Reverse poly(butylene oxide)–poly(ethylene oxide)–poly(butylene oxide) block copolymers with lengthy hydrophilic blocks as efficient single and dual drug-loaded nanocarriers with synergistic toxic effects on cancer cells

Abstract

Combination therapy appears as a very interesting alternative for solving some of the problems associated with single-drug therapies such as drug resistance and improvement of patients' survival. However, it also possesses a series of potential drawbacks, mainly associated with the complicated administration of several antineoplastic drugs, which use different excipients owing to compatibility and stability issues. Hence, the combination of two or more drugs in one polymeric micelle with sufficient loading capacity could solve this constraints as well as provide a suitable control of the release rate and protection for cargos. In this study, four different reverse poly(butylene oxide)–poly(ethylene oxide)–poly(butylene oxide) block copolymers, BO_nEO_mBO_n, with BO blocks ranging from 8 to 21 units and EO units from 90 to 411 were tested as potential single and dual nanocarriers of two antineoplastic drugs, namely, docetaxel and doxorubicin currently used in combination for the treatment of advanced/metastatic breast cancer. Polymeric micelles formed by these copolymers were shown to solubilise important amounts of these drugs, either alone or combined, in a single micelle with a good stability under serum-mimicking conditions. These polymeric nanocarriers were able to release drugs in a sustained manner; the release rate became slower as the hydrophobicity of copolymer chain increased. Drugs loaded in the polymeric micelles accumulated more slowly inside the cells than free DOXO due to their sustained release. Copolymers were found to be biocompatible over most of the concentration range tested. The in vitro cell cytotoxicity was found to be larger for dual DCX/DOXO-loaded micelles than for single-loaded ones, and free administered drugs in both cervical HeLa and breast MDA-MB-231 cancer cell lines were tested by exerting a synergistic effect. Therefore, poly(butylene oxide)–poly(ethylene oxide) block copolymers offer important features as efficient nanocarriers for dual combination therapy, which on combining with the ability of some of the present copolymer varieties inhibit efflux pump mechanisms and allow the development of interesting nanoformulations with unparalleled therapeutic efficacies in cancer treatment.

---

1. Introduction

The properties of amphiphilic copolymers combining hydrophilic poly(ethylene oxide) units with different types of hydrophobic blocks have been found to show suitable characteristics, which fulfill the requirements for a relatively efficient therapeutic action of poorly-aqueous soluble drugs by enhancing their solubilization, allowing their sustained release, improving their pharmacokinetics and facilitating their access to the site of action while providing “stealthiness” to evade scavenging by the mononuclear phagocyte system.^1–3 These beneficial properties can originate from the spontaneous self-assembly of copolymer chains into nanoscopic core–shell micellar structures in a solution.⁴ The micellar cores offer an excellent platform for the solubilization of the poorly water-soluble therapeutic agents;^5,6 this is particularly interesting with respect to hydrophobic anticancer agents in preclinical development, e.g. 17-allylamino-17-demethoxygeldanamycin and 17-AAG; and in clinical practice, e.g. placlitaxel, docetaxel, doxorubicin and etoposide, which require safe vehicles for solubilization and intravenous infusion since current vehicles for intravenous drug infusion are often toxic, e.g. Cremophor EL. Moreover, the polymeric micellar carrier also offers protection for cargos, which is provided by the hydrophilic shell that minimizes nonspecific uptake through the reticuloendothelial system (RES), thereby enhancing drug circulation times and passive accumulation in solid tumors.⁷

Single drug therapy for cancer is rarely successful due to the inherent and developing drug resistance to tumors, the heterogeneity of cancer cells, and the existence of leakages and/or burst release phases from the nanocarriers, which do not allow the drug to reach sustained effective therapeutic concentrations for the payloads. Different attempts have been made to solve these issues, for example, the development of core–shell nanostructures (polymeric fibers and micelles), which enable a biphasic sustained release of the cargo molecules to ensure optimal therapeutic concentrations.^8,9 Another alternative is the combined use of different drugs, the so-called combination chemotherapy, which has become a standard regimen to treat cancer patients.¹⁰ Such therapy regimens commonly involve the sequential administration of multiple drugs that can act along different synergistic pathways and kill cancer cells better while retarding the occurrence of resistant cell lines within acceptable toxicity.¹¹ Despite the advantages of combination chemotherapy, one of the main challenges associated with its clinical application is the complicated administration of several antineoplastic drugs, which use different excipients owing to compatibility and stability issues. For example, the solubilization of one drug inside a polymeric micelle embedded into the core of a electrospun fiber, which at the same time, serves as an additional depot for an additional drug has been shown to be an effective way to achieve a codelivery and multistep sustained release of several drugs.¹² Another option can be simply combining two or more drugs into one single polymeric micelle with sufficient loading capacity, which could simplify the fabrication processes and treatments, making the latter less hazardous to patients. A mediocre loading might be an impediment for even single drug micelle formulations since administration would require prohibitively high doses of the polymer, which could result in vehicle-derived toxicity as observed, for example, in the case of Taxol and Taxotere, clinical formulations of placitaxel and docetaxel, which contain Cremophor EL and Polysorbate 80, respectively, as excipients can induce hypersensitivity reactions and toxicity during intravenous infusions.¹³

Polymeric micelles, which combine poly(oxyethylene) and poly(oxypropylene) blocks [EO = oxyethylene, OCH₂CH₂ and PO = oxypropylene, OCH₂CH(CH₃)] in a triblock structure, either direct, EO_mPO_nEO_m, or reverse, PO_nEO_mPO_n (where the subscripts m and n denote number-average block lengths), have been the most extensively studied for antineoplastic drug administration due to their commercial availability in a very broad range of compositions, a sustained release pattern, good biocompatibility of most varieties, and approval of some varieties by regulatory agencies to be used in pharmaceutical formulations.¹⁴ Nevertheless, EO_mPO_nEO_m, or PO_nEO_mPO_n copolymers possess several drawbacks, for example, very limited solubilization capacity, low stability upon dilution in the bloodstream, and changes in the micellization behavior from batch to batch due to polydispersity as a consequence of the transfer reaction from hydrogen abstraction during the polymerization of PO blocks.¹⁵ Hence, during the last few years, different block copolymer counterparts with similar architecture but with the PO segment replaced by a more hydrophobic one have been proposed with the aim of improving drug solubilization capacities and release profiles.^16–21 Special attention has been paid to copolymers with 1,2-butylene oxide (BO) as the hydrophobic monomer due to its structural similarity to PO; moreover, the transfer is not a problem in the laboratory polymerization of BO.²² This monomer (as does PO) adds to the growing chain to give a secondary oxyanion, and the slow initiation of EO chains at the secondary termination might lead to a broadened EO-block length distribution.²³ To avoid such deleterious effects, BO blocks can be polymerized last, forming EO_mBO_m diblock and BO_nEO_mBO_n triblock copolymers. In addition, the larger relative hydrophobicity of BO blocks compared to PO (six-fold as estimated from the ratio of the logarithms of the molar critical micellar concentrations, cmcs)¹⁶ allows the formation of polymeric micelles, transient micelle clusters and polymer networks by bridging of extended chains between micelles^24,25 at much lower concentrations than PO_nEO_mPO_n does. Provided that these copolymers have been proved to be biocompatible in some preliminary experiments,²⁶ their use as anticancer nanocarriers would allow the solubilization of higher concentrations of poorly water soluble drugs at much lower copolymer concentrations²⁷ in the form of injectable solutions, oral suspensions and/or (sub)dermal gelling depots^27,28 while exerting a complementary role as “cell response modifiers”, for example, by inhibiting the P-glycoprotein efflux pump in multidrug-resistant (MDR) tumoral cells. We have recently shown that BO_nEO_mBO_n copolymers can efficiently incorporate doxorubicin (DOXO), exerting an enhanced and sustained therapeutic effect against multidrug resistant ovarian cancer cells.²⁷

This paper addresses two questions: firstly, whether different water-insoluble antineoplastic drugs can be incorporated in such polymeric micelles with suitable loading capacities, and secondly, whether multiple drugs can be simultaneously incorporated inside micellar cores to form pharmacologically synergistic chemotherapeutic combinations of high enough potency to kill cancer cells. Therefore, four different reverse BO_nEO_mBO_n triblock copolymers (BO₈EO₉₀BO₈, BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁) were tested as effective nanocarriers for single/dual drug delivery of hydrophobic antineoplastic compounds such as docetaxel (DCX) and DOXO, which are used nowadays in combination chemotherapy for the treatment of advanced/metastatic breast cancer through parenteral administration. The results suggest that BO_nEO_mBO_n micelles can provide an attractive, biocompatible platform for co-solubilization and delivery of multiple hydrophobic molecules with minimal vehicle-associated side effects while avoiding the limitations of organic solvents and/or surfactants in current chemotherapeutic formulations to enhance the solubilization of this class of drugs in terms of side effects.

2. Materials and methods

2.1. Materials

Four BO_nEO_mBO_n copolymers with narrow chain length distributions (BO₈EO₉₀BO₈, BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁) were prepared and characterized as previously described.²⁹ The critical micelle concentrations (cmcs) in an aqueous solution were estimated from pyrene fluorescence measurements, as previously reported.³⁰ Table 1 summarizes the molecular characteristics of the copolymers.

