Research Article

Disruption of IFN-I Signaling Promotes HER2/NeuTumor Progression and Breast Cancer Stem CellsLuciano Castiello1, Paola Sestili1, Giovanna Schiavoni1, Rosanna Dattilo2,Domenica M. Monque1, Fiorella Ciaffoni3, Manuela Iezzi4, Alessia Lamolinara4,Antonella Sistigu2,5, Federica Moschella1, Anna Maria Pacca1, Daniele Macchia1,Maria Ferrantini1, Ann Zeuner1, Mauro Biffoni1, Enrico Proietti1, Filippo Belardelli1,and Eleonora Aric�o1

Abstract

Type I interferon (IFN-I) is a class of antiviral immunomod-ulatory cytokines involved in many stages of tumor initiationand progression. IFN-I acts directly on tumor cells to inhibitcell growth and indirectly by activating immune cells to mountantitumor responses. To understand the role of endogenousIFN-I in spontaneous, oncogene-driven carcinogenesis, wecharacterized tumors arising in HER2/neu transgenic (neuT)mice carrying a nonfunctional mutation in the IFNI receptor(IFNAR1). Such mice are unresponsive to this family of cyto-kines. Compared with parental neuþ/� mice (neuT mice),IFNAR1�/� neuþ/� mice (IFNAR-neuT mice) showed earlieronset and increased tumor multiplicity with marked vasculari-

zation. IFNAR-neuT tumors exhibited deregulation of geneshaving adverse prognostic value in breast cancer patients, includ-ing the breast cancer stem cell (BCSC) marker aldehyde dehy-drogenase-1A1 (ALDH1A1). An increased number of BCSCswere observed in IFNAR-neuT tumors, as assessed by ALDH1A1enzymatic activity, clonogenic assay, and tumorigenic capacity.In vitro exposure of neuTþ mammospheres and cell lines toantibodies to IFN-I resulted in increased frequency of ALDHþ

cells, suggesting that IFN-I controls stemness in tumor cells.Altogether, these results reveal a role of IFN-I in neuT-drivenspontaneous carcinogenesis through intrinsic control of BCSCs.Cancer Immunol Res; 6(6); 658–70. �2018 AACR.

IntroductionType I interferon (IFN-I) are pleiotropic cytokines

that exert multiple biological effects in viral infections andcancer (1, 2). Injection of IFN-I–blocking antibody resulted

in enhanced transplantable tumor growth, suggesting thatendogenous IFN-I controls tumor development and progres-sion in immunocompetent mice (3). Endogenous IFN-Ifunctions in cancer immune surveillance (4) and in the innaterecognition of tumors by several immune cells (5).

Studies in mouse models show that IFN-I inhibits growth oftransplantable tumors and exerts biological effects on immunecells, including NK cells, dendritic cells (DCs), T, and B cells (2).Thus IFN-I links innate immunity and development of the anti-tumor immune response (5).

HER2 is an oncogene overexpressed in 15% to 30% breastcancers (6). HER2þ breast cancers are characterized by amolecular signature distinguishing these cancers from otherbreast cancer types (7). HER2 overexpression is a negativeprognostic factor, associated with poorly differentiated, high-grade tumors, shorter disease-free and overall survival (6).Mice engineered to express an activated rat neu gene (Erbb2,ortholog of HER2) mimic most of the features observed inhuman HER2þ breast cancers (8, 9). In fact, during the firstpostnatal weeks, foci of atypical hyperplasia can be observedin mammary glands of neu transgenic mice. Gradually, thesefoci coalesce to form large carcinomas in situ, which progressto invasive carcinomas (10).

Despite the extensive characterization of IFN signaling intumor biology and studies on the mechanisms behind neu-driven tumorigenesis, little is known about endogenous IFN-Isignaling in neu transformation and progression and in spon-taneous carcinogenesis. Here, we studied neu-driven tumori-genesis in mice unresponsive to IFN-I by generating a mouse

1Department of Oncology and Molecular Medicine, Istituto Superiore di Sanit�a,Rome, Italy. 2Unit of Tumor Immunology and Immunotherapy, Department ofResearch, Advanced Diagnostics and Technological Innovation Regina ElenaNational Cancer Institute, Rome, Italy. 3Department of Cell Biology andNeurosciences, Istituto Superiore di Sanit�a, Rome, Italy. 4Department of Med-icine and Aging Science, Center of Excellence on Aging and TranslationalMedicine (CeSi-Met), G. D'Annunzio University, Chieti-Pescara, Italy. 5Depart-ment of General Pathology and Physiopathology, Universit�a Cattolica del SacroCuore, Rome, Italy.

Note: Supplementary data for this article are available at Cancer ImmunologyResearch Online (http://cancerimmunolres.aacrjournals.org/).

L. Castiello and P. Sestili contributed equally to this article.

Current affiliation for L. Castiello: Istituto Pasteur Italia—Fondazione CenciBolognetti, Rome, Italy; current affiliation for F. Belardelli: Institute of Transla-tional Pharmacology, CNR, Rome, Italy; and current affiliation for E. Aric�o:FaBioCell, Core Facilities, Istituto Superiore di Sanit�a, Rome, Italy.

Corresponding Authors: Filippo Belardelli, Istituto Superiore di Sanita,Viale Regina Elena, 288, Rome 00161, Italy. Phone: þ390649902414; Fax:þ3949902140; E-mail: filippo.belardelli@iss.it; and Eleonora Aric�o, FaBio-Cell, Core Facilities, Istituto Superiore di Sanit�a, Viale Regina Elena 299,00161, Rome, Italy. Phone: þ390649902414; E-mail: eleonora.arico@iss.it

doi: 10.1158/2326-6066.CIR-17-0675

�2018 American Association for Cancer Research.

CancerImmunologyResearch

Cancer Immunol Res; 6(6) June 2018658

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

strain transgenic for HER2/neu and lacking the IFN-I receptorIFNAR1 (IFNAR-neuT mice).

Materials and MethodsMice

NeuT and IFNAR-neuT transgenic mice were generated on129sv background and bred in the animal facility of the Depart-ment of Oncology and Molecular Medicine at the IstitutoSuperiore di Sanit�a (Rome, Italy). Pure neuT and IFNAR-neuTmice were obtained by at least 12 backcrosses of BALB-neuTtransgenic males (kindly provided by Dr. Guido Forni, Univer-sity of Torino, Torino, Italy) with 129Sv IFNAR1�/� female mice(A129, originally purchased from B&K Universal Ltd, nowMarshall BioResources; ref. 11) or the IFNAR1þ/þ counterpart(Charles River Laboratories). The BALB-neuT strain originatedfrom a transgenic CD1 random-bred breeder male mouse(no. 1330) carrying the mutated rat HER2/neu oncogene drivenby the MMTV promoter (9). Mice were maintained under strictinbreeding conditions. The presence of the rat HER2 transgenewas routinely checked by polymerase chain reaction (PCR) ontail DNA using primers hybridizing to vector (5-ATCGGT-GATGTCGGCGATAT-3) and to MMTV sequences (5-GTAACA-CAGGCAGATGTAGG-3). The mammary glands of all transgenicvirgin female mice were inspected once a week for tumorappearance. Individual neoplastic masses were measured withcalipers in 2 perpendicular diameters and the mean value wasrecorded. Progressively growing masses with mean diameter >1mm were regarded as tumors. Mice bearing tumor massesexceeding 20 mm mean diameter or necrotic lesions or miceshowing signs of distress were euthanized. Tumor multiplicitywas calculated as the cumulative number of incident individualtumors/total number of mice and reported as mean � SE.

NOD-SCID-g (NSG) mice from Charles River Laboratorieswere maintained in accordance with the European Community.The experimental protocols were approved by the Ethical Com-mittee for Animal Experimentation of the Istituto Superiore diSanit�a (Rome, Italy).

