A Novel ACKR2-Dependent Role Of

Published OnlineFirst March 8, 2019; DOI: 10.1158/1078-0432.CCR-18-1294

Translational Cancer Mechanisms and Therapy Clinical Cancer Research A Novel ACKR2-Dependent Role of Fibroblast- Derived CXCL14 in Epithelial-to-Mesenchymal Transition and Metastasis of Breast Cancer Elin Sjoberg€ 1, Max Meyrath2, Laura Milde3, Mercedes Herrera1, John Lovrot€ 1, Daniel Hagerstrand€ 1, Oliver Frings1, Margarita Bartish1, Charlotte Rolny1, Erik Sonnhammer4, Andy Chevigne2, Martin Augsten1, and Arne Ostman€ 1

Abstract

Purpose: Fibroblasts expressing the orphan chemokine graft model. Furthermore, tumor cells primed by CXCL14 CXCL14 have been previously shown to associate with poor fibroblasts displayed enhanced lung colonization after tail- breast cancer prognosis and promote cancer growth. This study vein injection. By loss-of function experiments, the atypical explores the mechanism underlying the poor survival associa- G-protein–coupled receptor ACKR2 was identified to mediate tions of stromal CXCL14. CXCL14-stimulated responses. Downregulation of ACKR2, or Experimental Design: Tumor cell epithelial-to-mesenchymal CXCL14-induced NOS1, attenuated the pro-EMT and migra- transition (EMT), invasion, and metastasis were studied in in tory capacity. CXCL14/ACKR2 expression correlated with EMT vitro and in vivo models together with fibroblasts overexpres- and survival in gene expression datasets. sing CXCL14. An approach for CXCL14 receptor identification Conclusions: Collectively, the findings imply an autocrine included loss-of-function studies followed by molecular and fibroblast CXCL14/ACKR2 pathway as a clinically relevant functional endpoints. The clinical relevance was further stimulator of EMT, tumor cell invasion, and metastasis. The explored in publicly available gene expression datasets. study also identifies ACKR2 as a novel mediator for CXCL14 Results: CXCL14 fibroblasts stimulated breast cancer EMT, function and thereby defines a pathway with drug target migration, and invasion in breast cancer cells and in a xeno- potential.

Introduction expression and shorter survival in a population-based breast cancer cohort. Notably, epithelial expression of CXCL14 did not Death of patients with breast cancer is almost exclusively due have an impact on breast cancer outcome (1). to metastatic disease. Metastasis develops through a multistep During the early steps of metastasis development, tumor cells process, involving tissue invasion, intravasation, survival in the lose cell-to-cell contacts and epithelial characteristics and instead bloodstream and lymph system, extravasation, and tissue colo- gain mesenchymal traits that allow them to invade the surround- nization. This study develops recent correlative studies that have ing tissue and metastasize; a process termed epithelial-to-mesen- indicated clinical relevance, in breast cancer, of stroma-derived chymal transition (EMT; ref. 2). expression of the chemokine CXCL14 by demonstrating signifi- EMT is controlled by distinct transcriptional programs activated cant and independent associations between high stromal CXCL14 by specific transcription factors, including Snail, Slug, Twist, Zeb, and Gsc. Activation of these factors ultimately leads to the loss of epithelial markers including E-cadherin and cytokeratins, and the 1 Department of Oncology-Pathology, Cancer Center Karolinska (CCK), Karo- upregulation of mesenchymal markers, such as vimentin, alpha- linska Institutet, Stockholm, Sweden. 2Department of Infection and Immunity, smooth muscle actin (aSMA), and matrix-degrading enzymes (3). Immuno-Pharmacology and Interactomics, Luxembourg Institute of Health (LIH), Esch-sur-Alzette, Luxembourg. 3Division for Vascular Oncology and Although the classical paradigm attributing EMT a crucial role in Metastasis, German Cancer Research Center (DKFZ), Heidelberg, Germany. the process of metastasis has been recently challenged by studies 4Stockholm Bioinformatics Center, Department of Biochemistry and Biophysics, in genetic mouse models, other recent studies including in vivo Stockholm University, Science for Life Laboratory, Stockholm, Sweden. imaging approaches demonstrated that cancer cells displaying an Note: Supplementary data for this article are available at Clinical Cancer EMT phenotype give rise to metastases (4–6). Research Online (http://clincancerres.aacrjournals.org/). Induction of EMT can occur in a paracrine manner by secreted M. Augsten and A. Ostman€ contributed equally to this article. factors from cells of the tumor stroma, as for example, the cancer- associated fibroblasts (CAF; refs. 7, 8). CAFs constitute a hetero- Current address for M. Augsten: amcure GmbH, Eggenstein-Leopoldshafen, geneous cell population that contributes to cancer growth and Germany. spread by secretion of a variety of protumorigenic factors, includ- € Corresponding Author: Arne Ostman, Karolinska Institutet, Cancer Center ing soluble factors. Among these CAF-secreted factors implicated Karolinska, R8:03, Karolinska University Hospital, Stockholm 17176, Sweden. in EMT and metastasis are chemokines, proteins of a size between Phone: 468-5177-0232; Fax: 468-339-031; E-mail: [email protected] 8 and 14 kDa that stimulate directed cell migration by creating a doi: 10.1158/1078-0432.CCR-18-1294 gradient along which cell types expressing the corresponding 2019 American Association for Cancer Research. receptor travel (7, 9). Chemokines bind to the pertussis-sensitive

