pH-Responsive supramolecular hydrogels for codelivery of hydrophobic and hydrophilic anticancer drugs

Abstract

Codelivery of multiple drugs with one kind of drug carrier provides a promising strategy to suppress the drug resistance and achieve the enhanced therapeutic effect in cancer treatment. In this work, we successfully developed multifunctional supramolecular hydrogels based on in situ host–guest inclusion between polymer-drug conjugates and α-cyclodextrin to codeliver hydrophobic and hydrophilic anticancer drugs with pH-trigged release properties. Taking advantage of the strong hydrophobicity of 4β-aminopodophyllotoxin (NPOD), a derivative of podophyllotoxin (POD), the NPOD molecule was conjugated to low-molecular-weight methoxypoly (ethylene glycol) (mPEG) chain via a pH-responsive imine bond, forming an amphiphilic polymer-drug conjugates (NPOD-PEG). After adding α-cyclodextrin (α-CD) into the NPOD-PEG solutions, the stable supramolecular hydrogels were formed based on a combination of the partial inclusion complexation between one end of the mPEG blocks and α-CD and the hydrophobic aggregation of NPOD groups. The formed hydrogels could further efficiently load another hydrophilic anticancer drug doxorubicin (DOX) for combination therapy purposes. The hydrogel demonstrated unique gel–sol transition properties and pH-dependent dual drug release behavior due to the hydrolysis of imine bond at acidic environments. Furthermore, the cytotoxicity results suggested that the DOX loaded NPOD-PEG/α-CD hydrogels showed an enhanced cytotoxicity in cancer cells in comparison with single modality treatment and the resulting hydrogels are characterized by producing an additive cytotoxicity to cancer cells. In fact, the codelivery of two anticancer drugs with different physicochemical properties and anticancer mechanisms was a key to opening the door to their controlled drug delivery and enhanced anticancer effect. Therefore, DOX loaded NPOD-PEG/α-CD hydrogels as pH-trigged drug codelivery systems might have important potential for combination cancer chemotherapy.

---

Introduction

Known as soft intelligent materials, hydrogels can exhibit a unique gel–sol or sol–gel transition under external stimuli, such as temperature, pH, pressure, and solvent composition.^1–4 Due to their excellent and attractive stimuli-responsive properties, intelligent hydrogels have gained diverse applications in controlled drug delivery, tissue engineering, artificial muscles, soft machines, etc.^5–8 For example, tumor and inflammatory tissues present an acidic environment relative to normal tissue, requiring a gel that could release drug molecules in response to a change in pH.^9–11 Research on pH-responsive hydrogels has mainly focused on the controlled release of a single drug in response to an external stimulus.¹² However, for some multifactor disease, such as cancer usually composed of mixed populations of malignant cells, it may require treatment with compounds that could target multiple intracellular components. Anticancer drugs are rarely used singly, because only a few tumors are sensitive enough to be cured by a single drug.¹³ So, for a specific type of tumors, effective chemotherapy usually depends on suitable combinations.^14,15 Accordingly, modulating multiple targets simultaneously can be achieved by the combination of multiple drugs with different mechanisms that could modulate several targets of a multifactorial disease.^16–19 Therefore, it will be beneficial to develop effective pH-responsive hydrogels loading multiple-anticancer drugs as a co-delivery system.

Among hydrogel systems, the most widely studied ones having great application prospects are hydrogels based on supramolecular self-assembly between cyclodextrin (CD) and polymeric chains.^20,21 In this area, one of the most classic and successful model is pseudopolyrotaxane (PPR), formed by a linear poly(ethylene glycol)s (PEGs) chain threading a series of CD cavities.^22–24 Strong hydrogen bonds between the adjacent threaded CDs result in microcrystalline aggregation and then promote physical gel formation. Such a hydrogel can be sensitive to the environmental stimuli and exhibit reversible thixotropism, paving the way for its potential use as a injectable carrier for drug delivery in view of the excellent biocompatibility of both PEG and α-CD.^25,26 However, it is known that high molecular weight (MW) PEG (normally M_n > 10 000 Da) are not suitable for filtration through the human kidney membrane because of their large hydrodynamic radius.²⁷ Notably, most strategy make hydrogels difficult to load diverse drugs with different physicochemistry property and anticancer mechanism owing to their highly hydrated, hydrophilic microstructures.

