In vivo β-catenin attenuation by the integrin α5-targeting nano-delivery strategy suppresses triple negative breast cancer stemness and metastasis

Accepted Manuscript In vivo β-catenin attenuation by the integrin α5-targeting nano-delivery strategy suppresses triple negative breast cancer stemness and metastasis Yunfei Li, Yajuan Xiao, Hsuan-Pei Lin, Derek Reichel, Younsoo Bae, Eun Y. Lee, Yiguo Jiang, Xuefei Huang, Chengfeng Yang, Zhishan Wang PII:

S0142-9612(18)30733-6

DOI:

10.1016/j.biomaterials.2018.10.019

Reference:

JBMT 18936

To appear in:

Biomaterials

Received Date: 7 September 2018 Revised Date:

16 October 2018

Accepted Date: 17 October 2018

Please cite this article as: Li Y, Xiao Y, Lin H-P, Reichel D, Bae Y, Lee EY, Jiang Y, Huang X, Yang C, Wang Z, In vivo β-catenin attenuation by the integrin α5-targeting nano-delivery strategy suppresses triple negative breast cancer stemness and metastasis, Biomaterials (2018), doi: https://doi.org/10.1016/ j.biomaterials.2018.10.019. 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.

ACCEPTED MANUSCRIPT In vivo β-Catenin Attenuation by the Integrin α5-targeting Nano-delivery Strategy Suppresses Triple Negative Breast Cancer Stemness and Metastasis Yunfei Li a, Yajuan Xiao a, Hsuan-Pei Lin a, Derek Reichel b, Younsoo Bae b, Eun Y. Lee c, Yiguo

a

RI PT

Jiang d, Xuefei Huang e, Chengfeng Yang a,* and Zhishan Wang a,*

Department of Toxicology and Cancer Biology, and Center for Research on Environment Disease,

b

SC

College of Medicine, University of Kentucky, Lexington, Kentucky 40536, United States

Department of Pharmaceutical Sciences, College of Pharmacy, University of Kentucky, Lexington,

c

Department of Pathology and Laboratory Medicine, College of Medicine, University of Kentucky,

Lexington, Kentucky 40536, United States d

M AN U

Kentucky 40536, United States

Institute for Chemical Carcinogenesis, State Key Laboratory of Respiratory Diseases, Guangzhou

e

TE D

Medical University, Guangzhou, Guangdong, People’s Republic of China Departments of Chemistry and Biomedical Engineering, Institute for Quantitative Health Science

EP

and Engineering, Michigan State University, East Lansing, Michigan 48824, United States * Corresponding Authors: Dr. Chengfeng Yang, E-mail: [email protected]

AC C

Dr. Zhishan Wang, E-mail: [email protected]

Acknowledgements: This study was supported in part by a Research Scholar Grant (RGS-15-026-01-CSM) from the American Cancer Society to C.Y. and a research grant from Elsa U. Pardee Foundation to Z.W. This research was also supported by the Shared Animal Imaging and Histology Resources of the University of Kentucky Markey Cancer Center (P30CA177558).

1

ACCEPTED MANUSCRIPT Abstract Cancer stem cells (CSCs) play pivotal roles in cancer metastasis, and strategies targeting cancer stemness may greatly reduce cancer metastasis and improve patients’ survival. The canonical

RI PT

Wnt/β-catenin pathway plays critical roles in CSC generation and maintenance as well as in normal stem cells. Non-specifically suppressing the Wnt/β-catenin pathway for cancer therapy could be deleterious to normal cells. To achieve specific β-catenin attenuation in cancer cells, we report an

SC

integrin α5 (ITGA5)-targeting nanoparticle for treating metastatic triple negative breast cancer (TNBC).

M AN U

We found that ITGA5 is highly expressed in strongly migratory and invasive TNBC cells as well as their lung metastatic foci, which rationalizes active-targeted drug delivery to TNBC cells via ITGA5 ligands such as a commercialized ligand-RGD motif (Arg-Gly-Asp). We modified lipid-polymer hybrid (LPH) nanoparticle for TNBC-targeted delivery of diacidic norcantharidin (NCTD), a potent

TE D

anti-cancer compound but with short half-life. Notably, in vivo imaging analysis showed that RGD-decorated LPH (RGD-LPH) accumulated more significantly and remained much longer than LPH in nude mouse orthotopic mammary TNBC tumor and lung metastatic tumor, which implicated

EP

the feasibility of ITGA5-targeting strategy for treating metastatic TNBC. Moreover, systemic

AC C

administration of NCTD-loaded RGD-LPH (RGD-LPH-NCTD) reduced nude mouse orthotopic mammary TNBC tumor growth and metastasis more effectively than free NCTD and LPH-NCTD via down-regulating β-catenin. These findings suggest that ITGA5-targeting nanoparticles may provide a facil and unique strategy of specially attenuating β-catenin in vivo for treating metastatic TNBC.

Keywords: Triple negative breast cancer (TNBC), metastasis, cancer stem cell (CSC), β-catenin, integrin α5 (ITGA5), NCTD, lipid-polymer hybrid (LPH) nanoparticle 2

ACCEPTED MANUSCRIPT 1. Introduction Breast cancer is the second leading cause of cancer death among women in the United States, and the high mortality is often attributed to triple negative breast cancer (TNBC), a breast cancer

RI PT

subtype that does not express receptors for estrogen and progesterone and lacks the amplification of human epidermal growth factor receptor 2. TNBC accounts for 15~20% of newly diagnosed breast cancer and shows more aggressive metastasis and worse prognosis than other breast cancer subtypes

SC

[1, 2]. Indeed, the worst survival among all breast cancer patients with metastatic disease was seen in

M AN U

TNBC with metastasis cases with a median survival of about 7 months[3, 4]. Unfortunately, the only currently approved treatment option for TNBC patients is chemotherapy, which is often ineffective and causes serious side effects. No FDA-approved targeted chemotherapy is clinically available for TNBC, representing an unmet, critical clinical need.

TE D

Over the past decades, one of the critical goals in cancer therapy is to eradicate cancer stem cell (CSC) population but not damage the endogenous somatic stem cells (SSCs). β-Catenin is a key mediator of the canonical Wnt pathway, a signaling pathway that plays crucial roles in development

EP

and homeostasis of adult tissue by regulating embryonic and tissue stem cell functions[5, 6].

AC C

Importantly, the Wnt/β-catenin pathway is also critically implicated in the generation and maintenance of CSCs that play pivotal roles in cancer initiation, metastasis, relapse and therapeutic resistance [5-7]. In particular, the Wnt/β-catenin pathway is highly activated and enriched in TNBC contributing to TNBC initiation and metastasis [8-10], representing a valid target for developing TNBC targeted therapy [11, 12]. However, targeting of the Wnt/β-catenin pathway for cancer therapy remains challenging due to a lack of strategies for efficient and specific delivery of drugs targeting the Wnt/β-catenin pathway to TNBC tumor cells, as non-specific inhibition of the Wnt/β-catenin 3

ACCEPTED MANUSCRIPT pathway can cause devastating effects on embryonic development and adult tissue homeostasis [13, 14]. Indeed, because of the safety profiles, although more than 25 drugs/biologics candidates demonstrated anti-Wnt signaling potentials, only one drug (PRI-724) entered phase 2 clinical trial

RI PT

[15, 16]. Since free small-molecules/biologics are likely to have similar distribution problems in vivo, an alternative strategy is desperately needed to provide the desired targeting/specificities. The targeting nanotechnology may provide a unique approach. For nanoparticle-mediated β-catenin

SC

attenuation that specifically targets TNBC stemness, one of the desired strategies is the use of

M AN U

distinct membrane molecules that are highly expressed in TNBC cells. However, there have been few reports concerning on the unique molecular marks of TNBC cells yet.

Norcantharidin as well as its diacid form (hereinafter referred as NCTD have strong inhibitory effects against a broad range of cancer cells including TNBC cells [17-19]. Recent in vitro studies

TE D

showed that NCTD could impair the stemness of pancreatic and other cancer cells probably by repressing the β-catenin pathway through its potent inhibitory effect on protein phosphatases [19, 20]. Unfortunately, NCTD have been reportedly hindered for cancer treatment by its short half-life (0.26

EP

h) in blood [21]. This issue could be solved by nanoparticle-mediated drug delivery strategy.

AC C

Compared to traditional hydrophobic drugs, nanoparticle encapsulation of low molecular weight (M.W.) hydrophilic drugs has made a very limited progress due to drugs’ fast partitioning to the external aqueous phase [22-24]. Four ways to prepare drug-loaded nanoparticles are emulsification solvent diffusion, emulsification solvent evaporation, nanoprecipitation and the double emulsion, and in most cases double emulsion method was the only option for highly-hydrophilic drugs’ encapsulation [24]. Although double-emulsion based techniques is extensively utilized for encapsulating hydrophilic drugs [22-25], stability issue is an intrinsic problem in the formulating 4

ACCEPTED MANUSCRIPT process. Particularly, susceptibility to emulsion coalescence restricts the removal of free hydrophilic drugs in the external aqueous phase by ultra-centrifugation or column chromatography methods [25]. Furthermore, few possibilities for ligand conjugation at the surface of emulsion droplets and

RI PT

micrometer-sized particles also hinder its extensive application for encapsulating low M.W. hydrophilic drugs.