www.aacrjournals.org OF1

Sjoberg€ et al.

Hs578t were cultured in DMEM (Hyclone), supplemented with Translational Relevance 10% FBS (Hyclone), 1% glutamine (Hyclone) and 1% penicillin/ Autocrine fibroblast CXCL14/ACKR2 signaling is shown to streptomycin (Hyclone). DMEM-F12 (Hyclone), supplemented induce EMT, migration, and metastasis and to correlate with with horse serum (Biochrom), 1% glutamine, and 1% penicillin/ worse survival in patients with breast cancer. The identification streptomycin was used for culturing the MCF10-DCIS cell line. of ACKR2 as a novel component in the signaling of the orphan Starvation was performed in medium without serum. All cells chemokine CXCL14 is relevant for further biomarker studies were maintained at 37 C in humidified air with 5% CO2. NIH-ctr and suggests novel targeting opportunities. and NIH-CXCL14 fibroblasts have been characterized earlier, and fibroblasts secrete physiologic levels of CXCL14 (22). Cell lines were purchased from ATCC or received from collaboration part- ners and continually tested for Mycoplasma infection during the Gai-subfamily of G-protein–coupled receptors (GPCR) that study. The identity of the cell lines used was confirmed by short engage different signaling pathways including ERK1/2, PI3K/AKT, tandem repeat (STR) profiling at Uppsala Genome Center. All and calcium signaling (10). Besides the classical chemokine experiments were performed with cells of passage 3–20. receptors, there is a subfamily of atypical chemokine receptors Western blot analyses used antibodies against p-ERK (#9101), (ACKR) that are predominantly involved in sequestration of ERK (#9102), E-cadherin (#3195), Snail (#3879), NOS1 (#4234) chemokines (11). (Cell Signaling Technology), b-actin (A1978), and a-tubulin In cancer, chemokines are involved in the recruitment of (T5201; Sigma-Aldrich). Primary antibodies detecting E-cadherin various cell types into tumors and thereby affecting inflammation, (#3195), Cytokeratin 8/18 (#4546), and PDGFbR (#3169; Cell angiogenesis, tumor growth, invasion, and metastasis (12). A Signaling Technology) together by fluorescent-linked secondary paracrine chemokine cross-talk between stromal cells and tumor antibodies, were used for immunofluorescence staining of xeno- cells, involving effects on EMT, has been demonstrated to enhance graft tumors. formation of metastases (13–16). Expression of certain chemo- Pertussis toxin was purchased from Sigma-Aldrich and recom- kines in distant tissues has also been reported to determine binant CXCL14 and CCL5 from R&D Systems and PeproTech. metastasis formation in specific organs, a process termed meta- Alexa Fluor 647–coupled chemokines were purchased from static tropism (17). Almac. The orphan chemokine CXCL14, earlier designated BRAK, MIP-2g, BMAC, or KS1 stimulates migration of various immune RNA isolation, cDNA synthesis, and qRT-PCR analysis cells, including B cells, NK cells, and monocytes, but not RNA was isolated from xenograft tumors or overnight starved T cells (18–21). CXCL14 expression has been shown to be cells using GeneElute Mammalian Total RNA Miniprep Kit upregulated in CAFs, as compared with normal fibroblasts, in (Sigma-Aldrich). cDNA synthesis was performed with the Super- human breast and prostate cancer (22, 23). Experimental studies Script III First-Strand Synthesis System for RT-PCR (Invitrogen), exploring the function of CXCL14 during tumor progression and using PolydT primers, in accordance with the manufacturer's metastasis formation have demonstrated context-dependent pro- instructions. The qRT-PCR reaction using SYBR Green Universal and antitumoral effects. The reasons for these effects remain PCR Master Mix (Applied Biosystems) was performed with the largely unknown and could possibly dependent on the cell type 7500 Real-Time PCR System (Applied Biosystems). The concen- that express the chemokine and on the profile of chemokine tration of primers was 200 nmol/L, and expression levels receptors and ACKRs expressed. Some studies with CXCL14 over- were normalized to the housekeeping gene glyceraldehyde-3- expression in tumor cells have demonstrated antitumoral effects phosphate dehydrogenase (GAPDH). The primer sequences for of this chemokine (24). In contrast, tissue culture and mouse primers obtained from Sigma-Aldrich are shown in Supplemen- cancer model studies of breast and prostate cancer have demon- tary Table S1. Other primers were QuantiTect primers obtained strated protumoral effects of CXCL14 expressed by stromal fibro- from Qiagen. blasts, associated with CXCL14-induced changes in fibroblast secretomes (22, 23, 25). Immunoblotting, immunofluorescence, and secretome The tumor-promoting roles of CAF-derived CXCL14 have analyses been shown to depend on nitric oxide synthase 1 (NOS1) Immunoblotting and immunofluorescence analyses were per- and involve stimulation of angiogenesis and recruitment of formed as described previously (22). In short, for analysis of macrophages (25). CXCL14-induced p-ERK signaling by immunoblotting, overnight Continued exploration of the roles of CXCL14 in tumor biol- serum-starved cells were stimulated with recombinant CXCL14 ogy and possible exploitation of this chemokine as a therapeutic (R&D Systems or PeproTech) for 7 minutes. For experiments with target depend on the identification of critical mediators of pertussis toxin, cells were treated with the toxin for 1 hour at 37C CXCL14 signaling, including receptors. This study therefore and 5% CO2 prior to CXCL14 stimulation. Cell lysates were aimed at providing a better understanding of the molecular prepared and SDS/PAGE was performed followed by transfer to mechanism underlying the documented poor survival association polyvinylidene difluoride membranes (Millipore). Immunoblot- of stroma-derived CXCL14. ting with p-ERK and ERK antibodies (Cell Signaling Technology), diluted 1:1,000, were performed and signals were detected with ImageQuant LAS4000 (GE Healthcare) and quantified using Materials and Methods ImageJ (http://imagej.nih.gov.proxy.kib.ki.se/ij). Cell lines and chemicals Analysis of EMT markers was performed for 48–72 hours after The mouse fibroblast cell line NIH3T3 (and derivatives), the stimulation with fibroblast-conditioned medium or coculture of breast cancer cell lines MCF7, SKBR3 MDA-MB-231, 4T1, and breast cancer cells and fibroblasts at a 1:1 ratio. The conditioned

