Introduction

Cancer has been being a prevailing threat to human health owing to high mortality and severe incidence [1]. The therapeutic strategy of tumor mainly includes chemotherapy, immunotherapy, surgical resection and radiation therapy [2, 3]. Compared with other therapeutics, chemotherapy is a preferential strategy due to the less body injuries and remarkable clinical curative effect [4, 5]. However, conventional anticancer drugs are always limited by their instability, poor solubility in water and severely toxic side effects to normal cells/tissues including emesis, fever, anaphylaxis, baldness and other clinical manifestation [6]. Nanosized drug carriers (NSDCs) (with 10–100 nm size) have attracted more attention due to those NSDCs can effectively overcome those chemotherapeutic obstacles by regulating the entering way and distribution of drugs in body [7,8,9,10]. However, the severe side effects to normal cells and tissues remain a serious issue, which is mainly attributed to the undesired and nonspecific release of drugs.

Especially, polymeric NSDCs with popular amphiphilic structure, high drug loading and controlled drug release behavior have obtained comprehensive interest and a series of polymeric nanostructures based on liposomes [11, 12], polymersomes [13], polymeric micelles [14,15,16,17,18,19,20,21,22,23,24] and nanogels [25, 26] have been reported. Notably, stimuli-responsive NSDCs in responded to the intrinsic microenvironment, such as low pH [27,28,29,30], high glutathione (GSH) concentration [31,32,33] and other stimulus [34], may offer an effective method by improving the drug efficacy while reducing the nonspecific toxicity. For this purpose, our group have previously reported a novel pH-responsive doxorubicin (DOX) prodrug which can effectively improve the drug loading content of polymeric prodrug and restrained the early release of drugs [28]. Recently, Hu and co-workers designed one self-reporting theranostic platform with reduction-sensitive drug release and intensive antitumor-capacity against tumor cells [31]. Duan and his co-workers recently reported a dual responsive drug delivery system which could effectively inhibit the propagation of tumor through controlling the drug release behaviors [35]. To render these theranostic platform with target function, targeting moieties, such as sugars, folate acid and peptides, could further realize targeted cancer therapy [36,37,38,39]. Especially, folic acid (FA) showed the specific binding interaction between FA molecules, and FA-receptor on the surface of cells has widely studied for targeted drug delivery since it firstly was extracted from yeast [40,41,42,43,44]. Guo and his co-workers designed a polymeric prodrug by attaching FA to the polymer of DOX-conjugated poly(ethylene glycol)-poly(ε-caprolactone), which presented favorable site-specific anticancer therapy [45]. These aforementioned therapeutic nanoplatforms could minimize the primitive drug-leakage at normal cells and selectively released drugs at tumor site, resulting a high promising strategy for cancer therapy.

To explore more resultful NSDCs, we presented herein the synthesis of acid-activatable amphiphilic polymeric prodrug via reversible addition–fragmentation chain transfer polymerization (RAFT). Compared with other previously reported method about polymeric micelles derived from amphiphilic block copolymer, the present design possesses some unique advantages as drug delivery systems: (1) the obtained prodrug have a ultra-high drug loading amount as high as 61.5 wt%; (2) the prodrug micelles show robust stability in physiological environment and pH-triggered drug release pattern; (3) the obtained prodrug show a selectively high toxicity against tumor cells than normal cells, indicating an instinctive tumor-specific targeting ability. Specifically, the resulting prodrug had a hydrophilic block of poly[(ethylene glycol) methyl ether methacrylate] (POEGMA) with/without FA decoration and hydrophobic block of DOX drug and was denoted as OM@DOX and FA-OM@DOX (Scheme 1), respectively. Obviously, the hydrophilic POEGMA block presented a potential to enable these prodrugs to form micelles in aqueous solution. The subsequent results also demonstrated the micelles possessing a high micellar stability and pH-triggered drug release due to the acid milieu-responsive hydrazine bond existed in the prodrugs. The collective cytotoxicity results revealed the OM@DOX prodrug showed a higher toxicity for the tumor cell of HeLa than the normal cell of L929. Excitingly, FA-OM@DOX further presented preferentially cytotoxicity for KB cells than A549 cells, and which might be due to the rich FA-receptor on membrane of the former. The subsequent in vitro cellular internalization and cellular uptake demonstrated the micelles could selectively realize the drug accumulation in tumor cells, resulting a promising nanoplatform for cancer therapy.

