Gadolinium/DOTA functionalized poly(ethylene glycol)-block-poly(acrylamide-co-acrylonitrile) micelles with synergistically enhanced cellular uptake for cancer theranostics

Abstract

The combination of diagnostic and therapeutic functions into a nano-carrier could achieve a delivery system with both accurate diagnosis and delivery capabilities. Herein, a functionalized polymer of 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid-poly(ethyleneglycol)-block-poly(acrylamide-co-acrylonitrile) (DOTA-PEG-b-poly(AAm-co-AN)) was designed and synthesized. Doxorubicin (DOX) and gadolinium ions (Gd³⁺) were loaded into the hydrophobic core and chelated on the shell of the micelles during the self-assembling process to endow the prepared micelles with drug delivery and magnetic resonance imaging functionalities. Such theranostics micelles exhibited noticeable accelerated DOX release with elevated temperature and high proton relaxivity r₁ (25.88 mM⁻¹ s⁻¹). Moreover, the Gd³⁺/DOX loaded micelle presented a synergistically enhanced cellular uptake efficiency (2.12 times that of free DOX) as compared with free drug or that loaded in micelles without Gd³⁺. The micelles provide promising application for chemotherapy and simultaneous enhanced magnetic resonance imaging (MRI) of cancer.

---

Introduction

In recent years, theranostic nanocarriers for simultaneous cancer imaging and therapy have attracted increasing attention.¹ Theranostic nanocarriers combine both diagnostic and therapeutic factors/agents into a single nanoparticle and may achieve specific, efficient, and personalized cancer management.^2,3 Therapeutic and diagnostic agents may be loaded in or conjugated to the nanoparticles, endowing the nanoparticle with multi-functionalities. Among the widely used imaging techniques in clinical practice, magnetic resonance imaging (MRI) is one of the most promising modalities in theranostics due to its non-invasiveness and high spatial resolution.^4,5

Longitudinal relaxivity (r₁) and transverse relaxivity (r₂) are applied to evaluate the ability to alter spin–lattice relaxation (T₁) and spin–spin relaxation (T₂) in the MRI, respectively. Paramagnetic complexes such as the T₁-type gadolinium Gd(III) chelate complexes can be harnessed in clinical application to amplify the signals of MRI tomography and improve the contrast between pathologic and normal tissues.⁶ Besides their high coordination constant and stability,⁷ these Gd-based complexes are also characterized by some disadvantages such as low relaxivity, short blood circulation time and low specificity to tissues.⁶ However, when loaded in the nanoscale carrier systems, the tumbling of the systems is slower, resulting in a higher proton longitudinal relaxation rate r₁ and long residential periods in the bloodstream.⁸ A variety of nanoscale carriers, including macromolecules, dendrimers, liposomes, vesicles, and micelles carrying with payload of contrast agents have been developed as MRI contrast agents.⁹ In addition, nanoparticles with defined size enjoy the advantage of long circulation time and are able to passively target tumor tissues through the enhanced permeability and retention (EPR) effect, such increasing the tumor uptake of the payload and making them useful delivery vehicles in cancer therapy.^10,11 In the case of micelles, diagnostic agents are usually conjugated to the shell. Li et al. first reported biodegradable micelles with Gd³⁺ ions chelated to the shell. These micelles exhibited two times higher relaxivity than that of the DTPA–Gd complex.⁹

We have previously reported an amphiphilic copolymer with upper critical solution temperature (UCST) behavior, namely, methoxy-poly(ethylene glycol)-block-poly(acrylamide-co-acrylonitrile) [mPEG-b-poly(AAm-co-AN)], which could self-assemble into polymeric micelles at temperature lower than the critical temperature and dissociate at higher temperature. The UCST of such block copolymers ranged from 43 °C to 65 °C, depending on their molar mass and concentration. Ideally, such thermo-responsive characteristic could be harnessed to fabricate anti-cancer drug carriers that would dissociate and increase the release of encapsulated anti-cancer drug at tumor site, where the temperature is higher than the normal tissues.¹³ However, the terminal of the polymer was end-capped with methoxy groups and will not favor further functionalization to conjugate or load imaging moieties and achieve accurate diagnosis. To this end, the polymer was designed to bear 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) group at its hydrophilic terminal for chelating of Gd³⁺ ions.

In this study, we fabricated micelles from the self-assembly of PEG-b-poly(AAm-co-AN) block copolymers. The hydrophilic PEG terminal was conjugated with the macrocyclic DOTA group to chelate the T₁ contrast agent, namely, Gd³⁺; and the therapeutic drug, DOX, was encapsulated in the core of the micelles for combined tumor imaging and therapy (Scheme 1). The DOX release behavior was studied on different types of micelles at varied temperatures. Cytotoxicity of these micelles against human embryonic hepatocytes (L02) and human hepatic carcinoma (Bel 7402) cells were evaluated. In vitro cellular uptake of different types of DOX-loaded micelles against Bel-7402 cells was investigated. The in vitro MR imaging of Gd³⁺-chelated micelles and micelles taken up by Bel-7402 cells were conducted.

Scheme 1 Schematic illustration of the self-assembly and thermally enhanced drug release of theranostic micelles.

Experimental

Materials

Poly(AAm-co-AN) end-capped with chain transfer agent (CTA-poly(AAm-co-AN)) exhibiting upper critical solution temperature (UCST) were synthesized by reversible addition–fragmentation chain-transfer polymerization according to previous reports,¹³ and poly(ethylene glycol) diacrylate (PEGDA, M_n = 2000 g mol⁻¹) were synthesized with minor modification according to published protocol.¹³ NHS-ester of 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) (NHS-DOTA, Macrocyclics, Texas, USA),^10,14 gadolinium(III) chloride hexahydrate (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), 3-(4,5-dimethyl-thiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT), and doxorubicin hydrochloride (DOX·HCl) (Aldrich, 98%) were used as received. Dichloromethane (CH₂Cl₂), dimethylsulfoxide (DMSO), and N,N′-dimethylformamide (DMF) were of analytical grade and distilled before use.