Cell lines and mammosphere cultureOne tumor cell line was isolated from a spontaneous mam-

mary tumor developed in a neuT virgin female, stabilizedin vitro and named 676-1-25 as previously described (11).N202.1A and N202.1E are cloned cell lines derived from anFVB mice (H-2q) transgenic for the rHER2/neu proto-oncogene(12). N202.1A cells express high levels of HER2/neu, whereasN202.1E cells are HER2�. Both cell lines were kindly providedby Dr. Claudia Curcio ("Gabriele d'Annunzio" University,Chieti, Italy) and tested by flow cytometry for HER2 expression.All cells were cultured in RPMI 1640 supplemented with 10%FCS. Cells were authenticated by morphology and growthin vitro and in vivo and routinely tested for mycoplasma; cellswere used for experiments within three passages after thawingmaster cell bank aliquots stored in liquid nitrogen.

Mammary gland tissues were collected at the designed timepoints for flow cytometric analysis, in vitro clonogenic assays orin vivo tumorigenic assay. Briefly, tumor samples were freed fromhemorrhagic and necrotic parts, washed in phosphate-bufferedsaline (PBS), minced with scissors, and digested in serum-freemedium supplemented with collagenase (Sigma Blend F, Sigma-Aldrich) 400 U/mL and DNAse type I 200 U/mL at 37�C for

30 minutes. After incubation, 5 mL RPMI 10% FCS EDTA2 mmol/L was added, and cells were then filtered in cell strainersand washed twice in PBS.

For mammosphere culture, cells recovered from transgenicmice (13) tumors were cultured at clonal density in serum-freemedium supplemented with 20 ng/mL epidermal growth factor(EGF) and 10 ng/mL basic fibroblast growth factor (bFGF).Ultralow attachment cell culture flasks were used. The mediumwas replaced or supplemented with fresh growth factors once aweek until cells started forming floating aggregates. Cultures wereexpanded by mechanical dissociation of spheres, followed byreplating of both single cells and residual small aggregates incomplete fresh medium.

Immunophenotypic analysisImmunophenotypic analysis of cells isolated from spleens,

lymph nodes, bone marrow, blood, and mammary carcinomasof mice was carried out as follows: cell suspensions were treatedwith lysis buffer (140 mmol/L NH4Cl, 17 mmol/L Tris HCl,pH 7.2) to eliminate red blood cells and then stained with thefollowing fluorescently labeled mAbs: anti-CD45 (30-F11), anti-CD11c (N418), anti-CD3 (145-2C11), anti-CD8a (53-6.7), anti-CD25 (PC61.5), anti-CD357/GITR (DTA-1), anti-Foxp3 (FJK-16s)from eBioscience; anti-CD4 (GK1.5), anti-Ly6C (AL-21), anti-Ly6G (1A8) from BD Pharmingen; anti-CD11b (M1/70), anti-CD19 (6D5), anti-Gr-1 (RB6-8C5), and anti-mPDCA1 (JF05-1C2.4.1) from Miltenyi Biotec. For determination of HER2/Neuexpression in tumors, cells isolated from mammary tumorswere stained with PerCP labeled anti-CD45 coupled to anti-HER2/Neu (NEU 7.16.4, Calbiochem) followed by anti-mouseAlexafluor488 and by propidium iodide (PI). The percentage ofHER2/Neu expression was determined in the PI�CD45� cellpopulation (bona fide tumor and stromal cells). For the analysisof putative CSC markers, tumor cells isolated from 24-week-oldtransgenicmice as described abovewere stainedwith the followingfluorescently labeled mAbs: anti–lineage markers: anti-Ter119(TER-119), anti-CD31 (MEC 13.3), anti-CD140a (APA5) fromBiolegend; anti-Ly6A/Sca1 PE (D7), anti-CD133 APC (13A4);anti-CD24 (M1/69) from e-Bioscience; anti-CD44 (IM7), anti-CD49f (GoH3), anti-CD61 (2C9.G2) from BioLegend; anti-LGR(GPR49) from Abgent; and anti-EPCAM (G8.8) from SantaCruz Biotechnology. For the immunophenotypic analysisof spleen, lymph node, bone marrow, blood, and mammarycarcinomas, cell populations were defined as follows: CD4 Tcells: CD45þCD3þCD19�CD4þCD8�; CD8 T cells: CD45þ

CD3þCD19�CD4�CD8þ; B cells: CD45þCD3�CD19þ; Treg cells:CD45þCD3þCD4þCD8�CD25þFoxp3þGITRþ; MDSC:CD45þCD11bþGr-1þCD124/IL4Rþ; Mo-MDSC: CD45þCD11bþ

Ly6ChiLy6G�; Gr-MDSC:CD45þCD11bþLy6CþLy6Gþ; pDC:CD45þmPDCA1þCD11b�CD11clow. All samples were run ona Gallios flow cytometer and analyzed with Kaluza AnalysisSoftware (Beckman Coulter).

In vitro clonogenic assays and tumorsphere formation assaysSoft agar colony-forming assays were carried out as previously

described (13). For tumorsphere formation assay, cells isolatedfrom tumors were plated in ultra-low attachment six-well plates(Corning) at a density of 10,000 cells/mL in serum-free mediumsupplemented with 20 ng/mL epidermal growth factor (EGF,Sigma-Aldrich), 10 ng/mL basic fibroblast growth factor (b-FGF,

IFN-I Signaling Regulates HER2 Carcinogenesis and Stemness

www.aacrjournals.org Cancer Immunol Res; 6(6) June 2018 659

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

Sigma-Aldrich). Cells were cultured under 5% CO2 at 37�C for1 week.

In vivo tumorigenesis assaysTumors were excised from 23-week-old neuT and IFNAR-neuT,

dissociated and the cells were counted and resuspended in amixture of PBS and Matrigel (1:1). A total of 1 � 106 cells wereinjected subcutaneously into the right flank of each mouse ororthotopically injected into the right fat pad. At day 30, tumorswere excised and prepared for immunohistochemistry analysis(see below).

Morphological analysis of the mammary glandsGroups of four mice were sacrificed at the indicated times,

and mammary tissue was processed as previously describ-ed for histologic, immunohistochemical, and whole-mountanalysis (14).

For whole-mount preparation, the pelt of euthanized trans-genic and control mice was fixed overnight in 10% neutralbuffered formalin. The mammary fat pads were scored intoquarters and gently scraped from the pelt. The quarters wereimmersed in acetone overnight and then rehydrated andstained in ferric hematoxylin (Sigma-Aldrich), dehydrated inincreasing concentrations of alcohol, cleared with histo-lemon,and stored in methyl salicylate (Sigma-Aldrich). Digital photoswere acquired with a Nikon Coolpix 995 (Nital SpA) mountedon a stereoscopic microscope (MZ6; Leica Microsystems) with a0.63 objective giving a total magnification of �6.3. The reso-lution was 1,600 � 1,200 pixels. Images were acquired with anAdobe PhotoShop v. 6.0 graphic software (Adobe Systems).Mammary glands that exceeded the size of a single-imagingarea were captured by photographing contiguous fields in araster pattern. Each captured image was merged using the layertechnique in Adobe PhotoShop to form a single-compositepicture for analysis. Spatial calibration was determined byphotographing a 1-mm stage using the same parameters asthose for image capturing of whole-mount preparations. Thedistance drawn on the 1-mm calibration image was dividedby 1,000 to find the number of pixels per micrometer. In eachimage, �100 discrete points were randomly chosen on theduct surface, the width in micrometers of any lesion in thatpoint was measured perpendicular to the duct direction, and avalue of zero was assigned if no lesion were present.