OF2 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

medium was generated by seeding the same number of NIH-ctr 100 nmol/L of Alexa Fluor 647–labeled CCL5 or CXCL14 for and NIH-CXCL14 cells. The next day, medium was changed to 90 minutes on ice. After a washing step, Zombie Green Fixable DMEM containing low serum (1% FBS) and the conditioned Viability Kit (BioLegend) was used to gate on living cells only. The medium was collected for 48 hours, sterile filtered with a 0.2-mm experiments were performed in PBS containing 1% BSA and 0.1% filter. Antibodies for immunoblotting, including E-cadherin NaN3 (FACS buffer). Chemokine binding was quantified by mean and Snail, were diluted 1:1,000 and NOS1 1:500. For immunofluorescence intensity (MFI) of 10,000 gated cells on a BD FACS fluorescence analysis, E-cadherin, Cytokeratin 8/18, and PDGFbR Fortessa Cytometer (BD Biosciences). antibodies were diluted 1:100. To analyze the secretome from CXCL14 fibroblasts, a protein b-arrestin recruitment assay array was performed. A total of 7.0 105 NIH-CXCL14 and Chemokine-induced b-arrestin-1 recruitment to ACKR2 was NIH-ctr cells were seeded in 10-cm dishes. The next day, monitored by Nanoluciferase complementation assay (NanoBiT, medium was changed from medium with 10% FBS to low Promega) as described previously (26, 27). A total of 5 106 HEK serum (1% FBS). Conditioned medium was collected after cells were plated in 10-cm culture dishes and 24 hours 48-hour incubation at 37C, sterile filtered, and stored at later transfected with plasmids containing human b-arrestin-1 20C. Aliquots of conditioned medium (1 mL) from N-terminally fused to LgBiT and ACKR2 C-terminally fused to NIH-CXCL14 and NIH-ctr fibroblasts were subjected to the SmBiT. Twenty-four hours posttransfection, cells were harvested, proteome profiler (mouse angiogenesis array kit ary015, R&D incubated 30 minutes at 37C with 200-fold diluted Nano- Systems), performed according to the manufacturer's protocol. Glo Live Cell substrate, and distributed into a white 96-well plate Array images obtained were analyzed using the ImageJ soft- (1 105 cells per well). b-arrestin-1 recruitment was measured ware. For all spots, the average background signal from negative over 25 minutes with a Mithras LB940 luminometer (Berthold control spots was subtracted. The average signal from positive Technologies). For each experiment, signal measured with a control spots of each membrane was used to normalize the two saturating concentration (300 nmol/L) of the full agonist (i.e., different types of fibroblast conditions. The relative differences CCL5) was set as 100%. in protein expression between NIH-CXCL14 and NIH-ctr cells were expressed as ratio (fold of NIH-ctr). Bioinformatic analyses Sequence analyses for novel chemokine receptors started from In vitro growth, migration, and invasion assays Pfam family PF00001 (7tm_1), using the 1679 human domains To study the effect of siRNA and short hairpin RNA (shRNA)- from 1,664 human sequences in the full alignment. A neighbor- mediated knockdown of ACKR2 on growth of NIH-ctr and NIH- joining tree of these was built using scoredist distances with CXCL14 fibroblasts, 2 104 cells were seeded per well of a 24-well Belvu (28). A subtree of 114 cytokines was cut out. After removing plate (Sarstedt), in quadruplicates in serum-reduced media. After fragment sequences and >99% identical sequences, 32 sequences 3 days of culture, AlamarBlue (Bio-Rad) was used to determine the were left. Following exclusion of DUFFY, not being part of the cell number. A total of 350 mL of a 1:10 dilution of AlamarBlue dye 23,760 homologs in the Pfam family, a candidate list of 31 in DMEM was added to each well and cells were incubated at 37C candidates was established (Supplementary Fig. S1). with 5% CO2 for 2.5 hours. To measure the conversion of the dye, 100 mL was transferred into each well of a white 96-well plate Transfection of siRNA and generation of stable cell lines (Costar) and absorbance was measured at a wavelength of For transfection of siRNA, 1.0–2.0 105 cells were seeded 570 nm. in 6-well plates (Sarstedt) and transfected for 48 hours with To study cell migration of breast cancer cells, a transwell 100 nmol/L siRNA (ACKR2 target sequence; 50-CTCAATTAGCGT- migration assay was used. A total of 2.5 104 breast cancer cells TATTGGCAA-30; Qiagen) using HiPerFect transfection reagent were seeded in transwell inserts (Corning) with an 8.0-mm pore- (Qiagen). To determine the efficiency of siRNA-mediated knock- sized membrane and placed in a 24-well plate (Corning), in down of genes of interest, RNA extraction, cDNA synthesis, and duplicates. For analysis of CXCL14 fibroblast–induced migration, qRT-PCR analysis was performed. 2.5 104 CXCL14- or control fibroblasts were seeded in the Fibroblast derivatives with stable knockdown of ACKR2 were bottom chamber. established using shRNA procedures, as described previously (25). For migration experiments, lasting for 16–24 hours, the inside In brief, phoenix cells were transfected with 2 unique 29mer of the insert was wiped with cotton swabs, washed with PBS, and shRNA constructs against ACKR2 (gene ID ¼ 59289; shACKR2: the cells were fixed in ice-cold methanol. The membrane was cut A and shACKR2:B) or nontargeting control shRNA (shCtr) in out and placed on Superfrost Plus slides (Menzel-Gl€aser) and retroviral vectors (Origene). After 48 hours, the supernatant was stained with Vectashield Mounting Medium with DAPI (Vector collected, filtered, and added to NIH-ctr and NIH-CXCL14 cells Laboratories). Quantification of cell migration was performed by for 5 hours. Cells were subsequently selected in 30 mg/mL blas- counting cell nuclei of the migrated cells. The same principles ticidin for 2 weeks. The knockdown of ACKR2 was confirmed by were used for the invasion assays, with inserts containing a qRT-PCR (Fig. 1A). Generation of NIH-ctr and NIH-CXCL14 thin Matrigel layer (Corning). Invasion was allowed to occur for fibroblasts with a stable knockdown of NOS1 have been described 72 hours. previously (25).