Human embryonic hepatocytes (L02 cell) and human hepatic carcinoma cell (Bel 7402 cell) were obtained from the Laboratory Animal Centre of Sun Yat-sen University (Guangzhou, China), and the related protocol was approved by the Laboratory Animal Centre of Sun Yat-sen University. Roswell Park Memorial Institute 1640 (RPMI-1640, Hyclone) medium supplemented with 10% of fetal bovine serum (FBS) and 1% of penicillin–streptomycin was used throughout this study.

Synthesis of acryloyl-PEG-b-poly(AAm-co-AN)

Acryloyl-PEG-b-poly(AAm-co-AN) was synthesized via Michael-type reaction between one end of PEGDA and CTA-poly(AAm-co-AN). Briefly, 0.1 mmol of CTA-poly(AAm-co-AN) and excessive PEGDA (1 mmol) were dissolved in 15 mL of DMF in a Schlenk flask. The mixture solution was degassed and added with 6 mmol of n-hexylamine and 2.67 mmol of triethylamine (TEA) to cleave the trithiol group and allow the Michael-type addition to proceed. The reaction mixture was stirred at 50 °C for 48 h, the product was collected via precipitation in cold methanol. Then the crude product was dissolved in small amount of DMF and dialyzed against deionized (DI) water for 72 hours in a dialysis tube (MWCO 7000 Da) and lyophilized.

Synthesis of cysteine-PEG-b-poly(AAm-co-AN)

0.2 mmol of cysteine and 0.1 mmol of acryloyl-PEG-b-poly(AAm-co-AN) were dissolved in 10 mL of DI water in a Schlenk flask. Then the mixture solution was degassed, to this solution was added 6 mmol of n-hexylamine and 2.67 mmol of triethylamine (TEA) through a syringe. The reaction was carried for 24 h at 25 °C. After reaction, the reaction mixture was then dialyzed (MWCO 7000 Da) against DI water for 72 hours and lyophilized.

Synthesis of DOTA-PEG-b-poly(AAm-co-AN)

0.2 mmol of DOTA-NHS and 0.1 mmol of cysteine-PEG-b-poly(AAm-co-AN) was dissolved in 2 mL of DMSO, then 30 μL of TEA was added to this solution. The mixture was stirred for 24 h at 25 °C. The product was dialyzed against DI water for 72 hours and lyophilized.

Preparation of Gd³⁺/DOX loaded micelles

The Gd³⁺/DOX loaded polymer micelles was prepared by a solvent switching method. DOX·HCl (2.0 mg, 3.4 μmol) and TEA (35 μL, 44 μmol) were dissolved in 500 μL of DMSO, and the mixture solution was incubated for 7 h in the dark. Then DOTA-PEG-b-poly(AAm-co-AN) (0.01 mmol, 20.0 mg) in 500 μL of DMSO was added to this mixture solution. Subsequently, the mixture solution was added drop wise to 3 mL of aqueous solution containing 2 mg of GdCl₃ under sonication (amplitude: 160 μm; 3 s on/3 s off) for 100 duty cycles. The solution was then transferred into a dialysis tube (MWCO 7000 Da) and dialyzed against 1 mM ethylenediaminetetraacetic acid (EDTA) solution for 48 hours to remove DMSO and free Gd³⁺, and another 24 hours' dialysis against pure water was performed to remove EDTA.

The drug-loading efficiency (DLE) and drug encapsulation efficiency (DEE) was determined using an UV-vis spectrophotometer according to reported protocol.¹³ Briefly, 300 μL of the dialysate was lyophilized and re-dissolved in 3 mL of DMSO to measure the absorbance at 480 nm. The mass of encapsulated DOX in the DMSO was then derived from established standard curve.

The DLE and DEE were calculated according to the following equations:

The amount of Gd³⁺ of the micelles was quantified by inductively coupled plasma optical emission spectroscopy (ICP-OES). Briefly, 10 mg of powder of the lyophilized micelles was digested in 69% HNO₃ at 120 °C, HNO₃ was continuously added until the solution turned grey, then the solution was evaporated to dryness; the solute was dissolved in 10 mL of 5% HNO₃ and subjected to ICP-OES measurement, and the content of Gd³⁺ of the micelles as found to be 0.274 wt%.

Critical micelle concentration

Pyrene solution in acetone was evaporated to obtain a certain mass of pyrene. To this evaporation residue was added DOTA-PEG-b-poly(AAm-co-AN) solution in DI water at different concentration to reach the final pyrene concentration of 6.0 × 10⁻⁷ M. The solution was then incubated under gentle stirring for 24 h. Fluorescence emission spectra of the solutions were carried on a FluoroMax-4 (HORIBA Jobin Yvon, France) spectrometer. The pyrene/micelle solution was excited at 335 nm and the emission spectrum in the range of 350–440 nm was recorded. The ratio of the fluorescence intensity at 393 nm to that at 373 nm (I₃₉₃/I₃₇₃) was derived to determine the critical micelle concentration.

Dynamic light scattering

The hydrodynamic diameter and zeta potential of micelles were determined on a Malvern Zetasizer (ZS 90, UK) at a fixed scattering angle of 90°. The results of each measurement were presented as intensity mean ± standard deviation.

Transmission electron microscopy (TEM)

The size and morphology of micelles were observed on a JEM1400 instrument (120 kV). 10 μL of the micelle suspension was dropped onto the surface of 400 mesh carbon-coated copper grids, then stained with phosphotungstic acid aqueous solution (0.05 wt%) for one minute and dried in air before observation.

