Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted...

15
Therapeutics, Targets, and Chemical Biology Bone-Induced Expression of Integrin b3 Enables Targeted Nanotherapy of Breast Cancer Metastases Michael H. Ross 1 , Alison K. Esser 1 , Gregory C. Fox 1 , Anne H. Schmieder 2 , Xiaoxia Yang 2 , Grace Hu 2 , Dipanjan Pan 3 , Xinming Su 1 , Yalin Xu 1 , Deborah V. Novack 4,5 , Thomas Walsh 6 , Graham A. Colditz 6 , Gabriel H. Lukaszewicz 1 , Elizabeth Cordell 1 , Joshua Novack 1 , James A. J. Fitzpatrick 7 , David L. Waning 8 , Khalid S. Mohammad 9 , Theresa A. Guise 9 , Gregory M. Lanza 2 , and Katherine N. Weilbaecher 1 Abstract Bone metastases occur in approximately 70% of metastatic breast cancer patients, often leading to skeletal injuries. Current treatments are mainly palliative and underscore the unmet clin- ical need for improved therapies. In this study, we provide preclinical evidence for an antimetastatic therapy based on target- ing integrin b3(b3), which is selectively induced on breast cancer cells in bone by the local bone microenvironment. In a preclinical model of breast cancer, b3 was strongly expressed on bone metastatic cancer cells, but not primary mammary tumors or visceral metastases. In tumor tissue from breast cancer patients, b3 was signicantly elevated on bone metastases relative to primary tumors from the same patient (n ¼ 42). Mechanistic investigations revealed that TGFb signaling through SMAD2/ SMAD3 was necessary for breast cancer induction of b3 within the bone. Using a micelle-based nanoparticle therapy that recog- nizes integrin avb3(avb3-MPs of 12.5 nm), we demonstrated specic localization to breast cancer bone metastases in mice. Using this system for targeted delivery of the chemotherapeutic docetaxel, we showed that bone tumor burden could be reduced signicantly with less bone destruction and less hepatotoxicity compared with equimolar doses of free docetaxel. Furthermore, mice treated with avb3-MP-docetaxel exhibited a signicant decrease in bone-residing tumor cell proliferation compared with free docetaxel. Taken together, our results offer preclinical proof of concept for a method to enhance delivery of chemother- apeutics to breast cancer cells within the bone by exploiting their selective expression of integrin avb3 at that metastatic site. Cancer Res; 77(22); 6299312. Ó2017 AACR. Introduction Bone metastases occur in approximately 70% of metastatic breast cancer patients (1), often leading to the development of signicant skeletal complications, such as pathologic fracture, spinal cord compression, or bone pain (2). Current treatments, radiation, surgery, chemotherapy, and antiresorptive drugs, can improve the quality of life but are rarely curative, with limited effect on overall survival (35). Treating breast cancer bone metastases has proven difcult due to frequent dissemination of metastases throughout the skeleton, a lack of tumor-specic targets expressed across breast cancer subtypes, and the chemo- protective nature of the bone microenvironment (2, 6, 7). One approach to overcome these problems has been to enhance drug delivery to bone, most commonly by using bone matrixtargeted hydroxyapatite-avid bisphosphonates (810). Bisphosphonates are standard of care in patients with bone metastases or osteoporosis for their ability to inhibit osteoclast function and formation (1, 2, 10). While osteoclast inhibition can attenuate tumor-associated bone destruction with second- ary effects on bone tumor growth, it does not improve overall survival in metastatic breast cancer patients (11, 12). In line with this, bisphosphonate-targeted chemotherapy delivery can reduce mouse models of breast cancerassociated bone loss more effectively than free-drug chemotherapy (13, 14). 1 Department of Medicine, Division of Molecular Oncology, Washington Univer- sity School of Medicine, St. Louis, Missouri. 2 Department of Medicine, Division of Cardiology, Washington University School of Medicine, St. Louis, Missouri. 3 Department of Bioengineering, University of Illinois at Urbana-Champaign, Champaign, Illinois. 4 Department of Medicine, Division of Bone and Mineral Diseases, Washington University School of Medicine, St. Louis, Missouri. 5 Department of Pathology, Washington University School of Medicine, St. Louis, Missouri. 6 Department of Surgery, Division of Public Health Sciences, St. Louis Breast Tissue Registry, Washington University School of Medicine, St. Louis, Missouri. 7 Departments of Cell Biology & Physiology and Neuroscience, Washington University Center for Cellular Imaging, Washington University School of Medicine, St. Louis, Missouri. 8 Department of Cellular and Molecular Physiology, Pennsylvania State University College of Medicine, Hershey, Penn- sylvania. 9 Department of Medicine, Division of Endocrinology, Indiana University School of Medicine, Indianapolis, Indiana. Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). A.K. Esser and G.C. Fox contributed equally to this work. G.M. Lanza and K.N. Weilbaecher are co-senior authors of this article. Corresponding Author: Katherine N. Weilbaecher, Division of Molecular Oncol- ogy, Washington University School of Medicine, Campus Box 8069, 330 South Euclid Avenue, St. Louis, MO 63110. Phone: 314-454-8858; Fax: 314-454-8979; E-mail: [email protected] doi: 10.1158/0008-5472.CAN-17-1225 Ó2017 American Association for Cancer Research. Cancer Research www.aacrjournals.org 6299 on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Transcript of Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted...

Page 1: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

Therapeutics, Targets, and Chemical Biology

Bone-Induced Expression of Integrin b3 EnablesTargeted Nanotherapy of Breast CancerMetastasesMichael H. Ross1, Alison K. Esser1, Gregory C. Fox1, Anne H. Schmieder2, Xiaoxia Yang2,Grace Hu2, Dipanjan Pan3, Xinming Su1, Yalin Xu1, Deborah V. Novack4,5, Thomas Walsh6,Graham A. Colditz6, Gabriel H. Lukaszewicz1, Elizabeth Cordell1, Joshua Novack1,James A. J. Fitzpatrick7, David L.Waning8, Khalid S. Mohammad9, Theresa A. Guise9,Gregory M. Lanza2, and Katherine N.Weilbaecher1

Abstract

Bone metastases occur in approximately 70% of metastaticbreast cancer patients, often leading to skeletal injuries. Currenttreatments are mainly palliative and underscore the unmet clin-ical need for improved therapies. In this study, we providepreclinical evidence for an antimetastatic therapy based on target-ing integrin b3 (b3), which is selectively induced on breast cancercells in bone by the local bonemicroenvironment. In a preclinicalmodel of breast cancer, b3 was strongly expressed on bonemetastatic cancer cells, but not primary mammary tumors orvisceral metastases. In tumor tissue from breast cancer patients,b3 was significantly elevated on bone metastases relative toprimary tumors from the same patient (n ¼ 42). Mechanisticinvestigations revealed that TGFb signaling through SMAD2/SMAD3 was necessary for breast cancer induction of b3 within

the bone. Using a micelle-based nanoparticle therapy that recog-nizes integrin avb3 (avb3-MPs of �12.5 nm), we demonstratedspecific localization to breast cancer bone metastases in mice.Using this system for targeted delivery of the chemotherapeuticdocetaxel, we showed that bone tumor burden could be reducedsignificantly with less bone destruction and less hepatotoxicitycompared with equimolar doses of free docetaxel. Furthermore,mice treated with avb3-MP-docetaxel exhibited a significantdecrease in bone-residing tumor cell proliferation comparedwith free docetaxel. Taken together, our results offer preclinicalproof of concept for a method to enhance delivery of chemother-apeutics to breast cancer cells within the bone by exploitingtheir selective expression of integrin avb3 at that metastatic site.Cancer Res; 77(22); 6299–312. �2017 AACR.

IntroductionBone metastases occur in approximately 70% of metastatic

breast cancer patients (1), often leading to the development ofsignificant skeletal complications, such as pathologic fracture,spinal cord compression, or bone pain (2). Current treatments,radiation, surgery, chemotherapy, and antiresorptive drugs, canimprove the quality of life but are rarely curative, with limitedeffect on overall survival (3–5). Treating breast cancer bonemetastases has proven difficult due to frequent dissemination ofmetastases throughout the skeleton, a lack of tumor-specifictargets expressed across breast cancer subtypes, and the chemo-protective nature of the bone microenvironment (2, 6, 7).

One approach to overcome these problems has been toenhance drug delivery to bone, most commonly by using bonematrix–targeted hydroxyapatite-avid bisphosphonates (8–10).Bisphosphonates are standard of care in patients with bonemetastases or osteoporosis for their ability to inhibit osteoclastfunction and formation (1, 2, 10). While osteoclast inhibitioncan attenuate tumor-associated bone destruction with second-ary effects on bone tumor growth, it does not improve overallsurvival in metastatic breast cancer patients (11, 12). In linewith this, bisphosphonate-targeted chemotherapy deliverycan reduce mouse models of breast cancer–associated boneloss more effectively than free-drug chemotherapy (13, 14).

1Department of Medicine, Division of Molecular Oncology, Washington Univer-sity School of Medicine, St. Louis, Missouri. 2Department of Medicine, Division ofCardiology, Washington University School of Medicine, St. Louis, Missouri.3Department of Bioengineering, University of Illinois at Urbana-Champaign,Champaign, Illinois. 4Department of Medicine, Division of Bone and MineralDiseases, Washington University School of Medicine, St. Louis, Missouri.5Department of Pathology,Washington University School of Medicine, St. Louis,Missouri. 6Department of Surgery, Division of Public Health Sciences, St. LouisBreast Tissue Registry, Washington University School of Medicine, St. Louis,Missouri. 7Departments of Cell Biology & Physiology and Neuroscience,Washington University Center for Cellular Imaging, Washington UniversitySchool of Medicine, St. Louis, Missouri. 8Department of Cellular and MolecularPhysiology, Pennsylvania State University College of Medicine, Hershey, Penn-sylvania. 9Department ofMedicine, Division of Endocrinology, IndianaUniversitySchool of Medicine, Indianapolis, Indiana.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

A.K. Esser and G.C. Fox contributed equally to this work.

G.M. Lanza and K.N. Weilbaecher are co-senior authors of this article.

Corresponding Author: Katherine N. Weilbaecher, Division of Molecular Oncol-ogy, Washington University School of Medicine, Campus Box 8069, 330 SouthEuclid Avenue, St. Louis, MO 63110. Phone: 314-454-8858; Fax: 314-454-8979;E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-17-1225

�2017 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 6299

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 2: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarilytargets osteoclasts and neighboring marrow cells near the bonematrix (8, 15). To improve drug uptake by bone metastases, wesought to target tumor cells in the bone microenvironmentmore directly.

