A Self-assembled Nanoparticle Platform Based on Amphiphilic Oleanolic Acid Polyprodrug for Cancer Therapy

Oleanolic acid (OA) is a pentacyclic triterpenoid compound with extensive biological effects, such as anti-inflammatory and anticancer activities. However, the application of OA in chemotherapy is hampered by its poor solubility and severe adverse effects. To solve the problems, we developed a self-assembled nanoparticle platform based on amphiphilic oleanolic acid polyprodrug, poly[oligo(ethylene glycol) methyl ether methacrylate]-b-poly[oleanolic acid methacrylate] (POEGMA-b-POAMA), encapsulating 10-hydroxycamptothecin (HCPT) to achieve efficient cancer therapy. The polyprodrug was prepared via reversible addition-fragmentation chain transfer polymerization (RAFT), and could self-assemble to prepare POEGMA-b-POAMA/HCPT nanoparticles (NPs). The obtained nanoparticles exhibited appropriate particle size, excellent drug stability, good drug loading capacity, and high drug loading efficiency. In vitro drug release indicated that the drug release was prolonged to 132 h. The POEGMA-b-POAMA/HCPT NPs enhanced cell cytotoxicity in 4T1 cells and MCF-7 cells and could be efficiently uptaken by 4T1 cells. Furthermore, in vivo antitumor efficiency showed that the POEGMA-b-POAMA/HCPT NPs had great antitumor efficiency with considerably low adverse effects in the treatment of the 4T1 mouse breast tumor xenograft tumor. Therefore, POEGMA-b-POAMA/HCPT NPs provide great potential as a platform for drug delivery applications.


INTRODUCTION
Polymeric drug delivery systems for cancer therapy have received remarkable attention and are under development to overcome the drawbacks of traditional cancer therapy, such as severe side effects, poor water solubility, and low antitumor activities. [1−4] Furthermore, polymeric drug delivery systems have advantages in diverse chemical structures, flexible synthesis methods, and easy functionalization. [5−7] Numerous polymeric drug deliveries such as nanogels, nanoparticles, and nanospheres have played a significant role in medical applications. [8−10] Therefore, polymeric drug delivery systems are considered as a potential strategy in the drug delivery field.
Prodrug-based polymeric nanoparticles, as one type of polymeric drug delivery systems, are advantageous owing to the combined desirable advantages of both prodrugs and nanocarriers. [11,12] Traditional polymeric nanoparticles usually introduce a great number of inert hydrophobic materials which lead to the drawbacks of low drug payloads and undesired side effects. Prodrug-based polymeric nanoparticles could overcome the drawbacks of traditional polymeric nanoparticles and exhibit extra benefits such as improving drug solubility, prolonging drug circulation, reducing the drug resistance, and enhancing drug accumulation in tumor sites because of the enhanced permeability and retention (EPR) effect. [13−16] Typical methods for the synthesis of polymeric prodrug nanoparticles are conjugation of the drug with amphiphilic or hydrophilic polymers and drug-initiated polymerization. [17,18] A few polymers were conjugated with doxorubicin (DOX), betulinic acid (BA), paclitaxel (PTX), or camptothecin (CPT), [19−22] which prolonged the drug release and decreased side effects. As for drug-initiated polymerization, reactive controlled radical polymerization is usually used to produce amphiphilic polymers, such as radical ring-opening polymerization (ROP), reversible addition-fragmentation chain transfer polymerization (RAFT), and atom transfer radical polymerization (ATRP). [23−25] Thanks to these methods, polymeric prodrug nanoparticles were successfully fabricated. However, the potential of polymeric prodrug nanoparticles has not been well developed due to the low drug loading content and suboptimal therapeutic effects.
To overcome these drawbacks, an efficient polymeric drug delivery system is satisfactory with a high drug loading content and low side effects. [26] Polyprodrug nanoparticles, which have been firstly proposed and developed since 2013, [27] are desirable strategies by polymerizing therapeutic drugs with repeating prodrug units and possess tremendous advantages such as high drug dosage, controlled drug loading, selfassembling morphologies, and flexible structure design. [28−32] In addition, combination therapy also plays a significant role to increase the efficiency of the nanoparticles and reduce drug resistance which cancer therapy often suffers from. [33−36] Herein, polyprodrug nanoparticles which could co-deliver diverse therapeutic agents are of great potential to be developed.
Oleanolic acid (OA) is a ubiquitous triterpenoid in many natural plants which possesses a broad series of biological effects including anti-inflammatory, antioxidant, and antitumor activities. [37−39] It has been confirmed that OA exhibits antitumor activity in the treatment of various cancers, such as hepatocellular carcinoma, thymic carcinoma, and breast carcinoma. [40−42] However, the clinical use of OA is limited by the poor aqueous solubility (< 1 μg·mL −1 ) and low bioavailability. [43] At present, there are several methods for enhancing the physicochemical properties of OA. [44−46] Among them, polyprodrug nanoparticles are considered as a highly efficient approach to overcome the drawbacks.
In the present work, a novel amphiphilic polyprodrug, poly[oligo(ethylene glycol) methyl ether methacrylate]-bpoly(oleanolic acid methacrylate) (POEGMA-b-POAMA), was synthesized via RAFT method, consisting of hydrophobic blocks of OA prodrug monomer and hydrophilic poly[oligo (ethylene glycol) methyl ether methacrylate]. Then, 10-hydroxycamptothecin (HCPT), another natural anticancer hydro-phobic drug, was encapsulated into the self-assembled POEGMA-b-POAMA nanoparticles (NPs), which formed the POEGMA-b-POAMA/HCPT NPs. The nanoparticles could be accumulated at the tumor site via EPR effect and endocytosed into cancer cells. Meanwhile, the nanoparticles were disassembled with the release of anticancer drugs owing to the hydrolysis of ester linkage for cancer treatment (Scheme 1). The characteristics of the POEGMA-b-POAMA/HCPT NPs including particle size, zeta-potential, particle stability, and morphology were investigated. The drug release was confirmed by in vitro drug release study. In vitro cell cytotoxicity and cell uptake were studied in 4T1 cells and MCF-7 cells. Moreover, in vivo anticancer efficacy and the adverse effects of the POEGMA-b-POAMA/HCPT NPs were systematically evaluated. The results indicated that the self-assembled nanoparticle platform based on amphiphilic oleanolic acid polyprodrug encapsulating 10-hydroxycamptothecin was a promising drug carrier for enhanced cancer therapy.