In vitro drug release

DOX loaded micelles was dialyzed for 72 hours at room temperature to remove free DOX prior to the release study. Then, 3 mL of the dialysate in a dialysis tube (MWCO 14 000 Da) was immersed into 10 mL of phosphate buffered saline (PBS) at 37 °C. 3 mL of the external buffer was taken and replenished with the same amount of fresh PBS at predetermined time points. The UV intensities of the release media was measured on UV spectrophotometer at 480 nm. The cumulative release percentage was defined by the following equation:
C_n, concentration of drug of the external phase at each time point (μg mL⁻¹); V, volume of the external phase (mL); C_x, concentration of drug in the x-th sample solution taken out (μg mL⁻¹); V′, volume of each sample solution (mL); m, mass of the initial drug (μg).¹⁵

Cytotoxicity

In vitro cytotoxicity of the micelles against L-02 cells and Bel-7402 cells was evaluated by MTT assay. Cells were seeded at initial density of 5000 cells per well in 96-well plate for 24 h. Then the culture medium was removed and 200 μL of corresponding culture medium containing nanoparticles with varied concentrations (10 μM–60 μM) was added and incubated with the cells for 24 hours. RPMI-1640 medium was used as control. After incubation, the wells were washed with PBS and refilled with 100 μL of RPMI-1640. Subsequently, 20 μL of MTT solution (5 mg mL⁻¹) was added to each cell and incubated for another 4 h and aspirated, then 150 μL of DMSO was added to each well to dissolve the formazan, and the absorbance was measured at 570 nm on a Synergy 4 microplate reader (BioTek).

Half maximal inhibitory concentration (IC50)

The anti-tumor efficiency of DOX loaded micelles was evaluated in term of the IC₅₀ value. Bel-7402 cells were seeded at initial density of 5000 cells per well in 96-well plate and incubated at 37 °C in 5% CO₂ atmosphere and 95% humidity for 24 h. Then the culture medium was removed and 200 μL of micelle solution in culture medium was added to each well and cultured for 24 hours at varied DOX dosages (0.024 μg mL⁻¹ to 75 μg mL⁻¹). Free DOX at the same concentrations were used as control. After incubation, the wells were washed with PBS and refilled with 100 μL of RPMI-1640 medium. Subsequently, 20 μL of MTT solution (5 mg mL⁻¹) in RPMI-1640 medium was added to each cell and incubated for additional four hours. The MTT solution was aspirated and each well was refilled with 150 μL of DMSO to dissolve the formazan, and the absorbance of each well at 570 nm was read on a Synergy 4 microplate reader (BioTek, USA). The value of IC₅₀ was acquired by inquire to the confidence limit generated by probit fitting in SPSS (IBM, USA).

Cellular uptake

DOX-loaded micelles in 500 μL of culture medium with concentration of DOX of 5 μg mL⁻¹ were added to Bel-7402 cells (5 × 10⁴ cells per well in a 24 well plate) and incubated for two hours. Cells was washed with PBS three times and imaged under IX71-inverted fluorescent microscope (Olympus, Japan). The cellular uptake was also quantified via flow cytometry analysis. Bel-7402 cells were seeded at a concentration of 2 × 10⁵ cells per well in a 24-well plate. Free DOX and DOX loaded micelle at DOX concentration of 10 μg mL⁻¹ were added to each well and incubated for 2 h. Then the medium was removed and the cells were washed three times with fresh PBS. The cells were trypsinized, collected, and re-suspended in 0.5 mL of PBS before being subjected to measurement. The mean fluorescence was measured on a FACS Calibur flow cytometer (Becton Dickinson, San Jose, CA, USA).

In vitro MR imaging

Diluted solutions of micelles were prepared from the original Gd-loaded micelles with initial concentration of micelles of 4.75 mg mL⁻¹ and concentration of Gd³⁺ of 0.074 mM. Samples with Gd³⁺ concentrations of 0.074, 0.0370, 0.0185, 0.0092, 0.0046, and 0.0023 mM were prepared and transferred to 2 mL centrifuge tubes, named sample 1–6, respectively. Spin–lattice (T₁) was measured at a MAGNETOM Verio 3T MRI system (SIEMENS AG FWB) using inversion-recovery and multi-echo pulse sequences on a knee joint coil. The MR images from the top view were acquired. Relaxivities (r₁ in mM⁻¹ S⁻¹) were obtained using region of interest (ROI) analysis of the acquired images.

The in vitro MRI of Gd³⁺ chelated micelles that taken up by Bel-7402 cells were conducted. Bel-7402 cells were seeded at a density of 2 × 10⁶ cells per well in a 6-well plate. Cell culture medium containing different concentrations of Gd³⁺ chelated micelles were added to each well and cultured for two hours. Cells cultured with culture medium only were set as control. Then the medium was removed and the cells were washed three times with fresh PBS to remove the micelles that were not taken up by the cells. The cells were collected in 1.5 mL centrifuge tubes and re-suspended in 0.1 mL of PBS at a density of 2 × 10⁷ cells per mL. Then the side view MR images were acquired on a MAGNETOM Verio 3T MRI system using the parameter mentioned above.

Results and discussion

Synthesis of diblock copolymer

CTA-poly(AAm-co-AN) bears phenyl group on one end and trithiocarbonate group on the other end with different molar masses (8 × 10³, 12 × 10³, and 16 × 10³ g mol⁻¹) was synthesized. GPC curves of the polymers indicated their unimodal and narrow distribution (Fig. S1†). The chemical structure of the acryloyl-PEG-b-poly(AAm-co-AN) was characterized by ¹H-NMR. The chemical shift at 1.30–2.71 ppm can be assigned to the methylene protons (–CH̲₂–CH–) in the backbone of poly(AAm-co-AN) segment, and signals at 6.68–7.84 ppm belong to phenyl and amide protons. Strong chemical shift at 3.52 ppm can be attributed to the ethylene oxide repeating unit (–CH̲₂–CH̲₂–O–) of the PEG segment (Fig. S2a†). After completion of the thiol–ene reaction of cysteine with the acryloyl-PEG-b-poly(AAm-co-AN), the peaks at 5.75, 6.03, and 6.11 ppm in the ¹H NMR spectrum of the product, which belong to the acryloyl groups, were no longer visible, suggesting the successful conjugation of cysteine to the acryloyl-PEG. Then DOTA-NHS ester was conjugated to the amino group of the cysteine, and the appearance of methylene peaks at 2.90 and 2.95 ppm. (Fig. S2b†), which belongs to methylene protons of DOTA, indicated that DOTA was conjugated to the terminus of the PEG-b-poly(AAm-co-AN).