One targeting candidate of interest was the integrin avb3.Integrins are heterodimeric cell surface receptors, composed ofan a and b subunit from a large family of subunits, that bind tocomponents of the extracellular matrix or to other cells (16).Integrins are expressed in a cell-specific and context-dependentmanner, and are critical to all aspects of cancer progression andmetastasis through effects on migration, invasion, and cell sur-vival (16–18). Specifically, integrin avb3 is composed of thetightly regulated integrin subunit b3 and the more widelyexpressed av subunit. Expression of avb3 on most cells in thebody is typically low, but is elevated on several cancer types (17),aswell as a variety of cell types that are importantwithin the tumormicroenvironment, such as osteoclasts (11, 12), neoangiogenicendothelium (19), and tumor-promotingmacrophages (20). Theexpression of avb3 on these cells led to the investigation of avb3as a therapeutic target for cancer treatment (21). In several clinicaltrials, pharmacologic inhibitionwithavb3 antagonists as a single-agent cancer therapy did not demonstrate a significant clinicaleffect (22, 23). We, and others, have demonstrated that theinefficiency of avb3 inhibitors in advanced cancer may be dueto the inherent complexities of integrin signaling and off-targeteffects of avb3 antagonism (20, 24).

Rather than aiming to antagonize avb3 function, we sought toutilize avb3 as a molecular target for enhancing drug delivery tocells of interest within the bone. Integrin b3 and avb3 expressionhas been observed on human breast cancer bone metastases(25–27), and ectopic overexpression of tumoral b3 on breastcancer has been demonstrated to enhance tumor establishment inbone (26, 28). However, for the purpose of targeting avb3,physiologic expression on breast cancer during tumor growthandmetastasis has not beendefined.Here,we report that thebonemicroenvironment preferentially induces integrin b3 on breastcancermetastases, as comparedwith the primary tumor or visceralmetastases, and we identify TGFb to be responsible for thisinduction. Utilizing this information, we evaluated nanoparti-cle-mediated drug delivery targeted against integrin avb3. Werecently developed phospholipid/polysorbate-80 micelle nano-particles (MPs, �12.5 nm) for their small size and unique mech-anism of "contact-facilitated drug delivery" (29); in this study, wedemonstrate that integrin avb3–targeted micelle nanoparticles(avb3-MPs) carrying the chemotherapeutic docetaxel reducebone metastases and tumor-associated bone destruction moreeffectively than free docetaxel. Collectively, we demonstrate thatavb3 is a tumor target on breast cancer bone metastases andprovides support for safer, more effective therapies against thisoften incurable disease by targeting integrin avb3.

Materials and MethodsAnimals

Animal studies were performed according to the guidelinesestablished by the Washington University Institutional AnimalCare and Use Committee. PyMT-Bo1 tumor cells were implantedinto female C57BL/6 mice, or 4T1 tumor cells were implantedinto female BALB/c mice. All mice were obtained from The

Jackson Laboratory, and injected at 6–7 weeks of age. All rodentswere housed according to the guidelines of the Division ofComparative Medicine, Washington University School of Medi-cine (St. Louis, MO). In collaboration with Dr. T.A. Guise, histo-logic bone sections from female athymic nude mice injected withMDA-MB-231 human breast cancer cells were obtained from anexperiment described previously (30).

Cell linesThe murine C57BL/6 PyMT-Bo1 luminal B breast cancer cell

line (stably expressing GFP and firefly luciferase genes) wasoriginally isolated from a transgenic MMTV-PyMT breasttumor, as validated and described previously (20). The murineBALB/c 4T1 triple-negative breast cancer cell line was purchasedfrom ATCC, as described previously (31). Human MDA-MB-231 breast cancer cells (HTB-26) were purchased from ATCC(30). All cells were maintained at subconfluence in DMEM with10% FBS and 0.5% penicillin–streptomycin, in a humidifiedchamber at standard culture conditions. Low-passage stockswere used and regularly tested forMycoplasma and maintenanceof growth characteristics.

Murine cancer modelsTo establish orthotopicmammary fat pad (MFP) tumors, 0.1�

106 tumor cells in 50-mL PBS were injected into MFP tissue of7-week-old female mice. To establish experimental secondarymetastases, 0.05 � 106 tumor cells in 50-mL PBS were intracar-dially injected into the left ventricular chamber of 6-week-oldfemale, with one exception; in collaboration with Dr. T.A. Guise,human MDA-MB-231 tumor cells were intracardially injected(0.1 � 106 tumor cells in 100-mL PBS) into 4-week-old femaleathymic nude mice, as described previously (30).

Synthesis of avb3-targeted micelle nanoparticlesPhospholipid/polysorbate-80micelle nanoparticles (MP)were

prepared as amicrofluidized suspension of 20% (v/v), combiningpolysorbate Tween80 (SigmaAldrich, Inc.)with a 2.0% (w/v) of asurfactant comixture, and 1.7% (w/v) glycerin in pH 6.5 carbon-ate buffer, as described previously (29). Optionally, the surfactantcomixture included 2.28 mol% of docetaxel-prodrug (DTX-PD),and/or 0.15 mol% of avb3-targeted quinolone nonpeptide cou-pled tophosphatidylethanolamine-PEG2000,with the remainingmol% lecithin. Docetaxel was modified into an Sn2 lipase-labilephosphatidylcholine DTX-PD as described previously (32). Forfluorescent labeling, rhodamine-conjugated to phosphatidyleth-anolamine (0.1mol%)was incorporated into the lipid surfactant.The surfactant components for each formulation were combinedwith the polysorbate, buffer, and glycerin mixtures and werehomogenized at 20,000 psi for 4 minutes at 4�C with a micro-fluidics (M110s; Microfluidics, Inc). The nanoparticles were pre-served under inert gas in sterile sealed vials until use. Dynamiclight scattering showed a nominal particle size of 12.5 � 0.8 nm,withpolydispersities 0.290�0.03, and an average electrophoreticzeta potential of �3.82 � 1.23 mV.

The nanoparticle-targeting quinolone nonpeptide specific forintegrin avb3 was originally developed by Bristol–Myers SquibbMedical Imaging (US patent 6,511,648 and related patents) andcoupled to phosphatidylethanolamine-PEG2000, as previouslydescribed (29). The quinolone nonpeptide was initially charac-terized as the 111In-DOTA conjugate RP478 and cyan 5.5 homo-log TA145 (33). This avb3-targeting ligand is selective for cells

Ross et al.

Cancer Res; 77(22) November 15, 2017 Cancer Research6300

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 3: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

expressing avb3 (IC50¼ 12 nmol/L), as compared with IC50 > 10mmol/L for anb5, a5b1, or aIIbb3 (34). Furthermore, the affinityof the anb3-targeting ligand increases 15-fold for activated integ-rin avb3 receptor in the presence of Mn2þ (33). In vitro, pretreat-ing human endothelial with the human-specific integrin avb3antibody (LM609) competitively inhibited binding of this avb3-targeting ligand (33). This avb3-targeted ligand has been appliedto the surface of large, vascular-constrained perfluorocarbonnanoparticles (�250 nm), which displayed a specific affinity foravb3-expressing angiogenic blood vessels and avb3-expressingmelanoma cells, in contrast to nanoparticle controls (32, 35, 36).The in vivo specificity of thisavb3-targeting ligandhasbeen furtherdemonstrated through competitive pretreatment with unlabeledavb3-targeting ligand, which inhibited avb3-targeted nanoparti-cle binding (35).

Bioluminescence imagingIn vivo and ex vivo bioluminescence imaging (BLI) was per-

formed on IVIS50 (PerkinElmer) as described previously (20).Total photon flux (photons/sec) wasmeasured from fixed regionsof interest (ROI) over entire mouse or manually around ex vivoorgans using Living Image 2.6, as indicated. Mice with outstand-ing chest BLI intensity indicative of a failed intracardiac injectionor with ineffective D-luciferin administration were excluded fromall analyses. Investigators were blinded to treatment groupsduring all BLI analyses.

Rhodamine-labeled MP colocalization with breastcancer bone metastases

Tumor-bearing mice day 8 postintracardially injected withPyMT-Bo1 cells, or age-matched tumor-free mice, were treatedwith a single dose of rhodamine-labeled MPs (either nontargetedor avb3-targeted) at a nanoparticle dose of 2 nmol/g mouseweight. Investigators were blinded to treatment groups duringintravenous MP and avb3-MP injections. MPs were allowed tocirculate for 3 hours in vivo, before sacrifice and tissue collection.UnboundMPswere cleared from circulation via cardiac perfusionwith 30 mL of PBS.

For fluorescent analysis of nanoparticle colocalization withinbone, fresh-frozen long bones in optimum cutting temperatureembeddingmedium (Tissue-Tek)were sectioned 5-mmthick ontocryofilm tape (Section Lab Co., Ltd) at the histology core of theWashington University Musculoskeletal Research Center. Cryo-film tape sections were fixed in�20�C acetone for 5 minutes andair-dried for 10 minutes. Slides were briefly rehydrated with PBS,and then mounted with ProLong Gold with DAPI (Invitrogen).

Fluorescent imaging and analysisFluorescent images were captured on a Photometrics Cool-

SNAPMYOcamera connected to aNikon Eclipse Ti-Emicroscope,acquired with the 4� Plan Fluor PhL DL objective through aTxRED HYQ filter cube and DAPI filter cube. The TxRED fluores-cent channel was equally set using fluorescent look-up-table(LUT) to the same minimum and maximum values, gamma ¼1. Fluorescent colocalization analysis was performed using NIS-Elements AR (Nikon Canada, Inc.). Bone metastatic region ofinterest (ROI) was drawn at the tumor/marrow boundary, asdefined by DAPI visualization of nuclei density. Using the objectcount function in NIS-Elements AR, fluorescent nanoparticlecolocalization was calculated via the ratio of rhodamine-positivepixel area by the total tumor ROI pixel area [(number of pixels of

positively labeled objects within the fixed tumor area)/(totalnumber of pixels within the fixed tumor area) � 100].

Treatment of murine breast cancer metastasesUsing the experimental metastasis model, PyMT-Bo1 metasta-

ses were established within female C57BL/6 mice. BLI analysisconfirmed the establishment of bone metastases on day 3 post-intracardiac injection, andmetastatic-bearing mice were random-ly sorted and treatment started. Docetaxel (LC Laboratories) wasprepared at 10 mg/mL in a Tween 80/ethanol/saline (20:13:67,v/v/v) solution for drug solubility, and diluted to 0.5 mg/mLdocetaxel or 1.0 mg/mL in saline for intravenous administration.For nanoparticle treatment, mice received either a cumulativedose of 5.55mg/kg docetaxel, an equimolar dose of nanoparticle-encapsulated docetaxel-prodrug (2.2 mmol/kg), or an equalamount of saline or cargo-free avb3-MP, fractionally adminis-tered intravenously every three days. A priori comparisons ofinterest were between saline-treated mice as compared withdocetaxel or avb3-MP/DTX-PD, or between saline-treated micecompared with the nanoparticle control treatments.