Synthesis of MacroRAFT Agent POEGMA
The macroRAFT agent POEGMA was prepared by using DDMAT as the chain transfer agent (Scheme 2b). OEGMA (1.8 g, 6.0 mmol), AIBN (2.63 mg, 0.016 mmol), and DDMAT (47.4 mg, 0.13 mmol) were dissolved in 6 mL of DMF inside a Schlenk flask. After degassed with nitrogen for 20 min, the Schlenk flask was immersed into the oil bath at 70 °C under nitrogen protection. The polymerization was carried out for 9 h and stopped by exposing to the air and cooling the flask with an ice bath. The polymer was purified with excess diethyl ether for twice and dried under vacuum overnight.

Synthesis of POEGMA-b-POAMA Polyprodrugs
The POEGMA-b-POAMA polyprodrugs were synthesized using POEGMA as the chain transfer agent (Scheme 2b). Monomer OAMA (50 mg, 0.06 mmol), DMF (4 mL), AIBN (1.9 mg, 0.012 mmol), and POEGMA (140 mg, 0.01 mmol) were added into a Schlenk flask. After degassed with nitrogen for 20 min, the reaction was allowed to proceed in nitrogen at 90 °C for 11 h. Then, the solution was quenched by cooling the flask with an ice bath and exposing to the air, and the mixture was purified by co-solvent precipitation. The crude products were dissolved in DMF and precipitated into diethyl ether for three times. The resulting polymer was dried under vacuum overnight and obtained as POEGMA-b-POAMA polyprodrugs. Japan) to analyze the absolute molecular weights using DMF as the eluent solvent.

Characterization of POEGMA-b-POAMA NPs and POEGMA-b-POAMA/HCPT NPs
Transmission electron microscopy (TEM) observations were carried out on a JEM-1010 microscope (100 kV) to determine the morphology of the nanoparticles. Dynamic light scattering (DLS) was analyzed on a Zetasizer Nano-ZS instrument (Malvern, UK). The stability of the nanoparticles was evaluated by particle size and PDI measurements in pH 7.2 for 72 h of storage at 4 °C. Drug loading capacity (DC) and drug loading efficiency (DE) of the drug in nanoparticles were calculated using high performance liquid chromatography (HPLC). The mobile solvent was acetonitrile/water, and the volume ratios for OA and HCPT were 95/5 and 70/30, respectively. DC and DE were quantitated as follows: DC (%) = (the weight of free drug in the samples/the weight of total samples) × 100, DE (%) = (the weight of free drug in the nanoparticles/the weight of feeding drug) × 100.

In Vitro Drug Release
The in vitro drug release of POEGMA-b-POAMA/HCPT NPs was examined by a dialysis method. [48] Briefly, POEGMA-b-POAMA/ HCPT NPs (10 mg) were dissolved in PBS (10 mL) and then removed to the dialysis bag. The dialysis bag was placed into a beaker filled with 100 mL of PBS (pH 5.0, 6.0 or 7.2, respectively) under gentle shaking (100 r) at 37 °C. Subsequently, the external medium (200 μL) was collected and replaced with PBS at the specific time. The samples were analyzed to determine the release of OA and HCPT via the HPLC method. The mobile solvent was acetonitrile/water at a flow rate of 1 mL·min −1 with the C18 column, and the volume ratios for OA and HCPT were 95/5 and 70/30 respectively. The UV detectors were 210 nm for OA and 254 nm for HCPT.

Cellular Uptake Study
Cellular uptake study of POEGMA-b-POAMA/HCPT NPs was assessed using 4T1 cells. In the flow cytometric (FCM) study, 4T1 cells (2 × 10 5 cells, 1.5 mL media per well) were separately plated in 6-well plates and incubated overnight. Then, the cells were treated with 1 mL of fresh culture media containing POEGMA-b-POAMA/HCPT NPs and free HCPT with equivalent concentration of HCPT (2 μg·mL −1 ). After a certain time, the cells were rinsed with PBS for three times to remove the supernatant and trypsinized from the plates. Finally, the cells were measured by a flow cytometer (Becton Dickinson, USA).

In Vivo Antitumor Efficiency
The

Evaluation of Adverse Effects
The mice bearing 4T1 tumors were injected with PBS, free HCPT, POEGMA-b-POAMA NPs or POEGMA-b-POAMA NPs at the equivalent HCPT dose of 10 mg·mL −1 via tail-vein injection. The mice blood was separated and centrifuged at 10 day. The samples were analyzed using the mouse IgE ELISA procedure to evaluate the type I hypersensitivity reactions, which is one of the most common hypersensitivity reactions.

Statistical Analysis
All the experiments were performed at least three times. The results are presented as mean ± standard deviation (SD).

Synthesis and Characterization of POEGMA-b-POAMA Polyprodrugs
We have designed and synthesized a novel drug carrier based on an amphiphilic POEGMA-b-POAMA polyprodrug for cancer therapy with good drug loading efficiency and biocompatibility. The amphiphilic POEGMA-b-POAMA polyprodrug prepared by a multiple synthetic approach consisted of the hydrophilic POEGMA and hydrophobic OAMA (Scheme 1). The structures and compositions of the monomers and the intermediates were calculated by 1 H-NMR and GPC. The characteristic resonance signals of OA were observed at 0.76−2.01 ppm (Fig. 1a). After the reaction between OA and MAC, new signals appeared at 6.74 ppm (b) and 6.89 ppm (c) and the chemical shift δ 3.22 (1H, t) of OA moved to 3.27 (1H, t), indicating the successful formation of OAMA monomers (Fig. 1A). As shown in Fig. 1(B), the signals at 3.46−3.93 ppm are ascribed to the PEG ethylene protons. The GPC analysis indicated that the number-average molecular weight (M n ) and polydispersity index (PDI) of POEGMA were 1.32 × 10 4 g·mol −1 and 1.12, respectively (Table 1). Fig. 1(B) shows that POEGMA-b-POAMA polyprodrug displayed the characteristic signals of both hydrophilic POEGMA and hydrophobic OAMA segments, suggesting that the copolymers were synthesized successfully. The GPC results demonstrated that the POEGMA-b-POAMA copolymer had a M n of 1.52 × 10 4 g·mol −1 and a PDI of 1.19. The degree of polymerization (m) of the POEGMA block was calculated by comparing the integral areas of the characteristic signal of POEGMA at 3.46 ppm (d) to that at 1.24 ppm (g). The repeating segments of POEGMA and OAMA were calculated by comparing the integral areas of the characteristic signal of POEGMA-b-POAMA at 3.46 ppm (d) to that at 5.3 ppm (e) (Fig.  1B). The POEGMA-b-POAMA copolymer contained 6 units of OAMA and 45 units of OEGMA. The POEGMA-b-POAMA copolymer in the study was POEGMA 45 -b-POAMA 6 . The GPC results agreed well with the M n calculated by 1 H-NMR and the feeding ratio, showing a well-controlled polymerization reaction.

Synthesis and Characterization of the Nanoparticles
POEGMA-b-POAMA NPs and POEGMA-b-POAMA/HCPT NPs were synthesized via the nanoprecipitation method. The POEGMA-b-POAMA copolymer was an amphiphilic polymer which could self-assemble into nanoparticles and encapsulate HCPT in aqueous solution with hydrophilic POEGMA corona and hydrophobic drug core. [49,50] The morphology of the POEGMA-b-POAMA NPs and the POEGMA-b-POAMA/HCPT NPs was investigated by TEM. Particle size, zeta potential, and size distribution of the nanoparticles were measured by DLS. As shown in Table 2, the average particle size of the POEGMA-b-POAMA NPs was 164.23 ± 4.21 nm with narrow PDI of 0.18, and the zeta potential of 9.03 ± 0.24 mV. The OA content in POEGMA-b-POAMA NPs was calculated as 17.35% ± 0.21%. The POEGMA-b-POAMA/HCPT NPs displayed average particle sizes of 267.84 ± 5.82, 190.12 ± 6.15, and 283.68 ± 7.64 nm, and the zeta potentials of 7.96 ± 0.53, 8.05 ± 0.36, and 8.19 ± 0.36 mV at the carrier/drug mass ratios of 20/5, 20/10, and 20/15, respectively. The POEGMA-b-POAMA/HCPT NPs displayed a prominent encapsulation performance with the highest DC HCPT of 32.64% ± 0.82% and DE HCPT of 92.34% ± 1.12%. The carrier/drug mass ratio reached as low as 20/15. DC HCPT in the POEGMA-b-POAMA NPs was improved with the decrease of carrier/drug weight ratio, which might be due to enhanced interaction between the carrier and drug.
In order to deliver the nanoparticles with an appropriate size below 200 nm, which is essential for EPR effect, the POEGMA-b-POAMA/HCPT NPs at the carrier/drug mass ratio of 20/10 were chosen for the subsequent study. Both of the nanoparticles exhibited spherical shapes (Figs. 2a and 2b) and equally distributed (Figs. 2c and 2d). To determine the colloidal stability of the nanoparticles, the particle size and the PDI were detected in pH 7.2 for 72 h at 4 °C. As shown in Figs. 2(e) and 2(f), there was no distinct change in particle size and PDI, indicating that the nanoparticles exhibited high redispersion stability. Therefore, these results demonstrated that the POEGMA-b-POAMA/HCPT NPs could be effective for the delivery of the two antitumor drugs.

In Vitro Drug Release
The pH (< 6.8) value in the tumor site is lower than that in normal cells (~ 7.4). The pH value is more acidic in the lysosomal  Therefore, in vitro release performances of OA and HCPT from the nanoparticles were investigated in PBS at pH 5.0, 6.0, and 7.2 (Fig. 3). There were two stages that could be distinguished for all drug release profiles. The POEGMA-b-POAMA NPs and POEGMA-b-POAMA/HCPT NPs showed an initial burst release due to the concentration gradient between the medium and the nanoparticles. After 12 h, the cumulative release of OA from   POEGMA-b-POAMA NPs was 20.3% (pH 5.0), 17.6% (pH 6.0), and 12.3% (pH 7.2) and that from the POEGMA-b-POAMA/HCPT NPs was 22.7% (pH 5.0), 18.4% (pH 6.0), and 13.9% (pH 7.2) (Figs. 3a and 3c), while the cumulative release of HCPT from the POEGMA-b-POAMA/HCPT NPs was about 36.2% (pH 5.0), 28.8% (pH 6.0), and 18.5% (pH 7.2) (Fig. 3b). Then, the nanoparticles degraded and released slowly at a low level, which indicated that the POEGMA-b-POAMA NPs and the POEGMA-b-POAMA/HCPT NPs significantly prolonged the release of OA and HCPT. After 132 h, the cumulative release of OA from the POEGMA-b-POAMA NPs reached up to 63.9% (pH 5.0), 54.5% (pH 6.0), and 49.3% (pH 7.2) and that from the POEGMA-b-POAMA/HCPT NPs reached 78.0% (pH 5.0), 61.7% (pH 6.0), and 44.7% (pH 7.2) (Figs. 3a and 3c). Meanwhile, the drug OA from the POEGMA-b-POAMA NPs and POEGMA-b-POAMA/HCPT NPs was slowly hydrolyzed and released at weakly acidic or neutral pH without the burst release phenomenon. The OA release rate in acidic solution (pH 5.0) was higher than that in slightly acidic solution (pH 6.0) or neutral solution (pH 7.2) (Figs. 3a and 3c), which was due to the re-protonation of the hydroxyl group of OA. [51,52] The HCPT encapsulated by POEGMA-b-POAMA/HCPT NPs displayed a similar release profile at the acid condition (Fig. 3b). This was attributed to the disassembly of nanoparticles in the acid environment. These results indicated that the prepared POEGMA-b-POAMA NPs and the POEGMA-b-POAMA/ HCPT NPs significantly prolonged the drug release under the weakly acid condition of the tumor microenvironment.

In Vitro Cell Cytotoxicity
In vitro cell cytotoxicity of the POEGMA-b-POAMA NPs and the POEGMA-b-POAMA/HCPT NPs was measured on the 4T1 cell lines and MCF-7 cell lines using free OA and free HCPT as the control. Figs. 4(a) and 4(b) show that the cell viability was significantly reduced with the increasing sample concentration.  In comparison to free OA, the POEGMA-b-POAMA NPs showed low cytotoxicity against 4T1 cells and MCF-7 cells, which might be attributed to the limited OA cleavage from the POEGMA-b-POAMA polyprodrug. The IC 50 values of different samples toward 4T1 cells and MCF-7 cells are displayed in Table 3. The POEGMA-b-POAMA/HCPT NPs exhibited the highest cytotoxicity against 4T1 cells and MCF-7 cells compared with free HCPT, free OA, and the POEGMA-b-POAMA NPs, which might be due to the enhanced cytotoxicity depending on the release of OA, and co-delivered HCPT. Furthermore, the CI values were 0.74 for 4T1 cells and 0.69 for MCF-7 cells, both of which were < 1, suggesting the synergistic effect between OA and HCPT in 4T1 cells and MCF-7 cells. [53,54] In summary, the POEGMA-b-POAMA/HCPT NPs displayed high therapeutic efficacy to 4T1 cells and MCF-7 cells.

In Vitro Cellular Uptake Study
The amount of cellular internalization was investigated by FCM (Figs. 5a and 5b). After incubation for 4 h, the HCPT fluorescence intensity became stronger, which might be due to the increasing nanoparticle concentration. The mean fluorescence intensity (MFI) of the POEGMA-b-POAMA/HCPT NPs was also increased in 4T1 cells with prolonged incubation time, which indicated that the POEGMA-b-POAMA/HCPT NPs could be internalized into the 4T1 cells successfully. Therefore, it could be concluded that the POEGMA-b-POAMA/HCPT NPs were good drug carriers across the cell membranes of the 4T1 cells.

In Vivo Antitumor Efficiency
The in vivo antitumor efficiencies of the POEGMA-b-POAMA/ HCPT NPs were evaluated using the 4T1 cell tumor model. As shown in Fig. 6(a), the curves of relative tumor volume in the POEGMA-b-POAMA/HCPT NPs, the POEGMA-b-POAMA NPs, and free HCPT were increased more slowly than the saline control group. The mean tumor volume of the POEGMA-b-POAMA/HCPT NPs group was 1004.6 ± 527.4 mm 3 , which was smaller than that of POEGMA-b-POAMA NPs (2458.4 ± 1124.3 mm 3 ) and free HCPT (1852.1 ± 1022.4 mm 3 ) at the end of the study ( Table 4). As shown in Fig. 6(b), the body weights in POEGMA-b-POAMA/HCPT NPs and POEGMA-b-POAMA NPs displayed no significant change, indicating that the nanoparticles were safe without systemic toxicity. Table 4 shows that the average tumor weights in POEGMA-b-POAMA/HCPT NPs, free HCPT, POEGMA-b-POAMA NPs, and the control group were 0.63 ± 0.08, 1.18 ± 0.12, 1.89 ± 0.09, and 2.23 ± 0.10 g on day 12. Tumors of the POEGMA-b-POAMA/HCPT NPs group were the smallest (Figs. 6c and 6d), which was in agreement with the results of the tumor volume, suggesting that POEGMA-b-POAMA/HCPT NPs showed the highest antitumor efficiency among all the treatment groups. Overall, the POEGMA-b-POAMA/HCPT NPs with excellent antitumor efficiency were promising for future clinical treatment.

Evaluation of Adverse Effects
To further investigate the adverse effects, the lgE levels and the WBC count of the mice blood were evaluated on the 10 th day. Fig. 7(a) presents that compared to the control group, the lgE levels of the POEGMA-b-POAMA/HCPT NPs and the POEGMA-b-POAMA NPs displayed no significant difference, while the lgE levels of free HCPT had a great increase, which might be due to the low blood solubility. The results demonstrated that the nanoparticles displayed no obvious adverse effects on mice. To further assess the safety of the nanoparticles, the WBC counts of the mice blood were also examined after treatment with the samples (Fig. 7b). No significant change was observed in the POEGMA-b-POAMA/HCPT NPs and the POEGMA-b-POAMA NPs group in comparison to the control group, while the apparent decrease occurred in free HCPT group. In conclusion, the nanoparticles could avoid adverse effects to confirm prominent for drug delivery.