Calcium phosphate nanoparticles functionalized with alendronate-conjugated polyethylene glycol (PEG) for the treatment of bone metastasis

Accepted Manuscript Title: Calcium phosphate nanoparticles functionalized with alendronate-conjugated polyethylene glycol (PEG) for the treatment of bone metastasis Author: Weijing Chu Yanjuan Huang Chanzhen Yang Yunhui Liao Xuefei Zhang Mina Yan Shengmiao Cui Chunshun Zhao PII: DOI: Reference:

S0378-5173(16)31114-0 http://dx.doi.org/doi:10.1016/j.ijpharm.2016.11.051 IJP 16255

To appear in:

International Journal of Pharmaceutics

Received date: Revised date: Accepted date:

14-9-2016 13-11-2016 22-11-2016

Please cite this article as: Chu, Weijing, Huang, Yanjuan, Yang, Chanzhen, Liao, Yunhui, Zhang, Xuefei, Yan, Mina, Cui, Shengmiao, Zhao, Chunshun, Calcium phosphate nanoparticles functionalized with alendronate-conjugated polyethylene glycol (PEG) for the treatment of bone metastasis.International Journal of Pharmaceutics http://dx.doi.org/10.1016/j.ijpharm.2016.11.051 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Calcium phosphate nanoparticles functionalized with alendronate-conjugated polyethylene glycol (PEG) for the treatment of bone metastasis

Weijing Chua,b,1, Yanjuan Huanga,1, Chanzhen Yanga, Yunhui Liaoa, Xuefei Zhanga, Mina Yana, Shengmiao Cuic,*, Chunshun Zhaoa,*

a

School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, People’s Republic

of China, 510006 b

Department of Pharmacy, Quanzhou Medical College, Quanzhou, People’s Republic of

China, 362100 c

School of Traditional Medicine, Guangdong Pharmaceutical University, Guangzhou,

People’s Republic of China, 51006 * Corresponding author: Shengmiao Cui, Tel: +86 20 39352169; Fax: +86 20 39352174; Email address: [email protected]. Chunshun Zhao, Tel: +86 20 39943118; Fax: +86 20 39943118; E-mail address: [email protected]. 1

Weijing Chu and Yanjuan Huang contributed equally to this work and should be considered

as co-first authors.

1

Graphical Abstract

Abstract Because of the peculiarity of the bone microstructure, the uptake of chemotherapeutics often happens at non-targeted sites, which induces side effects. In order to solve this problem, we designed a bone-targeting drug delivery system that can release drug exclusively in the nidus of the bone. Alendronate (ALN), which has a high ability to target to hydroxyapatite, was used to fabricate double ALN-conjugated poly (ethylene glycol) 2000 material (ALNPEG2k-ALN). The ALN-PEG2k-ALN was characterized using 1H NMR and 31P NMR and FTIR. ALN-PEG2k-ALN-modified calcium phosphate nanoparticles (APA-CPNPs) with an ALN targeting moiety and hydrophilic poly (ethylene glycol) arms tiled on the surface was prepared for bone-targeted drug delivery. The distribution of ALN-PEG2k-ALN was tested by X-ray photoelectron spectroscopy. Isothermal titration calorimetry data indicated that similar to free ALN, both ALN-PEG2k-ALN and APA-CPNPs can bind to calcium ions. The bonebinding ability of APA-CPNPs was verified via ex vivo imaging of bone fragments. An in vitro release experiment demonstrated that APA-CPNPs can release drug faster in an acid environment than a neutral environment. Cell viability experiments indicated that blank APA2

CPNPs possessed excellent biocompatibility with normal cells. Methotrexate (MTX) loaded APA-CPNPs have the same ability to inhibit cancer cells as free drug at high concentrations, while they are slightly weaker at low concentrations. All of these experiments verified the prospective application of APA-CPNPs as a bone-targeting drug delivery system. Keywords: Bone targeting; Alendronate; Bone metastasis; Calcium phosphate nanoparticle

1. Introduction Because of the pecularity of the physiological environment, the skeleton is the most common organ to be affected by metastatic cancer (Breuksch et al., 2016; Mundy, 2002; Roodman, 2004; Suva et al., 2011). According to the 2011 statistic data by the American Academy of Orthopedic Surgeons (AAOS), about half of the 1.2 million cancer cases that occur every year come with bone metastasis (H. and MD., 2011). Among these cancers, breast and prostate cancers have the highest prevalence in bone metastasis (Miller et al., 2013). The clinical therapies for bone metastasis concentrate mainly on surgical resection, radiotherapy and chemotherapy (Chen et al., 2012). However, these therapies are not being fully utilized because of the uptake of chemotherapy agents in non-targeted sites and their related side effects (Clines and Guise, 2008). Therefore, designing a bone-targeting drug delivery system that can release drug exclusively in the nidus of the bone would be significant. Hydroxyapatite (Ca10(PO4)6(OH)2) is specific to the skeleton (Choi and Kim, 2007b; Uskokovic and Uskokovic, 2011). Bone diseases such as cancer metastasis result in the exposure of hydroxyapatite to the blood. The hydroxyapatite that is exposed around the 3

metastasis site can be regarded as an ideal target for drug carriers. Because of their unique affinity for the inorganic component hydroxyapatite, molecules such as tetracyclines, bisphosphonates, and acidic oligopeptides have been used as a targeted group to target bone tissue (Fu et al., 2014). When compared to acidic oligopeptides, Wang et al. found that bisphosphonates, with a hydroxyl group and two phosphate groups, can bind to all types of hydroxyapatite, while acidic oligopeptides binds preferentially to a higher crystalline hydroxyapatite (Wang et al., 2006). Meanwhile, bisphosphonates are widely used for their binding capability and for their profound effect on osteoporosis. Nitrogen-containing bisphosphonates, such as alendronate (ALN) and zoledronate, inhibit farnesyl pyrophosphate synthase activity and disrupt the mevalonic acid pathway. Osteoclasts lose their function and undergo apoptosis without the mevalonic acid pathway (Low and Kopeček, 2012). The antiosteoporosis property makes bisphosphonates important drugs for the clinical therapy of cancer metastasis-induced bone osteoporosis or facture. Additionally, bisphosphonates are in demand in the design of targeted drug carriers. Kiran et al. built zoledronate-anchored PLGAPEG nanoparticles with docetaxel inside. The PLGA-PEG-zoledronate NPs showed enhanced apoptotic activity and higher retention at the bone site (Ramanlal Chaudhari et al., 2012). Laura et al. synthesized poly(γ-benzyl-l-glutamate)-PEG6k-alendronate (PBLG10k-PEG6kALN), which can form nanoparticles spontaneously and has a strong affinity for hydroxyapatite (de Miguel et al., 2014b). For clinical applications, amorphous calcium phosphate sediments exhibit similar properties to hydroxyapatite. With good biocompatibility, biodegradability and osteoconductivity, it can stimulate tissue regeneration and has been widely used as a bone 4