In collaboration with Dr. T.A. Guise, histologic bone sectionswere obtained from female athymic nudemice bearingMDA-MB-231 bonemetastases, treated dailywith the TGFb receptor I kinaseinhibitor, 60 mg/kg/d SD-208 (Epichem Pty Ltd, Murdoch Uni-versity, Australia) or vehicle (1% methylcellulose), from anexperiment published previously (30).

Serum chemistry and hematologic analysisBlood was collected by submandibular vein puncture into

Microtainer EDTA tubes (BD Biosciences) for hematologic anal-ysis (Hemavet 950 FS; Drew Scientific, Inc.) or Microtainer serumseparator tubes (BD Biosciences) for serum chemistry analysis ona Liasys 330 AMS Diagnostic liquid chemistry analyzer. Investi-gators were blinded to treatment groups during analysis.

MTT assayThe MTT assay was performed as described previously (31).

Signal intensity normalized to 0% for the viability of cells at timeof drug addition, and 100% for the viability of vehicle controltreated cells at 72 hours.

RadiographyOsteolytic lesions were imaged by X-Ray imaging system (Fax-

itron), and lesion area within the tibiofemoral joint was quan-tified using ImageJ (NIH, Bethesda, Maryland).

Immunohistologic stainingFreshly removed tissue was fixed in 10% neutral buffered

formalin for 24 hours. Bone was decalcified in 14% EDTA for10 days. Tissue was paraffin embedded and sectioned 5-mm thickat the histology core of the Washington University Musculoskel-etal Research Center. Standard tartrate-resistant acidic phospha-tase (TRAP) staining or hematoxylin and eosin (H&E) stainingwas performed by the musculoskeletal histology core of theWashington University Musculoskeletal Research Center.

For IHC, all slides were stained in parallel, using identicalstaining conditions. Paraffin tissue slides were prepared byimmersing slides in xylene and rehydrating tissue in 100% eth-anol, 95% ethanol, 70% ethanol, 50% ethanol, and deionizedwater washing steps. Slides were immersed in EDTA antigenretrieval buffer (1 mmol/L EDTA, 0.05% Tween 20, PH 8.0) at50�C overnight. Slides were treated with dual endogenous

Drug Targeting to Breast Cancer via Bone-Induced Integrin b3

www.aacrjournals.org Cancer Res; 77(22) November 15, 2017 6301

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 4: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

enzyme block (Dako), TBS/0.1% Tween-20 (TBST) wash buffer,and 10 minutes of serum-free protein block (Dako). Slides werestained with the following primary antibodies: anti-integrin b3antibody (D7 � 3P, 1:200, Cell Signaling Technology), anti-integrin av (ab179475, 1:500, Abcam), isotype control rabbitIgG (ab27472, Abcam) antibody, or biotinylated anti-PCNAantibody (PC10, 1:100, eBioscience). Following primary anti-body incubation, slides were extensively washed in TBST. EitherAnti-Rabbit EnVisionþ System-HRP (Dako) or Vectastain EliteABC HRP kit (Vector Laboratories) was used as the secondaryantibody, followed by Liquid DABþ (3,30-Diaminobenzidine)Substrate system (Dako), according to the manufacturer's proto-col. Nuclear hematoxylin counterstain was applied, followed bydehydration through 70% ethanol, 95% ethanol, 100% ethanol,and xylene. Slides were mounted with Cytoseal XYL (ThermoScientific).

Histologic imaging and analysisHistologic slides were imaged on either an Olympus Nano-

Zoomer 2.0-HT System or on a Zeiss AxioScan.Z1. In each exper-iment, postimage analysis was limited to changes in brightness orcontrast, gamma ¼ 1, which were applied equally to all images.Sections stained with integrin b3 or proliferating cell nuclearantigen (PCNA) were quantified using Visiopharm software,which allows for recognition and quantification of DAB-stainedtissue areas. A supervised Bayesian pixel classifier was used toclassify an image based on three distinct categories: DAB staining,hematoxylin staining (nuclei), and unstained tissue and otherbackground structures. Integrin b3 expression from each samplewas calculated within 5–10 random high-powered fields of 100–200 cells within the tumors. Values expressed as the percentage ofintegrin b3 expression (area of total DAB-positive staining) divid-ed by the tumor area in the high-powered field. PCNA prolifer-ation index was quantified by calculating the number of PCNA-positive cells divided by the total number of cells within the entirebone metastatic region. ImageJ software (NIH, Bethesda, Mary-land) was used to measure tumor area, and to quantitate osteo-clasts (OC; defined as a TRAP positive, multinucleated cell on abone surface) per millimeter of bone surface (mmB.S.) at thetumor/bone interface (N.OC/mmB.S.). Histologic analysis ofbone metastatic tumor burden (tumor area/total bone marrowarea) was calculated by measuring the tumor area within themetaphysis of the tibiofemoral joint, divided by the total marrowarea of the metaphysis.

Flow cytometry and FACSIn vitro tumor cells were lifted with 1� Versene (Invitrogen).

Ex vivo tumor cells were collected from the MFP or the bone,and prepared into single-cell suspensions for flow cytometryanalysis as described previously (20). Ex vivo cells were stainedwith PE-conjugated anti-mouse integrin b3 (1:200, clone:2C9.G2, BD Pharmingen), fixed, and permeabilized using theCytofix/Cytoperm kit (BD Biosciences) according to the man-ufacturer's protocol, and then with AlexaFluor488–conjugatedanti-human/mouse cytokeratin 18 (1:100, clone LDK18,eBioscience). Data acquisition was performed on the LSRFor-tessa (BD Biosciences) and FlowJo software version 10.1 (TreeStar) was used for analysis, and fluorescence compensationusing UltraComp eBeads (eBioscience) according to the man-ufacturer's protocol. All flow cytometry data are presented asmedian fluorescent intensity. Ex vivo flow cytometry analysis

of bone samples with insufficient number of tumor cells(<500 events) were excluded.

For FACS, in vitro tumor cells were lifted with 1� Versene(Invitrogen) and stained for surface expression of integrin b3as described above. Tumor cells were sorted into two popula-tions using a BD FACSAria-II cell sorter (BD Biosciences):integrin b3 negative (b3�) cells (based on the fluorescentintensity of unstained cells) and integrin b3-expressing(b3þ) cells. In addition, control cells sorted without b3 dis-crimination were also collected (b3-all). After sorting, eachpopulation was counted for live/dead cells, and 0.05 � 106

live tumor cells in 50-mL PBS were intracardially injected intothe 6-week-old female mice.

Pharmacologic inhibition of signaling pathwaysTumor cells were pretreated for 1 hour with pharmacologic

inhibitors: cells were pretreated for 1 hour with the followingpharmacologic inhibitors: TGFb-receptor I kinase inhibitor,specific for the site necessary for SMAD2/SMAD3 phosphory-lation (SMAD2/3i, SB431542, 20 mmol/L, Sigma-Aldrich); p38MAP kinase (MAPK) inhibitor (p38i, SB203580, 20 mmol/L,Cell Signaling Technology); MEK1/2 (MAPK/ERK Kinase)inhibitor (MEK1/2i, U0126, 20 mmol/L, Cell Signaling Tech-nology); c-Jun N-terminal kinase (JNK) inhibitor (JNKi,SP600125, 50 mmol/L, Sigma-Aldrich). After 1 hour of pre-treatment, cells were treated with 2 ng/mL of murine TGF-b1(R&D Systems) or vehicle treatment. Cells were all cultured in0.1% DMSO.

Western blot analysisWhole cell lysates from tumor cells were collected in RIPA

buffer (Cell Signaling Technology) in the presence of Halt phos-phatase inhibitor cocktail (Thermo Scientific) at 4�C. Proteinsamples were separated on 10% Mini-PROTEAN TGX polyacryl-amide gels (Bio-Rad) and transferred onto an Immobilon-Ppolyvinylidene difluoride membrane (EMD Millipore). Mem-branes were incubated with phosphorylated-SMAD2/phos-phorylated-SMAD3 (pSMAD2/pSMAD3, D27F4), totalSMAD2/SMAD3 (D7G7), integrin b3 (D7X3P), or b-actin(13E5), followed by horseradish peroxidase–conjugated anti-rabbit secondary antibody (all from Cell Signaling Technolo-gy). All antibodies were diluted and used according to themanufacturer's protocol. Bands were developed by enhancedchemiluminescence.

qPCR analysisTotal RNA from cells was isolated with the RNeasy Mini Plus

Kit (Qiagen). Complementary DNA was made using the Super-Script II first-strand synthesis system for qPCR (Invitrogen).qPCR was performed using SYBR Advantage mix (Bio-Rad) asdescribed previously (20), withmouse-specific primers formRNAgenes of interest: Itgb1 forward: 50-GCAGGTGTCGTGTTTGTGA-ATGCT-30, Itgb1 reverse: 50-ACAAGTTGGCCCTTGAAACTTGGG-30, Itgb3 forward: 50-TTCAATGCCACCTGCCTCAACAAC-30, Itgb3reverse: 50-ACGCACCTTGGCCTCGATACTAAA-30, Itgb5 forward:50-TGTTCAGCTACACAGAACTGCCCA-30, Itgb5 reverse: 50-TTT-GGAACTTGGCAAACTCTCGGC-30, Itgb6 forward: 50-TCAAAC-CTCCTTCACTCCTGCCAT-30, Itgb6 reverse: 50-TGAATCTGCTG-GTTGCTCCAGACT-30, Itgb8 forward: 50-CAACTGCATCCAG-GAGCTGA-30, Itgb8 reverse: 50-AACCAAGACGGAAGTCACGG-30, Itga2 forward: 50-AGCCCGTGATCTTTCCTAAAC-30, Itga2reverse: 50-GCAGCCACAGAGTAACCTAAA-30, Itga2b forward:

Ross et al.

