Sigma receptor-mediated targeted delivery of anti-angiogenic multifunctional nanodrugs for combination tumor therapy

Journal of Controlled Release 228 (2016) 107–119

Contents lists available at ScienceDirect

Journal of Controlled Release journal homepage: www.elsevier.com/locate/jconrel

Sigma receptor-mediated targeted delivery of anti-angiogenic multifunctional nanodrugs for combination tumor therapy Yuanke Li a, Yuanyuan Wu a, Leaf Huang b, Lei Miao b, Jianping Zhou a, Andrew Benson Satterlee b,c, Jing Yao a,⁎ a b c

State Key Laboratory of Natural Medicines, Department of Pharmaceutics, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, China Division of Molecular Pharmaceutics, Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill 27599, USA University of North Carolina and North Carolina State University Joint Department of Biomedical Engineering, Chapel Hill, NC 27599, USA

a r t i c l e

i n f o

Article history: Received 10 October 2015 Received in revised form 31 January 2016 Accepted 27 February 2016 Available online 3 March 2016 Keywords: Low molecular weight heparin Ursolic acid Nanodrugs Anti-angiogenesis Combination therapy Sigma receptor

a b s t r a c t The potential of low molecular weight heparin (LMWH) in anti-angiogenic therapy has been tempered by poor in vivo delivery to the tumor cell and potentially harmful side effects, such as the risk of bleeding due to heparin's anticoagulant activity. In order to overcome these limitations and further improve the therapeutic effect of LMWH, we designed a novel combination nanosystem of LMWH and ursolic acid (UA), which is also an angiogenesis inhibitor for tumor therapy. In this system, an amphiphilic LMWH-UA (LHU) conjugate was synthesized and self-assembled into core/shell nanodrugs with combined anti-angiogenic activity and significantly reduced anticoagulant activity. Furthermore, DSPE-PEG-AA-modified LHU nanodrugs (A-LHU) were developed to facilitate the delivery of nanodrugs to the tumor. The anti-angiogenic activity of A-LHU was investigated both in vitro and in vivo. It was found that A-LHU significantly inhibited the tubular formation of human umbilical vein endothelial cells (HUVECs) (p b 0.01) and the angiogenesis induced by basic fibroblast growth factor (bFGF) in a Matrigel plug assay (p b 0.001). More importantly, A-LHU displayed significant inhibition on the tumor growth in B16F10-bearing mice in vivo. The level of CD31 and p-VEGFR-2 expression has demonstrated that the excellent efficacy of antitumor was associated with a decrease in angiogenesis. In conclusion, A-LHU nanodrugs are a promising multifunctional antitumor drug delivery system. © 2016 Elsevier B.V. All rights reserved.

1. Introduction It is widely accepted that tumor growth and metastasis depend on angiogenesis, as these neovasculatures must provide the nutrition necessary for the dividing tumor cells [1,2]. Angiogenesis plays a fundamental role in physiological and pathological conditions such as cancer and chronic inflammation, which is regulated by a number of growth factors such as vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), and platelet derived growth factor (PDGF). These growth factors bind to heparan sulfate proteoglycans (HSPGs) present endothelial cells (ECs), as well as in the extracellular matrix (ECM), and thereby trigger the proliferation and migration of ECs [3–7]. In many human tumors, the molecule heparanase is more highly expressed, and can partially digest these HSPGs, producing fragments which seem even more effective than the native heparin sulfate in potentiating the activity of bound growth factors. As compared to traditional cancer therapy, anti-angiogenic therapy can inhibit tumor growth by targeting blood vessels more effectively and stably by overcoming impediments such as drug resistance and inadequate drug delivery

⁎ Corresponding author. E-mail address: [email protected] (J. Yao).

http://dx.doi.org/10.1016/j.jconrel.2016.02.044 0168-3659/© 2016 Elsevier B.V. All rights reserved.

[8]. Therefore, antiangiogenesis has represented a potential target for cancer therapy. Low molecular weight heparin (LMWH) is a non-cytotoxic, biodegradable, and water-soluble natural glycosaminoglycan. It has been reported that LMWH can reduce angiogenesis induced by bFGF and VEGF when administered systemically by competitively inhibiting the binding of growth factors to their endothelial receptors [3,9–11]. Collen et al. demonstrated that LMWH inhibited both bFGF and VEGF-induced proliferation of human microvascular endothelial cells (HUVECs) [12]. Moreover, heparin and some chemically modified heparins can inhibit tumor cell heparanase activity, which correlates with a lower metastatic potential. The effects of heparins on the outgrowth of primary tumors, angiogenesis, and metastasis have been studied in several animal models. Nathan et al. have found that heparin significantly inhibited the growth of transplanted rodent Murphy–Sturm lymphosarcoma by the 15th day of therapy [13]. However, a long-term administration of a high concentration of heparin was necessary to produce an effect on the primary tumor growth, which would unfortunately put patients at risk of hemorrhage [14]. Various chemically modified heparins have therefore been synthesized to minimize its anticoagulant activity, such as periodate-oxidized, N-acetylated, N-desulfated, O-desulfated or carboxyl-reduced heparin [15]. It has been described that periodatetreated, non-anticoagulant heparin carrying polystyrene exhibited

108

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

stronger anti-angiogenic properties and much lower anti-coagulant activity than heparin itself [16]. In addition, LMWH chemically modified with hydrophobic segments also exhibited this reduced anticoagulant activity and enhanced tumor growth inhibition via anti-angiogenesis [15,17,18]. Park et al. have also developed the chemical conjugation LMWH-deoxycholic acid (LHD), which exhibits no anticoagulant activity and high inhibition of tumor growth via anti-angiogenic effect [15]. Ursolic acid (UA) is a pentacyclic triterpenoid derived from berries, leaves, flowers, and fruits of medicinal plants [19]. It has been described that hydrophobic UA possesses pleiotropic biological effects such as antibacterial, hepatoprotective, antitumor, anti-inflammatory and antiangiogenic activities [20–23]. The antitumor effect of UA is related to its ability to influence the activity of several enzymes, which modulate the process of tumor growth. For example, the MAPK/ERK and PI3K/ AKT/mTOR signaling cascades play critical roles in the transmission of signal from growth factor receptors to regulate gene expression, which are responsible for anti-apoptotic and drug resistance effect in cells [24]. UA has the ability to suppress communication through these routes to inhibit the tumor growth. In addition, forkhead box (FOX) M1 is a kind of transcription factors that is able to cross-talk with other molecules in cancer development such as NF-κB, COX-2, ERK and MMPs [25,26]. The study has reported that UA inhibited the Forkhead box M1 expression on MCF-7 human breast cancer cells [26]. UA has demonstrated the capability to inhibit key steps of angiogenesis in vitro, including endothelial cell proliferation, migration, and differentiation [27]. It was found that UA inhibited angiogenesis in a dose-dependent manner, with the dose required for half-maximal inhibition in a chick embryo chorioallantoic membrane being a low dose of 5 μg [23]. The potential of UA to inhibit angiogenesis in vivo was also investigated. Kanjoormana et al. found that UA inhibited tumorassociated capillary formation in B16F10 melanoma grown in C57BL/6 mice. Meanwhile, as compared to the control, the VEGF, NO, and proinflammatory cytokines were highly reduced and serum TIMP-I (tissue inhibitor of metalloproteinase-I) and IL-2 (interleunkin-2) levels were significantly increased in UA-treated mice [19]. The anti-angiogenic abilities of UA are usually attributed to the inhibition of the downregulation of matrix metalloproteinase (MMP) activity, which are group of the enzymes involved in degradation of extracellular matrix. The studies reported that UA inhibited the activity of MMP-9 and MMP-2 [28,29]. Taking the anti-angiogenic advantages of both LMWH and UA, we have successfully prepared a LMWH-UA (LHU) conjugate as a polymeric drug by covalently binding UA to LMWH via an amide linker for combination cancer therapy. In this way, an additive inhibition of tumor angiogenesis could be achieved over the single anti-angiogenic effect of LMWH or UA alone. More importantly, compared to common chemical drugs, the amphiphilic LHU conjugate is able to form nano-sized particles in aqueous condition, thereby possessing the excellent properties of polymeric nanoparticles, including good stability, improved pharmacokinetic and distribution characteristics (e.g. EPR effect-based passive tumor targeting), and reduced side effects [30,31]. Moreover, the water solubility of UA can also be increased by binding to the hydrophilic LMWH, which facilitates the intravenous administration of insoluble UA. Sigma receptors are well known membrane-bound proteins which are over-expressed on many types of cancer cells, such as melanoma, non-small cell lung carcinoma, breast tumors of neural origin, and prostate tumors [32–37]. The PEG-lipid containing anisamide (DSPE-PEGAA) exhibits a high affinity to sigma receptors over-expressed on the tumor cells [37]. Therefore, the DSPE-PEG-AA was used to modify LHU nanodrugs in this study to facilitate the targeted delivery of nanodrugs to the tumor. Based on sigma receptor-mediated endocytosis and the tumor's EPR effect, DSPE-PEG-AA-modified LHU (A-LHU) nanodrugs should show increased accumulation at the tumor site. PEG, a biocompatible hydrophilic polymer, is known to contribute to the long circulation time of nanoparticles, and it was added to A-LHU nanodrugs to