repairing material in the areas of orthopedics and dentistry (Boskey, 1997; Zhao et al., 2011). Peter et al. found that the peri-implant bone density was increased after grafting zoledronate to a hydroxyapatite coating on titanium implants (Peter et al., 2005). Grover et al. similarly concluded that the presence of pyrophosphate in calcium phosphate cements appeared to stimulate the mineralization of the healing bone around the implant (Grover et al., 2013). Because of its excellent biocompatibility and unique physical properties, today it is used as drug carrier for DNA (Tobin et al., 2013; Zhang et al., 2010), siRNA (Xu et al., 2014), antibacterial drugs (Chen et al., 2014) and anti-cancer drugs (Iafisco et al., 2009; Mukesh et al., 2009). Inspired by these ideas, a novel bone-targeting nanoparticle with high biocompatibility and stability was designed and prepared in this study. We conjugated ALN to each side of polyethylene glycol 2000 (PEG2k) via an amide bond to obtain the novel hydrophilic material ALN-PEG2k-ALN. This material, with good biocompatibility (Pedraza et al., 2008), was designed to anchor on the surface of calcium phosphate nanoparticles in order to control the diameter (less than 40 nm) and give the nanoparticles the ability to target bone as well as prolong the circulation time of NPs in the vascular system. These ALN-PEG2k-ALN (APA)modified calcium phosphate (CP) nanoparticles (NPs) were designated APA-CPNPs. With ALN-PEG2k-ALN anchored on the surface, this nanoparticle showed obvious binding ability to bone tissue, which made it a promising drug delivery system. An in vitro release experiment showed that APA-CPNPs slowly release drug in normal physiological conditions while releasing drug quickly in cancer metastasis sites due to the acidic environment induced by cancer cells. The anticancer effect of APA-CPNPs was similar to free chemotherapeutics. 5

2. Experiment 2.1 Materials Polyethylene glycol 2000 (PEG2k, Mw=2000, BioUltra), hydroxyapatite (powder, Sigma), 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide (MTT) and 2morpholinoethanesulfonic acid (MES, BioUltra) were purchased from Sigma-Aldrich (St. Louis, USA). Alendronate (ALN) sodium was purchased from TCI (Tokyo, Japan). Methotrexate (MTX) was provided by Amresco, LLC (Solon, OH, USA). Succinic anhydride (SA, 99%), 4-dimethylaminopyridine (DMAP, 99%), triethylamine (TEA, 99%), 1-Ethyl-3(3-dimethylaminopropyl) carbodiimide hydrochloride (EDCI, 98%) and Nhydroxysuccinimide (NHS, 98%) were purchased from Aladdin (Shanghai, China). Sephadex® G25 was purchased from GE (Fairfield, USA). Dulbecco's Modified Eagle Medium (DMEM), pancreatic enzymes and newborn calf serum (NCS) were purchased from Gibco. Ultrapure water (18.2 Ω) was provided by PURELAB (High Wycombe, UK). MCF-7 cells and HeLa cells were purchased from ATCC (Manassas, VA). All other reagents were purchased as analytical reagent grade and used as received. Sprague–Dawley rats (200±5 g) and Kunming mice (20±2 g) were supplied by the Laboratory Animal Center of Sun Yat-sen University. All experimental procedures were approved and supervised by the Institutional Animal Care and Use Committee of Sun Yat-sen University.

2.2 Synthesis of ALN-PEG2k-ALN 6

2.2.1

COOH-PEG2k-COOH

COOH-PEG2k-COOH was synthesized by reacting PEG2k with succinic anhydride. Briefly, PEG2k (10 mmol, 20.66 g), succinic anhydride (25 mmol, 2.50 g), 4dimethylaminopyridine (20 mmol, 2.44 g) and triethylamine (20 mmol, 2.75 mL) were dissolved in anhydrous dioxane. The reaction was stirred at room temperature overnight under inert atmospheric conditions. Next, the mixture was dropped into CCl4 and filtered to isolate triethylamine hydrochloride from the product. After redundant CCl4 was removed, the mixture was dropped into cold diethyl ether. After being stored at -4 °C overnight, the precipitation of crude COOH-PEG2k-COOH was recovered by filtration and dried under vacuum. Two grams of crude COOH-PEG2k-COOH was placed into a NaHCO3 saturated solution and filtered. Upon cooling, the pH of the filtrate was adjusted to 1 using concentrated HCl. Extraction with CHCl3 (3 × 5 mL) was performed, and the white waxy solid of pure COOH-PEG2k-COOH was obtained by precipitation in cold diethyl ether. The structure was confirmed via NMR with the following parameters: COOH-PEG2k-COOH 1H NMR (400 MHz, CDCl3) δH: 4.214.24 ppm (m, 4H), 3.40-3.90 ppm (m, 176H), 2.60-2.62 ppm (m, 8H). 2.2.2

ALN-PEG2k-ALN

The synthesis of ALN-PEG2k-ALN was achieved based on a slightly modified carbodiimide chemistry approach (Chung et al., 2006; de Miguel et al., 2014b). Solution A: To pre-activate the carboxylic group and obtain the NHS-PEG2k-NHS solution, a mixture of COOH-PEG2k-COOH (0.063 mmol), EDCI (1.25 mmol) and NHS (1.25 mmol) was dissolved in 2-morpholinoethanesulfonic acid buffer (0.1 M, pH 5.5), and stirred first on an ice bath for 30 min and then at room temperature for another 30 min. 7

Solution B: ALN (300 mg) was dissolved in 1 mL of 2 M NaOH and 800 μL of H2O. Upon cooling, the pH was adjusted to 7-8 using 1 M HCl. Solution A was added slowly into solution B under vigorous stirring, and the pH of the reaction was maintained at 7-8 using 1 M NaOH. The mixture was stirred for 1 h on ice and the 24 h on a water bath (37 °C). Next, the water was evaporated in vacuo. Purification of ALN-PEG2k-ALN from free ALN was performed via size exclusion chromatography using Sephadex® G25. The PEG-containing fraction was identified using cobalt thiocyanate and was collected (Giger et al., 2013). The white cottony solid of pure ALN-PEG2k-ALN was obtained by lyophilization of the collected fractions. The structure was confirmed via NMR with the following parameters: ALN-PEG2k-ALN 1H NMR (400 MHz, D2O) δH: 4.31 ppm (m, 4H), 3.50-4.00 ppm (m, 176H), 1.70-2.10 ppm (m, 8H); 31P NMR (400 MHz, D2O) δP: 18.11 ppm.