Fabrication and characterization of micelles

Hydrophobic–hydrophilic balance plays a vital role in the self-assembly of amphiphilic block copolymers. The thermodynamic stability of the micelles in aqueous solution was evaluated in term of the critical micelle concentration (CMC). Increasing the length of the hydrophobic block in the block copolymer would alter the hydrophobicity–hydrophilicity balance of the micelles, leading to noticeable decrease in the CMC value.¹⁶ In our case, the CMC values of PEG_2k-b-poly(AAm-co-AN)_8k, PEG_2k-b-poly(AAm-co-AN)_12k, and PEG_2k-b-poly(AAm-co-AN)_16k were determined to be 13.2 μg mL⁻¹, 5.5 μg mL⁻¹, and 2.4 μg mL⁻¹, respectively (Fig. S3†), suggesting the increase in the length of the hydrophobic poly(AAm-co-AN) block led to the decrease of the CMC value. In addition to the length of hydrophobic block, the conjugation of hydrophilic DOTA group on the hydrophilic end of the block copolymer also affect the hydrophilicity–hydrophobicity balance, the CMC value of the DOTA-PEG_2k-b-poly(AAm-co-AN)_8k (26.5 μg mL⁻¹) was twice of that of CTA-PEG_2k-b-poly(AAm-co-AN)_8k; on the other hand, the steric hindrance of DOTA group may possibly hinder the self-assembly of the block copolymers, leading to increased CMC value.¹⁷

Due to the hydrophobicity of DOX, the hydrophobic interaction between DOX and the hydrophobic segment of the amphiphilic block copolymer may be a governing factor in the formation of micelles with high drug loading content; and it has been reported that higher loading content of DOX could be achieved in the presence of longer hydrophobic block length of the amphiphilic block copolymer.¹⁸ For micelles fabricated from PEG_2k-b-poly(AAm-co-AN)_8k, PEG_2k-b-poly(AAm-co-AN)_12k, and PEG_2k-b-poly(AAm-co-AN)_16k, the DLE values were found to be 3.88%, 4.55%, and 4.89%, respectively (Table 1); obviously, longer hydrophobic segment would facilitate its hydrophobic interaction with DOX and lead to elevated DLE. As compared with that of micelles from PEG_2k-b-poly(AAm-co-AN)_8k, the DLE of micelles from DOTA-PEG_2k-b-poly(AAm-co-AN)_8k was found to be 2.89%, which is lower than that of PEG_2k-b-poly(AAm-co-AN)_8k micelles. This could be probably attributed to the DOTA group on the shell of the micelles. Upon the introduction of Gd³⁺ ions in the process of fabrication of micelles, the resulting Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles showed increased DLE of 4.0%, we hyperthesize that Gd³⁺ ions may affect the core structure of the micelle via coordination with AN groups in the core and help increase the DLE.

Table 1 Characterizations of micelles

Polymer	DOX loaded micelles
	DLE (wt%)	Size (nm)	PDI	ζ potential
PEG2k-b-poly(AAm-co-AN)8k	3.88%	222.2 ± 4.1	0.10 ± 0.04	−31.6 ± 0.66
PEG2k-b-poly(AAm-co-AN)12k	4.55%	97.2 ± 0.5	0.19 ± 0.01	−20.85 ± 1.34
PEG2k-b-poly(AAm-co-AN)16k	4.89%	80.2 ± 0.5	0.18 ± 0.03	−21.65 ± 0.63
DOTA-PEG2k-b-poly(AAm-co-AN)8k	2.89%	262.7 ± 3.1	0.12 ± 0.03	−26.2 ± 1.12
Gd^3+/DOTA-PEG2k-b-poly(AAm-co-AN)8k	4.0%	129.7 ± 3.2	0.10 ± 0.01	−21.6 ± 0.43

---

The size of the DOX-loaded micelles also was dependent on the length of hydrophobic block (Table 1). With the molar mass of the poly(AAm-co-AN) blocks of 8000, 12 000, and 16 000 g mol⁻¹, the hydrodynamic diameters of corresponding micelles were 222.2 ± 4.1, 97.2 ± 0.5, and 80.2 ± 0.5 nm, respectively; there was no doubt that the longer hydrophobic segment would result in micelles with smaller size. However, the conjugation of DOTA group onto the hydrophilic end of the block copolymer (PEG_2k-b-poly(AAm-co-AN)_8k) caused an increase in the size of micelles (262.7 ± 3.1 nm), this could probably be attributed to the steric hindrance between DOTA groups that would prevent the polymer chains to form compact self-assembly. Further coordination of Gd³⁺ ions with DOTA groups led to a much smaller size of micelles, this could partially be due to the reduced hydrophilicity of the hydrophilic block and the enhanced hydrophobicity of the core through coordination of Gd³⁺ ions with the poly(AAm-co-AN) segment.

