Functional block copolymer nanocarriers for anticancer drug delivery

Abstract

Polymer-based nanocarriers for anticancer drug delivery bearing “clickable”, biodegradable, pH-sensitive and subcellular targeting functions were designed and successfully obtained. Firstly, well-defined functional amphiphilic diblock copolymers were synthesized applying a multistep controlled polymerization and modification procedure. As a result, copolymers comprising alkyne-end functionalized biodegradable poly(D,L-lactide) and polycationic poly(N,N-dimethylaminoethyl methacrylate) blocks were obtained. The latter blocks were additionally functionalized with subcellular targeting triphenylphosphonium cations. The amphiphilic block copolymers self-associated in aqueous media into nanosized functional micelles that were able to incorporate the natural anticancer drug curcumin into their biodegradable cores. The in vitro cytotoxicity evaluation of block copolymer micelles indicated a low intrinsic inhibitory potential on the proliferation of different human cell lines. More importantly, the drug-loaded nanocarriers demonstrated an obvious ability to induce apoptosis and exhibited more prominent inhibition of the NF-κB transcription factor in cancer cell lines and their drug-resistant analogues, as compared to the free drug. The obtained results are optimistic for potential application of the functional block copolymers in nanomedicine.

---

1 Introduction

In recent years, much attention has been paid to developing new nanoparticle (NP) based drug-delivery systems with obvious advantages compared to conventional forms of dosage, such as enhanced bioavailability, higher efficiency, lower toxicity and controlled release.¹ The rational design of drug delivery nanovehicles is based on two targeting approaches. Passive targeting relies on the characteristics of the delivery vehicle itself and the pathology of the specific disease that leads to a preferential accumulation of nanoparticles at the site of interest. The passive mechanism of nanoparticle accumulation and uptake is known as the enhanced permeability and retention (EPR) effect and was first reported by Maeda et al.^2,3 Usually, the NP are surface-modified with nonionic hydrophilic polymer, thus avoiding elimination by the reticuloendothelial system (RES) and assuring longer circulation time.⁴ The use of stimuli responsive carriers is another option for accumulating active substances to the desired sites.^5,6 The second approach is based on active targeting strategies that direct the drug delivery systems to the sites of interest on cellular and subcellular level.⁷ Various targeting moieties (ligands) with high affinity for proteins selectively overexpressed on the surface of cancer cells, as compared to normal cells, such as anitibodies,⁸ small molecules,⁹ peptides¹⁰ and polysaccharides^11,12 are used to decorate the drug delivery nanovehicles. Moreover, there is an increasing interest in drug delivery systems targeted to specific intracellular organelles. Thus, many different approaches have been proposed for attaining efficient intracellular drug delivery to an organelle of interest.¹³

Up to now, many nanomaterials including polymeric micelles,¹⁴ dendrimers,¹⁵ liposomes,¹⁶ and various inorganic nanoparticles,^17–19 have been utilized as carriers in drug delivery systems. Among them, polymer-based nanoparticles are one of the most promising candidates for targeted drug delivery. They are able to encapsulate a wide variety of drugs, to release them over prolonged periods and exhibit excellent stability, both in vitro and in vivo.²⁰ Controlled polymerization methods combined with highly efficient modification techniques such as “click” chemistry approach, offer the possibility for the development of multifunctional nanoparticles with unique properties and functions for biomedical applications depending on the target disease and site of delivery (organs, tissues, cells or subcellular organelles).^21–23 The most commonly used hydrophobic polymers for controlled drug delivery applications include biodegradable polyesters such as poly(lactic acid) (PLA), poly(glycolic acid), and their copolymers, as well as poly(ε-caprolactone).^24,25 A variety of amphiphilic block copolymers comprising hydrophilic polymers such as poly(ethylene glycol) (PEG), poly(amino acids) or other charged polymers are used for the preparation of self-assembled therapeutic carriers.²⁶ In addition, stimuli-responsive polymeric micelles are also extensively used in drug delivery.²⁷

Curcumin (Curc), is a natural polyphenol derived from Curcuma longa and possesses a plethora of attractive properties that include anticancer, anti-amyloid, antioxidant, antidiabetic, anti-inflammatory, antibiotic, and antiviral activities.^28,29 As a result, it has attracted an enormous interest as a potential multifunctional anticancer agent. The hurdles for Curc biomedical applications are the poor solubility and chemical instability in aqueous medium as well as the low cellular uptake.^30,31 In order to overcome those problems Curc has been encapsulated in various copolymer micelles such as poly(ε-caprolactone)-based block copolymers,^32–34 PEG-monoacrylate, poly(N-isopropylacrylamide) and poly(N-vinyl-2-pyrrolidone) random copolymers,³⁵ or poly(PEG) methyl ether methacrylate and polystyrene block copolymers.³⁶ Most of those systems lack either targeting functions or if there are any, they cannot be easily replaced by other ligands according to a specific use. Since there is no universal drug nanocarrier it is very important to have in hand a drug delivery system that can be relatively easy adjusted to a specific target (cell and/or cellular organelle). Designing such a modular polymer system would make it easy to finely tune the properties of the drug delivery vehicle without starting the synthesis each time from the beginning.

Herein, we present a new synthetic strategy for the preparation of functional polymeric drug nanocarrier bearing clickable alkyne end-group and ligands for intracellular targeting. Novel, alkyne end-functionalized amphiphilic poly(D,L-lactide)-b-poly(N,N-dimethylaminoethyl methacrylate) (A-PLA-b-PDMAEMA) block copolymers of desired hydrophilic/hydrophobic balance were obtained via controlled polymerization techniques. The copolymers were further decorated with triphenylphosphonium cations for potential intracellular targeting. The nanocarriers were designed to possess easily biodegradable polyester core stabilized by an outer shell of PDMAEMA-blocks accommodating the targeting ligands. Moreover, PDMAEMA contains weakly basic amino groups that appears to have sufficient buffering capacity for nanocarrier's endosomal escape.³⁷ Finally, the copolymers are functionalized with alkyne end-groups that makes them a useful platform for further modular modifications with various additional functions. The copolymers' self-association and loading with Curc as an anticancer drug was investigated. Initial in vitro biological evaluations were performed showing promising results for potential application of the functional polymeric drug carrier in nanomedicine.

2 Materials and methods

2.1 Reagents and materials

All chemicals were purchased from Sigma-Aldrich. Dichloromethane (DCM, ≥99.5%) and tetrahydrofuran (THF, >99%) were distilled from calcium hydride prior to use. D,L-Lactide (LA) was recrystallized from toluene/ethyl acetate mixture (95 : 5 v/v). N,N-Dimethylaminoethyl methacrylate (DMAEMA, 98%) was passed through a column containing neutral aluminum oxide. Propargyl alcohol (PrOH, 99%) was distilled under reduced pressure. Triethylamine (TEA, ≥99%) was distilled from potassium hydroxide. Methanol (≥99.8%), hexane (≥99%), acetone (≥99.5%), copper(I) bromide (CuBr, 99.999%), 2-bromo-2-methylpropionyl bromide (BIBB, 98%), 4-dimethylaminopyridine (DMAP, >99%), 1,1,4,7,10,10-hexamethyltriethylenetetramine (HMTETA, 97%), (4-bromobutyl)triphenylphosphonium bromide (Br-Bu-TPP-Br, 98%), and Curc were used as received.

2.2 Synthesis of alkyne end-functionalized poly(D,L-lactide) (A-PLA-OH)

Typically, D,L-lactide (2.6 g, 18.04 mmol) and DMAP (0.18 g, 1.44 mmol) were dried in vacuo for 1 h. Then, 9 mL of DCM were added under Ar atmosphere. The temperature was adjusted to 35 °C and 42 μL (0.72 mmol) from the initiator (PrOH) were injected into the solution. The reaction proceeded for 24 h followed by concentration and precipitation of the product in cold methanol.

Yield: 1.8 g (69%). ¹H NMR (600 MHz, CDCl₃, δ, ppm): 5.15 (CH–(CH₃)–O), 4.74 (CH₂–C≡H), 4.36 (CH–(CH₃)–OH), 2.50 (C≡H), 1.57 (CH–(CH₃)–O), 1.49 (CH–(CH₃)–OH). FTIR (1/λ, cm⁻¹): 1082 (ν_s C–O–C), 1182 (ν_as C–O–C), 1452 (δ_as CH₃), 1748 (ν C=O).

2.3 Synthesis of an atom transfer radical polymerization (ATRP) macroinitiator (A-PLA-Br)

The functional polyester A-PLA-OH was quantitatively converted into an ATRP macroinitiator applying a previously described procedure.³⁸ Briefly, A-PLA-OH (1.5 g, 0.52 mmol) was dissolved in 11 mL of DCM followed by the addition of 0.28 mL (2.03 mmol) TEA. The solution was cooled down to 0 °C (ice bath) and a three-fold excess of BIBB (0.19 mL, 1.56 mmol), dissolved in 12 mL of DCM was added dropwise under Ar. The reaction proceeded at room temperature for 24 hours. The solution was stirred overnight with small amount of charcoal. The salt precipitate and the charcoal were removed by filtration followed by solution concentration and product precipitation in cold methanol.

Yield: 1.2 g (80%). ¹H NMR (600 MHz, CDCl₃, δ, ppm): 5.15 (CH–(CH₃)–O), 4.73 (CH₂–C≡H), 2.50 (C≡H), 1.98 (C–(Br)–(CH₃)₂), 1.57 (CH–(CH₃)–O).

2.4 Synthesis of functional amphiphilic poly(D,L-lactide)-b-poly(N,N-dimethylaminoethyl methacrylate) diblock copolymer (A-PLA-b-PDMAEMA)

The macroinitiator, A-PLA-Br (1.1 g, 0.38 mmol) and CuBr (0.054 g, 0.38 mmol) were dissolved in 5 mL of THF under Ar. Then, DMAEMA (0.96 mL, 5.7 mmol) and HMTETA (0.21 mL, 0.76 mmol) were added. The mixture was degassed and backfilled with argon thrice and stirred at 60 °C for 5 h. The polymerization was stopped in liquid nitrogen. The reaction mixture was diluted with THF and passed through a column containing neutral Al₂O₃. The clear solution was concentrated and the product was precipitated in chilled hexane.

Yield: 1.78 g (86.4%). ¹H NMR (600 MHz, CDCl₃, δ, ppm): 5.15 (CH–(CH₃)–O), 4.74 (CH₂–C≡H), 4.06 (O–CH₂–CH₂–N), 2.58 (O–CH₂–CH₂–N), 2.50 (C≡H), 2.29 (N–(CH₃)₂), 1.90–1.80 (CH₂–C–(CH₃)), 1.57 (CH–(CH₃)–O), 1.05–0.87 (C–(CH₃)₂ + CH₂–C–(CH₃)). FTIR (1/λ, cm⁻¹): 1086 (ν_s C–O–C), 1132 (ν C–N), 1182 (ν_as C–O–C), 1454 (δ_as CH₃), 1726 (ν C=O DMAEMA), 1755 (ν C=O lactide), 2772 and 2821 (ν –N(CH₃)₂), 2945 (ν –CH₃ and –CH₂).

2.5 Block copolymer modification with triphenylphosphonium cations (A-PLA-b-PDMAEMA-TPP⁺)

In a typical reaction, A-PLA-b-PDMAEMA (1.45 g, 4.40 mmol (CH₃)₂N-groups) and Br-Bu-TPP-Br (0.25 g, 0.52 mmol) were dissolved in 12 mL of dry DCM under argon. The temperature was increased to 35 °C and the reaction solution was stirred for 24 h. The solvent was evaporated, the residue was redissolved in methanol and dialyzed against methanol (dialysis membrane with molecular weight cut-off (MWCO) 1200 Da) for 24 h. The product was recovered through a solvent evaporation.

Yield: 1.28 g (75.3%). ¹H NMR (600 MHz, CDCl₃, δ, ppm): 7.90–7.65 ((C₆H₅)₃P⁺), 5.15 (CH–(CH₃)–O), 4.74 (CH₂–C≡H), 4.06 (O–CH₂–CH₂–N), 3.44 (⁺N–(CH₃)₂), 2.57 (O–CH₂–CH₂–N), 2.50 (C≡H), 2.28 (N–(CH₃)₂), 1.90–1.80 (CH₂–C–(CH₃)), 1.57 (CH–(CH₃)–O), 1.05–0.87 (C–(CH₃)₂ + CH₂–C–(CH₃)).

2.6 Characterization

¹H NMR spectra were recorded in CDCl₃ on a Bruker Avance II+ 600 MHz instrument. Gel permeation chromatography (GPC) was performed in THF at a flow rate of 1.0 mL min⁻¹ using Shimadzu Nexera XR HPLC chromatograph, equipped with quaternary pump, degasser, automatic injector, column heater, UV/Vis (SPD-20A) detector, differential refractive index (RID-20A) detector, 10 μm PL gel mixed-B, 5 μm PL gel 500 Å and 50 Å columns. The system was calibrated versus polystyrene narrow molar mass standards. Infrared spectra were recorded on a IRAffinity-1 Shimadzu Fourier Transform Infrared (FTIR) spectrophotometer with MIRacle Attenuated Total Reflectance Attachment. UV/Vis spectra were taken on a DU 800 Beckman Coulter spectrometer. Transmission electron microscope (TEM) images were obtained using HRTEM JEOL JEM-2100 (200 kV) instrument. Dynamic light scattering (DLS) measurements for particles' size and size-distribution determination were carried out at 37 °C on a Zetasizer Nano-ZS instrument (Malvern Instruments), equipped with a He–Ne laser (λ = 633 nm) with a scattering angle of 173°. The ζ-potentials were calculated from the obtained electrophoretic mobility by the Smoluchowski equation:where η is the solvent viscosity, μ is the electrophoretic mobility, and ε is the dielectric constant of the solvent.

2.7 Block-copolymer micelles formation

Typically, the block copolymer (A-PLA-b-PDMAEMA-TPP⁺) was dissolved in acetone (10 mg mL⁻¹). Then, 0.5 mL of the solution was added dropwise to approx. 3 mL of ultrapure water (18.2 MΩ cm) under stirring. The mixture was stirred vigorously for 12 h at room temperature in order to remove the organic solvent. The concentration was adjusted to 1 mg mL⁻¹ by the addition of ultrapure water.

2.8 Critical micelle concentration (CMC) determination

The CMC of the block copolymers in aqueous media was determined applying the dye solubilization method described by Alexandridis et al.³⁹ UV measurements on increasing concentrations of block copolymers (0.005–1.0 mg mL⁻¹) in the presence of the hydrophobic dye 1,6-diphenyl-1,3,5-hexatriene (DPH, 10 μL from 0.4 mM solution in methanol) in aqueous media were performed. While DPH does not dissolve in water it solubilizes into the hydrophobic micellar core, giving a characteristic spectrum with absorption maximum at 356 nm. By plotting the intensity of this maximum vs. block copolymer concentration the CMC was estimated as the cross-point of the obtained two straight lines.

2.9 In vitro drug loading and release

Curcumin was dissolved in acetone (1 mg mL⁻¹). Then, 1 mL from the Curc solution was used to dissolve 10 mg of the block copolymer. As described above, 0.5 mL from the copolymer/Curc solution was added to 3 mL of water under vigorous stirring. After organic solvent evaporation the concentration was adjusted to 1 mg mL⁻¹ (micelles to Curc ratio – 10 : 1 w/w). In order to determine the drug loading efficiency (DLE) and the drug loading capacity (DLC) the micellar dispersion was filtered (0.45 μm), lyophilized and the residue was redissolved in acetone. The solution was analyzed by UV/Vis spectroscopic measurement. A standard curve (λ_max = 418 nm, ε = 61 882 M⁻¹ cm⁻¹) was constructed from different concentrations of Curc in acetone. The DLE and DLC were calculated according to the following equations:

The in vitro Curc release from the loaded copolymer micelles was evaluated by regular membrane dialysis at 37 °C against phosphate buffered saline (PBS pH 7). Typically, 1 mL of the tested formulations was placed in a Spectra/Por® dialysis membrane tubing (MWCO 10 000 Da). The dialysis bag was then placed in a temperature-controlled vessel, containing 100 mL of PBS with an addition of 5% (v/v) ethanol. At predetermined time intervals aliquots were taken from the release medium and assayed for curcumin by UV/Vis spectroscopy at λ_max = 427 nm. Alternatively, 2 mL of aqueous dispersion from the Curc-loaded micelles (1 mg mL⁻¹) were placed in a vial at 37 °C, followed by the addition of 1 mL of non-mixing solvent chloroform. At predetermined time intervals 0.5 mL from the organic phase were withdrawn and subjected to UV/Vis spectroscopy (λ_max = 415 nm, ε = 53 703 M⁻¹ cm⁻¹) while the release medium volume was kept constant by the addition of fresh chloroform.

2.10 Cell lines and culture conditions

The biocompatibility of the non-loaded micelles was tested in a panel of cell lines, namely HEK-293 (human embryonal kidney, non-tumorigenic), HL-60 (acute myeloid leukemia) and HEP-G2 (human hepatocellular carcinoma). The cytotoxic activity of free and formulated curcumin was assessed against the acute myelocyte leukemia-derived HL-60 cell line, and its multidrug-resistant (HL-60/DOX) and cisplatin-resistant (HL-60/CDDP) sublines, as well as in the non-tumorigenic cell line HEK293 to allow determination of the anticancer selectivity indices. HEK-293, HEP-G2, HL-60, and HL-60/DOX were purchased from the German Collection of Microorganisms and Cell Cultures (DSMZ GmbH, Braunschweig, Germany). HL-60/CDDP was established in the authors' lab by continuous selection in growth medium with gradually increasing concentrations of cisplatin. The cells were grown in controlled environment – cell culture flasks at 37 °C in an incubator “BB 16-Function Line” Heraeus (Kendro, Hanau, Germany) with humidified atmosphere and 5% CO₂. The growth medium was 90% RPMI-1640 + 10% FBS. HL-60/CDDP were cultivated in the presence of 25 μM cisplatin, and HL-60/DOX were grown in medium containing 0.2 μM doxorubicin in order to maintain their drug resistance phenotype. Three days before the cytotoxicity screening and the mechanistic assays, cells were re-cultured in drug-free medium in order to evade possible synergistic interactions with the tested active treatments.

2.11 Cytotoxicity assessment (MTT-dye reduction assay)

The cellular viability after exposure to the non-loaded micelles, free curcumin or its micellar formulations was assessed using the MTT-dye reduction assay as previously described elsewhere⁴⁰ with minor modifications.⁴¹ The method is based on the reduction of the yellow tetrazolium salt MTT to a violet MTT-formazan by the mitochondrial succinate dehydrogenase in viable cells. Briefly, exponentially growing cells were plated in 96-well flat-bottomed microplates (100 μL per well) at a density of 1 × 10⁵/mL (leukemic cell lines) or 1 × 10⁴ cells per mL (HEK-293, HEP-G2) and after 24 h of incubation at 37 °C, they were exposed to various concentrations of the tested compounds for 72 h. For each concentration a set of at least 8 wells were used. After the exposure period, 10 μL MTT solution (10 mg mL⁻¹ in PBS) aliquots were added to each well. Thereafter, the microplates were incubated for 4 h at 37 °C and the MTT-formazan crystals formed were dissolved through addition of 100 μL per well 5% (formic acid-acidified 2-propanol). The MTT-formazan absorption was recorded using a Labexim LMR-1 microplate reader at 580 nm. Cell survival fractions were calculated as percentage of the untreated control. In addition, IC₅₀ values were derived from the concentration–response curves.

Bioassay data processing and statistics. The MTT-bioassay survival data were normalized as percentage of the untreated control (set as 100% viability), were fitted to sigmoidal dose response curves and the corresponding IC₅₀ values (concentrations causing 50% suppression of cellular viability) were calculated using non-linear regression analysis (GraphPad Prizm Software for PC). The statistical processing of biological data included the paired Student's t-test whereby values of p ≤ 0.05 were considered as statistically significant. The MTT-bioassay derived IC₅₀ values of free curcumin were divided by those derived with the corresponding formulations to yield the modulation indices (MI). In addition, the resistance indices as a relative merit for the level of resistance in HL-60/DOX and HL-60/CDDP were determined as the ratio between the IC₅₀ in the multidrug-resistant or HL-60/CDDP sublines and the corresponding IC₅₀ in the sensitive parent line HL-60. Selectivity indices were derived for free curcumin and its micellar formulations as the ratio between the IC₅₀ values on the non-tumorigenic kidney cell line HEK293 and the arithmetic mean of the IC₅₀ values for the malignant cell lines.

2.12 Apoptosis assay

The apoptotic DNA fragmentation was examined using a commercially available ‘Cell-death detection’ ELISA kit (Roche Applied Science). This assay allows semi-quantitative determination of the characteristic for the apoptotic process histone-associated mono- and oligonucleosomal DNA-fragments using ‘sandwich’ ELISA. Exponentially proliferating HL-60, HL-60/DOX or HL-60/CDDP cells were exposed to equieffective concentrations (IC₅₀) of the tested formulations for 24 h and thereafter cytosolic fractions of 1 × 10⁴ cells per group (treated or untreated) served as antigen source in a sandwich ELISA, utilizing a primary anti-histone antibody-coated microplate and a secondary peroxidase-conjugated anti-DNA antibody. The spectrophotometric immunoassay for histone-associated DNA fragments was performed according to the manufacturers' instructions at 405 nm, using a microprocessor-controlled microplate reader (Labexim LMR-1). The results are expressed as the oligonucleosome enrichment factor (representing a ratio between the absorption in the treated vs. the untreated control samples). The assay was run in quadruplicate.

2.13 NF-κB (p65) inhibition assay

Inhibition of the transcription factor NF-κB (p65) in HL-60, HL-60/DOX or HL-60/CDDP cells treated with equieffective concentrations (IC₅₀) of the tested formulations for 24 h was quantitatively determined using a commercially available NF-κB p65 ELISA (Enzo Life Science; EKS-446) according to the manufacturer's instructions and expressed in relative units per mg of protein. The assay was run in quadruplicate.

3 Results and discussion

3.1 Controlled synthesis of functional amphiphilic block copolymers

The synthetic route to functional block copolymers intended for biomedical applications is depicted in Scheme 1. A metal-free organocatalyzed controlled ring-opening polymerization proposed by Nederberg et al.⁴² was applied for the synthesis of PLA biodegradable block. For the first time, a commercially available propargyl alcohol was used as a heterobifunctional initiator for D,L-lactide controlled polymerization. Thus, while the hydroxyl group initiated the polymerization process, each of the formed polymer chains was quantitatively functionalized with the “clickable” alkyne end-group with no need of additional protection/deprotection steps. The heterobifunctional polyesters obtained were characterized by ¹H NMR and FTIR-spectroscopy. The average degree of LA polymerization was estimated from the relative intensities of methyne protons from the polyester repeating units at 5.15 ppm and those corresponding to the methylene protons next to the terminal alkyne group at 4.75 ppm (Fig. S1†). Moreover, the presence of strong bands at 1082, 1182 and 1748 cm⁻¹, corresponding to C–O–C (symmetric and asymmetric) and C=O stretching vibrations in the products' FTIR spectra further confirmed the formation of polyester block (Fig. S2†). The polymers' molar-mass distributions were obtained from the GPC performed in THF. The GPC-elugrams show monomodal molar-mass distribution with dispersity in the 1.15–1.20 range (Fig. S3†).

Scheme 1 Synthetic route to functional amphiphilic diblock copolymers.

The second synthetic step involved a quantitatively conversion of the polyester into an ATRP-macroinitiator (A-PLA-Br) through a reaction of its terminal hydroxyl group with BIBB (Scheme 1). A key point at this stage is the thorough purification of the product from the excess of the esterification reagent. That was achieved through extended treatment of the reaction mixture with charcoal. The complete conversion of polymer's terminal hydroxyl groups was confirmed by ¹H NMR analysis. The resonance at 1.49 ppm corresponding to methyl protons next to the terminal OH-group has completely disappeared, while a new resonance at 1.98 ppm corresponding to the six methyl protons of bromoisobutyryl end-group appeared (Fig. 1a). The second polymer block of desired length was obtained through a controlled ATRP of DMAEMA initiated by the macroinitiator A-PLA-Br (Scheme 1). The polymerization proceeded in THF and was completed within 5 hours. The copper-containing catalytic system was removed by passing the reaction mixture through Al₂O₃-containing column. The amphiphilic block copolymers were recovered in high yield (above 75%). Since the molar mass of the PLA-macroinitiator is known, the average degree of DMAEMA polymerization was calculated from the relative intensities of methylene protons (next to the ester group of the DMAEMA-units) at 4.06 ppm and those corresponding to the methyne protons (from A-PLA-Br) at 5.15 ppm (Fig. 1b). The calculated degrees of DMAEMA polymerization were close to the theoretical ones (defined from the monomer/macroinitiator molar ratio) suggesting good control over the process. The presence of DMAEMA units in the products was detected also by FTIR analyses. Additional bands at 1132, 1726, 2772 and 2821 cm⁻¹ corresponding to C–N, C=O and –N(CH₃)₂ stretching vibrations of the second polymer block appeared in the products' spectra (Fig. S4†). The formation of block-copolymer architecture was evidenced by GPC (Fig. 2). The molar masses were obtained vs. polystyrene standards and differ from the real ones but the most important is that there is a clear shift to lower elution volume (higher molar mass) for the block copolymer as compared to the corresponding macroinitiator. Moreover, the block copolymers still eluted as monomodal species with dispersities in the 1.35–1.40 range.

Fig. 1 ¹H NMR (600 MHz) spectrum in CDCl₃ of: (a) A-PLA-Br macroinitiator, and (b) amphiphilic diblock copolymer A-PLA-b-PDMAEMA.

Fig. 2 GPC elugram of poly(D,L-lactide)-macroinitiator (A-PLA-Br) and the corresponding diblock copolymer (A-PLA-b-PDMAEMA) in THF (vs. polystyrene standards).

The final step in the synthetic procedure was the introduction of subcellular targeting ligands into the PDMAEMA-block. Partial quaternization of the dimethylamino-groups was performed with bromobutyl-derivative of TPP⁺ ligand (Scheme 1). The presence of alkyl group in the quaternization reagent was intended to reduce the steric hindrance of the bulky aromatic rings. The degree of quaternization was determined after the functional block copolymer purification from the relative intensities of the aromatic protons at 7.65–7.90 ppm and the methyne protons (from the PLA-block) at 5.15 ppm (Fig. S5†). Furthermore, a new resonance at 3.44 ppm corresponding to the protons of the dimethyl ammonium groups appeared in the spectrum of A-PLA-b-PDMAEMA-TPP⁺. It is worth mentioning that during all polymerization and modification stages the alkyne end-groups remain intact and are clearly visible in all ¹H NMR spectra (4.75 ppm). The macroinitiators and block copolymers molar-mass characteristics are summarized in Table 1.

Table 1 Characteristics of macroinitiators and the corresponding functional diblock copolymers

Macroinitiator			Block copolymer				TPP^+ ligand^d
Code	Mn^a (g mol^−1)	ĐM^b	Code	DPn^a	MIeff^c	ĐM^b	Code
a Number-average molar masses (Mn) and degrees of polymerization (DPn), as determined by ^1H NMR analyses.b Molar-mass dispersity (ĐM), as determined by GPC in THF vs. polystyrene standards.c Macroinitiator efficiency (MIeff = DP^targetedn/DP^NMRn).d TPP^+-modified amphiphilic diblock copolymers (10 mol% from the corresponding DMAEMA-units).
A-PLA10-Br	1650	1.20	A-PLA10-b-PDMAEMA19	19	0.79	1.39	B1
A-PLA19-Br	2900	1.19	A-PLA19-b-PDMAEMA26	26	0.86	1.37	B2
			A-PLA19-b-PDMAEMA17	17	0.88	1.40	B3
A-PLA15-Br	2350	1.18	A-PLA15-b-PDMAEMA21	21	0.84	1.37	B4
A-PLA25-Br	3750	1.15	A-PLA25-b-PDMAEMA14	14	0.71	1.35	B5

---

3.2 Self-association of A-PLA-b-PDMAEMA-TPP⁺ amphiphilic block copolymers

Above their CMC the amphiphilic diblock copolymers can spontaneously form micelles comprising hydrophobic PLA cores and hydrophilic PDMAEMA coronas decorated with TPP⁺. Polymer micelles in aqueous media were obtained from A-PLA-b-PDMAEMA-TPP⁺ copolymers by the nanoprecipitation technique. Initially, the block copolymer was dissolved in acetone (good solvent for both blocks) and added dropwise to water. After the organic solvent evaporation polymer micelles were formed and subjected to analyses.

The CMC values for the amphiphilic diblock copolymers were obtained applying the dye solubilization method in which the absorption intensity of DPH at 356 nm was plotted as a function of polymer concentration (Fig. S6†). The block copolymers' CMC values were in the 0.1–0.2 mg mL⁻¹ range. Thus, for further analyses micelles with concentration of 1 mg mL⁻¹ were prepared. DLS measurements showed formation of particles with average sizes between 80 and 165 nm with monomodal and relatively narrow size-distribution depending on the constituent polymer blocks length and the hydrophilic/hydrophobic balance (Fig. 3a). The copolymer with the shortest hydrophobic PLA-block (B1) forms particles with the biggest average sizes most likely due to the higher number of macromolecules needed for the stabilization of the core–shell structure.

Fig. 3 DLS measurements: (a) size-distribution curves for copolymers B1 (d = 165 nm; PdI: 0.079) and B2 (d = 76 nm; PdI: 0.270), and (b) zeta-potential distribution curves (ζ = 64.5 mV for B1 and ζ = 55.7 mV for B2).

The zeta-potential determination confirmed that the micelles' surface is positively charged due to the exposure of DMAEMA-units and TPP⁺ ligands to the aqueous media (Fig. 3b).

The formation of spherical nanoparticles was visualized by TEM analysis (Fig. 4). The data concerning average particle sizes are in good agreement with those obtained from the DLS measurements.

Fig. 4 TEM-images of micelles obtained from functional block copolymers: (a) B1 (mean particle diameter 187 (±14) nm), and (b) B2 (mean particle diameter 84 (±10) nm).

3.3 In vitro drug loading and release

To investigate the potential application of the functional micelles in drug delivery systems two of the synthesized block copolymers (B1 and B2) were chosen for drug loading and the other in vitro evaluations due to their different composition, hydrophilic/hydrophobic balance and micelles properties (micelles size and size distribution) (Table S1† and Fig. 3). The drug loading was preformed in similar manner as the micelles preparation. The only difference is that Curc was dissolved together with the block copolymer in acetone and the solution was added dropwise to water in order to form drug-loaded micelles through co-association. The DLE was almost quantitative for both copolymer micelles (94 and 98 wt% for B1/Curc and B2/Curc respectively) and DLC was close to 10 wt% (9.4 and 9.8 wt% for B1/Curc and B2/Curc respectively). DLS and zeta-potential measurements of the loaded particles did not show any significant changes in their average sizes, size-distribution and surface charge. The similar positive values of particles zeta-potentials before and after drug loading indicate that Curc is located in the hydrophobic core of the micelles.

The in vitro release of Curc from the polymer carriers was followed in phosphate buffered saline (PBS pH 7) for 24 h at 37 °C. Additionally, 5 wt% of ethanol was introduced to the release media in order to solubilize and detect the released Curc via UV/Vis spectroscopy. The cumulative release percentages of Curc loaded functional block-copolymer micelles (B1/Curc and B2/Curc) are presented in Fig. 5. The initial accelerated Curc release during the first hours was followed by a much slower drug release. At the 24^th hour of the experiment less than 25% of the loaded Curc has left the particles' core. Alternatively, the Curc release from aqueous drug-loaded micellar dispersion was followed using a non-mixing organic solvent (chloroform) as a release media (Fig. S7†). Even at these conditions (release media that readily dissolves Curc) the cumulative Curc release reached a plateau of 43% at the 24^th hour of the experiment. The results obtained are indicative for the ability of the drug delivery system to preserve its cargo into the micelles' core during the transport to the target cells.

Fig. 5 In vitro curcumin release from A-PLA-b-PDMAEMA-TPP⁺ (B1 and B2) copolymer micelles in PBS (pH 7) at 37 °C.

3.4 Cytotoxicity assessment of non-loaded block copolymer micelles

The low cytotoxicity and pharmacological inertia is a hallmark requirement for all carriers and excipients, used for compounding drug delivery systems. For this reason, we sought to determine the intrinsic inhibitory potential of the non-loaded functional polymer micelles on the proliferation of human cell lines of different cell type and origin. The tested systems were assayed against the non-tumorigenic cell line HEK-293 (embryonal kidney) and two tumor cell lines – HL-60 (myeloid) and HEP-G2 (hepatocellular), chosen as representative of important cellular populations, that would be exposed upon systemic delivery of the micelles. The concentration response curves are depicted in Fig. 6a. The micelles displayed relatively low cytotoxic effects, whereby they failed to induce 50% inhibition of cellular viability/proliferation. Even at the highest concentration employed (100 μg mL⁻¹) the cell growth inhibition was less than 40%. The comparative evaluation of the tested functional diblock copolymers (B1 and B2) shows that the longer cationic PDMAEMA-block in B2 leads to some increase of its inhibitory potential. The established negligible antiproliferative effects of the tested non-loaded micelles against a panel of human cell lines with different origin is a favorable characteristic as far as their potential application as drug carriers is concerned.

Fig. 6 Cytotoxicity of: (a) non-loaded A-PLA-b-PDMAEMA-TPP⁺ (B1 and B2) copolymer micelles after 72 h continuous exposure at 37 °C. Each data point represents the arithmetic mean ± SD of 6 separate experiments, and (b) free Curc and the loaded A-PLA-b-PDMAEMA-TPP⁺ copolymer micelles (B1/Curc and B2/Curc) after 72 h continuous exposure at 37 °C. Each data point represents the arithmetic mean ± SD of 8 separate experiments.

3.5 In vitro comparative assessment of antiproliferative effect of free vs. micellar curcumin against human tumor cell lines

The cytotoxicity bioassay of the presented systems was conducted to address the relative inhibitory activity of free, non-encapsulated Curc (introduced as an ethanol solution), vs. its copolymer-micellar formulations (B1/Curc and B2/Curc). The in vitro antineoplastic activity was evaluated against the acute promyelocyte leukemia derived HL-60 cell line, and its two resistant variants, namely HL-60/DOX (multidrug-resistant) and HL-60/CDDP (cisplatin resistant). The concentration–response curves are shown in Fig. 6b. The equieffective inhibitory concentrations, as well as the modulation and the resistance indices derived thereof are summarized in Table 2. The results obtained in the chemosensitive and the resistant cells indicate that the encapsulation of Curc into copolymer micelles was invariably consistent with a significant augmentation of the cytotoxic activity and shifting of the concentration–response curves towards lower concentrations. Similar superior in vitro therapeutic efficiency of nanoparticle encapsulated drugs was previously demonstrated by Tan et al.⁴³ when evaluating the cytotoxicity of docetaxel and tamoxifen loaded polylactide-based amphiphilic drug-delivery systems versus free drugs in MCF7 breast cancer cell lines.

Table 2 Equieffective concentrations (IC₅₀), modulation (MI) and resistance indices (RI) of curcumin-loaded functional polymer micelles (B1/Curc and B2/Curc), vs. the free drug

Cell line	IC50 (μg mL^−1)			MI^a		RI^b
	Curc	B1/Curc	B2/Curc	B1/Curc	B2/Curc	Curc	B1/Curc	B2/Curc
a MI = IC50(free curcumin)/IC50(encapsulated curcumin).b RI = IC50(resistant HL-60 variant)/IC50(chemosensitive HL-60).
HL-60	8.02	1.62	1.57	4.95	5.1	—	—	—
HL-60/DOX	11.8	1.42	1.33	8.3	8.87	1.47	0.88	0.85
HL-60/CDDP	25.4	1.51	1.46	16.82	17.4	3.16	0.93	0.94

---

The IC₅₀ values for both systems (B1/Curc and B2/Curc) were several fold lower than those obtained for the free drug with identical magnitude of modulation – by factors of approximately 5 (HL-60), 8–9 (HL-60/DOX) and 17 (HL-60/CDDP). Moreover, while the resistant variants were less responsive to free Curc relative to the chemosensitive parent cell line, the Curc-loaded micelles displayed collateral sensitivity phenomenon, i.e. bypassing of the resistance mechanisms and lower IC₅₀ values in HL-60/DOX and HL/CDDP, relative to HL-60 yielding resistance indices <1.

Another parameter that was addressed within the MTT-bioassay was the selectivity index (SI) of Curc-loaded functional polymer micelles vs. the free drug, i.e. the ratio between the IC₅₀ values on the non-tumorigenic kidney cell line HEK293 and the arithmetic mean of the IC₅₀ values for the cancer cell lines. Evident from the SI values ploted in Fig. 7, Curc is approximately two fold less toxic on normal vs. malignant cell lines (SI = 2.2), whereas the antineoplastic selectivity of its micellar formulations is further increased with SI values of 4.8 (for B1/Curc) and 4.2 (for B2/Curc). These findings demonstrate that the drug-loaded polymer systems are characterized by favorable modulation of the antiproliferative activity of encapsulated drug as a tumor-targeted natural compound.

Fig. 7 Selectivity indices of free (Curc) and loaded into A-PLA-b-PDMAEMA-TPP⁺ copolymer micelles (B1/Curc and B2/Curc) curcumin.

3.6 Apoptogenic effect of free and encapsulated circumin

The ability of Curc – free or formulated to trigger programmed cell death was monitored using a commercially available ELISA kit for quantitative determination of histone-associated DNA fragmentation, which is a key feature of apoptosis. Following a 24 h exposure of the chemosensitive and resistant cell lines to equieffective concentrations of the tested systems (Curc, B1/Curc and B2/Curc) there was a statistically significant increase of the content of mono- and oligonucleosomal DNA fragments. The apoptogenic effects of micellar formulations, and especially that of B1/Curc, were more pronounced than those of the free drug (Fig. 8). The most prominent apoptotic response was documented with B1/Curc in HL-60/CDDP. These findings further account for the pronounced effects of the formulated drug, especially taking into consideration the treatment of cells with lower concentrations than those for free Curc, based on the IC₅₀ values.

Fig. 8 Apoptotic DNA-fragmentation as assessed by a commercially available ELISA kit, following a 24 h exposure to equieffective concentrations of free or micellar curcumin (n = 4). Statistically significant differences are presented, as follows: * – significantly different vs. the control at p ≤ 0.05; ** – significantly different vs. the control at p ≤ 0.01; # – significantly different vs. the free curcumin at p ≤ 0.05.

3.7 NF-κB (p65) inhibition assay

The effects of free and formulated curcumin on the cellular levels of p65 – the active form of the transcriptional factor NF-κB were compared, using a commercially available ELISA kit (Fig. 9). Corresponding to the cytotoxicity and apoptogenic bioassays Curc loaded into functional copolymer micelles exhibited more prominent NF-κB inhibition, as compared to the free drug. These findings are worth discussing because although the anticancer effects of Curc are mediated via pleiotropic mechanisms NF-κB is considered as its ultimate pharmacological target.^44–46 This transcriptional factor is upregulated in a variety of tumors and promotes the expression of a battery of genes, crucial for the malignant transformation, and mediating hyper-proliferation and inhibition of apoptosis.^44,47–49 The used read-out system monitors the levels of the active form of NF-κB (p65), which is capable of nuclear translocation and exhibits transcription-activating properties, and hence the presented data clearly indicate that the micellar formulations of Curc (B1/Curc and B2/Curc) are superior to the free drug at its upmost pharmacological action. Taken together the mechanistic studies on the apoptogenic and NF-κB-modulating activities of free vs. formulated Curc imply that the augmented activity and especially the bypassing of the resistance mechanisms in HL-60/DOX and HL-60/CDDP established with the loaded micelles are at least partly mediated by enhanced curcumin-induced modulation of apoptotic and p65-dependent cell signaling.

Fig. 9 NF-κB (p65) inhibition as assessed by a commercially available ELISA kit, following a 24 h exposure to equieffective concentrations of free or micellar curcumin (n = 4). Statistically significant differences are presented, as follows: * – significantly different vs. the control at p ≤ 0.05; ** – significantly different vs. the control at p ≤ 0.01; # – significantly different vs. the free curcumin at p ≤ 0.05.

4 Conclusions

In this study, a synthetic strategy for controlled synthesis of functional amphiphilic block copolymers comprising biodegradable PLA- and polycationic PDMAEMA-blocks, decorated with clickable end-groups and targeting ligands is presented. The copolymers self-associate into nanosized core–shell micelles in aqueous media. The initial studies concerning thier potential application in drug delivery systems is demonstrated through micelle-core loading with natural anticancer drug curcumin and initial in vitro evaluations. The low intrinsic cytotoxicity of the tested micellar systems and their ability to retain the generic anticancer properties of curcumin, give us a reason to consider them as feasible drug-delivery systems for this important natural antineoplastic agent. Moreover, the presence of clickable end-groups in the copolymers opens new possibilities for additional modifications that would lead to further functionalization of the drug delivery systems. Currently, we are working on the preparation of multifunctional polymer system for cellular and subcellular-targeted drug delivery based on the presented block copolymer.

Acknowledgements

This research was financially supported by the National Science Fund of Bulgaria through project DFNI T02-21/2014. I.D. thanks Prof. Katerina Goracinova for the fruitful discussions.

Notes and references

1. E. Blanco, H. Shen and M. Ferrari, Nat. Biotechnol., 2015, 33, 941–951.
2. Y. Matsumura and H. Maeda, Cancer Res., 1986, 46, 6387–6392.
3. H. Maeda, J. Controlled Release, 2012, 164, 138–144.
4. K. Miyata, R. Christie and K. Kataoka, React. Funct. Polym., 2011, 71, 227–234.
5. D. Shenoy, S. Little, R. Langer and M. Amiji, Pharm. Res., 2005, 22, 2107–2114.
6. A. Hoffman, Adv. Drug Delivery Rev., 2013, 65, 10–16.
7. V. Torchilin, Adv. Drug Delivery Rev., 2012, 64, 302–315.
8. H. Tada, H. Higuchi, T. Watanabe and N. Ohuchi, Cancer Res., 2007, 67, 1138–1144.
9. D. Pan, J. Turner and K. Wooley, Chem. Commun., 2003, 2400–2401.
10. Y. Zhang, J. Wang, D. Bian, X. Zhang and Q. Zhang, Eur. J. Pharm. Biopharm., 2010, 74, 467–473.
11. B. Davis and M. Robinson, Curr. Opin. Drug Discovery Dev., 2002, 5, 279–288.
12. J. Bergen, H. von Recum, T. Goodman, A. Massey and S. Pun, Macromol. Biosci., 2006, 6, 506–516.
13. A. Maity and D. Stepensky, Mol. Pharmaceutics, 2016, 13, 1–7.
14. Z. Ahmad, A. Shah, M. Siddiq and H.-B. Kraatz, RSC Adv., 2014, 4, 17028–17038.
15. Z. Zhou, X. Ma, C. J. Murphy, E. Jin, Q. Sun, Y. Shen, E. Van Krik and J. Murdoch, Angew. Chem., Int. Ed., 2014, 53, 10949–10955.
16. L. Hosta-Rigau, R. Chandrawati, E. Saveriades, P. Odermatt, A. Postman, F. Ercole, K. Breheney, K. Wark, B. Stadler and F. Caruso, Biomacromolecules, 2010, 11, 3548–3555.
17. M. Epple, K. Ganesan, R. Heumann, J. Klesing, A. Kovtun, S. Neumann and V. Sokolova, J. Mater. Chem., 2010, 20, 18–23.
18. J.-H. Park, G. von Maltzahn, L. Zhang, M. Schwartz, E. Ruoslahti, S. Bhatia and M. Sailor, Adv. Mater., 2008, 20, 1630–1635.
19. E. Boisselier and D. Astruc, Chem. Soc. Rev., 2009, 38, 1759–1782.
20. N. Kamaly, Z. Xiao, P. Valencia, A. Radovic-Moreno and O. Farokhzad, Chem. Soc. Rev., 2012, 41, 2971–3010.
21. J. Nicolas, S. Mura, D. Brambilla, N. Mackiewicz and P. Couvreur, Chem. Soc. Rev., 2013, 42, 1147–1235.
22. V. Rostovtsev, L. Green, V. Fokin and K. Sharpless, Angew. Chem., Int. Ed., 2002, 41, 2596–2599.
23. S. Zhang, K. Chan, R. Prud'homme and A. Link, Mol. Pharmaceutics, 2012, 9, 2228–2236.
24. Y. Ohya, A. Takahashi and K. Nagahama, Adv. Polym. Sci., 2012, 247, 65–114.
25. T. Dash and V. Konkimalla, Mol. Pharmaceutics, 2012, 9, 2365–2379.
26. X.-B. Xiong, Z. Binkhathlan, O. Molavi and A. Lavasanifar, Acta Biomater., 2012, 8, 2017–2033.
27. M. Nakayama, J. Akimoto and T. Okano, J. Drug Targeting, 2014, 22, 584–599.
28. M. Egan, M. Pearson, S. Weiner, V. Rajendran, D. Rubin, J. Glöckner-Pagel, S. Canny, K. Du, G. Lukacs and M. Caplan, Science, 2004, 304, 600–602.
29. T. Esatbeyoglu, P. Huebbe, I. Ernst, D. Chin, A. Wagner and G. Rimbach, Angew. Chem., Int. Ed., 2012, 51, 5308–5332.
30. G. Garcea, D. Jones, R. Singh, A. Dennison, P. Farmer, R. Sharma, W. Steward, A. Gescher and D. Berry, Br. J. Cancer, 2004, 90, 1011–1015.
31. P. Anand, A. Kunnumakkara, R. Newman and B. Aggarwal, Mol. Pharmaceutics, 2007, 4, 807–818.
32. K. Letchford, R. Liggins and H. Burt, J. Pharm. Sci., 2008, 97, 1179–1190.
33. K. Yoncheva, K. Kamenova, T. Perperieva, V. Hadjimitova, P. Donchev, K. Kaloyanov, S. Konstantinov, M. Kondeva-Burdina, V. Tzankova and P. Petrov, Int. J. Pharm., 2015, 490, 298–307.
34. G. Grancharov, V. Gancheva, M. Kyulavska, D. Momekova, G. Momekov and P. Petrov, Polymer, 2016, 84, 27–37.
35. S. Bisht, G. Feldmann, S. Soni, R. Ravi, C. Karikar, A. Maitra and A. Maitra, J. Nanobiotechnol., 2007, 5, 3.
36. T. Chang, D. Trench, J. Putnam, M. Stenzel and M. Lord, Mol. Pharmaceutics, 2016, 13, 924–932.
37. S. Agarwal, Y. Zhang, S. Maji and A. Greiner, Mater. Today, 2012, 15, 388–393.
38. M. Spasova, L. Mespouille, O. Coulembier, D. Paneva, N. Manolova, I. Rashkov and P. Dubois, Biomacromolecules, 2009, 10, 1217–1223.
39. P. Alexandridis, J. Holzwarth and T. Hatton, Macromolecules, 1994, 27, 2414–2425.
40. T. Mosmann, J. Immunol. Methods, 1983, 65, 55–63.
41. S. Konstantinov, H. Eibl and M. Berge, Br. J. Haematol., 1999, 107, 365–374.
42. F. Nederberg, E. Connor, M. Möller, T. Glauser and J. Hedrick, Angew. Chem., Int. Ed., 2001, 40, 2712–2715.
43. G.-R. Tan, S.-S. Feng and D. Leong, Biomaterials, 2014, 35, 3044–3051.
44. S. Singh and B. Aggarwal, J. Biol. Chem., 1995, 270, 24995–25000.
45. C. Buhrmann, A. Mobasheri, F. Busch, C. Aldinger, R. Stahlmann, A. Montaseri and M. Shakibaei, J. Biol. Chem., 2011, 286, 28556–28566.
46. J. Marquardt, L. Gomez-Quiroz, L. Camacho, F. Pinna, Y.-H. Lee, M. Kitade, M. Domínguez, D. Castven, K. Breuhahn, E. Conner, P. Galle, J. Andersen, V. Factor and S. Thorgeirsson, J. Hepatol., 2015, 63, 661–669.
47. S. Shishodia, P. Potdar, C. Gairola and B. Aggarwal, Carcinogenesis, 2003, 24, 1269–1279.
48. A. Bharti, Y. Takada and B. Aggarwal, J. Immunol., 2004, 172, 5940–5947.
49. S. Shishodia, H. Amin, R. Lai and B. Aggarwal, Biochem. Pharmacol., 2005, 70, 700–713.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19236j

---