In this study, a lipid-polymer hybrid (LPH) nanoparticle was chosen as our fundamental

SC

delivery vehicles, which was recently reported to overcome many intrinsic drawbacks of both

M AN U

conventional liposome and polymer nanoparticles such as stability or biocompatibility in vivo [26, 27]. Meanwhile, LPH with intermediate rigidity displays superior permeability against tumor cells and multiple interstitial and tumor barriers to conventional liposomes or polymer nanoparticles [28]. In addition to these advantages, the surface of LPH could be decorated with targeting ligands [26,

TE D

27], which fulfills our special need to effectively deliver therapeutics to desired sites. We found that expression level of a cellular membrane protein-integrin α5 (ITGA5) is significantly high in TNBC cells. Aiming to investigate the feasibility of ITGA5-targeting strategy to attenuate β-catenin in

EP

TNBC treatment, a commercialized RGD (Arg-Gly-Asp) motif, reportedly able to recognize and

AC C

bind ITGA5 [29, 30], was adopted as a targeting ligand of LPH. The generated RGD-LPH-NCTD showed high and long-term accumulation in both primary TNBC tumor and its lung metastatic foci, and thus drastically reduced spontaneous lung metastasis of TNBC via β-catenin attenuation.

2. Materials and Method 2.1. Cell lines and Chemicals Human TNBC cell lines MDA-MB-231, LM2 and SUM159 cells were obtained and cultured as 5

ACCEPTED MANUSCRIPT described in our recent publications [31, 32]. The LM2 cell is a derivative of MDA-MB-231 cells, which was selected for its strong ability to metastasize to lung in vivo [33]. β-Catenin stably overexpressing cells were generated using lentiviral expression (pLenti6.3) technique as described in

technologies

inc.

(Huntsville,

AL)

and

cloned

RI PT

our recent publications [31, 32]. Human β-catenin full length cDNA was obtained from transOMIC into

pLenti6.3.

(1,2-distearoyl-sn-glycero-3-phophoethanolamine-N-carboxy(polyethylene from

NOF

corporation

(Tokyo,

glycol)2000)

Japan).

SC

purchased

PEG-DSPE was

Mal-PEG-DSPE

M AN U

(1,2-distearoyl-sn-glycero-3-phophoethanolamine-N-carboxy(polyethylene glycol)2000-maleimide) was purchased from Jenkem Technology USA Inc. (Allen, TX). Thiazolyl Blue Tetrazolium Bromide (MTT), norcantharidin, poly(D,L-lactide-co-glycolide) (PLGA, easter terminated, lactide: glycolide 75: 25, M.W. 76000~115000) was purchased from Sigma-Aldrich (St Louis, MO). Lecithin was

TE D

purchased from Alfa Aesar (Ward Hill, MA). RGD peptide with a terminal cysteine (c(RGDfC), M.W. 578.65, RGD-SH) was purchased from Peptides International, Inc. (Louisville, KY). The fluorescence dye Cy5, Cy7.5 and Cy7.5 amine were from Lumiprobe Corporation (Hunt Valley, MD).

EP

Fluorescence and bioluminescence of lung metastases were quantitatively measured using IVIS

AC C

Spectrum live animal imaging system (PerkinElmer Co., Waltham, MA).

2.2. Preparation and characterization of LPH-NCTD and RGD-LPH-NCTD RGD-PEG-DSPE was synthesized as previously described [34]. Briefly, Mal-PEG-DSPE and RGD-SH (feed molar ratio 1:1.5) was dissolved in 0.05M HEPES/0.01 M EDTA aqueous solution. The mixed solution was then rotated overnight in darkness at room temperature, followed by ultrafiltration to remove the excessive free RGD-SH peptide. Next, the RGD-LPH was fabricated 6

ACCEPTED MANUSCRIPT from PLGA, lecithin, PEG-DSPE and RGD-PEG-DSPE using a modified nanoprecipitation method. Lecithin (2mg), PEG-DSPE (18mg) and RGD-PEG-DSPE (2mg) were dissolved in 4% ethanol aqueous solution and heated to 65

. NCTD (1.25 mg) was mixed with PEI10K (375µg) in aqueous to get NCTD/PEI10K bundles, then followed by mixing with

RI PT

solution and incubated 5 min at 65

PLGA (1.875 mg) in 600 µL acetone solution. The NCTD/PEI10K/PLGA solution was then added into the preheated aqueous solution containing lipids, which was then incubated 2~3 h at room

SC

temperature to self-assemble RGD-LPH nanoparticles and allow acetone to evaporate. The

M AN U

remaining free molecules (NCTD and acetone) were removed using an Amicon Ultra-4 centrifugal filter (Molecular cut-off 100,000, Millipore, Billerica, MA) and then resuspended in PBS to obtain a final desired concentration. The size (diameter, nm) and surface charge (ζ-potential, mV) of RGD-LPH were obtained using a Malvern Instruments Zetasizer ZS-90 instrument. The morphology

TE D

was observed utilizing transmission electron microscopy (TEM, JEM-2200FS operated at 200kV) and the staining was done using 1% uranyl acetate. RGD-LPH blank and LPH without

AC C

ingredients.

EP

RGD-anchoring nanoparticles were prepared by a similar method without the use of corresponding

2.3. Labeling LPH and RGD-LPH with Cyanine 5 (Cy5) or Cyanine 7.5 (Cy7.5) The Cy5-labled RGD-LPH [RGD-LPH (Cy5)] was fabricated by the similar procedure to that of RGD-LPH-NCTD. Briefly, Lecithin (2 mg), PEG-DSPE (18 mg) and RGD-PEG-DSPE (2 mg) were dissolved in 4% ethanol aqueous solution and heated to 65

. The Cy5 (100 µg in 100 µL DMSO)

and PEI10K (375 μg) was mixed with PLGA (1.875 mg) in 600 µL acetone solution thoroughly and then incubated in 65

water bath for about 1 min. The Cy5/PEI10K/PLGA solution was then added 7

ACCEPTED MANUSCRIPT into the preheated lipid aqueous solution, which was incubated overnight at room temperature to self-assemble RGD-LPH(Cy5) nanoparticles and allow acetone to evaporate. The remaining free molecules were removed using an Amicon Ultra-4 centrifugal filter (Millipore, Billerica, MA) and

RI PT

then resuspended in PBS to obtain a final theoretical concentration of 50 µg/mL. The Cy5-labeled LPH [LPH(Cy5)] was fabricated by the similar procedure and recipe but without the use of RGD-PEG-DSPE.

The

Cy7.5-labled

LPH

[LPH(Cy7.5)]

or

Cy7.5-labled

RGD-LPH

M AN U

SC

[RGD-LPH(Cy7.5)] were fabricated by the similar procedure but using Cy7.5 instead of Cy5.

2.4. Determination of NCTD encapsulation efficiency (EE) and drug loading (DL) in LPH-NCTD and RGD-LPH-NCTD

NCTD EE and DL in nanoparticles was quantified by LC-MS/MS method (Varian instrument,

TE D

ILC pumps Model ProStar 410 interfaced by Negative ESI, Mass Spectrometer 1200 L, Phenomenex Synergi Hydro-RP Column (100 mm×3.0 mm×2.5 μm), water/acetonitrile with 0.1% formic acid mobile phase, 0.2 mL/min, precursor-to-product ion transition m/z 185

141). Briefly, nanoparticle

EP

solutions were diluted at 1:1 ratio with acetone to break up the PLGA core and then 1M NaOH

AC C

solution was added at 1:1 ratio after complete evaporation of acetone. The solutions were finally vortexed to release free drug from the ionic complexes after addition of 0.5 mg/mL heparin solution (50 μL). After centrifugation (10000 rpm × 5 min), supernatant was diluted by mobile phase and injected to the LC-MS/MS for quantification. NCTD peaks from nanoparticle solutions were quantified by comparing to a calibration curve of known NCTD concentration standards and corrected to account for sample dilution. The weight ratio of NCTD to nanoparticles was defined as the drug loading efficiency (DL), while the weight ratio of entrapped NCTD to added NCTD was 8

ACCEPTED MANUSCRIPT defined as the encapsulation efficiency (EE). Drug release over time was characterized by dialyzing samples against 1×PBS (pH 5.5 and 7.4) in 10 kDa MWCO Slide-A-Lyzer MINI dialysis units (Thermo Scientific) according to the reported

RI PT

methods [26]. The amount of drug retained in the nanoparticle samples was assessed at 0, 1, 4, 8 and 12 hours. Samples were collected and destroyed by acetone, 1M NaOH and 0.5 mg/mL heparin sequentially. After acetone evaporation and centrifugation, the supernatant was analyzed by

M AN U

SC

LC-MS/MS as described above.

2.5. MTT assay

The MTT assay was carried out as described previously [35]. Briefly, cells were seeded on 96-well plates (3-5 ×103 cells/well in 100 µL of complete culture medium). Different concentrations

TE D

(0.25, 0.50 and 1.00 µg/mL) of LPH-NCTD, RGD-LPH-NCTD and RGD-LPH blank were added into the wells 24 h after cell seeding and incubated with cells for 48 h, meanwhile, free NCTD was used as a positive control (2.50, 5.00 and 10.00 µg/mL),. At the end of incubation, 50 µL of the MTT

EP

reagent (5 mg/mL) was added to each well and incubated for 4 h. Then, 200 µL of dimethyl sulfoxide

AC C

(DMSO) was added to each well and incubated for another h. The plate was read using a microplate reader (SpectraMax i3x, Molecular Devices, Sunnyvale, CA) at the wavelength of 570 nm. The inhibition on relative cell growth was determined by the following formula: Cell Viability (%) = (Absorbance at 570 nm of nanoparticle-treated cells) / (Absorbance at 570 nm of control-treated cells) × 100

2.6. Clonogenic assay 9

ACCEPTED MANUSCRIPT The clonogenic assay was performed followed the published protocol [36]. Briefly, cells were seeded on 60-mm culture dishes at a concentration of 100 cells per dish. After overnight culture, cells were treated with 0.5 µg/mL of LPH-NCTD, RGD-LPH-NCTD and RGD-LPH blank for 72 h. PBS

RI PT

was set as a negative control and 5 µg/mL of free NCTD was set as a positive control. Cells were then washed with PBS twice and cultured in drug-free media for 10 more days. At the end of the culture, cell clones were stained with 0.01% (w/v, in 4% paraformaldehyde) crystal violet and

2.7. Soft agar colony formation assay

M AN U

SC

counted.