Cancer Res; 77(22) November 15, 2017 Cancer Research6302

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 5: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

50-ACATTGAGGGCTTTGAGAGGCT-30, Itga2b reverse: 50-TTGC-CACAGGCAACATCACG-30, Itga3 forward: 50-GCGGAAGGA-CTGGGATTTAT-30, Itga3 reverse: 50-GATGATGTCCGTGGGATG-TAG-30, Itgav forward: 50- AGAAGACGTTGGGCCTATTG-30, Itgavreverse: 50-TGTAAGGCCACTGGAGATTTAG-30, Gapdh forward:50-AGGTCGGTGTGAACGGATTTG-30, Gapdh reverse: 50-TGTA-GACCATGTAGTTGAGGTCA-30. Target gene expression was nor-malized against the housekeeping gene GAPDH (Gapdh), anddata were analyzed using the DDCt method.

Panel of cytokines and growth factorsTumor cells were cultured for 24 hours with the following

murine factors: 2 ng/mL TGFb1 (R&D Systems), 2 ng/mLTGFb2 (R&D Systems), 50 ng/mL Sonic Hedgehog (Shh; Pepro-Tech, #315-22), 50 ng/mL WNT-3A (PeproTech, #315-20),100 ng/mL insulin-like growth factor 1 (IGF-1; PeproTech,#250-19), 50 ng/mL epidermal growth factor (EGF; PeproTech,#315-09), 50 ng/mL fibroblast growth factor 2 (FGF2 or bFGF;PeproTech, #450-33), 100 ng/mL osteopontin (OPN; Leinco,#O121), 10 ng/mL IL4 (R&D Systems), 200 ng/mL stromalcell–derived factor 1a (SDF-1a or CXCL12; Biolegend), 10 ng/mLIL6 (R&D Systems).

Patient breast cancer and matching bone metastatic biopsiesMatching primary breast and bone metastatic biopsies were

both taken at the time of metastatic diagnosis, from patientswithout detectable bone metastases at diagnosis but detectablebone metastases at least 6 months after diagnosis. Data wereobtained in accordance with the guidelines established by theWashington University Institutional Review Board (IRB#201102394) and WAIVER of Elements of Consent per 45 CFR46.116 (d). All patient information was deidentified prior tosharing with investigators. All of the human research activitiesand all activities of the IRBs designated in the WashingtonUniversity (WU) Federal Wide Assurance (FWA), regardless ofsponsorship, are guided by the ethical principles in "The BelmontReport: Ethical Principles and Guidelines for the Protection ofHuman Subjects Research of the National Commission for theProtection of Human Subjects of Biomedical and BehavioralResearch."

Tissue samples all displayed detectable tumor cells, as pre-viously determined by examination of cellular morphology bya board-certified pathologist. IHC for integrin b3 was per-formed as described in the previous section. Tumor-associatedb3 expression was semiquantitatively scored in a blindedmanner, in which the scorer did not have access to diseaseclassification or clinical annotation, using the histoscore(H-score) system: H-score ¼P

(i x %), where "i" is the stainingintensity (0–3 scale), and "%" is the percentage of tumor cellsstained at each intensity, ranging 0% to 100%. We observed anexpression pattern of tumoral b3 consistent with a previousliterature description of b3 expression in human breast cancer:primarily localized along the cell plasma membrane at sites oftumor–tumor and tumor–stroma contact (37). In addition tothe b3 expression on bone metastases, we also observed b3staining on some host cells within the primary and bone tumormicroenvironments, including tumor-infiltrating immune cells(b3 expression colocalized with CD68-stained cells, predomi-nately expressed on monocytes and macrophages), tumor-associated blood vessels, and osteoclasts, consistent with pre-vious observations (11, 12, 19, 20).

Breast cancer subtyping was based on ER,PR,HER2 status,where luminal A was ERþ,PR�,HER2� (n ¼ 18), luminal B wasERþ,PR�,HER2þ (n ¼ 7), HER2-enriched was ER�,PR�,HER2þ

(n ¼ 2), and triple-negative was ER�,PR�,HER2� (n ¼ 7). Eightbreast cancer carcinomas had incomplete genetic subtyping andwere excluded from molecular subtype analysis.

Statistical analysisAll data shown as mean with error bars representing SEM. All

sample sizes reported in the study are the minimum number ofsamples. For animal studies, sample sizes were estimated accord-ing to our previous experience. Statistical differences were ana-lyzed using either a two-tailed t test, ANOVA with Tukey test for aposteriori (post hoc)multiple comparisons, or a two-tailed unpairedt test with Bonferroni correction for a priori comparisons betweena control group and experimental treatment groups of interest.Assumptions for ANOVA and t test (independent samples,approximately normal distributions) for samples n > 5 weresufficientlymet, or used if a random sample of n� 5were selectedfrom an approximately normally distributed population. Non-normally distributed data was analyzed using a two-tailedMann–Whitney U test or a two-tailed Wilcoxon signed-rank test formatched pairs. All tests were considered significant at P < 0.05,or in case of k comparisons, P < 0.05/k. Data analyses werecompleted using Prism 6 (GraphPad Software).

ResultsBreast cancer cells overexpress integrin b3 within the bonemicroenvironment

To determine the potential of targeting avb3 at differentmetastatic sites, we evaluated breast cancer expression of integrinav or b3 at the primary site as compared with secondary metas-tases. Using the murine breast cancer cell line PyMT-Bo1 thatmodels the luminal B subtype, primary MFP tumors were estab-lished by orthotopic injection, or metastases within the bone andvisceral organs were established by intracardiac injection, intoseparate cohorts of mice. Tumors were removed 12 days postin-jection and stained for integrin b3 or integrin av expression usingIHC. Bonemetastases expressed elevated levels ofb3, as comparedwith MFP tumors or visceral metastases within the lung or kidney(Fig. 1A; Supplementary Fig. S1A). Unlike the selective upregula-tion of integrin b3 on bonemetastases, we observed expression ofintegrin av across all primary and metastatic tumors (Supple-mentary Fig. S1B). To determine whether the observed b3 expres-sion was tumor cell–specific, we used flow cytometry to quantifycell surface b3 expression. Using PyMT-Bo1 cells or murine 4T1triple-negative breast cancer cells,MFP tumors or bonemetastaseswere established by orthotopic injection or intracardiac injection,respectively. Isolated tumor cells were identified on the basis ofcytokeratin 18 (CK18þ) expression (Supplementary Fig. S1C). Inboth cell lines, tumor cells from bone metastases expressedsignificantly elevated b3 levels, as compared with MFP tumorcells (Fig. 1B and C).

To assess the translational implications of this observation,we examined b3 expression on biopsies from 42 breast cancerpatients, comparing patient-matched primary tumors to bonemetastases. IHC for b3 was semiquantitatively scored on thebasis of the extent and intensity of tumor-associated b3 expres-sion. Across nearly all patients, human bonemetastases expressedsignificantly higher levels of tumor-associated b3, as compared

Drug Targeting to Breast Cancer via Bone-Induced Integrin b3

www.aacrjournals.org Cancer Res; 77(22) November 15, 2017 6303

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 6: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

with the matching primary tumor (Fig. 2A; SupplementaryFig. S1D). Furthermore, this elevated expression of bone meta-static b3 was observed across all breast cancer patient subtypes:luminal A, luminal B, HER2-enriched, and triple-negative (Fig.2B).

Tumoral b3 is induced in the bone microenvironment andmediated by TGFb signaling

Wenext askedwhether bonemetastases express elevatedb3dueto preferential seeding and colonization of a b3hi tumor cellsubpopulation in the bone. Using FACS, PyMT-Bo1 cells weresorted into three different subpopulations based on surface b3expression: b3-negative cells (b3�), b3-expressing cells (b3þ),and control cells sorted without b3 discrimination (b3-all). ThesePyMT-Bo1 subpopulations were collected and injected intracar-dially into separate groups of mice, which all developed bonemetastases (Fig. 3A). Flow cytometry revealed that regardless ofinitial b3 status, each FACS subpopulation developed into bonemetastases that strongly expressed tumoral b3 (Fig. 3B). Thisresult demonstrates that elevated expression of bone metastaticb3 is not due to preferential seeding of b3hi tumor cells to thebone, and suggests that tumoral b3 is induced in the bonemicroenvironment.

To ascertain pathways that might be responsible for this induc-tion, we evaluated a panel of cytokines and growth factors presentin the bone microenvironment for their effect on b3 expression.Of the tested factors, only members of the TGFb family upregu-lated integrin b3 expression, observed in both PyMT-Bo1 and 4T1cells (Fig. 3C and D). To assess the specificity of TGFb-mediatedupregulation of integrin b3 as compared with other relatedintegrin subunits, we evaluated how TGFb alters expression ofintegrin av, as well as a subfamily of b-integrin subunits that can

also heterodimerizewith integrinav (integrin subunitsb1,b5,b6,and b8). In addition, we evaluated integrin aIIb (glycoprotein-IIb), which is largely restricted to platelets and megakaryocyte-lineage cells but can also heterodimerize with integrin b3 (16),and integrin a2, which facilitates adhesion to type I collagen andis upregulated on prostate cancer lines in response to TGFb (38).

Collectively, we evaluated gene expression of integrin subunitsb1, b3, b5, b6, b8, av, aIIb, and a2 on PyMT-Bo1 and 4T1 cells,following 24 hours of in vitro TGFb stimulation. Integrin b3 wasthe most upregulated subunit in both PyMT-Bo1 and 4T1 cells(Supplementary Fig. S2A and S2B). In PyMT-Bo1 cells, integrina2was the next most upregulated subunit, although 5-fold less thanthe level of b3 induction (Supplementary Fig. S2A). In 4T1 cells,b5 and a2 were the next most upregulated subunits, although 4-fold less than the level of b3 induction (Supplementary Fig. S2A).TGFb-mediated changes in av, aIIb, and the other b-integrinsubunits were either unchanged, reduced, or displayed a less-than2-fold upregulation (Supplementary Fig. S2A and S2B). Theseresults provide support for TGFb as a potent inducer of integrin b3as compared with these other integrin subunits.

TGF-b signals through TGFb receptor 1 (TGFbR1), whichcanonically phosphorylates the transcription factors SMAD2/SMAD3. In addition, TGFbR1 can activate "noncanonical" sig-naling pathways,most commonly p38MAP kinase (p38),MAPK/ERKkinase-1 and -2 (MEK1/2), and c-JunN-terminal kinase (JNK;ref. 39). To evaluate how TGFb induces b3, tumor cells weretreated with TGFb in combination with pharmacologic inhibitorsfor p38, MEK1/2, JNK, or TGFbRI kinase activity at the sitespecific for SMAD2/SMAD3 phosphorylation (see Fig. 4 fordetails). In both PyMT-Bo1 and 4T1, only pharmacologic inhi-bition of SMAD2/SMAD3 phosphorylation completely ablatedb3 upregulation. This was observed by flow cytometry for surface

Figure 1.