Chemokine binding assay Animal experiments HEK-293 cells or HEK-293 cells stably expressing ACKR2 The animal experiments were conducted in accordance with (under 200 mg/mL hygromycin selection) were distributed national guidelines and approved by the Stockholm North Ethical into 96-well plates (2 105 cells per well) and incubated Committee on Animal Experiments. The generation of xenograft with increasing concentrations ranging from 10 pmol/L to tumors was performed as described previously (25).

www.aacrjournals.org Clin Cancer Res; 2019 OF3

Sjoberg€ et al.

AB )

r 1.5 ls t e C v e sh

l *** *** / r t A 1 ** * -c RN H I m N

2 0.5 of R K old f AC ( 0 r r t :A t :A :B C 2 2:B C 2 2 h R s R /sh R R r/ K K 4 K K t C C 1 C C -c A A L A A IH h h C h h /s /s X /s /s N tr tr 4 4 -c c -C 1 1 - H L L H IH I C C NI N N X X -C -C IH IH N N

C D

) 35 * ** 2.5 ***** ls e 30 hCtr

2.0 NIH-ctr/shctr s

r/ 25 t NA lev /ERK 1.5 20

NIH-ctr/ R m RK 1.0 shACKR2:A 15 E - f unstim control) p 10 0.5 o NIH-ctr/ 5 shACKR2:B NOS1 0.0 (fold of NIH-c 0 (fold 0 100 200 400 r A :B t :A :B Ctr 2: C 2 2 rCXCL14 (ng/mL) h R h R2 /s R R /s K K 4 K K C C 1 C C -ctr A A L A A h h C h h IH /s /s /s N tr r/s X 4 4 ct -C 1 -c L L1 IH IH- C N N NIH XC X -C C IH IH- N N E F

2.5 *** ** ** n

o 2 i at

r 1.5 fe 1 NIH-ctr of NB shCtr) roli d P 0.5 NIH-CXCL14 (fol 0 r t :A B C 2 2: h R s KR K C C A A h h s s

Figure 1. ACKR2 mediates CXCL14-stimulated signaling in fibroblasts. A, The suppression of ACKR2 expression following introduction of two different ACKR2-targeting shRNA (A and B) in NIH-ctr and NIH-CXCL14 fibroblasts was analyzed by qRT-PCR. Results are obtained from three independent experiments. B, The activation of ERK1/2 signaling following CXCL14 stimulation was monitored by Western blot analysis in NIH-3T3 fibroblasts with and without stable downregulation of ACKR2. C, Quantifications of three independent experiments from B. unstim, unstimulated. Analysis of Nos1 transcript (qRT-PCR; D) and Nos1 protein (Western blot) levels (E) in NIH-ctr and NIH-CXCL14 derivatives with or without stable downregulation of ACKR2. E, One representative blot together with the quantification of three independent experiments. F, The growth of NIH-ctr and NIH-CXCL14 fibroblasts with or without stable downregulation of ACKR2 was evaluated by the AlamarBlue assay (see details in "Materials and Methods") after culture for 3 days in serum-reduced medium. The results of three independent experiments are summarized in the figure. P values were derived from unpaired two-sided Student t tests. , P < 0.001; , P < 0.01; and , P < 0.05. Error bars, SD or SEM.