The soft agar colony formation experiment was performed in 60-mm tissue culture dishes following our previous protocol [37]. Briefly, cells were pretreated with 0.5 µg/mL of LPH-NCTD,

TE D

RGD-LPH-NCTD, or RGD-LPH blank for 3 h. PBS was set as a negative control and 5 µg/mL of free NCTD was set as a positive control. Cells were then collected by trypsinization and suspended in culture media at a concentration of 0.5 × 103 cells/mL. Four milliliter of 0.6% normal melting

EP

point agar was placed into each 60-mm tissue culture dish as the bottom agar. Two milliliter of cell

AC C

suspension (0.5 × 103 cells/mL) and 2 mL of lower melting point agar (0.8%) were mixed together and poured over the bottom agar. After solidification of the upper agar, 3 ml of culture media were added, and dishes were incubated at 37°C in a humidified 5% CO2 atmosphere. Colony formation in the agar was stained with 0.003% (w/v) crystal violet solution, photographed and counted after 4-week culture.

2.8. Suspension serum-free culture tumorigenic sphere formation assay 10

ACCEPTED MANUSCRIPT The suspension serum-free culture tumorigenic sphere formation assay was performed following the published protocol [38]. Briefly, SUM159 cells were pretreated with 0.5 µg/mL of LPH-NCTD, RGD-LPH-NCTD or RGD-LPH blank for 3 h. PBS was set as a negative control and 5 µg/mL of free

RI PT

NCTD was set as a positive control. Then cells were collected by trypsinization and resuspended in serum-free F-12 culture medium containing B-27 (1×), epidermal growth factor (20 ng/mL), basic fibroblast growth factor (20 ng/mL), heparin (5 µg/mL) and 1% of penicillin-streptomycin. The

SC

suspended cells were plated into the ultra-low adhesion 24-well cell culture plates (2500 cells per

M AN U

well) and cultured for 10 days for sphere formation. At the end of culture, spheres with a diameter bigger than 50 μm were photographed and counted.

2.9. Western blot analysis, immunohistochemistry (IHC) and Immunofluorescence (IF) staining

TE D

Cells were lysed using Tris-sodium dodecyl sulfate (SDS) for Western blot analysis as described in our previous publication [39]. The following primary antibodies were used: anti-β-catenin, anti-non-phospho β-catenin (Cell Signaling Technology, Beverly, MA); and anti-β-actin (Sigma).

AC C

EP

The IHC and IF staining were performed following the protocols in our previous publications [40].

2.10. The Cy7.5-labled LPH and RGD-LPH in vivo distribution study The LPH(Cy7.5) and RGD-LPH(Cy7.5) in vivo distribution was studied and compared using the following three nude mouse models. All aspects of the animal studies were performed in accordance with the guidelines defined by the Institutional Animal Care and Use Committee of University of Kentucky. Animal study #1 was designed to compare the distribution of Cy7.5-labeled LPH and RGD-LPH in nude mouse orthotopic primary mammary tumor and spontaneous lung metastatic 11

ACCEPTED MANUSCRIPT tumor. Briefly, the nude mouse orthotopic mammary tumor was established by directly injecting LM2 cells (1 × 106) into 7-week old female Nu/Nu nude mouse (Charles River Laboratories) 4th mammary fat pad as described in our recent publications [31, 32]. When mammary tumor volume

RI PT

reached about 1000 mm3, mice were administered LPH(Cy7.5) or RGD-LPH(Cy7.5) via tail vein (5 µg of Cy7.5 per mouse). Mouse fluorescence imaging (λex/λem = 745 nm/800 nm) was performed using the IVIS live animal imaging system (PerkinElmer, Waltham, MA) at 24, 48, and 72 h

SC

post-injection, respectively. To verify the finding of live animal imaging, mice were sacrificed, then

M AN U

liver, heart, spleen, kidney, mammary tumors and lungs were collected and imaged ex vivo as well. Animal study #2 was designed to detect and compare the distribution of Cy7.5-labeled LPH and RGD-LPH in experimentally-induced lung metastatic tumor. Briefly, 1 × 106 LM2 cells were inoculated via tail vein injection. The establishment of lung metastatic tumor was confirmed by IVIS

TE D

live animal bioluminescence imaging 25 days post-inoculation. The LPH(Cy7.5) and RGD-LPH(Cy7.5) were injected via tail vein into mice bearing metastatic lung tumors (2.5 µg of Cy7.5 per mouse), and mice were imaged at 24, 48, and 72 h post-injection, respectively. Animal

EP

study #3 was designed to detect and compare the long-term distribution of Cy7.5-labeled LPH and

AC C

RGD-LPH in experimentally-induced lung metastatic tumor. This study was performed in the same way as the animal study #2 except that mice were imaged for 10 days. To compare the biodistribution behavior varieties between NCTD and RGD-LPH-NCTD, we labeled NCTD molecules by Cy7.5 amine to get Cy.7.5-NCTD, which was further encapsulated into RGD-LPH (RGD-LPH-Cy7.5-NCTD) according to the method described in Section 2.2.. Briefly, Cy7.5 amine (10 mg), norcantharidin (10 mg) and K2CO3 anhydrous (10 mg) were put into 2 mL of Dimethylformamide (DMF) and stirred overnight at the room temperature. Next day, the reaction 12

ACCEPTED MANUSCRIPT solution was poured into 10 mL distilled water and then purified by dialysis. After lyophilization, Cy7.5-NCTD was characterized by 1H NMR (400 MHz Bruker NEO equipped with a smartphobe, Billerica, MA). Finally, free Cy7.5-NCTD and RGD-LPH-Cy7.5-NCTD were intravenously

RI PT

administered at the dosage of 5 µg Cy7.5 per mouse, and mice were imaged at 12, 24, 48 and 72 h post-injection, respectively.

SC

2.11. Evaluation on the therapeutic effect of systemic administration of free NCTD,

M AN U

LPH-NCTD and RGD-LPH-NCTD on TNBC using a nude mouse orthotopic mammary xenograft tumor model

Nude mouse orthotopic mammary xenograft tumors were established by directly injecting LM2 cells into the 4th mammary gland fat pad as described above. The treatment started at day 10 post

TE D

LM2 cell inoculation when mammary tumor volumes reached about 100 mm3 and terminated at day 44 post LM2 cell inoculation. LPH-NCTD and RGD-LPH-NCTD were administered thrice a week at the dosage of 0.5, 0.5 and 0.75 mg NCTD/kg body weight; PBS was set as a negative control and

EP

free NCTD at the dosage of 5.0, 5.0 and 7.5 mg NCTD/kg body weight was set as a positive control.

AC C

Tumor volumes were measured weekly. At day 46 post LM2 cell inoculation, all mice were sacrificed. The lungs were harvested and imaged by IVIS imaging individually using the same settings: exposure time 60s; F/Stop 1; Binning Small, Field of View D. The mammary tumors were collected, weighted, fixed in 10% neutral formalin, embedded in paraffin, cut into 5 µm sections and stained with hematoxylin and eosin (HE). β-catenin and E-cadherin immunofluorescence staining (IF), E-cadherin immunohistochemistry staining (IHC) in tissue sections was performed using a protocol of heat induced antigen epitope retrieval. 13

ACCEPTED MANUSCRIPT

2.12. Statistical analysis The numerical data are expressed as mean ± SD, tested for different treatment effects by

RI PT

one-way analysis of variance (ANOVA); and the differences between treatment groups were determined using two-tailed student t-test or chi-square test. A p value of <0.05 was considered

SC

statistically significant.

M AN U

3. Results and Discussion 3.1. RGD-LPH fabrication and characterization

The drug-loaded nanoparticle fabrications are dominated by polymer-drug interactions, one of which is hydrophobic interaction as the major driving force for lipophilic drug encapsulation [41].

TE D

Therefore, the poor physicochemical property of the drug could impede its encapsulation into nanoparticles. LPH is an excellent platform for delivery and controlled release of highly hydrophobic drugs (e.g. docetaxel and paclitaxel) [26, 27, 42, 43]. However, LPH loaded with not very

EP

hydrophobic and low M.W. drugs (e.g. cisplatin, fluorouracil) did not demonstrate controlled-release

AC C

profiles [27]. NCTD is a hydrophilic small molecule compound (M.W.=186). In our preliminary experiments, traditional LPH entrapped few free norcantharidin or NCTD molecules. To enhance NCTD affinity to PLGA, PEI10K-NCTD ionic complex was prepared, and in self-assembly process the hydrophobic PLGA core could have a strong affinity with PEI10K [44, 45] as well as it attached NCTD molecules (Fig. 1A). By employing the nanoprecipitation procedure, the PEI10K-NCTD bundles were then incorporated into LPH cloaked with lecithin and PEG-DSPE (Fig. 1A). Further dynamic light scattering (DLS) analysis showed that RGD-LPH displays a unimodal size distribution 14

ACCEPTED MANUSCRIPT with a mean diameter of 60.85 nm (Fig. 1A), which is within the diameter range of typical nanoparticle drug carriers. Moreover, the surface of RGD-LPH is negatively charged (-42.3 mV) (Fig. 1A), which is good for RGD’s active-targeting effect and could avoid the non-specific interaction

of RGD-LPH-NCTD were further characterized by TEM (Fig. 1A).