benefit its targeted delivery in vivo [38]. A-LHU nanodrugs are therefore multifunctional ternary antitumor drug delivery systems used for synergistic angiogenetic inhibition via LMWH and UA, as shown in Fig. 1. In this study, the capacity of A-LHU nanodrugs to inhibit bFGFinduced angiogenesis was evaluated in vitro and in vivo. In vitro cellular uptake of A-LHU nanodrugs was monitored on two different cell lines: B16F10 cells (high level of sigma receptor) and HUVECs (low level of sigma receptor). Antitumor efficacy of A-LHU nanodrugs was also investigated. 2. Materials and methods 2.1. Materials LMWH (100 IU/mg), average molecular weight near 4500 Da, was obtained from Nanjing University. UA was purchased from Wuhan Yuan Cheng Co-created Technology Co. Ltd. (Wuhan, China). DSPE– PEG–anisamide (DSPE–PEG–AA) was synthesized in our lab as described [39].DSPE-PEG-OCH3 was purchased from Xiamen Sinopeg Biotech Co. Ltd. (Xiamen, China). 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) and N-Hydroxysuccinimide (NHS) were purchased from Aladdin Industrial Corporation (Shanghai, China). N, Ndicyclohexylcarbodiimide (DCC) and pyrene were from Sinopharm Chemical Reagent Co. Ltd. (Nanjing, China). Anhydrous dimethylformamide (DMF) and anhydrous formamide were from Shanghai Lingfeng Chemical Reagent Co. Ltd. (Shanghai, China). Acetone was purchased from Nanjing Chemical Reagent Co. Ltd. (Nanjing, China). All other chemicals were of analytical grade and were used without further purification. 2.2. Synthesis and characterization of LHU conjugate The LHU conjugate was obtained by coupling aminated LMWH with UA, as shown in Fig. 2A. Firstly, LMWH dissolved in phosphate-buffered solution (PBS) (0.01 M, pH 7.4) was reacted with EDC and NHS (molar ratio 1:3:3) for 4 h at room temperature, followed by adding ethylenediamine as the linker. The reaction of activated LMWH and ethylenediamine was for 24 h at room temperature. After reaction with ethylenediamine, the mixture was dialyzed against deionized water for 48 h using a dialysis membrane (MWCO 3500). The pure LMWH-NH2 was obtained followed by lyophilization. The degree of substitution (DS) of ethylene diamine on LMWH was 17.4% according to the 1H NMR assay. Secondly, UA dissolved in THF was reacted with NHS and DCC (molar ratio 1:1.5:1.2) for 24 h at room temperature. The precipitated side product dicyclohexylurea (DCU) was removed by filtration. The activated UA was precipitated in n-hexane and then filtered, followed by vacuum drying at room temperature. Finally, the activated UA dissolved in DMF was added into the mixture of LMWHNH2 dissolved in formamide and EDC (molar ratio 4 ∶ 1 ∶ 4), and then reacted for 24 h at room temperature. The mixture was then precipitated in excess cold acetone and the precipitate was carefully washed three times with cold acetone. The dried LHU conjugate was suspended in water and dialyzed against deionized water for 24 h using a dialysis membrane (MWCO 3500), followed by lyophilization. The chemical structure of LHU conjugate was characterized by 1H NMR spectra. The hydroxyl of UA can be easily dehydrated with the presence of sulfuric acid resulting in the formation of the chromophores in the visible range. Therefore, the DS of UA covalently attached to LMWH was determined spectrophotometrically after a reaction with sulfuric acid [15,40]. The particle size and zeta potential of LHU nanodrugs were determined by dynamic light scattering (DLS) measurements (BI-200SM, Brookhaven Instruments Corp., USA). The morphology of LHU nanodrugs was observed by transmission electron microscopy (H-600, Hitachi, Japan). The critical micelle concentration (CMC) of the LHU conjugate was investigated by fluorescence spectroscopy, using pyrene as a probe as described previously [41]. Briefly, 1 mL

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

109

Fig. 1. Schematic design of A-LHU nanodrug formation, simultaneous drug delivery, and accumulation at tumor site.

of 6.0 × 10− 6 M pyrene solution in acetone was added to a series of 10 mL volumetric flasks and then acetone was evaporated. 10 mL of different concentrations of LHU conjugate solutions (0.4-2000 μg/mL) was added to the volumetric flasks followed by sonicating for 30 min. The samples were incubated at 65 °C for 1 h, and then left to cool down overnight at room temperature. Fluorescence spectra were recorded with a RF-5301 PC florescence spectrophotometer (Shimadzu, Japan) with the emission wavelength at 390 nm. Both of the excitation and emission slit-widths were 3 nm. The CMC was estimated as the cross-point when extrapolating the intensity ratio I338/I333 at low and high concentration regions.

2.3. Preparation and characterization of A-LHU nanodrugs The A-LHU nanodrugs with different mass ratios of DSPE-PEG-AA and LHU (1:19, 1:10, 1:9, 1:7 and 1:4) were prepared as follows: 1.05, 2, 2.2, 2.9, or 5 mL DSPE-PEG-AA solution(1 mg/mL) was added to 20 mg LHU conjugate dissolved in 2 mL distilled water. The resulting mixture was incubated at room temperature for 60 min. The nontargeted DSPE-PEG-OCH3/LHU nanodrugs were prepared using this same method. The particle size and zeta potential of A-LHU nanodrugs were determined by DLS measurements. The morphology of A-LHU nanodrugs was observed by atomic force microscopy (Nano Scope IIIa, Veeco, USA) and transmission electron microscopy.

2.4. In vitro endothelial tubular formation assay Matrigel (50 μL, growth factor reduced and phenol red-free, BD Bioscience) was loaded into each well of 96-well plate and allowed to polymerize for 1 h at 37 °C. Human umbilical vein endothelial cells (HUVECs, 3 × 105 cells/mL) were added onto Matrigel and cultured in 5% Fetal Bovine Serum-Endothelial Cell Medium containing 10% bFGF only or free LMWH, free UA, LMWH plus UA, LHU and A-LHU nanodrugs (at equivalent LMWH concentration of 50 μg/mL, UA concentration of 2.6 μg/mL) in 5% CO2 at 37 °C. After 6 h incubation, cell growth and organization were observed through an OLYMPUS inverted microscope, and tubular structures were quantified by counting and averaging tubular length from 5 high-power fields (×100).

2.5. In vivo matrigel plug assay The Matrigel plug assays were performed to assess the antiangiogenic properties of LHU nanodrugs. Six hundred and thirty seven μL of a Matrigel-PBS mixture was injected subcutaneously into the flanks of 5 to 6 week-old male mice as a negative control. Matrigel (500 μL) with 12.5 μL bFGF (500 ng/mL) to promote angiogenesis was used as a positive control. To test anti-angiogenic treatment efficacy, Matrigel and bFGF were mixed with 125 μL free LMWH, free UA, LMWH plus UA, LHU, or A-LHU nanodrugs (at equivalent LMWH concentration of 500 μg/mL, UA concentration of 22.5 μg/mL) and injected subcutaneously into the mice. After 10 days, mice were sacrificed, and Matrigel plugs were removed and photographed. To measure hemoglobin contents in the new blood vessels in the Matrigel plugs, the removed Matrigel plugs were homogenized in hypotonic lysis buffer (2 mL of 0.1% Brij-35 per plug) and centrifuged for 10 min at 10,000 rpm. A constant volume of supernatant was incubated in 0.5 mL of Drabkin's solution (Sigma, St. Louis, MO.) for 15 min at room temperature and the absorbance was measured at 540 nm with Drabkin's solution as a blank. The relative hemoglobin content was calculated versus the negative (0%) and positive (100%) control.