OF4 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

The analyses of lung metastasis formation were performed after R package nlme [R syntax lme(groupedData(ScaledEMTscore tail-vein injection of 2 105 breast cancer cells in three groups of scaledCXCL14 | cohort, data))]. All gene expression data 8-week-old female SCID mice, without any further randomiza- analysis was performed in R/Bioconductor and SPSS 21.0. The tion. The sample size of 10 mice in each group was determined by EMT score and CXCL14 data are first centered and scaled to unit the 3R criteria together with previous experience. After 4 weeks, SD within cohorts to facilitate comparison between cohorts. mice were sacrificed and lungs were collected, washed in PBS Hence, the slope of a linear regression in each cohort in Supple- (Hyclone), and snap frozen (for qRT-PCR analysis) or embedded mentary Fig. S2A is mathematically equivalent to the Pearson in optimal cutting temperature medium (OCT) and snap frozen correlation coefficient. This equivalence does not hold in sub- (for histologic analysis). No animals were excluded from the groups as in Supplementary Fig. S2B. study. The TCGA gene expression data (Fig. 3A; Supplementary Figs. For RNA extraction, 1 mL of TRIzol (Life Technologies) was S3 and S4) are displayed as Z-scores obtained from cBioportal. For added to the lung tissue and homogenized with a polytron 3 the breast cancer gene expression dataset, CXCL14low and times, each for 10 seconds. A volume of 0.2-mL chloroform was CXCL14high groups were divided according to the 50 percentile, added and the samples were shaken for 15 seconds and incubated whereas ACKR2low and ACKR2high expression was determined by for 2 minutes at room temperature. The samples were centrifuged fitting a mixture of two normal distributions using the R package at 12,000 g for 15 minutes at 4C and the aqueous phase was mixtools, resulting in a dichotomization below and above the placed in a new tube. RNA was precipitated by addition of 0.5 mL 90.7 percentile. For the TCGA ovarian and prostate cancer gene 100% isopropanol (Merck) to the samples and incubation for expression datasets, the ACKR2low and ACKR2high subgroups were 10 minutes at room temperature. Following centrifugation at divided by the 4th quartile. 12,000 g for 10 minutes at 4C, the RNA pellet was washed with 75% ethanol and air dried before resuspension in nuclease- Statistical analysis free water (Ambion). The RNA concentration of each sample was Statistical calculations were performed using Excel 2011 for determined using the Nanodrop ND-1000 spectrophotometer Mac (Microsoft Office), R/Bioconductor, or the statistical package (NanoDrop Technologies). cDNA were synthetized and subse- SPSS 21.0 (SPSS Inc.). All data are expressed as mean or median quent qRT-PCR analysis was performed as described previously, values, and error bars represent the SD or SEM. Data that are being with human- and mouse-specific primers. Percentages of human statistically compared, relevant for the conclusions, exhibit sim- cells in mice lungs were quantified as described in Malek and ilar variation. Statistical differences between groups were deter- colleagues (29). mined using two-sided, unpaired Student t test or Mann–Whitney Ten frozen sections (10 mm) were made from the OCT-embed- U test. Pearson correlation was used to analyze correlations ded lungs. Five sections were thrown in between each saved