For histology and immunohistochemistry, samples werefixed in 10% neutral buffered formalin and embedded intoparaffin or fixed in 4% PFA and frozen in a cryo-embeddingmedium (OCT, BioOptica); 5 mm slides were cut and stainedwith hematoxylin (BioOptica) and eosin (BioOptica) for his-tological examination. For immunohistochemistry, slides weredeparaffinized, serially rehydrated and, after the appropriateantigen retrieval procedure, stained with rabbit polyclonalanti-human HER2 antibody (A0485, Dako) followed by sec-ondary antibody (Jackson ImmunoResearch, LiStarFish).Immunoreactive antigen was detected using streptavidin per-oxidase (Thermo Scientific; Lab Vision Corp.) and the DABChromogen System (Dako). After chromogen incubation,slides were counterstained in hematoxylin (BioOptica) andimages were acquired by Leica DMRD optical microscope(Leica). For immunofluorescence, 6 mm cryostat sections wereair-dried and fixed in ice-cold acetone for 10 minutes. Slideswere incubated with rat monoclonal anti-CD31 mixed with rat

monoclonal anti-CD105 (BD Pharmingen), followed by sec-ondary antibody conjugated with Alexa 546 (Invitrogen, LifeTechnologies). Nuclei were stained with DRAQ5 (Alexis, LifeTechnologies). Image acquisition and analysis were performedusing Zeiss LSM 510 META confocal microscope. The numberand area of CD31þCD105þ vessels were evaluated on thedigital images of 2 tumors per group (6 � 400 microscopicfields per tumor) in a double-blind fashion, and digital imagesof representative areas were taken. Vessels area (in pixels) wasevaluated with Adobe Photoshop by selecting vessels with thelasso tool and reporting the number of pixels indicated in thehistogram window.

RNA isolation, labeling, and hybridizationInguinal mammary glands were isolated from 19-week-old

mice, tissues were submerged with RNA later (Qiagen), andstored at �20�C. For RNA extraction, the mammary glandswere washed twice in PBS, immersed in TRIzol (QIAzol Lysisreagent). Tissues from individual mice were disrupted by 30"homogenization (Ika Homogenizer system T10), the resultinghomogenate underwent high-speed centrifugation to removeparticulate and total RNAs were eventually extracted by chloro-form followed by isopropanol precipitation. Total RNAswere then spectrophotometrically quantified and quality wasassessed by using the Agilent Model 2100 Bioanalyzer and theEukaryoteTotal RNA Pico kit (Agilent Technologies). Low-quality RNAs were discarded. Forty nanograms of sampleRNA underwent cRNA preparation and cyanin (Cy)5 labelingby Low Input Quick Amp Labeling Kit Standard DualColor (Agilent Technologies), according to the manufacturer'sinstruction. The same amount of universal reference mouseRNA (Stratagene) was processed and Cy3-labeled as referencecontrol. Labeled amplified RNAs were purified with Qiagen'sRNeasy mini spin columns and cRNA yields and specificactivities were checked before hybridization. A total of 825 ngof Cy3-labeled, linearly amplified reference cRNA and 825 ngof Cy5-labeled, linearly amplified cRNA were fragmented,mixed, and cohybridized to mouse gene expression 4 � 44Kv2 microarray slides (G4646A), in a final volume of 55 mL.Microarray slides were assembled into the appropriate cham-bers and placed in rotisserie in a hybridization oven set to65�C for 17 hours. Slides were then disassembled and washedaccording to the manufacturer's instructions.

Microarray data analysisScanning and image file processing were performed with Gen-

ePix 4200A instrument (Axon Instruments). Images were ana-lyzed using ScanArray Express software where spots with badquality were omitted allowing the rejection of spurious signalsdue to artifacts. Slide quality assessmentwas performed accordingto the signal intensity obtained in the presence of Spike Insamples. Data have been deposited in NCBI's Gene ExpressionOmnibus (GEO) and are accessible throughGEO Series accessionnumber GSE110350 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE110350).

The resulting data files were analyzed via BRB ArrayTool(http://linus.nci.nih.gov/BRB-ArrayTools.html) developed atthe National Cancer Institute, Biometric Research Branch,Division of Cancer Treatment and Diagnosis. The raw dataset was filtered according to standard procedure to excludespots with minimum intensity, arbitrarily set to <20 in both

Castiello et al.

Cancer Immunol Res; 6(6) June 2018 Cancer Immunology Research660

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

fluorescence channels. If the fluorescence intensity of onechannel was >20 and that of the other below <20, the fluo-rescence of the low-intensity channel was arbitrarily set to20. Spots with diameters <10 mm and flagged spots werealso excluded from the analysis. The resulting database, con-sisting of 18,361 transcripts, was used for further statisticalanalysis. Filtered data were then normalized in log spaceusing locally weighted linear regression (Lowess). All statisti-cal analyses were done using the log2-based ratios. Supervisedclass comparison between sample classes was performed withthe BRB ArrayTool using the random-variance model withparametric P value threshold set to 0.005. Significantly differ-entially expressed genes resulting from the analysis were fur-ther analyzed using cluster and TreeView software. Gene ratioswere average corrected across experimental samples and dis-played according to the central method for display using anormalization factor as recommended by Ross (15). Genefunction and classification was assigned by means of theDatabase for Annotation, Visualization, and Integrated Dis-covery (DAVID) (https://david.ncifcrf.gov/). Overlaps withpublished gene signatures were assessed using the MolecularSignature Database (MSigDB; https://www.broadinstitute.org/data-software-and-tools; 16). To evaluate the prognosis ofgene signatures in breast cancer patients, the dataset publiclyavailable on the KM plotter website was used (www.kmplot.com; 17).

Aldefluor assay by FACSThe ALDEFLUOR kit (StemCell Technologies) was used to

analyze the population with a high ALDH enzymatic activity.Cells obtained from freshly dissociated breast tissue or mam-mosphere cultures were first surface-stained with antibodiesdirected against lineage markers (Ter119, CD31, CD140a, andCD45), then resuspended in ALDEFLUOR assay buffer con-taining ALDH substrate (BAAA, 1 mmol/L per 1 � 106 cells)and incubated for 40 minutes at 37�C. As a negative control foreach sample of cells, an aliquot was treated with 50 mmol/Ldiethylaminobenzaldehyde (DEAB), an ALDH inhibitor. Cellswere then labeled with propidium iodide (PI) and analyzed byflow cytometry. Positivity for ALDH was assessed in cellsnegative for PI and lineage markers expression. DEAB-treatedsamples served as controls for background fluorescence.

Statistics and reproducibilityResults are represented as means � SEM. For group com-

parisons, we used nonparametric two-tailed Mann–Whitney–Wilcoxon U test. For free-survival analysis, Kaplan–Meier plotsand significance with P log-rank test were used. All statisticswere calculated with Openstat and Microsoft Excel software.The experiments based on images (Whole-Mount analysis,FACS, IF, H&E staining) were repeated at least 3 times andrepresentative images are shown. Microarray data statisticalanalysis, as detailed in the microarray section, was performedusing BRB ArrayTool (http://linus.nci.nih.gov/BRB-ArrayTools.html). For supervised analysis, the random variance modelwas used, and the parametric P value threshold was set to0.005. For gene ontology analysis, biological process classeswere ranked according to the �log (P) of the modified Fishertest. The overlaps with gene signatures present in the MSigDBwere calculated by the false discovery rate of q values and by theratio between the number of genes of the two sets.

ResultsIFN-I signaling disruption accelerates neuT-driventumorigenesis

To examine the role of endogenous IFN-I signaling on mam-mary tumorigenesis, we generated a mouse strain carrying theHER2/neu oncogene and lacking a functional IFNAR1. 129SvIFNAR1�/� mice (18) were backcrossed with Balb/C HER2/neutransgenic mice for more than 12 generations, until a colony ofneuþ/� IFNAR1�/� mice on a 129Sv background was obtained(IFNAR-neuTmice). Control 129Sv neuþ/�mice expressing func-tional IFN-I receptor gene (IFNAR1þ/þ) were also generated forcomparison (neuT mice). The results of tumor growth monitor-ing in both strains revealed that, despite nonstatistically signi-ficant earlier onset of first palpable lesion (Fig. 1A), IFNAR-neuTmice showed significantly higher mean tumor multiplicity(P < 0.01; Fig. 1B), achieving tumor formation in all 10 glandsapproximately 4 weeks earlier as compared with control neuTmice. A comparative whole-mount microscopy analysis of themammary glands showed similar progression from hyperplasticlesions to large microscopic lesions (equivalent to in situ carci-nomas) in neuT and IFNAR-neuT mice until the 15th week ofage (Fig. 1C). In contrast, the progression to invasive solidcarcinomas was accelerated in IFNAR-neuT mice with respectto controls. In fact, by week 19, IFNAR1�/�, but not parentalNeuT mice, displayed enlarged and partially clumped lesions,and, by week 26, large invasive carcinomas occupied the wholemammary gland, whereas control mice showed fewer invasivecarcinomas (Fig. 1C). Hematoxylin–eosin staining of tumor sec-tions confirmed larger lesions in IFNAR-neuT than in neuT micefromweek15 to26, althoughweobservednodifferences in tumordifferentiation grade at these stages (Fig. 1D). Altogether, theseresults indicate that endogenous IFN-I signaling disruption doesnot influence the early stages of tumor transformation but accel-erates the growth of large invasive carcinomas without affectingdifferentiation grade. No significant increase in lung metastaticburden was observed in IFNAR-NeuT versus NeuT mice (80% vs.69% of lungs with metastasis, respectively). Indeed, dissemina-tion of cancer cells in HER2 transgenic mice has been reported tooccur mostly from early lesions (7–9 weeks), before IFNAR-neuTtumor progression accelerates (19).