Integrinb3 is preferentially inducedonbreast cancer cellswithin the bonemicroenvironment.A, IHC forb3onPyMT-Bo1mammary tumors ormetastases, establishedby orthotopic or intracardiac injection into separate cohorts of mice, respectively. Tumors were removed 12 days postinjection. Representative images (left),quantification of DAB-stained tumors (right), n ¼ 4 per group. Scale bar, 50 mm. One-way ANOVA with Tukey post hoc test; ���� , P < 0.0001. B and C, Mammarytumors or bonemetastaseswere established as described previously. Isolated tumor cells were identified based on cytokeratin 18 expression (CK18þ) and evaluatedby flow cytometry for surface b3 expression. Left, representative samples. B, PyMT-Bo1 cells, 13 days postinjection, n ¼ 3 mammary, n ¼ 4 bone. C, 4T1 cells,15 days postinjection, n ¼ 5 mammary, n ¼ 4 bone. Two-tailed unpaired t test; �� , P < 0.01; ��� , P < 0.001. Data presented as mean � SEM. MFI, median fluorescentintensity.

Ross et al.

Cancer Res; 77(22) November 15, 2017 Cancer Research6304

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 7: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

b3 expression (Fig. 4A and B) and qPCR analysis for b3 mRNAlevels (Supplementary Fig. S3A). Western blot analysis confirmedselective inhibitionof SMAD2/SMAD3phosphorylation (Fig. 4C)and suppression of total b3 protein (Fig. 4D) with TGFbRI kinaseinhibition. To determine whether TGFb signaling was responsiblefor tumoral b3 induction in vivo, mice bearing human MDA-MB-231 triple-negative breast cancer bone metastases were treateddaily with a pharmacologic TGFbRI kinase inhibitor. TGFbRIkinase inhibition significantly suppressed b3 expression onMDA-MB-231 breast cancer bone metastases (Fig. 4E), demon-strating the role of TGFb signaling inb3 inductiononbreast cancercells within the bone microenvironment. TGFbRI kinase inhibi-tion resulted in an approximately 1.6-fold reduction inMDA-MB-231 bone metastatic tumor burden (Supplementary Fig. S3B),consistent with a previous report (30).

Integrin avb3–targeted micelle nanoparticles colocalize withbreast cancer bone metastases

To evaluate whether bone metastatic induction of b3 couldbe therapeutically exploited, we sought to examine the potentialof nanotherapy targeting integrin avb3. We used single-lipidphospholipid/polysorbate-80 micelle nanoparticles (MPs,�12.5 nm), coated with a well-characterized quinolone nonpep-tide targeting ligand with enhanced specificity for activated avb3(Supplementary Fig. S4A). The specificity of this targeting ligandfor integrin avb3 has been demonstrated through a variety ofassays both in vitro and in vivo (see Materials and Methods for

details; refs. 32–36). To test the effectiveness of integrin avb3 as atarget for enhancing drug delivery to bone metastases, rhoda-mine-labeledavb3-targetedMPs (avb3-MPs) or nontargetedMPswere administered intravenously into mice bearing PyMT-Bo1bonemetastases. Immunofluorescent analysis of bonemetastasesrevealed strong colocalization ofavb3-MPswith bonemetastases,while nontargeted MPs displayed only minor colocalization(Fig. 5A). Furthermore, accumulation of either avb3-MPs or MPswithin tumor-free bone marrow was negligible (SupplementaryFig. S4B). Together, these results demonstrate the targeting poten-tial and specificity of avb3-MPs for bone metastases.

Next, we sought to evaluate the therapeutic efficacy of avb3-MP–mediateddrug delivery.Nanotherapeutics often suffer fromapremature loss of drug payloads during circulation and dimin-ished intracellular drug bioavailability due to endosomal entrap-ment (40). To overcome these problems, a lipase-labile phos-pholipid–prodrug concept was employed, in which drug cargo iscoupled to the Sn2 acyl-chain of phosphatidylcholine. This phos-pholipid-prodrug can be stably incorporated into the MP mem-brane during self-assembly and provides drug retention duringcirculation (35). Upon avb3-MP binding to activated avb3 ontarget cells, a hemifusion complex forms between MPs and theplasma membrane, enabling endocytosis-independent "contact-facilitated drug delivery" (see Fig. 5B for details; refs. 36, 41).

The chemotherapeutic agent selected for nanoparticle deliverywas docetaxel, a potentmicrotubule inhibitor employed as a first-line agent against breast cancer (42).Wemodified docetaxel as an

Figure 2.

IHC for b3 on patient-matched primary breast cancer and bone metastatic biopsies. A, Semiquantitative analysis of the extent and intensity of tumor-associatedb3 expression using the histoscore (H-score) system. n ¼ 42 matched-pairs, two-tailed Wilcoxon signed-rank test, ���� , P < 0.0001. B, Subdivision of A bymolecular subtype (see Materials and Methods for details). Right, representative images of patient-matched primary tumors and bone metastases. Scalebar, 50 mm. Two-tailed paired t test; � , P < 0.05; �� , P < 0.01; ���, P < 0.001; ���� , P < 0.0001. Data presented as mean � SEM.

Drug Targeting to Breast Cancer via Bone-Induced Integrin b3

www.aacrjournals.org Cancer Res; 77(22) November 15, 2017 6305

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 8: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

Sn2 lipase-labile phospholipid-prodrug (DTX-PD) to enableavb3-MP delivery (Fig. 5B; ref. 32). In vitro, PyMT-Bo1 cells weremarkedly sensitive to both docetaxel and DTX-PD, with a bio-equivalent IC50 of 5.5 nmol/L (Fig. 5C). Unfortunately, che-motherapeutics such as docetaxel rarely eradicate bonemetastasesin human patients (3, 4) and can induce off-target toxicity such ashair loss, neuropathy, pancytopenia, and liver toxicity (42). Totest the efficacy of docetaxel against PyMT-Bo1 cells in vivo, micewere intracardially injected with PyMT-Bo1 cells. BLI analysisconfirmed equivalent establishment of bone metastases on day3 post-intracardiac injection, and mice bearing PyMT-Bo1 metas-tases received intravenous docetaxel treatments beginning on day4 post-intracardiac injection (see Fig. 5D). Animals experienceddose-dependent docetaxel-induced hematologic toxicity (Supple-mentary Fig. S5A), yet metastatic tumor burden in the liver,kidney, or bone was not attenuated by docetaxel treatment, andlung metastatic tumor burden was only significantly reduced atthe highest docetaxel dose of 20 mg/kg/week (Fig. 5E). Thus,while docetaxel was potently suppressive against PyMT-Bo1 invitro, docetaxel displayed limited efficacy against PyMT-Bo1metastases in vivo.

avb3-MP–Mediated drug delivery of docetaxel attenuatesbone metastases

To evaluate the therapeutic potential of avb3-targeted deliveryof DTX-PD against PyMT-Bo1 bonemetastases, mice were treatedwith docetaxel, or an equimolar amount of DTX-PD encapsulatedby avb3-MP (avb3-MP/DTX-PD). In parallel, two control nano-

particle treatmentswere evaluated to assess the in vivo specificity ofavb3-MP targeting, and the specificity of docetaxel-mediatedtumor-suppressive effects: nontargeted MP/DTX-PD and cargo-free avb3-MP without DTX-PD cargo, respectively.

As before, BLI analysis on day 3 post-intracardiac injectionconfirmed PyMT-Bo1 bone metastatic establishment and treat-ment began on day 4, with a cumulative dose of 5.55 mg/kgdocetaxel or equimolar dose of MP-encapsulated DTX-PD(see Fig. 5D). On day 12, ex vivo BLI analysis of bone metastasesrevealed that avb3-MP/DTX-PD significantly attenuated bonemetastatic tumor burden, while no significant attenuation wasobserved with docetaxel treatment (Fig. 6A and B). Consistentwith this result, histologic analysis of bone tumor burden (Fig.6C) and X-ray analysis of tumor-associated bone loss (Fig. 6D)revealed a significant reductionwithavb3-MP/DTX-PD treatmentas compared with saline, and no significant attenuation withdocetaxel. BLI analysis of visceral metastases within the liver,lungs, kidneys, and brain revealed that neither avb3-MP/DTX-PD nor docetaxel resulted in significant attenuation (Supplemen-tary Fig. S5B), supporting our previous observations concerningbone-specific b3 expression and resistance of PyMT-Bo1 metas-tases to docetaxel.

Hepatotoxicity was tested by serum liver function tests foraspartate transaminase (AST) and alanine transaminase (ALT).docetaxel treatment resulted in elevated levels of AST and ALToutside of the normal ranges, while avb3-MP/DTX-PD treatmentdisplayed no evidence of hepatotoxicity (Fig. 6E). Both renalfunction, assessed by blood urea nitrogen (BUN), and whole

Figure 3.

Bone microenvironment and TGFb stimulation induce tumoral b3. A, In vitro PyMT-Bo1 cells were FACS sorted into three groups based on basal b3 expression:b3�, b3þ, and control b3-all. Immediately after collection, cells were injected (intracardially) into separate groups of mice. B, Thirteen days postinjection,isolated bone metastatic PyMT-Bo1 cells were identified by CK18þ and evaluated by flow cytometry for surface b3 expression. Left, representative samples.Right, n¼ 4 b3�; n¼ 4 b3-all; n¼ 5 b3þ. One-way ANOVA with Tukey post hoc test; ns, not significant. C and D, qPCR analysis of b3 (Itgb3)mRNA expression byPyMT-Bo1 or 4T1 cells cultured in vitro, following 24 hours stimulation with the listed factors (see Materials and Methods for details). One of two biological replicates,each in technical duplicate. One-way ANOVA with Tukey post hoc test; ����, P < 0.0001. Data presented as mean � SEM.

Ross et al.

Cancer Res; 77(22) November 15, 2017 Cancer Research6306

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 9: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

blood counts were within the normal range among mice treatedwith saline, docetaxel, oravb3-MP/DTX-PD (Fig. 6E; Supplemen-tary Fig. S5C).

Neither nontargeted MP/DTX-PD nor cargo-free avb3-MPtreatment significantly altered bone metastatic tumor burden orbone destruction, demonstrating the necessity ofavb3-MP target-ing and the specificity of docetaxel in mediating the tumor-suppressive effects of this nanotherapy (Supplementary Fig.S6A–S6D). Control avb3-MP–treated mice displayed elevatedlevels of AST, but ALT, renal, and hematologic toxicities werewithin the normal range for avb3-MP and nontargeted MP/DTX-PD treatments (Supplementary Fig. S6E and S6F).