frozen between different parameters. The Kaplan–Meier method and section. Stainings were performed with the human-specific anti- log-rank test method was performed to estimate overall survival. body Stem121 (Takara Bio) and the positive cells were counted in Cox proportional hazards model was used to compare HRs in each section and results are displayed as average number per lung both uni- and multivariate analyses. The multivariate analysis (Fig. 2D). The tail-vein injections were performed blinded and the included known clinical relevant parameters, including T-stage, analyses of lung metastasis were performed unblinded. N-stage, M-stage, and molecular subtypes of breast cancer. Results are presented in the multivariate analysis as HRs Clinical cohorts including 95% confidence intervals (95% CI). For all analyses, The relation between CXCL14 transcript abundance and EMT P values below 0.05 were considered significant (, P < 0.05; as assessed by an EMT gene expression signature was investigated , P < 0.01; , P < 0.001). All relevant data are available from in clinical breast cancer cohorts with publicly available transcrip- the authors. tome data: the Uppsala (30), Stockholm (31), Rotterdam (32), and METABRIC (33) cohorts, as well as The Cancer Genome Atlas (TCGA; ref. 34). Each study site in METABRIC is treated as a Results separate cohort. TCGA gene expression datasets for breast cancer, CXCL14 fibroblasts induce loss of epithelial markers and ovarian cancer, and prostate cancer was used to investigate the enhance expression of mesenchymal markers and EMT levels of EMT markers and survival associations in CXCL14/ transcription factors in breast tumor xenografts ACKR2 subgroups was additionally performed in gene expression The findings of stromal CXCL14 as a poor prognostic marker in datasets for bladder cancer, clear cell renal cell carcinoma breast cancer (1), prompted analyses of the potential effects of (ccRCC), colorectal cancer, esophageal cancer, glioblastoma mul- CAF-derived CXCL14 on tumor cell invasion and metastasis. tiforme (GBM), low-grade glioma, head and neck cancer (HNCC), We first studied the expression levels of EMT markers as an lung adenocarcinoma, pancreas cancer, and stomach cancer (34). indicator of a proinvasive phenotype in xenograft tumors formed following coinjection of the epithelial breast cancer cell line MCF7 Gene expression data analysis and either control fibroblasts or CXCL14 fibroblasts (25). An EMT gene expression signature score was derived for each Immunofluorescence staining of xenograft tumor sections tumor in the panel of clinical cohorts as described previously (35). demonstrated a significant loss of tumor cell E-cadherin and In brief, the signature was identified from the changes in gene Cytokeratin 8/18 in CXCL14 breast tumors, as compared with expression shared by upregulation of Gsc, Snail, Twist, and control tumors (Fig. 4A; Supplementary Fig. S5A; Supplementary TGFb1 and by downregulation of E-cadherin. PAM50 intrinsic Table S2). NOS1, an oxidative stress–induced enzyme, was pre- subtype classification was performed as described previously (36). viously shown to functionally contribute to the protumorigenic Individual patient data meta-analysis (IPDMA) of all cohorts actions of CXCL14-expressing fibroblasts (25). EMT markers with a linear mixed-effects model was performed using the were therefore also analyzed in MCF7/NIH-CXCL14 tumors