2.6. In vitro cellular uptake of A-LHU nanodrugs Before determining the cellular uptake of A-LHU nanodrugs, we measured the expression level of sigma receptor in both B16F10 cells and HUVECs. The non-small cell lung cancer (NSCLC), H460, known to have abundant sigma receptor expression, was used as a positive control. In brief, cells were lysed with a Radio-Immunoprecipitation Assay (RIPA) buffer supplemented with 1% protease inhibitor cocktail. Protein concentration of the cell lysates was measured using bicinchoninic acid (BCA) protein assay reagent following manufacturer's instruction (Invitrogen). Same amount of protein from each cell lysates was separated by 4–12% SDS-PAGE electrophoresis (Invitrogen) before being transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad). 5% skim milk was used to block these membranes for 1 h at 37 °C followed by addition of anti-sigma receptor primary antibody (Santa Cruz biotechnology, Inc.). The membranes were then incubated with the primary antibody at 4 °C overnight. After extensive washing, the membranes were then probed with the horseradish peroxidase-conjugated

110

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

Fig. 2. Preparation and characterization of LHU conjugate: (A) The synthetic scheme of LHU conjugate; (B) 1H NMR spectra of (a) LMWH and (b) LHU conjugate in D2O, respectively.

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

secondary antibodies and detected using the Pierce ECL western blotting substrate (Thermo Fisher Scientific, Rockford, IL). GAPDH antibody was probed as the loading control. Confocal laser scanning microscopy was used to visualize the cellular uptake of LHU nanodrugs in vitro. Coumarin-6 was incorporated into LHU as a fluorescence marker. Coumarin-6-loaded LHU was prepared by a dialysis method. Eighteen mg of LHU conjugate was dissolved in 3 mL of water. Coumarin-6 at the concentration of 1 mg/mL in ethanol was then added into the LHU solution, followed by ultra-sonication for 30 min in an ice bath by a probe-type ultra-sonicator (JY92-2D, Ningbo Scentz Biotechnology Co., Ltd., China). The solution was dialyzed against excess distilled water with a dialysis membrane (MWCO 3500) overnight, followed by filtration through a 0.8 μm pore-sized microporous membrane and lyophilization. B16F10 cells and HUVECs were seeded at a density of 106 cells/ well and incubated, respectively. Then cells were incubated with free coumarin-6, coumarin-6-loaded LHU nanodrugs, coumarin-6-loaded A-LHU nanodrugs containing 10 wt% DSPE-PEG-AA (1:9), or coumarin-6-loaded A-LHU nanodrugs containing 5 wt% DSPE-PEG-AA (1:19) diluted in the medium. After 2 h or 6 h incubation, the cells were washed three times with PBS and fixed with 4% paraformaldehyde for 30 min, followed by adding Hoechst 33258 to stain the cell nuclei. Finally, the cells were observed using the Olympus Flowview FV 1000. Flow cytometry was also used to quantitatively determine the cellular uptake of the nanodrugs. B16F10 cells were seeded at a density of 5 × 105 cells/well and incubated. The cells were then incubated with free coumarin-6, coumarin-6-loaded LHU nanoparticles, coumarin-6loaded A-LHU nanodrugs (1:9) orcoumarin-6-loaded A-LHU nanodrugs (1:19) diluted with the medium. After 6 h incubation, the cellular uptake of the nanodrugs was quantified for coumarin-6-positive cells by flow cytometry (BD, Biosciences, USA). 2.7. In vitro cytotoxicity studies The cytotoxicity of LHU and A-LHU nanodrugs was performed using the MTT assay. The B16F10 cells were seeded at a density of 5 × 103 cells/well in a 96 well plate and incubated for 24 h. The cells were then incubated with LMWH plus UA, LHU, and A-LHU nanodrugs (at equivalent LMWH concentration of 0.02, 0.2, 2, 20, 200 μg/mL, UA concentration of 1.25 × 10−3, 1.25 × 10−2, 0.125, 1.25, 12.5 μg/mL, respectively) for 72 h, respectively. After incubation, MTT solution (20 μL, 5 mg/mL in PBS) was added to each well and the cells were incubated further for 4 h at 37 °C. The media was then carefully removed, and the resulting formazan crystals were dissolved in DMSO. Absorbance was quantified at 490 nm using a microplate reader. Percent cell viability (%) was calculated as (OD of test group/OD of control group) × 100 (n = 5). 2.8. In vivo anti-tumor activity The anti-tumor activity of the LHU, DSPE-PEG-OCH3/LHU, and ALHU nanodrugs was evaluated in B16F10-bearing mice, all of which were female C57BL/6 mice (5–6 weeks). Subcutaneous injection of 9 × 106 cells in 100 μL of PBS into the right flanks of the mice were used to establish the tumor model. After each tumor became palpable (150–200 mm3), mice were divided randomly into five groups (n = 5) and were injected intravenously with either a 5% glucose solution as an untreated control, a solution of LMWH plus UA, LHU, DSPE-PEGOCH3/LHU, or A-LHU nanodrugs. The drug dose of all groups was 20 mg/kg/day of equivalent LMWH concentration and 1.05 mg/kg/day of equivalent UA concentration. Therapy was continued 4 times at 2day intervals using tail vein injections. The tumor volume was calculated using the following equation: V = 0.5 × a × b2 (a: the lengths of the longest tumor axis; b: the lengths of the vertical axis). After 8 days of treatment, all of the mice were sacrificed, and then their tumors and spleens were dissected and weighted. A pathologic examination was

111

performed after Hematoxylin and Eosin (HE) staining to investigate the anti-tumor effect of the formulations. 2.9. Immunohistochemical analysis CD31, a platelet/endothelial cell adhesion molecule, is a panendothelial cell antigen, distributing in the endothelial cells, platelets and lymphocytes. An immunohistochemical analysis was performed to determine the level of CD31, which evaluated the presence of endothelial cells and regeneration of tumor microvessels. The paraffinembedded sections were dewaxed, rehydrated, and microwaved for antigen retrieval. After cooling and washing in PBS, endogenous peroxidase was blocked with 3% hydrogen peroxide in methanol for 10 min, followed by incubation with 10% normal goat serum for 30 min. The specimens were incubated with anti-CD-31 antibody at a dilution of 1:200 for 2 h at 37 °C. The slides were pre-incubated and rinsed with PBS and then incubated with a secondary antibody for 30 min at 37 °C. Staining of the specimens was performed by adding fresh diaminobenzidine (DAB) solution into each slide. Counter staining was performed using hematoxylin for 10 min and then the slides were rinsed with distilled water. Vessels with a clearly defined lumen or well-defined linear vessel shape rather than single endothelial cells were considered as microvessels. The area of tissue with the largest number of distinctly highlighted microvessels was selected by scanning the immunostained sections at low magnification (40×). Microvessels were then counted in each selected area at high magnification (400×).The final microvessel density (MVD) was calculated as the average of five areas. 2.10. Western blot analysis Tumors were separated from mice after treatment and homogenized in modified lysis buffer (50 mM Tris–HCl with pH 7.5, 150 mM NaCl, 1% NP-40, 0.1% SDS, 1 mM EDTA, 0.2 mM phenylmethylsulfonyl fluoride, 0.1 mM Na3VO4 and 0.1 mM NaF). Protein was separated by 10% SDS-PAGE electrophoresis before being transferred to nitrocellulose membranes. 5% skim milk was used to block these membranes for 1 h at 37 °C followed by addition of an anti-VEGFR-2 antibody and an antiphospho-VEGFR-2 antibody, which were incubated at 4 °C overnight. A β-actin antibody was probed as the loading control. The membranes were washed three times and secondary antibodies were added and incubated at 37 °C for 1 h. Finally, the membranes were washed 3 times and the proteins were visualized by enhanced chemiluminescence (Odyssey). 2.11. Hemolysis assay A hemolysis assay was performed to evaluate the safety of the nanodrugs in vivo. Rabbit red blood cells (RBCs) were obtained by centrifuging rabbit whole blood at 3500 ×g for 10 min, removing the supernatant, washing the RBC pellet with the normal saline, and repeating until the supernatant was clear. Following the last wash, the RBCs were resuspended with normal saline to a concentration of 2% (w/v). Subsequently, UA, LMWH, LMWH plus UA and A-LHU were diluted with normal saline to different concentrations, incubated with 2% RBCs at 37 °C for 1 h, and then centrifuged at 3000 ×g for 10 min. Absorbance of hemoglobin in the supernatant was measured at 540 nm by UV–Vis spectroscopy. The observed hemolysis of RBCs in normal saline and distilled water were used as negative (0% hemolysis) and positive (100% hemolysis) controls, respectively. 2.12. Statistical analysis Data were expressed as mean ± SD. The statistical significance of group differences was analyzed using one-way unweighted mean

112

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

analysis of variance (ANOVA) and a value of p b 0.05 was considered significant.