The morphologies of different micelles were observed under TEM (Fig. 1). The PEG_2k-b-poly(AAm-co-AN)_8k micelles displayed typical core–shell structure with average diameter of approximately 280 nm (Fig. 1a). As compared with that of the PEG_2k-b-poly(AAm-co-AN)_8k micelles, more distinct shell structure was visible with the DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles (Fig. 1b). We hyperthesize that the distinct shell structure may be attribute to the chelation of WO₄²⁻ ion with DOTA groups, which contain carboxyl groups that may complex with tungsten(VI).¹⁹ Subsequent chelation of Gd³⁺ also gave similar distinct core–shell structure micelles with smaller size (Fig. 1c). Upon chelating of Gd³⁺ and loading of DOX, the micelles displayed more intense cores (Fig. 1d).

Fig. 1 TEM images of micelles. (a) PEG_2k-b-poly(AAm-co-AN)_8k, (b) DOTA-PEG_2k-b-poly(AAm-co-AN)_8k, (c) Gd³⁺/DOTA-b-PEG_2k-poly(AAm-co-AN), (d) Gd³⁺/DOX loaded DOTA-PEG_2k-b-poly(AAm-co-AN)_8k (scale bar: 200 nm).

In vitro release of DOX

The in vitro release of DOX from the micelles in PBS (pH 7.4) was evaluated at varied temperatures (Fig. 2). Burst release was observed in the first 24 hours in all types of micelles, while a sustained release lasted till 144 hours. In our previous work, it was found that the PEG-b-poly(AAm-co-AN) micelles exhibited UCST-type phase transition behavior; the temperature responsive release of DOX from PEG_2k-b-poly(AAm-co-AN)_8k micelles was also studied at 14, 37, and 60 °C (Fig. 2a). The drug release rate was accelerated at elevated temperature; and after 144 hours, the cumulative releases of DOX from PEG_2k-b-poly(AAm-co-AN)_8k micelles at different temperatures gradually plateaued off at 12.0%, 31.7%, and 48.9%, respectively, indicating more drug was released at elevated temperature. The thermo-responsive drug release was in consistent with our previously work, in which 23.3% and 47.0% of the encapsulated drug was released within 120 h at 42 °C and 70 °C, respectively.¹³ Interestingly, for the DOX-loaded nanoparticles which were incubated at 14 and 37 °C, a sudden increase in temperature to 60 °C would facilitate additional release; after 216 hours, these three micelles showed cumulative release of DOX of 41.3%, 46.6%, and 52.7%, respectively. Such phenomenon strongly suggested the dissociation/swelling of micelle at elevated temperature and most of the DOX in the core of the PEG-b-poly(AAm-co-AN) micelles could only be released at temperature much higher than their UCST.

Fig. 2 Cumulative release of DOX from different types micelles (concentrations of micelles: approx. 1.40 mg mL⁻¹).

The release of DOX from micelles with different length of hydrophobic block was studied with PEG_2k-b-poly(AAm-co-AN)_8k, PEG_2k-b-poly(AAm-co-AN)_12k, and PEG_2k-b-poly(AAm-co-AN)_16k at 37 °C (Fig. 2b). The cumulative releases of DOX from PEG_2k-b-poly(AAm-co-AN)_8k, PEG_2k-b-poly(AAm-co-AN)_12k, and PEG_2k-b-poly(AAm-co-AN)_16k micelles within 144 hours were 31.7%, 27.5%, and 28%, respectively. It could be explained by the stronger hydrophobic interaction of DOX with longer hydrophobic segment, and the enhanced hydrophobic interaction between the hydrophobic segment and DOX would in turn retard the release of drug from the matrix. Moreover, we hypothesized that the cumulative drug release was not only associated with the hydrophobic length of the block copolymer, but also may be correlated to the UCST of the micelles. In the other word, the swelling and dissociation of the micelles at temperature higher than the UCST may contribute to the enhanced diffusion rate, i.e., the accelerated release of DOX. It was discovered in our previous work that longer hydrophobic block leads to higher UCST of micelles, and the UCSTs of both PEG_2k-b-poly(AAm-co-AN)_12k, and PEG_2k-b-poly(AAm-co-AN)_16k micelles would be higher than 37 °C at the studied concentration, implying that both micelles may not experience significant swelling and dissociation and exhibited comparable cumulative release. A second release of DOX was also recorded with all the three micelles after the temperature was raised up to 60 °C.

The size change of DOX-loaded PEG_2k-b-poly(AAm-co-AN)_8k micelles in term of release time was also evaluated (Fig. S6†). The hydrodynamic diameter of the drug-loaded micelles increased slightly upon the release of encapsulated DOX. This may be explained by the weakened hydrophobic interactions between DOX and hydrophobic core and the core of the micelles became loose. The morphology of the DOX-loaded PEG_2k-b-poly(AAm-co-AN)_8k micelles at different drug release time points was also observed under TEM (Fig. S5†); the dark area in the center of the nanoparticles, which corresponds to the DOX-loaded hydrophobic core of the micelles, became lighter with prolonged release time, suggesting the release of encapsulated drug.

It was quite surprising to observe a 9% higher release of the loaded DOX at 144th hour from the DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles (40.7%) as compared with its non-conjugated counterpart (31.7%). It was noted that DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles exhibited relatively larger diameter than their non-conjugated counterparts, this may probably be attributed to the presence of DOTA group, i.e., the thicker hydrophilic shell of the micelles. Since DOX is a hydrophobic drug, increase in the thickness of the hydrophilic shell of the micelles would inevitably lower down the diffusion rate of DOX through the hydrophilic shell of the micelles; as a consequence, the cumulative release of DOX from the DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles was significantly higher than that from their non-conjugated counterparts even if the later ones contained higher content of DOX. Moreover, a cumulative DOX release of 62.6% from the Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles after 144 hours was obtained (Fig. 2c).