2.3 Preparation and Characterization of APA-CPNPs 2.3.1

Preparation of APA-CPNPs

APA-CPNPs were obtained based on a slightly modified hydrothermal method (Lai et al., 2014). ALN-PEG2k-ALN and (NH4)2HPO4 were used as the source of organic phosphate ions and inorganic phosphate ions, respectively. Meanwhile, Ca(NO3)2·4H2O was used as the source of Ca2+ ions. The total Ca/P molar ratio was 1:1. Among all the phosphate ions, the percentage of organic phosphate ions derived from ALN-PEG2k-ALN ranged from 10% to 40%. The pH of both solutions was adjusted to 11 by aqueous ammonia. In order to prepare APA-CPNPs, the Ca2+ ion solution was added dropwise into the phosphate ion solution under 8

vigorous stirring at room temperature (26 °C), followed by ultrasonic treatment (40 kHZ) for 30 min. Then, the mixture was moved into a 25 mL autoclave with Teflon linen and heated for 3 h at 85 °C. The resulting solution was exposed to ultrasonic treatment (40 kHZ) for 30 min. For the preparation of non-targeted calcium phosphate nanoparticles (P-CPNPs), COOHPEG2k-COOH was used as an alternative material to ALN-PEG2k-ALN during the procedure. Except for the changing of material, the preparation was the same as the procedure described above. The APA-CPNPs solution was purified via centrifugation (16,000 rpm, 30 min, 10 °C) and washed with ultrapure water (18.2 Ω). This purification procedure was repeated three times to obtain pure APA-CPNPs. The nanoparticle precipitates were re-suspended with a certain volume of ultrapure water (18.2 Ω) by vigorous vortexing. In order to obtain the drugloaded APA-CPNPs or fluorescence-labeled APA-CPNPs, a certain amount of methotrexate (MTX) or calcein, respectively, was dissolved in the phosphate ion solution before mixing. The other steps were the same as the procedure described above. 2.3.2

Particle Size and Zeta Potential

The average particle size, distribution and zeta potential of the nanoparticles was obtained by dynamic light scattering (DLS) using a Malvern Zetasizer Nano ZS90 (Malvern Instruments, Malvern, UK). The temperature of the cell was kept constant at 25 °C. APACPNPs were diluted in distilled water and measured at least three times. The size results were given as a number distribution of the mean diameter with the standard deviation.

2.4 In Vitro Drug Release Assay 9

The drug release assay was carried out in glass bottles at 37 °C in phosphate buffer (0.01 M, pH = 7.4, 6.5) and acetate buffer (0.01 M, pH = 4.7) solutions. Before the experiment, the calcein loaded APA-CPNPs were re-suspended with ultrapure water (18.2 Ω). Firstly, 1 mL of the calcein loaded APA-CPNPs (10 mg/ml) were placed in a dialysis tube with a molecular weight cutoff of 10 kDa. Next, the dialysis tube was immersed in 45 mL of the buffer, which was preheated to 37 °C for approximately 30 min in a shaker, keeping the stirring (100 rpm) and temperature (37 °C) constant. At time points of 10 min, 20 min, 30 min,1 h, 2 h, 4 h, 6 h, 8 h, 12 h, and 24 h, 2 mL of release medium was withdrawn and an equal volume of fresh medium was replenished. The amount of released calcein was detected via fluorescence at 490 nm excitation and 515 nm emission. The release study was performed in triplicate. All the results were calculated according to the following formula:

𝑅𝑒𝑙𝑒𝑎𝑠𝑒 (%) = +

𝐶𝑠𝑎𝑚𝑝𝑙𝑒.(𝑛) × 45 × 100% 𝐶𝑁𝑃 × 1 ∑𝑛−1 𝑖=1 [(𝐶𝑠𝑎𝑚𝑝𝑙𝑒.(𝑛−1) × 45)⁄(𝐶𝑁𝑃 × 1)] × 2 45

1. Csample.(n) refers to the calcein concentration of the samples at each time point that was detected by the fluorospectro photometer. 2. CNP refers to the initial calcein concentration of calcein-loaded APA-CPNPs.

2.5 X-ray Photoelectron Spectroscopy To measure the atomic composition of the surface of the APA-CPNPs, the calcium phosphate sediment (prepared without the addition of ALN-PEG2k-ALN) and the hydroxyapatite powder, analysis was performed by X-ray photoelectron spectroscopy (Thermo Fisher Scientific, ESCALAB 250) using a monochromatized Al Kα radiation (hν = 10

1486.6 eV). The XPS BE value was corrected by measuring the reference peak of C1s (284.8 eV) from the surface contamination. The elements detected were observed from the survey spectrum over a range of 0 to 1100 eV with an energy resolution of 1.0 eV. The P2p region were curve fitted using the XPS peak fitting program (XPSPEAK, Version 4.1, 2000).

2.6 Isothermal Titration Calorimetry (ITC) The isothermal titration experiment was carried out to test the binding affinity between calcium ions and both ALN-PEG2k-ALN and APA-CPNPs. ITC was performed at 25 °C on a thermal activity monitor calorimeter (VP-ITC, Microcal Inc., USA). The experiment was designed according to Velazquez-Campoy et al. (Velazquez-Campoy and Freire, 2006) and de Miguel et al. (de Miguel et al., 2014b) with slight modifications. Prior to each measurement, each solution and suspension was degassed to remove air bubbles and was filtered with a 0.45 μm filtrate membrane. 1.4 mL of the sample solution, such as the ALN solution, ALN-PEG2kALN solution, COOH-PEG2k-COOH solution or the APA-CPNPs suspension (CALN=1.5 mM, CALN-PEG2k-ALN=0.75 mM, CCOOH-PEG2k-COOH =0.75 mM), was titrated by adding 30 × 5 μL of 30 mM CaCl2 solution in the same buffer. The first aliquot was used to correct volume errors on the first injection so that the corresponding heat was excluded from evaluation. The procedure was set as follows: duration of each injection (~10 s), spacing between injections (~200 s), and stirring speed (~307 rpm). Additionally, the interaction of an aqueous solution of ALN with calcium ions was performed and used as a reference. The data were analyzed using Origin 7.5 (OriginLab Co., US).

11

2.7 Bone Mineral Binding Ability In Vitro. The affinity of APA-CPNPs toward bone fragments (in vitro) was investigated in comparison with non-targeted P-CPNPs. Skull fragments were obtained from fetal rats. Firstly, fetal rats were disinfected with a 75% ethanol solution. Then, the skull fragments were removed via craniotomy. The harvested skull fragments were washed with normal saline three times. Secondly, the periost of the skull were removed under inverted microscope. Finally, the treated skulls were dried in an oven and collected in a desiccator. Calcein-loaded APACPNPs, which were stable in phosphate buffer (0.01 M, pH 7.4，37 °C), were incubated with the skull fragments (4 mm × 4 mm) of fetal mice with a stirring rate of 100 rpm (ZHWY200D shaker, Zhicheng Co. Shanghai). The binding of calcein-loaded APA-CPNPs were compared with P-CPNPs and free calcein solution by imaging under a fluorescent microscope. Bone fragments were incubated with nanoparticles for 10 min, 20 min, 30 min, 1 h, 2 h and 4 h. After incubation, the bone fragments were washed gently with PBS three times and air dried in the dark. Photography was taken by inverted fluorescent microscope (Nikon, Japan) with a settled exposure time of 60 ms.