RI PT

with normal tissues often provoked by cationic particles. The apparent size and spherical morphology

To identify a ligand for decorating the surface of our modified LPH, we screened and compared

SC

expression levels of several membrane proteins in various subtypes of breast cancer cells. We found

M AN U

that only highly migratory and invasive basal mesenchymal-like TNBC cells express a higher integrin α5 (ITGA5) level (Fig. 1B), and the metastatic foci in mouse lung sections also featured an elevated ITGA5 level compared to the healthy mouse lung sections (Fig. 1C). These findings from both in vitro and in vivo studies rationalize active-targeted drug delivery to TNBC cells via ITGA5

TE D

ligands such as RGD motif (Arg-Gly-Asp) [29, 30]. RGD-modified LPH (RGD-LPH) were then prepared for active targeting of NCTD to TNBC cells. The nanoparticles were prepared following the LPH method with the addition of 5 mol% RGD-conjugated PEG-DSPE.

EP

We next established a LC-MS/MS method to determine the EE and DL. The LC-MS/MS

AC C

method could avoid the low specificity in HPLC. A calibration curve was successfully established as shown in Fig.S1. Based on the regression equation derived from the calibration curve, the EE of LPH and RGD-LPH were revealed as 13.04±2.79% and 11.41±2.79%, respectively. The DL of LPH and RGD-LPH were revealed as 0.80±0.16% and 0.66±0.12%, respectively. The previously reported EEs of double-emulsions method were higher (50~80%), but without the removal of free hydrophilic drug in the aqueous phase [24]. To our best knowledge, this is a first try to use nanoprecipitation method to encapsulate hydrophilic drugs. Binding cords other than PEI may be investigated to 15

ACCEPTED MANUSCRIPT improve EE and DL in the future studies. An appropriate release profile is one of key goals of nanocarriers. An extended drug release rate at pH 7.4 was observed, 30.47±5.37%, 73.27±1.33% and 86.64±0.67% of the drug were released,

RI PT

after 1 h, 4 h and 8 h, respectively. The release was almost completed (97.31±0.30%) after 12 h (Fig. S2). However, the release profile is still much slower than ever reported double-emulsion solvent diffusion technique, which released 76% of the drug within 1 h at pH 7.0 [24]. The slower drug

SC

release rate may be ascribed to two factors in our LPH design: (1) the increased apparent size and

M AN U

steric hindrance of NCTD in the NCTD bundles prevent the rapid diffusion of hydrophilic NCTD to the surrounding aqueous phase; (2) as a potent barrier in the nanostructure of already formed LPH/RGD-LPH, the amphiphilic lipid monolayer greatly decreases the possibility of interactions between the PLGA/PEI10K/NCTD core and the outside aqueous environment. A substantial

TE D

difference of release profiles between pH 5.5 and pH 7.4 was not found, indicating that pH conditions barely affected the release behavior. The temporal stability of LPH and RGD-LPH was evaluated over time in both 10% BSS and 10% FBS. An increase in Z-average diameter was not

EP

observed for both LPH-NCTD and RGD-LPH-NCTD (Fig. S3), indicating particle stability in the

AC C

presence of serum proteins. Taken together, the stability features indicated RGD-LPH-NCTD’s suitability for further in vitro and in vivo evaluation. We physically labeled LPH and RGD-LPH with the fluorescent dye Cy5 to compare their uptake by TNBC cells. Cy5 has similar labeling efficiency for LPH and RGD-LPH, as no substantial differences of Cy5 fluorescence intensities were detected between the same amount of Cy5-labeled LPH and RGD-LPH nanoparticles(Fig. S4A). By using epifluorescence scanning analysis, the Cy5 fluorescence was significantly brighter in TNBC cells (LM2) that express high levels of ITGA5 and 16

ACCEPTED MANUSCRIPT were treated with RGD-LPH(Cy5) than the TNBC cells treated with an identical amount of LPH(Cy5) (Fig. 1D). However, no significant difference of Cy5 fluorescence intensity was observed in LPH(Cy5)- and RGD-LPH(Cy5)-treated TNBC cells (MDA-MB-468) that does not express a high

RI PT

level of ITGA5 (Fig. S5). These results suggest that RGD decoration of LPH could markedly improve the recognition and uptake of LPHs by TNBC cells expressing high levels of ITGA5.

SC

3.2. In vitro evaluation of RGD-LPH-NCTD therapeutic effect against TNBC cells.

M AN U

We used the MTT assay (a short-term cell viability and proliferation assay), clonogenic assay (a long-term cell proliferation assay), soft agar colony formation assay (for determining anchorage-independent growth, a feature of tumor cells), and suspension serum-free culture tumorigenic sphere formation assay (for characterizing stem cell-/CSC-like property) to evaluate and

TNBC cells.

TE D

compare the in vitro therapeutic effects of free NCTD, LPH-NCTD and RGD-LPH-NCTD against

MTT assays showed that free NCTD and RGD-LPH-NCTD treatment for 48 h exhibit

EP

consistent and dose-dependent inhibitory effects on three TNBC cells lines, but RGD-LPH-NCTD

AC C

had a stronger inhibitory effect on cell growth than free NCTD and LPH-NCTD (Fig. 2A). This is consistent with our cellular uptake studies, which demonstrate that RGD on the LPH surface facilitates efficient trafficking of LPH nanoparticles to the ITGA5-expressing cells (Fig. 1C). In a long term (10-day) cell growth clonogenic assay, RGD-LPH-NCTD treatment also displays a significant stronger inhibitory effect on TNBC cell growth as evidenced by greatly reduced colony formation (Fig. 2B). Importantly, one single treatment with RGD-LPH-NCTD causes a strong inhibition of colony formation by TNBC cells. In contrast, 4 free NCTD treatments were needed to 17

ACCEPTED MANUSCRIPT achieve a similar inhibitory effect to that of one single treatment with RGD-LPH-NCTD (Fig. 2B). Similarly, RGD-LPH-NCTD also exhibits a stronger inhibitory effect on soft agar colony formation by TNBC cells (Fig. 2C and D). These results suggest that RGD facilitates the in vitro trafficking to

RI PT

TNBC cells, and the LPH platform confers RGD-LPH-NCTD extended inhibitory capability against TNBC cell short-term and long-term growths compared to free NCTD.

Cancer is now considered as a stem cell disease and CSC or CSC-like cells are thought to be

SC

responsible for cancer initiation, metastasis, relapse and therapeutic resistance [46, 47]. It is thus

M AN U

proposed that targeting CSC or CSC-like cells is a critical strategy to reduce cancer metastasis, relapse and drug resistance. We used a suspension and serum free culture tumorigenic sphere formation assay, a well-established and widely-used assay for examining the presence of stem cell/CSC or CSC-like cells [38, 48], to compare the effect of free NCTD, LPH-NCTD and

TE D

RGD-LPH-NCTD treatment on the stemness of TNBC cells. As shown in Fig. 3A-B, only a single 3 h pretreatment with RGD-LPH-NCTD drastically reduces the numbers of tumorigenic suspension spheres formed by TNBC cells. Similar to the effects of free NCTD treatment on the long-term

EP

TNBC cell growth determined by the clonogenic assay (Fig. 2B), 4 free NCTD treatments display a

AC C

strong inhibitory effect on tumorigenic suspension sphere formation by TNBC cells (Fig. 3C). It is reported that nanosize allows for easy cellular uptake, and offers nanoparticles an endocytosis internalization pathway, overcoming multidrug-resistance caused by drug efflux mechanisms on the cellular level [49]. Probably due to this reason, our results demonstrate that RGD-LPH-NCTD being endocytosed into cells has an extended inhibitory effect on TNBC cell cancer stem cell-like property compared to free NCTD molecules.

18

ACCEPTED MANUSCRIPT 3.3. RGD-LPH-NCTD reduces TNBC cell growth and CSC-like property by down-regulating β-catenin We next determined a mechanism by which RGD-LPH-NCTD suppresses TNBC cell growth

RI PT

and CSC-like property. Although NCTD has shown cytotoxic effects against various cancer cells including breast cancer cells, the underlying mechanism has not been well defined. NCTD displays a potent inhibitory effect on the activity of protein phosphatase type 1/2A (PP1/2A) [50], whose

SC

activity plays an important role in stabilizing and activating the β-catenin pathway [51, 52], an

M AN U

important pathway in embryonic development and cancer initiation and metastasis [5, 6]. We thus hypothesize that RGD-LPH-NCTD reduces TNBC cell growth and CSC-like property by down-regulating the β-catenin signaling pathway, which is highly activated and enriched in TNBC contributing to TNBC initiation and metastasis [8-10].