To investigate the effect of Gd³⁺ on the DOX loading in and release from the micelles, DOX-loaded PEG_2k-b-poly(AAm-co-AN)_8k micelles as a control was prepared in the presence of GdCl₃ experiment (Fig. S4†). In addition to its higher DLE (4.41%), about 57.2% of the encapsulated DOX was released within 144 hours, which was comparable to that of the Gd³⁺ chelated micelles (62.6%), and was much higher than that of PEG_2k-b-poly(AAm-co-AN)_8k micelles (30.7%) (Fig. 2d). It was reported that Gd³⁺ ion can coordinate to the CN groups in the poly(AAm-co-AN) segment.²⁰ In this case, the core structure of the micelles may be effected, leading to the variation in drug loading content.

In vitro cytotoxicities and anti-tumor activities

Bel-7402 cells were incubated with free DOX, DOX-loaded PEG_2k-b-poly(AAm-co-AN)_8k micelles, DOX-loaded DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles, and Gd³⁺/DOX loaded DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles. The characteristic red fluorescence in the cells under demonstrated the effective cellular uptake of DOX, and all these groups incubated with the DOX-loaded micelles showed significantly higher fluorescence intensity than that of free DOX group at the same DOX concentration (5 μg mL⁻¹) (Fig. 3b). Drug-loaded micelles are known to enter the cells via endocytosis with much higher cellular uptake efficiency as compared with free drug that enters cell via free diffusion.²¹ Although these micelles only release may it was also noticed that the DOX-loaded PEG_2k-b-poly(AAm-co-AN)_8k micelles exhibited comparable fluorescence intensity to that of the DOX-loaded PEG_2k-PLGA_16k micelles, indicating the PEG_2k-b-poly(AAm-co-AN)_8k micelles have similar performance to the extensively studied nano-carriers (Fig. 3a).

Fig. 3 Cellular uptake by Bel-7402 cells of (a1) PEG_2k-b-poly(AAm-co-AN)_8k micelles, (a2) PEG_2k-b-PLGA_16k micelles, (b1) free DOX, (b2) PEG_2k-b-poly(AAm-co-AN)_8k, (b3) DOTA-PEG_2k-b-poly(AAm-co-AN)_8k, and (b4) Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles (DOX concentration 5 μg mL⁻¹).

The fluorescence images of cellular uptake were analyzed using Image Pro Plus software. The integrated optical density of groups including the free DOX, PEG_2k-b-poly(AAm-co-AN)_8k micelles, PEG_2k-b-PLGA_16k micelles, DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles, and Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles was 765, 2205, 1485, 2010 and 2209, respectively. PEG_2k-b-poly(AAm-co-AN)_8k and Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles showed the highest optical density, suggesting the PEG_2k-b-poly(AAm-co-AN)_8k and Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles could be more easily taken up than other micelles.

Flow cytometry was employed to quantify the efficiency of cellular uptake of different DOX-loaded micelles compared with free DOX. All the DOX-loaded micelles showed enhanced uptake than free DOX. The DOX-loaded PEG–PLGA micelles showed a fluorescence intensity of 1.14 times of that of free DOX. The fluorescence intensities of the PEG_2k-b-poly(AAm-co-AN)_8k, Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles, PEG_2k-b-poly(AAm-co-AN)_12k micelles were found to be 1.49, 2.12, and 1.64 times of that of free DOX, respectively, exhibiting much higher fluorescence intensity than that of the PEG–PLGA micelles. The fluorescence intensity of the PEG_2k-b-poly(AAm-co-AN)_8k group was comparable to that of mPEG_2k-b-poly(AAm-co-AN)_8k, which was 1.43 times of free DOX.¹³ This may be explained by their similarity in structure. The fluorescence intensities of DOTA-PEG_2k-b-poly(AAm-co-AN)_8k and PEG_2k-b-poly(AAm-co-AN)_16k micelles were 1.17 and 1.08 times of that of free DOX, and are comparable to that of the PEG–PLGA micelles. Interestingly, the Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles outperformed other micelles in cellular uptake (Fig. 4). It was believed that cells prefer to attach to more hydrophobic substrate, or to particle with the same hydrophilic configuration but smaller size.¹² In this case, the chelation of Gd³⁺ ions on the surface of micelles may not only reduced the hydrophilicity of the micelles, but also resulted in much smaller size as compared with their counterparts, leading to the synergistically enhanced cellular uptake.

Fig. 4 Flow cytometric analysis of BEL-7402 cells after 2 h incubation with (a–c) PEG_2k-b-poly(AAm-co-AN)_8k, DOTA-PEG_2k-b-poly(AAm-co-AN)_8k, Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles, (d) free DOX, (e) PEG_2k-b-PLGA_16k micelles, (f and g) PEG_2k-b-poly(AAm-co-AN)_12k, PEG_2k-b-poly(AAm-co-AN)_16k micelles, and (h) blank (DOX concentration 10 μg mL⁻¹).

In the case of Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_16k and its counterpart, similar phenomenon was observed to that of Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k (Fig. S8†); such findings indicated unambiguously that the chelation of Gd³⁺ ions may not only help reduce particle size but also facilitate the cellular uptake. The actual amount of DOX taken up by BEL-7402 cells may vary slightly because of the different rates of release of the drug from different micelles. But the effect would be minor in a short release time of two hours. During which less than 14% of the loaded drug was released. One can conclude the same tendency even if takes the difference in the release behavior into consideration.

The cytotoxicity of PEG_2k-b-poly(AAm-co-AN)_8k, DOTA-PEG_2k-b-poly(AAm-co-AN)_8k, and Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles at concentrations from 10 μM to 60 μM against L02 cells (Fig. 5a) and Bel-7402 cells (Fig. 5b) were evaluated via MTT assay. The corresponding concentrations of Gd³⁺ ion in the case of Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles were 0.00188 mM, 0.00376 mM, 0.00563 mM, 0.00751 mM, 0.00939 mM, and 0.0113 mM, respectively. The PEG_2k-b-poly(AAm-co-AN)_8k micelles was found to be compatible to both cell types. The DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles and Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles showed moderate toxicity to the cells at polymer concentration up to 60 μM, with cell viability of above 80% (Fig. 5b). The unanimous tendency of the cytotoxicity of these two types of micelles suggest that Gd³⁺ were tightly chelated to the DOTA moieties and did not significantly release from the micelles and cause severe toxicity to cells.²²

Fig. 5 Cytotoxicity of micelles toward (a) L-02 cells and (b) BEL-7402 cells; and (c) the IC₅₀ of DOX-loaded micelles on BEL-7402 cells.