Scheme1
scheme 1

a Schematic diagram of amphiphilic FA-OM@DOX polymeric prodrug and the self-assembly behavior; b The proposed mechanism of pH-triggered DOX release; c The folic acid (FA) receptor-mediated cellular uptake and pH-triggered drug release mechanism

Full size image

Experimental section

Regents and materials

All biological reagents including Dulbecco’s Modified Eagle’s Medium (DMEM), TrypLE™ Express Enzyme, Penicillin–Streptomycin Solution, Triton X-100,Alexa Fluors®633 Phalloidin,DAPI and Lyso Tracker®Red were acquired from the Life Technologies of China. The chemicals embracing folate (FA), 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC·HCl), N–BOC-1,6-hexanediamine hydrochloride (NH2–BOC·HCl), methyl glycolate, N-Hydroxysuccinimide, methacryloyl chloride, trifluoroacetic acid (TFA), trimethylamine (TEA) and anhydrous solvent including dimethyl sulfoxide (DMSO), 1,4-dioxane, N, N-dimethylformamide (DMF) were acquired from Sigma-Aldrich (USA) except for doxorubicin hydrochloride (DOX·HCl), which was attained from Beijing HuaFeng United Technology CO. Ltd (China). The deionized (DI) water obtained from a water purification system of Millipore. The other solvents with analytical grade were used directly without any further treatment.

Instruments

A Bruker AV 600 NMR spectrometer (Rheinstetten, Germany) was used to record the 1H NMR spectra of the polymers; Gel permeation chromatography (GPC, Agilent 1260, USA) was employed to notarize the molecular weights (M w) and Polydispersity (PDI); Dynamic light scattering (DLS, Nano-ZS90, Malvern, UK) was used to determine the size distribution and stability of micelles. Transmission electron microscopy (TEM, JEM-2100, Japan) were employed to measure the size and morphology of micelles. A Tensor 27 Fourier transform infrared spectroscopy (FT-IR) was used to study the molecular structure and chemical bonds; A Cary Eclipse fluorescence spectrophotometer (RF-5301PC, Shimadzu, Japan) was used to measure fluorescence emission spectra and fluorescence excitation spectra, the latter could be used to analyze the critical micelle concentration; A Lambda 35spectrometer (UV-2550, Shimadzu, Japan) was used to record UV–vis absorbance spectra; A synergy H1 Microplate reader (Tecan SPARK-10M) was used to measure the accumulation of drug release and cell viability; a confocal laser scanning microscope (CLSM, Zeiss LSM 880, Germany) was used to record cellular imaging; A flow cytometry from ACEA biosciences (FAC, NovoCyte 2060R) was used to detect the cellular uptake at a certain concentration and different time points.

Synthesis of FA–NH2

Typically, folic acid (FA, 1.0 g, 2.3 mmol) was dispersed in a solvent of 40 mL anhydrous DMF under the protection of argon gas, and then a mixture of 1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC·HCl, 440 mg, 2.4 mmol) and N-hydroxysuccinimide (NHS, 260 mg, 2.3 mmol) was added, and the mixture was stirred for 0.5 h under an ice bath. After that, a mixture of NH2–BOC.HCl (615 mg, 2.5 mmol) and TEA (2.4 mmol, 336 uL) in 5.0 mL DMF was added, and the reaction was further stirred for 12 h at room temperature. Finally, 200 mL deionized water was added slowly to precipitate the resultant, yielding a dark yellow solid (denoted as FA–NH2–BOC) (282 mg, 19.2% yields).

In order to remove the protection of t-butyloxycarbonyl, FA–NH2–BOC (282 mg, 0.4 mmol) was dispersed in a mixture solvent of 6 mL of DMF and 6 mL of TFA and stirred for 0.5 h. After evaporation of the solvent, methanol was added and further co-evaporated three times for removing residual solvent. The mixture was precipitated into the excess ether; after dried under vacuum oven, a dark yellow solid of FA–NH2 was obtained (48.8 mg, 17.3% yields).

Synthesis of POEGMA block polymer

The RAFT agent of CPPA-NHS (38 mg, 0.1 mmol), OEGMA (1.425 g, 2.8 mmol) and AIBN (3.2 mg, 0.02 mmol) were dissolved in 5 mL of anhydrous 1, 4-dioxane. Then, the formative mixture was stirred for 24 h at 70 °C after three degassed cycles. The reaction was terminated using liquid nitrogen and the obtained crude product were dialyzed against methanol to remove the unreacted OEGMA monomer. The liquid in dialysis tube was concentrated by a rotary evaporator and obtained the pure POEGMA block polymer, which would be as a RAFT agent for the subsequent reaction. The degree of polymerization (DP) of OEGMA was calculated as 30 based on the 1H NMR results (307.8 mg, 21.1% yield).

Synthesis of POEGMA-b-PMGMA diblock copolymer

Firstly, the ethyl glycinate methacrylamide (MGMA) monomer was synthesized according to the previous reference with some modification [45]. The POEGMA-b-PMGMA diblock copolymer were synthesized using POEGMA as the macroRAFT agent and MGMA as the monomer, concretely, POEGMA (170 mg, 0.01 mmol), MGMA (81 mg, 0.5 mmol) and AIBN (0.825 mg, 0.005 mmol) were dissolved in a mixture solvent of 1.5 mL DMSO and 1.5 mL dioxane; the formative mixture was degassed three times and stirred for 24 h at 70 °C. The reaction was terminated using liquid nitrogen, and the obtained crude product was dialyzed against methanol to remove the unreacted OEGMA monomer. The liquid in dialysis tube was concentrated by a rotary evaporator and obtained the pure P(OEGMA)30-b-P(MGMA)22 diblock copolymer. Similarly, the P(OEGMA)30-b-P(MGMA)60 copolymer was synthesized with the same method while the molar quantity of MGMA was 1.0 mmol.

Synthesis of FA-POEGMA-b-PMGMA diblock copolymer

Typically, a mixture of P(OEGMA)30-b-P(MGMA)60 (300 mg, 0.01 mmol) and FA–NH2 (6.7 mg, 0.01 mmol) was dissolved in 2 mL anhydrous DMSO, and then, 11.2 uL TEA was added into the above mixture and stirred 48 h at room temperature. At the end of reaction, excess diethyl ether was added and precipitated the resultant. After dried in vacuum, a viscous solid was obtained (101.3 mg, 33.0% yields).

Synthesis of amphiphilic polymeric prodrug

A two-step procedure was used to prepare the DOX polymeric prodrug. 100 mg of POEGMA-b-PMGMA or FA-POEGMA-b-PMGMA was dispersed in a mixture anhydrous solvent of 6 mL of methanol and 3 mL of DMF, and hydrazine hydrate was added and stirred for 12 h at 25 °C. After the reaction, the mixture was dialyzed (MWCO = 3500) against DI water (200 ml × 3) for 24 h. After a freeze–drying process, a pale solid was obtained and further used to graft DOX drug.

Next, 47 mg of POEGMA-b-hydrazide or FA-POEGMA-b-hydrazide and DOX·HCl (2.0 equivalents to the molar amount of the NH2 group in POEGMA-b-hydrazide or FA-POEGMA-b-hydrazide) was added in a mixture solvent of 3 mL DMF and 3 mL methanol. After adding one drop of TFA, the mixture was stirred for 48 h in the dark, and then, the mixture were dialyzed against methanol (200 ml × 3) for 48 h to remove the Free DOX drug. Finally, the dark red solid denoted as POEGMA-b-PDOX (OM@DOX) or FA-POEGMA-b-PDOX (FA-OM@DOX) was obtained after a freeze–drying procedure.

Measurement of the critical micelle concentration (CMC)

The value of CMC is a crucial parameter to verify the stability of micelles. Typically, 2.0 mg of pyrene was dissolved in 10 mL acetone, after volatilizing the acetone completely, 3 mL of OM@DOX-1 and OM@DOX-2 micelles with different concentrations were added, respectively. After stirring in the dark for 12 h, the fluorescence of the solution were detected with a fixed emission wavelength at 390 nm, further plotted the concentration of micelles and fluorescence intensity ratio I338/I333; the CMC value was obtained based on the point of the curve.

Preparation and micellar stability of prodrug micelles

Briefly, OM@DOX-1, OM@DOX-2 or FA-OM@DOX prodrug micelles were prepared using a dialysis method. A mixture of prodrug (5 mg) and TEA (10 uL) was dissolved in 1 mL of DMSO and stirred about 1 h. Then, the prodrug solution was added dropwise into 8 mL deionized water and stirred for 0.5 h, the micellar solution was dialyzed against deionized water (500 ml × 3) for 48 h and the concentration was adjusted to 500 µg/mL for further experiment.

Dynamic light scattering (DLS) instrument was employed to access the micellar stability of OM@DOX-1 and OM@DOX-2 micelles by recording the size distribution of micelles. Specifically, 500 µg/mL of OM@DOX-1 and OM@DOX-2 micelles were placed at refrigerator set at 4 °C, and the average diameter at pre-determined time ranged from 1 day to 5 days, respectively.

Drug release of DOX from prodrug micelles

To estimate the loading content of DOX of OM@DOX-1 and OM@DOX-2 micelles, 1.0 mg of prepared OM@DOX-1 and OM@DOX-2 prodrug was added into 3.0 mL of diluted hydrochloric acid (0.01M) and stirred for 48 h in the dark. The fluorescent intensity of the above solution was measured, and the grafting DOX amount was calculated based on the standard cure of DOX.

The in vitro release of DOX from OM@DOX-1 and OM@DOX-2 micelles was measured using a dialysis method described as follow: Briefly, 1.0 mL of the prepared OM@DOX-1 or OM@DOX-2 micelles in water was transferred into a dialysis tube (MWCO3500) and dialyzed against different pH buffer solution (80 mL) including pH 7.4, pH 6.8 and pH 5.0 at 37 °C. Then, 3.0 mL of buffer solution was taken out and were replaced with 2.0 mL freshly buffer solution at a specific time interval. The corresponding fluorescent intensity at 555 nm was measured using a 488 excitation wavelength. The amounts of DOX in dialysate were determined using fluorescence spectroscopy based on the standard cure of DOX.

In vitro cytotoxicity

Cytotoxic analysis of OM@DOX-1 and OM@DOX-2 micelles was studied by PrestoBlue assay with HeLa cells. Firstly, 1 × 104 HeLa cells were added into the 96-well plates per well and cultured at 37 °C with 5% CO2 for 12 h, then phosphate buffered saline PBS (1×) was used to wash the metabolites of the cells, and 200 uL different concentrations of Free DOX,OM@DOX-1 and OM@DOX-2 micelles ranging from 0 to10 ug/mL were added, respectively. After being cultured for 24, 48 and 72 h at 37 °C with 5% CO2, the cells were washed with PBS (1×) and added 100 uL pre-prepared PrestoBlue solution, which were diluted 10 times with DMEM, after cultivated for 1 h, a Synergy H1 Microplate Reader was used to record the absorbance at 570 and 600 nm, the untreated cells were used as control and the measured data were analyzed. With the same method, the cytotoxicity of OM@DOX micelles to normal cells (L929) was detected, and the cells treated with Free DOX were used as control. To further study the selectively cytotoxicity of FA-OM@DOX micelles, the cell viability of Free DOX, OM@DOX-2 and FA-OM@DOX micelles to KB cells and A549 cells were measured and analyzed with the protocol.

Cellular uptake and intracellular localization

To examine the internalization of DOX in tumor cells, HeLa cells were incubated into 8-well plate with a density of 1 × 104 per well at 37 °C with 5% CO2, after 12 h adhesion of cells on the plates, Free DOX, OM@DOX-1 and OM@DOX-2 micelles (final DOX concentration: 30 μg/mL) were added and incubated with HeLa cells for 0.5, 2 and 6 h, respectively. Then, the drug outside of cells was washed for five times by PBS (1×) and treated with formalin solution for 0.5 h along with thrice washing using PBS (1×). After that, 0.1% (vol/vol) Triton X-100 in PBS (1×) was added to permeabilize the cells for 5 min, and then, the cells were fixed with 1% (wt/vol) BSA in PBS. Then, filamentous actin (F-actin) cytoskeleton was stained with Alexa Fluor® 633 phalloidin for 1 h. After thrice washing with PBS, DAPI was further used to stain the cell nucleus for 1 min. After removing residual DAPI using PBS washing, a few fresh PBS (1×) was left in the plate for fluorescent imaging. Lasers of 633, 488 and 405 nm were employed to excite Alexa Fluor®633 phalloidin, DOX and DAPI, respectively. The corresponding fluorescence emissions were recorded by a confocal laser scanning microscopy (LSM 800, Carl Zeiss, Germany) using a band-pass filter combination including 638–747 nm, 493–634 nm and 410–507 nm for imaging in three individual channels (Objective: EC Plan-Neofluar 10x/0.30 M27; dimension is 1024 × 1024.).

To obtain the FA-medicated cellular uptake, OM@DOX and FA-OM@DOX micelles were detected, and the procedure was similar to the above-narrated without any staining. Concretely, KB cells and A549 cells were incubated into 8-well plate with a density of 1 × 104 per well at 37 °C with 5% CO2; after 12 h adhesion of cells on the plates, OM@DOX-2 and FA-OM@DOX micelles (final DOX concentration: 30 μg/mL) were added and incubated with HeLa cells for 0.5, 1 and 2 h, respectively. Then, the drug outside of cells were washed for five times by PBS (1×) and a few fresh PBS (1×) was left in the plate for fluorescent imaging. Laser of 488 nm was employed to excite DOX, the fluorescence emissions were recorded by CLSM.

To evaluate the intracellular localization of cells in vitro, Lyso Tracker® Red was used to label lysosomes. Briefly, 8-well plates were used to add 1 × 104 HeLa cells per well, after 12 h adhesion of cells on the plates at 37 °C with 5% CO2, Free DOX, OM@DOX-1 and OM@DOX-2 micelles (final DOX concentration: 30 μg/mL) were added and incubated for 0.5, 1 and 2 h, respectively, then the drugs had not entered the cells were washed for five times with PBS(1×); 400uL Lyso Tracker® Red was added and cultured for 0.5~1 h. After that, the residual Lyso Tracker® Red was washed three times with PBS(1×) and replaced with 100 uL DMEM to record the fluorescence imaging with CLSM and lasers of 577 nm were employed to excite Lyso Tracker®Red.

Quantification of cellular uptake by flow cytometry

six-well plates were used to add 1 × 105 HeLa cells per well and cultured for 12 h at 37 °C with 5% CO2, then OM@DOX-1 and OM@DOX-2 micelles (final DOX concentration: 30 μg/mL) were added 2 mL per well and incubated with HeLa cells for 0.5, 2 and 6 h, respectively, then the drugs had not entered the cells were washed and substituted with 500uL PBS (1×), after that the cells were collected and used a flow cytometry to detect the fluorescence intensity of DOX (PE channel was used and the incident was set as 1 × 105 cells with intermediate speed), among which, the cells without any treatment were served as control.

Results and discussion

Synthesis and characterization of pH-responsive DOX prodrug

The synthesis routes of amphiphilic OM@DOX and FA-OM@DOX prodrug were depicted in Fig. 1. Firstly, the block of POEGMA was synthesized by reversible addition–fragmentation chain transfer (RAFT) reaction of OEGMA using the CPPA-NHS as the RAFT agent and AIBN as the initiator at 70 °C. The diblock copolymer of POEGMA-b-PMGMA was further prepared using POEGMA as the macroinitiator. After reacted with FA–NH2, the targeted prodrug of FA-POEGMA-b-PMGMA was harvested. The obtained POEGMA-b-PMGMA or FA-POEGMA-b-PMGMA copolymer was further cultivated with hydrazine hydrate and subsequently conjugated with DOX, and then the OM@DOX and FA-OM@DOX polymeric prodrugs were obtained, respectively.

Figure 1
figure 1

Synthesis routes of amphiphilic POEGMA-b-PDOX (OM@DOX) and FA-POEGMA-b-PDOX (FA-OM@DOX) polymeric prodrugs

Full size image

The representative 1HNMR spectra of polymer including POEGMA, POEGMA-b-PMGMA, FA-POEGMA-b-PMGMA and FA-OM@DOX were monitored and showed in Fig. 2. The marked peak 1 (3.24 ppm) and peak 2 (3.60 ppm) were the typical peaks of methoxy and methylene group of POEGMA block, while the characteristic peak of CPPA-NHS were also present at 3 (7.83~7.18 ppm). The degree of polymerization (DP) of OEGMA was determined on the base of the integral ratio of peak 3 (phenyl group) and 1 (methyl group). After extending the chain using MGMA monomer, the peaks at 4 (3.73 ppm) and 5 (4.55 ppm) were present in the 1HNMR spectra of POEGMA-b-PMGMA as shown in Fig. 2b. The DP of MGMA was determined on the base of the integral ratio of peak 1 (methyl group) and 5 (methylene group). After reacted with FA-NH2 (Figure S1) by an amidation reaction, FA signals (6.63~8.60 ppm) were present in the 1HNMR spectra of FA-POEGMA-b-PMGMA, indicating the successful synthesis. Finally, FA-POEGMA-b-PMGMA were treated with hydrazine hydrate and further conjugated with DOX drug to obtain amphiphilic polymeric prodrug of FA-OM@DOX, and its 1HNMR spectrum also confirmed the existence of DOX drug.

Figure 2
figure 2

1 H NMR spectra of a POEGMA, b POEGMA-b-PMGMA, c FA-POEGMA-b-PMGMA and d FA-OM@DOX polymer

Full size image

The molecular weights (M n ,NMR, M n,GPC, g mol−1) of aforementioned polymers were traced using gel permeation chromatography (GPC) technology. As shown in Fig. 3, after further polymerization of POEGMA, the molecular weights present obviously increase from 7100 to 8000 or 10800 for P(OEGMA)30-b-P(MGMA)22 or P(OEGMA)30-b-P(MGMA)60 diblock copolymer, respectively. Additionally, the PDI of above-mentioned polymers were all below 1.2 and indicated a satisfying controlled polymerization. Table 1 summarized the structural information of two targeted polymers and their precursors. In addition, the FT-IR spectra of POEGMA-b-PMGMA, Free DOX and OM@DOX were detected and showed in Figure S2, the stretch-vibration of C–O (OEGMA) and C=O (MGMA) appeared at 1109 and 1731 cm−1, respectively. After grafting DOX, a strong vibrating peak at 1670 cm−1 corresponding to hydrazide group were present in the FT-IR spectrum of OM@DOX, indicated the successful conjugated of DOX.

Figure 3
figure 3

GPC traces of P(OEGMA)30, P(OEGMA)30-b-P(MGMA)22 (OM-1) and P(OEGMA)30-b-P(MGMA)60 (OM-2) copolymer

Full size image
Table 1 Structural characterization of the polymers
Full size table

Morphology and size of prodrug micelles

The amphiphilic OM@DOX prodrug provided a capacity to form polymeric micelles in water solution. We used dynamic light scattering (DLS) and transmission electron microscopy (TEM) to verify the potential micellization. As shown in Fig. 4, OM@DOX micelles exhibited spherical morphology and had an average diameter of 37.8 ± 3.2 and 24.5 + 2.8 nm, which were revealed by TEM results as shown in Fig. 4b and e. Due to the relatively compact morphology, OM@DOX-2 had a small diameter in comparison with OM@DOX-1, although the former had a high more molecular weight. Furthermore, the DLS results of OM@DOX-1 and OM@DOX-2 micelles were determined as 44.5 and 28.8 nm, which were all larger than that of TEM, this might be due to the hydration of the amphiphilic structure of OM@DOX prodrug. The diameters detected by DLS were larger about 5 nm than TEM observed, this appearance maybe due to the swelling of micelles in water media or contraction of micelles in the absence of water media [46, 47]. Additionally, the DLS result indicated that FA-OM@DOX micelles showed an average diameter of 32.8 nm (Figure S3), indicating a desirable micellar size for biological application. The stability of micelles was a crucial parameter to evaluate its in vivo application. Herein, the dynamic change of diameter of OM@DOX-1 and OM@DOX-2 micelles were monitored(the time interval was 24 h)and showed in Fig. 4(c and f), obviously, no distinct change appeared within 5 days, which demonstrated the excellent stability of OM@DOX-1 micelles and OM@DOX-2 micelles in water media, leading to an enhanced drug efficiency.

Figure 4
figure 4

DLS (a and d), TEM (b and e) and micellar stabilities(c and f) of OM@DOX-1 and OM@DOX-2 micelles in aqueous media, the data are shown as mean ± SD (n = 3)

Full size image

As presented in Fig. 5a, the UV–vis spectra of Free DOX, OM@DOX-1 and OM@DOX-2 micelles in water showed a representative absorption of DOX at 488 nm, while the peak in OM@hydrazide was absent. Additionally, the fluorescence intensity of representative polymers was tested and presented in Fig. 5b, and it was obvious that Free DOX, OM@DOX-1 and OM@DOX-2 micelles in water with a proper emissive fluorescence corresponding to DOX at 557 nm while OM@hydrazide was absent once again. The above results provided sufficient evidence to support the grafting of DOX molecules. After FA molecules grafted on the end of POEGMA-b-PMGMA polymer, the obtained FA-POEGMA-b-PMGMA polymer in DMF shown the feature absorption peak of FA at 358 nm (Fig. 5c), demonstrating the successful modification of FA in FA-POEGMA-b-PMGMA polymer. In Fig. 5d, pyrene was used as fluorescence probe to monitor the critical micelles concentration (CMC) values of OM@DOX-1 and OM@DOX-2, and the CMC values of which were 0.0087 and 0.0102 mg/mL, respectively, indicating that the prodrugs could assemble into micelles at a low concentration (above the corresponding CMC value).

Figure 5
figure 5

aUV–vis spectra and b fluorescence spectra of Free DOX, OM@hydrazide, OM@DOX-1 and OM@DOX-2 micelles in water. The insets of a, b and c were the digital photographs of Free DOX, OM@DOX-1 and OM@DOX-2 micelles in water. c UV–vis spectra of POEGMA-b-PMGMA and FA-POEGMA-b-PMGMA in DMF; d The fluorescence intensity ratio I3/I1 of pyrene as a function of OM@DOX-1 and OM@DOX-2 concentration

Full size image

In vitro release of DOX from prodrug micelles

To obtain the information about the potential of OM@DOX prodrug as the drug carrier,the in vitro release of DOX from OM@DOX-1 and OM@DOX-2 micelles was detected under three simulated buffer solution (pH = 7.4, 6.8 or 5.0) at 37 °C. As shown in Fig. 6, the cumulative release of DOX was only 20.3% from OM@DOX-1micelles while was 11.4% from OM@DOX-2 micelles after 96 h. however, at an acid buffer of pH 5.0,the release rate of DOX within 36 h increased sharply to 63.2 and 71.5% for OM@DOX-1 and OM@DOX-2 micelles, respectively, ultimately, the cumulative release rate were 75.8 and 90.0% within 96 h. In brief, the release results could endow the potential of OM@DOX prodrug as a pH-responsive drug carrier delivering DOX drug into cells with a controlled drug release model.

Figure 6
figure 6

DOX release from OM@DOX-1 and OM@DOX-2 micelles in different PBS solution (pH = 7.4, 6.8 and 5.0). Data are shown as Mean ± SD (n = 3)

Full size image

In vitro cytotoxicity

The cytotoxicity of OM@DOX-1, OM@DOX-2 and FA-OM@DOX micelles were evaluated against HeLa cells (cancer cells) and L929 cells (normal cells) using PrestoBlue assay, and the DOX concentration was ranged from 0 to 10.0 μg/mL. In Fig. 7a–c, after being incubated in HeLa cell for 72 h, the cytotoxicity of OM@DOX-2 micelles with DOX concentration of 10.0 ug/mL presented a higher cytotoxicity than OM@DOX-1 micelles, and the cell viability were 25.6 and 16.2%, respectively, indicating a higher toxicity against tumor cells. Interestingly, the cytotoxicity of OM@DOX-1 and OM@DOX-2 micelles against the normal cells of L929 were 36.4 and 37.1% after 72 h incubation, which is infrequently reported in previously DOX drug delivery systems [16, 30]. The cytotoxicity results further demonstrated that the OM@DOX prodrug shown an obviously high cytotoxicity against tumor cells (HeLa) than normal cells (L929), suggesting a spontaneous tumor-specific targeting ability. In consideration of high drug loading feature and pH-triggered drug release process, the obtained prodrug could simultaneously meet high therapeutic efficiency and low side effect in one system.

Figure 7
figure 7

Cell viability of HeLa cells (a, b and c) and L-929 cells (d) incubated with Free DOX, OM@DOX-1 and OM@DOX-2 micelles with different drug concentrations. Cell viability of (e) KB and (f) A549 cell lines treated with Free DOX (1), OM@DOX-2 (2) and FA-OM@DOX (3) micelles for 72 h. The data were presented as mean ± SD (n = 3)

Full size image

To confirm the FA-receptor specificity of OM@DOX prodrug, the cytotoxicity of FA-OM@DOX, OM@DOX and Free DOX were also calculated against the KB cells (FA-receptor-positive) and A549 cells (FA-receptor-negative). As shown in Fig. 7e and f, FA-OM@DOX shown a preferably cytotoxicity against KB cells in comparison with A549 cells, although the cytotoxicity of two cells was parallel under the high concentration of 10 µg/mL, this might be due to the maximum DOX release amount directed by the acid environment in term of tumor cells. Furthermore, OM@DOX prodrug showed high toxicity against tumor cells with passive-targeting pathway, after armed with FA targeting, the prodrug could further realize the tumor-targeting ability through the specific interaction with FA-receptor-positive tumor cells, which was beneficial to targeted cancer therapy [48, 49].

In vitro cellular uptake

The subcellular internalization of Free DOX, OM@DOX-1, OM@DOX-2 and FA-OM@DOX were investigated with a laser confocal scanning microscopy (LCSM). The green fluorescence of DOX, red fluorescence of Alexa Fluors®633 Phalloidin (F-actin) and blue fluorescence of DAPI were used to dye the F-actin and nuclei of cells. As shown in Fig. 8 and Figure S4, the fluorescence intensity of DOX was distinctly increased from 0.5 to 6 h, which suggested that OM@DOX- micelles with high permeability and further delivered DOX into the cells nuclei. Furthermore, OM@DOX-2 micelles with evidently increased fluorescence than OM@DOX-1 micelles at the same time points (0.5, 2 and 6 h), this might be related to the smaller diameter of OM@DOX-2 than OM@DOX-1 micelles. However, after 2 h incubation, Free DOX strong aggregated in the nuclei and shown brighter fluorescence intensity than that of OM@DOX-1 and OM@DOX-2 micelles.

Figure 8
figure 8

CLSM images of HeLa cells incubated with Free DOX and OM@DOX-2 micelles (DOX equivalent amount of 10 μg/mL) for different time. The fluorescence of DOX, AF-633 and DAPI were labeled with as green, red and blue. Scale bars: 50 μm

Full size image

The subcellular internalization of FA-OM@DOX micelles was also investigated and the fluorescence of DOX showed as red, OM@DOX micelles were used as a control against KB cells (Fig. 9). The fluorescence intensity of DOX delivered by FA-OM@DOX micelles with a sustained increasing from 0.5 to 2 h than that of OM@DOX micelles, which indicated that the FA molecular of FA-OM@DOX micelles could effectively interact with FA-receptors on the membrane of KB cells and then entered into the cells.

Figure 9
figure 9

CLSM images of KB cells treated with OM@DOX and FA-OM@DOX micelles (DOX equivalent amount of 10 μg/mL) at 0.5, 1 and 2 h, respectively. The fluorescence of DOX was labeled with as red. Scale bars: 50 μm

Full size image

To further study the procedure of OM@DOX-1 and OM@DOX-2 micelles entered into cells, we estimated the subcellular co-localization of OM@DOX-1 and OM@DOX-2 micelles with Lyso Tracker@Red. As presented in Fig. 10a and Figure S5, the micelles could cross the cell membrane via endocytosis, and then released the drug triggered by the low pH of lysosome in cells. After incubated for 0.5 to 6 h, the increased fluorescence intensity of DOX derived from OM@DOX micelles delivered suggested a sustained entering process of micelles.

Figure 10
figure 10

a CLSM images of HeLa cells treated with OM@DOX-2 micelles for 0.5, 2 and 6 h, respectively. For each row, images from left to right display the internalized DOX (green), lysosome (Lyso Tracker@Red, red), cell morphology in bright filed and the overlays of three images, respectively. Scale bars: 50 μm. b Quantitative analysis of intracellular uptake of OM@DOX-1 and OM@DOX-2 micelles incubated with control, 0.5, 2 and 6 h in HeLa cells using flow cytometry; c the mean fluorescence intensity (MFI) of OM@DOX-1 and OM@DOX-2 micelles at different time points, and the data were presented as mean ± SD (n = 3)

Full size image

Quantitative cellular uptake of DOX at different time points was monitored with Flow cytometry. In Fig. 10b and Figure S6, HeLa cells treated with OM@DOX-2 micelles showed a considerable increase of fluorescence intensity due to the internalized DOX. Furthermore, the median fluorescence intensity (MFI) of which reached up to 8028 (87.04%) after 0.5 h, which suggested most drugs had been phagocytosed by cells, while OM@DOX-1 micelles with a lower intracellular uptake rate and after 0.5 h with the MFI of 5714 (58.50%). After 6 h incubation, the fluorescence of swallowed OM@DOX-2 micelles (MFI = 15765, 96.29%) was slightly higher than OM@DOX-1 micelles (MFI = 16914, 93.01%), and these results were consistent with the CLSM results and were partially related to its smaller diameter.

Conclusion

In conclusion, we have synthesized one new type of acid-activatable doxorubicin (DOX) polymeric prodrug with ultra-high drug loading for targeted cancer therapy. The amphiphilic FA-OM@DOX or OM@DOX prodrug could self-assemble into prodrug micelles in water solution and expressed high micellar stability even a 5 days storage. The size of prodrug micelles could control by adjusting the length of hydrophilic and hydrophobic chain. Additionally, the designed structure also endowed the prodrug a high drug loading up to 61.5 wt%. The following cytotoxicity results demonstrated that OM@DOX micelles exhibited a higher toxicity against HeLa cells (Cancer cells) than L929 cells (Normal cells), indicating a spontaneous tumor-specific targeting ability. Moreover, the targeted form of FA-OM@DOX micelles present a selective toxicity for FA-receptor-positive KB cells rather than the FA-receptor-negative A549 cells, showing a tumor-targeting ability. In consideration of high drug loading content, pH-triggered drug release behavior and fast cell-penetrating ability. We believe that the present prodrug could offer constructive idea for preparation of stimuli-responsive nanomedicine platform and further extruded its mission in the field of targeted cancer therapy.