To evaluate the potential mechanism of avb3-MP/DTX-PDanti-bone metastatic activity, we quantified osteoclast number atthe tumor–bone interface, which was unchanged between saline,docetaxel, and avb3-MP/DTX-PD treatments (Fig. 6F), demon-strating that avb3-MP/DTX-PD attenuates tumor-induced boneloss independent of inhibition of osteoclast formation or osteo-clast killing. To assesswhether tumor cells responded to the effectsof the chemotherapy, we measured tumor proliferation by IHCfor proliferating cell nuclear antigen (PCNA). Bone metastatictumor cells treated with avb3-MP/DTX-PD displayed a signifi-cantly lower proliferation index than saline or docetaxel (Fig. 6G).Collectively, this work provides support for safer, more effective

drug delivery against breast cancer bone metastases throughexploitation of avb3 expression.

DiscussionBreast cancer bone metastases often irreversibly damage the

skeleton, severely impacting quality of life and overall survival(2). Treating bone metastases has proven difficult, in part dueto the bone microenvironment's status as a chemoprotectiveniche (7), and coupled with inefficient delivery of drug totumor cells in bone (43). While drug or nanoparticle conju-gation to hydroxyapatite-avid molecules like bisphosphonateshave provided positive results against mouse models of bonemetastases (13, 14), they lack direct specificity for tumor cells(8, 15). In this study, we provide new evidence for integrin b3as a molecular target on breast cancer bone metastases in miceand humans, and demonstrate the potential for improvingtherapeutic efficacy against this metastatic site via avb3-tar-geted approaches.

We first investigated differences in the physiologic expressionof integrin b3onbreast cancer tumors growing in different organs,in search of a tumor target on cells within the bone microenvi-ronment. We demonstrated that orthotopic injection of murinebreast cancer cell lines into the MFP establishes primary tumors

Figure 4.

TGFb induces tumoral b3 through TGFbRI-mediated phosphorylation of SMAD2/SMAD3. A and B, In vitro tumor cells treated with either 2 ng/mL TGF-b1 orDMSOcontrol in the presence of a pharmacologic inhibitor SMAD2/3i (20mmol/L SB431542),MEK1/2i (20mmol/LU0126), p38i (20mmol/L SB203580), or JNKi (50mmol/LSP600125). After 48 hours, flow cytometry for surface b3 expression was evaluated on PyMT-Bo1 cells (A) or 4T1 cells (B). Left, representative experiment; right,n ¼ 3 biological replicates. One-way ANOVA with Tukey post hoc test, with denoted significance in relation to DMSO control; � , P < 0.05; ��� , P < 0.001;���� , P < 0.0001.C andD,Western blot analysis of in vitro PyMT-Bo1 cells treated as described previously, for 3 hours (C) or 24 hours (D). E,Mice intracardially injectedwithMDA-MB-231 cells were treated daily with a TGFbRI kinase inhibitor of SMAD2/3 phosphorylation (SD-208, 60 mg/kg/d) or vehicle control (1% methylcellulose)for 28 days. IHC for b3 with representative images (left) and quantification of DAB-stained bone metastases, n¼ 4 (right). Scale bar, 100 mm. Two-tailed unpaired t test;��� , P < 0.001. Data presented as mean � SEM.

Drug Targeting to Breast Cancer via Bone-Induced Integrin b3

www.aacrjournals.org Cancer Res; 77(22) November 15, 2017 6307

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 10: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

withweak b3 expression. Intracardiac injection establishesmetas-tases within the lung and kidney with similarly weak b3 expres-sion, but establishes bone metastases with significantly strongerb3 expression, observed by both IHC and flow cytometry oftumoral surface expression. In contrast, integrin av was stronglyexpressed on tumors at all evaluated sites. We employed thewidely published intracardiac injection model to establish exper-imental bone and visceral metastases in these experiments. Whilethismodel does not recapitulate all steps of themetastatic cascade,

circulating tumor cells still need to extravasate from the vessels toestablish within the bone and other organs.

We also evaluated b3 expression on 42 human bonemetastatictissue samples, as compared withmatching primary breast cancerbiopsies from the same patient. Across nearly all patients, tumor-associated b3 expression was significantly higher on bone metas-tases. Furthermore, we show for the first time that b3 is highlyexpressed on breast cancer bone metastases of all subtypes,including those with the worst overall survival that have proven

Figure 5.

Micelle nanoparticles targeting integrin avb3 colocalize with breast cancer bone metastases and are designed to carry the chemotherapeutic docetaxel. A, Micebearing PyMT-Bo1 bone metastases were intravenously injected with rhodamine-labeled (red) nontargeted MPs or avb3-MPs. After 3 hours, bones were removedand fresh-frozen sections were immunofluorescently imaged. DAPI nuclear counterstain, blue. Left, representative images; right, quantification of fluorescentcolocalizationwith the bonemetastases, n¼ 3 per group. Scale bar, 200 mm. Two-tailed unpaired t test; ��� , P <0.001.B, Schematic ofavb3-MP–mediated "contact-facilitated drug delivery." Upon hemifusion, phospholipid-conjugated docetaxel-prodrug (DTX-PD) transfers to the target cell's plasmamembrane, where bioactivedocetaxel is enzymatically liberated by phospholipases and released directly into the cytoplasm. C, In vitro PyMT-Bo1 cell viability via MTT assay after 72 hours ofdocetaxel or DTX-PD treatment. n ¼ 2 biological replicates, each in technical triplicate. D, Treatment schematic of mice intracardially injected with PyMT-Bo1cells. Representative ex vivo BLI of tumor-bearing organs on day 12. E, Following the schematic in Fig. 5D, mice were treated with increasing concentrationsof docetaxel, with the cumulative dose as indicated. Day 12 ex vivo PyMT-Bo1 metastatic tumor burden by BLI analysis, n ¼ 6. Bone, kidney, lung: two-tailedunpaired t test. Liver, two-tailed Mann–Whitney U test. All with Bonferroni correction for a priori multiple comparisons between control 0 mg/kg and eachdocetaxel treatment group, � , P < (0.05/4); �� , P < (0.01/4); ns, not significant. Data presented as mean � SEM.

Ross et al.

Cancer Res; 77(22) November 15, 2017 Cancer Research6308

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 11: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

Figure 6.

Integrin avb3–targeted nanotherapy attenuates breast cancer bone metastases. Following the schematic described in Fig. 5D, mice were treated with either acumulative dose of 5.55 mg/kg docetaxel, an equimolar dose of docetaxel-prodrug encapsulated by avb3-MP (avb3-MP/DTX-PD), or saline. Analysescompleted on samples collected on day 12 postinjection. A, Representative in vivo BLI of mice bearing PyMT-Bo1 bone metastases, matching day 3 andday 12 postinjected mice from each group B, Ex vivo PyMT-Bo1 bone metastatic tumor burden by BLI analysis, n ¼ 9. C, Histologic analysis of tumorburden within the tibiofemoral joint, n ¼ 4. D, X-ray analysis of osteolytic bone destruction, n ¼ 9. B–D, Scale bar, 1 mm, two-tailed unpaired t test withBonferroni correction for a priori multiple comparisons between saline and each experimental treatment, � , P < (0.05/2). E, Serum chemistry analysis of BUN,AST, and ALT, n¼ 5. Gray range illustrates the normal reference range (mean� 2SD) for female C57BL/6 mice bearing PyMT-Bo1 metastases. F, TRAP stainingfor quantification of osteoclast number per millimeter of bone surface (N.OC/mmB.S.) at the tumor (T), bone (B) interface, n ¼ 4. G, Quantification ofPCNA-positive cells within the bone metastatic region, n ¼ 3. F and G, Scale bar, 50 mm, one-way ANOVA with Tukey post hoc test; �, P < 0.05; �� , P < 0.01;ns, not significant. All images are representative and data presented as mean � SEM.

Drug Targeting to Breast Cancer via Bone-Induced Integrin b3

www.aacrjournals.org Cancer Res; 77(22) November 15, 2017 6309

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 12: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

difficult to target (44), supporting the translational potential oftargeting b3 in bone metastatic breast cancer patients.

On the basis of previous studies demonstrating that ectopicoverexpression of b3 enhances tumor cell establishment in bone(26, 28), we considered the possibility that breast cancer bonemetastases display elevated b3 due to preferential bone coloni-zation of a tumor cell subpopulation with high basal b3 expres-sion. We established bone metastases from FACS-sorted tumorcell subpopulations, which were initially positive or negative forbasal b3 expression. Here, we demonstrated that initial b3 expres-sion is dispensable for establishing breast cancer bone metastaseswith elevated b3 expression. This finding does not contradict therole of tumoral avb3 in enhancing bone colonization (26, 28), asthere are numerous ligands for avb3 within the bone microen-vironment (17, 18); rather, it indicates that tumoral b3 can beinduced during the establishment of bone metastases, suggestingthat b3 expression might be regulated by the bone microenvi-ronment itself.

The bone is a natural reservoir for TGFb, which is stored in thematrix as an inactive latent complex that must be liberated forbioactivation (45). Bioactive TGFb plays a critical role in both thehomeostatic and pathologic bone microenvironment, regulatingosteolytic destruction and pathologic muscle weakness (30, 46).TGFb has been shown to induce b3 expression on normal andtransformed mammary epithelial cells (47–49), but the physio-logic regulation of tumoral b3 in vivo has not been explored. Here,we demonstrated that TGFb strongly induces tumoral b3, andanalysis of other integrin subunits of interest revealed that TGFbinduces integrin b3 more strongly than any other. Furthermore,we determined that TGFb induces b3 through TGFbRI phosphor-ylation of SMAD2/SMAD3, evaluated at the transcriptional, totalprotein, and cell surface expression level. And for the first time, wedemonstrate that TGFb signaling is required for integrin b3induction by breast cancer cells within the bone.

Wewere surprised by the lack of tumoral b3upregulation in theMFP tumor or visceral metastases, where TGFb can be present.This observation suggests that other microenvironments eitherlack a sufficient amount of bioactive TGFb necessary for tumoralb3 induction, or contain inhibitors of TGFb or SMAD2/SMAD3signaling that prevent TGFb-mediated induction of tumoral b3.Studies are in progress to define themolecular mechanism under-lying the low expression of tumoral b3 outside of the bone.