IFNAR-neuT tumors show increased angiogenesis but nodifference in the immune infiltrate

Given that tumor vasculature sustains tumor growth, inva-sion, and metastasis, and that IFN-I has antiangiogenic pro-perties (20), we analyzed tumor microvascular density byendothelial cell staining in neuT and IFNAR-neuT tumors ofequivalent size (Fig. 2A). Although similar numbers of intra-tumoral vessels were found in both groups of carcinomas(Fig. 2B), tumor-associated vessels appeared to be 25% largerin IFNAR-neuT than in neuT lesions (P � 0.05; Fig. 2C).

Disruption of IFN-I signaling did not alter the percentage andcomposition of tumor-infiltrating leukocytes in neuT lesions.No significant differences in the frequency of total infiltratingimmune cells, tumor-infiltrating T lymphocytes, B cells, andboth granulocytic and monocytic myeloid-derived suppressorcells (MDSC) were found in tumors of similar sizes arising inIFNAR-neuT and neuT mice (Fig. 2D and E). The two strainsalso shared a similar percentage of cells expressing the HER2/neu antigen (Fig. 2D). Similarly, immune cell phenotype was

IFN-I Signaling Regulates HER2 Carcinogenesis and Stemness

www.aacrjournals.org Cancer Immunol Res; 6(6) June 2018 661

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

not altered in the blood, spleen, bone marrow, and lymphnodes of IFNAR1�/� mice, whether or not they were transgenicfor HER2/neu (Supplementary Fig. S1). These findings suggestthat the increased tumor growth observed in IFNAR-neuT micemay be related not to immunosurveillance but rather to theincreased vessel size that resulted from absence of inhibition byendogenous IFN-I.

Mammary gland transcriptional program is affected by neuTcarcinogenesis

Weanalyzed gene expression profiles of 19thweek IFNAR-neuTand neuT inguinal mammary gland tumors. Although not alltransgenic mice showed palpable tumors at 19 weeks (Fig. 1C),tumor masses were visible within those inguinal mammaryglands chosen for the analysis at sacrifice. Mammary glands of129svWT and 129sv IFNAR1�/� mice were also examined tobetter assess tumor-specific molecular changes.

Class comparison analysis among all groups (P � 0.005)revealed 642 differentially expressed genes. A large set of genes

was similarly expressed in neuT and IFNAR-neuT tumors (blueboxes in Fig. 3A), suggesting that some of the neuT-driventranscriptional changes are not affected by IFNAR status. Weinitially focused on the modulations shared by IFNAR-neuT andneuT tumors. To strengthen the statistical power of our analysis,we compared neuT tumors with nontransgenic samples (neuTand IFNAR-neuT vs. 129svWT and 129sv IFNAR1�/�; P� 0.005).Of the 742 differentially expressed genes, 513 were upregulatedand 229 genes were downregulated in neuT tumor. Gene onto-logy analysis (Fig. 3B) revealed an overall overexpression, in neuTtumors, of genes related with response to stress (such as ATF3,ATF4, EGLN3, and PLSCR1), cell cycle (such as CDC25A, CDCA5,CDCA8, and CDK1), immune genes (NMI, STAT2 and STAT5A),and epithelial cell proliferation (such as ITGB3, TGFB2, andTHBS1). Genes downregulated in tumor tissue mainly belongedto metabolic pathways such as those that drive the metabolicchanges occurring upon transformation (Fig. 3B; ref. 21).

The genes upregulated by neuT tumorigenesis in our mousemodel overlapped with published signatures of human breast

Figure 1.

Earlier onset, higher mean tumor multiplicity and accelerated growth of mammary tumors in IFNAR-neuT mice. Incidence (A), and multiplicity (B) ofmammary carcinomas in neuT (blue line, n ¼ 25) or IFNAR-neuT (red line, n ¼ 28) female mice. Statistically significant differences in tumormultiplicity at the indicated time points are shown. � , P � 0.05; x, P � 0.01 (n ¼ 25, Mann–Whitney U test). C, Representative images of whole-mountmicroscopic analysis. The large central oval black spot in each image represents the intramammary lymph node. D, Photomicrograph of hematoxylin/eosin-stained slides of 15- and 26-week tumors, displaying typical growing adenocarcinoma. The scale bar, 100 mm (magnification 20�).

Castiello et al.

Cancer Immunol Res; 6(6) June 2018 Cancer Immunology Research662

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

tumors (such as SOTIRIOU_BREAST_CANCER_GRADE_1_VS_3_UP) and with neuT specific signatures in mice (LANDIS_ERBB2_BREAST_TUMORS; MSigDB; http://www.broadinstitute.org/gsea/msigdb; Fig. 3C). Overall, gene expression profilingdata of the spontaneous carcinogenesis in HER2/neu transgenicmammary glands suggested that shared molecular pathwayscould be responsible for the similarities observed in IFNAR-neuT and neuT tumors.

IFNAR-neuT tumors share gene expression profile withaggressive human breast cancers

To better characterize the effect of IFN-I disruption on 19-week-old HER2/neu spontaneous tumors, we compared ex-pression profiles of neuT and IFNAR-neuT samples. We found118 differentially expressed genes (P � 0.005; Fig. 4A), ofwhich 60 were more expressed and 58 were less expressed inIFNAR-neuT versus neuT neoplasms. IFNAR-neuT tumorsoverexpressed transcripts encoding proteins involved in tumorinvasiveness and epithelial-to-mesenchymal transition, such

as Laminin C and fibronectin (22, 23), as well as ALDH1A1, amarker of cancer stem cells (CSC; ref. 24) and a negativeprognostic factor for breast cancer patients (25). We, therefore,tested the prognostic value of the IFNAR-neuT overexpressedgenes in breast cancer patients from the KM plotter data set(17). Although the gene signature has no prognostic valueon relapse-free survival in the whole set of breast cancerpatients, we observed adverse prognostic value in HER2þ

breast cancer patients on relapse-free survival and distantmetastases-free survival (Fig. 4B–E). No prognostic value ofthe signature was observed by restricting the analysis accord-ing to other breast cancer subtypes except for progesterone-receptor negative and triple-negative breast cancers (Supple-mentary Fig. S2). Altogether, these results indicate that theIFNAR-neuT cancers are characterized by an expression profilethat is observed in breast cancer patients with a worse prog-nosis, thus suggesting that the lack of IFN-I signaling increasestumor aggressiveness through pathways shared with certainsubtypes of breast cancer.

Figure 2.

IFNAR-neuT mammary carcinomas showed larger vessels than neuT lesions but minimal changes of immune infiltrate. A, Confocal microscopy analysisof carcinoma samples stained with anti-CD31 and anti-CD105 (red) in neuT and IFNAR-neuT carcinomas. Nuclei are stained with DAPI. Single colorsand merged images are shown. Quantification of the number of vessel (B) and vessel area (C) in neuT (blue bars) and IFNAR-neuT (red bars) carcinomas.Data are representative of three independent experiments. Results are represented as means� SEM from 7 � 100 microscopic fields of 10 tumors. � , P � 0.05.D, Expression of HER2/neu and total infiltrating leukocytes (CD45þ) on spontaneous neuT (blue bars) and IFNAR-neuT (red bars) digested tumors,isolated from 26-week-old transgenic mice. E, Flow cytometry analysis of tumor-infiltrating cells in NeuT and IFNAR-NeuT carcinomas. Data arerepresentative of three independent experiments and are expressed as mean percentages of the indicated immune cell subsets among gated CD45þ

total leukocytes in individual mice � SEM. �, P � 0.05 (n ¼ 10, Mann–Whitney U test).