TE D

Phosphorylation of β-catenin by serine/threonine protein kinases promotes its ubiquitination and subsequent proteasomal degradation; and non-phospho-β-catenin represents its active form [53]. We first demonstrated that multiple NCTD treatments are capable of reducing the active (non-phospho-

EP

β-catenin) and total β-catenin protein levels in TNBC cells (Fig. S6). We then performed Western

AC C

blot to compare the effects of a single treatment of free NCTD, LPH-NCTD or RGD-LPH-NCTD on the levels of active and total β-catenin in TNBC cells. As shown in Fig. 4A, only a single treatment with RGD-LPH-NCTD consistently reduced both active (non-phospho-β-catenin) and total β-catenin protein levels in all three TNBC cells, which was consistent with the strongest inhibitory effects of RGD-LPH-NCTD treatment on TNBC cell growth and CSC-like property shown in Fig. 2 and Fig. 3. Accumulation of non-phospho-β-catenin in cytoplasm leads to its nuclear translocation and 19

ACCEPTED MANUSCRIPT activates the expression of Wnt/β-catenin target genes [53]. So β-catenin nuclear localization is an indication of β-catenin activation. We also performed β-catenin immunofluorescence (IF) staining to determine and compare the effects of free NCTD, LPH-NCTD and RGD-LPH-NCTD treatment on

RI PT

β-catenin nuclear localization. As shown in Fig. 4B, the merged pinkish color from β-catenin IF staining (red) and nuclear DNA DAPI fluorescent staining (blue) indicates β-catenin nuclear localization. While free NCTD and LPH-NCTD treatments partially reduced nuclear β-catenin IF

SC

staining, RGD-LPH-NCTD treatment displayed the most drastic effect in decreasing both

M AN U

cytoplasmic and nuclear β-catenin IF staining, which was consistent with the results from Western blot analysis (Fig. 4A). Similar β-catenin IF staining patterns were also observed in SUM-159 cells treated with free NCTD, LPH-NCTD, RGD-LPH-NCTD, or RGD-LPH blank, respectively (Fig. S7).

TE D

To further elucidate that RGD-LPH-NCTD treatment reduces TNBC cell growth and CSC-like property by down-regulating β-catenin, TNBC cells that stably overexpress β-catenin were generated. As shown in Fig. 4C, the increased protein levels of both active (non-phospho- β-catenin) and total

EP

β-catenin in β-catenin stable expressing cells are confirmed by Western blot. In contrast to the drastic

AC C

effect of RGD-LPH-NCTD treatment in reducing the active and total β-catenin in TNBC cells as shown in Fig. 4A, RGD-LPH-NCTD treatment showed no significant effect on the active and total β-catenin levels in β-catenin stably overexpressing cells (Fig. 4C). As a result, RGD-LPH-NCTD treatment displayed no significant inhibitory effect on β-catenin stably overexpressing TNBC cell growth determined by the MTT assay (Fig. 4D) and clonogenic assay (Fig. 4E) as well as cancer stem cell-like property determined by suspension culture tumorigenic sphere formation assay (Fig. 4F). Collectively, these results indicated that RGD-LPH-NCTD reduced TNBC cell growth and 20

ACCEPTED MANUSCRIPT CSC-like property by down-regulating β-catenin.

3.4. RGD-LPH displays a high and long-term accumulation in nude mouse orthotopic

RI PT

mammary TNBC tumor, spontaneous and experimental lung metastatic tumor because of ITGA5-targeting strategy

We next labeled our LPH and RGD-LPH nanoparticles with the fluorescent dye Cy7.5 to detect

SC

and compare their accumulation in nude mouse orthotopic mammary TNBC tumor, spontaneous and

M AN U

experimental lung metastatic tumor. Similar Cy7.5 labeling efficiencies for LPH and RGD-LPH were observed as no substantial differences of Cy7.5 fluorescence intensities were detected between the same amount of LPH(Cy7.5) and RGD-LPH(Cy7.5) nanoparticles (Fig. S4B). We established nude mouse orthotopic mammary TNBC tumors by surgically opening the 4th mammary gland and

TE D

directly injecting LM2 cells into the mammary fat pad. The LM2 cells, human TNBC cells with high lung metastatic capability [33], also stably express the luciferase reporter facilitating the detection of lung metastasis using the IVIS Spectrum in vivo imaging system. When mammary tumors reached

EP

about 1000 mm3, mice were given the same amount of LPH(Cy7.5) or RGD-LPH(Cy7.5) via tail

AC C

vein injection. As shown in Fig. 5A, strong Cy7.5 fluorescent signals were detected using IVIS live animal imaging system in the livers of both mice administered with LPH(Cy7.5) and RGD-LPH(Cy7.5) 24 h post injection. This is an expected outcome given that liver has been widely considered to play an important role in nanoparticle clearance [54]. While the Cy7.5 fluorescence signals were also detected in the primary mammary tumors and the lungs of both mice, they were much stronger in the mouse injected with RGD-LPH(Cy7.5) than that in the LPH(Cy7.5)-injected mouse at 24 h post injection. Moreover, strong Cy7.5 fluorescent signals in the liver, primary 21

ACCEPTED MANUSCRIPT mammary tumor and lungs were only seen in the mouse injected with RGD-LPH(Cy7.5) at the time of 48 h post injection. At 72 h post injection, no significant amount of Cy7.5 fluorescent signal was detected in the mouse injected with LPH(Cy7.5). In contrast, strong Cy7.5 fluorescence in primary

RI PT

mammary tumor and lungs could still be detected in the mouse injected with RGD-LPH(Cy7.5) (Fig. 5A). The ex vivo images for the other major organs were consistent with the in vivo images (Fig. S8). We speculate that the strong Cy7.5 fluorescent signal in the lungs in the mouse administered

SC

with RGD-LPH(Cy7.5) is due to its targeting and accumulation in the metastatic lung. Due to the

M AN U

strong bioluminescent signals from the primary mammary tumor as imaging background, IVIS instrument was not able to detect the bioluminescence signals from the lungs simultaneously while mice were alive. We then sacrificed the mice at the end of 72 h fluorescence live-animal imaging and performed ex vivo bioluminescence and fluorescence imaging for the primary mammary tumors and

TE D

the lungs. Similar and strong bioluminescent signals were detected in primary mammary tumors of both mice administered with LPH(Cy7.5) or RGD-LPH(Cy7.5), indicating that both mice have similar primary mammary tumors. However, much stronger Cy7.5 fluorescent signals were detected

EP

in the mammary tumor of the mouse injected with RGD-LPH(Cy7.5), indicating more

AC C

RGD-LPH(Cy7.5) accumulations in the mammary tumor than LPH(Cy7.5) (Fig. 5A). Meanwhile, bioluminescent signals were detected in the lungs of both mice, indicating the occurrences of spontaneous lung metastasis from the primary mammary tumors. Although the bioluminescent signals in the lungs of the mouse injected with LPH(Cy7.5) were significantly stronger (indicating more lung metastasis), the Cy7.5 fluorescent signals in the lungs of the mouse administered with RGD-LPH(Cy7.5) were much stronger than that in LPH(Cy7.5)-administered mouse lungs (Fig. 5A). Together, these results suggest that both LPH(Cy7.5) and RGD-LPH(Cy7.5) accumulate rapidly and 22

ACCEPTED MANUSCRIPT efficiently in the primary mammary tumor and lung metastatic tumor. However, RGD-LPH accumulation in those sites shows a higher efficiency and lasts for a longer time than LPH. As for the reason of more RGD-LPH(Cy7.5) accumulating in liver than LPH(Cy7.5) (Fig. 5A and S8), RGD receptors on hepatic stellate cells located

RI PT

ligand was reportedly able to bind to the collagen type

in liver and thus offer an unfavorable nanoparticle accumulation at liver [55]. The increased accumulations of RGD-LPH in the primary mammary tumor and metastatic lung were likely due to

SC

the targeting ability of the surface-anchored RGD peptides to the ITGA5 at LM2 cell membrane. The

M AN U

high affinity between RGD and ITGA5 facilitates the rapid accumulation of RGD-LPH(Cy7.5) at the desired sites and thus reduces its in vivo clearance.

The targeting capability of RGD-LPH to metastatic lungs was further investigated by using an experimentally-induced lung metastatic tumor model in nude mouse. The nude mouse lung

TE D

metastasis was first established by tail vein injection of LM2 cells. The successful establishment of lung metastasis was confirmed by the presence of bioluminescence in the lungs as revealed by IVIS live animal imaging (Fig. 5B). The mice with lung metastasis were given LPH(Cy7.5) or

EP

RGD-LPH(Cy7.5); and the negative control mouse (without lung metastasis) was administered with

AC C

RGD-LPH(Cy7.5) via tail vein injection. It was found that strong Cy7.5 fluorescent signals were only detected in the mouse with lung metastasis and administered with RGD-LPH(Cy7.5) at 24, 48 and 72 h post injection (Fig. 5B), confirming the observations in Fig. 5A showing superior accumulation of RGD-LPH in spontaneous lung metastasis to LPH. A 10-day monitoring experiment was also performed to further assess and compare the in vivo retention time of LPH(Cy7.5) and RGD-LPH(Cy7.5). The mouse lung metastasis was established by tail vein injection of LM2 cells. As shown in Fig. S9, mouse #1 and #2 developed similar lung 23

ACCEPTED MANUSCRIPT metastasis and mouse #3 was the negative control without lung metastasis as revealed by the IVIS bioluminescence analysis. Mouse #1 was administered with LPH(Cy7.5) and the mouse #2 and #3 were administered with RGD-LPH(Cy7.5) via tail vein. At the 1st day post injection, nanoparticles

RI PT

distributed through whole mouse bodies in all 3 mice. Once again, the strongest Cy7.5 fluorescence in the lungs was detected in the mouse #2 that developed lung metastasis and was injected with Cy7.5-labeled RGD-LPH. At the 9th and 10th day after injection, Cy7.5 fluorescent signals were still

SC

strong in the lungs of mouse #2, while only weak fluorescent signals were detected in the lungs of

M AN U

mouse #1 and #3. It should be stressed that all three mice were imaged simultaneously at each time point rather than individually, so that the fluorescence intensity from each mouse could be directly compared at each time point. In conclusion, the active-targeting capability originating from the RGD motif promotes fast in vivo targeting of RGD-LPH and hence results in significantly more

TE D

accumulation and longer retention time at the desired sites than untargeted LPH. These findings provide a visual demonstration for the feasibility of ITGA5-targeting strategy in the treatment of

EP

aggressive TNBC.