www.aacrjournals.org Clin Cancer Res; 2019 OF5

Sjoberg€ et al.

A − NIH-ctr NIH-CXCL14

Cancer cells MCF7

Fibroblasts

Figure 2. DCIS CXCL14-expressing ﬁbroblasts enhance migration and stimulate lung colonization of breast cancer cells. A, MCF7, DCIS, and SKBR3 cells were allowed to migrate toward NIH-ctr or NIH-CXCL14 SKBR3 ﬁbroblasts in a transwell migration assay for 24 hours. Migration was determined by counting DAPI- ** stained cells that had moved ** *** through an 8-mm pore size * membrane of the transwell (see ) 3 2 * 30 *** details in "Materials and 2 20 Methods"). Results are derived 1 from three independent experiments and are presented as Migration Migration Migration 1 Migration 10 fold of MCF7, DCIS, or SKBR3 cells (fold of control (fold of control) control) of (fold

(fold of control) alone. B, MCF7 cells primed for 0 0 0 MCF7 + ++ DCIS + ++ SKBR3 + ++ 72 hours in a transwell coculture NIH-ctr − + − NIH-ctr − + − NIH-ctr − + − assay with NIH-ctr, or NIH-CXCL14 NIH-CXCL14 − − + NIH-CXCL14 − − + NIH-CXCL14 − − + fibroblasts were injected into the tail-vein of 8-week-old SCID mice (n ¼ 10). C, Lungs were harvested B C 4 weeks after injection of the ** cancer cells. The number of MCF7 0.4 *** cells (human origin) in mouse 0.3 lungs was semiquantitatively Fibroblasts ** assessed by qRT-PCR using 0.2 human- and mouse-specific 0.1 primers (see details in "Materials in mice lungs in % Human cells % Human cells 0 and Methods"). D, Lung sections Cancer cells 7 tr 4 from the tail-vein experiment were F -c H CL1 MC NI stained for the human-specific + -CX 7 IH marker Stem121. The number of CF N M + 7 MCF7 cells in the lung was counted CF M in 10 sections/lung and is depicted as average (n ¼ 10). Arrowheads D MCF7 NIH-ctr + MCF7 NIH-CXCL14 + MCF7 indicate tumor cells. Scale bar, *** 100 mm. nr, number. P values were 30 *** derived from unpaired two-sided 25 Student t tests. , P < 0.001; 20 , P < 0.01; and , P < 0.05. 15 Error bars, SEM. 10 5

Mean nrof cells/lung MCF7 0 ) ) 7 tr 4 F c 1 C - L IH C M N X ( C 7 - F IH C (N M 7 F C M

with a stable NOS1 downregulation in the fibroblasts. Impaired Analyses of mRNA levels of EMT markers substantiated these expression of NOS1 was sufficient to reverse the decrease in findings and uncovered a reduction of epithelial markers includ- epithelial markers induced by CXCL14 fibroblasts in vivo ing E-cadherin (CDH1), Cytokeratin 18 (KRT18), and Cytokeratin (Fig. 4A; Supplementary Fig. S5A; Supplementary Table S2). 8(KRT8), and increase in mesenchymal markers including Furthermore, CXCL14 fibroblasts also reduced cancer cell Cyto- Vimentin (VIM), a-SMA (ACTA2), and MMP2 (MMP2), and an keratin 8/18 levels, in a NOS1-dependent manner, in a xenograft increase in EMT transcription factors including Slug (SNAI2), and coinjection model of prostate cancer (Supplementary Fig. S5B; Twist (TWIST1) in CXCL14-breast tumors (Fig. 4B). Furthermore, Supplementary Table S2). analyses of MCF7/NIH-CXCL14 tumors with a stable NOS1