The amphiphilic LHU conjugate was synthesized by chemically conjugating UA to a water-soluble LMWH backbone using ethylenediamine as the linker. As shown in Fig. 2B, the composition of synthesized LHU conjugate was analyzed by 1H NMR. The new amide linkage between LMWH and UA showed from 7.8 to 8.4 ppm and the intrinsic sulfonamide in LMWH appeared at 5.3 ppm [15,41]. The characteristic peaks of UA appeared at 1.0–1.5 ppm [42]. These results indicated that the LHU conjugate was synthesized successfully. The calculated average DS of UA in LHU conjugate was 13.2 ± 4.5%. The calculated molecular weight of LHU is approximately 5000 Da according to the DS of UA. The particle size of LHU nanodrugs were around 194.7 ± 3.7 nm. The CMC of the LHU conjugate was 257 μg/mL determined by a fluorescence technology using pyrene as the probe, suggesting that LHU nanodrug can maintain stability in dilute conditions, which may result in good stability after intravenous injection into the blood stream [43,44]. Moreover, LHU conjugate was relatively stable in the plasma; while its hydrolysis in the tumor homogenate was much more than in the plasma (in Table S1). It may be due to some enzymes and acidic environment in the tumor tissue.

increasing mass ratio of DSPE-PEG-AA to LHU. A-LHU nanodrugs exhibited large and non-uniform particle sizes at a ratio N 1:7. Therefore, the A-LHU nanodrugs containing 10% DSPE-PEG-AA (mass ratio of 1:9) with a particle size of 235.5 ± 6.1 nm and a negative charge were used for the following in vitro and in vivo studies. As shown in Fig. 3B, the AFM images revealed the spherical shape of A-LHU nanodrugs and exhibited the uniform size distribution of A-LHU nanodrugs. Moreover, Fig. 3C showed the TEM images of LHU and ALHU nanodrugs with 2% phosphotungstic acid as a contrast agent to investigate whether the A-LHU nanodrugs were formed successfully. Compared to the TEM image of LHU alone, the TEM image of A-LHU nanodrugs showed that the LHU (dark spot) was surrounded by a white disk of unstained lipids (DSPE-PEG-AA) (red arrows in the figure) that stands out against the stained background [45,46]. This may indicate the successful formation of an A-LHU nanodrug with the DSPE-PEG-AA in the outer shell, which increases the chance for targeted delivery of nanodrugs to tumor cells. The particle sizes of both LHU and A-LHU nanodrugs displayed in the AFM and TEM images were smaller than their hydrodynamic sizes determined by DLS since AFM and TEM images were obtained under a dehydrated condition. Overall, by conjugating UA to LMWH, the self-assembled nanodrugs could be formed in aqueous conditions. In this way, the simultaneous administration of UA and LMWH could be achieved by overcoming the insolubility of UA. In addition, the A-LHU nanodrugs coated by DSPE-PEG-AA were expected to target the tumor site via the EPR effect and enter cells via sigma receptor-mediated endocytosis.

3.2. Preparation and characterization of A-LHU nanodrugs

3.3. Anti-angiogenic effect of A-LHU nanodrugs in vitro

A-LHU nanodrugs were obtained by incubating the LHU conjugate with a DSPE-PEG-AA solution at room temperature for 60 min. As shown in Fig. 3A, the particle size of A-LHU nanodrugs increased with

To evaluate the anti-angiogenic inhibitory effect of the A-LHU nanodrugs, we grew HUVECs on polymerized Matrigel in a 96-well plate, treated them with UA, LMWH, LMWH plus UA, LHU, or A-LHU

3. Results 3.1. Synthesis and characterization of the LHU conjugate

Fig. 3. Characterization of A-LHU nanodrugs: (A) The particle size and zeta potential of A-LHU nanodrugs with different mass ratios of DSPE-PEG-AA:LHU; (B) AFM image of A-LHU nanodrugs; (C) TEM images of (a) LHU and (b) A-LHU nanodrugs with negative staining (Red arrows indicated unstained lipids of DSPE-PEG-AA surrounded LHU).

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

113

nanodrugs for 6 h, and observed the capillary-like tubular formation by optical microscopy (Fig. 4A). The endothelial cells in the control group were connected closely and well organized where large numbers of tubular formation occurred. DMSO, the solvent of UA, did not influence tubular formation. As compared to the control, LHU and A-LHU nanodrugs were able to disrupt the tubular formation, exhibiting more than 35% inhibition as compared to the control (p b 0.01) (Fig. 4B). Among all the tested groups, LMWH plus UA showed the maximum inhibition to tube formation, decreasing tubular formation by 45% (p b 0.01). Free UA and free LMWH were only able to inhibit tubular formation by 25% (p b 0.05) and 23% (p b 0.05), respectively. Therefore, the combination administration of LMWH and UA enhanced the antiangiogenesis in vitro than single free drug. The enhanced ability to disrupt the tubular formation by treatment with LMWH plus UA and A-LHU nanodrugs may be attributed to the additive inhibitory effect to angiogenesis by the combined administration of LMWH and UA.

those plugs. The plugs treated with LHU or A-LHU nanodrugs, however, looked pale in color, which was attributed to low neovascularization within the plugs, whereas there was no significant difference in the plugs between LHU and A-LHU. To quantitatively evaluate the in vivo anti-angiogenic effect of different groups, the hemoglobin content of each plug was determined after isolation and compared to the positive and negative controls. As shown in Fig. 5B, the hemoglobin contents of plugs with LHU or ALHU were only 11.3% and 10.1% of the positive control, respectively (p b 0.001). Therefore, the LHU and A-LHU exhibited the marked antiangiogenic capacity in vivo. Although the free combination of LMWH and UA showed the most effective tubular inhibition in vitro, they exhibited less effective blood vessel inhibition than LHU and A-LHU nanodrugs in vivo. This may be because their smaller size caused them to be quickly cleared from the Matrigel plug in a similar way to how they would be cleared from a tumor, lessening their sustained effect.

3.4. Anti-angiogenic effect of A-LHU nanodrugs in vivo

3.5. In vitro cellular uptake assay

The Matrigel plug assay was carried out to investigate whether ALHU nanodrugs inhibit the angiogenesis induced by bFGF in vivo. As shown in Fig. 5A, the Matrigel plug of the positive control group as well as the LMWH group exhibited a dark red color, which suggested that new blood vessels were formed within the plugs. In contrast, the plugs containing free UA or LMWH plus UA with bFGF showed light red color, indicating that new vessel formation had been reduced in

We have investigated the sigma receptors expression level of B16F10 cells, HUVECs and H460 cells by western blot assay. As shown in Fig. 6, HUVECs hardly expressed the sigma receptors; while B16F10 cells and H460 cells expressed much more sigma receptors than HUVECs. Meanwhile, the B16F10 cells expressed the highest level of sigma receptors among these three cell lines. Waterhouse et al. also reported that B16 melanoma cells express relatively high sigma receptors

Fig. 4. Tubular formation assay of HUVECs. (A) Representative morphologies of formed tubules of HUVECs treated with bFGF only (control), DMSO, UA, LMWH, LMWH plus UA, LHU, or ALHU nanodrugs (×100). (B) Quantitative analysis of the extent of tubular formation was obtained by counting the length of all tubes in the field. Data were expressed as mean ± SD (n = 5). *p b 0.05 and ***p b 0.01 vs. control.