Selective elevation of b3 on bone metastases prompted us toexplore avb3-targeted nanotherapy as a means to overcome thebarriers limiting effective treatment. We selected phospholipid/polysorbate-80 MPs for their small size (�12.5 nm) and uniquemechanism of drug delivery (see Fig. 5B for details), whichovercomes two problems that often hinder effective nanotherapy:poor intracellular drug bioavailability and premature loss of drugduring circulation (40). Micelle nanoparticles deliver phospho-lipid–prodrug cargo directly to the target cell's plasmamembrane,where bioactive free-drug is enzymatically liberated directly intothe cytoplasm by phospholipase activity (36, 41). Furthermore,phospholipid–prodrug is stably incorporated into the MP-singlelipid layer, thereby minimizing passive drug loss during circula-tion (35). Administering rhodamine-labeled MPs or avb3-MPsinto mice bearing breast cancer bone metastases resulted instronger avb3-MP colocalization with bone metastases, com-pared with a 6.5-fold reduction in colocalization with nontar-getedMPs. To examine the importance of MP's small size in bonemetastatic colocalization, we also evaluated the colocalization

potential of significantly larger avb3-targeted perfluorocarbonnanoparticles (�250 nm) (35). We observed that fluorescentlylabeled avb3-targeted perfluorocarbon nanoparticles were muchless effective at localizing to the center of bone metastases (datanot shown). These observations prompted us to proceed withavb3-MP–mediated drug delivery, and we selected the chemo-therapeutic docetaxel as cargo for lipase-labile phospholipid–prodrug modification (DTX-PD).

While PyMT-Bo1 breast cancer cells are equally sensitive toboth docetaxel and DTX-PD in vitro, PyMT-Bo1 metastases werestrikingly resistant in vivo, with bone metastatic tumor burdenessentially unaffected by docetaxel treatment even at the highest,most toxic dose tested (20 mg/kg/week). These observationshighlight the difficulty of achieving an inhibitory dose of che-motherapy at metastatic sites in vivo. To test the efficacy of MP-encapsulated DTX-PD, mice bearing PyMT-Bo1 metastases weretreated with a suboptimal docetaxel dose (5.55 mg/kg/week),3.6-fold lower than the previously tested 20 mg/kg/week doce-taxel dose, which did not significantly attenuate bonemetastases.At this suboptimal dose, we observed a significant attenuation ofbone metastatic tumor burden and tumor-associated bone losswith avb3-MP/DTX-PD treatment, as compared with no signif-icant attenuationby equimolar docetaxel treatment. Furthermore,neither cargo-free avb3-MP treatment nor nontargeted MP/DTX-PD treatment significantly altered bone metastatic tumor burdenor osteolytic bone destruction. The observation that micellecoating with the avb3-targeting ligand was necessary for bonemetastatic colocalization, coupled with fact that avb3-MP/DTX-PD failed to attenuate metastases outside of b3-expressingbone metastases, provides additional support for the in vivospecificity of the avb3-targeting ligand for integrin avb3, in linewith previous reports (32–36).

To examine the potential mechanism of avb3-MP–mediatedactivity, we quantified osteoclast number at the tumor/boneinterface, which is reflective of osteoclast formation and servesas a useful indicator of bone resorption. Previous studies havefound that osteoclast formation is markedly sensitive to inhibi-tion by docetaxel, even at nanomolar concentrations (50). Oste-oclast number at the tumor/bone interface was similar betweenavb3-MP/DTX-PD and saline treatment groups, demonstratingthatavb3-MP/DTX-PD treatment attenuates bonemetastases andtumor-associated bone loss independent of inhibiting osteoclastformation or direct osteoclast killing. It is possible that thereremains an effect of avb3-MP/DTX-PD treatment on osteoclastfunction that was not revealed by histomorphometric analyses.

We evaluated tumor cell proliferation through expressionquantification of PCNA, a DNA polymerase processivity factorrequired for DNA synthesis during replication. Docetaxel inhibitstumor proliferation through effects on microtubule stabilizationand disruption of mitotic spindle assembly (42). We found thatavb3-MP/DTX-PD treatment significantly decreased PCNAexpression on bone-residing tumor cells, as compared with salineand docetaxel treatment. Taken together with the osteoclasthistomorphometry result, these observations suggest that avb3-MP/DTX-PD attenuated bone metastases due to enhanced ther-apeutic efficacy against bone-residing tumor cells.

We observed that docetaxel administration resulted inincreased hepatotoxicity, while hematologic values and liverfunction tests remained within normal limits following avb3-MP/DTX-PD treatment. This work thus provides support for safer,more effective drug delivery against bone metastases through

Ross et al.

Cancer Res; 77(22) November 15, 2017 Cancer Research6310

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 13: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

exploitation of avb3 expression on breast cancer cells withinbone.Going forward,avb3-targeted drug cargo could bemodifiedto fit specific treatment needs.

We recognize the limitation of avb3-targeted nanotherapyagainst breast cancer metastases outside of the bone. Never-theless, bone metastases occur in approximately 70% of met-astatic breast cancer patients and represent the only metastaticsite in approximately 30% of patients, suggesting that a sub-stantial number of patients could benefit from this approach(1, 44). Collectively, we provide support for integrin avb3-targeted drug delivery as a bone-specific therapeutic strategy toaddress the unmet clinical need for effective treatments againstbreast cancer bone metastases.

Disclosure of Potential Conflicts of InterestD. Pan is a cofounder and CTO at Innsight Tech, Inc. and Kalocyte and has

ownership interest (including patents) in Innsight Tech andKalocyte. T.A.Guiseis a consultant/advisory board member for Bayer and has received experttestimony from Novartis. No potential conflicts of interest were disclosed bythe other authors.

Authors' ContributionsConception and design: M.H. Ross, D. Pan, X. Su, T. Walsh, G.M. Lanza,K.N. WeilbaecherDevelopment of methodology: M.H. Ross, A.K. Esser, A.H. Schmieder, G. Hu,D. Pan, X. Su, Y. Xu, D.V. Novack, T. Walsh, G.M. Lanza, K.N. WeilbaecherAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): A.K. Esser, G.C. Fox, A.H. Schmieder, X. Yang, X. Su,Y. Xu, D.V. Novack, T. Walsh, G.A. Colditz, G.H. Lukaszewicz, J.A.J. Fitzpatrick,D.L. Waning, K.S. Mohammad, T.A. Guise, K.N. WeilbaecherAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):M.H. Ross, A.K. Esser, G.C. Fox, A.H. Schmieder,J. Novack, J.A.J. Fitzpatrick, D.L. Waning, T.A. Guise, G.M. Lanza,K.N. WeilbaecherWriting, review, and/or revision of the manuscript: M.H. Ross, A.K. Esser,G.C. Fox, A.H. Schmieder, G.Hu,D. Pan, X. Su, Y. Xu,D.V.Novack,G.A. Colditz,G.H. Lukaszewicz, J.A.J. Fitzpatrick, K.S. Mohammad, T.A. Guise, G.M. Lanza,K.N. Weilbaecher

Administrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): M.H. Ross, G. Hu, X. Su, Y. Xu, T. Walsh,G.H. Lukaszewicz, E. Cordell, K.N. WeilbaecherStudy supervision: D. Pan, K.N. Weilbaecher

AcknowledgmentsWe thank Sheila Stewart, Daniel Link, Roberta Faccio, Fanxin Long, David

DeNardo, and Vivek Arora, Jingyu Xiang, Takayuki Kobayashi, and BhavnaMurali for their valuable suggestions. We thank Lynne Marsala, Julie Prior,Crystal Idleburg, and Samantha Coleman for their expert technical assistance.

Grant SupportThis research was supported in whole or part by grants from the CA154737

(K.N. Weilbaecher/D. Pan/G.M. Lanza), CA100730 (K.N. Weilbaecher),CA097250 (K.N. Weilbaecher), HL122471 (G.M. Lanza), HL112518(G.M. Lanza), HL113392 (G.M. Lanza), HHSN26820140042C (G.M. Lanza),P30CA091842 (G.A. Colditz), CA143057 (T.A. Guise), CA69158 (T.A. Guise),and training grants 5T32GM007067-39 (M.H. Ross), T32AR060719 (M.H.Ross), 5T32CA113275-07 (A.K. Esser), and GM07200 (G.C. Fox). Additionalfunding support provided by grants from the St. Louis Men's Group AgainstCancer and the Siteman Cancer Center. Imaging and analysis of human breastcancer breast and bone biopsies was performed on the Zeiss Axio Scan.Z1through the use of Washington University Center for Cellular Imaging(WUCCI), supported by Washington University School of Medicine, the Chil-dren's Discovery Institute of Washington University and St. Louis Children'sHospital (CDI-CORE-2015-505), the National Institute for Neurological Dis-orders and Stroke (NS086741), and by the Foundation for Barnes JewishHospital. Technical support was provided by the Washington University Mus-culoskeletal Research Center (P30AR057235), the Hope Center Alafi Neuro-imaging Lab (Shared Instrumentation Grant S10 RR027552), the MolecularImaging Center at Washington University (P50 CA094056), and the St. LouisBreast Tissue Registry (funded by The Department of Surgery at WashingtonUniversity School of Medicine).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 3, 2017; revised July 26, 2017; accepted August 24, 2017;published OnlineFirst August 30, 2017.

References1. Buijs JT, van der Pluijm G. Osteotropic cancers: from primary tumor to

bone. Cancer Lett 2009;273:177–93.2. ColemanRE.Clinical features ofmetastatic bonedisease and risk of skeletal

morbidity. Clin Cancer Res 2006;12:6243s–9s.3. Niikura N, Liu J, Hayashi N, Palla SL, Tokuda Y, Hortobagyi GN, et al.

Treatment outcome and prognostic factors for patients with bone-onlymetastases of breast cancer: a single-institution retrospective analysis.Oncologist 2011;16:155–64.

4. Ahn SG, Lee HM, Cho SH, Lee SA, Hwang SH, Jeong J, et al. Prognosticfactors for patientswithbone-onlymetastasis in breast cancer. YonseiMed J2013;54:1168–77.

5. Suva LJ, Washam C, Nicholas RW, Griffin RJ. Bone metastasis:mechanisms and therapeutic opportunities. Nat Rev Endocrinol 2011;7:208–18.

6. Croucher PI,McDonaldMM,Martin TJ. Bonemetastasis: the importance ofthe neighbourhood. Nat Rev Cancer 2016;16:373–86.

7. Meads MB, Hazlehurst LA, Dalton WS. The bone marrow microenviron-ment as a tumor sanctuary and contributor to drug resistance. Clin CancerRes 2008;14:2519–26.

8. Dang L, Liu J, Li F, Wang L, Li D, Guo B, et al. Targeted delivery systems formolecular therapy in skeletal disorders. Int J Mol Sci 2016;17:428.

9. Tautzenberger A, Kovtun A, Ignatius A. Nanoparticles and their potentialfor application in bone. Int J Nanomedicine 2012;7:4545–57.

