y. Y. Li, J. Liu, Z. C. Wang, S. X. Cheng, X. Z. Zhang and R. X. Zhuo, J. Polym. Sci., Part A: Polym. Chem., 2008, 46, 3048–3057.
47. C. N. Zhang, W. Wang, T. Liu, Y. K. Wu, H. Guo, P. Wang, Q. Tian, Y. M. Wang and Z. Yuan, Biomaterials, 2012, 33, 2187–2196.
48. G. L. Ma, C. Zhang, L. H. Zhang, H. F. Sun, C. X. Song, C. Wang and D. L. Kong, J. Mater. Sci.: Mater. Med., 2016, 27, 17.
49. H. S. Yoo and T. G. Park, J. Controlled Release, 2011, 70, 63–70.
50. E. R. Gillies and J. M. J. Fréchet, Bioconjugate Chem., 2005, 16, 361–368.
51. C. Chittasupho, S. X. Xie, A. Baoum, T. Yakovleva, T. J. Siahaan and C. J. Berkland, Eur. J. Pharm. Sci., 2009, 37, 141–150.
52. M. Prabaharan, J. J. Grailer, S. Pilla, D. A. Steeber and S. Q. Gong, Biomaterials, 2009, 30, 3009–3019.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed amount of each reagent used for the synthesis of 3s- and 6s-PCL–POEGMA. ¹H NMR spectra of 3s-, and 6s-polymer synthesized, GPC and ¹H NMR data of 4s-PCL–POEGMA in an acid-triggered degradation study. Tables S1–S3, and Fig. S1–S4. See DOI: 10.1039/c6ra21408h

---