IFN-I Signaling Regulates HER2 Carcinogenesis and Stemness

www.aacrjournals.org Cancer Immunol Res; 6(6) June 2018 663

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

IFNAR-neuT tumors show increased frequencies ofbreast CSC

The observed modulation of genes with an adverse prog-nostic value and involved in CSC, together with the report-ed correlation between HER2 overexpression and tumor-initiation capacity in vitro and in vivo (26, 27), prompted usto further investigate the role of ALDH1A1 in IFNAR-neuTmammary lesions.

Therefore, we analyzed the enzymatic activity of ALDH inlesions freshly isolated from IFNAR-neuT tumors as comparedwith the neuT counterparts. IFNAR-neuT lesions showed a higherpercentage ofALDHþ cells as comparedwithneuT tumors (Fig. 5Aand B). We did not observe altered expression of moleculespreviously reported to be CSC markers in HER2/neu transgenicmice (CD61, CD49f, CD24, Sca1, and CD44; refs. 28–30) orin other tumors (CD133, EPCAM, and LGR5; SupplementaryTable S1; ref. 31).

To assess whether the increased presence of ALDHþ cells wasparalleled by higher number of BCSCs in IFNAR-neuT tumors,we evaluated tumor cell self-renewal and tumorigenic capacityin vitro and in vivo. A soft agar clonogenic assay revealed that more

colonies could be generated from IFNAR-neuT than from neuTtumor cells (Fig. 5C). The increased numbers of BCSCs werealso tested by tumorigenic ability potential in vivo. Tumor cellsisolated from IFNAR-neuT mice and orthotopically transplantedin the mammary gland of NOD/SCID/Gamma (NSG) mice ledto tumor lesions that were palpable 7 days earlier than tumorsderived from neuT mice (Fig. 5D). Moreover, IFNAR-neuTtumors grew faster, resulting in tumor nodules 6-fold largerthan the control counterpart 30 days after transplant (Fig. 5D).Similar results were observed when freshly isolated tumorcells were transplanted subcutaneously in NSG mice (Supple-mentary Fig. S3), where IFNAR-neuT tumors became palpablesooner than neuT tumors and continued to grow faster. Histo-logical analysis showed a similarmorphology of neuT and IFNAR-neuT tumors in NSG xenografts (Fig. 5E), retaining the histo-logical features of primary cognate tumors (Fig. 1D).

Altogether, these results indicate that, in the absence offunctional IFN-I signaling, spontaneously arising neuT tumorsare characterized by increased frequencies of BCSCs, endowedwith increased self-renewal and tumorigenic capacity in vitroand in vivo.

Figure 3.

Gene expression profiles reveal tumorigenic signatures in neuT and IFNAR-neuT carcinomas. A, Heatmap of the 642 genes differentially expressedamong neuT, IFNAR-neuT, 129sv WT, and 129sv IFNAR1�/� mammary tissue at 19 weeks (P � 0.005, random-variance F test). Average corrected geneexpression levels are shown. Downregulated genes are in blue, upregulated ones in yellow. Highlighted by blue squares are cluster of genes induced insamples harboring neuT gene irrespectively of IFNAR1 status. B, Gene Ontology biological process classes, ranked according to the �log (P) of themodified Fisher test, are shown for the 513 upregulated genes (top) and for the 229 downregulated ones (bottom) as analyzed by David [https://david.ncifcrf.gov/]. C, Schematic representation of gene set overlaps of published gene signatures and the 513 neuT upregulated genes. Significance of overlap iscalculated by false discovery rate of q values and shown by red bars, whereas the percentage of overlap is shown by the red line showing the ratio of k(the number of genes of the specified gene set belonging to the 513 upregulated genes) and K (the total number of genes belonging to the gene set).

Castiello et al.

Cancer Immunol Res; 6(6) June 2018 Cancer Immunology Research664

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

In vitro exposure to anti–IFN-I increased numbers of ALDHþ

cells in mammospheresTo analyze in vitro the effect of the lack of endogenous IFN-I

signaling on neuT tumors, we cultured neuT tumor cells innonadherent nondifferentiating conditions (mammospheres)and measured variations in the percentage of ALDHþ cells inthe presence of a neutralizing anti–IFNa/b sheep immunoglob-ulin (anti-IFN; Fig. 6A and B; ref. 32). Mammospheres derivedfrom neuT tumors showed more ALDHþ cells 24 hours afteranti-IFN treatment (Fig. 6A and B). No modulation wasobserved after IFN-I neutralization in mammospheres derivedfrom IFNAR-neuT tumors, which already showed more ALDHþ

cells than their neuT counterparts (Fig. 6A and B). Anti-IFNtreatment induced formation of bigger mammospheres afterculturing in soft agar from neuT-derived tumor cells thanfrom IFNAR-neuT–derived tumor cells, which are unable torespond to IFN-I (Fig. 6C). A similar increase in the number ofcells expressing the CSC marker ALDH was observed in otherneuT-adherent tumor cell lines when treated with anti-IFN

(Fig. 6A and B). In particular, IFN-I neutralization elicitedan increase in the frequencies of ALDHþ cells in the neuT-expressing cell lines 676 and N202.1A, isolated from tumorsspontaneously developed in 129Sv and FVB transgenic mice,respectively (Fig. 6A and B; refs. 11, 12). By contrast, no effectcould be observed in N202.1E, the FVB-derived HER2/neu�

counterpart (Fig. 6A and B). These data suggest an involvementof IFN-I signaling in regulating the number of ALDHþ cellsin neuT-expressing tumors in a manner dependent on theneuT oncogene.

DiscussionHere, we examined how the impairment of IFN-I signaling

due to absence of a functional IFN-I receptor affects neuTtumorigenesis and its progression. Our results show enhancedtumor growth and altered tumor vasculature in IFNAR-neuTmice, but no differences in immune infiltrate composition.IFN-I signaling impairment affected the presence of BCSCs

Figure 4.

IFNAR-neuT tumors show molecular signature shared with certain subtypes human breast cancer patients. A, Clustered heatmap of the 118 genesdifferentially expressed (P � 0.005, random-variance T test) between neuT (green bar) and IFNAR-neuT (blue bar) tumor lesions, isolated from 19-week-oldtransgenic mice. B to E, Kaplan–Meier plots showing the prognostic value of the genes overexpressed in IFNAR-neuT on relapse-free survival (B andD) or distant metastases-free survival (C and E) in the overall set of breast cancer patients (B and C) or when restricting the analysis on the only HER2þ

tumor patients (D and E). The analysis was performed using publicly available database KM plotter.

IFN-I Signaling Regulates HER2 Carcinogenesis and Stemness

www.aacrjournals.org Cancer Immunol Res; 6(6) June 2018 665

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

within tumor and altered expression of the CSC marker ALDH,thus highlighting a role for IFN-I in BCSC biology.

The results shown in the present paper imply that, in line withwhat observed in other in vivo mouse models by our group andothers (33–36), endogenous IFN-I is constitutively expressedin neuTmice at low levels. Although below the limits of detectionin our assay, endogenous IFN-I is in fact responsible for the effectson spontaneous tumors observed in the absence of a functionalIFN receptor in vivo and in the presence of anti–IFN-I in vitro.

Given the pleiotropic activities of IFN-I, a variety of mechan-isms, including the lack of antiproliferative effects on tumorcells and an indirect impact on tumor angiogenesis (37), mayaccount for the observed acceleration of neuT-driven tumorigen-esis and progression in IFNAR-neuT versus neuTmice. Our resultsshowed that tumor vasculature was increased in IFNAR-neuTmice. Rather than an increase in the number of vessels, weobserved an enlargement of vessel area in IFNAR-neuT mice,which may improve permeability and diffusion rates, and thusimprove oxygen and nutrient supply to the tumor.

IFN-I signaling disruption impairs immunosurveillance there-fore boosting tumor growth (38). Some studies have shown that

implantable and induced tumors grow faster in IFNAR1�/� micedue to the reduced ability of the immune system of these miceto develop an antitumor immune response (4). Our results didnot reveal any change in immune cell composition betweenneuT and IFNAR-neuT mice at the tumor site and in the mainlymphoid organs, thus suggesting that immunosurveillancemechanisms play a limited role in our setting. In transgene-driventumorigenesis, tolerance mechanisms usually surge becausedominant clones are selectively deleted in neuT transgenic mice(39). Therefore, under normal conditions, spontaneous HER2tumors show limited immune cell infiltration (8), which can bepotentiated by immunotherapies (11). In the absence of a spon-taneous antitumor immune response (i.e., in neuT-tolerizedmice), immunosurveillance dysfunctions are rarely observed.Therefore, we conclude that, in this setting, IFN-I–dependentimmune impairment has a reduced role in tumorigenesis.Enhanced expression of IFNg and other inflammatory mediatorshas been reported in IFNAR1�/� mice with respect to controlanimals (40), suggesting that redundant biological responsesand immune mechanisms might compensate for lack of endog-enous IFN-I responses in our neuT model.

Figure 5.

IFNAR-neuT tumors are enriched in BCSCs. A, Percentages of ALDHþ cells assessed by flow cytometric analysis of tumor cells freshly isolated from neuTand IFNAR-neuT spontaneous tumors. Data are representative of three independent experiments. The histogram shows the mean percentage (�SEM)of ALDHþ cells, calculated for five mice per group and representative of two independent experiments. x, P � 0.01 (n ¼ 5, Mann–Whitney U test). B,Representative plots for Aldefluor control (þ DEAB, a specific ALDH inhibitor, left) and test (right) samples from tumor cells isolated from neuT andIFNAR-neuT mice are shown. C, Mean colony-forming efficiency of neuT (blue bars) and IFNAR-neuT (red bars) tumor cells assessed by soft-agarclonogenic assay at the indicated time points. The number of colonies was evaluated at 14, 23, and 33 days after seeding 5,000 cells per well. Data arerepresentative of three independent experiments. Error bars represent standard errors x, P � 0.01 (n ¼ 5, Mann–Whitney U test). D, Tumor growth(expressed as mean tumor diameter, mm) after orthotopic transplantation of 1 � 106 cells neuT (blue line) or IFNAR-neuT (red line) tumor cells in NSG mice.Data are representative of two independent experiments. Error bars represent standard errors and statistical differences are indicated � , P � 0.05; x, P � 0.01(n ¼ 5, Mann–Whitney U test). E, Photomicrograph of hematoxylin/eosin-stained slides of NSG xenografts, excised 30 days after injection andshowing similarities with cognate primary tumors (magnification 20�).

Castiello et al.

Cancer Immunol Res; 6(6) June 2018 Cancer Immunology Research666

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

Figure 6.

IFN-I neutralization in vitro increases the number of ALDHþ cells in mammospheres derived from neuT tumors and other HER2þ tumor cell lines. A, FACSprofiles showing the effect of anti-IFN-I (50U, IFN-neutralizing units of antibody/mL) 24 hours treatment on the percentage of ALDHþ cells in (i)mammosphere derived from neuT tumor; (ii) mammosphere derived from IFNAR-neuT tumor, (iii) neuT conventional tumor cells line isolated fromneuT tumor; (iv) neuþ/� tumor cells line isolated from HER2/neu FVB tumor-bearing mouse, and (v) neu�/� tumor cells line isolated from HER2/neu FVBtumor-bearing mouse. Representative plots for Aldefluor control (þDEAB, a specific ALDH inhibitor, left) and test (right) samples are shown for eachcondition. B, Mean percentage of ALDHþ cells in control (white bar) and test (gray bar) samples for anti-IFN-I–treated samples versus untreated (NT) ormock-treated controls. Data are representative of at least 3 independent experiments, all performed with four replicates per group. Error barsrepresent standard errors and statistical differences are indicated x, P � 0.01 (n ¼ 4, Mann–Whitney U test). C, Effect of anti-IFN-I (50 U/mL) for 24 hourstreatment on the morphology of neuT and IFNAR-neuT mammospheres, as shown by phase-contrast microscopy. Data representative of threedifferent experiments. Magnification, 20�.

IFN-I Signaling Regulates HER2 Carcinogenesis and Stemness

www.aacrjournals.org Cancer Immunol Res; 6(6) June 2018 667

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

Downstream mechanisms thought to be controlled byHER2 during neoplastic transformation (41) include cell-cyclederegulation, loss of cell polarity and cell adhesion, inductionof invasive phenotype, and cellular metabolism. These pro-cesses are surely affected by HER2/neu (41). Our analysis ofthe gene expression data showed induction of transcriptsinvolved in regulation of cell cycle, apoptosis, and metabolicpathways compared with nontransformed mammary gland.The observed molecular changes are representative of onco-genic signatures reported in other in vitro and in vivo breastcancer settings. Most of the HER2-transformation-inducedchanges were shared between neuT and IFNAR-neuT tumors,thus corroborating the impact of carcinogenesis on mammarygland transcriptional program. Nonetheless, IFNAR-neuTtumors showed upregulation of a limited set of transcriptsincluding, among the others, genes involved in tumor aggres-siveness and epithelial-to-mesenchymal transition, such aslaminin C and fibronectin, as well as ALDH1A1. This latterenzyme is a marker of stemness in several solid tumors includ-ing breast cancer (42) and is associated with higher invasive-ness and poor prognosis (43). An increased expression ofALDH1A1 in normal and malignant breast carcinoma celllines has been reported in response to HER2 overexpression,thus implying that HER2 carcinogenesis and tumor invasioninvolve the modulation of the stem cell compartment, and thatthe expression of ALDH1A1 can play a role in this process (26).The signature of overexpressed transcripts in IFNAR-neuTlesions showed an adverse prognostic value in breast cancerpatients, suggesting that the downstream pathways leading toincreased aggressiveness observed in IFNAR-neuT tumors areshared with human breast cancers.

Many cytokines are known to regulate BCSC homeostasis(44). Exogenous IFNa exerts antiproliferative and proapoptoticeffects on CSCs when administered per se or in combinationwith epigenetic drugs (45). The CSC-inducing activity of CD95is mediated by STAT1 and involves IFN-I pathway (46), andBCSC biology depends on the mir-199-LCOR-IFN-I axis (35).In fact, BCSCs through the overexpression of mir-199 and theconsequent downregulation of LCOR, escape IFN-I signaling,thus leading to a more aggressive tumor phenotype. Our resultsimply that basal concentrations of endogenous IFN-I act asnegative regulators of BCSC homeostasis in neuT mice. IFNAR-neuT tumors showed more BCSCs as assessed not only byALDH1A1 cells expression and enzymatic activity, but also byclonogenic assays and in vivo tumorigenicity after orthotopictransplant in NGS mice. The possibility that BCSC homeostasisis affected by IFN-I signaling rather than being mediated byimmune-related effects is supported by the finding that in vitroexposure to IFN-I–blocking antibody increased ALDH expres-sion in neuT-expressing BCSC and oncogene-positive tumorcell lines. This effect was visible on IFN-competent cells andabsent in their knockout counterparts. No alteration of theALDHþ compartment was induced by IFN-I neutralization onbreast tumor cells negative for HER2/neu, thus suggesting thatthe oncogene can be involved in IFN-mediated regulation ofALDH expression.

CSCs represent key players in malignancy and are respon-sible for reduced success of cancer therapies (47–49).Increased expression of the detoxifying enzyme ALDH1A1may provide CSCs a means to resist chemotherapy, particu-larly cyclophosphamide (50). We show here that the disrup-

tion of IFN-I signaling affects ALDH expression in tumor cellsand BCSCs, indicating a role of endogenous IFN-I in therestriction of BCSCs development and phenotype. Other thanjust highlighting a role of IFN-I signaling in BCSC biology,these findings suggest involvement of endogenous IFN-I inbreast as well as other cancers. The clinical efficacy of anti-HER2-specific immunotherapy combined with trastuzumab isdependent on the IFN-I pathway without requiring lympho-cyte-mediated cytotoxicity (51). It would be of interest toevaluate whether the IFN-I contribution to trastuzumab effi-cacy also involves BCSC biology or chemo-sensitivity, per-haps through modulations of cancer cell ALDH expression inpatients. Activation of IFN-induced genes in residual tumorcells is an early marker of chemotherapy responsiveness inHER2þ breast cancers (52). Thus, our results provide a ratio-nale for using these cytokines in HER2þ breast cancerpatients, perhaps in combination with other treatments tolower the BCSC fraction.

Our results may also suggest mechanisms by which IFN cancontrol tumor development and progression in other humanmalignancies. As an example, in chronic myeloid leukemia(CML), IFNa therapy is receiving renewed interest due to itslong-term efficacy, suggesting that these cytokines can act notonly as immunomodulators, but also can restrain CML progeni-tors responsible for tumor relapse (53). Even in this setting, IFNamay regulate ALDH expression and thus might restrict CSCpopulation.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: P. Sestili, R. Dattilo, M. Ferrantini, A. Zeuner,F. Belardelli, E. Aric�oDevelopment of methodology: L. Castiello, P. Sestili, G. Schiavoni, R. Dattilo,A.M. Pacca, D. MacchiaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): L. Castiello, P. Sestili, G. Schiavoni, R. Dattilo,D.M. Monque, A. Lamolinara, A. SistiguAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): L. Castiello, P. Sestili, R. Dattilo, M. Iezzi, A. Lamo-linara, F. Moschella, E. Aric�oWriting, review, and/or revision of the manuscript: L. Castiello, P. Sestili,G. Schiavoni, R. Dattilo, A. Sistigu, M. Biffoni, F. Belardelli, E. Aric�oAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): P. SestiliStudy supervision: P. Sestili, E. Proietti, E. Aric�o

AcknowledgmentsThe research was supported in part by grants from the Italian Association

for Research against Cancer (AIRC IG16891 and IG 14297). We thankM. Venditti and M. Spada for assistance with preliminary experiments onmice. We are grateful to Paola Di Matteo for kindly providing us themammosphere medium and to Dr. Marta Baiocchi for scientific advice onNSG mice experiments.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received November 24, 2017; revised February 13, 2018; accepted March 29,2018; published first April 5, 2018.

Castiello et al.

Cancer Immunol Res; 6(6) June 2018 Cancer Immunology Research668

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

References1. Vilcek J. Fifty years of interferon research: aiming at a moving target.

Immunity 2006;25:343–8.2. Rizza P, Moretti F, Capone I, Belardelli F. Role of type I inter-

feron in inducing a protective immune response: perspectivesfor clinical applications. Cytokine Growth Factor Rev 2015;26:195–201.

3. Gresser I, Belardelli F,MauryC,MaunouryMT, ToveyMG. Injectionofmicewith antibody to interferon enhances the growth of transplantable murinetumors. J Exp Med 1983;158:2095–107.

4. Dunn GP, Bruce AT, Sheehan KCF, Shankaran V, Uppaluri R, Bui JD, et al.A critical function for type I interferons in cancer immunoediting.Nat Immunol 2005;6:722–9.

5. Fuertes MB, Woo S-R, Burnett B, Fu Y-X, Gajewski TF. Type I interferonresponse and innate immune sensing of cancer. Trends Immunol 2013;34:67–73.

6. SlamonDJ, Clark GM,Wong SG, LevinWJ, Ullrich A,McGuireWL. Humanbreast cancer: correlation of relapse and survival with amplification ofthe HER-2/neu oncogene. Science 1987;235:177–82.

7. Perou CM, Sørlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, et al.Molecular portraits of human breast tumours. Nature 2000;406:747–52.

8. Lollini P-L, Cavallo F, Nanni P, Quaglino E. The promise of preventivecancer vaccines. Vaccines Multidisciplinary Digital Publishing Institute(MDPI); 2015;3:467–89.

9. Muller WJ, Sinn E, Pattengale PK, Wallace R, Leder P. Single-step inductionof mammary adenocarcinoma in transgenic mice bearing the activatedc-neu oncogene. Cell 1988;54:105–15.

10. Pannellini T, Forni G,Musiani P. Immunobiology of HER2/neu transgenicmice. Breast Dis 2004;20:33–42.

11. Aric�o E, Sestili P, Carpinelli G, Canese R, Cecchetti S, Schiavoni G,et al. Chemo-immunotherapy induces tumor regression in a mousemodel of spontaneous mammary carcinogenesis. Oncotarget 2016;7:59754–65.

12. Nanni P, Nicoletti G, De Giovanni C, Landuzzi L, Di Carlo E, Cavallo F,et al. Combined allogeneic tumor cell vaccination and systemic interleukin12 prevents mammary carcinogenesis in HER-2/neu transgenic mice. J ExpMed 2001;194:1195–205.

13. Bartucci M, Svensson S, Romania P, Dattilo R, Patrizii M, Signore M, et al.Therapeutic targeting of Chk1 in NSCLC stem cells during chemotherapy.Cell Death Differ 2012;19:768–78.

14. Cappello P, Triebel F, Iezzi M, Caorsi C, Quaglino E, Lollini P-L, et al.LAG-3 enables DNA vaccination to persistently prevent mammarycarcinogenesis in HER-2/neu transgenic BALB/c mice. Cancer Res2003;63:2518–25.

15. Ross DT, Scherf U, Eisen MB, Perou CM, Rees C, Spellman P, et al.Systematic variation in gene expression patterns in human cancer celllines. Nat Genet 2000;24:227–35.

16. SubramanianA, TamayoP,Mootha VK,Mukherjee S, Ebert BL,GilletteMA,et al. Gene set enrichment analysis: a knowledge-based approach forinterpreting genome-wide expression profiles. Proc Natl Acad Sci USA2005;102:15545–50.

17. Gy€orffy B, Lanczky A, Eklund AC, Denkert C, Budczies J, Li Q, et al. Anonline survival analysis tool to rapidly assess the effect of 22,277 genes onbreast cancer prognosis using microarray data of 1,809 patients. BreastCancer Res Treat 2010;123:725–31.

18. Muller U, Steinhoff U, Reis L, Hemmi S, Pavlovic J, Zinkernagel R, et al.Functional role of type I and type II interferons in antiviral defense. Science1994;264:1918–21.

19. Hosseini H, Obradovic MMS, Hoffmann M, Harper KL, Sosa MS, Werner-KleinM, et al. Early dissemination seedsmetastasis in breast cancer. Nature2016;540:552–8.

20. Mccarty MF, Bielenberg D, Donawho C, Bucana CD, Fidler IJ. Evidence forthe causal role of endogenous interferon-a/b in the regulation of angio-genesis, tumorigenicity, and metastasis of cutaneous neoplasms. Clin ExpMetastasis 2002;19:609–15.

21. Yoshida GJ.Metabolic reprogramming: the emerging concept and associ-ated therapeutic strategies. J Exp Clin Cancer Res 2015;34:111.

22. Sponziello M, Rosignolo F, Celano M, Maggisano V, Pecce V, De Rose RF,et al. Fibronectin-1 expression is increased in aggressive thyroid cancer andfavors the migration and invasion of cancer cells. Mol Cell Endocrinol2016;431:123–32.

23. Kashima H, Wu R-C, Wang Y, Sinno AK, Miyamoto T, Shiozawa T, et al.Laminin C1 expression by uterine carcinoma cells is associated with tumorprogression. Gynecol Oncol 2015;139:338–44.

24. Charafe-Jauffret E, Ginestier C, Bertucci F, Cabaud O, Wicinski J, Finetti P,et al. ALDH1-positive cancer stem cells predict engraftment of primarybreast tumors and are governed by a common stem cell program. CancerRes 2013;73:7290–300.

25. Liu Y, Lv D, Duan J, Xu S, Zhang J, Yang X, et al. ALDH1A1 expressioncorrelates with clinicopathologic features and poor prognosis of breastcancer patients: a systematic review and meta-analysis. BMC Cancer2014;14:444.