114

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

Fig. 5. In vivo Matrigel plug assay. (A) Representative isolated Matrigel plugs from each group. (B) Relative hemoglobin content in isolated Matrigels: Matrigel containing bFGF only (positive control), Matrigel containing bFGF with LMWH, UA, LMWH plus UA, LHU, or A-LHU nanodrugs. ***p b 0.001 vs. control.

level and maximum number of binding sites of receptors (Bmax) was 1.8 pmol/mg protein [47]. In the cellular uptake assay, we investigate the cellular uptake of A-LHU nanodrugs by B16F10 cells (high sigma receptors expression) and HUVECs (low sigma receptors expression). To investigate the cellular uptake efficacy of A-LHU nanodrugs, coumarin-6 was used as the hydrophobic fluorescent marker by encapsulation into the LHU nanodrugs. Fig. 7 shows the LHU and A-LHU nanodrugs at 2 h and 6 h after incubation with B16F10 cells and HUVECs. The free coumarin-6 group showed little green fluorescence both at 2 h and 6 h after incubation in B16F10 cells and HUVECs. In B16F10 cells that were treated with coumarin-6-loaded LHU and ALHU for 2 h and 6 h, the green fluorescence became stronger as the incubation time increased, indicating the nanodrugs accumulated in the cells over time. As shown in Fig. 7B, after incubation for 6 h, A-LHU nanodrugs containing 10% DSPE-PEG-AA (1:9) exhibited the most efficient cellular uptake by B16F10 cells. The cellular uptake level of targeted A-LHU nanodrugs containing 5% DSPE-PEG-AA (1:19) was

reduced compared with those nanodrugs containing 10% DSPE-PEGAA. The LHU nanodrugs showed less uptake compared to A-LHU nanodrugs. These findings indicated that the enhanced cellular uptake of A-LHU nanodrugs was partially mediated by the sigma receptors. In addition, the flow cytometry results clearly show that A-LHU exhibited more efficient cellular uptake by B16F10 cells after 6 h incubation compared to LHU. As shown in Fig. 7C, B16F10 cells treated with coumarin-6-loaded A-LHU nanodrugs containing 10% DSPE-PEG-AA showed significantly higher fluorescence intensity than control and LHU (p b 0.05). Moreover, it had a 1.6 times higher fluorescence intensity than that of LHU nanodrugs. To further investigate if A-LHU nanodrugs specifically enter the cell via the sigma receptor, the cellular uptake of A-LHU nanodrugs in HUVECs, which express much less sigma receptor than B16F10 cells, was performed. As shown in Fig. 7D, the fluorescence intensity of coumarin-6 from A-LHU nanodrugs in HUVECs was much weaker than that in B16F10 cells after incubation for 6 h. Moreover, there was no significant difference between cellular uptake of targeted A-LHU nanodrugs and that of LHU nanodrugs. The findings further suggested that the A-LHU nanodrugs could readily internalize into the cancer cells via receptor-mediated endocytosis. 3.6. In vitro cytotoxicity studies

Fig. 6. (A) Western-blot assay image of sigma receptor expression in B16F10 cells, HUVECs and H460 cells. (B) The ratio of integrated density of sigma receptor to GAPDH. The integrated density of each band of the resulting image was analyzed by image analyzer (image J). **pb0.01.

The potential of LHU and A-LHU nanodrugs to kill the B16F10 cells was investigated by the MTT assay. As shown in Fig. 8, all three tested groups exhibited concentration-dependent cytotoxicity against B16F10 cells in vitro. The IC50 values for LMWH plus UA, LHU, and ALHU were 13, 84, and 63 μg/mL, respectively. The LMWH plus UA group exhibited the most effective cytotoxicity against B16F10 cells. We believe that the cytotoxicity of the three formulations was mainly attributed to the UA, as previous studies have indicated that LMWH shows no cytotoxicity at these concentrations [41,48,49]. In contrast, the obvious cytotoxicity of UA against the B16F10 cells have been reported in some studies [50]. The increased cytotoxicity of the free drug combination is not surprising, and may be because these free drugs are easily internalized by cells by passive diffusion due to their smaller molecular weight and more hydrophobic (for UA) nature. Because of this, they can quickly provide their therapeutic effect and kill the cells without requiring a drug release process [44]. In contrast, the lower in vitro cytotoxicity from the larger LHU and A-LHU nanodrugs

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

115

Fig. 7. In vitro cell uptake assay. (A) Confocal microscopy image of B16F10 cells after 2 h and (B) 6 h incubation in vitro. (C) Cellular uptake of different test groups by B16F10 cells after 6 h incubation by flow cytometry. (D) Confocal microscopy image of HUVECs after 6 h incubation in vitro.*p b 0.05.

was likely because they first need to be internalized via endocytosis before they could release their drug payload. The greater cytotoxicity of free drugs in vitro is usually reversed in vivo when long circulation times and residence times in the tumor become major factors. No significant cytotoxicity of nanodrugs was observed at relatively low concentrations (less than 20 μg/mL). In fact, properly low cytotoxicity of nanodrugs in the process of in vivo delivery might contribute to the reduced potential side effect while not affecting the in vivo efficacy of nanodrugs in the tumor site. A-LHU did show more cytotoxicity than LHU at the concentration of 200 μg/mL (p b 0.01), which can be attributed to improved cellular uptake by B16F10 cells [39,51]. 3.7. In vivo antitumor activity

Fig. 8. In vitro cytotoxicity of LMWH plus UA, LHU, and A-LHU as a function of LMWH concentration in B16F10 cells. Data are represented as mean ± SD (n = 5). ***p b 0.01.

The antitumor efficacy of A-LHU was investigated in a B16F10 murine tumor model. The non-targeted nanodrugs modified with DSPE-PEG-OCH3 (DSPE-PEG-OCH3/LHU) were also studied to discuss the role of DSPE-PEG-AA in the tumor growth inhibition of A-LHU nanodrugs. As shown in Fig. 9A, the tumor volume in mice treated with free LMWH plus UA exhibited no significant reduction when compared to the control. However, a marked decrease of tumor volume could be observed in the other three groups treated with LHU, DSPE-

116

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

PEG-OCH3/LHU, and A-LHU. Notably, A-LHU yielded the most significant antitumor activity in all tested groups. The tumor growth rate was reduced by 53% after a schedule of multiple doses as compared to the control (p b 0.01), while the tumor volume in the LHU group and DSPE-PEG-OCH3/LHU group decreased by ~ 30% compared to the control (p b 0.05). The tumors were excised for pathological examination after the therapy. As shown in Fig. 9B are the tumor sections stained by HE. The hematoxylin stains the nuclei blue or purple, while the eosin stains the cytoplasm red or pink [52]. The group injected with LHU or DSPEPEG-OCH3/LHU showed more necrotic tumor area than the control (glucose solution) and LMWH plus UA groups. The tumor section in the ALHU group exhibited the largest area of necrosis and thus possessed the most effective inhibition to tumor growth compared to other groups. Therefore, the pathological examination also demonstrated that A-LHU nanodrugs exhibited the high antitumor efficacy in vivo. To evaluate the potential toxicity of different treatment groups, the mouse body weight change and spleen index were monitored. As shown in Fig. 9C, no reduced body weight was observed in any group during the treatment. The spleen plays a major role for immune response and contains a large amount of lymphocytes. Immunopotentiator could increase the weight of spleen [53]. As compared to the control, the spleen in all of the treatment groups showed a significant enlargement (p b 0.05), indicating that LMWH and UA have fewer side effects on the immune system than the control (Fig. 9D). Overall, A-LHU nanodrugs exhibited excellent antitumor efficacy as well as low toxicity.