Table 1 Molecular weight and critical micelle concentration (cmc) of the copolymers

Copolymers	Mn^a (g mol^−1)	Mw/Mn^b	Mw (g mol^−1)	CMC^c (mg mL^−1)	N	rh (nm)
a Estimated by NMR.b Estimated by GPC; Mw calculated from Mn and Mw/Mn. Estimated uncertainty: Mn to ±3%; Mw/Mn to ±0.01.c Values from ref. 37.d Values from ref. 61.e Values from ref. 62.
BO8EO90BO8	5100	1.07	5460	0.33	38^d	13.0^d
BO14EO378BO14	18 600	1.12	20 832	0.058	18^e	18.5^e
BO20EO411BO20	21 000	1.08	22 680	0.012	17^d	18.9^d
BO21EO385BO21	20 000	1.10	22 000	0.025	9^e	20.4^e

---

Docetaxel and doxorubicin hydrochloride (DOXO·HCl) were acquired from Sigma-Aldrich. DOXO base was obtained by means of the aqueous precipitation of DOXO·HCl aqueous solution (1 mg mL⁻¹) by adding triethylamine and methylene chloride. The system was kept under vigorous stirring for 1 h and then the organic phase was evaporated in order to recover the DOXO base.³¹ Herein, DOXO refers to DOXO base. Water was double distilled and degassed before use. All other reagents were of analytical grade.

2.2. Drug solubilisation

Solubilization of DCX and DOXO (intrinsic solubilities in water were ca. 4.9 and 0.5 mg dm⁻³, respectively)³² in micellar copolymer solutions was tested in triplicate following the procedure of Elsabahy et al. with minor modifications.³³ Briefly, the desired amount of drug(s) (typically 40 μg in total) dissolved in dichloromethane (100 μM) was added to the weighed solid copolymer (typically 2 mg). The organic solution was stirred and the solvent evaporated to dryness. Then, distilled water was added dropwise to the dried sample and left stirring overnight. Solutions were centrifuged at 3000 rpm for 30 min and the supernatants were filtered (Millipore Millex filters, 0.45 μm pore size) to remove the non-solubilised drug. The filtered solutions were diluted (1/1000) with methanol to disrupt the self-assembled structures. The amounts of solubilised DCX and DOXO were determined using a UV-Vis (Cary 50 UV-Vis spectrophotometer, Agilent, Germany) at 227 and 480 nm, respectively, using solutions of each copolymer under the same dilution conditions as blanks, by fluorescence spectroscopy by excitation at 480 nm (only for DOXO) and/or by reverse phase high performance liquid chromatography (HPLC). For HPLC, an Agilent Technologies 1200 series HPLC system equipped with a Nucleosil C18 5 μm column (250 mm × 4.6 mm) was used. Samples were diluted using the mobile phase (specified below) and injected (20 μL) into the HPLC system. For single DCX, DOXO and dual DCX/DOXO solutions, a mixture of acetonitrile/water (55/45 v/v) was used as the mobile phase. The flow rate was 1.0 mL min⁻¹, and the column temperature was set to 30 °C. Detection wavelengths were 227 and 480 nm and retention times were 8.1 and 7.0 min for DCX and DOXO, respectively.

Drug loading, D.L., entrapment efficiency, E.E., and the solubilisation capacity per gram of copolymer in solution, S_CP (namely, the amount of drug dissolved at 37 °C in 100 mL of copolymer solution in excess of that dissolved in an equivalent volume of water) were calculated as follows:

2.3. Micellar sizes and physical stability of the drug-loaded micelles upon dilution

Sizes of unloaded and drug-loaded polymeric micelles were determined by dynamic light scattering (DLS) using an ALV-5000F (ALV-GmbH, Germany) instrument with vertically polarized incident light (λ = 488 nm) supplied by a diode-pumped Nd:YAG solid-state laser (Coherent Inc., CA, USA) operated at 2 W and combined with an ALV SP-86 digital correlator (sampling time of 25 ns to 100 ms; scattering angle θ = 90°). Experiment duration was in the range 5–10 min and each measurement was repeated at least twice. The correlation functions from DLS runs were analyzed by the CONTIN method to obtain the intensity distributions of decay rates (Γ), the apparent diffusion coefficients, and the apparent hydrodynamic radius (r_h,app) applying the Stokes–Einstein equation.¹⁶ Sizes and morphologies of drug-loaded polymeric micelles were also measured by transmission electron microscopy (TEM). Micellar drug-loaded polymer solutions were applied over carbon-coated copper grids, blotted, washed, negatively stained with 2 wt% phosphotungstic acid, air-dried and then examined with a Phillips CM-12 transmission electron microscope operating at an accelerating voltage of 120 kV.

The physical stability of the drug-loaded micelles was assessed by dilution of the samples (1/50) in PBS buffer (pH 7.4), supplemented with 10% fetal bovine serum (FBS) at 37 °C under moderate stirring, and the drug concentration was monitored over time by UV spectrophotometry, as described above. The experiments were performed in triplicate. Simultaneously, aliquots were taken, filtered (Triton free Millipore Millex, 0.22 μm porosity) into scattering cells and allowed to equilibrate at 37 °C for 30 min before recording changes in the size of drug-loaded micelles by DLS as described above.

2.4. In vitro drug release

To investigate the release profiles, the required amount of DCX-loaded micelles (4 mL) were placed into dialysis tubes (SpectraPore®, MWCO 3500), into which 100 mL PBS buffer supplemented with 10% FBS at pH 7.4 and 2% (v/v) ethanol was introduced to enhance the solubility of released free DCX and avoid its aggregation.³⁴ The whole assembly was kept at 37 °C under stirring and covered by Parafilm to avoid evaporation. At each sampling time, 1 mL of medium was withdrawn and replaced with the same volume of fresh buffer (containing 2% (v/v) ethanol) to maintain the required sink conditions. Quantification was done by the calibration curve of DCX at 293 nm after dilution with methanol. Assays were carried out in triplicate.

Drug release profiles from the micellar systems were fitted to the square-root kinetics³⁵

M_t/M_∞ = kt^0.5 (4)

and to the Fickian diffusion model, considering the micelles as perfect spheres,³⁶

M_t/M_∞ = k₁ + k₂t^0.5 − k₃t (5)

where M_t and M_∞ represent the drug amount released at time t and that was initially contained in the formulation, respectively, and k, k₁, k₂ and k₃ are the release rate coefficients.

2.5. Cellular uptake by fluorescence microscopy

Polymeric micelle uptake was followed by fluorescence microscopy by seeding HeLa cells on poly-L-lysine coated glass coverslips (12 × 12 mm²) placed inside 6-well plates (3 mL, 5 × 10⁴ cells per well) and grown for 24 h under standard culture conditions (5% CO₂ at 37 °C in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% (v/v) FBS, 2 mM L-glutamine, 1% penicillin/streptomycin, 1 mM sodium pyruvate, and 0.1 mM MEM non-essential amino acids (NEAA)). Then, 50 μL of either DOXO loaded-polymeric micelle dispersions or free DOXO (0.005 μM DOXO) were added to cells. After 1 h and 24 h of incubation, cells were washed three times with PBS pH 7.4 and then fixed with paraformaldehyde 4% (w/v) for 10 min, washed with PBS and permeabilized with 0.2% (w/v) (Triton X-100). The cells were washed again with PBS, mounted on glass slides stained with ProLong® Gold antifade DAPI (Invitrogen) and cured for 24 h at −20 °C. Samples were visualized with 20× and 63× objectives using an inverted fluorescence microscope, Leica DMI6000B (Leica Microsystems GmbH, Heidelberg Mannheim, Germany), whereby the blue channel corresponds to DAPI (λ_ex 355 nm) and the red channel to DOXO (λ_ex 475 nm).

2.6. In vitro copolymer cytocompatibility evaluation and cytotoxicity of drug-loaded polymeric micelles

MDA-MB-231 adenocarcinoma breast and HeLa cervical cancer cells from Cell Biolabs (San Diego, CA, USA) were used for in vitro studies. Cells were grown under standard culture conditions (5% CO₂ at 37 °C) in Dulbecco's modified Eagle's medium (DMEM), supplemented with 10% Fetal Bovine Serum (FBS), 2 mM L-glutamine, 1% penicillin/streptomycin, 1 mM sodium pyruvate, and 0.1 mM MEM Non-Essential Amino Acids (NEAA).

Breast MDA-MB-231 and cervical HeLa cancer cells with an optical confluence of 80–90% were seeded into 96-well plates (100 μL, 1.5 × 10⁴ cells per well) and grown for 24 h under standard culture conditions in 100 μL growth medium. After 24 h of incubation at 37 °C under 5% CO₂, free DCX and DOXO solutions and bare, single and dual-loaded polymeric micelles at different cargo loadings were added to the cultured cells for the cytocompatibility evaluation of bare copolymers and the cytotoxicity efficiency determination of the polymeric nanocarriers, and the cells were subsequently incubated from 24 to 72 h. Cells exposed to copolymer-free culture medium were used as a negative control (100% viability) in both types of experiments. Cytotoxicity was evaluated at different time points using the CCK-8 proliferation assay. After incubation, 10 μL of CCK-8 reagent was added to each well and after 2 h, the absorption at 450 nm of cell samples was measured with an UV-Vis microplate absorbance reader (BioRad model 689, USA). Cell viability was calculated as:

% viability = (Abs_sample/Abs_control) × 100 (6)

where Abs_sample is the absorbance at 450 nm for cell culture samples with either bare copolymer solutions or drug formulations (free drugs and drug-loaded micelles) and Abs_control is the absorbance for PBS controls. Assays were carried out in triplicate.