AC C

3.5. Systematic administration of RGD-LPH-NCTD suppresses TNBC tumor lung metastasis in a mouse orthotopic mammary xenograft tumor model We now successfully demonstrated: (i) strong inhibitory effects of RGD-LPH-NCTD against TNBC cell growth and CSC-like property; and (ii) efficient and prolonged accumulations of RGD-LPH in mouse orthotopic primary mammary TNBC tumor and metastatic lung tumor. We wanted to explore the in vivo therapeutic effect of RGD-LPH-NCTD on TNBC using a mouse orthotopic mammary xenograft tumor model. The LM2 cells were directly injected into the 4th 24

ACCEPTED MANUSCRIPT mammary gland fat pad of nude mouse to produce orthotopic mammary TNBC tumor. When mammary tumors reached about 100 mm3, mice were given LPH-NCTD, RGD-LPH-NCTD, or RGD-LPH-blank via tail vein injection three times a week for 5 consecutive weeks (Fig. 6A). By a

RI PT

similar schedule, the negative and positive control groups of mice were treated with PBS and free NCTD respectively. During the course of the treatment, no significant changes in mouse body weight were observed among different treatment groups (Fig. 6B), indicating no significant toxicity of the

SC

treatments.

M AN U

Compared to the negative control group, only RGD-LPH-NCTD treatment caused a significant decrease in tumor volume and tumor weight (Fig. 6C-E). Consistent with the result from our recent study [31], all 6 mice in the negative control group developed lung metastasis as revealed by the IVIS bioluminescence analysis (Fig. 6F). Similarly, all 6 mice in the RGD-LPH-blank (with no

TE D

NCTD) treatment group also developed lung metastasis. Four and five mice developed lung metastasis from the group administered with free NCTD and LPH-NCTD, respectively. In contrast, only 2 mice in the group treated with RGD-LPH-NCTD developed limited lung metastasis as the

EP

bioluminescent signals in the lungs appear focally. Further statistical analysis indicates that only

AC C

RGD-LPH-NCTD treatment significantly reduces the spontaneous lung metastasis of mammary TNBC tumor (Fig. 6G). These findings were consistent with the strong inhibitory effects of RGD-LPH-NCTD on TNBC cell growth and CSC-like property. The significant suppression of lung metastasis by RGD-LPH-NCTD treatment could be attributed to its efficient and prolonged accumulation in both primary mammary tumors and lung metastatic tumor. Another imaging study disclosed that the random distributing pattern of free NCTD was partially responsible for its inferior therapeutic efficiency to RGD-LPH-NCTD. In this study, NCTD 25

ACCEPTED MANUSCRIPT molecules was labelled by Cy7.5, in which the anhydride ring of norcantharidin could easily react with the amino group of Cy7.5 amine to produce the NCTD-Cy7.5 conjugate (Fig. S10A and B). Both free Cy7.5-NCTD and RGD-LPH(Cy7.5-NCTD) were i.v. administered and imaged at

RI PT

determined time points (Fig. S11A). Both in vivo and ex vivo (Fig. S11) results indicated the random distribution behavior of free NCTD. The extended NCTD release rate from the nanoparticles (Fig. S2) could help protect NCTD in circulation for a relatively longer period, while ITGA5-targeting

SC

strategy facilitated the rapid accumulation to the desired sites (Fig. S11). As a whole, the

M AN U

RGD-LPH-delivered NCTD not only targets tumor cells in the primary tumor, but may also act on the tumor cells in the metastatic lungs, thus significantly suppressing the development of lung metastasis.

TE D

3.6. Systematic administration of RGD-LPH-NCTD reduces β -catenin nuclear localization in mouse primary mammary tumor and improves mammary tumor differentiation To further investigate the mechanism by which RGD-LPH-NCTD treatment significantly

EP

reduces primary mammary tumor growth and lung metastasis, we further analyzed mouse primary

AC C

mammary tumor histology and performed immunohistochemistry (IHC) and IF staining to determine the expression of the epithelial marker E-cadherin and β-catenin nuclear localization, respectively. Mammary tumors in the mice treated with free NCTD, LPH-NCTD or RGD-LPH-NCTD showed much better differentiation than the tumors from negative control and RGD-LPH-blank treatment groups (Fig. 7). Further IHC and IF staining of the epithelial marker E-cadherin (Fig. 7 and S12) demonstrate that tumors from free NCTD-, LPH-NCTD- and RGD-LPH-NCTD-treated mice showed E-cadherin expression; but the expression of E-cadherin in the mammary tumor from 26

ACCEPTED MANUSCRIPT RGD-LPH-NCTD-treated

mice was

much stronger than

that

from

free NCTD- and

LPH-NCTD-treated mice. No E-cadherin staining was detected in mammary tumors from negative control- and RGD-LPH-blank-treated mice. The expression of E-cadherin in tumor tissue is an

RI PT

indication of increased tumor differentiation, which is usually associated with better prognosis and response to drug treatment. The RGD-LPH-NCTD treatment induced significantly higher E-cadherin expression and mammary tumor better differentiation, which likely contributed to its strong

SC

inhibitory effect on lung metastasis.

M AN U

As we found that RGD-LPH-NCTD treatment inhibits TNBC cell growth and CSC-like property by down-regulating β-catenin and reducing its nuclear localization (Fig. 4), we further analyzed the effect of different treatment on β-catenin in mouse mammary tumors. It was found that free NCTD, LPH-NCTD and RGD-LPH-NCTD treatments all reduce mammary tumor β-catenin IF staining

TE D

fluorescence intensity, but the RGD-LPH-NCTD treatment displayed a much stronger effect (Fig. 7). Moreover, RGD-LPH-NCTD treatment also showed a stronger effect in reducing β-catenin nuclear localization in mouse mammary tumors, which was consistent with the effect observed in our cell

EP

culture studies (Fig. 4). Given the reported role of β-catenin signaling in cancer stemness, tumor

AC C

growth and metastasis, the strong effect of RGD-LPH-NCTD in reducing β-catenin level and its nuclear localization contributes to its significant inhibitory effect on mammary tumor growth and lung metastasis.

4. Conclusions In principle, developing a targeted therapy for metastatic TNBCs represents a unique challenge. The therapeutic studies targeting β-catenin are at a very initial stage, and the suboptimal distributions 27

ACCEPTED MANUSCRIPT of small-molecules or biologics greatly restrict their uses in vivo. To date, only PRI-724 has entered into Phase 2 clinical trial due to its very acceptable toxicity profile [15]. There have not been any reports using targeting nanomedicine to specifically down-regulate cancer cell β-catenin level yet.

RI PT

Herein, we demonstrated that it is possible to utilize ITGA5-targeting nanoparticles to attenuate β-catenin and significantly reduce TNBC metastasis. This strategy leverages the inhibitory effect of free NCTD molecules and helps them achieve a therapeutic window. While fluorescence imaging

SC

and NCTD treatment were used in the present study as proof-of-concept modalities, given the

M AN U

generalizability of the nanoparticle-encapsulating capacity, it can be envisioned that the ITGA5-targeting strategy outlined here can be adapted toward a variety of other active therapeutic agents against β-catenin.

TE D

Acknowledgements: We thank Dr. Justin Mobley of NMR Center at University of Kentucky for the assistance to the 1H NMR characterization of Cy7.5-NCTD. This study was supported in part by a Research Scholar Grant (RGS-15-026-01-CSM) from the American Cancer Society to C.Y. and a

EP

research grant from Elsa U. Pardee Foundation to Z.W. This research was also supported by the

AC C

Shared Animal Imaging and Histology Resources of the University of Kentucky Markey Cancer Center (P30CA177558).

Conflict of interest disclosure statement: The authors declare no potential conflict of interest.

28

ACCEPTED MANUSCRIPT Figure legends Fig. 1. RGD-anchoring lipid-polymer hybrid (LPH) nanoparticle encapsulating NCTD (RGD-LPH-NCTD) schematic illustration and characterization. (A) Fabricating scheme and

RI PT

characterization of RGD-LPH. (B) Western blot analysis of the expression level of integrin α5 (ITGA5) in different subtypes of breast cancer cells. (C) Representative images of mouse lung section IF staining of ITGA5 (red), fluorescence staining of nuclear DNA DAPI (blue), and direct

SC

fluorescence imaging of GFP (green). The metastatic tumor cells express GFP. Scale bar = 100 μm.

M AN U

The areas within white circles indicate the metastatic foci formed in the lung. (D) The representative images and quantification (mean ± SD, n = 30 fields) of Cy5-labeled LPH and RGD-LPH uptake by LM2 cells. The LM2 cells were treated with Cy5-labeled LPH or RGD-LPH for 3 h, then fixed with 4% paraformaldehyde, stained with DAPI, viewed and photographed under a phase-contrast

TE D

fluorescence microscope. The quantification was calculated based the mean Cy5 fluorescence intensity per cell in randomly-selected 30 fields of view (FOV). * p < 0.05. Scale bar = 50 μm.