10. Cole LE, Vargo-Gogola T, Roeder RK. Targeted delivery to bone andmineral deposits using bisphosphonate ligands. Adv Drug Deliv Rev2016;99:12–27.

11. Bakewell SJ, Nestor P, Prasad S, Tomasson MH, Dowland N, Mehrotra M,et al. Platelet and osteoclast beta3 integrins are critical for bonemetastasis.Proc Natl Acad Sci U S A 2003;100:14205–10.

12. Morgan EA, Schneider JG, Baroni TE, Uluckan O, Heller E, HurchlaMA, et al. Dissection of platelet and myeloid cell defects byconditional targeting of the beta3-integrin subunit. FASEB J 2010;24:1117–27.

13. Miller K, Clementi C, Polyak D, Eldar-Boock A, Benayoun L, Barshack I,et al. Poly(ethylene glycol)-paclitaxel-alendronate self-assembled micellesfor the targeted treatment of breast cancer bone metastases. Biomaterials2013;34:3795–806.

14. Ye WL, Zhao YP, Li HQ, Na R, Li F, Mei QB, et al. Doxorubicin-poly(ethylene glycol)-alendronate self-assembled micelles for targeted therapyof bone metastatic cancer. Sci Rep 2015;5:14614.

15. Carbone EJ, Rajpura K, Allen BN, Cheng E, Ulery BD, Lo KW. Osteotropicnanoscale drug delivery systems based on small molecule bone-targetingmoieties. Nanomedicine 2016;13:37–47.

16. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell2002;110:673–87.

17. Desgrosellier JS, Cheresh DA. Integrins in cancer: biological implicationsand therapeutic opportunities. Nat Rev Cancer 2010;10:9–22.

18. Schneider JG, Amend SR, Weilbaecher KN. Integrins and bone metas-tasis: integrating tumor cell and stromal cell interactions. Bone2011;48:54–65.

19. Brooks PC, Clark RA, ChereshDA. Requirement of vascular integrin alpha vbeta 3 for angiogenesis. Science 1994;264:569–71.

Drug Targeting to Breast Cancer via Bone-Induced Integrin b3

www.aacrjournals.org Cancer Res; 77(22) November 15, 2017 6311

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 14: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

20. Su X, Esser AK, Amend SR, Xiang J, Xu Y, Ross MH, et al. Antagonizingintegrin beta3 increases immunosuppression in cancer. Cancer Res 2016;76:3484–95.

21. Steeg PS. Targeting metastasis. Nat Rev Cancer 2016;16:201–18.22. Kim KB, Prieto V, Joseph RW, Diwan AH, Gallick GE, Papadopoulos NE,

et al. A randomized phase II study of cilengitide (EMD121974) in patientswith metastatic melanoma. Melanoma Res 2012;22:294–301.

23. Alva A, Slovin S,Daignault S, CarducciM,Dipaola R, PientaK, et al. Phase IIstudy of cilengitide (EMD 121974, NSC 707544) in patients with non-metastatic castration resistant prostate cancer, NCI-6735. A study by theDOD/PCF prostate cancer clinical trials consortium. Invest New Drugs2012;30:749–57.

24. Ley K, Rivera-Nieves J, SandbornWJ, Shattil S. Integrin-based therapeutics:biological basis, clinical use and new drugs. Nat Rev Drug Discov 2016;15:173–83.

25. Liapis H, Flath A, Kitazawa S. Integrin alpha V beta 3 expression by bone-residing breast cancer metastases. Diagn Mol Pathol 1996;5:127–35.

26. Zhao Y, Bachelier R, Treilleux I, Pujuguet P, Peyruchaud O, Baron R, et al.Tumor alphavbeta3 integrin is a therapeutic target for breast cancer bonemetastases. Cancer Res 2007;67:5821–30.

27. Page JM, Merkel AR, Ruppender NS, Guo R, Dadwal UC, Cannonier SA,et al. Matrix rigidity regulates the transition of tumor cells to a bone-destructive phenotype through integrin beta3 and TGF-beta receptor typeII. Biomaterials 2015;64:33–44.

28. Sloan EK, Pouliot N, Stanley KL, Chia J, Moseley JM, Hards DK, et al.Tumor-specific expression of alphavbeta3 integrin promotes spontaneousmetastasis of breast cancer to bone. Breast Cancer Res 2006;8:R20.

29. Soodgupta D, Pan D, Cui G, Senpan A, Yang X, Lu L, et al. Small moleculeMYC inhibitor conjugated to integrin-targeted nanoparticles extends sur-vival in a mouse model of disseminated multiple myeloma. Mol CancerTher 2015;14:1286–94.

30. Waning DL, Mohammad KS, Reiken S, Xie W, Andersson DC, John S, et al.Excess TGF-beta mediates muscle weakness associated with bone metas-tases in mice. Nat Med 2015;21:1262–71.

31. Xiang J, Hurchla MA, Fontana F, Su X, Amend SR, Esser AK, et al. CXCR4protein epitope mimetic antagonist POL5551 disrupts metastasis andenhances chemotherapy effect in triple-negative breast cancer. Mol CancerTher 2015;14:2473–85.

32. Pan D, Schmieder AH, Wang K, Yang X, Senpan A, Cui G, et al. Anti-angiogenesis therapy in the Vx2 rabbit cancermodelwith a lipase-cleavableSn 2 taxane phospholipid prodrug using alpha(v)beta(3)-targeted ther-anostic nanoparticles. Theranostics 2014;4:565–78.

33. Sadeghi MM, Krassilnikova S, Zhang J, Gharaei AA, Fassaei HR, Esmailza-deh L, et al. Detection of injury-induced vascular remodeling by targetingactivated alphavbeta3 integrin in vivo. Circulation 2004;110:84–90.

34. Harris TD, Kalogeropoulos S, Nguyen T, Liu S, Bartis J, Ellars C, et al.Design, synthesis, and evaluation of radiolabeled integrin alpha v beta 3receptor antagonists for tumor imaging and radiotherapy. Cancer BiotherRadiopharm 2003;18:627–41.

35. PanD, PhamCT,Weilbaecher KN, TomassonMH,Wickline SA, LanzaGM.Contact-facilitated drug delivery with Sn2 lipase labile prodrugs optimizetargeted lipid nanoparticle drug delivery. Wiley Interdiscip Rev NanomedNanobiotechnol 2016;8:85–106.

36. Soman NR, Lanza GM, Heuser JM, Schlesinger PH, Wickline SA. Synthesisand characterization of stable fluorocarbonnanostructures as drug deliveryvehicles for cytolytic peptides. Nano Lett 2008;8:1131–6.

37. Zutter MM, Mazoujian G, Santoro SA. Decreased expression of integrinadhesive protein receptors in adenocarcinoma of the breast. Am J Pathol1990;137:863–70.

38. Kostenuik PJ, Singh G, Orr FW. Transforming growth factor beta upregu-lates the integrin-mediated adhesion of humanprostatic carcinoma cells totype I collagen. Clin Exp Metastasis 1997;15:41–52.

39. Akhurst RJ, Hata A. Targeting the TGFbeta signalling pathway in disease.Nat Rev Drug Discov 2012;11:790–811.

40. Moros M,Mitchell SG, Grazu V, de la Fuente JM. The fate of nanocarriers asnanomedicines in vivo: important considerations and biological barriersto overcome. Curr Med Chem 2013;20:2759–78.

41. KanedaMM, Sasaki Y, LanzaGM,Milbrandt J,Wickline SA.Mechanisms ofnucleotide trafficking during siRNA delivery to endothelial cells usingperfluorocarbon nanoemulsions. Biomaterials 2010;31:3079–86.

42. Jones SE, Erban J, Overmoyer B, Budd GT, Hutchins L, Lower E, et al.Randomized phase III study of docetaxel compared with paclitaxel inmetastatic breast cancer. J Clin Oncol 2005;23:5542–51.

43. Soundrapandian C, Sa B, Datta S. Organic-inorganic composites for bonedrug delivery. AAPS PharmSciTech 2009;10:1158–71.

44. Diessner J, Wischnewsky M, Stuber T, Stein R, Krockenberger M, Hausler S,et al. Evaluation of clinical parameters influencing the development ofbone metastasis in breast cancer. BMC Cancer 2016;16:307.

45. Horner A, Kemp P, Summers C, Bord S, Bishop N, Kelsall A, et al.Expression and distribution of transforming growth factor-b isoforms andtheir signaling receptors in growing human bone. Bone 1998;23:95–102.

46. Chiechi A, Waning DL, Stayrook KR, Buijs JT, Guise TA, Mohammad KS.Role of TGF-beta in breast cancer bone metastases. Adv Biosci Biotechnol2013;4:15–30.

47. Parvani JG, Galliher-Beckley AJ, Schiemann BJ, Schiemann WP. Targetedinactivation of beta1 integrin induces beta3 integrin switching, whichdrives breast cancer metastasis by TGF-beta. Mol Biol Cell 2013;24:3449–59.

48. Galliher AJ, Schiemann WP. Beta3 integrin and Src facilitate transforminggrowth factor-beta mediated induction of epithelial-mesenchymal transi-tion in mammary epithelial cells. Breast Cancer Res 2006;8:R42.

49. Desgrosellier JS, Lesperance J, Seguin L, Gozo M, Kato S, Franovic A, et al.Integrin alphavbeta3 drives slug activation and stemness in the pregnantand neoplastic mammary gland. Dev Cell 2014;30:295–308.

50. Takahashi M, Mizoguchi T, Uehara S, Nakamichi Y, Yang S, Naramoto H,et al. Docetaxel inhibits bone resorption through suppression of osteoclastformation and function in different manners. J Bone Miner Metab2009;27:24–35.

Cancer Res; 77(22) November 15, 2017 Cancer Research6312

Ross et al.

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225

Page 15: Bone-Induced Expression of Integrin b3 Enables Targeted … · However, bisphosphonate-targeted drug delivery indiscrimi-nately accumulates on all skeletal bone matrix and primarily

2017;77:6299-6312. Published OnlineFirst August 30, 2017.Cancer Res   Michael H. Ross, Alison K. Esser, Gregory C. Fox, et al.   Nanotherapy of Breast Cancer Metastases

3 Enables TargetedβBone-Induced Expression of Integrin

  Updated version

  10.1158/0008-5472.CAN-17-1225doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://cancerres.aacrjournals.org/content/suppl/2017/08/30/0008-5472.CAN-17-1225.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/77/22/6299.full#ref-list-1

This article cites 50 articles, 12 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/77/22/6299.full#related-urls

This article has been cited by 1 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/77/22/6299To request permission to re-use all or part of this article, use this link

on April 10, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst August 30, 2017; DOI: 10.1158/0008-5472.CAN-17-1225