3. Results and discussion

Marketed intravenous formulations of chemotherapeutic drugs involve the use of organic solvents and/or surfactants to enhance their solubilization for injection in order to achieve the required therapeutic doses. This is the case, for example, with placlitaxel (PCX, Taxol®) and DCX (Taxotere®), which utilize Cremophor El and ethanol or polysorbate 80, respectively, for solubilization⁷ and avoidance of drug degradation in protic solvents.³⁷ In particular, DCX is a taxane compound that displays a broad spectrum of antitumor activity, interferes with microtubule formation during cell division,³⁸ and is currently approved for the treatment of breast, non-small-cell lung, prostate, stomach, head and neck cancers.^37,39 DCX is also used in combination with DOXO (supplied in saline formulation, Andryamicin®), an anthracycline antibiotic, which intercalates into nuclear DNA and interacts with topoisomerase II to cause DNA cleavage and cytotoxicity⁴⁰ in the treatment of advanced or metastatic breast cancer,^41,42 given their different mechanisms of action and partially non-overlapping toxicity profiles. However, there still exist severe side effects associated with the administration of these two drugs. Earlier studies have shown that the polysorbate 80 formulation of DCX causes severe allergic reactions and peripheral neuropathy in up to 40% of patients;^43,44 after dilution with the hydroalcoholic vehicle provided, the Taxotere® formulation is physically unstable and must be administered to the patient within 8 h. On the other hand, DOXO binding to cell membranes ultimately results in the production of active oxygen species that attack myocytes, which is the main cause of severe DOXO cardiotoxicity.⁴⁵ Therefore, co-encapsulation of these two drugs in BO_nEO_mBO_n polymeric micelles is expected to improve the efficiency and safety of the combinatorial treatment by acting on multiple pathways,⁴⁶ and even to provide some types of synergistic therapeutic effects while reducing acute toxicity and side effects.^20,21,47

Four BO_nEO_mBO_n copolymers (BO₈EO₉₀BO₈, BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁) that cover a wide range of molecular weights and cmc values were chosen. These copolymers form spherical micelles ranging from 17 to 35 nm in diameter and have association numbers between 20 and 43 (see Table 1).^48,49 At higher polymer concentrations, viscous (from 1 to 6 wt%) and immobile gels (>5 wt%, depending on copolymer type and solution temperature) appear as a consequence of the cross-linking, originating from the residence of BO blocks in one polymer chain in two adjacent micelles promoting a progressively denser open network structure.

3.1 Cytocompatibility of BO_nEO_mBO_n copolymers

To take into account potential differences in toxicity due to different cell phenotypes and sensitivity, we performed a comparative study to quantify how bare BO_nEO_mBO_n polymeric micelles at different concentrations and incubation times (from 24 to 72 h) affect the in vitro viabilities of two tumoral cells lines of cervical (HeLa) and metastatic breast (MDA-MB-231) cancer by means of a cell-counting kit-8 cell proliferation assay (CCK-8). This test is based on the bioreduction of 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt (WST-8), which produces a water-soluble formazan dye in the presence of an electron carrier, 1-methoxy phenazinemethosulfate (PMS).

For HeLa cells, cell viabilities (CVs) between 90% and 100% were found for the copolymer, BO₈EO₉₀BO₈, over the whole concentration range. For the remaining copolymers, concentrations below 2 mg mL⁻¹ led to viability extents above ca. 80% after 24 h and 72 h of incubation, except for BO₂₁EO₃₈₅BO₂₁, which produced slightly lower values (Fig. 1a). A certain cell viability loss was also noted as the copolymer concentration increased but in general, CVs were >60%, larger than the limit of 50% considered as a reference value for a nanomaterial to be cytocompatible;⁵⁰ there was an exception for the copolymer, BO₂₁EO₃₈₅BO₂₁, at the highest concentrations tested (4 mg mL⁻¹), for which CVs of ca. 40–42% denoted cell toxicity after 72 h of incubation (Fig. 1a). Moreover, a certain decrease in CVs was found as the hydrophobic block length of the copolymers increased; this was especially noted for the copolymer, BO₂₁EO₃₈₅BO₂₁, which displayed lower CVs. This behavior could stem from the greater affinity of the amphiphile for cellular membrane structures, increasing the cell permeability.

Fig. 1 Cell viabilities of BO_nEO_mBO_n polymeric micelles in HeLa cells after (a) 24 and (b) 72 h of incubation; and in MDA-MB-231 cells after (c) 24 and (d) 72 h of incubation.

Conversely, MDA-MB-231 cells were more sensitive to the presence of polymeric micelles exhibiting lower CVs, particularly after extensive incubation (72 h). A decrease in CV after incubation with MDA-MB-231 cells was also noted. For example, cells exposed to 4 mg mL⁻¹ BO₈EO₉₀BO₈ micelles underwent a sharp viability loss from ca. 80% at 24 h of incubation to ca. 25% after 72 h. In contrast, under similar conditions, the same copolymer produced viability extents of ca. 100% after 72 h of incubation in HeLa cells. BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₇₈BO₂₁ copolymers were toxic to MDA-MB-231 cells at the highest concentrations tested (>3 mg mL⁻¹) after 72 h of incubation, the CVs being below 45% and 24% at 3 and 4 mg mL⁻¹, respectively (Fig. 1b). At lower concentrations, copolymer biocompatibility was observed, with CVs above 75% and 50% after 24 and 72 of incubation, respectively. Nevertheless, it is necessary to bear in mind that the cytotoxicity of BO_nEO_mBO_n copolymers may be overestimated in vitro as cells are not protected by the anatomical barriers present in vivo.⁵¹ The present findings stressed the relevance of doing cytotoxicity tests in different cell lines in order to consider possible influences of cell phenotype on the cytocompatibility of a given material.

3.2. Solubilization capacity

Different studies reported that the major factor influencing the solubilization capacity of polymeric micelles is the compatibility between the drug and the core-forming block.⁵² As mentioned previously, BO blocks are six times more hydrophobic than PPO units present in commercial Pluronic and Tetronic block copolymers (on the basis of their molar cmcs) so a better compatibility between BO blocks and the present antineoplastic hydrophobic drugs might be expected in agreement with previous observations made for much shorter diblock EO_mBO_n and triblock EO_mBO_nEO_m copolymers.^48,49

Single (DCX) and dual (DCX/DOXO) encapsulation experiments were carried out to evaluate the impact of the feeding drug amount on the entrapment efficiency and the total loaded drug amount inside micelles by means of UV-Vis spectrophotometry and HPLC. Preliminary attempts at dissolving DCX and DOXO into preformed micelles resulted in low drug loading, probably because of the slow drug diffusion into the viscous micellar core (data not shown). Accordingly, it was decided to codissolve the drug(s) and polymer in a pharmaceutically acceptable organic solvent (i.e. dichloromethane) to form a homogeneous polymeric matrix, followed by the evaporation of the organic solvent and subsequent addition of the water phase. Hence, different amounts of drug(s) to solid copolymer (final polymer concentration usually 2 mg mL⁻¹, more than 10 times higher than the respective cmc ensuring complete micellization) were mixed to achieve different initial feeding ratios (see Table 2). For DCX-loaded and DOXO-loaded micelles²⁷ under the present solution conditions, a maximum D.L. of ca. 0.9 and 1.1% (w/w) was obtained. The solubility of DCX and DOXO per gram of copolymer (S_CP) was concentration-dependent, reaching maximum values of up to ca. 9.2 and 10.2 mg g⁻¹, with solubility increments of ca. four and forty-fold the solubility of the pristine free drug(s) in water (5.5 and 0.5 mg L⁻¹, respectively).³² Among the different copolymers, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁ exhibited a slightly larger solubilization capacity with DCX and DOXO, which can be attributed to its longer hydrophobic blocks; in this respect, their extremely lengthy hydrophilic PEO blocks can also contribute to the copolymer bearing a more hydrophobic character. Conversely, BO₈EO₉₀BO₈ displayed lower D.L. values, possibly resulting from its smaller polymeric sizes, which do not allow the solubilization of large amounts of drug. Focusing on DCX-loaded micelles developed in the present manuscript, as the drug/copolymer weight ratio increased, the entrapment efficiency decreased as a consequence of the progressive DCX saturation of micelles (see Table 2) in agreement with previous observations of DOXO-loaded BO_nEO_mBO_n micelles.²⁷ This leads to the assumption that micellar formulations could enhance the solubility of poorly soluble drugs, but to a maximum limit, after which any increase in the drug concentration can bring about drug precipitation. Nevertheless, full micellar saturation with DCX was not achieved in the light of D.L. data within the drug/copolymer ratios analyzed. In addition, the observed D.L. and E.E. for single DCX-loaded micelles were somewhat larger than those previously obtained for other block copolymers, for example, PEO–PPO-based block copolymers such as Pluronics®⁵³ and Tetronic®,⁵⁴ or PEO–poly[N-(2-hydroxypropyl)methacrylamide]-lactate copolymers,³⁶ but were still lower than maximum values found for PEO-based block copolymers with other hydrophobic blocks such as poly(styrene oxide),³³ poly(lactic) acid,^21,55 poly(caprolactone)⁵⁶ or poly(2-oxazoline)-based copolymers.⁵⁷

Table 2 Docetaxel loading, D.L., entrapment efficiency, E.E. and solubility per gram of copolymer, S_CP, of BO_nEO_mBO_n copolymers

Copolymers	DCX/copolymer (w/w %)	D.L.^a (wt%)	E.E.^a (wt%)	SCP^b (mg g^−1)
a Estimated uncertainty ±0.2%.b ±1 mg g^−1.
BO8EO90BO8	0.2	0.10	49.5	0.99
	0.5	0.24	46.1	2.31
	1.0	0.30	29.7	2.97
	2.0	0.40	20.0	4.01
	3.0	0.47	15.7	4.70
BO14EO378BO14	0.5	0.34	69.0	3.45
	1.0	0.49	49.5	4.95
	1.5	0.66	44.1	6.62
	2.0	0.70	35.5	7.10
	2.5	0.81	32.7	8.18
	3.0	0.89	30.0	9.00
	3.5	0.93	26.9	9.29
BO20EO411BO20	0.5	0.36	72.0	3.60
	1.0	0.49	49.5	4.95
	1.5	0.70	47.0	7.05
	2.0	0.77	38.8	7.75
	2.5	0.91	36.7	9.17
	3.0	1.01	34.2	10.25
	3.5	1.06	30.7	10.75
BO21EO385BO21	0.5	0.43	86.5	4.33
	1.0	0.48	48.5	4.85
	1.5	0.61	41.2	6.18
	2.0	0.68	34.4	6.87
	2.5	0.79	32.0	8.00
	3.0	0.87	29.2	8.75
	3.5	0.99	28.6	10.00

---

On the other hand, dual DCX/DOXO-loaded micelles were also prepared at different DCX/DOXO weight ratios while keeping the total drug and polymer concentrations constant during preparation. The drug solubilities are shown in Table 3. The presence of two drugs within BO_nEO_mBO_n micelles did not adversely affect the apparent solubility of the individual drugs; in fact, an increase in the D.L. of copolymer micelles of up to. ca. 1.3% (w/w) was observed. Hence, it seems that the capacity of these BO_nEO_mBO_n micelles to incorporate drugs slightly increases when co-loading process takes place, which points to a more suitable environment for drug solubilization inside the micellar cores possibly due to enhanced drug(s)–polymer interactions as previously observed, for example, for poly(2-oxazoline)-based copolymers⁵⁷ and sterarate-grafter chitosan oligosaccharide micelles⁵⁸ to much larger extents. Moreover, the amount of the two drugs inside dual-loaded micelles was rather similar (when not a bit larger) to the single-loaded one with similar initial loading concentrations of the single drug (data not shown). D.L. capacities in dual-loaded systems followed the same trend as in the case of single-loaded drugs, i.e., the levels of drug encapsulation were higher for the most hydrophobic copolymers with the largest micellar cores, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO_21. Maximum D.L. values were also noted for most of the copolymer micelles at DCX/DOXO weight ratios of 50/50. In addition, the E.E. of DCX increased in the presence of DOXO inside the micellar core, reaching values of ca. 94.5% for the copolymer, BO₂₀EO₄₁₁BO_20. The ability of the present micelles to load somewhat larger amounts of two anticancer drugs into the core is a behavior that needs further study but these findings are consistent with earlier investigations.^21,59

Table 3 Drug loading, D.L., entrapment efficiency, E.E. and solubility per gram of copolymer, S_CP, of the copolymers of dual (DCX/DOXO) loaded BO_nEO_mBO_n copolymer micelles

Copolymers	DCX/DOXO (w/w %)	D.L.^a (wt%)	E.E.^a (wt%)	SCP^b (mg g^−1)
a Estimated uncertainty ±0.2%.b ±1 mg g^−1.
BO8EO90BO8	75/25	0.31	45.1	3.16
	50/50	0.34	42.5	3.40
	25/75	0.32	35.6	3.20
	0/100	0.26	26.0	2.60
BO14EO378BO14	75/25	1.07	61.9	10.83
	50/50	1.16	58.8	11.75
	25/75	0.94	42.2	9.50
	0/100	0.79	32.0	8.00
BO20EO411BO20	75/25	1.06	61.2	10.71
	50/50	1.26	60.8	12.75
	25/75	1.22	54.7	12.32
	0/100	1.01	41.0	10.25
BO21EO385BO21	75/25	0.93	53.5	9.37
	50/50	1.06	53.8	10.75
	25/75	1.14	51.1	11.50
	0/100	1.03	41.5	10.38

---

3.3. Micellar size and stability of loaded polymeric micelles

Size distribution and dispersion stability of drug-loaded micelles are crucial factors for their successful parenteral application. Since particle size will not only affect endocytosis by tumor cells but also influences longevity during systemic circulation, micelles must be small enough to evade detection and destruction by the RES. Sizes and size distributions of DCX and DCX/DOXO loaded micelles were measured by DLS (for DOXO-loaded micelles see ref. 27). Micellar sizes ranged from 20 to 50 nm depending on the copolymer type with relatively monodispersed distributions (Fig. 2a). No important differences in sizes were observed between non-loaded, single and dual-loaded systems and only an extremely slight size increase was observed as the loaded drug concentration was increased; however, a broader population size distribution could be observed for either single or dual-loaded micelles compared to non-loaded ones (Fig. 2b). The absence of size increases in single and dual-loaded micelles could be attributed to the effect of hydrophobic interactions between the aromatic rings of the drugs inside the micellar core and hydrogen bonds and van der Waals forces between hydroxy groups of drug(s) and block copolymers.⁶⁰ (Co)-loaded micelles could be readily freeze-dried and their initial size distribution recovered upon reconstitution in an aqueous solution, as observed previously²⁷ (data not shown).

Fig. 2 (a) Intensity fraction size distributions of bare polymeric micelles of copolymers BO₁₄EO₃₇₈BO₁₄ (·····), BO₂₀EO₄₁₁BO₂₀ (- - -) and BO₂₁EO₃₈₅BO₂₁ (—) at 25 °C in PBS buffer (pH 7.4) and (b) of unloaded (·····) and DCX/DOXO-loaded BO₈EO₉₀BO₈ polymeric micelles under similar conditions.

Formulation challenges such as stability and drug–drug compatibility need to be considered in multiple drug delivery in a single dosage form. Hence, the stability over time of single DCX and dual-loaded DCX/DOXO polymeric micelles was tested under serum mimicking conditions by monitoring the micellar size evolution. In particular, for dual DCX/DOXO-loaded micelles, sizes were observed to increase ca. 20 to 35 nm after 3–4 days of incubation, depending on the copolymer type (Fig. 3a), the larger increase being found for the copolymer, BO₂₁EO₃₈₅BO₂₁ (ca. 35 nm), and relatively similar increases being observed for the three remaining ones (ca. 20–25 nm). Similar behavior was found for single DCX-loaded micelles (not shown). In the present case, a synergistic improvement of micelle stability in the present multidrug-loaded micelles is not achieved, in contrast to what was observed by Kwon et al., by whom an enhanced stability in dual-loaded polyethylene glycol–poly(D,L-lactide) (PEG–PDLA) copolymer micelles was noted as a consequence of possible favorable drug–drug interactions through H-bonding inside micelles.^21,57 Moreover, our present data point to the fusion of adjacent loaded micelles into larger ones (see Fig. 3b) in order to provide a more suitable environment for the cargo and minimize serum protein binding. These changes can be a consequence of the reverse structure of the present copolymers: upon micellization, copolymer chains must display two tight-junctions at the micellar core–corona interface, which may lead to the preclusion of some hydrophobic blocks outside the micellar core configuring a less compact micellar interior and favoring the formation of intermicellar bridges;^48,49 hence, to avoid drug contact with the biological external environment and subsequent protein binding, a structural rearrangement of micelles can take place. This picture is corroborated by TEM images, wherein the spherical micellar morphology can be observed to be retained; however, drug loaded-micelles (either single or dual ones) upon extended incubation seem to be formed by a more loosely thicker corona than as-prepared loaded copolymer micelles (Fig. 3c and d). Micellar sizes observed by TEM are also slightly smaller than those obtained by DLS but are still comparable, despite TEM analysis being performed under ambient (dry) conditions, while DLS determines the hydrodynamic diameter (“equivalent sphere diameter”), i.e. the size of the swollen and hydrated particle in an aqueous phase.¹⁷ Whatever the case, after extensive incubation under the present conditions, the single and dual loaded-polymeric micelles were still below ca. 100–120 nm in size, which enable their tumor-specific accumulation via the EPR effect.

Fig. 3 (a) Temporal evolution of micellar sizes of dual-loaded BO_nEO_mBO_n copolymers. TEM images of dual DCX/DOXO loaded BO₂₁EO₃₈₅BO₂₁ micelles (b) during and (c) after restructuration. (d) TEM image of just-prepared dual DCX/DOXO loaded BO₂₁EO₃₈₅BO₂₁ micelles.

To ensure the delivery of the carried drug(s) to the site of action, the micellar carrier must be able to resist rapid dissociation upon dilution and exposure to blood plasma conditions. Hence, the physical stability of DCX-loaded and DCX/DOXO-loaded micelles was tested upon high dilution (1/50) in PBS buffer with pH 7.4 containing 10% FBS at 37 °C to simulate in vivo parenteral administration by monitoring the free DCX concentration in solution over time (final copolymer concentrations were well below the cmc). DCX-loaded polymeric micelles remained stable upon extensive incubation, at least for 20 days, in the protein rich medium (Fig. 4a). All the systems remained physically stable until day 3, the DCX solubility being above 90% of the initial value. As opposed to BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₇₈BO₂₁, micelles showing high stability over time (above 75% of the initial value after 20 days of incubation), BO₈EO₉₀BO₈ micelles gradually lost DCX, 60% remaining at day 8 and ca. 35% after day 20. BO₁₄EO₃₇₈BO₁₄ micellar nanocarriers also showed a slight DCX concentration loss, with 85% remaining at day 8 and ca. 60% after 20 days (Fig. 4a).

Fig. 4 Temporal evolution of micelle stability in terms of DCX retained in (a) single and (b) dual-loaded BO_nEO_mBO_n micelles over time at 37 °C in PBS buffer, pH 7.4, supplemented with 10% FBS.

For dual-loaded micelles, similar observations as for single-loaded micelles were noted (Fig. 4b). By monitoring the DCX release, it could be observed that all the copolymers displayed stabilities larger than 85% after 3–4 days of incubation and for BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₇₈BO₂₁ micelles, DCX remained above 73% after 20 days. BO₈EO₉₀BO₈ micelles gradually lost DCX, 60% remaining at day 8 and ca. 32% after day 20, while BO₁₄EO₃₇₈BO₁₄ micelles showed better stability than the former, with 79% remaining at day 8 and ca. 56% after 20 days. Hence, the largest resistance to disintegration was observed for BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₇₈BO₂₁ loaded micelles; this is probably because the larger hydrophobic chains would favor a larger core, providing a more suitable environment for the drug while reducing interactions with water. The observed DCX retention values inside BO_nEO_mBO_n micelles were larger than those observed for other structurally related polyethylene oxide (PEO)-based copolymers such as polyethylene oxide–propylene oxide (PEO–PPO)-based, including Pluronics and Tetronics,^36,61 poly(N-vinylpyrrolidone)–poly(D,L-lactide) (PVP–PDLL)⁶² or PEG–PDLA ones,²¹ and rather similar to those observed for more hydrophobic copolymers such as polystyrene oxide–polyethylene oxide (PSO–PEO)⁶³ and polyethylene glycol–polycaprolactone (PEG–PCL),⁶⁴ or mixed polymeric micelles of polyethylene glycol–poly(D,L-lactic acid/D-α-tocopheryl) polyethylene glycol 1000 succinate/stearic acid-modified chitosan polymers.⁶⁰

3.4 In vitro release kinetics

The release of DCX from polymeric micelles was assayed using dialysis tubing (Spectra Pore, cellulose ester membrane cutoff 3500 Da), which ensured that no micellar diffusion occurred. Release profiles for single and dual-drug loaded micelles in physiological mimicking medium (pH 7.4) are shown in Fig. 5a and b. For single DCX-loaded micelles, the release profiles were similar for all the copolymers and were characterized by the existence of an initial burst phase (Fig. 5a), in which ca. 41%, 38%, 40%, and 38% DCX was released during the first 5 h of incubation from BO₈EO₉₀BO₈, BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁ copolymers, respectively. The former phase was then followed by a sustained release pattern with ca. 92%, 86%, 81% and 79% DCX released after 30 h from each of the above polymeric micelles. The release profiles and the individual drug release rates did not change significantly for dual-loaded micelles compared to the single-loaded ones. Dual-loaded micelles containing DCX and DOXO also showed an initial burst phase, wherein ca. 39–45% and 34–42% of DCX and DOXO, respectively, were released after 5 h of incubation, followed by a sustained pattern, wherein 71–90% of DCX and 48–61% of DOXO were released after 30 h depending on the copolymer (Fig. 5b).

Fig. 5 In vitro drug release of (a) DCX and (b) DCX (closed symbols) and DOXO (open symbols) from single and dual-loaded BO_nEO_mBO_n polymeric micelles over time at 37 °C. Fit lines are not shown for clarity. In (b) only release from BO₈EO₉₀BO₈ and BO₁₄EO₃₇₈BO₁₄ is shown for a better visualization.

The observed release profiles for both single and dual-loaded micelles might be the result of a certain disruption of the micellar system due to cohesion, higher concentration gradients, and/or sink conditions.⁵⁵ The amount of drug released from BO₈EO₉₀BO₈ micelles was larger than for the other polymeric micelles, probably as a consequence of the slightly lower affinity between the drug and the shorter hydrophobic chains of this copolymer leading to smaller cores; conversely, a relatively higher retention of the drug inside the micellar structure was observed for the copolymers with longer BO blocks, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁.

The in vitro release profiles were fitted to the Higuchi³⁵ and Fickian diffusion models.³⁶ Tables 4 and 5 show that the latter model best fitted the experimental data on the basis of the correlation coefficient (R² > 0.98). This model takes into account drug diffusion, conformational changes in the micellar structure during release and partial transfer of drug from one micelle to another. At the beginning, with the uptake of water, micelles might swell and allow the drug within to diffuse through the pores. However, the hydrophobic cores could retard the diffusion of water into the core, hence decreasing the diffusion rate. The constant associated with drug diffusion (k₂) was rather similar (within uncertainty) for all copolymers investigated for both single and dual-loaded micelles, which points to a common mechanism of drug release, which is independent of the type of cargo inside the polymeric micelles.

Table 4 Coefficients of DCX release from single-loaded BO_nEO_mBO_n micelles

Copolymer	Higuchi			Fickian
	k	χ^2 reduced	R^2	k1	k2	k3	χ^2 reduced	R^2
BO8EO90BO8	17.74 (0.54)	44.93	0.951	−15.60 (4.66)	32.32 (3.38)	2.32 (0.50)	16.00	0.983
BO14EO378BO14	16.44 (0.46)	33.22	0.954	−8.16 (4.50)	25.953 (3.26)	1.62 (0.48)	14.92	0.980
BO20EO411BO20	16.26 (0.71)	78.55	0.872	−12.40 (3.44)	32.31 (2.50)	2.81 (0.37)	8.73	0.986
BO21EO385BO21	15.73 (0.64)	62.29	0.900	−14.30 (3.30)	31.81 (2.40)	2.72 (0.35)	8.05	0.987

---

Table 5 Coefficients of DCX and DOXO release from dual-loaded BO_nEO_mBO_n micelles

Copolymer	Higuchi			Fickian
	k	χ^2 reduced	R^2	k1	k2	k3	χ^2 reduced	R^2
DOCETAXEL
BO8EO90BO8	16.58 (0.54)	6.22	0.971	−13.26 (0.94)	30.88 (0.95)	2.35 (0.16)	0.36	0.998
BO14EO378BO14	16.00 (0.50)	4.54	0.958	−11.26 (0.47)	29.42 (0.41)	2.36 (0.07)	0.05	0.999
BO20EO411BO20	14.76 (0.51)	48.15	0.894	−9.41 (0.64)	27.60 (0.45)	2.32 (0.06)	0.33	0.999
BO21EO385BO21	14.21 (0.54)	55.41	0.869	−11.61 (0.59)	28.84 (0.42)	2.60 (0.06)	0.28	0.998
DOXORUBICIN
BO8EO90BO8	12.94 (0.68)	88.31	0.634	−1.52 (1.63)	23.37 (1.16)	2.23 (0.17)	2.17	0.991
BO14EO378BO14	11.55 (0.62)	73.59	0.594	−0.53 (1.65)	19.78 (1.17)	1.86 (0.17)	2.22	0.988
BO20EO411BO20	10.88 (0.64)	75.56	0.623	−0.67 (1.45)	17.34 (1.23)	1.65 (0.18)	2.33	0.987
BO21EO385BO21	10.34 (0.55)	67.89	0.644	−0.43 (1.22)	16.88 (1.11)	1.58 (0.15)	1.88	0.991

---

3.5. Internalization of drug-loaded micelles

To determine the potential use of these typess of polymeric micelles as potential drug delivery systems, in vitro cell internalization experiments were performed by exploiting the intrinsic fluorescence emission of the DOXO chromophore observe by fluorescence microscopy. As observed in Fig. 6a, after 1 h of incubation, cell exposure to free DOXO caused a rapid drug accumulation inside cells, particularly in cell nuclei. In contrast, the fluorescence intensity was much lower after 24 h incubation as a result of certain cell death, together with drug metabolization and subsequent excretion of the drug from the cells (Fig. 6b).

Fig. 6 : Fluorescence microscopy images of cellular uptake and intracellular distribution of drug-loaded BO_nEO_mBO_n micelles. (a) Bright field; (b) DOXO-fluorescence (red-coloured, λ_ex = 488 nm); (c) blue fluorescence from cell nuclei stained with DAPI (λ_ex = 355 nm); (d) merged images of free DOXO after (1) 1 h and (2) 24 h of administration and dual-loaded micelles after (3) 1 h and (4) 24 h of administration. Scale bar is 10 μm.

Conversely, the distribution pattern changed significantly for drug administration inside single and dual drug-loaded micelles. Drug accumulation in cells was observed to increase with incubation time. The initial lower accumulation may be caused by the slower release of DOXO (and, also DCX) from micelles (see Fig. 6c). The self-quenching effect of DOXO inside micelles makes fluorescence observable only when DOXO is released.⁶⁵ After 24 h of incubation, DOXO fluorescence inside the cells was particularly intense, indicating that large drug amounts had been released from the micellar cores and were localized inside the cells (Fig. 6d), mainly in the nuclei but with some remaining in the cytoplasm. These findings support the hypothesis of a sustained drug release inside the cells. Preliminary experiments suggested that the cell uptake of drug-loaded BO_nEO_mBO_n micelles would take place by an endocytosis-mediated mechanism, micelles being initially located within endosome vesicles enabling drug release in the cytosol in a sustained manner due to the acidic environment in the endosome.

3.6 Proliferation assay studies

The antiproliferative effects of single DCX and DOXO-loaded and dual DCX/DOXO-loaded BO_nEO_mBO_n micelles at different polymer concentrations were evaluated by means of the CCK8 proliferation assay in the breast MDA-MB-231 tumoral cell line, in which the sequential combinatorial administration of DCX and DOXO is used in advanced or metastatic breast cancer. An MTT assay could not be used because DOXO interferes with the formation of formazan crystals.⁶⁶ In addition, the efficacy of treatments was further evaluated in cervical HeLa cancer cells in order to account for differences due to different cell phenotypes.

Single DCX and DOXO-loaded polymeric micelles exhibited a dose-dependent cytotoxic activity, i.e. the cell cytotoxicity increased with the DOXO or DCX concentration released from the micelles, in agreement with the observations for free drugs (Fig. 7a and b). Moreover, they displayed a larger cell toxicity than free drugs at the same dose and incubation time, and thus larger IC₅₀ values, as observed in Fig. 7a and b. Provided that the copolymer was shown to be cytocompatible at the concentration used in the experiments (<2 mg mL⁻¹ for BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁ and <5 mg mL⁻¹ for BO₈EO₉₀BO₈), the enhancement in cell toxicity observed upon incubation with drug-loaded polymeric micelles can be ascribed to the cytotoxic effect of the loaded drug. This is compatible with an enhanced micelle accumulation inside the cell and a subsequent sustained drug release from the micelles, in agreement with fluorescence uptake data. This difference in toxicity can stem from the high stability of the micelles in vitro and the sustained drug release, which results in more drug being available to exert its therapeutic effect for longer times on both cancer cell lines. In this regard, in spite of free drugs being able to rapidly enter cells, they may be subsequently diffused out from cells, for example, through efflux pump mechanisms, decreasing their residence time.⁶⁷ Among the different single drug-loaded polymeric micelles, those made of the copolymer, BO₈EO₉₀BO₈, displayed the largest cell toxicities, especially in the low to moderate drug concentration range (<10 μM), probably as a consequence of their faster release from the micelle interior. For example, viabilities of only 13%, 23%, 28% and 32% were found for micelles of the copolymers BO₈EO₉₀BO₈, BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁, respectively, at a DCX concentration of 9 μM in MDA-MB-231 cancer cells after 72 h of incubation (Fig. 7a). At larger concentrations, cell toxicities became similar for all the copolymers since the amount of released drug is high enough to cause a huge cell death (>90%).

Fig. 7 Cell viabilities of (a) DCX and (b) DOXO-loaded polymeric micelles in MDA-MB-231 breast cancer cells after 72 h of incubation and of DCX-loaded polymeric micelles in HeLa cells after (c) 24 and (d) 72 h of incubation.

Cell proliferation was also observed to be enhanced with incubation time as a consequence of the progressive drug release (Fig. 7c). For example, growth inhibition caused by polymeric micelles loaded with 17.5 μM DCX decreased from 35%, 42%, 47% and 44% at 24 h to 11%, 14%, 16% and 18% at 72 h of incubation for BO₈EO₉₀BO₈, BO₁₄EO₃₇₈BO₁₄, BO₂₀EO₄₁₁BO₂₀ and BO₂₁EO₃₈₅BO₂₁, respectively. Similar behavior was observed previously, for example, for DCX-loaded poly(ethylene oxide)–poly(styrene oxide)³³ and poly(N-vinylpyrrolidone)–poly(D,L-lactide) micelles upon cargo release.^33,68 Moreover, despite both cell lines being greatly affected by the presence of the antineoplastic drugs, HeLa cells were observed to be slightly more sensitive (Fig. 7d), especially to DOXO.

On the other hand, the cytotoxicity of DCX/DOXO loaded in the present polymeric micelles was evaluated at different drug weight ratios. DCX/DOXO weight ratios close to 50 : 50 were observed to be the most effective in inhibiting cell proliferation (see Fig. 8a as an example). To further analyze whether DCX and DOXO combinations are synergistic, additive, or antagonistic against HeLa and MDA-MB-231 proliferation, the combination indices for the various dosing ratios were calculated using Compusyn⁶⁹ software. The calculated CI values of some of the free drug combinations in DMSO as well as most of the dual drug-loaded micelles were well below 1.0 (Fig. 8b), indicating a synergistic antiproliferative effect against both types of cancer cells, especially for MDA-MB-231 cancer cells at ratios close to 50 : 50 in weight. Similar findings were also reported, for example, upon dual administration of placlitaxel and rapamycin drug-loaded combinations in poly(ethylene glycol)–poly(lactide) (PEG–PLA) micelles to different tumor cell lines,⁵⁶ and dual-loaded etoposide/17-allylamino-17-demothoxygeldanamycin and bortezomib/17-allylamino-17-demothoxygeldanamycin poly(2-oxazoline) micelles.⁵⁴ Nevertheless, it is worth pointing out that further optimization of drug ratios is required, especially in in vivo models, for which an understanding of the pharmacokinetics and pharmacodynamics of each individual drug in the multidrug composition is needed.

Fig. 8 (a) Cell viabilities and (b) combination indices of free combined drugs and dual-loaded polymeric micelles at different weight ratios in MDA-MB-231 cancer cells after 24 h of incubation. Inset in (b) denotes combination indices in HeLa cells. Cell viabilities of single free drugs, free combined drugs, single and dual-loaded micelles of (c) copolymersBO₁₄EO₃₇₈BO₁₄ and BO₂₀EO₄₁₁BO₂₀ in MDA-MB-231 cells and of (d) copolymer BO₂₀EO₄₁₁BO₂₀ in HeLa ones after 24 h of incubation.

Cytotoxicity assays of dual DCX/DOXO-loaded polymeric micelles of copolymers BO₁₄EO₃₇₈BO₁₄ and BO₂₀EO₄₁₁BO₂₀ (50/50 weight ratio) were performed in both HeLa and MDA-MB-231 cells. We chose these copolymers since they provided intermediate release rates and high cell cytotoxicities while requiring a lower polymer concentration than BO₈EO₉₀BO₈ to form micelles. We observed that cell cytotoxicities after administration of dual-loaded micelles were larger than those obtained with free DCX/DOXO combined drugs or single-drug loaded micelles at similar doses, as observed in Fig. 8c and d, especially for the breast cancer cell line. For example, dual drug-loaded micelles of the copolymer, BO₁₄EO₃₇₈BO₁₄ (9 μM total drug concentration), showed a cytotoxicity of ca. 44% in MDA-MB-231 cells in contrast to ca. 55% and 69% for single DOXO and DCX-loaded ones (Fig. 8c). Similar results were also found for the copolymer, BO₂₀EO₄₁₁BO₂₀ (Fig. 8d).

The observed larger proliferative inhibition effect of dual-loaded micelles again confirms the protection role and progressive release of drugs exerted by polymeric micelles and their subsequent intracellular accumulation leading to enhanced cell death. However, given that DCX and DOXO act by different mechanisms, combination therapy with these drugs within a single micelle can offer a new available therapeutic option to treat tumoral cells.

4. Conclusions

The existence of resistance to classical mono-chemotherapeutic cancer treatments and the clinical evidence of synergistic responses and reduced toxicity associated with higher doses of individual drugs in tumor treatments make combination therapy of chemotherapeutics an unparalleled approach to overcome the limitations of single-drug treatments. However, for the analyzed chemotherapeutics, DOXO and DCX, currently used in the treatment of advanced and metastatic breast cancer, multiple drug combinations in a single delivery system are not yet commercially available. On the basis of these findings, we have selected BO_nEO_mBO_n polymeric micelles as a vehicle for single and dual drug delivery of DCX and DOXO. We have shown that BO_nEO_mBO_n micelles can effectively incorporate important amounts of both single and coloaded drugs. When coloaded, solubility extents are somewhat larger than in the case of single solubilized drugs. In addition, the present nanocarrier systems also provide good stability, with a relatively slight micelle size increase after 3–4 days of incubation, related to the fusion of adjacent micelles as a consequence of polymeric chain restructuration. The single- and dual-loaded micelles can effectively deliver drugs to cancer cells. Both single- and dual-loaded micelles displayed larger cell toxicities than free drugs administered at the same doses. In particular, DCX/DOXO-loaded micelles displayed synergistic effects against the two cancer cell lines (MDA-MB-231 and HeLa) tested. These synergistic effects were dependent on drug ratios and require further optimization of the corresponding drug formulations. The combination of a slight increase in drug loading and a decrease in the amount of both the total drug and polymeric excipients to achieve similar cell cytotoxicities while eliminating the need of organic solvents/excipients for DCX/DOXO parenteral administration make the present BO_nEO_mBO_n polymeric micelles appear an attractive drug delivery system and may have great advantages over current methods, potentially increasing the safety of clinical interventions while minimizing adverse side effects. Moreover, the inhibition ability of the P-gp efflux pump of some of the present copolymer varieties would also provide a further complement to configure an “active” nanocarrier with a great potential therapeutic efficacy.

Acknowledgements

The authors thank Ministerio de Economía y Competitividad (MINECO) and Xunta de Galicia for research projects MAT 2013-40971-R and EM2013-046, respectively. The authors also especially thank the staff of Instituto de Ortopedia y Banco de Tejidos Musculoesqueléticos of the Universidad de Santiago de Compostela, and especially Maite Silva, for helpful assistance during in vitro cell experiments. S.B. gratefully acknowledges MINECO for her Ramon y Cajal fellowship. E.V.A. is grateful to the Spanish Ministerio de Economia y Competitividad for her FPU (AP2012-2921) fellowship.

References

1. J. L. Arias, Drug targeting strategies in cancer treatment: an overview, Mini-Rev. Med. Chem., 2011, 11, 1–17.
2. P. Y. Lee and K. K. Y. Wong, Nanomedicine: a new frontier in cancer therapeutics, Curr. Drug Delivery, 2011, 8, 245–253.
3. J. Heidel and M. Davis, Clinical developments in nanotechnology for cancer therapy, Pharm. Res., 2011, 28, 187–199.
4. M. C. Jones and J. C. Leroux, Polymeric micelles: a new generation of colloidal drug carriers, Eur. J. Pharm. Biopharm., 1999, 48, 101–111.
5. V. P. Torchilin, Structure and design of polymeric surfactant-based drug delivery systems, J. Controlled Release, 2001, 73, 137–172.
6. V. P. Torchilin, Micellar nanocarriers: pharmaceutical perspectives, Pharm. Res., 2007, 24, 1–16.
7. G. Gaucher, M. H. Dufresne, V. P. Sant, N. Kang, D. Maysinger and J. C. Leroux, Block copolymer micelles: preparation, characterization and application in drug delivery, J. Controlled Release, 2005, 109, 169–188.
8. D. G. Yu, G. R. Williams, X. Wang, X.-K. Liu, H.-L. Li and S. W. A. Bligh, Dual drug release nancomposites prepared using a combination of electrospraying and electrospinning, RSC Adv., 2013, 3, 4652–4658.
9. D.-G. Yu, F. Liu, L. cui, Z.-P. Liu, X. Wang and S. W. Bligh, Coaxial electrspinning using a concentric Teflon spinneret to prepare biphasic-release nanofibers of helicid, RSC Adv., 2013, 3, 17775–17783.
10. S. N. Ramaswamy, Rational design of cancer-drug combinations, N. Engl. J. Med., 2007, 357, 299–300.
11. H. Zhao, Z. Wei, S. Xu, M. Tang, L. Deng, J. Zhang and A. Dong, Preparation and properties of polymer micelle-embedded nanofibrous membranes for dual-drug co-delivery and multistep release, RSC Adv., 2015 10.1039/c5ra03345d.
12. R. S. Finely and C. Balmer, Optimizing chemotherapy outcome, in Concepts in Oncology Therapeutics, ed. R. S. Finleyand and C. Balmer, American Society of Health-System Pharmacist, Bethesda, MD, 1998, pp. 87–89.
13. H. Gelderblom, J. Verweij, K. Nooter and A. Sparreboom, Cremophor EL: the drawbacks and advantages of vehicle selection for drug formulation, Eur. J. Cancer, 2001, 37, 1590–1598.
14. C. Alvarez-Lorenzo, A. Sosnik and A. Concheiro, PEO–PPO block copolymers for passive micellar targeting and overcoming multidrug resistance in cancer therapy, Curr. Drug Targets, 2011, 12, 1112–1130.
15. K. Mortensen, W. Batsberg and S. Hvidt, Effects of PEO–PPO diblock impurities on the cubic structure of aqueous PEO–PPO–PEO pluronics micelles: fcc and bcc ordered structures in F127, Macromolecules, 2008, 41, 1720–1727.
16. P. Taboada, G. Velasquez, S. Barbosa, Z. Yang, S. K. Nixon, Z. Zhou, F. Heatley, M. Ashford, V. Mosquera, D. Attwood and C. Booth, Micellization and drug solubilization in aqueous solutions of a diblock copolymer of ethylene oxide and phenyl glycidyl ether, Langmuir, 2006, 22, 7465–7470.
17. S. Barbosa, M. A. Cheema, P. Taboada and V. Mosquera, Effect of copolymer architecture on the micellization and gelation of aqueous solutions of copolymers of ethylene oxide and styrene oxide, J. Phys. Chem. B, 2007, 111, 10920–10928.
18. M. Crothers, Z. Zhou, N. M. P. S. Ricardo, Z. Yang, P. Taboada, C. Chaibundit, D. Attwood and C. Booth, Solubilisation in aqueous micellar solutions of block copoly(oxyalkylene)s, Int. J. Pharm., 2005, 293, 91–100.
19. M. E. N. P. Ribeiro, S. M. de Oliverira, N. M. P. S. Ricardo, S.-M. Mai, D. Attwood, S. G. Yeates and C. Booth, Diblock copolymers of ethylene oxide and 1,2-butylene oxide in aqueous solution: Formation of unimolecular micelles, Int. J. Pharm., 2008, 362, 193–196.
20. H. C. Shin, A. W. Alani, H. Cho, Y. Bae, J. M. Kolesar and G. S. Kwon, A 3-in-1 polymeric micelle nanocontainer for poorly water soluble drugs, Mol. Pharm., 2011, 8, 1257–1265.
21. H. C. Shin, A. W. Alani, D. A. Rao, N. C. Rockich and G. S. Kwon, Multi-drug loaded polymeric micelles for simultaneous delivery of poorly soluble anticancer drugs, J. Controlled Release, 2009, 140, 294–300.
22. F. Heatley, G.-E. Yu, W. B. Sun, E. J. Pywell, R. H. Mobbs and C. Booth, Triad sequence assignment of the ¹³C NMR spectra of copolymers of ethylene oxide and 1,2-butylene oxide, Eur. Polym. J., 1990, 26, 583–592.
23. G.-E. Yu, Z. Yang, M. Ameri, C. Attwood, J. H. Collett, C. Price and C. Booth, Diblock copolymers of ethylene oxide and 1,2-butylene oxide in aqueous solution. Effect of E-block length distribution on self-association properties, J. Phys. Chem. B, 1997, 101, 4394–4401.
24. A. Kelarakis, X.-F. Yuan, S.-M. Mai, Y.-W. Yang and C. Booth, Aqueous micellar solutions of a triblock copolymer of ethylene oxide and1,2-butylene oxide, B₁₂E₁₁₄B₁₂. Scaling the viscoelasticity of fluid and gel, Phys. Chem. Chem. Phys., 2003, 5, 2628–2634.
25. Z. Zhou, Y.-W. Yang, C. Booth and B. Chu, Association of a triblock ethylene oxide (E) and butylene oxide (B) copolymer (B₁₂E₂₆₀B₁₂) in aqueous solution, Macromolecules, 1996, 29, 8357–8361.
26. A. Cambón, J. Brea, M. I. Loza, C. Alvarez-Lorenzo, A. Concheiro, S. Barbosa, P. Taboada and V. Mosquera, Cytocompatibility and P-glycoprotein inhibition of block copolymers: Structure–activity relationship, Mol. Pharm., 2013, 10, 3232–3241.
27. A. Cambón, A. Rey-Rico, D. Mistry, J. Brea, M. I. Loza, D. Attwood, S. Barbosa, C. Alvarez-Lorenzo, A. Concheiro, P. Taboada and V. Mosquera, Doxorubicin-loaded micelles of reverse poly(butylene oxide)–poly(ethylene oxide)–poly(butylene oxide) block copolymers as efficient “active” chemotherapeutic agents, Int. J. Pharm., 2013, 445, 47–57.
28. M. E. N. P. Ribeiro, I. M. Cavalcante, N. M. P. S. Ricardo, S.-M. Mai, D. Attwood, S. G. Yeates and C. Booth, Solubilisation of griseofulvin in aqueous micellar solutions of diblock copolymers of ethylene oxide and 1,2-butylene oxide with lengthy B-blocks, Int. J. Pharm., 2009, 369, 196–198.
29. D. Mistry, T. Annable, X.-F. Yuan and C. Booth, Rheological behavior of aqueous micellar solutions of a triblock copolymer of ethylene oxide and 1,2-butylene oxide: B₁₀E4₁₀B₁₀, Langmuir, 2006, 22, 2986–2992.
30. A. Cambón, M. Alatorre-Meda, J. Juárez, A. Topete, D. Mistry, D. Attwood, S. Barbosa, P. Taboada and V. Mosquera, Micellisation of triblock copolymers of ethylene oxide and 1,2-butylene oxide: Effect of B-block length, J. Colloid Interface Sci., 2011, 361, 154–158.
31. J. Kim, J. E. Lee, S. H. Lee, J. H. Yu, J. H. Lee, T. G. Park and T. Hyeon, Designed fabrication of a multifunctional polymer nanomedical platform for simultaneous cancer-targeted imaging and magnetically guided drug delivery, Adv. Mater., 2008, 20, 478–483.
32. S. H. Yalkowsky and Y. He, Handbook of aqueous solubility data, CRC Press, Boca Raton, FL, 2003.
33. M. Elsabahy, M. E. Perron, N. Bertrand, G.-E. Yu and J. C. Leroux, Solubilization of docetaxel in poly(ethylene oxide)-block-poly(butylene/styrene oxide) micelles, Biomacromolecules, 2007, 8, 2250–2257.
34. K. K. Upadhyay, A. N. Bhatt, E. Castro, A. K. Mishra, K. Chuttani, B. S. Dwarakanath, C. Schatz, J. F. Le Meins, A. Misra and S. Lecommandoux, In vitro and in vivo evaluation of docetaxel loaded biodegradable polymersomes, Macromol. Biosci., 2010, 10, 503–512.
35. H. Wang, J. Xu, J. Wang, T. Chen, Y. Wang, Y. W. Tan, H. Su, K. L. Chan and H. Chen, Probing the kinetics of short-distance drug release from nanocarriers tonanoacceptors, Angew. Chem., Int. Ed., 2010, 49, 8426–8430.
36. Z. Wei, J. Hao, S. Yuan, Y. Li, W. Juan, X. Sha and X. Fang, Paclitaxel-loaded Pluronic P123/F127 mixed polymeric micelles: formulation, optimization and in vitro characterization, Int. J. Pharm., 2009, 376, 176–185.
37. B. M. Rao, A. Chakraborty, M. K. Srinivasu, M. L. Devi, P. R. Kumar, K. B. Chandrasekhar, A. K. Srinivasan, A. S. Prasad and J. Ramanatham, A stability-indicating HPLC assay method for docetaxel, J. Pharm. Biomed. Anal., 2006, 41, 676–681.
38. A. Sparreboom, O. van Tellingen, W. J. Nooijen and J. H. L. Bejinen, Preclinical pharmacokinetics of paclitaxel and docetaxel, Anti-Cancer Drugs, 1998, 9, 1–17.
39. R. A. Walker, J. L. Jones, S. Chappell, T. Walsh and J. A. Shaw, Molecular pathology of breast cancer and its application to clinical management, Cancer Metastasis Rev., 1997, 16, 5–27.
40. F. Zunino and G. Capranico, DNA topoisomerase II as the primary target of antitumoranthracyclines, Anticancer Drug Design, 1990, 5, 307–317.
41. K. D. Miller, W. McCaskill-Stevens, J. Sisk, D. M. Loesch, F. Monaco, R. Seshadri and G. W. Sledge Jr, Combination versus sequential doxorubicin and docetaxel as primary chemotherapy for breast cancer: A randomized pilot trial of the Hoosier Oncology Group, J. Clin. Oncol., 1999, 17, 3033–3037.
42. I. C. Smith, S. D. Heys, A. W. Hutcheon, I. D. Miller, S. Payne, E. J. Gilbert, A. K. Ah-See, O. Eremin, L. G. Walker, T. K. Sarkar, S. P. Eggleton and K. N. Ogston, Neoadjuvant chemotherapy in breast cancer: Significantly enhanced response with docetaxel, J. Clin. Oncol., 2002, 20, 1456–1466.
43. A. Ardavanis, D. Tryfonopoulos, I. Yiotis, G. Gerasimidis, N. Baziotis and G. Rigatos, Non-allergic nature of docetaxel-induced acute hypersensitivity reactions, Anti-Cancer Drugs, 2004, 15, 581–585.
44. A. J. Tije, J. Verweij, W. J. Loos and A. Sparreboom, Pharmacological effects of formulation vehicles: implications for cancer chemotherapy, Clin. Pharmacokinet., 2003, 42, 665–685.
45. J. M. Berthiaume and K. B. Wallace, Adriamycin-induced oxidative mitochondrial cardiotoxicity, Cell Biol. Toxicol., 2007, 23, 15–25.
46. Y. Bae, A. W. Alani, N. C. Rockich, T. S. Lai and G. S. Kwon, Mixed pH-sensitive polymeric micelles for combination drug delivery, Pharm. Res., 2010, 27, 2421–2432.
47. J. R. Hasenstein, H.-C. Shin, K. Kasmerchak, D. Buehler, G. S. Kwon and K. R. Kozak, Antitumor activity of triolimus: a novel multidrug-loaded micelle containing paclitaxel, rapamycin, and 17-AAG, Mol. Cancer Ther., 2012, 11, 2233–2242.
48. A. Cambon, E. Figueroa-Ochoa, M. Blanco, S. Barbosa, J. F. A. Soltero, P. Taboada and V. Mosquera, Micellar self-assembly, bridging and gelling behaviour of two reverse triblock poly(butylene oxide)–poly(ethylene oxide)–poly(butylene oxide) copolymers with lengthy hydrophilic blocks, RSC Adv., 2014, 4, 60484–60496.
49. A. Cambon, E. B. Figueroa-Ochoa, J. Juárez, E. M. Villar-Alvarez, A. Pardo, S. Barbosa, J. F. A. Soltero, P. Taboada and V. Mosquera, Complex self-assembly of reverse poly(butylene oxide)–poly(ethylene oxide)–poly(butylene oxide) triblock copolymers with long hydrophobic and extremely lengthy hydrophilic blocks, J. Phys. Chem. B, 2014, 4, 5258–5269.
50. M. E. Cavet, K. L. Harrington, K. R. Vandermeid, K. W. Ward and J.-Z. Zhang, Comparison of the effect of multipurpose contact lens solutions on the viability of culturedcorneal epithelial cells, Contact Lens and Anterior Eye, 2009, 32, 171–175.
51. J. S. Palmer, W. J. Cromie and R. C. Lee, Surfactant administration reduces testicularischemia-reperfusion injury, J. Urol., 1998, 159, 2136–2139.
52. C. Allen, D. Maysinger and A. Eisenberg, Nano-engineering block copolymer aggregates for drug delivery, Colloids Surf., B, 1999, 16, 3–27.
53. Z. Liu, D. Liu, L. Wang, J. Zhang and N. Zhang, Docetaxel-loaded pluronic P123 polymeric micelles: in vitro and in vivo evaluation, Int. J. Mol. Sci., 2011, 12, 1684–1696.
54. C. Alvarez-Lorenzo, J. Gonzalez-Lopez, M. Fernandez-Tarrio, I. Sanchez-Macho and A. Cocheiro, Tetronic micellization, gelation and drug solubilization: Influence of pH and ionic strength, Eur. J. Pharm. Biopharm., 2007, 66, 244–252.
55. Y. Li, M. Jin, S. Shao, W. Huang, F. Yang, W. Chen, S. Zhang, G. Xia and Z. Gao, Small-sized polymeric micelles incorporating docetaxel suppress distant metastases in the clinically-relevant 4T1 mouse breast cancer model, BMC Cancer, 2014, 14, 329.
56. A. S. Mikhail and C. Allen, Poly(ethylene glycol)-b-poly(ε-caprolactone) micelles containing chemically conjugated and physically entrapped docetaxel: Synthesis, characterization, and the influence of the drug on micelle morphology, Biomacromolecules, 2010, 11, 1273–1280.
57. Y. Han, Z. He, A. Schulz, T. K. Bronich, R. Jordan, R. Luxenhofer and A. V. Kabanov, Synergistic combinations of multiple chemotherapeutic agents in high capacity poly(2-oxazoline) micelles, Mol. Pharm., 2012, 9, 2302–2313.
58. M. D. Zhao, F. Q. Hu, Y. Z. Du, H. Yuan, F. Y. Chen, Y. M. Lou and H. Yu, Coadministration of glycolipid-like micelles loading cytotoxic drug with different action site for efficient cancer chemotherapy, Nanotechnology, 2009, 20, 055102.
59. G. P. Mishra, D. Nguyen and A. W. G. Alani, Inhibitory effect of paclitaxel and rapamycin individual and dual drug-loaded polymeric micelles in the angiogenic cascade, Mol. Pharm., 2013, 10, 2071–2078.
60. J. Dou, H. Zhang, X. Liu, M. Zhang and G. Zhai, Preparation and evaluation in vitro and in vivo of docetaxel loaded mixed micelles for oral administration, Colloids Surf., B, 2014, 114, 20–27.
61. D. A. Chiappetta, C. Alvarez-Lorenzo, A. Rey-Rico, P. Taboada, A. Concheiro and A. Sosnik, N-alkylation of poloxamines modulates micellar assembly and encapsulation and release of the antiretroviral efavirenz, Eur. J. Pharm. Biopharm., 2010, 76, 24–37.
62. D. Le Garrec, S. Gori, L. Luo, D. Lessard, D. C. Smith, M.-A. Yessine, M. Ranger and J. C. Leroux, Poly(N-vinylpyrrolidone)-block-poly(D,L-lactide) as a new polymeric solubilizer for hydrophobic anticancer drugs: in vitro and in vivo evaluation, J. Controlled Release, 2004, 99, 83–101.
63. A. Cambón, S. Barbosa, A. Rey-Rico, E. B. Figueroa-Ochoa, J. F. A. Soltero, S. G. Yeates, C. Alvarez-Lorenzo, A. Concheiro, P. Taboada and V. Mosquera, Poly(ethylene oxide)–poly(styrene oxide)–poly(ethylene oxide) copolymers: micellization, drug solubilization, and gelling features, J. Colloid Interface Sci., 2012, 387, 275–284.
64. J. Jin, B. Sui, J. Gou, J. Liu, X. Tang, H. Xu, Y. Zhang and X. Jin, PSMA ligand conjugated PCL–PEG polymeric micelles targeted to prostate cancer cells, PLoS One, 2014, 9, e112200.
65. D. C. Rees, E. Johnson and O. Lewinson, ABC transporters: the power to change, Nat. Rev. Mol. Cell Biol., 2009, 10, 218–227.
66. A. C. Collier and C. A. Pritsos, The mitochondrial uncoupler dicumarol disrupts the MTT assay, Biochem. Pharmacol., 2003, 66, 281–287.
67. S. Kunjachan, B. Rychlik, G. Storm, F. Kiessling and T. Lammers, Multidrug resistance: physiological principles and nanomedical solutions, Adv. Drug Delivery Rev., 2013, 65, 1852–1865.
68. V. Alakhov, E. Klinski, P. Lemieux, G. Pietrzynski and A. Kabanov, Block copolymeric biotransport carriers as versatile vehicles for drug delivery, Expert Opin. Biol. Ther., 2001, 1, 583–602.
69. T.-C. Chou and N. Martin, CompuSyn software for drug combinations and for general dose-effect analysis, and user's guide, ComboSyn, Inc, Paramus, NJ, 2007, http://www.combosyn.com.

---

Footnote

† These authors contribute equally to this work.

---