2.8 Cell Assays 2.8.1

Cell Culture

The NIH/3T3 cell line, HeLa cell line and the MCF-7 cell line were purchased from the American Type Culture Collection (ATCC). Cells were cultured in DMEM supplemented with 10% NBCS and antibiotics (streptomycin 100 U/mL and penicillin 100 U/mL). Cells were grown at 37 °C, 5% CO2, 100 units penicillin and 100 mg/mL streptomycin (Life 12

Technologies). 2.8.2

Cell Viability Assays

The relative cytotoxicity of free MTX (10 mM stock solution in H2O), MTX loaded APACPNPs, ALN-PEG2k-ALN and blank APA-CPNPs against cells was measured using an MTT assay. Briefly, cells were seeded in 96-well plates at a density of 1.0 × 104 cells per well and incubated for 24 hours. Then, the cells were treated with serial concentrations of free MTX, MTX loaded APA-CPNPs, free ALN-PEG2k-ALN material or blank APA-CPNPs. After 48 hours of incubation, 10 μL of MTT solution (5 mg/mL) was added to each well and

incubated for another 4 h. Finally, the medium was replaced with 150 μL of dimethylsulfoxide (DMSO), and the optical density was determined with a microplate reader at a wavelength of 490 nm in triplicate.

2.9 In Vivo Biocompatibility Assay for APA-CPNPs 2.9.1

Blood Biochemical Index Study

Six female Sprague–Dawley rats (200±5 g) were randomly assigned to two groups for the investigation of the biochemical indexes. Tail veins of the rats in each of the two groups were injected with 1 mL of normal saline or APA-CPNPs (10 mg/mL in NS) at scheduled time points (1, 4, and 7 days after the first injection). At scheduled points in time (2 and 8 days), 0.4 mL of blood was collected from the periorbital vein without anticoagulant and allowed to sit at room temperature for 30-90 minutes. The samples were centrifuged using the 535-1 type centrifuge (3000 rpm,15 min,4 °C), and the supernatant was separated out. Alanine aminotransferase (ALT) and aspartate aminotransferase (AST), albumin (ALB), and 13

creatinine (CREA) levels were determined to evaluate liver function, nephrotoxicity and tissue damage or inflammation, respectively. The detection methods for the ALT and AST, ALB, and CREA are the rate method, the bromocresol green method and the sarcosine oxidase method, respectively. These biochemical indexes were all assayed by the automatic

biochemical analyzer (BeckmanCX5). 2.9.2

Tissue Biopsy Study

Six female Kunming mice (20±2 g) were randomly assigned to two groups. The tail veins of the mice in each of the two groups were injected with 0.1 mL of normal saline or APACPNPs (10 mg/mL in normal saline) at scheduled time points (1, 4, and 7 days after the first injection). At the second day after the first injection, a mouse from the normal saline and APA-CPNPs group was dissected and the heart, liver, spleen, lung and kidney were harvested, then fixed in 10% neutral formalin for more than 72 h. The samples were cut into small specimens and later transferred to the embedding box. The specimens were dehydrated using the Leica TP1020 dehydration machine, and then soaked in molten wax. The specimens were embedded with paraffin using the Leica EG1160 embedding machine. The specimens were sliced using the RM2155 Leica paraffin slicing machine. The slice thickness was approximately 3 micrometers. The slices were placed at 60 °C overnight, dewaxed with a transparent agent, hydrated with anhydrous ethanol and gradient alcohol, stained with hematoxylin and 1% eosin solution, dehydrated with gradient alcohol, made transparent with a transparent agent, and finally sealed with neutral balata. On the eighth day after the first injection, the remaining mice were processed according to the description above. The tissue biopsy analyses were performed through observation under a light microscope (Leica 14

DM5000B) where the tissue sections were blinded to the pathologist.

2.10 Statistical Analysis Statistical analyses were performed using SPSS Statistics for Windows, Software Version 13.0 (SPSS Inc., Chicago, IL, USA). A one-way analysis of variance (ANOVA) and the least significant difference test were employed to analyze the data.

3. Results and Discussion 3.1 Synthesis and Characterization of ALN-PEG2k-ALN The synthesis of ALN-PEG2k-ALN was performed in three steps: carboxylation of PEG2k, activation of COOH-PEG2k-COOH via EDCI and NHS, and finally binding of ALN to activated PEG polymer (see Scheme 1.). Polymers were characterized by 1H NMR, 31P NMR and FTIR. The 1H NMR spectra of COOH-PEG2k-COOH showed a significant signal approximately 4.31 ppm due to the carboxylation of PEG2k. The complete substitution of the hydroxyl groups was proved by 1H NMR (Figure 1C), showing that both side of the PEG2k were carboxylated. Initial attempts at completely grafting the ALN onto COOH-PEG2k-COOH via esterification reaction under a wide variety of conditions in aqueous conditions proved unsuccessful. However, it was found that an efficient linkage between ALN and the carboxyl could be achieved by varying the pH of the solutions of both the activation of COOH-PEG2kCOOH (pH = 5.5) and the binding of ALN to the activated PEG polymer (pH = 7-8). Additionally, controlling the reacting temperature of each step was as important. The 1H NMR 15

spectra of ALN-PEG2k-ALN show the specific signal of ALN approximately 1.9 ppm and the integration of the peak approximately 1.9 ppm and 4.3 ppm show that a majority of the – COOH were substituted. (Figure 1A). The 31P NMR spectra show the specific signal of P-C-P of ALN approximately 18.11 ppm (Figure 1B) (Kieczykowski et al., 1995). The FTIR spectra give proof for the successful esterification of COOH-PEG2k-COOH because of the appearance of the –C=O signal approximately 1730 cm-1(Figure 1D). And from Figure 1D we find that the –C=O signal was weak while the -NH-C=O (1640 cm-1) signal was strong. Thus, we can summarize that most of the –COOH on COOH-PEG2k-COOH participated in the amidation reaction with alendronate. In summary, the FTIR and NMR measurements demonstrated that the synthesis of ALNPEG2k-ALN was feasible.