26. Korkaya H, Paulson A, Iovino F, Wicha MS. HER2 regulates the mammarystem/progenitor cell population driving tumorigenesis and invasion.Oncogene 2008;27:6120–30.

27. Magnifico A, Albano L, Campaner S, Delia D, Castiglioni F, Gasparini P,et al. Tumor-initiating cells of HER2-positive carcinoma cell linesexpress the highest oncoprotein levels and are sensitive to trastuzumab.Clin Cancer Res 2009;15:2010–21.

28. Grange C, Lanzardo S, Cavallo F, Camussi G, Bussolati B. Sca-1 identifiesthe tumor-initiating cells in mammary tumors of BALB-neuT transgenicmice. Neoplasia 2008;10:1433–43.

29. Lo P-K, Kanojia D, Liu X, Singh UP, Berger FG, Wang Q, et al. CD49fand CD61 identify Her2/neu-induced mammary tumor-initiating cellsthat are potentially derived from luminal progenitors and maintained bythe integrin-TGFb signaling. Oncogene 2012;31:2614–26.

30. Liu JC, Egan SE, Zacksenhaus E. A Tumor initiating cell-enriched prognosticsignature for HER2þ: ERa� breast cancer; rationale, new features, con-troversies and future directions. Oncotarget 2013;4:1317–28.

31. Nguyen LV, Vanner R, Dirks P, Eaves CJ. Cancer stem cells: an evolvingconcept. Nat Rev Cancer 2012;12:133.

32. Gresser I, Tovey MG, Bandu ME, Maury C, Brouty-Boy�e D. Role ofinterferon in the pathogenesis of virus diseases in mice as demonstratedby the use of anti-interferon serum. I. Rapid evolution of encephalomyo-carditis virus infection. J Exp Med 1976;144:1305–15.

33. Belardelli F, Vignaux F, Proietti E, Gresser I. Injection ofmicewith antibodyto interferon renders peritoneal macrophages permissive for vesicularstomatitis virus and encephalomyocarditis virus. Proc Natl Acad Sci USA1984;81:602–6.

34. Gresser I, Vignaux F, Belardelli F, Tovey MG, Maunoury MT. Injection ofmice with antibody to mouse interferon alpha/beta decreases the levelof 20-50 oligoadenylate synthetase in peritoneal macrophages. J Virol1985;53:221–7.

35. Celi�a-Terrassa T, Liu DD, Choudhury A, Hang X, Wei Y, Zamalloa J, et al.Normal and cancerous mammary stem cells evade interferon-inducedconstraint through the miR-199a–LCOR axis. Nat Cell Biol 2017;19:711–23.

36. Sistigu A, Yamazaki T, Vacchelli E, Chaba K, Enot DP, Adam J, et al. Cancercell-autonomous contribution of type I interferon signaling to the efficacyof chemotherapy. Nat Med 2014;20:1301–9.

37. Ferrantini M, Capone I, Belardelli F. Interferon-alpha and cancer: mechan-isms of action and new perspectives of clinical use. Biochimie 2007;89:884–93.

38. Dunn GP, Koebel CM, Schreiber RD. Interferons, immunity and cancerimmunoediting. Nat Rev Immunol 2006;6:836–48.

39. Rolla S, Nicol�o C, Malinarich S, Orsini M, Forni G, Cavallo F, et al. Distinctand non-overlapping T cell receptor repertoires expanded by DNA vacci-nation in wild-type and HER-2 transgenic BALB/c mice. J Immunol 2006;177:7626–33.

40. Cousens LP, Peterson R, Hsu S, Dorner A, Altman JD, Ahmed R, et al. Tworoads diverged: interferon alpha/beta- and interleukin 12-mediated path-ways in promoting T cell interferon gamma responses during viral infec-tion. J Exp Med 1999;189:1315–28.

41. Moasser MM. The oncogene HER2: its signaling and transformingfunctions and its role in human cancer pathogenesis. Oncogene 2007;26:6469–87.

42. Rodriguez-Torres M, Allan AL. Aldehyde dehydrogenase as a marker andfunctional mediator of metastasis in solid tumors. Clin Exp Metastasis2016;33:97–113.

43. Ginestier C, Hur MH, Charafe-Jauffret E, Monville F, Dutcher J, BrownM, et al. ALDH1 is a marker of normal and malignant human mammary

IFN-I Signaling Regulates HER2 Carcinogenesis and Stemness

www.aacrjournals.org Cancer Immunol Res; 6(6) June 2018 669

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

stem cells and a predictor of poor clinical outcome. Cell Stem Cell2007;1:555–67.

44. Korkaya H, Liu S, Wicha MS. Breast cancer stem cells, cytokine networks,and the tumor microenvironment. J Clin Invest 2011;121:3804–9.

45. Buoncervello M, Romagnoli G, Buccarelli M, Fragale A, Toschi E, Parlato S,et al. IFN-a potentiates the direct and immune-mediated antitumor effectsof epigenetic drugs on both metastatic and stem cells of colorectal cancer.Oncotarget 2016;7:26361–73.

46. Qadir AS, Ceppi P, Brockway S, Weichselbaum RR, Yu J, Peter ME, et al.CD95/Fas increases stemness in cancer cells by inducing a STAT1-dependent Type I interferon response. Cell Rep 2017;18:2373–86.

47. Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat RevCancer 2005;5:275–84.

48. Creighton CJ, Li X, Landis M, Dixon JM, Neumeister VM, Sjolund A, et al.Residual breast cancers after conventional therapy display mesenchymalas well as tumor-initiating features. Proc Natl Acad Sci USA 2009;106:13820–5.

49. Khoury T, Ademuyiwa FO, Chandraseekhar R, Jabbour M, DeLeo A,Ferrone S, et al. Aldehyde dehydrogenase 1A1 expression in breast canceris associated with stage, triple negativity, and outcome to neoadjuvantchemotherapy. Mod Pathol 2012;25:388–97.

50. Tomita H, Tanaka K, Tanaka T, Hara A. Aldehyde dehydrogenase 1A1 instem cells and cancer. Oncotarget 2016;7:11018–32.

51. Stagg J, Loi S, Divisekera U, Ngiow SF, Duret H, Yagita H, et al. Anti-ErbB-2 mAb therapy requires type I and II interferons and synergizeswith anti-PD-1 or anti-CD137 mAb therapy. Proc Natl Acad Sci USA2011;108:7142–7.

52. Legrier ME, Bi�eche I, Gaston J, Beurdeley A, Yvonnet V, D�eas O, et al.Activation of IFN/STAT1 signalling predicts response to chemotherapyin oestrogen receptor-negative breast cancer. Br J Cancer 2016;114:177–87.

53. Talpaz M, Hehlmann R, Quint�as-Cardama A, Mercer J, Cortes J. Re-emergence of interferon-a in the treatment of chronic myeloid leukemia.Leukemia 2013;27:803–12.

Cancer Immunol Res; 6(6) June 2018 Cancer Immunology Research670

Castiello et al.

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675

2018;6:658-670. Published OnlineFirst April 5, 2018.Cancer Immunol Res   Luciano Castiello, Paola Sestili, Giovanna Schiavoni, et al.   Progression and Breast Cancer Stem CellsDisruption of IFN-I Signaling Promotes HER2/Neu Tumor

  Updated version

  10.1158/2326-6066.CIR-17-0675doi:

Access the most recent version of this article at:

  Material

Supplementary

  http://cancerimmunolres.aacrjournals.org/content/suppl/2018/04/05/2326-6066.CIR-17-0675.DC1

Access the most recent supplemental material at:

   

   

  Cited articles

  http://cancerimmunolres.aacrjournals.org/content/6/6/658.full#ref-list-1

This article cites 53 articles, 15 of which you can access for free at:

  Citing articles

  http://cancerimmunolres.aacrjournals.org/content/6/6/658.full#related-urls

This article has been cited by 2 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

  .pubs@aacr.orgat

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

  Permissions

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

.http://cancerimmunolres.aacrjournals.org/content/6/6/658To request permission to re-use all or part of this article, use this link

on October 24, 2020. © 2018 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst April 5, 2018; DOI: 10.1158/2326-6066.CIR-17-0675