EP

Fig. 2. RGD-LPH-NCTD displays a significant stronger inhibitory effect on TNBC cell

AC C

short-term and long-term growth than free NCTD and LPH-NCTD. (A) Effect of LPH-NCTD, RGD-LPH-NCTD or RGD-LPH blank on TNBC cell short-term growth assessed by MTT assay, which was performed as described in the Methods. Free NCTD was set as a positive control group. Results are expressed as mean ± SD (n = 7). *p < 0.05. (B) Effect of LPH-NCTD, RGD-LPH-NCTD or RGD-LPH blank on TNBC cell long-term growth assessed by clonogenic assay, which was performed as described in the Methods. PBS and free NCTD (5 μg/mL) were set as negative and positive control groups respectively. Results are expressed as mean ± SD (n = 3). *p < 0.05. (C) 29

ACCEPTED MANUSCRIPT Representative images showing the effect of LPH-NCTD, RGD-LPH-NCTD or RGD-LPH blank treatment on soft agar colony formation by MDA-MB-231 cells. PBS and free NCTD (5 μg/mL) were set as negative and positive control groups respectively. Cells were pretreated with each

RI PT

individual treatment for 3 h (calculated by NCTD, 0.5 μg/mL) before being embedded into top agar, and then cultured for 30 days without any further treatment. (D) Quantification of soft agar colony

SC

formation by MDA-MB-231 cells. Results are expressed as mean ± SD (n = 3). *p < 0.05.

M AN U

Fig. 3. RGD-LPH-NCTD displays a significant stronger inhibitory effect on cancer stem cell-like property of TNBC cells than free NCTD and LPH-NCTD. (A) Representative images of suspension serum free culture tumorigenic spheres formed by TNBC SUM159 cells, which were pretreated for 3 h with LPH-NCTD, RGD-LPH-NCTD (calculated by NCTD, 0.5 μg/mL) or

TE D

RGD-LPH blank. PBS and free NCTD (5 μg/mL) were set as negative and positive control groups respectively. After 3 h treatment, cells were collected by trypsinization and used for suspension serum free culture tumorigenic sphere formation assay as described in the Methods. Scale bar=100

EP

μm. (B) and (C) Quantification of suspension spheres formed without and with additional free

AC C

NCTD treatment (5 μg/mL), respectively. Results are expressed as mean ± SD (n = 3). *p < 0.05.

Fig. 4. RGD-LPH-NCTD inhibits TNBC cell growth and cancer stem cell-like property by downregulating the level of β-catenin. (A) Representative images of Western blot analysis of the effect of free NCTD, LPH-NCTD and RGD-LPH-NCTD treatment on the level of non-phospho-β-catenin and total β-catenin in TNBC cells. Cells were treated with LPH-NCTD, RGD-LPH-NCTD (calculated by NCTD, 0.5 μg/mL) or RGD-LPH blank for 48 h and collected for 30

ACCEPTED MANUSCRIPT Western blot analysis. PBS and free NCTD (5 μg/mL) were set as negative and positive control groups respectively. (B) Representative overlaid images of β-catenin (red) and nuclear DNA DAPI (blue) IF staining in LM2 cells treated with LPH-NCTD, RGD-LPH-NCTD (calculated by NCTD,

RI PT

0.5 μg/mL) or RGD-LPH blank for 48 h. PBS and free NCTD (5 μg/mL) were set as negative and positive control groups respectively. Scale bar = 50 µm. (C). Western blot analysis of non-phospho-β-catenin and total β-catenin levels in vector control and β-catenin stably

SC

overexpressing cells; and in β-catenin stably overexpressing cells treated with RGD-LPH blank or

M AN U

RGD-LPH-NCTD (0.5 µg/mL, 48 h). (D-F) β-Catenin overexpression reverses the inhibitory effect of RGD-LPH-NCTD on cell growth (D and E) and cancer stem cell-like property (F). The MTT (D) and clonogenic assay (E) and suspension culture sphere formation assay (F) were performed as

TE D

described in Fig. 2 and Fig. 3. The quantitative data are presented as mean ± SD (n = 3-7).

Fig. 5. RGD-LPH accumulates significantly more than LPH in nude mouse orthotopic mammary TNBC tumor, spontaneous lung metastatic tumor and experimentally-induced lung

EP

metastatic tumor. (A) Representative images showing the accumulation of Cy7.5-labeled LPH and

AC C

RGD-LPH in nude mouse orthotopic mammary TNBC tumor and spontaneous lung metastatic tumor. The nude mouse orthotopic mammary TNBC tumor was established by directly injecting LM2 cells into the 4th mammary gland fat pad as described in the Methods. When the volumes of mammary tumors reached about 1000 mm3, mice were administered with LPH(Cy7.5) or RGD-LPH(Cy7.5) via tail vein injection. Live animal whole-body fluorescence imaging was performed at 24, 48 and 72 h post iv injection, respectively. At the end of 72 h live animal imaging, mice were sacrificed; primary mammary tumors and lungs were harvested for ex vivo fluorescence and bioluminescence imaging 31

ACCEPTED MANUSCRIPT analysis. Similar imaging results were obtained from repeated experiments. (B) Representative images showing the accumulation of Cy7.5-labeled LPH and RGD-LPH in experimentally-induced lung metastatic tumor in nude mice. The nude mouse lung metastatic tumor was established by tail

RI PT

vein injection of LM2 cells as described in the Methods. Twenty five days after LM2 cell injection, the establishment of lung metastatic tumors was confirmed by live animal bioluminescence imaging. Mice were then administered with LPH(Cy7.5) or RGD-LPH(Cy7.5) via tail vein injection. Live

SC

animal fluorescence imaging was carried out at 24, 48 and 72 h post iv injection, respectively. The

M AN U

mouse abdomen part was covered with black board for acquiring better fluorescence imaging at lung areas. Similar imaging results were obtained from repeated experiments.

Fig. 6. Systemic administration of RGD-LPH-NCTD suppresses nude mouse orthotopic

TE D

mammary tumor spontaneous lung metastasis. (A) The drug administration schedule. PBS (for negative control group), free NCTD, LPH-NCTD, RGD-LPH-NCTD and RGD-LPH blank were administered via tail vein injection three times per week for 5 weeks, and mice were sacrificed 24 h

EP

after the final drug injection. (B) and (C) Effect of different treatment group on mouse body weight

AC C

and mammary tumor volume, respectively. Results are presented as mean ± SD (n = 6). *p < 0.05. (D) and (E) Mammary tumor images and average tumor weight from different treatment group, respectively. Results in panel E are presented as mean ± SD (n = 6). *p < 0.05. (F) The ex vivo bioluminescence images of lungs harvested from mice at different treatment group at the end of experiment. (G) The statistical analysis of lung metastasis revealed by the presence of bioluminescence signals in panel F. *p < 0.05.

32

ACCEPTED MANUSCRIPT Fig. 7. Systemic administration of RGD-LPH-NCTD reduces mammary tumor β-catenin level and improves tumor differentiation. Pictures are representative images of mammary tumor H&E staining (scale bar = 10 μm), IHC staining of E-cadherin (scale bar = 10 μm) and IF staining of β-catenin

RI PT

(scale bar = 50 μm). The images for β-catenin IF staining are merged images of β-catenin staining (red) and nuclear DNA DAPI staining (blue).

SC

References

M AN U

[1] Bianchini G, Balko JM, Mayer IA, Sanders ME, Gianni L. Triple-negative breast cancer: challenges and opportunities of a heterogeneous disease. Nature reviews Clinical oncology. 2016;13:674. [2] Hurvitz S, Mead M. Triple-negative breast cancer: advancements in characterization and treatment approach. Current Opinion in Obstetrics and Gynecology. 2016;28:59-69.

TE D

[3] Kast K, Link T, Friedrich K, Petzold A, Niedostatek A, Schoffer O, et al. Impact of breast cancer subtypes and patterns of metastasis on outcome. Breast cancer research and treatment. 2015;150:621-9. [4] Kennecke H, Yerushalmi R, Woods R, Cheang MCU, Voduc D, Speers CH, et al. Metastatic behavior of

EP

breast cancer subtypes. Journal of clinical oncology. 2010;28:3271-7.

AC C

[5] Nusse R, Clevers H. Wnt/β-catenin signaling, disease, and emerging therapeutic modalities. Cell. 2017;169:985-99.

[6] Zhan T, Rindtorff N, Boutros M. Wnt signaling in cancer. Oncogene. 2017;36:1461. [7] Holland JD, Klaus A, Garratt AN, Birchmeier W. Wnt signaling in stem and cancer stem cells. Current opinion in cell biology. 2013;25:254-64. [8] Pohl S-G, Brook N, Agostino M, Arfuso F, Kumar AP, Dharmarajan A. Wnt signaling in triple-negative breast cancer. Oncogenesis. 2017;6:e310. 33

ACCEPTED MANUSCRIPT [9] Rangel MC, Bertolette D, Castro NP, Klauzinska M, Cuttitta F, Salomon DS. Developmental signaling pathways regulating mammary stem cells and contributing to the etiology of triple-negative breast cancer. Breast cancer research and treatment. 2016;156:211-26.

RI PT

[10] Khramtsov AI, Khramtsova GF, Tretiakova M, Huo D, Olopade OI, Goss KH. Wnt/β-catenin pathway activation is enriched in basal-like breast cancers and predicts poor outcome. The American journal of pathology. 2010;176:2911-20.

SC

[11] Jamdade VS, Sethi N, Mundhe NA, Kumar P, Lahkar M, Sinha N. Therapeutic targets of triple

negative

M AN U

breast cancer: a review. British journal of pharmacology. 2015;172:4228-37.

[12] O'toole SA, Beith JM, Millar EK, West R, McLean A, Cazet A, et al. Therapeutic targets in triple negative breast cancer. Journal of clinical pathology. 2013:jclinpath-2012-201361. [13] Takebe N, Miele L, Harris PJ, Jeong W, Bando H, Kahn M, et al. Targeting Notch, Hedgehog, and Wnt

TE D

pathways in cancer stem cells: clinical update. Nature reviews Clinical oncology. 2015;12:445-64. [14] Kahn M. Can we safely target the WNT pathway? Nature reviews Drug discovery. 2014;13:513. [15] Lenz HJ, Kahn M. Safely targeting cancer stem cells via selective catenin coactivator antagonism. Cancer

EP

science. 2014;105:1087-92.