3.2 Preparation and Characterization of APA-CPNPs The APA-CPNPs were formed by mixing a calcium solution (derived from Ca(NO3)2) and phosphate solution (derived from Na2HPO4 and ALN-PEG2k-ALN) in an alkaline condition (pH = 11). The molar ratio of Ca/P was 1:1. The phosphate ions derived from ALNPEG2k-ALN ranged from 10% to 40% of all the phosphate ions (10%, 20% and 40%). The corresponding nanoparticles were designated APA-CPNPs 10% to APA-CPNPs 40%. All of the added ALN-PEG2k-ALN could not take part in the forming of the calcium phosphate nanoparticles, however, they build a “self-restrained” environment for the nanoparticles to form. As the calcium and phosphate ions were mixed in an alkaline solution, they aggregated rapidly. The growing calcium phosphate nanoparticles were stabilized by the addition of 16

ALN-PEG2k-ALN because of the strong affinity between calcium phosphate sediment and ALN. The concentration of ALN-PEG2k-ALN did have an influence on particle size and polydispersity index (PDI). As shown in Table 1. The mean diameter ranged from 17 nm to 200 nm while the content of ALN-PEG2k-ALN ranged from 40% to 10%, as determined by DLS. The size and PDI of the particles covered with COOH-PEG2k-COOH reached 1.6 μm and 1.000, respectively. Lower ALN-PEG2k-ALN concentrations led to larger aggregates because particle growth was not efficiently inhibited. The size is an important parameter for the systemic administration of bone-targeting nanoparticles. It is known that the sinusoids in the bone marrow have pores of approximately 80-100 nm (Choi and Kim, 2007a; De Miguel et al., 2014a; Wang et al., 2003). As a result, we choose APA-CPNPs 20% to complete the remaining experiments due to their favorable particle size (33.617 ± 1.849 nm) and PDI (0.212 ± 0.006). The stability of APA-CPNPs were evaluated in PBS (pH = 7.4, 37 °C) at different time intervals by DLS. As shown in Figure 2B, we found that with the help of ALN-PEG2k-ALN, the size of APA-CPNPs remained relatively constant in a mimetic physiological environment over 1 week, confirming the physical stability of the system. From the zeta potential result (Figure 2A) we hypothesized that the ALN-PEG2k-ALN covered the surface of the nanoparticles. As shown in the picture, the zeta potential of the ALN-PEG2k-ALN-modified nanoparticles was about -25 mV, while the zeta potential of the COOH-PEG2k-COOH-modified nanoparticles was approximately 5 mV. That might be due to the absorption of the phosphate group of ALN-PEG2k-ALN to the surface of nanoparticle. Without the modification of ALN, the COOH-PEG2k-COOH could not absorb onto the 17

nanoparticle successfully, so the zeta potential was relatively higher. TEM images of APACPNPs 10% and APA-CPNPs 20% verified this hypothesis. As shown in Figure 2C, 2D, we obtained spherical nanoparticles. Compared with the XRD spectrum of hydroxyapatite, the results showed that the calcium phosphate sediment without adding ALN-PEG2k-ALN was identified as the hydroxyapatite crystal form. All characteristic peaks of the calcium phosphate sediment were consistent with previous data (Piccirillo et al., 2015). However, as shown in Figure 3, APA-CPNPs showed different peaks compared to the two samples above, which demonstrated that the crystal form of the calcium phosphate core was amorphous. This was an unexpected but interesting phenomenon to us. Since it has been discovered, hydroxyapatite has been the most documented material in surgical applications due to its biocompatibility, osteoconductivity and osseointegration (Ambard and Mueninghoff, 2006; Fu et al., 2004). It has not replaced autologous tissue as the surgical ‘gold-standard’ because of its dense form, which causes it to remain in the body for a prolonged period following implantation, and its fragility (Grover et al., 2013). However, it has been demonstrated that amorphous calcium phosphate has better biodegradability than hydroxyapatite in vivo (Zhao et al., 2011). This phenomenon, for us, is beneficial because APA-CPNPs, with good biodegradability, can release drug slowly in the bone marrow. And with part of the ALN dangling on the surface, the APA-CPNPs can bind to bone spontaneously. Additionally, we attributed the transformation of the crystal form to the changing of Ca/P when preparing the APA-CPNPs. The Ca/P ratio was important in the preparation of calcium phosphate sediment, which could lead to the transformation of the crystal form (Piccirillo et 18

al., 2015). The changing of Ca/P means that a portion of the phosphate ions might not participate in the forming of the calcium phosphate core. So we could hypothesize that a portion of the ALN-PEG2k-ALN were anchored on the surface by only one ALN molecule, with the other one dangling on the surface of nanoparticle, so that the phosphate ions that took part in the forming of the calcium phosphate core were reduced accordingly. Regarding the state of the ALN-PEG2k-ALN, there were two possibilities. On one hand, ALN-PEG2k-ALN could take part in the formation of the calcium phosphate core such that most of the material was immersed inside. On the other hand, only a portion of the ALN might be anchored on the surface of the calcium phosphate core with the PEG chains and ALN dangling outside. To determine which scenario was taking place, we investigated the elemental distribution of APA-CPNPs via X-ray photoelectron spectroscopy (XPS). XPS is a qualitative and quantitative surface-sensitive technique that can be used to evaluate the elemental and chemical composition of material surface. It has been applied previously in the surface characterization of calcium phosphates (Chusuei et al., 1999; Lu et al., 2000; Zeng and Lacefield, 2000). By taking survey scans for the surface of APA-CPNPs, calcium phosphate core and the hydroxyapatite powder, the surface atomic ratios of Ca/P and Ca/O, as well as the P2p peak, were investigated in this paper. A typical survey XPS spectra is shown in Figure 4A and the binding energy data were collected in Table 2. A small N1s peak was observed approximately 399.92 eV in the sample APA-CPNPs. This N1s peak, which was the same as the binding energy of the amide bond, indicated the appearance of ALN-PEG2kALN on the surface of the nanoparticles.

19

The Ca/P and Ca/O were calculated directly from the XPS data and are presented in Figure 4B. As shown in the picture, the chemical composition of the calcium phosphate core was similar to the hydroxyapatite powder. The Ca/P value of hydroxyapatite was 1.39, which agrees with the previous data (Lu et al., 2000). However, the Ca/P and Ca/O ratio of the APACPNPs were lower than those two groups, which means that ALN-PEG2k-ALN covered the surface of the calcium phosphate core. As we examined the P2p peak in more detail, we found that there was data that verified the discussion above. The P2p regions were curve fitted using commercial software (XPSPEAK, Version 4.1, 2000). Figure 4 presents the P2p spectra for hydroxyapatite (C), calcium phosphate sediment (D) and the APA-CPNPs (E). The fitting result of hydroxyapatite

was similar to the previous result that showed two peaks approximately 132.5 eV and 133.4 eV (Landis and Martin, 1984; Massaro et al., 2001; McLeod et al., 2006). The sediment group showed an analogous result. Comparing the APA-CPNPs group with the hydroxyapatite and sediment groups, we found that the integral area of the peak approximately 132.5 eV was larger, which was due to the phosphorous component of the bisphosphonate (McLeod et al., 2006), indicating that the P2p was composed primarily of bisphosphonate. All of the fitted values and the fitting area ratio are presented in Table 3. . From the XRD and XPS result, we concluded that only part of the ALN of ALN-PEG2kALN anchored on the surface of the calcium phosphate core with PEG chains and the rest of the ALN dangling outside.

20

3.4 In Vitro Drug Release Assay Next we determined the drug release properties of drug-loaded APA-CPNPs. In this experiment we used calcein as a model drug. Figure 5 showed the release profile of calcein from APA-CPNPs 20% in three buffer solutions with a pH of 4.7, 6.5 and 7.4 at 37 °C. At physiological conditions the cumulative release of drug was less than 10% within 30 min, while the combining rate of ALN to the bone was so high that approximately 37% of the ALN injected into the vein could bind to calcified tissues within 5 min and increase to 95% in 1 h (Porras et al., 1999). This phenomenon indicated that APA-CPNPs are a promising drug delivery system that could reduce the concentration of free chemotherapeutics in non-targeted sites. However, the drug release rate was improved with the reduction of the pH value. The amount of calcein released in acidic surroundings pH 6.5 and pH 4.7) was more than 50% and 90%, respectively, within 12 h. This is due to the the acid-sensitive feature of calcium phosphate, which contributes to the dissolution of drug in acidic environments (Kester et al., 2008; Uskokovic and Uskokovic, 2011). In the bone microenvironment, the resorbing osteoclasts secrete hydrogen ions to lower the pH to as low as 4.5-4.7 (Clarke, 2008; Silver et al., 1988). According to the result above, we considered that the APA-CPNPs would release drug rapidly as soon as it arrived at the resorbing site in the bone marrow.

3.5 In Vitro Bone Targeting of NPs It is well-documented that bisphosphonates, because of the phosphonate group and hydroxyl group, show a strong affinity to bone mineral tissue (Chen et al., 2012; Clementi et al., 2011; de Miguel et al., 2014b). We used isothermal titration calorimetry (ITC) to generate 21

thermodynamic data to assess the binding properties of ALN-PEG2k-ALN and APA-CPNPs to calcium ions. In this experiment, ALN and COOH-PEG2k-COOH were used as the positive and negative control, respectively. ALN-PEG2k-ALN and APA-CPNPs clearly showed a specific interaction with calcium ions, which was the same as free ALN molecules (Figure 6). However, little interaction between COOH-PEG2k-COOH and calcium ions existed. Those results verified the conclusion that ALN-PEG2k-ALN and APA-CPNPs did have the ability to target to bone fragments, similar to free ALN molecules. According to previous research, ALN can form complexes with Ca2+ with a ratio of 1:2 (Ca2+: ALN) because of the participation of the amino and phosphonate groups (Fernandez et al., 2003). ALN-PEG2k-ALN, with an amino link to the PEG chain, can form complexes with Ca2+ with a ratio of 2:1 (Ca2+: ALN-PEG2k-ALN). The fitted data were collected in Table 4, where we found that the thermodynamic parameters of ALN-PEG2k-ALN and ALN molecules were similar. These data demonstrated a similar interaction between ALN-PEG2k-ALN and calcium ions as ALN.

To verify the bone targeting ability of APA-CPNPs convincingly, we used mouse bone fragments (mouse skull bone with the periosteum stripped) with fluorescently labeled APACPNPs and observed the samples using fluorescence microscopy. As shown in Figure 7, the binding ability of APA-CPNPs to bone fragments rose significantly as the content of ALNPEG2k-ALN in NPs was increased from 20% to 40%. P-CPNPs refers to nanoparticles prepared with COOH-PEG2k-COOH. Without ALN spreading on the surface, the P-CPNPs did not absorb to the mouse skull bone surface. Thus, those results highlight the potential of 22

our engineered nanoparticles for targeting bone.

3.6 In Vitro Biocompatibility of ALN-PEG2k-ALN Material and APA-CPNPs The cytotoxicity of blank material and APA-CPNPs was assessed in vitro by MTT against NIH/3T3 cells, HeLa cells and MCF-7 cells. Figure 8A gave the relative cell viability after 48 h of incubation with the ALN-PEG2k-ALN material and a blank APA-CPNP solution in a series of concentrations. The results illustrated that both of the ALN-PEG2k-ALN material and blank APA-CPNPs did not show significant cytotoxicity to those cells even when the concentrations were as high as 6.25 mg/mL (ALN-PEG2k-ALN) and 5.00 mg/mL (blank APACPNPs), which is much higher than the maximum therapeutic concentration. This verified the potential for APA-CPNPs to be a good drug delivery system.

3.7 In Vivo Biocompatibility of APA-CPNPs To verify the biocompatibility of APA-CPNPs, we detected the blood biochemical indexes (ALT: alanine aminotransferase, AST: aspartate aminotransferase, ALB: albumin, CREA: creatinine) and performed tissue biopsies after dosing with APA-CPNP solution (10 mg/mL in normal saline) continuously for 7 days via tail vein injection. From Figure 8B we demonstrated that the APA-CPNP solution, which did not have an influence on the blood biochemical indexes, possesses favorable biocompatibility with rat liver and kidney. This assumption was certified by the following result. As shown in Figure 8C, no hyperemia, edema, hyperplasia, or atrophy were observed in the heart, liver, spleen, lung and kidney of the APA-CPNP treatment group; in addition, the tissues showed well-organized cell structure 23

and did not exhibit any morphological changes when compared with the control group. These results all indicated that the APA-CPNPs showed a great biocompatibility in animals.

3.8 In Vitro Anticancer Assay The MTT assay was used to evaluate the anti-proliferative and cytotoxic effects of MTXloaded APA-CPNPs using the human HeLa and MCF-7 cancer cell lines in vitro. The concentration of MTX ranged from 0 to 1 μg/mL. The results of HeLa cell relative viability are shown in Figure 9A. The result showed cellular inhibition when the loaded MTX concentration was 64 ng/mL, which was the same as the free MTX group. The results of the MCF-7 cell viability assay are shown in Figure 9B. Those results revealed that MTX-loaded APA-CPNPs have the ability to inhibit cell growth similarly to free drug at high concentrations while being slightly weaker at low concentrations. The lower cancer cell proliferation inhibition of MTX-loaded nanoparticles may be due to the time-consuming drug release from the nanoparticles and the delayed cellular uptake because of the strong polar surface of APA-CPNPs. All of these results showed that APA-CPNPs will not disrupt the anticancer effect of MTX.

4. Conclusions A new material was designed and synthesized in the current study, which bound to a calcium phosphate core and formed nanoparticles (APA-CPNPs) with well-controlled particle size (approximately 30 nm), polydispersity index and bone-targeting ability. The results showed that this nanoparticle could release chemotherapeutics in an acidic environment more 24

rapidly, while remaining relatively stable in a physiological environment. The MTX loaded APA-CPNPs show cellular inhibition of cancer cells similar to free drug. Additionally, the APA-CPNPs showed a favorable biocompatibility in animals. These data all verified the prospective application of APA-CPNPs as a drug delivery system for bone cancer therapy. Acknowledgment We are grateful for the financial support provided by the National Natural Science Foundation of China (grant numbers 81302715/H3008).

25

References Ambard, A.J., Mueninghoff, L., 2006. Calcium phosphate cement: review of mechanical and biological properties. Journal of prosthodontics : official journal of the American College of Prosthodontists 15, 321-328. Boskey, A.L., 1997. Amorphous calcium phosphate: the contention of bone. Journal of Dental Research 76, 1433-1436. Breuksch, I., Weinert, M., Brenner, W., 2016. The role of extracellular calcium in bone metastasis. Journal of Bone Oncology. Chen, C., Weir, M.D., Cheng, L., Lin, N.J., Lin-Gibson, S., Chow, L.C., Zhou, X., Xu, H.H.K., 2014. Antibacterial activity and ion release of bonding agent containing amorphous calcium phosphate nanoparticles. Dental Materials 30, 891-901. Chen, H., Li, G., Chi, H., Wang, D., Tu, C., Pan, L., Zhu, L., Qiu, F., Guo, F., Zhu, X., 2012. Alendronateconjugated amphiphilic hyperbranched polymer based on Boltorn H40 and poly(ethylene glycol) for bone-targeted drug delivery. Bioconjugate chemistry 23, 1915-1924. Choi, S.-W., Kim, J.-H., 2007a. Design of surface-modified poly (D, L-lactide-co-glycolide) nanoparticles for targeted drug delivery to bone. Journal of Controlled Release 122, 24-30. Choi, S.W., Kim, J.H., 2007b. Design of surface-modified poly(D,L-lactide-co-glycolide) nanoparticles for targeted drug delivery to bone. Journal of controlled release : official journal of the Controlled Release Society 122, 24-30. Chung, H.J., Kim, H.K., Yoon, J.J., Park, T.G., 2006. Heparin immobilized porous PLGA microspheres for angiogenic growth factor delivery. Pharmaceutical research 23, 1835-1841. Chusuei, C.C., Goodman, D.W., Stipdonk, M.J., Van, Justes, D.R., Schweikert, E.A., 1999. Calcium Phosphate Phase Identification Using XPS and Time-of-Flight Cluster SIMS. Analytical Chemistry 71, 149-153. Clarke, B., 2008. Normal bone anatomy and physiology. Clinical journal of the American Society of Nephrology : CJASN 3 Suppl 3, S131-139. Clementi, C., Miller, K., Mero, A., Satchi-Fainaro, R., Pasut, G., 2011. Dendritic poly(ethylene glycol) bearing paclitaxel and alendronate for targeting bone neoplasms. Molecular Pharmaceutics 8, 10631072. Clines, G.A., Guise, T.A., 2008. Molecular mechanisms and treatment of bone metastasis. Expert Reviews in Molecular Medicine 10, null-null. De Miguel, L., Noiray, M., Surpateanu, G., Iorga, B.I., Ponchel, G., 2014a. Poly (γ-benzyl-l-glutamate)PEG-alendronate multivalent nanoparticles for bone targeting. International journal of pharmaceutics 460, 73-82. de Miguel, L., Noiray, M., Surpateanu, G., Iorga, B.I., Ponchel, G., 2014b. Poly(γ-benzyl-l-glutamate)PEG-alendronate multivalent nanoparticles for bone targeting. International journal of pharmaceutics 460, 73-82. Fernandez, D., Vega, D., Goeta, A., 2003. Alendronate zwitterions bind to calcium cations arranged in columns. Acta Crystallographica Section C 59, m543-m545. Fu, Q., Zhou, N., Huang, W., Wang, D., Zhang, L., Li, H., 2004. Preparation and characterization of a novel bioactive bone cement: glass based nanoscale hydroxyapatite bone cement. Journal of materials science. Materials in medicine 15, 1333-1338. Fu, Y.C., Fu, T.F., Wang, H.J., Lin, C.W., Lee, G.H., Wu, S.C., Wang, C.K., 2014. Aspartic acid-based

26

Accepted Manuscript Title: Calcium phosphate nanoparticles functionalized with alendronate-conjugated polyethylene glycol (PEG) for the treatment of bone metastasis Author: Weijing Chu Yanjuan Huang Chanzhen Yang Yunhui Liao Xuefei Zhang Mina Yan Shengmiao Cui Chunshun Zhao PII: DOI: Reference:

S0378-5173(16)31114-0 http://dx.doi.org/doi:10.1016/j.ijpharm.2016.11.051 IJP 16255

To appear in:

International Journal of Pharmaceutics

Received date: Revised date: Accepted date:

14-9-2016 13-11-2016 22-11-2016

Please cite this article as: Chu, Weijing, Huang, Yanjuan, Yang, Chanzhen, Liao, Yunhui, Zhang, Xuefei, Yan, Mina, Cui, Shengmiao, Zhao, Chunshun, Calcium phosphate nanoparticles functionalized with alendronate-conjugated polyethylene glycol (PEG) for the treatment of bone metastasis.International Journal of Pharmaceutics http://dx.doi.org/10.1016/j.ijpharm.2016.11.051 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

of particle size and DNA condensation salt for calcium phosphate nanoparticle transfection. Biomaterials 29, 3384-3392. Peter, B., Pioletti, D.P., Laïb, S., Bujoli, B., Pilet, P., Janvier, P., Guicheux, J., Zambelli, P.Y., Bouler, J.M., Gauthier, O., 2005. Calcium phosphate drug delivery system: influence of local zoledronate release on bone implant osteointegration. Bone 36, 52-60. Piccirillo, C., Pullar, R.C., Costa, E., Santos-Silva, A., Pintado, M.M., Castro, P.M., 2015. Hydroxyapatitebased materials of marine origin: a bioactivity and sintering study. Materials science & engineering. C, Materials for biological applications 51, 309-315. Porras, A.G., Holland, S.D., Gertz, B.J., 1999. Pharmacokinetics of Alendronate. Clinical Pharmacokinetics 36, 315-328. Ramanlal Chaudhari, K., Kumar, A., Megraj Khandelwal, V.K., Ukawala, M., Manjappa, A.S., Mishra, A.K., Monkkonen, J., Ramachandra Murthy, R.S., 2012. Bone metastasis targeting: a novel approach to reach bone using Zoledronate anchored PLGA nanoparticle as carrier system loaded with Docetaxel. Journal of controlled release : official journal of the Controlled Release Society 158, 470-478. Roodman, G.D., 2004. Mechanisms of Bone Metastasis. New England Journal of Medicine 350, 16551664. Silver, I.A., Murrills, R.J., Etherington, D.J., 1988. Microelectrode studies on the acid microenvironment beneath adherent macrophages and osteoclasts. Experimental Cell Research 175, 266–276. Suva, L.J., Washam, C., Nicholas, R.W., Griffin, R.J., 2011. Bone metastasis: mechanisms and therapeutic opportunities. Nature reviews. Endocrinology 7, 208-218. Tobin, L.A., Xie, Y., Tsokos, M., Chung, S.I., Merz, A.A., Arnold, M.A., Li, G., Malech, H.L., Kwong, K.F., 2013. Pegylated siRNA-loaded calcium phosphate nanoparticle-driven amplification of cancer cell internalization in vivo. Biomaterials 34, 2980-2990. Uskokovic, V., Uskokovic, D.P., 2011. Nanosized hydroxyapatite and other calcium phosphates: chemistry of formation and application as drug and gene delivery agents. Journal of biomedical materials research. Part B, Applied biomaterials 96, 152-191. Velazquez-Campoy, A., Freire, E., 2006. Isothermal titration calorimetry to determine association constants for high-affinity ligands. Nature protocols 1, 186-191. Wang, D., Miller, S., Sima, M., Kopečková, P., Kopeček, J., 2003. Synthesis and Evaluation of WaterSoluble Polymeric Bone-Targeted Drug Delivery Systems. Bioconjugate chemistry 14, 853-859. Wang, D., Sima, M., Mosley, R.L., Davda, J.P., Tietze, N., Miller, S.C., Gwilt, P.R., Kopečková, P., Kopeček, J., 2006. Pharmacokinetic and Biodistribution Studies of a Bone-Targeting Drug Delivery System Based on N-(2-Hydroxypropyl)methacrylamide Copolymers. Molecular Pharmaceutics 3, 717-725. Xu, Z., Wang, Y., Zhang, L., Huang, L., 2014. Nanoparticle-Delivered Transforming Growth Factor-β siRNA Enhances Vaccination against Advanced Melanoma by Modifying Tumor Microenvironment. ACS Nano 8, 3636-3645. Zeng, H., ., Lacefield, W.R., 2000. XPS, EDX and FTIR analysis of pulsed laser deposited calcium phosphate bioceramic coatings: the effects of various process parameters. Biomaterials 21, 23–30. Zhang, X., Kovtun, A., Mendoza-Palomares, C., Oulad-Abdelghani, M., Fioretti, F., Rinckenbach, S., Mainard, D., Epple, M., Benkirane-Jessel, N., 2010. SiRNA-loaded multi-shell nanoparticles incorporated into a multilayered film as a reservoir for gene silencing. Biomaterials 31, 6013-6018. Zhao, J., Liu, Y., Sun, W.B., Zhang, H., 2011. Amorphous calcium phosphate and its application in dentistry. Chemistry Central journal 5, 40.

28

Figure captions Figure 1. 1H NMR spectra of (A) ALN-PEG2k-ALN, (B) COOH-PEG2k-COOH, (C) 31P NMR spectra of ALN-PEG2k-ALN, (D) FTIR spectra of PEG2k, COOH-PEG2k-COOH and ALNPEG2k-ALN.

Figure 2. (A) Zeta Potential plot of APA-CPNPs and P-CPNPs. (B) The stability of APACPNPs in PBS (0.1M, pH 7.4, 37 °C) determined via DLS. TEM images of nanoparticles for (C) APA-CPNPs 10% (×30,000), (D) APA-CPNPs 20%, (×60,000).

Figure 3. PXRD spectra of APA-CPNPs, Calcium phosphate sediment and Hydroxyapatite.

Figure 4. (A) XPS spectra of calcium phosphate sediment and APA-CPNPs. (B) Surface chemical composition of hydroxyapatite, calcium phosphate sediment and APA-CPNPs. And XPS P2p spectra in (C) hydroxyapatite, (D) calcium phosphate sediment and (E) APACPNPs.

Figure 5. Release profiles of APA-CPNPs 20% in three kinds of buffers (pH 4.7, pH 6.5 and pH 7.4, n=3).

Figure 6. ITC integrated heat data profiles obtained from the binding interaction of CaCl2 with: (a) (up-left) ALN solution; (b) (up-right) COOH-PEG2k-COOH solution (c) (down-left) 29

ALN-PEG2k-ALN solution (d) (down-right) APA-CPNPs solution.

Figure 7. Bone targeting ability of APA-CPNPs: fluorescence image of bone fragments after incubation with calcein labeled APA-CPNPs (20% and 40%) and P-CPNPs. Images taken by inverted fluorescence microscope.

Figure 8. Cell viability determined by MTT. (A) Effect of ALN-PEG2k-ALN and blank APACPNPs on the viability of 3T3/NIH cells, Hela cells and MCF-7 cells. The data represent the mean ± SD of three wells and are representative of three independent experiments. (B)Blood biochemical indexes determined after I.V. injection of PBS and blank APA-CPNPs solution (10 mg/mL in NS) for 1 day (blue) and 7 days (red) respectively. The data represent the mean ±SD of three animals. (C)Tissue biopsies of heart, liver, spleen, lung and kidney detected after I.V. injection of PBS and blank APA-CPNPs solution (10 mg/mL in NS) for 1 day (1D) and 7 days (7D) respectively. Figure 9. Cell viability determined by MTT. (A) Effect of MTX loaded APA-CPNPs on the viability of HeLa cells, (B) effect of MTX loaded APA-CPNPs on the viability of MCF-7 cells. The data represent the mean ± SD of three wells and are representative of three independent experiments.

Scheme 1. Overall synthesis steps used to prepare COOH-PEG2k-COOH and ALN-PEG2kALN bone targeting material

30

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Scheme 1

Table 1. Average sizes, PDI of the five types of APA-CPNPs and P-CPNPs. Material Name

Ca/P

Material

Diameter (nm)

PDI

10%

216.067 ± 25.581

0.258 ± 0.121

20%

33.617 ± 1.849

0.212 ± 0.006

40%

17.203 ± 1.378

0.243 ± 0.079

40%

1612.67 ± 230.81

1.000 ± 0.000

Perc. APA-CPNPs

ALN-PEG2k1.00

10%

ALN

APA-CPNPs

ALN-PEG2k1.00

20%

ALN

APA-CPNPs

ALN-PEG2k1.00

40%

ALN COOH-PEG2k-

P-CPNPs

1.00 COOH

* Data were presented as Mean ± SD (n=3)

Table 2. The binding energy data of elements deposit on the surface of calcium phosphate sediment and APA-CPNPs.

Name

C1s

O1s

N1s

P2p

Ca2p

hydroxyapatite Peak BE (eV)

284.8

530.89

\

133.01

346.99

Sediment

Peak BE (eV)

284.8

530.85

\

132.92

346.92

APA-CPNPs

Peak BE (eV)

284.79

530.61

399.92

132.66

346.8

\ : Not detected.

31

Table 3. Atomic composition of hydroxyapatite, calcium phosphate sediment and APA-CPNPs calculated from the XPS spectra.

Sample

Peak

Position (eV)

1

133.47

Peak 1 : Peak 2

1.634

hydroxyapatite 2

132.63

1

133.35 1.707

Sediment 2

132.51

1

133.37

APA-CPNPs

0.600 2

132.40

Table 4. The binding parameters for ALN and ALN-PEG2k-ALN with calcium ions.

a

Alendronate

ALN-PEG2k-ALN

Chi^2/DoF

20.45

62.91

N (Ca2+/compound)

0.500 ± 0.00 a

2.00 ± 0.00 a

Kb (L.mol-1)

2.35×103 ± 138

1.97×103 ± 165

ΔH (cal.mol-1)

352.2 ± 5.953

328.4 ± 7.368

ΔS (cal.mol-1.K-1)

16.6

16.2

ΔG (Kj.mol-1)

-4.597×103

-4.501×103

Imposed stoichiometry for calculations.

32