Novel cyclic azobenzene-containing vesicles: photo/reductant responsiveness and potential applications in colon disease treatment

Abstract

A cyclic azobenzene was chosen as a pendant in the hydrophobic segments of amphiphilic copolymers, and novel nanoparticles were constructed with dual photo and reductant responsiveness. To investigate the topological effects of the cyclic azobenzene architecture on the properties, an analogue of the amphiphilic copolymer with linear azobenzene units was also synthesized. Two kinds of amphiphilic copolymers with cyclic azobenzene (PEG₄₅-b-PCAzo₁₇) and linear azobenzene pendants (PEG₄₅-b-PLAzo₁₉) assembled into stable vesicles in PB solution (pH = 7.4) and also show unique dual sensitivities to ultraviolet radiation and dithionate derivatives. We have investigated the differences of the vesicles obtained from the two kinds of copolymers in the encapsulation and release of Nile Red (NR) or doxorubicin (DOX). The vesicles formed by PEG₄₅-b-PCAzo₁₇ exhibited higher drug loading content and better reversibility of NR fluorescence variation under alternating irradiation with UV/Vis light than those formed by PEG₄₅-b-PLAzo₁₉; meanwhile, neither disruption of the vesicles nor leakage of Nile Red molecules was detected. Moreover, the CAzo-containing vesicles showed a higher release rate and larger release amount of DOX molecules from the membrane under the reduction of Na₂S₂O₄. Because dithionites can act as a mimic of azoreductase, the amphiphilic copolymer with cyclic azobenzene revealed competitive performance as a drug carrier for colon disease treatment.

---

Introduction

In the past few decades, stimuli-responsive systems for drug delivery have received tremendous attention from scientists, especially in the biomedicine field,^1–4 where nanoparticles must not only be nontoxic to cells, but also may be sensitive to specific endogenous or external stimuli, such as light with different wavelengths, lowered interstitial pH, reductive/oxidative conditions or an increased level of certain enzymes.^5,6

Recently, in order to improve in vitro and/or in vivo drug release profiles and performances, researchers are attempting to prepare novel polymeric nanoparticles that respond to dual or multi signals such as redox/pH,⁷ temperature/enzyme,⁸ pH/temperature/redox,⁹ etc.^10–12 Also, more remarkably, amphiphiles with functional diblock or triblock copolymers have become progressively important for the development of stimuli-responsive nanostructures thanks to their easily tunable chemical structures/properties and smart self-assembly. Furthermore, the introduction by an easy synthesis process of multi-responsive chemical species into polymer chains, such as DMAEMA and methyl viologen (MV) segments, that are temperature/pH responsive and redox/guest responsive,^13–15 has addressed a number of important issues.

Here, we note that azobenzene-based monomers also can be used to fabricate multi-responsive biomedical materials. Azobenzenes are well-known photoresponsive chromophores that show unique sensitivity to light (or heat), which leads to reversible trans–cis photoisomerization and dramatic changes in their shapes and dipole moments.¹⁶ Moreover, azobenzene is sensitive to certain reductants or enzymes due to its ability to undergo cleavage of the azo bond upon treatment with hydrazine, Na₂S₂O₄ or bacterial azoreductase existing in the colon.^17–20 These versatilities have made azobenzene-based molecules into a powerful tool for fabricating new types of materials for a wide range of applications including fluorescent probes,²¹ therapy of colon diseases^22,23 and degradable materials.^24,25 Gao et al. have fabricated a new kind of multiple responsive vesicle via host–guest complexation between cyclodextrin and azobenzene-grafted polyglycidyl methacrylates; they then investigated its cargo encapsulation/release behavior in the presence of Na₂S₂O₄.¹⁸ Recently, Gillies et al. have synthesized a library of azobenzenes with electron-withdrawing groups and studied their rates of reduction with hydrazine. Moreover, azobenzene units were incorporated into the copolymers; then, UV light and hydrazine were used together to stimulate the breakdown of the assemblies.²⁶ Additionally, the investigation of cyclic polymers^27–29 and cyclic functional molecules^30–33 has captured increasing interest owing to their more topological configuration, which is useful in the construction of various nanostructures toward manufacturing nanomaterials with unique properties.

In previous studies, we have successfully introduced cyclic azobenzene moieties into polymer skeletons and found that these polymers exhibit unique or enhanced properties compared with those containing linear moieties in the areas of surface relief grating, self-assembled nanostructures and chiroptical switches.^34–38 Considering that introducing the cyclic azobenzene segment into the macrocyclic architecture can endow topological effects on the properties of polymers, it is thus highly desirable to further extend their performance in biological applications.

Herein, we report the synthesis and dual photo/reductant responsiveness of a novel cyclic azobenzene-containing an amphiphilic copolymer, PEG₄₅-b-PCAzo₁₇. For purposes of comparison with the above polymer, an analogue with linear azobenzene units, PEG₄₅-b-PLAzo₁₉, was also prepared successfully according to the synthesis route in Scheme 1. All the chemical structures of the obtained copolymers were characterized by ¹H NMR, GPC and FT-IR (Fig. 1 and S1†). In addition, the differences between the two copolymers with different topological azobenzene segments in fluorescent probes and drug delivery are discussed in detail. The proposed mechanism of the dual photo/reductant responsiveness is shown in Scheme 2. This study may afford a new avenue to tailor stimuli-responsive nanostructures and finally expand the versatility of azobenzene-derived polymers.

Scheme 1 Synthesis of the proposed linear/cyclic azobenzene-containing block copolymer.

Fig. 1 (A) ¹H NMR spectrum of PEG₄₅-b-PCAzo₁₇; (B) ¹H NMR spectrum of PEG₄₅-b-PLAzo₁₉; (C) GPC curves of PEG₄₅-b-PCAzo₁₇ (red dot) and PEG₄₅-b-PLAzo₁₉ (blue dot) prepared via PEG–Br (black line) initiated ATRP.

Scheme 2 Schematic of the dual photo/reductant responsiveness of polymeric vesicles based on linear and cyclic azobenzene segments.

Experimental

Chemicals and materials

Sodium azide (99.5%, Aldrich), 2,2′-dihydroxybiphenyl (98.5%, J&K), 3-ethynylaniline (>98%, Aldrich), 1,6-dibromohexane (analytical reagent, Shanghai Chemical Reagent Co. Ltd), 11-bromo-1-undecanol (98%, J&K), methacryloyl chloride (analytical grade, Shanghai Chemical Reagent Co. Ltd), triethylamine (TEA) (analytical grade, Shanghai Chemical Reagent Co. Ltd.), o-methoxyaniline (98%, Energy), potassium carbonate (K₂CO₃) (≥99%, Shanghai Sinopharm Chemical Reagent Co. Ltd.), α-bromoisobutyryl bromide (98%, Energy), and phenol (97%, Energy) were used as received. CuBr was purified via washing with acetic acid and then dried in vacuum at 40 °C. N,N,N′,N′′,N′′′-pentamethyldiethylenetriamine (PMDETA, 98%; J&K) was distilled under reduced pressure. Poly(ethylene glycol)methyl ether (PEG) with an average molecular weight of ca. 2000 Da was stored in a drying oven before use. Nile Red was purchased from J&K. Doxorubicin hydrochloride (DOX·HCl, 99%) was purchased from Sigma. All other reagents were used without further purification. L929 fibroblast cells were seeded in 24-well plates at a cell density of 1.0 × 10⁴ cells per well and maintained in a medium supplemented with 10% FBS at 37 °C in a humidified atmosphere containing 5% CO₂. Human lung adenocarcinoma A549 cells were routinely grown at 37 °C in RPMI 1640 medium supplemented with 5% fetal calf serum (FCS) under standard conditions of humidity and CO₂ atmosphere.

Synthesis of PEG₄₅-b-PCAzo₁₇

The amphiphilic diblock copolymer of PEG₄₅-b-PCAzo₁₇ was synthesized by atom-transfer radical polymerization (ATRP) as follows. PEG–Br (29.5 mg, 0.012 mmol) and CAzo (160 mg, 0.24 mmol) were dissolved in 1.50 mL of THF and then degassed and filled with argon. CuBr (5.20 mg, 0.036 mmol) and PMDETA (15.0 μL, 0.072 mmol) were successively added. The mixture was degassed by three freeze–pump–thaw cycles, sealed under vacuum, and placed in an oil bath preheated at 80 °C for 18 h. Then the resulting solution was passed through a neutral alumina column to remove the copper catalyst. The filtrate was concentrated to 1 mL and then precipitated into 50 mL of methanol two times. The product was collected and dried in a vacuum oven at 25 °C for 24 h. Yield: 150 mg (75%). M_{n (NMR)} = 13 900 Da, M_{n (GPC)} = 10 800 Da, M_w/M_n = 1.18.

Synthesis of PEG₄₅-b-PLAzo₁₉

PEG₄₅-b-PLAzo₁₉ was synthesized in a manner similar to that of PEG₄₅-b-PCAzo₁₇, with the exception that LAzo was used as the monomer. M_{n (NMR)} = 11 300 Da, M_{n (GPC)} = 12 800 Da, M_w/M_n = 1.14.

The calculation of the number of repeated units of CAzo (m) in PEG₄₅-b-PCAzo_m and LAzo (n) in PEG₄₅-b-PLAzo_n

From the ¹H NMR spectra of the PEG₄₅-b-PCAzo_m in Fig. 1(A), the number of the repeated units (m) of the CAzo blocks could be calculated according to eqn (1).

m = I_8.31/I_3.72–3.46 × 180 (1)

I_8.31: the integrations at 8.31 ppm in ¹H NMR relative to the protons of 1,2,3-triazole, while the integrations at 3.72 to 3.46 ppm belong to the protons on the –OCH₂CH₂O– segments.

From the ¹H NMR spectra of the PEG₄₅-b-PLAzo_n in Fig. 1(B), the number of the repeated units (n) of the LAzo blocks could be calculated according to eqn (2).

n = [(I_7.83 + I_6.95)/8]/I_3.72–3.46 × 180 (2)

I_7.83 and I_6.95: the integrations at I_7.83 and I_6.95 ppm in ¹H NMR relative to the Ar-H protons, while the integrations at 3.72 to 3.46 ppm belong to the protons on the –OCH₂CH₂O– segments.

The self-assembly of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ in phosphate buffered (PB) solution (pH = 7.4)

The amphiphilic copolymers were capable of self-assembly into vesicles in PB solution (pH = 7.4). The typical procedure was conducted as follows: PEG₄₅-b-PCAzo₁₇ was first dissolved in THF with an initial concentration of 0.5 mg mL⁻¹, and dust was removed by filtering through a 0.22 μm filter. Then, PB solution was added gradually until the final water volume fraction reached 50%. The stable vesicles were dialyzed against PB solution to remove the organic solvent using a dialysis membrane (MWCO 3500) for 2 days. PB solution of the vesicles with a concentration around 0.25 mg mL⁻¹ was obtained. The self-assembly of PEG₄₅-b-PCAzo₁₉ was also conducted according to a similar procedure.

Determination of the critical aggregation concentration (CAC)

The critical aggregation concentration of the amphiphilic copolymers was determined using Nile Red as the probe. 20 μL of a solution of Nile Red in acetone (5 × 10⁻⁶ M) was added to a series of bottles, and the solvent was evaporated. Subsequently, 3 mL of PB solution of the vesicles with concentrations ranging from 1.0 × 10⁻³ to 0.125 mg mL⁻¹, prepared by diluting an initial concentration of 0.5 mg mL⁻¹, was added to each flask. After stirring overnight, the fluorescence of these solutions was measured. The emission spectra of Nile Red were registered from 580 to 700 nm with excitation at 550 nm.

Loading of DOX into the vesicle

In order to encapsulate the model drug DOX, firstly, DOX·HCl (5.0 mg) was stirred with TEA (2.4 μL) in 1 mL of DMSO overnight to obtain the DOX base. Then, 60 μL of DOX base solution was added to the fresh vesicle solution and stirred overnight. Afterwards, the mixture was dialyzed against PB solution for 48 h to remove unencapsulated DOX and the organic solvent. The amount of encapsulated DOX in the vesicles was determined using the fluorescence standard curve. First, the calibration curve of DOX/DMSO solutions with different DOX concentrations was obtained by fluorescence measurement (excitation at 480 nm). Then, DOX-loaded vesicles in DMSO were analyzed by fluorescence spectroscopy according to the calibration curve. The drug loading content (DLC) and drug loading efficiency (DLE) were calculated using the following formula:

DLC (wt%) = (weight of loaded drug/(weight of polymer + weight of loaded drug)) × 100%

DLE (wt%) = (weight of loaded drug/weight of drug in feed) × 100%

Characterization techniques

All ¹H NMR spectra were collected using a Bruker nuclear magnetic resonance instrument (300 MHz) using tetramethylsilane (TMS) as the internal standard at room temperature. NMR samples were prepared with concentrations of 10 to 20 mg mL⁻¹ in CDCl₃. The number-average molecular weight (M_n) and polydispersity (Đ = M_w/M_n) of the polymers were determined using a size exclusion column (TOSOH HLC-8320) equipped with refractive-index and UV detectors using two TSKgel Super Mutipore HZ-N (4.6 × 150 mm, 3 μm bead size) columns arranged in series; this system can separate polymers in the molecular weight range of 500 to 1.9 × 10⁵ g mol⁻¹. THF was used as the eluent at a flow rate of 0.35 mL min⁻¹ at 40 °C. Data acquisition was performed using EcoSEC software, and molecular weights were calculated with polystyrene (PS) standards. Ultraviolet visible (UV-Vis) absorption spectra of the samples were recorded on a Shimadzu UV-2600 spectrophotometer at room temperature. The transmission electron microscopy (TEM) measurement was performed on a Hitachi HT7700 instrument (Japan) operating at a voltage of 120 kV. Before observation by TEM, the samples were prepared by dropping the freshly dialyzed assembly solution onto a carbon-covered copper grid. After drying in air at room temperature, the samples were negatively stained with 1.0 wt% phosphotungstic acid. FT-IR spectra were recorded on a Nicolette-6700 FT-IR spectrometer. Hydrodynamic diameter (D_h) was measured by dynamic light scattering (DLS) using Brookhaven's NanoBrook 90Plus PALS instrument at 25 °C at a scattering angle of 90°. The fluorescence emission spectra were obtained on a PerkinElmer LS-50B fluorescence spectrophotometer.

Results and discussion

Synthesis and characterization of the amphiphilic diblock copolymers

As shown in Scheme 1, firstly, the monomers, CAzo,³⁴ LAzo,³⁹ and PEG₄₅–Br⁴⁰ were prepared according to previous literature reports. Then, the corresponding amphiphilic polymers, PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉, were facilely synthesized using PEG₄₅–Br as the macro-initiator via ATRP in tetrahydrofuran (THF). The ¹H NMR, FT-IR and GPC curves confirmed the successful synthesis of the copolymers (Fig. 1 and S1†). Moreover, their molecular weights were determined by a combination of GPC with ¹H NMR via end group analysis. The repeated CAzo units (m) and LAzo units (n) in PEG₄₅-b-PCAzo_m or PEG₄₅-b-PLAzo_n were calculated based on the characteristic proton signals in their ¹H NMR spectra, as described in the ESI,† as well as using eqn (1) or (2) above; the results demonstrated that the two copolymer structures had similar molecular weights and hydrophilic/hydrophobic ratios (m = 17, n = 19).

Self-assembly of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ in PB solution (pH = 7.4)

The prepared amphiphilic copolymers can be assembled into stable vesicles in a mixture solution of PB and THF by carefully tuning the experimental parameters. The critical aggregation concentration (CAC) in PB was measured using Nile Red as a hydrophobic fluorescent probe according to a procedure previously reported.⁴¹ In the cases of both copolymers, the emission spectra registered from 580 to 700 nm were measured. As shown in Fig. 2, the emission intensity increased with increasing polymer concentration, and the CAC value for PEG₄₅-b-PCAzo₁₇ was around 25 μg mL⁻¹, which is lower than that corresponding to PEG₄₅-b-PLAzo₁₉ containing linear azobenzene moieties (38 μg mL⁻¹). The lower CAC value for the cyclic structure can be attributed to the poor solubility of cyclic azobenzene in common solvents, which is potentially caused by π-stacking interactions.⁴²

Fig. 2 Fluorescence intensity of Nile Red at 628 nm (λ_exc = 550 nm) versus copolymer concentration (mg mL⁻¹): PEG₄₅-b-PCAzo₁₇ (A) and PEG₄₅-b-PLAzo₁₉ (B).

Transmission electron microscopy (TEM) was then utilized to determine the morphologies of the formed self-assemblies. Obviously, it was confirmed that undisturbed vesicles with a deflated appearance were obtained in both cases (Fig. 3). The hydrophobic blocks containing the azobenzene moieties were aggregated in the membranes of the vesicles, and the hydrophilic PEG arrangements remained on the outer and inner surface of the vesicles. Further, the DLS measurement showed that the average diameters of the vesicles formed through PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ were 306 nm and 249 nm, respectively (Fig. 4).

Fig. 3 TEM images of vesicles formed by PEG₄₅-b-PCAzo₁₇ (A) and PEG₄₅-b-PLAzo₁₉ (B) in PB solution (pH 7.4).

Fig. 4 (A) DLS results for the hydrodynamic diameter (D_h,DLS) distributions of the self-assembled vesicles formed by PEG₄₅-b-PCAzo₁₇ in PB solutions at the initial state (D_h = 306 nm, PDI = 0.241), the PSS_UV state (D_h = 321 nm, PDI = 0.282) and the PSS_Vis state (D_h = 319 nm, PDI = 0.323). (B) DLS results for the hydrodynamic diameter (D_h,DLS) distributions of the self-assembled vesicles formed by PEG₄₅-b-PLAzo₁₉ in PB solutions at the initial state (D_h = 249 nm, PDI = 0.166), the PSS_UV state (D_h = 251 nm, PDI = 0.164) and the PSS_Vis state (D_h = 254 nm, PDI = 0.191). PSS: photo-stationary states.

Light-responsive behavior of Nile Red-loaded vesicles

Nile Red dye is one of the most popular model molecules for use in drug delivery research. As mentioned above, the hydrophobic Nile Red molecules can be encapsulated in the inner part of the membranes of vesicles. Before studying the light-responsive behavior of Nile Red-loaded vesicles, the photoisomerization performances of two kinds of vesicles without encapsulated substances that were formed by the different copolymers in PB solution were examined via UV/Vis absorption spectra (Fig. S2†). As shown in Fig. S2,† upon irradiation with UV light at 365 nm for 40 s or with visible light at 435 nm for 30 s, distinct trans-to-cis or reverse cis-to-trans isomerization behaviors of azobenzene moieties were observed based on the absorbances around 350 nm and 450 nm, which means that both the vesicles with cyclic azobenzene and linear azobenzene exhibit excellent photo-sensitivities. The DLS results demonstrated that the assemblies were not dissociated under UV light stimulation, as shown by their basically unchanged aggregation sizes before and after irradiation (Fig. 4). After Nile Red was encapsulated into the vesicles, an intense emission peak registered at 630 nm appeared in the fluorescence emission spectra for both assembly solutions (Fig. 5), indicating that Nile Red was in a hydrophobic environment. It is noteworthy that the fluorescence intensity of PEG₄₅-b-PCAzo₁₇ vesicle solution at the initial state was higher than that of PEG₄₅-b-PLAzo₁₉, indicating that the former can accommodate more dye molecules in the membrane than the latter. Moreover, the photos of dye-loading vesicle solutions also confirmed this result, because the vesicle solution containing cyclic azobenzenes and Nile Red was darker in color than the other solution, as the insets show in Fig. 5. The rigid configuration of small rings containing azobenzene may restrict the close packing of hydrophobic segments in PEG₄₅-b-PCAzo₁₇ so that vesicles with more loose membranes were formed compared with those formed by PEG₄₅-b-PLAzo₁₉, which has a linear structure.⁴³ O'Reilly et al. have also confirmed that cyclic graft copolymers show larger volumes of hydrophobic polycarbonate cores for Nile Red storage than linear graft copolymers.⁴⁴ Next, both vesicle solutions containing Nile Red were irradiated by UV light. After irradiation, a gradual decrease of the emission was observed in both cases (Fig. 5(A) and (B)) although the vesicle size showed no distinct variation, as Fig. 4 shows. This possibly resulted from the polarity change of azobenzene in the membrane portion, because Nile Red is an uncharged hydrophobic probe whose fluorescence is strongly influenced by the polarity of its environment.¹⁰ Higher polarity is induced and the hydrophilic/hydrophobic balance in the inner membrane is broken as the trans-azobenzene segments change to the cis-isomers, which may cause the fluorescence intensity of Nile Red to weaken.⁴⁵ As shown in Fig. 5, the fluorescence emission of the two kinds of solutions was almost constant after irradiation at 120 s and 90 s under UV light, respectively. Reversibly, after irradiation under visible light, the fluorescence emission was again enhanced, resulting from the recovery of the hydrophobicity of the vesicle membrane as the cis-azobenzene moieties return to the trans-isomers. After irradiating for 180 s, about 65% Nile Red emission intensity was recovered to the initial value for PEG₄₅-b-PCAzo₁₇. In comparison, 90 s was used for the recovery process of the PEG₄₅-b-PLAzo₁₉ system, while only about 33% fluorescence was recovered. Furthermore, we investigated by UV-Vis spectra whether Nile Red molecules could leak from the membrane into the PB solution. In Fig. 6, the UV-Vis spectra of the vesicle solutions loaded with Nile Red under alternating 365 nm light and 435 nm light irradiation for various times are given. The absorption maximum of Nile Red in PB solution is at about 550 nm. Before UV light irradiation, the absorption values of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ at 550 nm are 0.15 and 0.09, which also means that more storage can be achieved by the cyclic graft copolymers. However, during the reversible isomerization process in all cases, the absorption intensity of Nile Red changed little, indicating no leakage of Nile Red. Based on the results of the above studies, these light-responsive nanoparticles may be used in sensor technologies with no environmental pollution because the fluorescence of Nile Red in the membrane could change reversibly under altering UV-Vis irradiation, and a larger variation of fluorescence intensity was achieved for the cyclic azobenzene-containing materials.

Fig. 5 Emission spectra changes of PB solutions of vesicles of PEG₄₅-b-PCAzo₁₇ (A) and PEG₄₅-b-PLAzo₁₉ (B) with encapsulated Nile Red under UV/Vis light irradiation until the photo-stationary states were achieved. Insets are images of the corresponding vesicle solutions at their initial states.

Fig. 6 UV-Vis spectra of the CAzo-containing vesicles loaded with Nile Red under the irradiation of UV light (A) and visible light (B). UV-Vis spectra of the LAzo-containing vesicles loaded with Nile Red under the irradiation of UV light (C) and visible light (D).

Encapsulation and reductant-triggered release of DOX

It was verified that azoreductase or dithionite can cleave the N=N bond of the azobenzene group, resulting in the formation of an aniline group at each end of the remaining segments.^46–48 This special property has shown promise for use as carriers for novel anti-cancer drug delivery in colon disease systems.^49,50 Therefore, apart from the photo-responsive behavior of the two amphiphilic copolymer vesicles, we also paid attention to their reductant-responsive properties and their potential application in the encapsulation and reductant-induced release of hydrophobic substances. First, a typical experiment was carried out as follows: 0.1 mL of Na₂S₂O₄ solution (0.15 M in PB solution) was added dropwise to 0.5 mL of PEG₄₅-b-PCAzo₁₇ or PEG₄₅-b-PLAzo₁₉ vesicle solution in a 2 mL ampoule. Then, the effective immolation of the N=N bonds in the LAzo/CAzo-based vesicles was confirmed via various characterization approaches. As shown in Fig. 7(B), (C), (F) and (G), the colors of both vesicle solutions faded distinctly after treatment with dithionite at 37 °C for 12 h. TEM microscopy was used to gain further information about the morphological changes that occurred during the reduction process. In the presence of Na₂S₂O₄, though most aggregates still maintained the vesicle morphology, the polymeric vesicles were found to be disassembled in some degree based on the fact that their sizes and outlines showed obvious changes in the TEM images (Fig. 7(A), (D), (E) and (H)). The vesicles before and after adding Na₂S₂O₄ were also confirmed by DLS measurements (Fig. 7(I) and (J)), further showing the size of the vesicles with increased Na₂S₂O₄ stimulation. This interesting behavior may be mainly attributed to the surface tension change of the vesicles. After Na₂S₂O₄ treatment, the azobenzene segments in the polymeric side chains were converted to aniline groups. As we know, the surface tension of aniline is greater than that of azobenzene, which made a great contribution to the deformation of the vesicle membrane.⁵¹

Fig. 7 TEM images of untreated PEG₄₅-b-PCAzo₁₇ (A) and PEG₄₅-b-PLAzo₁₉ (E) vesicles; TEM images of PEG₄₅-b-PCAzo₁₇ (D) and PEG₄₅-b-PLAzo₁₉ (H) vesicles after cultivation in 0.15 M Na₂S₂O₄ (PB, pH = 7.4) at 37 °C for 36 h; images of untreated PEG₄₅-b-PCAzo₁₇ (B) and PEG₄₅-b-PLAzo₁₉ (F) vesicle solutions; images of PEG₄₅-b-PCAzo₁₇ (C) and PEG₄₅-b-PLAzo₁₉ (G) vesicle solutions after cultivation in 0.15 M Na₂S₂O₄ (PB, pH = 7.4) at 37 °C for 36 h; (I) DLS results for the hydrodynamic diameter (D_h,DLS) distributions of self-assembled vesicles formed by PEG₄₅-b-PCAzo₁₇ in PB solutions in the absence of Na₂S₂O₄ (D_h = 306 nm, PDI = 0.241) and with Na₂S₂O₄ stimulation (D_h = 345 nm, PDI = 0.287); (J) DLS results for the hydrodynamic diameter (D_h,DLS) distributions of self-assembled vesicles formed by PEG₄₅-b-PLAzo₁₉ in PB solution in the absence of Na₂S₂O₄ (D_h = 249 nm, PDI = 0.166) and with Na₂S₂O₄ stimulation (D_h = 435 nm, PDI = 0.201).

Subsequently, the anticancer drug doxorubicin (DOX) was chosen as the hydrophobic probe in the investigation of reductant-responsive properties because Nile Red contains a quinoid fragment that can be easily reduced by dithionites, such as sodium dithionite.^52,53 Moreover, DOX has maximum absorbance at 480 nm in pH 7.4 medium, a wavelength at which the absorbance of both copolymers is negligible (Fig. 6). The encapsulation method is illustrated above, and the results showed that at a theoretical DOX loading content of 37.5 wt%, the drug loading efficiencies of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ vesicle were 21.9 ± 0.7% and 17.5 ± 1.0%, corresponding to DOX loading contents of 11.6 ± 0.4 wt% and 9.5 ± 0.5 wt%, respectively. Then, the controlled release of DOX under reductant stimuli was evaluated by UV-Vis absorbance spectra.⁵⁴ The vesicles loaded with DOX were charged with Na₂S₂O₄ solution and stirred in a water bath at 37 °C. After stimulation for a specific time, the absorbance spectra were recorded (Fig. S3†), where the absorbance intensity of DOX at 480 nm decreased gradually because the hydrophobic DOX was released from the polymeric nanoparticles. Fig. 8 shows the release profile of DOX encapsulated in the polymeric vesicles of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉, from which it can be seen that the release behaviors almost ceased for both vesicles after stimulation for 6 h, indicating that the cleavage of N=N bonds leads to the breakage that allows the cargo to pass. After 12 h, the total release amounts were 63% and 55%,⁵⁵ respectively. Then, to ensure the accuracy of the data, the controlled experiment of release was also investigated in the absence of Na₂S₂O₄ at 37 °C; from this experiment, it was found to be a small amount (<8%). These results revealed that both polymeric vesicles were stable in the absence of stimulation. Also, more importantly, the PEG₄₅-b-PCAzo₁₇ vesicles show a higher release rate and larger release amount than the linear analogues because the looser membrane formed from the cyclic architecture can more easily interact with dithionite.

Fig. 8 Release profiles of DOX encapsulated in the polymeric vesicle nanoparticles of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ with/without Na₂S₂O₄ stimulation.

Cytotoxicity tests

Cytotoxicity assessments of PEG₄₅-b-PCAzo₁₇ and PEG₄₅-b-PLAzo₁₉ vesicles before and after reduction by Na₂S₂O₄ were evaluated by MTT assay. In order to avoid the influence of residual dithionite on living cells, the vesicles that underwent reduction were dialyzed against PB solution to completely remove the dithionite before the MTT test. Unlike the copolymer PEG₄₅-b-PCAzo₁₇, dissociative anisidine molecules were generated in PEG₄₅-b-PLAzo₁₉ vesicles during the reduction step, but also would be removed from the solution during the dialysis procedure. Thus, appropriate anisidine molecules equal to the mole quantity of azo moieties were added to the resulting PEG₄₅-b-PLAzo₁₉ vesicles by dialysis to investigate whether the resulting trace of anisidine molecules has an influence on the cells. Here, two kinds of cells, normal L929 cells and A549 carcinoma cells, were used. After the MTT reagent (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) was added, the cells were incubated with vesicles for 72 h at varying concentrations from 10 to 100 μg mL⁻¹. As shown in Fig. 9, the results revealed that both LAzo-based vesicles and CAzo-based vesicles have low cytotoxicities (cell viability >90%) at all tested concentrations and conditions. Furthermore, though the vesicles containing anisidine show no toxicity to the above two cell types, it is well known that aniline derivatives are carcinogenic.⁵⁶ Also, organic small molecules are easily absorbed in living bodies; therefore, long-term intake of aniline or its analogues is dangerous to human health. Herein, we believe that one prominent advantage of the cyclic azo-based polymer materials over the linear materials is that the resulting aniline moieties remained on the polymer skeleton when the N=N bonds in the cyclic azobenzene structure were broken apart by dithionite; the undegradable polymers can then be eliminated from the body directly.⁵⁷ Therefore, PEG₄₅-b-PCAzo₁₇ may have greater application potential than PEG₄₅-b-PLAzo₁₉ in the biomedicine field.

Fig. 9 Cell cytotoxicity of polymeric vesicles against normal L929 cells and A549 carcinoma cells at various concentrations.

Conclusions

Novel responsive amphiphilic copolymers with cyclic azobenzene (PEG₄₅-b-PCAzo₁₇) and linear azobenzene pendants (PEG₄₅-b-PLAzo₁₉) were successfully synthesized via ATRP technology. Both polymers can assemble into stable vesicles with diameters of 100 to 300 nm in PB solution (pH 7.4). More importantly, these vesicles show unique dual sensitivities to UV/Vis light radiation and dithionite due to the existence of azobenzene segments; especially, better responsiveness of the cyclic azobenzene structure compared to the linear analogue was observed. The two different vesicles can effectively encapsulate common hydrophobic fluorescence probes, such as Nile Red and DOX. Furthermore, CAzo-containing vesicles exhibited higher drug loading content and better reversibility of NR fluorescence variation under irradiation with UV/Vis light, which has potential applications in the fields of sensors. Moreover, in the investigation of the reductant-triggered release of DOX, it is noteworthy that the PEG₄₅-b-PCAzo₁₇ vesicles show a higher release rate and a larger release amount (ca. 63%) than the linear counterpart PEG₄₅-b-PLAzo₁₉ (ca. 55%). Because Na₂S₂O₄ can act as an imitator to azoreductase, which exists in the colon, the vesicles generated from cyclic azobenzene-based copolymers have great potential for the treatment of colon disease based on their superiority to the linear azobenzene-based copolymers in drug loading and controlled release.

Acknowledgements

This work was supported by the National Science Foundation of China (21234005, 21574090), the State and Local Joint Engineering Laboratory for Novel Functional Polymeric Materials, the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and the Program of Innovative Research Team of Soochow University.

Notes and references

1. S. Ganta, H. Devalapally, A. Shahiwala and M. Amiji, J. Controlled Release, 2008, 126, 187–204.
2. E. Fleige, M. A. Quadir and R. Haag, Adv. Drug Delivery Rev., 2012, 64, 866–884.
3. S. Mura, J. Nicolas and P. Couvreur, Nat. Mater., 2013, 12, 991–1003.
4. D. Schmaljohann, Adv. Drug Delivery Rev., 2006, 58, 1655–1670.
5. C. Boyer, N. A. Corrigan, K. Jung, D. Nguyen, T.-K. Nguyen, N. N. M. Adnan, S. Oliver, S. Shanmugam and J. Yeow, Chem. Rev., 2016, 116, 1803–1949.
6. A. Bagheri, J. Yeow, H. Arandiyan, J. Xu, C. Boyer and M. Lim, Macromol. Rapid Commun., 2016, 37, 905–910.
7. Y.-J. Pan, Y.-Y. Chen, D.-R. Wang, C. Wei, J. Guo, D.-R. Lu, C.-C. Chu and C.-C. Wang, Biomaterials, 2012, 33, 6570–6579.
8. P.-F. Caponi, F. M. Winnik and R. V. Ulijn, Soft Matter, 2012, 8, 5127–5130.
9. L. Zha, Y. Zhang, W. Yang and S. Fu, Adv. Mater., 2002, 14, 1090–1092.
10. N. Feng, G. Han, J. Dong, H. Wu, Y. Zheng and G. Wang, J. Colloid Interface Sci., 2014, 421, 15–21.
11. G. Wang, D. Yuan, T. Yuan, J. Dong, N. Feng and G. Han, J. Polym. Sci., Part A: Polym. Chem., 2015, 53, 2768–2775.
12. T. Yuan, J. Dong, G. Han and G. Wang, RSC Adv., 2016, 6, 10904–10911.
13. X. Liu, P. Ni, J. He and M. Zhang, Macromolecules, 2010, 43, 4771–4781.
14. S.-i. Yamamoto, J. Pietrasik and K. Matyjaszewski, Macromolecules, 2008, 41, 7013–7020.
15. X. Yan, F. Wang, B. Zheng and F. Huang, Chem. Soc. Rev., 2012, 41, 6042–6065.
16. H. D. Bandara and S. C. Burdette, Chem. Soc. Rev., 2012, 41, 1809–1825.
17. A. D. Wong, T. M. Güngör and E. R. Gillies, ACS Macro Lett., 2014, 3, 1191–1195.
18. W.-X. Gu, Q.-L. Li, H. Lu, L. Fang, Q. Chen, Y.-W. Yang and H. Gao, Chem. Commun., 2015, 51, 4715–4718.
19. J. Rao and A. Khan, J. Am. Chem. Soc., 2013, 135, 14056–14059.
20. H. Mutlu and C. Barner-Kowollik, Polym. Chem., 2016, 7, 2272–2279.
21. A. Chevalier, C. Mercier, L. Saurel, S. Orenga, P.-Y. Renard and A. Romieu, Chem. Commun., 2013, 49, 8815–8817.
22. M. Chourasia and S. Jain, J. Pharm. Pharm. Sci., 2003, 6, 33–66.
23. S. S. Dhaneshwar, N. Gairola, M. Kandpal, G. Vadnerkar, L. Bhatt, B. Rathi and S. Kadam, Eur. J. Med. Chem., 2009, 44, 3922–3929.
24. H. Tian, Z. Tang, X. Zhuang, X. Chen and X. Jing, Prog. Polym. Sci., 2012, 37, 237–280.
25. L. Bromberg, M. Temchenko, V. Alakhov and T. A. Hatton, Langmuir, 2005, 21, 1590–1598.
26. A. D. Wong, A. L. Prinzen and E. R. Gillies, Polym. Chem., 2016, 6, 1871–1881.
27. P. S. Nunes, P. D. Ohlsson, O. Ordeig and J. P. Kutter, Microfluid. Nanofluid., 2010, 9, 145–161.
28. J. N. Hoskins and S. M. Grayson, Polym. Chem., 2011, 2, 289–299.
29. T. Yamamoto and Y. Tezuka, Polym. Chem., 2011, 2, 1930–1941.
30. Y. Wang and Q. Li, Adv. Mater., 2012, 24, 1926–1945.
31. J. Deng, N. Song, Q. Zhou and Z. Su, Org. Lett., 2007, 9, 5393–5396.
32. S. S. Zade and M. Bendikov, J. Org. Chem., 2006, 71, 2972–2981.
33. F. Zhang, G. Götz, H. D. Winkler, C. A. Schalley and P. Bäuerle, Angew. Chem., Int. Ed., 2009, 48, 6632–6635.
34. J. Li, N. Zhou, Z. Zhang, Y. Xu, X. Chen, Y. Tu, Z. Hu and X. Zhu, Chem.–Asian J., 2013, 8, 1095–1100.
35. J. Lu, A. Xia, N. Zhou, W. Zhang, Z. Zhang, X. Pan, Y. Yang, Y. Wang and X. Zhu, Chem.–Eur. J., 2015, 21, 2324–2329.
36. Y. Cai, J. Lu, F. Zhou, X. Zhou, N. Zhou, Z. Zhang and X. Zhu, Macromol. Rapid Commun., 2014, 35, 901–907.
37. B. Chen, Z. Wang, J. Lu, X. Yang, Y. Wang, Z. Zhang, J. Zhu, N. Zhou, Y. Li and X. Zhu, Polym. Chem., 2015, 6, 3009–3013.
38. J. Lu, G. Jiang, Z. Zhang, W. Zhang, Y. Yang, Y. Wang, N. Zhou and X. Zhu, Polym. Chem., 2015, 6, 8144–8149.
39. A. G. Cook, R. T. Inkster, A. Martinez-Felipe, A. Ribes-Greus, I. W. Hamley and C. T. Imrie, Eur. Polym. J., 2012, 48, 821–829.
40. R. Dong, B. Zhu, Y. Zhou, D. Yan and X. Zhu, Polym. Chem., 2013, 4, 912–915.
41. H. Hong, Y. Mai, Y. Zhou, D. Yan and J. Cui, Macromol. Rapid Commun., 2007, 28, 591–596.
42. V. A. Azov, J. Cordes, D. Schlüter, T. Dülcks, M. Böckmann and N. L. Doltsinis, J. Org. Chem., 2014, 79, 11714–11721.
43. X. Fan, B. Huang, G. Wang and J. Huang, Macromolecules, 2012, 45, 3779–3786.
44. R. J. Williams, A. Pitto-Barry, N. Kirby, A. P. Dove and R. K. O'Reilly, Macromolecules, 2016, 49, 2802–2813.
45. E. Blasco, J. L. Serrano, M. Piñol and L. Oriol, Macromolecules, 2013, 46, 5951–5960.
46. J. Rao and A. Khan, Polym. Chem., 2015, 6, 686–690.
47. J. Rao, C. Hottinger and A. Khan, J. Am. Chem. Soc., 2014, 136, 5872–5875.
48. S. H. Lee, E. Moroz, B. Castagner and J.-C. Leroux, J. Am. Chem. Soc., 2014, 136, 12868–12871.
49. S. R. Deka, S. Yadav, M. Mahato and A. K. Sharma, Colloids Surf., B, 2015, 135, 150–157.
50. S. Yadav, S. R. Deka, G. Verma, A. K. Sharma and P. Kumar, RSC Adv., 2016, 6, 8103–8117.
51. K. Han, W. Su, M. Zhong, Q. Yan, Y. Luo, Q. Zhang and Y. Li, Macromol. Rapid Commun., 2008, 29, 1866–1870.
52. C. Yi, D. Huang and M. Gratzl, Anal. Chem., 1996, 68, 1580–1584.
53. O. A. Kucherak, S. Oncul, Z. Darwich, D. A. Yushchenko, Y. Arntz, P. Didier, Y. Mély and A. S. Klymchenko, J. Am. Chem. Soc., 2010, 132, 4907–4916.
54. X. Zhang, L. Meng, Q. Lu, Z. Fei and P. J. Dyson, Biomaterials, 2009, 30, 6041–6047.
55. F. Yang, Z. Cao and G. Wang, Polym. Chem., 2015, 6, 7995–8002.
56. M. Kugler-Steigmeier, U. Friederich, U. Graf, W. K. Lutz, P. Maier and C. Schlatter, Mutat. Res., 1989, 211, 279–289.
57. P. Caliceti and F. M. Veronese, Adv. Drug Delivery Rev., 2003, 55, 1261–1277.

---

Footnote

† Electronic supplementary information (ESI) available: Polymer characterization data, DLS, UV-Vis, and FL spectra data were given. See DOI: 10.1039/c6ra12751g

---