Our group has recently developed a novel approach for preparing supramolecular hydrogels by the combination of hydrophobic anticancer drugs monoend-functionalized low-MW methoxypoly (ethylene glycol) (mPEG) and α-CD and the hydrophobic aggregation of the drug molecules.¹⁶ Meanwhile, the formed hydrogels could be further loaded with another water-soluble anticancer drug for combination therapy purpose.¹⁶ In this system, both the partial inclusion complexation and the hydrophobic aggregation acts as supra-cross-links playing important roles in supramolecular hydrogel formation, dual anticancer drug encapsulation and controlled release. Due to the reversible temperature-responsive supramolecular assembly of PPR between one end of the PEG blocks and α-CD, the gel–sol transition temperature of such a hydrogel could be successfully modulated from 30 to 60 °C by varying the length of PEG chain and the concentration of α-CD.²⁸ We thus proposed that using an imine bond to connect PEG chain and hydrophobic drug would provide an opportunity to further endow hydrogels with pH-responsive gel–sol transition property and dual drug-loading capacity.

Etoposide (VP-16) and teniposide (VM-26) as successful derivatives of podophyllotoxin are high activities against small cell lung cancer, testicular carcinoma, lymphoma, kaposi's sarcoma, neuroblastoma, soft tissue sarcoma and other serious cancers.^29,30 So far, a large amount of elegant approaches have been reported for the preparation of new podophyllotoxin derivatives.^31,32 4β-aminopodophyllotoxin (NPOD) as an intermediate plays an important role in the preparation of the new analogues because of the presence of amino in its structure. DOX is a water-soluble anti-cancer drug with topoisomerase II inhibiting activity, which is widely used in chemotherapy to treat cancers of the bladder, breast, stomach, lung, ovaries, thyroid, soft tissue sarcoma, and multiple myeloma.³³

On the basis of the above studies, 4β-aminopodophyllotoxin (NPOD) and doxorubicin (DOX) were selected as hydrophobic and hydrophilic model drugs with different anticancer mechanism, respectively. As shown in Scheme 1, NPOD were conjugated to a low MW PEG (MW = 2000) chain via an imine bond, forming the amphiphilic polymer-drug conjugates (NPOD-PEG). Dual drug-loaded supramolecular hydrogel was successfully constructed based on in situ host–guest interaction between NPOD-PEG and α-CD. It is interesting that by varying the pH of culture solutions, supramolecular hydrogel demonstrated unique pH-dependent controlled release behavior. Besides, in vitro cytotoxicity studies confirmed that dual drug-loaded supramolecular hydrogel exerted higher cytotoxicity on A549 human lung cancer cells in comparison with that of free NPOD. We also evaluated the toxicological interactions of two drugs by the method of combination index (CI)-isobologram equation, which produced a non-dose-dependent additive cytotoxicity to cancer cells.

Scheme 1 Schematic representation of the supramolecular hydrogel made of NPOD-PEG and α-CD and the dual phase drug release process.

Experimental section

Materials

mPEG (MW = 2000) was purchased from Alfa Aesar without further purification, α-cyclodextrin (α-CD) and 4-(dimethylamino) pyridine (DMAP) was purchased from Aladdin Chemical Co. (China). 4β-aminopodophyllotoxin (NPOD) was kindly given by Lanzhou University. DOX was purchased from Sigma-Aldrich. Dicyclohexylcarbodiimide (DCC) was purchased from Shanghai Kefeng Chemical Reagent Co., Ltd (China). Other reagents were analytical pure and used directly without further purification. All solvents and water were freshly redistilled.

Measurements

¹H NMR spectra were measured on a Bruker AVANCE III-400 spectrometers. The chemical shifts of ¹H NMR are expressed in parts per million downfield relative to the internal tetramethylsilane (δ = 0 ppm) or chloroform (δ = 7.26 ppm). The crystalline changes of the hydrogels were confirmed by X-ray diffraction measurements, which were performed by using Cu Kα irradiation with a PHILP X'Pert PRO. The rheological behavior of the hydrogels was investigated using a HAKKE RS6000 rotational rheometer. For the scanning electron microscopy (SEM) observations, the specimens were freeze-dried under a vacuum and ground to fine powder. The powder was placed on conducting glue and coated with gold vapor and then analyzed on a JSM-5600LV electron microscope.

Synthesis of NPOD-PEG conjugates

The synthesis of NPOD-PEG conjugates is shown in Fig. 1, and the brief procedure was as follows: mPEG 2000 (1.28 g, 0.64 mmol) and 4-formylbenzoic acid (0.11 g, 0.73 mmol) were dissolved in 20 mL dichloromethane, and the resulting solution was stirred in an ice bath. Then DCC (0.14 g, 1 mmol) and DMAP (0.10 g, 0.1 mmol) were added to the reaction solution. The solution was kept stirred overnight. After removing dicyclohexylurea (DCU) by filtration, the filtrate was concentrated using a rotavapor. The crude product was subjected to column chromatography on C₁₈ using methanol/water (6 : 4) to give the pure product as a white crude solid mPEG-CHO (0.99 g, 72% yield). ¹H NMR (400 MHz, CDCl₃): δ10.07 (1H, s, –CHO), 8.20–7.91 (4H, m, phenyl group), 4.47 (2H, t, –CH₂CH₂OOC–), 3.61 (H, m, –OCH₂CH₂O), 3.34 (3H, s, –OCH₃). The spectrum is shown in Fig. S1 (ESI†).

Fig. 1 Synthesis route of NPOD-PEG conjugates.

NPOD (0.20 g, 0.5 mmol) and mPEG-CHO (0.70 g, 0.34 mmol) were dissolved in 30 mL ethanol, and the reaction mixture was refluxed overnight, then evaporated in vacuum and followed by Sephadex LH-20 gel chromatography with methanol as eluent to give the pure yellow viscous oil product NPOD-PEG (0.78 g, 80% yield). ¹H NMR (400 MHz, CDCl₃): δ 8.47 (s, 1H), 8.06 (d, 2H), 7.78 (d, 2H), 6.56 (s, 1H), 6.54 (s, 1H), 6.35 (s, 1H), 5.94 (d, 1H), 5.88 (d, 1H), 5.46 (s, 1H), 4.67 (d, 1H), 4.62 (d, 1H), 4.45 (t, 2H), 4.22 (t, 1H), 3.96 (m, 1H), 3.78 (s, 6H), 3.60 (br, 186H), 3.45 (s, 3H), 3.35 (s, 3H), 3.09 (m, 1H). ¹³C-NMR (100 MHz, CDCl₃): δ 37.62, 41.73, 43.93, 56.46, 59.02, 64.38, 67.74, 69.11, 70.52, 71.88, 76.68, 101.43, 108.02, 108.56, 110.55, 128.37, 129.57, 129.99, 131.28, 139.18, 146.36, 147.21, 147.98, 159.54, 165.87. The spectrum is shown in Fig. S2 (ESI†).

pH responsiveness of NPOD-PEG conjugates

The pH responsiveness of NPOD-PEG with imine bond at different pH values (5.0 and 7.4) was carried out as follows: first, 5 mg NPOD-PEG was added to 10 mL buffer solution (pH 5.0, pH 7.4), respectively. Then, the solutions were incubated in water bath at 37 °C. 0.5 mL medium was removed at desired time intervals. Hydrolysis tendency of NPOD-PEG conjugates was detected by high performance liquid chromatographic (HPLC) at 265 nm. Chromatographic separation was performed on a Kromasil C₁₈ column (4.6 × 250 mm, 5 μm) with methanol and aqueous solutions (75 : 25, v/v) as mobile phase at a flow rate of 0.8 mL min⁻¹.

Formation of hydrogels

The general protocol for the hydrogel formation is as follows: 90 mg α-CD was added to a 1.0 mL aqueous solution of NPOD-PEG. Various concentrations of NPOD-PEG (5, 10, 15 and 30 mg mL⁻¹) were used to formulate different hydrogels. For all samples, the solution was thoroughly sonicated for 5 min followed by incubation for 72 h before measurements. For encapsulating DOX, α-CD (90.0 mg) and 1.0 mg DOX was added to 1.0 mL aqueous solution of NPOD-PEG (30.0 mg mL⁻¹), then the solution was mixed thoroughly by sonication for 5 min followed by incubation at room temperature for 72 h before measurements. Free DOX was carefully removed and determined by UV-vis spectrophotometer (Lambda 35) at 485 nm using the equation for calibration curve. The DOX encapsulation efficiency (EE) was calculated using the eqn (1):

EE = (m₁ − m₂)/m₁ × 100% (1)

where m₁ is the initial amount of drug loaded in the hydrogel, m₂ is the amount of free DOX.

In vitro release kinetics studies

90.0 mg α-CD and 1.0 mg DOX was separately added into 1.0 mL NPOD-PEG conjugates PBS (pH 7.4) or acetate buffer (pH 5.0) solution (30.0 mg mL⁻¹). The solution was thoroughly sonicated for 5 min followed by incubation for 72 h, allowing the mixture to form a viscous hydrogel. The cuvette was placed upside-down in a test tube with 30.0 mL buffer solution with different pH value and incubated in a 37 °C water bath. 5 mL medium was removed and replaced by 5 mL fresh buffer solution at desired time intervals. The concentrations of DOX released from hydrogels were determined using a Lambda 35 UV-vis spectrophotometer at 485 nm and calculated based on the equation for calibration curve, which is shown in Fig. S3.† Meanwhile, the cumulative percent release of NPOD was determined by HPLC with the same chromatographic conditions as described above.

Cytotoxicity assays

The cytotoxicity of DOX (1 mg)-loaded NPOD-PEG/α-CD hydrogel, NPOD-PEG/α-CD hydrogel, NPOD and NPOD-PEG against A549 human lung cancer cell line was evaluated by the MTT assay. Briefly, cells at the exponential growth phase were harvested and seeded into a flat-bottom 96-well plate at an initial density of 5 × 10⁴ cells/well, and cultured in a 5% humidified CO₂ incubator at 37 °C for 24 h. Thereafter, the cells were treated with the test samples at various concentrations. 20 μL MTT solution was added to each well to continue incubating for 4 h at 37 °C. The cell viability was obtained by scanning with a microplate reader at 570 nm. The relative cell viability (%) was expressed as a percentage of that of the control culture. The experiments were carried out six times in parallel. The results presented are the average data.

Median-effect principle for dose-effect analysis and the combination index studies

Median-effect principle for dose-effect analysis and the combination index studies was utilized to examine drug interactions.³⁴ This method involved plotting the dose-effect curves for each compound and their combinations in multiple diluted concentrations by using the median-effect equation:

f_a/f_u = (D/D_m)^m (2)

where D is the dose, D_m is the dose for 50% effect (e.g., 50% inhibition of cell growth), f_a is the fraction affected by dose D, f_u is the unaffected fraction (therefore, f_a = 1 − f_u), and m is the coefficient of the sigmoidicity of the dose-effect curve.

The D_m and m values for each fibrate are easily determined by the median-effect plot: x = log(D) versus y = log(f_a/f_u) which is based on the logarithmic form of eqn (2). In the median-effect plot, m is the slope and log(D_m) is the x-intercept. A combination index (CI) is then determined with the classic isobologram equation of Chou–Talalay:³⁵

CI = (D)₁/(D_x)₁ + (D)₂/(D_x)₂ (3)

where (D_x)₁ is the dose of agent 1 (NPOD) required to produce × percentage effect alone and (D)₁ is the dose of agent 1 required to produce the same × percentage effect in combination with (D)₂. Similarly, (D_x)₂ is the dose of agent 2 (DOX) required to produce × percentage effect alone and (D)₂ is the dose required to produce the same effect in combination with (D)₁. (D_x)₁ and (D_x)₂ as the denominators of the CI equation above can be determined by D_x = D_m[f_a/(1 − f_a)]^1/m. Different values of CI may be obtained for solving the equation for different values of f_a. CI values of <1 indicate synergy, >1 indicate antagonism and =1 indicate additive effect.

Results and discussion

Synthesis of NPOD-PEG conjugates

The imine bond was chosen as the pH responsive linking bond for preparation of polymer-drug conjugates (NPOD-PEG) because it is easy to be hydrolyzed in mildly acidic environment.³⁶ According to our previous work, supramolecular hydrogel can be successfully prepared through host–guest interaction of drug modified low-MW mPEG (MW ≤ 2000) and α-CD. Therefore, PEG with average molecular weight 2000 was selected as the hydrophilic segment. The synthetic procedures of NPOD-PEG are shown in Fig. 1. The mPEG with a terminal aldehyde group (mPEG-CHO) was first synthesized via 4-formylbenzoic acid with good yield of over 70%. The structure of mPEG-CHO was confirmed by ¹H NMR (Fig. S1 in the ESI†). Then, the NPOD-PEG with imine bond as linkage bond was synthesized by refluxing the mPEG-CHO and NPOD in ethanol for 12 h. The structure of the product was confirmed by ¹H and ¹³C NMR. From ¹H NMR (Fig. S2 in the ESI†), the chemical shift at δ 8.47 (s, 1H) represented the proton of the new formed imine bond, and all the other chemical shifts of the NMR and integral ratios could be attributed to the targeted structure. ¹³C NMR results also confirmed the successful synthesis of the NPOD-PEG conjugates.

Preparation of supramolecular hydrogels

When α-CD was added to low-MW mPEG (MW 2000) solution (30 mg mL⁻¹), as expected, a precipitate rather than a homogeneous hydrogel was formed (Fig. 2a). This precipitate has been proved to be comprised of crystalline complexes of the PEG/α-CD polypseudorotaxanes.^37,38 However, a homogeneous hydrogel was formed when adding α-CD to NPOD-PEG conjugates at low concentrations (10 mg mL⁻¹). It is worth noting that under the same conditions as those for mPEG, homogeneous hydrogels were obtained in several minutes over a broad concentrations range of NPOD-PEG (10–30 mg mL⁻¹, Fig. 2c–e). Therefore, the introduction of NPOD molecules into a low-MW mPEG chain to form amphiphilic blocks plays an important role in providing additional physical cross-links for the gelation process.

Fig. 2 Optical photo of the precipitate made of (a) mPEG (30 mg mL⁻¹), (b) NPOD-PEG (5 mg mL⁻¹) and the supramolecular hydrogels made of (c) NPOD-PEG (10 mg mL⁻¹), (d) NPOD-PEG (15 mg mL⁻¹), (e) NPOD-PEG (30 mg mL⁻¹), (f) NPOD-PEG (30 mg mL⁻¹) with HCl, (g) DOX (1 mg) and NPOD-PEG (30 mg mL⁻¹). For all samples [α-CD] = (90 mg mL⁻¹).

To further confirm the formation of the PPR complexes between NPOD-PEG conjugates and α-CD molecules, complementary XRD studies of freeze-dried hydrogels, pure α-CD and pure NPOD-PEG were carried out (Fig. 3). As shown in Fig. 3, after adding α-CD into NPOD-PEG solutions, the hydrogel sample was observed to have a new characteristic diffraction peak at 2θ = 19.8° (Fig. 3D and E) which is different from that of pure NPOD-PEG (Fig. 3B, 2θ = 19.2° and 23.3°) and α-CD (Fig. 3A, 2θ = 21.5°). The sharp diffraction peak at 2θ = 19.8° (Fig. 3C) is identical with the extended channel structure of α-CD, which corresponds to the rod channel structures of PEG/α-CD inclusion complex.^39,40 This implies that α-CD rings are stacked along the NPOD-PEG chains to form a channel-type crystalline structure in the hydrogel network. At the same time, the strong hydrogen-bond interaction among adjacent rod channel type crystalline structures of PPR can provide additional supra-cross-links which is favorable to the gelation process (Scheme 1).

Fig. 3 X-ray diffraction patterns for freeze-dried (A) pure α-CD, (B) pure NPOD-PEG, (C) mPEG/α-CD inclusion complexes, (D) NPOD-PEG/α-CD inclusion complexes [NPOD-PEG] = 5 mg mL⁻¹, (E) NPOD-PEG/α-CD hydrogels [NPOD-PEG] = 30 mg mL⁻¹. For all samples the concentrations of α-CD is 90 mg mL⁻¹.

The introduction of NPOD molecules into low-MW PEG chains shows remarkable positive effects not only on the strength but also on the viscosity of hydrogels. As shown in Fig. S4,† the values of the storage modulus (G′) are greater than that of the loss modulus (G′′) indicating that all samples are hydrogels. For a comparison, mPEG/α-CD inclusion complex was also measured as a blank. As shown in Fig. 4, at a fixed stress and α-CD concentration (90 mg mL⁻¹), both G′ and the viscosity of the hydrogels increased with an increase in NPOD-PEG content, while G′ and the viscosity of mPEG/α-CD inclusion complex with high mPEG concentration (30 mg mL⁻¹) were almost zero. For example, the G′ of NPOD-PEG at 30 mg mL⁻¹ is ∼6 times higher than that at 10 mg mL⁻¹ over a broad frequency range. Such a difference in the rheological behaviour might be attributed to the increase of supra-cross-links in the hydrogels. Furthermore, all NPOD-PEG hydrogels exhibit good shear-thinning behavior, a basic and valuable property required for injectable hydrogels.⁴⁰ This shear-thinning effect can be attributed to the supra-cross-links. Under shearing, partial dissociation of both the hydrophobic aggregates of the NPOD-PEG conjugate in the hydrogels and the inclusion complex between the mPEG chain and α-CD leads to a substantial decrease in the degree of cross-links.

Fig. 4 (a) Dynamic and (b) steady rheological behaviors of the precipitates made of mPEG (30 mg mL⁻¹), α-CD (A), NPOD-PEG (5 mg mL⁻¹), α-CD (B) and hydrogels made of α-CD, NPOD-PEG with various concentrations: (C) 10, (D) 30 mg mL⁻¹. For all samples [α-CD] = 90 mg mL⁻¹.

In biomedical applications, the existence of a porous structure of hydrogels is indispensable to allow for tissue growth and diffusion of drugs and nutrients.⁴¹ Therefore, to evaluate the network structure of the formed hydrogel, the freeze-dried hydrogel samples and crystalline complexes formed by α-CD and mPEG were examined by SEM. As shown in Fig. 5, the complexes formed by α-CD and mPEG show a typical disk-like crystalline structure, whereas all the NPOD-PEG/α-CD samples clearly demonstrate the presences of a typical porous structure (Fig. 5b–d). These might due to the fact that the introduction of NPOD would endow NPOD-PEG conjugates with amphiphilic property resulting in the increase of network density.

Fig. 5 SEM images of the crystal complex made of mPEG/α-CD (a) and hydrogels made of NPOD-PEG/α-CD with different NPOD-PEG concentrations, 5 mg mL⁻¹ (b), 10 mg mL⁻¹ (c) and 30 mg mL⁻¹ (d). For all samples [α-CD] = 90 mg mL⁻¹. Scale bar: 5 μm in (a), and 10 μm in (b)–(d).

pH responsiveness of NPOD-PEG conjugates and NPOD-PEG/α-CD hydrogel

The imine bond was chosen as the pH responsible linking bond for preparing polymer-drug conjugates (NPOD-PEG) because it is easy to be hydrolyzed in mildly acidic environment.³⁶ In this work, the pH responsiveness of NPOD-PEG with imine bond was investigated under two different pH values (5.0 and 7.4), and the variation tendency of NPOD-PEG conjugates was detected by HPLC. As shown in Fig. 6, the hydrolysis rate of NPOD-PEG conjugates is relatively slow at pH 7.4. At pH 5.0, however, the hydrolysis was dramatically accelerated, and NPOD-PEG conjugate was completely hydrolyzed and release NPOD within 0.5 h. These results indicate that NPOD-PEG are acid-sensitive and more likely cleavable at acidic endosome or tumor, while stable at physiological environment. As shown in Fig. 2f, the hydrogel became mobile and exhibited a gel–sol transition after the addition of a drop of diluted HCl. The rheological behavior of hydrogels treated with HCl was further investigated. As shown in Fig. S5,† the G′ and G′′ of hydrogel both dramatically decreased after being treated with HCl, and both G′ and viscosity were almost zero (Fig. S5†). It was might be accounted for the fact that the hydrolysis of imine bond between NPOD and PEG occurs under acidic environment destroying the supra-cross-links formed by hydrophobic aggregation of NPOD groups. Successful preparation of intelligent hydrogel provides an opportunity for site-specific drug delivery using pH changes as a trigger.

Fig. 6 The HPLC curves of NPOD-PEG at different pH values and time. (a) NPOD-PEG at pH 7.4, (A = 0 h; B = 0.5 h; C = 1.5 h); (b) NPOD-PEG at pH 5.0, (A = 0 h; B = 0.5 h).

In vitro release kinetics studies

The proposed hydrogels were further utilized for encapsulating another water-soluble anti-cancer drug DOX due to its highly hydrated, hydrophilic microstructures. The introduction of DOX has no effect on the hydrogel forming process (Fig. 2g). The DOX encapsulation and release behavior of NPOD-PEG/α-CD hydrogel were investigated under two different pH values (pH 5.0, 7.4). The encapsulation efficiencies (EE) of DOX were up to 93.36% (pH 5.0) and 92.18% (pH 7.4), respectively. Fig. 7 shows the in vitro release profiles of NPOD and DOX from NPOD-PEG/α-CD hydrogels under different pH values at 37 °C. The hydrogels showed a controlled release property, sustaining the release of DOX and NPOD at pH 7.4. As shown in Fig. 7a, only less than 40% of the DOX and NPOD were released from the supramolecular hydrogel within 23 h. NPOD was released from the hydrogel mainly by the slow hydrolysis of NPOD-PEG conjugates at pH 7.4, while for DOX, it is mainly caused by diffusion and the partial breakup of supra-cross-links.^16,28,42 When pH decreased to 5.0, the release of DOX was dramatically accelerated. As shown in Fig. 7b, about 100% of DOX was released from the hydrogels within 8 h. The release patterns suggested that the DOX release was not diffusion-controlled, but mainly caused by the internal structure change of the hydrogel. The erosion of the supramolecular hydrogels was induced by a significant hydrolysis of NPOD-PEG conjugates when the hydrogels were placed into a large amount of acidic buffer solution, which dramatically accelerates the release rate of encapsulated DOX. The release pattern of NPOD showed a slower and more steady release rate than those of DOX and NPOD releasing from free NPOD-PEG conjugates at pH 5.0 (Fig. 6 and 7b), suggesting that the formation of inclusion complexes between NPOD-PEG conjugates and α-CD could efficiently prevent the rapid hydrolysis of NPOD-PEG at acidic condition. Such pH-responsive dual drug encapsulation and release behavior seems to be generally feasible for site-specific drug combination with therapeutic windows.

Fig. 7 The release kinetics of NPOD and DOX at pH 7.4 (a); NPOD and DOX at pH 5.0 (b) from the NPOD-PEG/α-CD hydrogels at 37 °C.

In vitro cytotoxicity measurement

The cytotoxicities of NPOD, NPOD-PEG conjugates, NPOD-PEG/α-CD hydrogel and DOX-loaded NPOD-PEG/α-CD hydrogel were investigated in A549 human lung cancer cell line by MTT assay. As shown in Fig. 8, the results reveal that both NPOD-PEG conjugates and NPOD-PEG/α-CD hydrogel have a minimal effect on the cytotoxicity of NPOD. However, under the same conditions, the DOX-loaded NPOD-PEG/α-CD hydrogel shows an enhanced cytotoxicity on cancer cells in comparison with that of free NPOD in all the doses tested in A549 cells.

Fig. 8 In vitro cytoxicity of NPOD, NPOD-PEG, NPOD-PEG/α-CD and DOX-loaded NPOD-PEG/α-CD hydrogel to A549 lung cancer cells determined by MTT assay.

Furthermore, the combination index (CI) of combined drug interactions was analyzed using the median-effect/CI-isobologram equation which is based on the median-effect principle (Table S1 and S2) (ESI†).³⁵ Data were examined using median-effect analysis to determine the type of interactions between NPOD and DOX which occurred, different values of CI may be obtained for solving the equation for different values of f_a. CI < 1, CI > 1 and CI = 1 indicate synergy, antagonism and additive effect. As shown in Fig. 9, the interactions of DOX and NPOD in hydrogels showed non-dose-dependent properties. The combination index (CI) was ≈1 for A549 cells from f_a0.1 to f_a0.9 (1.02–1.09), indicating additive effect. The above results show that NPOD-PEG/α-CD formed hydrogels could produce additive anticancer effects for combination therapy and could serve as an effective multifunctional carrier to load two anticancer drugs.

Fig. 9 Combination index (CI) versus fraction affected (f_a) plots obtained from median-effect analysis of Chou-Talalay. CI < 1, CI > 1 and CI = 1 indicate synergy, antagonism and additive effect.

Conclusions

In summary, a pH-responsive supramolecular hydrogel was successfully prepared based on in situ host–guest interaction between a pH-sensitive amphiphilic NPOD-PEG and α-CD, which could load another water-soluble anticancer drug DOX to enhance their anticancer activity. The dual drug-loaded supramolecular hydrogel showed the pH-sensitive release behaviour and produced a non-dose-dependent additive cytotoxicity to cancer cells. It is believed that the multifunctional dual drug-loaded supramolecular hydrogel would provide an opportunity for site-specific drug delivery using pH changes as a trigger. Meanwhile, the study provided guidance for the combination therapy of podophyllotoxin analogues and DOX in clinical application.

Acknowledgements

We thank Dr. Yang Kang (Chengdu Insititute of Biology, CAS) for cytotoxicity analysis. This work was supported by the National Natural Science Foundation of China (nos 21105106, 21375136 and 21405164).

Notes and references

1. Y. Osada and J. Gong, Prog. Polym. Sci., 1993, 18, 187–226.
2. P. Gupta, K. Vermani and S. Garg, Drug Discovery Today, 2002, 7, 569–579.
3. F. Liu and M. W. Urban, Prog. Polym. Sci., 2010, 35, 3–23.
4. S. Sun and P. Wu, J. Mater. Chem., 2011, 21, 4095–4097.
5. Z. Liu, J. T. Robinson, X. Sun and H. Dai, J. Am. Chem. Soc., 2008, 130, 10876–10877.
6. X. Yang, X. Zhang, Z. Liu, Y. Ma, Y. Huang and Y. Chen, J. Phys. Chem. C, 2008, 112, 17554–17558.
7. J. Liu, Y. Li, Y. Li, J. Li and Z. Deng, J. Mater. Chem., 2010, 20, 900–906.
8. H. Wang and Z. Yang, Soft Matter, 2012, 8, 2344–2347.
9. J. H. Jeong, J. J. Schmidt, C. Cha and H. Kong, Soft Matter, 2010, 6, 3930–3938.
10. R. E. Kohman, C. Cha, S. C. Zimmerman and H. Kong, Soft Matter, 2010, 6, 2150–2152.
11. N. Murthy, Y. X. Thng, S. Schuck, M. C. Xu and J. M. Fréchet, J. Am. Chem. Soc., 2002, 124, 12398–12399.
12. V. Kozlovskaya, J. Chen, C. Tedjo, X. Liang, J. Campos-Gomez, J. Oh, M. Saeed, C. T. Lungu and E. Kharlampieva, J. Mater. Chem. B, 2014, 2, 2494–2507.
13. E. Frei, A. Elias, C. Wheeler, P. Richardson and W. Hryniuk, Clin. Cancer Res., 1998, 4, 2027–2037.
14. M. A. Azagarsamy and K. S. Anseth, Angew. Chem., Int. Ed., 2013, 52, 13803–13807.
15. S. W. Chen, R. Xiang, J. Liu and X. Tian, Bioorg. Med. Chem., 2009, 17, 3111–3117.
16. W. Ha, J. Yu, X. Y. Song, Z. J. Zhang, Y. Q. Liu and Y. P. Shi, J. Mater. Chem. B, 2013, 1, 5532–5538.
17. L. Zhao, Y. Cheng, J. Hu, Q. Wu and T. Xu, J. Phys. Chem. B, 2009, 113, 14172–14179.
18. L. Liao, J. Liu, E. C. Dreaden, S. W. Morton, K. E. Shopsowitz, P. T. Hammond and J. A. Johnson, J. Am. Chem. Soc., 2014, 136, 5896–5899.
19. L. Mao, H. Wang, M. Tan, L. Ou, D. Kong and Z. Yang, Chem. Commun., 2012, 48, 395–397.
20. N. Lin and A. Dufresne, Biomacromolecules, 2013, 14, 871–880.
21. J. Li, NPG Asia Mater., 2010, 2, 112–118.
22. H. Wei, H. Yu, A. Y. Zhang, L. G. Sun, D. Hou and Z. G. Feng, Macromolecules, 2005, 38, 8833–8839.
23. T. Zhao and H. W. Beckham, Macromolecules, 2003, 36, 9859–9865.
24. A. Harada, J. Li and M. Kamachi, Nature, 1994, 370, 126–128.
25. X. Liao, G. Chen, X. Liu, W. Chen, F. Chen and M. Jiang, Angew. Chem., Int. Ed., 2010, 122, 4511–4515.
26. R. Yuan and X. Shuai, J. Polym. Sci., Part B: Polym. Phys., 2008, 46, 782–790.
27. B. Jeong, Y. H. Bae, D. S. Lee and S. W. Kim, Nature, 1997, 388, 860–862.
28. W. Ha, J. Yu, X.-Y. Song, J. Chen and Y. P. Shi, ACS Appl. Mater. Interfaces, 2014, 6, 10623–10630.
29. C. P. Belani, L. A. Doyle and J. Aisner, Cancer Chemother. Pharmacol., 1994, 34, S118–S126.
30. K. R. Hande, Eur. J. Cancer, 1998, 34, 1514–1521.
31. Y. Zheng, Y. G. Sun, Y. Zhang, J. J. Lv, W. C. Bi, Y. W. Ma and H. Chen, Chin. Chem. Lett., 2009, 20, 1431–1434.
32. A. Kamal, J. R. Tamboli, M. V. P. S. Vishnuvardhan, S. F. Adil, V. L. Nayak and S. Ramakrishna, Bioorg. Med. Chem. Lett., 2013, 23, 273–280.
33. G. Takemura and H. Fujiwara, Prog. Cardiovasc. Dis., 2007, 49, 330–352.
34. T. C. Chou, J. Theor. Biol., 1976, 59, 253–276.
35. T. C. Chou and P. Talalay, Adv. Enzyme Regul., 1984, 22, 27–55.
36. S. Xu, Y. Luo and R. Haag, Macromol. Biosci., 2007, 7, 968–974.
37. A. Harada, J. Li and M. Kamachi, Macromolecules, 1993, 26, 5698–5703.
38. A. Harada and M. Kamachi, Macromolecules, 1990, 23, 2821–2823.
39. J. Li, X. Li, Z. Zhou, X. Ni and K. W. Leong, Macromolecules, 2001, 34, 7236–7237.
40. J. Li, X. Li, X. Ni, X. Wang, H. Li and K. W. Leong, Biomaterials, 2006, 27, 4132–4140.
41. J. He, M. Zhang and P. Ni, Soft Matter, 2012, 8, 6033–6038.
42. K. LiáLiu, Soft Matter, 2011, 7, 11290–11297.

---

Footnote

† Electronic supplementary information (ESI) available: ¹H NMR spectrum of PEG formylbenzoic acid ester; ¹H NMR and ¹³C NMR spectrum of NPOD-PEG; calibration curve of DOX; the dependence of the viscoelastic moduli on frequency for NPOD-PEG/α-CD hydrogel samples; dynamic and steady rheological behaviors of the diluted HCl-treated NPOD-PEG/α-CD hydrogel; dose-effect relationship parameters for NPOD and DOX in cancer model; combination indices at different effect levels. See DOI: 10.1039/c4ra11311j

---