3.8. Anti-angiogenic effect in tumor bearing mice CD31 is a transmembrane glycoprotein also designated as platelet endothelial cell adhesion molecule 1. CD31 plays an important role in the adhesion cascade between endothelial cells during angiogenesis. Therefore, it has been a commonly used cytoplasmic endothelial cell marker. To determine whether the antitumor efficacy of the nanoparticles was related to an anti-angiogenic effect, tumor tissue sections were stained with anti-CD31 antibody and then visualized under a microscope. As shown in Fig. 10A (a), the density of anti-CD31 positive microvessels (brown color) reduced after administration of LMWH plus UA, LHU, DSPE-PEG-OCH3/LHU and A-LHU. As shown in Fig. 10A (b), the MVD of the LMWH plus UA group was lower than the control (p b 0.05), indicating the combination of LMWH and UA could inhibit angiogenesis in vivo. An even lower MVD was displayed after administration of LHU, DSPE-PEG-OCH 3 /LHU, and A-LHU nanodrugs (p b 0.01). In particular, A-LHU exhibited a microvessel inhibition of 81% (p b 0.01), which was higher than all other groups. VEGF is an important regulator of vascular development that can induce phosphorylation of VEGF receptor-2 (VEGFR-2), which can then activate downstream signaling that results in endothelial cell migration, proliferation, and survival. The western blot assay was used to show the expression level of VEGFR-2 and p-VEGFR-2 in the tumor tissues after treatment with LMWH plus UA, LHU, and A-LHU. As shown in Fig. 10B, the expression level of VEGFR-2 in the three tested groups showed no significant difference with the control. However, it was found that LHU and A-LHU nanodrugs could significantly down-

Fig. 9. In vivo antitumor activity. (A) Tumor growth curve of mice bearing B16F10 tumors. Mice were treated every other day for four total treatments. *p b 0.05 and ***p b 0.01 vs. the control group, #p b 0.05 vs. LHU and DSPE-PEG-OCH3/LHU. (B) HE staining of tumor sections from B16F10 tumor-bearing mice treated with glucose solution, LMWH plus UA, LHU, DSPE-PEG-OCH3/LHU nanodrugs, and A-LHU nanodrugs. (C) Change of body weight and (D) Spleen index of B16F10 tumor-bearing mice in different groups. Data are represented as mean ± SD (n = 5).

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

117

Fig. 10. Anti-angiogenic effect of LMWH plus UA, LHU, DSPE-PEG-OCH3/LHU and A-LHU nanodrugs in mice bearing B16F10 cells. (A) Immunohistochemical analysis (a) Microphotographs of anti-CD31 antibody immunostaining against microvessels on the isolated tumor tissue sections. (b) The numbers of anti-CD31 immunostained microvessels in the tumor tissues. Data were expressed as mean ± SD (n = 5). ***p b 0.01 vs. the control. (B) Western blot assay for VEGFR-2 and p-VEGFR-2. Protein in the tumors from B16F10 tumor-bearing mice after a schedule of 4 doses was extracted and analyzed.

regulate the expression of p-VEGFR-2. In particular, A-LHU nanodrugs produced the greatest inhibition of p-VEGFR-2. This indicates that ALHU was most effective at competitively inhibiting the binding of VEGF-2 with VEGFR-2, which effectively inhibited angiogenesis and tumor vascular endothelial cell proliferation. 3.9. Hemolysis

low efficiency of in vivo tumor delivery in the complex physiologic environment. UA has also been reported as a potent anti-angiogenic agent that can retard invasion and migration of cancer cells [19]. To achieve a combined angiogenetic inhibition of both LMWH and UA and overcome the bottleneck of in vivo delivery of free drugs, we chemically conjugated UA to LMWH to construct a nanodrug formed by self-assembly of its amphiphilic dual-drug monomers. Furthermore, DSPE-PEG-AA

A safety evaluation of intravenous administration of the polymeric nanodrugs and controls was completed in vivo. Hemolysis from nanodrug administration was carried out to test for hemocompatibility. As shown in Fig. 11, within the range of measurement, all of the tested groups exhibited less than 2% hemolysis relative to the negative control, indicating that these drugs pose no risk of hemolysis-related toxicity, which would be reckoned at levels above 5%. LMWH showed an ability to inhibit hemolysis, which is consistent with the literature [54]. Even at the highest investigated concentration of 1.6 mg/mL, A-LHU nanodrugs only displayed a hemolysis of 1.85%, which is still negligible. These results show that A-LHU nanodrugs should not be toxic toward erythrocytes after intravenous injection. 4. Discussion LMWH has been given more attention as a cancer treatment based on its excellent anti-angiogenic activity by competitively inhibiting various angiogenic factors such as VEGF and bFGF. However, the antitumor efficacy of free LMWH in tumor-bearing mice is not high, possibly due to

Fig. 11. The hemolysis of different groups at different drug concentrations.

118

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119

was used as a targeting ligand for the sigma receptor through its addition on the surface of the nanodrugs to improve their cellular internalization and tumor accumulation. Our studies demonstrated that ALHU nanodrugs exhibited an improved anti-angiogenic effect both in vitro and in vivo compared to the free combination of LMWH and UA. The tubular formation of HUVECs was effectively inhibited by A-LHU nanodrugs in vitro. The Matrigel plug assay also showed significantly decreased hemoglobin content in Matrigel plugs in the A-LHU group, indicating that A-LHU nanodrugs could suppress angiogenesis through the migration inhibition of endothelial cells to the Matrigel plug in vivo [55]. It is known that N- and 2-O-sulfate groups in the heparin structure are essential to bind FGF-2 while stimulation of FGFR-1 and Erk2 kinases by FGF-2 also require the presence of 6-O-sulfate groups. It has been reported that N- and O-sulfate groups play an important role in the interaction of heparin or heparan sulfate with different angiogenic growth factors [56,57]. The above results show that A-LHU was the most successful formulation in vivo, which indicates that conjugating LMWH and UA did not significantly decrease their anti-angiogenic efficacy. In this study, LHU still maintained its anti-angiogenic activity because the hydrophobic UA molecule was introduced to the carboxylic groups of LMWH without changing the sulfation pattern of LMWH. Park and Lee et al. have described that the anti-angiogenic effect of LMWH is actually enhanced by increasing its hydrophobicity and reducing its chain flexibility [15,58]. In this study, conjugation of the hydrophobic UA molecule may have increased the potency of LMWH itself. Although small molecule drugs can be quickly internalized into cells and show profound effect in vitro, rapid inactivation, excretion, and non-specific distribution will all occur in vivo, resulting in little drug accumulation in the tumor and more side effects in other organs, which seriously limit the effect of their cancer therapy [59]. Our studies show that A-LHU nanodrugs overcome these obstacles and exhibit a higher antitumor effect than other tested groups in a B16F10 tumor-bearing mouse model. The A-LHU surface was modified with DSPE-PEG, a widely used moiety to prevent rapid uptake of nanoparticles by the mononuclear phagocyte system [60], and by conjugating the targeting ligand anisamide to the distal end of the PEG, A-LHU nanodrugs were more effectively accumulated in the tumor tissue. Soluble heparin competes with heparan sulfates (HS) on heparan sulfate proteoglycans (HSPG) for binding with growth factors to release these protein [40]. Due to HSPG are localized in the ECM and on the cell surface [15,40], A-LHU nanodrugs can induce the inhibition on angiogenesis before internalization into the cell. Part of A-LHU nanodrugs can be also internalized into the tumor cells mediated by sigma receptor. The hydrolysis of amide bond between UA and LMWH can be increased in the slightly acidic lysosomal environment and the lysosomal enzymes [61,62]. The released UA inhibited VEGF secretion by tumor cells to induce the inhibition on angiogenesis. Moreover, UA also induced the cell cytotoxicity in the cell, which can also increase the antitumor effect of A-LHU nanodrugs in vivo. Therefore, the A-LHU nanodrugs were most effective at suppressing tumor growth in vivo due to longer circulation conferred by PEG modification, EPR-related accumulation of the nano-scaled drug, and receptor-mediated endocytosis. A-LHU's multi-faceted mechanism of tumor growth inhibition may include cytotoxicity, CD-31 microvessel downregulation, and inhibition of VEGFR-2 phosphorylation. 5. Conclusion In this study, we successfully developed A-LHU nanodrugs with a UA inner core and a LMWH outer shell. DSPE-PEG-AA was incorporated into the nanodrugs to provide longer circulation and active targeting. A-LHU exhibited high angiogenic inhibition in vitro and in vivo, which resulted from the combined anti-angiogenic activity achieved by LMWH and UA. Based on the EPR effect and sigma receptor-mediated endocytosis, A-LHU nanodrugs dramatically suppressed tumor growth in vivo. Overall, the A-LHU delivery system can be regarded as a novel and effective cancer therapy.

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jconrel.2016.02.044. Acknowledgment This project was supported by the Jiangsu Natural Science Foundation of China (BK20131312), the 12th of Six Talent Peak Foundation of Jiangsu Province (YY-001), the National Natural Science Foundation of China (No. 81173006), the Project Program of State Key Laboratory of Natural Medicines, China Pharmaceutical University (JKGQ201107), and the Fundamental Research Funds for the Central Universities (JKPZ2013005). It was also sponsored by Qing Lan Project and Jiangsu Overseas Research and Training Program for University Prominent Young and Middle-aged Teachers and Presidents. It was also supported by NIH grants CA151652 and CA149387. References [1] J. Folkman, Role of angiogenesis in tumor growth and metastasis, Semin. Oncol. 29 (2002) 15–18. [2] A. Jones, A.L. Harris, New developments in angiogenesis: a major mechanism for tumor growth and target for therapy, Cancer J. Sci. Am. 4 (1998) 209–217. [3] I. Debergh, N. Van Damme, P. Pattyn, M. Peeters, W.P. Ceelen, The low-molecularweight heparin, nadroparin, inhibits tumour angiogenesis in a rodent dorsal skinfold chamber model, Br. J. Cancer 102 (2010) 837–843. [4] C.H. Lee, J.H. Kim, H.J. Lee, K. Jeon, H. Lim, H. Choi, E.R. Lee, S.H. Park, J.Y. Park, S. Hong, S. Kim, S.G. Cho, The generation of iPS cells using non-viral magnetic nanoparticle based transfection, Biomaterials 32 (2011) 6683–6691. [5] E. Ruoslahti, Y. Yamaguchi, Proteoglycans as modulators of growth factor activities, Cell 64 (1991) 867–869. [6] J.L. Andres, D. DeFalcis, M. Noda, J. Massague, Binding of two growth factor families to separate domains of the proteoglycan betaglycan, J. Biol. Chem. 267 (1992) 5927–5930. [7] S. Colin, J.C. Jeanny, F. Mascarelli, R. Vienet, S. Al-Mahmood, Y. Courtois, J. Labarre, In vivo involvement of heparan sulfate proteoglycan in the bioavailability, internalization, and catabolism of exogenous basic fibroblast growth factor, Mol. Pharmacol. 55 (1999) 74–82. [8] J. Folkman, Antiangiogenic gene therapy, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 9064–9066. [9] K. Norrby, Low-molecular-weight heparins and angiogenesis, APMIS 114 (2006) 79–102. [10] K. Norrby, 2.5 kDa and 5.0 kDa heparin fragments specifically inhibit microvessel sprouting and network formation in VEGF165-mediated mammalian angiogenesis, Int. J. Exp. Pathol. 81 (2000) 191–198. [11] K. Norrby, P. Ostergaard, Basic-fibroblast-growth-factor-mediated de novo angiogenesis is more effectively suppressed by low-molecular-weight than by highmolecular-weight heparin, Int. J. Microcirc. Clin. Exp. 16 (1996) 8–15. [12] A. Collen, S.M. Smorenburg, E. Peters, F. Lupu, P. Koolwijk, C. Van Noorden, V.W. van Hinsbergh, Unfractionated and low molecular weight heparin affect fibrin structure and angiogenesis in vitro, Cancer Res. 60 (2000) 6196–6200. [13] N. Back, R. Steger, Effect of aprotinin, EACA and heparin on growth and vasopeptide system of Murphy–Sturm lymphosarcoma, Eur. J. Pharmacol. 38 (1976) 313–319. [14] T.M. Niers, C.P. Klerk, M. DiNisio, C.J. Van Noorden, H.R. Buller, P.H. Reitsma, D.J. Richel, Mechanisms of heparin induced anti-cancer activity in experimental cancer models, Crit. Rev. Oncol. Hematol. 61 (2007) 195–207. [15] J.W. Park, O.C. Jeon, S.K. Kim, T.A. Al-Hilal, S.J. Jin, H.T. Moon, V.C. Yang, S.Y. Kim, Y. Byun, High antiangiogenic and low anticoagulant efficacy of orally active low molecular weight heparin derivatives, J. Control. Release 148 (2010) 317–326. [16] K. Ono, M. Ishihara, K. Ishikawa, Y. Ozeki, H. Deguchi, M. Sato, H. Hashimoto, Y. Saito, H. Yura, A. Kurita, T. Maehara, Periodate-treated, non-anticoagulant heparincarrying polystyrene (NAC-HCPS) affects angiogenesis and inhibits subcutaneous induced tumour growth and metastasis to the lung, Br. J. Cancer 86 (2002) 1803–1812. [17] K. Park, Y.S. Kim, G.Y. Lee, R.W. Park, I.S. Kim, S.Y. Kim, Y. Byun, Tumor endothelial cell targeted cyclic RGD-modified heparin derivative: inhibition of angiogenesis and tumor growth, Pharm. Res. 25 (2008) 2786–2798. [18] L. Hou, J. Yao, J. Zhou, Q. Zhang, Pharmacokinetics of a paclitaxel-loaded low molecular weight heparin-all-trans-retinoid acid conjugate ternary nanoparticulate drug delivery system, Biomaterials 33 (2012) 5431–5440. [19] M. Kanjoormana, G. Kuttan, Antiangiogenic activity of ursolic acid, Integr. Cancer Ther. 9 (2010) 224–235. [20] K. Subbaramaiah, P. Michaluart, M.B. Sporn, A.J. Dannenberg, Ursolic acid inhibits cyclooxygenase-2 transcription in human mammary epithelial cells, Cancer Res. 60 (2000) 2399–2404. [21] M.T. Huang, C.T. Ho, Z.Y. Wang, T. Ferraro, Y.R. Lou, K. Stauber, W. Ma, C. Georgiadis, J.D. Laskin, A.H. Conney, Inhibition of skin tumorigenesis by rosemary and its constituents carnosol and ursolic acid, Cancer Res. 54 (1994) 701–708. [22] S. Manez, M.C. Recio, R.M. Giner, J.L. Rios, Effect of selected triterpenoids on chronic dermal inflammation, Eur. J. Pharmacol. 334 (1997) 103–105. [23] K.H. Sohn, H.Y. Lee, H.Y. Chung, H.S. Young, S.Y. Yi, K.W. Kim, Anti-angiogenic activity of triterpene acids, Cancer Lett. 94 (1995) 213–218.

Y. Li et al. / Journal of Controlled Release 228 (2016) 107–119 [24] T.D. Gilmore, Introduction to NF-kappaB: players, pathways, perspectives, Oncogene 25 (2006) 6680–6684. [25] Z. Wang, A. Ahmad, Y. Li, S. Banerjee, D. Kong, F.H. Sarkar, Forkhead box M1 transcription factor: a novel target for cancer therapy, Cancer Treat. Rev. 36 (2010) 151–156. [26] J.S. Wang, T.N. Ren, T. Xi, Ursolic acid induces apoptosis by suppressing the expression of FoxM1 in MCF-7 human breast cancer cells, Med. Oncol. 29 (2012) 10–15. [27] C. Cardenas, A.R. Quesada, M.A. Medina, Effects of ursolic acid on different steps of the angiogenic process, Biochem. Biophys. Res. Commun. 320 (2004) 402–408. [28] H.J. Cha, S.K. Bae, H.Y. Lee, O.H. Lee, H. Sato, M. Seiki, B.C. Park, K.W. Kim, Antiinvasive activity of ursolic acid correlates with the reduced expression of matrix metalloproteinase-9 (MMP-9) in HT1080 human fibrosarcoma cells, Cancer Res. 56 (1996) 2281–2284. [29] M.Y. Lai, H.W. Leung, W.H. Yang, W.H. Chen, H.Z. Lee, Up-regulation of matrix metalloproteinase family gene involvement in ursolic acid-induced human lung nonsmall carcinoma cell apoptosis, Anticancer Res. 27 (2007) 145–153. [30] M.C. Garnett, Targeted drug conjugates: principles and progress, Adv. Drug Deliv. Rev. 53 (2001) 171–216. [31] F. Greco, M.J. Vicent, Combination therapy: opportunities and challenges for polymer-drug conjugates as anticancer nanomedicines, Adv. Drug Deliv. Rev. 61 (2009) 1203–1213. [32] B.J. Vilner, C.S. John, W.D. Bowen, Sigma-1 and sigma-2 receptors are expressed in a wide variety of human and rodent tumor cell lines, Cancer Res. 55 (1995) 408–413. [33] C.S. John, W.D. Bowen, T. Saga, S. Kinuya, B.J. Vilner, J. Baumgold, C.H. Paik, R.C. Reba, R.D. Neumann, V.M. Varma, et al., A malignant melanoma imaging agent: synthesis, characterization, in vitro binding and biodistribution of iodine-125-(2piperidinylaminoethyl)4-iodobenzamide, J. Nucl. Med. 34 (1993) 2169–2175. [34] C.S. John, B.J. Vilner, W.D. Bowen, Synthesis and characterization of [125I]-N-(Nbenzylpiperidin-4-yl)-4- iodobenzamide, a new sigma receptor radiopharmaceutical: high-affinity binding to MCF-7 breast tumor cells, J. Med. Chem. 37 (1994) 1737–1739. [35] C.S. John, B.J. Vilner, M.E. Gulden, S.M. Efange, R.B. Langason, T.W. Moody, W.D. Bowen, Synthesis and pharmacological characterization of 4-[125I]-N-(Nbenzylpiperidin-4-yl)-4-iodobenzamide: a high affinity sigma receptor ligand for potential imaging of breast cancer, Cancer Res. 55 (1995) 3022–3027. [36] C.S. John, M.E. Gulden, B.J. Vilner, W.D. Bowen, Synthesis, in vitro validation and in vivo pharmacokinetics of [125I]N-[2-(4-iodophenyl)ethyl]-N-methyl-2-(1piperidinyl) ethylamine: a high-affinity ligand for imaging sigma receptor positive tumors, Nucl. Med. Biol. 23 (1996) 761–766. [37] C.S. John, B.J. Vilner, B.C. Geyer, T. Moody, W.D. Bowen, Targeting sigma receptorbinding benzamides as in vivo diagnostic and therapeutic agents for human prostate tumors, Cancer Res. 59 (1999) 4578–4583. [38] R.R. Sawant, R.M. Sawant, V.P. Torchilin, Mixed PEG-PE/vitamin E tumor-targeted immunomicelles as carriers for poorly soluble anti-cancer drugs: improved drug solubilization and enhanced in vitro cytotoxicity, Eur. J. Pharm. Biopharm. 70 (2008) 51–57. [39] R. Banerjee, P. Tyagi, S. Li, L. Huang, Anisamide-targeted stealth liposomes: a potent carrier for targeting doxorubicin to human prostate cancer cells, Int. J. Cancer 112 (2004) 693–700. [40] É. Oganesyan, Mechanism of the reaction of triterpenoids with sulfuric acid, Chem. Nat. Compd. 16 (1980) 464–468. [41] L. Hou, Y. Fan, J. Yao, J. Zhou, C. Li, Z. Fang, Q. Zhang, Low molecular weight heparinall-trans-retinoid acid conjugate as a drug carrier for combination cancer chemotherapy of paclitaxel and all-trans-retinoid acid, Carbohydr. Polym. 86 (2011) 1157–1166. [42] Y. Kashiwada, T. Nagao, A. Hashimoto, Y. Ikeshiro, H. Okabe, L.M. Cosentino, K.H. Lee, Anti-AIDS agents 38. Anti-HIV activity of 3-O-acyl ursolic acid derivatives, J. Nat. Prod. 63 (2000) 1619–1622. [43] K.Y. Lee, W.H. Jo, I.C. Kwon, Y.-H. Kim, S.Y. Jeong, Structural determination and interior polarity of self-aggregates prepared from deoxycholic acid-modified chitosan in water, Macromolecules 31 (1998) 378–383.

119

[44] J. Yao, L. Zhang, J. Zhou, H. Liu, Q. Zhang, Efficient simultaneous tumor targeting delivery of all-trans retinoid acid and paclitaxel based on hyaluronic acid-based multifunctional nanocarrier, Mol. Pharm. 10 (2013) 1080–1091. [45] B. Dubertret, P. Skourides, D.J. Norris, V. Noireaux, A.H. Brivanlou, A. Libchaber, In vivo imaging of quantum dots encapsulated in phospholipid micelles, Science 298 (2002) 1759–1762. [46] J. Li, Y. Yang, L. Huang, Calcium phosphate nanoparticles with an asymmetric lipid bilayer coating for siRNA delivery to the tumor, J. Control. Release 158 (2012) 108–114. [47] R.N. Waterhouse, J. Chapman, B. Izard, A. Donald, K. Belbin, J.C. O'Brien, T.L. Collier, Examination of four 123I-labeled piperidine-based sigma receptor ligands as potential melanoma imaging agents: initial studies in mouse tumor models, Nucl. Med. Biol. 24 (1997) 587–593. [48] J. Park, J.Y. Kim, S.R. Hwang, F. Mahmud, Y. Byun, Chemical conjugate of low molecular weight heparin and suramin fragment inhibits tumor growth possibly by blocking VEGF165, Mol. Pharm. 12 (2015) 3935–3942. [49] D.H. Kim, U. Termsarasab, H.J. Cho, I.S. Yoon, J.Y. Lee, H.T. Moon, D.D. Kim, Preparation and characterization of self-assembled nanoparticles based on low-molecularweight heparin and stearylamine conjugates for controlled delivery of docetaxel, Int. J. Nanomedicine 9 (2014) 5711–5727. [50] M. Yoshida, M. Fuchigami, T. Nagao, H. Okabe, K. Matsunaga, J. Takata, Y. Karube, R. Tsuchihashi, J. Kinjo, K. Mihashi, T. Fujioka, Antiproliferative constituents from Umbelliferae plants VII. Active triterpenes and rosmarinic acid from Centella asiatica, Biol. Pharm. Bull. 28 (2005) 173–175. [51] S.D. Li, S. Chono, L. Huang, Efficient gene silencing in metastatic tumor by siRNA formulated in surface-modified nanoparticles, J. Control. Release 126 (2008) 77–84. [52] D. Berrington, N. Lall, Anticancer activity of certain herbs and spices on the cervical epithelial carcinoma (HeLa) cell line, Evid. Based Complement. Alternat. Med. 2012 (2012) 564927. [53] X. Chen, W. Nie, G. Yu, Y. Li, Y. Hu, J. Lu, L. Jin, Antitumor and immunomodulatory activity of polysaccharides from Sargassum fusiforme, Food Chem. Toxicol. 50 (2012) 695–700. [54] G.L. Logue, Effect of heparin on complement activation and lysis of paroxysmal nocturnal hemoglobinuria (PNH) red cells, Blood 50 (1977) 239–247. [55] J.Y. Kim, T.A. Al-Hilal, S.W. Chung, S.Y. Kim, G.H. Ryu, W.C. Son, Y. Byun, Antiangiogenic and anticancer effect of an orally active low molecular weight heparin conjugates and its application to lung cancer chemoprevention, J. Control. Release 199 (2015) 122–131. [56] S. Guimond, M. Maccarana, B.B. Olwin, U. Lindahl, A.C. Rapraeger, Activating and inhibitory heparin sequences for FGF-2 (basic FGF). Distinct requirements for FGF-1, FGF-2, and FGF-4, J. Biol. Chem. 268 (1993) 23906–23914. [57] L. Lundin, H. Larsson, J. Kreuger, S. Kanda, U. Lindahl, M. Salmivirta, L. ClaessonWelsh, Selectively desulfated heparin inhibits fibroblast growth factor-induced mitogenicity and angiogenesis, J. Biol. Chem. 275 (2000) 24653–24660. [58] E. Lee, Y.S. Kim, S.M. Bae, S.K. Kim, S. Jin, S.W. Chung, M. Lee, H.T. Moon, O.C. Jeon, R.W. Park, I.S. Kim, Y. Byun, S.Y. Kim, Polyproline-type helical-structured lowmolecular weight heparin (LMWH)-taurocholate conjugate as a new angiogenesis inhibitor, Int. J. Cancer 124 (2009) 2755–2765. [59] J. Yao, Y. Zhang, S. Ramishetti, Y. Wang, L. Huang, Turning an antiviral into an anticancer drug: nanoparticle delivery of acyclovir monophosphate, J. Control. Release 170 (2013) 414–420. [60] D.S. Kohane, Microparticles and nanoparticles for drug delivery, Biotechnol. Bioeng. 96 (2007) 203–209. [61] T. Etrych, J. Strohalm, L. Kovar, M. Kabesova, B. Rihova, K. Ulbrich, HPMA copolymer conjugates with reduced anti-CD20 antibody for cell-specific drug targeting. I. Synthesis and in vitro evaluation of binding efficacy and cytostatic activity, J. Control. Release 140 (2009) 18–26. [62] T. Etrych, V. Subr, R. Laga, B. Rihova, K. Ulbrich, Polymer conjugates of doxorubicin bound through an amide and hydrazone bond: impact of the carrier structure onto synergistic action in the treatment of solid tumours, Eur. J. Pharm. Sci. 58 (2014) 1–12.