AC C

[16] Katoh M, Katoh M. Molecular genetics and targeted therapy of WNT-related human diseases. International journal of molecular medicine. 2017;40:587-606. [17] Puerto Galvis CE, Vargas Méndez LY, Kouznetsov VV. Cantharidin

Based Small Molecules as

Potential Therapeutic Agents. Chemical biology & drug design. 2013;82:477-99. [18] Deng L, Dong J, Cai H, Wang W. Cantharidin as an antitumor agent: a retrospective review. Current medicinal chemistry. 2013;20:159-66. [19] Hsieh C-H, Chao K, Liao H-F, Chen Y-J. Norcantharidin, derivative of cantharidin, for cancer stem cells. 34

ACCEPTED MANUSCRIPT Evidence-Based Complementary and Alternative Medicine. 2013;2013. [20] Wang W-J, Wu M-Y, Shen M, Zhi Q, Liu Z-Y, Gong F-R, et al. Cantharidin and norcantharidin impair stemness of pancreatic cancer cells by repressing the β-catenin pathway and strengthen the cytotoxicity of

RI PT

gemcitabine and erlotinib. International journal of oncology. 2015;47:1912-22. [21] Ding X-Y, Hong C-J, Liu Y, Gu Z-L, Xing K-L, Zhu A-J, et al. Pharmacokinetics, tissue distribution, and metabolites of a polyvinylpyrrolidone-coated norcantharidin chitosan nanoparticle formulation in rats and

SC

mice, using LC-MS/MS. International journal of nanomedicine. 2012;7:1723.

M AN U

[22] Govender T, Stolnik S, Garnett MC, Illum L, Davis SS. PLGA nanoparticles prepared by nanoprecipitation: drug loading and release studies of a water soluble drug. Journal of Controlled Release. 1999;57:171-85.

[23] Tewes F, Munnier E, Antoon B, Ngaboni Okassa L, Cohen-Jonathan S, Marchais H, et al. Comparative

TE D

study of doxorubicin-loaded poly(lactide-co-glycolide) nanoparticles prepared by single and double emulsion methods. European Journal of Pharmaceutics and Biopharmaceutics. 2007;66:488-92. [24] Cohen-Sela E, Chorny M, Koroukhov N, Danenberg HD, Golomb G. A new double emulsion solvent

AC C

Release. 2009;133:90-5.

EP

diffusion technique for encapsulating hydrophilic molecules in PLGA nanoparticles. Journal of Controlled

[25] Pagels RF, Prud'homme RK. Polymeric nanoparticles and microparticles for the delivery of peptides, biologics, and soluble therapeutics. Journal of Controlled Release. 2015;219:519-35. [26] Zhang L, Chan JM, Gu FX, Rhee J-W, Wang AZ, Radovic-Moreno AF, et al. Self-assembled lipid− polymer hybrid nanoparticles: a robust drug delivery platform. ACS nano. 2008;2:1696-702. [27] Hadinoto K, Sundaresan A, Cheow WS. Lipid–polymer hybrid nanoparticles as a new generation therapeutic delivery platform: a review. European journal of pharmaceutics and biopharmaceutics. 35

ACCEPTED MANUSCRIPT 2013;85:427-43. [28] Yu M, Xu L, Tian F, Su Q, Zheng N, Yang Y, et al. Rapid transport of deformation-tuned nanoparticles across biological hydrogels and cellular barriers. Nature communications. 2018;9:2607.

Atomic details of the fibronectin receptor. J Cell Biol. 2012;197:131-40.

RI PT

[29] Nagae M, Re S, Mihara E, Nogi T, Sugita Y, Takagi J. Crystal structure of α5β1 integrin ectodomain:

[30] García AJ, Schwarzbauer JE, Boettiger D. Distinct activation states of α5β1 integrin show differential

SC

binding to RGD and synergy domains of fibronectin. Biochemistry. 2002;41:9063-9.

M AN U

[31] Humphries B, Wang Z, Li Y, Jhan J-R, Jiang Y, Yang C. ARHGAP18 downregulation by miR-200b suppresses metastasis of triple-negative breast cancer by enhancing activation of RhoA. Cancer research. 2017;77:4051-64.

[32] Humphries B, Wang Z, Oom AL, Fisher T, Tan D, Cui Y, et al. MicroRNA-200b targets protein kinase Cα

TE D

and suppresses triple-negative breast cancer metastasis. Carcinogenesis. 2014;35:2254-63. [33] Minn AJ, Gupta GP, Siegel PM, Bos PD, Shu W, Giri DD, et al. Genes that mediate breast cancer metastasis to lung. Nature. 2005;436:518-24.

EP

[34] Wang Y, Zhou K, Huang G, Hensley C, Huang X, Ma X, et al. A nanoparticle-based strategy for the

AC C

imaging of a broad range of tumours by nonlinear amplification of microenvironment signals. Nature materials. 2014;13:204.

[35] Wang Z, Humphries B, Xiao H, Jiang Y, Yang C. MicroRNA-200b suppresses arsenic-transformed cell migration by targeting protein kinase Cα and Wnt5b-protein kinase Cα positive feedback loop and inhibiting Rac1 activation. Journal of Biological Chemistry. 2014;289:18373-86. [36] Franken NA, Rodermond HM, Stap J, Haveman J, Van Bree C. Clonogenic assay of cells in vitro. Nature protocols. 2006;1:2315. 36

ACCEPTED MANUSCRIPT [37] Yang C, Wu J, Zhang R, Zhang P, Eckard J, Yusuf R, et al. Caffeic acid phenethyl ester (CAPE) prevents transformation of human cells by arsenite (As) and suppresses growth of As-transformed cells. Toxicology. 2005;213:81-96.

RI PT

[38] Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes & development. 2003;17:1253-70.

SC

[39] Yang C, Liu Y, Leskow FC, Weaver VM, Kazanietz MG. Rac-GAP-dependent inhibition of breast cancer

M AN U

cell proliferation by β2-chimerin. Journal of Biological Chemistry. 2005;280:24363-70. [40] Zhao Y, Tan YS, Haslam SZ, Yang C. Perfluorooctanoic acid effects on steroid hormone and growth factor levels mediate stimulation of peripubertal mammary gland development in C57BL/6 mice. Toxicological sciences. 2010;115:214-24.

TE D

[41] Ke X, Ng VWL, Ono RJ, Chan JM, Krishnamurthy S, Wang Y, et al. Role of non-covalent and covalent interactions in cargo loading capacity and stability of polymeric micelles. Journal of Controlled Release. 2014;193:9-26.

EP

[42] Chan JM, Zhang L, Yuet KP, Liao G, Rhee J-W, Langer R, et al. PLGA–lecithin–PEG core–shell

AC C

nanoparticles for controlled drug delivery. Biomaterials. 2009;30:1627-34. [43] Dehaini D, Fang RH, Luk BT, Pang Z, Hu C-MJ, Kroll AV, et al. Ultra-small lipid–polymer hybrid nanoparticles for tumor-penetrating drug delivery. Nanoscale. 2016;8:14411-9. [44] Yu D, Zhang Y, Zhou X, Mao Z, Gao C. Influence of surface coating of PLGA particles on the internalization and functions of human endothelial cells. Biomacromolecules. 2012;13:3272-82. [45] Bivas-Benita M, Romeijn S, Junginger HE, Borchard G. PLGA–PEI nanoparticles for gene delivery to pulmonary epithelium. European Journal of Pharmaceutics and Biopharmaceutics. 2004;58:1-6. 37

ACCEPTED MANUSCRIPT [46] Nguyen LV, Vanner R, Dirks P, Eaves CJ. Cancer stem cells: an evolving concept. Nature Reviews Cancer. 2012;12:133. [47] Shukla S, Meeran SM. Epigenetics of cancer stem cells: Pathways and therapeutics. Biochimica et

RI PT

Biophysica Acta (BBA)-General Subjects. 2014;1840:3494-502. [48] Nolte SM, Venugopal C, McFarlane N, Morozova O, Hallett RM, O’Farrell E, et al. A cancer stem cell model for studying brain metastases from primary lung cancer. Journal of the National Cancer Institute.

SC

2013;105:551-62.

M AN U

[49] Hu C-MJ, Zhang L. Therapeutic nanoparticles to combat cancer drug resistance. Current drug metabolism. 2009;10:836-41.

[50] Hill TA, Stewart SG, Gordon CP, Ackland SP, Gilbert J, Sauer B, et al. Norcantharidin analogues: synthesis, anticancer activity and protein phosphatase 1 and 2A inhibition. ChemMedChem. 2008;3:1878-92.

TE D

[51] Wlodarchak N, Xing Y. PP2A as a master regulator of the cell cycle. Critical reviews in biochemistry and molecular biology. 2016;51:162-84.

[52] Luo W, Peterson A, Garcia BA, Coombs G, Kofahl B, Heinrich R, et al. Protein phosphatase 1 regulates catenin degradation complex. The EMBO journal. 2007;26:1511-21.

EP

assembly and function of the β

AC C

[53] Stamos JL, Weis WI. The β-catenin destruction complex. Cold Spring Harbor perspectives in biology. 2013;5:a007898.

[54] Alexis F, Pridgen E, Molnar LK, Farokhzad OC. Factors affecting the clearance and biodistribution of polymeric nanoparticles. Molecular pharmaceutics. 2008;5:505-15. [55] Zhang Y-N, Poon W, Tavares AJ, McGilvray ID, Chan WC. Nanoparticle–liver interactions: Cellular uptake and hepatobiliary elimination. Journal of controlled release. 2016;240:332-48.

38

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT