Synthesis of high drug loading, reactive oxygen species and esterase dual-responsive polymeric micelles for drug delivery

Stimulus-responsive, controlled-release systems are of great importance in medical science and have drawn significant research attention, leading to the development of many stimulus-responsive materials over the past few decades. However, these materials are mainly designed to respond to external stimuli and ignore the key problem of the amount of drug loading. In this study, exploiting the synergistic effect of boronic esters and N-isopropylacrylamide (NIPAM) pendant, we present a copolymer as an ROS and esterase dual-stimulus responsive drug delivery system that has a drug loading of up to 6.99 wt% and an entrapment efficiency of 76.9%. This copolymer can successfully self-assemble into polymer micelles in water with a narrow distribution. Additionally, the measured CMC hinted at the good stability of the polymeric micelles in water solution, ensuing long circulation time in the body. This strategy for increasing the drug loading on the basis of stimulus response opens up a new avenue for the development of drug delivery systems.


Introduction
In recent decades, multifunctional polymeric materials have attracted growing interest due to their wide range of applications, especially in the biomedical eld. 1 These polymeric materials achieve better control of architecture 2 and functionality via the copolymerization of different monomers. Among the numerous polymerization methods, reversible additionfragmentation chain transfer (RAFT) polymerization has proven to be an excellent and convenient route to block copolymers via the use of the desired RAFT agents. 3 Meanwhile, by selectively incorporating certain functional groups, desired amphiphilic block copolymers or responsive materials can be synthesized using RAFT methods. In aqueous solutions, amphiphilic block copolymers can self-assemble into a variety of different nanostructures, including spherical micelles, cylindrical micelles, and drug-delivery vesicles. 4 These smart materials show mono-, dual-or multi-responsive abilities and can efficiently response to external stimuli. 5 For example, Qin et al. prepared a boronic acid block copolymer whose morphology is controlled by the regulation of the pH. 6 Sumerlin et al. prepared pH-or sugar-responsive block copolymers via RAFT polymerization. 7 Sumerlin's group also combined pHand sugar-responsiveness with the thermo-responsiveness of NIPAM to prepare block copolymers. 1 However, biological applications have not been explored in their studies. And according to literature, esterase are expressed in numerous tissues. 8 And a series of esterase-responsive probes have been designed. 9 In 2016, Shen's group taking advantage of the high esterase activity in HeLa cancer cells to design a esteraseresponsive charge-reversal polymer, whose polyplexes had a selective gene expression in the cancer cells high in esterases. 10 Generally, although pH, glucose-sensitive or multiplesensitivity boronic acid block polymers have already been reported, high drug loading, ROS and esterase-responsive materials for drug delivery have not been synthesized or investigated to date.
Herein, we constructed a di-block copolymer system containing boronic esters and N-isopropyl acrylamide via RAFT methods that offers genuine synergies between the esterase and ROS stimuli. Arylboronic esters could be oxidized by ROS (e.g., H 2 O 2 ) 11 and then undergo rearrangement to unmask the modied group, and the ester bonds could be hydrolysed by esterase. 12 Hydrogen peroxide (H 2 O 2 ) is one of the most important ROS and plays critical roles in a number of physiological and pathological processes. 13 This reactive strategy has been usually applied to design small-molecule-based uorescent probes in recent years. 14 In addition to dual-stimulus responsiveness, phenylboronic acid, pinacol ester pendent can function as an electron acceptor that affords donoracceptor coordination with doxorubicin to obtain micelles with a high drug-loading ability. 15 This kind of co-polymer can selfassemble into novel polymeric micelles that show low cytotoxicity, low critical micelle concentration, high drug-loading and dual response to external stimuli. Moreover, in vitro drugrelease studies revealed that this polymeric system can serve as a novel controlled-release system (Scheme 1).
Anti-HER2 Affibody was purchased from Affibody (Solna, Sweden). 1 H NMR and 13 C NMR spectra were measured using a Bruker AM 400 NMR spectrometer. Proton chemical shis of the NMR spectra were given in ppm relative to the internal TMS reference ( 1 H, 0.00 ppm). ESI-HRMS spectral data were recorded using a Finnigan LCQDECA mass spectrometer. Fluorescence emission spectra were obtained using a Hitachi F-7000 spectrometer at 298 K. Gel permeation chromatography (GPC) was carried to conrm the polymer molecular weights and their distributions using a Waters HPLC system equipped with a model 1515 isocratic pump, a 717 plus autosampler, and a 2424 refractive index (RI) detector with Waters Styragel® HT3 and HT4 columns in series. The eluting solvent was THF at a ow rate of 1.0 mL min À1 at 45 C. The retention times were calibrated against a poly(ethylene glycol) standard with a molecular weight range of 600-80 000 Da. Transmission electron microscopy (TEM, Hitachi H-600) at an acceleration voltage of 100 kV was performed to investigate the micelle morphology. Dynamic light scattering (DLS) measurements were carried out at 25 C using a Zetasizer Nano-ZS90 system from Malvern Instruments equipped with a 633 nm HeNe laser using backscattering detection with a xed detector angle of 90 . Confocal lasing scanning microscopic (CLSM) images of single-photo were obtained using LSM 780 (Zeiss) and Nikon A1R MP + , and FTIR spectra were recorded using a Shimadzu FTIR-4200 Scheme 1 Self-assembly properties and stimuli-responsive release properties of mPEG 2000 -PNIPAM-co-PBAPE (P3).

Synthetic procedures and characterization
Synthesis of the CTA-COOH. First, a small-molecule-based RAFT agent was synthesized according to the reported method. 16 Generally, a three-neck ask was cooled below 10 C with a cold circulating pump while removing the air contained in the ask with nitrogen. Then, 1-dodecanethiol (12.11 g, 59.85 mmol), acetone (28.86 g, 496.90 mmol), and Aliquot 336 (tricaprylylmethylammonium chloride, 0.97 g, 2.41 mmol) were added, followed by the dropwise addition of a sodium hydroxide solution (50 wt%, 5.03 g). The reaction mixture was stirred for another 15 min before carbon disulphide (4.563 g, 60 mmol) in acetone (6.054 g, 103.5 mmol) was added, and during this time, the mixture turned red. Ten minutes later, chloroform (10.688 g, 90 mmol) was added in one portion, followed by the dropwise addition of a 50% sodium hydroxide solution (24 g, 300 mmol) over 30 min. The reaction mixture was stirred overnight, and 90 mL of water was added, followed by 15 mL of concentrated HCl to acidify the aqueous solution. Nitrogen gas was purged through the reactor with vigorous stirring to assist the evaporation of acetone. The solid was collected through ltering and then was dissolved in 150 mL of 2-propanol. The undissolved solid was ltered off, and the rest of the 2-propanol solution was concentrated and dried, and the resulting solid was recrystallized from n-hexane to obtain 12.4 g of yellow crystalline solid (yield: 57% Synthesis of the marco-CTA-COOH (P1). The CTA-COOH was conjugated on the pendant of mPEG 2000 -OH through an ester-ication reaction. 17 Briey, mPEG 2000 -OH (1.9397 mmol) and DMAP (1.4218 g, 11.6382 mmol) were dissolved in anhydrous DCM (20 mL) and placed in a 50 mL round-bottom ask in an ice bath, followed by the addition of DCC (0.2397 g, 1.1638 mmol) with a magnetic stirring bar, and CTA-COOH (0.7073 g, 1.9397 mmol) was dissolved into anhydrous DCM and dropped slowly into the ask. The reaction was carried out for 2 h in an ice bath and allowed to reach room temperature for another 24 h with stirring. Then, the white precipitate was ltered out, and the remaining ltrate of the P1 was precipitated in cold diethyl ether (40 mL) twice in order to remove the unreacted small molecule compounds. Aer purication and lyophilisation, the macro-CTA-COOH (P1) was obtained as a light yellow solid (4.10 g). 1 H NMR (400 MHz, CDCl 3 ) d 4.28-4.21 (t, 2H), 3.28-3.21 (t, 2H), 1.69 (s, 6H), 1.24 (m, 18H), 0.88 (t, J ¼ 11.6, 5.0 Hz, 3H. Additionally, the synthesis of P1 was also conrmed by matrix-assisted laser desorption ionization time-of-ight mass spectrometry (MALDI-TOF MS), and the matrix was prepared at a concentration of 10 mg mL À1 .
Synthesis of the mPEG 2000 -PNIPAM (P2). The polymer mPEG 2000 -PNIPAM (P2) was synthesized by reversible additionfragmentation chain transfer polymerization. 18 Briey, N-isopropylacrylamide (1.107 g, 9.78 mmol) was initially dissolved in 20 mL of 1,4-dixone in a Schlenk tube under ultrasonication, followed by dissolving macro-CTA-COOH (1.2 g, 0.51 mmol), and AIBN (2 mg) into this mixture. Aer three freeze-pumpthaw cycles, the polymerization was allowed to proceed at 70 C for 24 hours. Then, the polymerization was quenched by cooling the system, and the resulting solution was precipitated with excess amount of cold diethyl ether three times. The precipitated copolymer was collected by ltration and was subjected to dialysis against ultrapure water for 48 h. A white powder was nally obtained with a yield of 2.04 g aer freeze drying.
Synthesis of the B1-AC. Acrylate-ended B1 (B1-AC) was synthesized according to the references with a few modications. 19 In brief, 4-hydroxymethylphenylboronic acid, pinacol  Paper ester monomer (7.39 mmol) was dissolved into anhydrous DCM at 20% solids concentration in a Schlenk ask. TEA (11.09 mmol, 1.5 eq.) was added with gentle stirring, and the ask was immersed in an ice bath, followed by the dropwise addition of acryloyl chloride (11.09 mmol, 1.5 eq.) for 10 min. The reaction was carried out for 2 h in an ice bath, and the reaction mixture was allowed to reach room temperature for further 24 h with stirring. Aer washing with dilute HCl solution three times and precipitating, the solids were dried under vacuum for 48 h, and then the crude product was further puried by column chromatography by using ethyl acetate/ Synthesis of the mPEG 2000 -PNIPAM-co-PBAPE (P3). The synthesis steps for obtaining P3 were very similar to those used to obtain P2. Briey, B1-AC (1.107 g, 9.78 mmol) was initially dissolved in 1,4-dixone (20 mL) in a Schlenk tube under ultrasonication, followed by dissolving P2 (1.2 g, 0.51 mmol), and AIBN (2 mg) into this mixture. Aer three freeze-pump-thaw cycles, the polymerization was allowed to proceed at 70 C for 24 hours. Then, the polymerization was quenched by cooling the system, and the resulting solution was precipitated with excess cold diethyl ether three times. The precipitated copolymer was collected by ltration and was subjected to dialysis against ultrapure water for 48 h. A white powder was nally obtained with a yield of 1.04 g aer freeze drying.
IR-characterization. The chemical structures of the assynthesized P2 and P3 were further analysed using a Shimadzu FTIR-4200 spectrometer.

Preparation of polymeric micelles
Preparation of polymeric micelles P3. Micelles of P3 were prepared using the dialysis method. The polymers (30 mg) were dissolved in 6 mL DMF. Then, 5 mL deionized water was added to the solution under gentle stirring. Aer stirring for 2 h at room temperature, the solution was transferred into a dialysis membrane (MWCO 2000 Da) and dialyzed for 48 h against deionized water. The outer phase was replaced with fresh deionized water every 6 h. The resulting solution was nally lyophilized. The preparation method of micelles P2 is very similar to that for P3, and specic experimental steps are shown in ESI. † Preparation of DOX-loaded polymeric micelles P3. Here, we identied a common anticancer drug, doxorubicin (DOX), as a model drug, and the water-insoluble DOX was encapsulated into P3 micelles by adding 1.5 mL of 2 mg mL À1 DOX in DMF to 6 mL of P3 solution in DMF (5 mg mL À1 ), followed by stirring at room temperature for two hours in absolute darkness condition to stabilize the system, and then, the resulting solution was placed in a dialysis membrane (MWCO 2000 Da) and dialyzed for 48 h against deionized water, with constant stirring and frequent water changes. Then, the solution was ltered through a 0.45 mm syringe lter, and dynamic light scattering (DLS) measurements were conducted at 25 C. The preparation method of DOX-loaded polymeric micelles P2 is very similar to that for P3, and specic experimental steps are shown in ESI. †

Critical micelle concentrations determination
The critical micelle concentration (CMC) was determined using Nile Red as the uorescence probe. The block copolymer concentration varied from 1.0 Â 10 À6 mg mL À1 to 0.2 mg mL À1 , and the Nile Red concentration was xed at 1.0 Â 10 À6 M. The uorescence spectra were recorded using a Hitachi F-7000 spectrometer at 298 K. Both the emission and excitation slit widths were 5 nm. The samples were excited at 560 nm, and the emission spectra were recorded from 580 to 800 nm. The emission uorescence values, I 631.6 at 631.6 nm was used for the subsequent calculations. The CMC was determined from the plots of the I 631.6 versus the logarithm of the polymer concentration using the intersection of the linear regression lines as the CMC value.

TEM and DLS characterization
The micelle solutions of P3 and DOX-loaded P3 at a concentration of 0.5 mg mL À1 were prepared as described above. To further study the formation of the micelles, their morphologies and sizes were also characterized by TEM and DLS measurements.
Entrapment efficiency and drug loading capacity determination. The standard curve for the determination of the free DOX content was measured by the uorescence method with a Hitachi F-7000 spectrometer. The concentration of free DOX was calculated by the standard curve (Fig. S8 †). The encapsulation efficiency and loading of free DOX were obtained by the following formulas. 20 LCð%Þ ¼ weight of loaded drug weight of drug loaded micelle Â 100% (1) EEð%Þ ¼ weight of loaded drug weight of drug in feed Â 100% (2) In vitro drug release experiments. The release test of the polymeric micelle nanocarrier platform was determined by the dialysis method (MW: cut-off: 3.5-5 kDa) at 37 C. In short, dialysis membrane carrying 2.0 mL of DOX-loaded polymeric micelle (P3/DOX) were immersed in 20 mL of PBS solution (0 mM H 2 O 2 , 10 mM H 2 O 2 , 10 mM H 2 O 2 and 30 mg mL À1 esterase, 0.2 mg mL À1 ) in a conical ask in the shaking incubator with the stirring speed of 100 rpm while the bulk polymeric micelle immersed in the PBS solution (0 mM H 2 O 2 and no esterase, 0.2 mg mL À1 ) under the same conditions were used as control groups. At predetermined intervals, aliquots of media (1.0 mL) were taken and quantied by measuring at 500 nm with the Hitachi F-7000 spectrometer. Throughout the entire process, the total volume was kept constant.

Cell imaging experiments
For uorescence imaging, cells (4 Â 10 3 /well) were passed on a 6-well plate and incubated for 48 h. Prior to the staining experiment, the cells were washed twice with physiological saline, incubated with 14 mg L À1 DOX-loaded polymeric micelle P3/DOX for different times (2 h, 4 h, 8 h, 12 h, 24 and 48 h) at 37 C, and then were washed twice with physiological saline. The confocal uorescent images were captured with an excitation light at 488 nm. At the same time, DOX and DOX-HCl under the same conditions were used as control groups (1 mg L À1 ).

In vitro cytotoxicity assay
The materials' toxicity towards HeLa cells was determined by MTS Cell Proliferation Colorimetric Assay Kit following the procedures described in the literature. Approximately 104 cells per well were seeded in 96-well plates and cultured overnight for 70-80% cell conuence. The medium was replaced with 100 mL of fresh medium with different concentration of the materials, to which 100 mL complexes were added at 200 mL. Aer 48 hours, 100 mL of 20% MTS solution in PBS was replaced with the old medium in each well for additional 0.5 h incubation. The metabolic activity of the probe-treated cells was expressed relative to untreated cell controls taken as 100% metabolic activity. The cytotoxicity study towards HL-7702 cells was carried out following the same procedure.

Results and discussion
Synthesis and characterization P1 was synthesized via esterication. As shown in Fig. S1, † the reversible addition-fragmentation chain transfer (RAFT) agent CTA-COOH was graed with mPEG 2000 -OH by the esterication reaction using DCC and DMAP as the coupling agent and catalyst, respectively. Fig. S2 † shows the matrix-assisted laser desorption ionization time-of-ight mass spectrum (MALDI-TOF MS) of P1. The MALDI-TOF spectrum of P1 has systematic peaks centred at $1990 m/z, extending from 1400 to 2500 m/ z. The difference in the m/z values between the neighbouring peaks is 44.036, corresponding to the molar mass of the mPEG 2000 unit. Then, the monoblock polymer P2 and diblock polymer P3 were successfully synthesized by using the macromolecular chain transfer reagent P1. P2, and P3 were synthesized and detected, as shown in Table 1. We can also calculate the degree of polymerization of polymer P2 and P3 according to the results of NMR. Fig. S3 † shows the particle size distributions of P2 micelles and we carried out the thermal responsive properties of materials P2 as it shown in Fig. S4 and S5 † shows the macro change of P2 temperature sensitivity. In addition, Fig. S6 † also examines the temperature-sensitive properties of P3. In the present study, P1 was graed as the water-soluble part by conjugating it to the N-isopropylacrylamide segment, and then the polymer was also modied by boronic acid, pinacol ester on the side chain introduced to obtain ROS sensitivity and increase the drug loading.

IR-characterization
As shown in Fig. 1, in the P2 spectrum, the peaks at 3465, 3420, and 3338 cm À1 correspond to the N-H stretching vibration of the -NHgroup, whereas the weak band at 3035 cm À1 is due to the aromatic C-H stretching vibration, the peaks at 2960-2850 cm À1 correspond to the alkane C-H vibration, the band at 1645 cm À1 is due to the C]O stretching vibration in amide I bonds, and the peaks at 1543 cm À1 can be ascribed to the N-H stretching vibrations of the amide II bonds. The peaks at 1389 and 1365 cm À1 are induced by the deformed vibration absorption peak of the two methyl in -CH (CH3) 2 -, while the two peaks at 1170 and 1130 cm À1 are associated with the C-O stretching mode in the ester and the absorption peak of the PEG ether bond at 1060 cm À1 covers the C]S bond absorption. These characteristic peaks prove that the N-isopropylacrylamide monomer was successfully polymerized onto the polymer chain. Additionally, in the P3 spectrum, the strong peaks at 833-810 cm À1 are derived from 1,4-two substituted benzene, while the peak at 1730 cm À1 is ascribe to the overtone bands.

Characterization of polymeric micelles P3 and DOX-loaded P3
Because the amphiphilic polymer P3 can self-assemble into micelles in aqueous solutions, to the best of our knowledge, the micelles can self-assemble above a certain threshold concentration, leading to a rapid increase in the emission intensity of Nile Red, as shown in Fig. S7. † The measured CMC values hinted at the good stability of these polymeric micelles in solution owing to their low CMC values, even aer extreme dilution by the larger volume of systemic circulation in vivo.
Furthermore, as shown in Fig. 1(c and d), TEM characterization reveals that most of the polymer particles were spherical, further supporting the conclusion that polymeric micelles were formed. The results of the DLS measurements are also presented in Table 1. The results showed that the P3 spherical micelles had average diameters of 60-70 nm whereas the DOXloaded P3 had the average diameters of 100-110 nm. These observations provided clear evidence not only for the micellization of P3 in water, similar to the results of the CMC analysis did, but also for their rather uniform morphology and size.

Entrapment efficiency and drug loading capacity determination
The entrapment efficiency and loading capacity of free DOX of P3 were calculated by the standard curve, according to the following equation. Entrapment efficiency ¼ 76.9%; loading capacity ¼ 6.99 wt%. We also investigated the entrapment efficiency and loading capacity of P2, and the data were shown in Fig. S9. †

In vitro release of DOX from prodrug DOX-loaded P3
In vitro drug release performances of the DOX-loaded P3 under physiological conditions (PBS, pH 7.4, control), oxidation environment only (PBS, 10 mM H 2 O 2 ) and both oxidation and esterase (PBS, 10 mM H 2 O 2 with 30 mg mL À1 esterase) conditions were investigated, as shown in Fig. 2. The drug release rate was signicantly inuenced by both H 2 O 2 and esterase, providing an additional indication of the dual-stimulussensitivity of DOX-loaded P3. At blank conditions, the prodrug DOX-loaded P3 show weak release, and the cumulative release amount of DOX was only 30% for 54 h, resulting from the stable and compact DOX-loaded P3 at normal physiological conditions. In 10 mM H 2 O 2 , the cumulated release of 54 h drugs was 30% higher than that in the blank group, as shown in Fig. 2(a), and the TEM image of the release uid presented in Fig. 2(c) shows that these micelles were adhesive and some particles are smaller than those observed in Fig. 2(b). In contrast, in the presence of both H 2 O 2 and esterase, the drug release rate was accelerated obviously, and the cumulative release amount reached 90%, and as shown in Fig. 2(a), these micelles were completely destroyed as observed from the TEM image presented in Fig. 2(d). Furthermore, as shown in Fig. S10 and Table  S1, † we investigated the particle size changes of P3 under different in vitro conditions. Compared to P3, the size of the particles increased with the addition of H 2 O 2 and the addition of both esterase and H 2 O 2 . Meanwhile, we investigated the in vitro release in the presence of different esterase concentrations, and as shown in Fig. S11(a), † it was found that compared to the control group, the amount of the release did not increase signicantly; this is in line with the TEM observations shown in Fig. S11 (b). † This may be due be to the rearrangement of ROSsensitive DOX-loaded P3 pendant and the cleavage of the esterase-sensitive main chain on the copolymer.

In vitro cellular uptake of DOX
The intracellular uptake of DOX-HCl, free DOX and DOX-loaded P3 was investigated using confocal lasing scanning microscopy (CLSM) (Fig. 3). When the cells were incubated with DOX-HCl, the cellular uptake of DOX-HCl was the same, regardless of the different time duration of 2, 4, 8, 12, 24 and 48 h. DOX-HCl was distributed in the nuclear region within 2 h of incubation, and strong uorescence was observed in the nucleus, as reported previously. 21 From column DOX-HCl (P)-DOX-HCl (R), aer 48 h incubation, it was clearly observed that the HeLa cell morphology was completely destroyed. However, when the cells were incubated with free DOX, the cellular uptake of free DOX was also the same, regardless of different time durations of 2 h, 4 h, 8 h, 12 h, 24 h and 48 h, and free DOX was distributed in the cytoplasm region during long-time incubation, with the aggregation of free DOX with increasing time. Moreover, it was difficult to for the free DOX to transfer into the nucleus even aer 48 h. When the cells were incubated with DOX-loaded polymeric micelle P3 in two hours, considerable uorescence intensity was detected mainly in the cytoplasm, suggesting novel cellular uptake ability of DOX-loaded polymeric micelle P3, and with the prolongation of the incubation time, the cellular uorescence was much higher. We found that aer 4 hours, the DOX molecules were localized in the cell nuclei, and as expected from the drug release in vitro experiment, DOX localized in the cell nuclei is likely to be intercalated into DNA strands, thereby showing its toxicity against tumour cells. This dual cellular uptake prole of DOX molecules is likely due to the rapid demicellization of the DOX-loaded polymeric micelles under cellular conditions, leading to the rapid release of DOX from polymeric micelles and its subsequent internalization into cell nuclei. As shown in Fig. S12, † the average uorescence intensity in the nucleus is a measure of the amount of DOX that enters the nucleus. When the cells were incubated with DOX-HCl, the intensity mean value was approximately 80% aer 24 h incubation, and with the incubation time extended to 48 h, the average uorescence intensity of HeLa nuclei was higher than 200%. However, even aer 48 hours, the average uorescence intensity of the cells incubated with free DOX remained at 20%. The value obtained for DOX-loaded P3 was between these values for free DOX and DOX-HCl and increased with incubation time, so that aer 48 h, the uorescence intensity increased signicantly to 80%. These ndings suggested that DOX-loaded polymeric micelles may be suitable for stimulus-responsive nanocarriers.

In vitro cell cytotoxicity
Cytotoxicity is a critical factor for a carrier, especially for in vivo drug delivery. Fig. 4 shows the in vitro cytotoxicity results for P3 using the MTS assay method in both HeLa and HL-7702 cells. As shown in Fig. 4(a), all of the concentration of the copolymer exhibited low cytotoxicity towards HeLa and HL-7702 cells. The cell viability was high even at a high polymer concentration (200 mg L À1 ), indicating that the amphiphilic gra copolymer of P3 is highly biocompatible and nontoxic. Furthermore, Fig. 4(b) shows that the in vitro cytotoxicity values for cell viability obtained using the MTS assay method in HeLa cells were decreased by DOX-loaded polymeric micelles P3, and the cytotoxicity of DOX-HCl and free DOX was examined by the same method, as shown in Fig. S13. † All of the concentration of the free DOX exhibited low cytotoxicity towards HeLa cells, while the cell viability decreased sharply by 95% for DOX-HCl, while for DOX-loaded polymeric micelles P3, the cell viability decreased sharply by 70% aer 48 h of incubation; these results provide further proof that DOX-loaded polymeric micelles P3 have sustained release effect on DOX.

Conclusions
In summary, we have successfully constructed a novel highdrug-loading, dual stimulus-responsive drug-release system based on a rearrangement of borate ester bond and a hydrolysis of an ester bond. Compared to the common drug control release systems, P3 exhibits high entrapment efficiency and drug loading capacity, and the synergetic characteristics of two stimuli. Such a dramatic loading capacity and synergetic characteristics is anticipated to achieve unprecedented advances in the development of stimuli-sensitive drug-delivery systems that are deemed to be a promising vehicle for drug-delivery nanocarriers.

Conflicts of interest
There are no conicts to declare.