The IC₅₀ value of the DOX-loaded micelles on the Bel-7402 cells was derived to evaluate their in vitro antitumor efficacies (Fig. 5c). The IC₅₀ values increased from 1.51 μg mL⁻¹ for free DOX to 7.16, 12.75, and 13.94 μg mL⁻¹ for Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k, DOTA-PEG_2k-b-poly(AAm-co-AN)_8k, and PEG_2k-b-poly(AAm-co-AN)_8k micelles, respectively. Generally, the DOX-loaded micelles can effectively suppress the viability of Bel-7402 cells at certain concentration of DOX. However, free DOX was found to impede the proliferation with higher efficiency. This can be partially explained by the much lower DOX concentration in the medium since the amount of DOX released from the micelles was inevitably lower than free DOX at the same initial nominal DOX concentration. It should also be noted that the cell viability in the presence of Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles was much lower than the other micelles, which could be attributed to the higher cumulative release of DOX at the same time period.

In vitro magnetic resonance imaging

It is well-known that Gd-based contrast agents can alter the spin–lattice relaxation and considerably improve the diagnostic sensitivity toward malignant tissues over normal ones.²³ In this case, DOTA groups conjugated on the surface of the micelles could effectively chelate the Gd³⁺ ion, leading to the improved positive MR imaging contrast enhancement. In vitro MR experiment of DOX-loaded Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k solution (Fig. 6a) reveals that, with the increase of concentration of Gd³⁺ from 0.0023 mM to 0.074 mM, the gradual positive contrast enhancement of MR signals was evidenced by the reinforcement of the spot brightness. The transverse relaxivity at 3.0 T of the Gd³⁺/DOX-loaded DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles reached 25.88 mM⁻¹ s⁻¹, demonstrating its potential as a desirable contrast agent in MR imaging.

Fig. 6 (a) In vitro MRI of Gd³⁺/DOX loaded PEG_2k-b-poly(AAm-co-AN)_8k micelle with varied concentrations (Gd³⁺ concentration from 1–6: 0.074, 0.0370, 0.0185, 0.0092, 0.0046, and 0.0023 mM). (b) In vitro MRI of BEL-7402 cell suspensions (cell density: 2 × 10⁷ cells per mL).

The MR image of the Bel-7402 cell suspension after internalization of the Gd³⁺/DOX-loaded PEG_2k-b-poly(AAm-co-AN)_8k micelles was also obtained to evaluate its applicability in cell imaging (Fig. 6b). Cells incubated with varied concentrations of nanoparticle (b1–b5) showed significant signal enhancement as compared with blank cells. It was demonstrated that Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles can easily enter the cells within two hours for enhanced MR imaging and can be further applied in vivo.

The above findings on both inhibitory activity on the carcinoma cells and MR imaging capability of the micelles unambiguously demonstrated the efficacy of these novel micelles as theranostic nano-carrier in potential clinical applications. However, it is still a major challenge to fine tune the UCST behavior of such micelles to achieve optimal curing effect. We believe the studies present here shall be a start and it lays the foundation for in-depth studies and in vivo research.

Conclusions

In summary, we have successfully prepared a biocompatible polymeric micelle based on DOTA-PEG_2k-b-poly(AAm-co-AN)_8k diblock copolymer, T₁-weighed MRI contrast agent and DOX as the anticancer drug was loaded in this micelle. Temperature responsive drug release was observed. The Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles displayed significantly higher cumulative release of DOX of 62% within 144 h as well as exhibiting superior cellular uptake efficiency to that of free DOX. High proton relaxivity r₁ (25.88 mM⁻¹ s⁻¹) was achieved in vitro and BEL-7402 cells that took up Gd³⁺/DOTA-PEG_2k-b-poly(AAm-co-AN)_8k micelles also showed enhanced MR imaging. This theranostic micelle showed great potential for chemotherapy and simultaneous enhanced magnetic resonance imaging (MRI) of cancer.

Acknowledgements

The authors are grateful to the financial support from the Program for New Century Excellent Talents in University of the Ministry of Education of China (Grant No. NCET-09-0818), the Guangdong Provincial Natural Science Foundation (Grant No. 2014A030311018), the Program for Industry, University & Research Institute Collaboration of Guangdong Province (Grant No. 2012B091100452), the Science and Technology Planning Project of Guangdong Province (Grant No. 2011A060901013, 2015B010125004), the Guangdong Innovative Research Team Program (Grant No. 2009010057), the Science and Technology Program of Guangzhou (Grant No. 201505041545195).

Notes and references

1. H. Yang, H. Mao, Z. Wan, A. Zhu, M. Guo, Y. Li, X. Li, J. Wan, X. Yang, X. Shuai and H. Chen, Micelles assembled with carbocyanine dyes for theranostic near-infrared fluorescent cancer imaging and photothermal therapy, Biomaterials, 2013, 34(36), 9124–9133.
2. F. Chen, E. B. Ehlerding and W. Cai, Theranostic nanoparticles, J. Nucl. Med., 2014, 55(12), 1919–1922.
3. Q. Liu, S. Chen, J. Chen and J. Du, An asymmetrical polymer vesicle strategy for significantly improving T₁ MRI sensitivity and cancer-targeted drug delivery, Macromolecules, 2015, 48(3), 739–749.
4. Y. Liu and N. Zhang, Gadolinium loaded nanoparticles in theranostic magnetic resonance imaging, Biomaterials, 2012, 33(21), 5363–5375.
5. J. Qin, Q. Liu, J. Zhang, J. Chen, S. Chen, Y. Zhao and J. Du, Rationally separating the corona and membrane functions of polymer vesicles for enhanced T₂ MRI and drug delivery, ACS Appl. Mater. Interfaces, 2015, 7(25), 14043–14052.
6. P. Mi, H. Cabral, D. Kokuryo, M. Rafi, Y. Terada, I. Aoki, T. Saga, I. Takehiko, N. Nishiyama and K. Kataoka, Gd–DTPA-loaded polymer–metal complex micelles with high relaxivity for MR cancer imaging, Biomaterials, 2013, 34(2), 492–500.
7. W. Mier, J. Hoffend, S. Kramer, J. Schuhmacher, W. E. Hull, M. Eisenhut and U. Haberkorn, Conjugation of DOTA using isolated phenolic active esters: the labeling and biodistribution of albumin as blood pool marker, Bioconjugate Chem., 2005, 16(1), 237–240.
8. K. Shiraishi, K. Kawano, Y. Maitani and M. Yokoyama, Polyion complex micelle MRI contrast agents from poly(ethylene glycol)-b-poly(L-lysine) block copolymers having Gd-DOTA; preparations and their control of T(1)-relaxivities and blood circulation characteristics, J. Controlled Release, 2010, 148(2), 160–167.
9. G. Zhang, R. Zhang, X. Wen, L. Li and C. Li, Micelles based on biodegradable poly(L-glutamic acid)-b-polylactide with paramagnetic Gd ions chelated to the shell layer as a potential nanoscale MRI-visible delivery system, Biomacromolecules, 2008, 9(1), 36–42.
10. A. I. Jensen, T. Binderup, P. Kumar Ek, A. Kjær, P. H. Rasmussen and T. L. Andresen, Positron emission tomography based analysis of long-circulating cross-linked triblock polymeric micelles in a U87MG mouse xenograft model and comparison of DOTA and CB-TE2A as chelators of copper-64, Biomacromolecules, 2014, 15(5), 1625–1633.
11. Y. Li, J. Ma, H. Zhu, X. Gao, H. Dong and D. Shi, Green synthetic, multifunctional hybrid micelles with shell embedded magnetic nanoparticles for theranostic applications, ACS Appl. Mater. Interfaces, 2013, 5(15), 7227–7235.
12. Y. Hu, J. Xie, Y. W. Tong and C. H. Wang, Effect of PEG conformation and particle size on the cellular uptake efficiency of nanoparticles with the HepG2 cells, J. Controlled Release, 2007, 118(1), 7–17.
13. G. Huang, H. Li, S.-T. Feng, X. Li, G. Tong, J. Liu, C. Quan, Q. Jiang, C. Zhang and Z. Li, Self-assembled UCST-type micelles as potential drug carriers for cancer therapeutics, Macromol. Chem. Phys., 2015, 216(9), 1014–1023.
14. W. Mier, J. Hoffend, S. Krämer, J. Schuhmacher, W. E. Hull, M. Eisenhut and U. Haberkorn, Conjugation of DOTA using isolated phenolic active esters: the labeling and biodistribution of albumin as blood pool marker, Bioconjugate Chem., 2004, 16(1), 237–240.
15. Y. Cong, C. Quan, M. Liu, J. Liu, G. Huang, G. Tong, Y. Yin, C. Zhang and Q. Jiang, Alendronate-decorated biodegradable polymeric micelles for potential bone-targeted delivery of vancomycin, J. Biomater. Sci., Polym. Ed., 2015, 26(11), 629–643.
16. Y. J. Li, J. Reeve, Y. L. Wang, R. K. Thomas, J. B. Wang and H. K. Yan, Microcalorimetric study on micellization of nonionic surfactants with a benzene ring or adamantane in their hydrophobic chains, J. Phys. Chem. B, 2005, 109(33), 16070–16074.
17. Y. He, D. Esteban-Fernández and M. W. Linscheid, Novel approach for labeling of biopolymers with DOTA complexes using in situ click chemistry for quantification, Talanta, 2015, 134, 468–475.
18. X. Zhai, W. Huang, J. Liu, Y. Pang, X. Zhu, Y. Zhou and D. Yan, Micelles from amphiphilic block copolyphosphates for drug delivery, Macromol. Biosci., 2011, 11(11), 1603–1610.
19. K. Majlesi and N. Hajali, Complexation of tungsten(VI) with methyliminodiacetic acid at different ionic strengths, J. Solution Chem., 2012, 41(11), 1889–1905.
20. R. Koner, M. G. B. Drew, A. Figuerola, C. Diaz and S. Mohanta, A new cyano-bridged one-dimensional GdIIIFeIII coordination polymer with o-phenanthroline as the blocking ligand: synthesis, structure, and magnetic properties, Inorg. Chim. Acta, 2005, 358(11), 3041–3047.
21. H. S. Yoo and T. G. Park, Biodegradable polymeric micelles composed of doxorubicin conjugated PLGA–PEG block copolymer, J. Controlled Release, 2001, 70(1–2), 63–70.
22. S. Zhang, Z. Jiang, X. Liu, L. Zhou and W. Peng, Possible gadolinium ions leaching and MR sensitivity over-estimation in mesoporous silica-coated upconversion nanocrystals, Nanoscale, 2013, 5(17), 8146–8155.
23. P. Mi, D. Kokuryo, H. Cabral, M. Kumagai, T. Nomoto, I. Aoki, Y. Terada, A. Kishimura, N. Nishiyama and K. Kataoka, Hydrothermally synthesized PEGylated calcium phosphate nanoparticles incorporating Gd–DTPA for contrast enhanced MRI diagnosis of solid tumors, J. Controlled Release, 2014, 174, 63–71.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra04038a

---