OF6 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

A CDH1 KRT8 KRT18 KRT19 TSPAN13 SNAI1 SNAI2 SNAI3 TWIST1 GSC ZEB1 ZEB2 CDH2 VIM ACTA2 FN1 MMP2 MMP3 MMP9 COL1A2 COL3A1 COL5A2 SPARC TCF4 WNT5A WNT5B CALD1 GNG11 ITGAV ITGA5 1

0.8

0.6

0.4 es) u l

nva 0.2

(mea 0 e

-0.2 Z-scor

-0.4

-0.6 CXCL14low CXCL14low CXCL14high CXCL14high ACKR2low ACKR2high ACKR2low ACKR2high -0.8

1.0 HR= 2.494 (95% CI = 1.218-5.104)

0.8

0.6

0.4 Survival probability

0.2

CXCL14high/ACKR2high 0.0 Rest P = 0.01

0.00 50.00 100.00 150.00 200.00 250.00 300.00 Months

Figure 3. Patients with breast cancer expressing high levels of CXCL14 and ACKR2 show enhanced EMT and adverse overall survival. A, Z-scores of EMT genes in the TCGA breast cancer gene expression dataset, in patients divided in different subgroups with high or low expression levels of CXCL14 and ACKR2. B, Kaplan–Meier analysis of the CXCL14high/ACKR2high subgroup compared with the rest of the population (n ¼ 1,100 patients). P value was derived from log-rank test, and HRs including conﬁdence intervals (CI) were derived from univariate Cox regression analyses.

www.aacrjournals.org Clin Cancer Res; 2019 OF7

Sjoberg€ et al.

MCF7 + NIH-ctr/shCtr

MCF7 + NIH-CXCL14 /shCtr

Figure 4. MCF7 + CXCL14 affects regulators of EMT and invasion in a xenograft tumor NIH-CXCL14 model of breast cancer. A, /shNOS1 Xenograft tumors of MCF7 cells coinjected with NIH-ctr or NIH-CXCL14 ﬁbroblasts were stained for E-cadherin. White B arrowheads indicate epithelial cells with weak E-cadherin expression. 60 Scale bar, 50 mm. B, qRT-PCR * * ** MCF7+ NIH-ctr/shCtr analysis of transcript levels of genes encoding EMT-regulated markers in 50 MCF7+NIH-CXCL14/shCtr xenograft tumors. The analysis comprises epithelial marker on MCF7 + NIH-CXCL14/shNOS1 si 40 (E-cadherin, Cytokeratin 18, and s e

r Cytokeratin 8), mesenchymal * ** marker [Vimentin, a-SMA (encoded 30 by ACTA2), and MMP2], and the eexp

iv ** EMT transcription factors Slug and at 20 * Twist (n ¼ 5). C, Staining of el * * R ** xenograft tumors formed following coinjection of MCF7 cells and NIH- (Fold of CXCL14 in10 NIH-ctr) ** * * ctr or NIH-CXCL14 fibroblasts with the human-specific antibody 0 Stem121. Budding cells (cluster of up 4 8 1 2 2 T 2 to three cells) in the border of the T H IM A I P L1 T18 R D V T A IS M C R K C N W tumor were counted in 10 vision X K C A S T M C fields, and results are shown as mean number (nr) of cells/vision field (n ¼ 5). Scale bar, 100 mm. C P values were derived from unpaired two-sided Student t tests. MCF + NIH-ctr MCF7 + NIH-CXCL14 *** , P < 0.001; , P < 0.01; and g 12 P n , < 0.05. Error bars, SEM. i

d 10 8 on field fbud i 6 ro vis n / 4 s l n a 2 cel Me 0 tr 4 -c 1 IH CL N X + 7 -C IH CF N + M 7 F C M

knockdown in fibroblasts demonstrated NOS1 dependency of a more invasive growth pattern. As shown in Fig. 4C, MCF7/NIH- these gene expression changes (Fig. 4B). However, mesenchymal CXCL14 tumors displayed a more invasive growth pattern with a markers, including Fibronectin (FN1), FAP (FAP), and MMP9 significantly higher number of budding cells in the tumor periph- (MMP9) were upregulated in CXCL14 tumors independently of ery, as compared with MCF7/NIH-ctr tumors. NOS1 expression (Supplementary Fig. S5C). These data collectively demonstrate that cancer cells coinjected Additional analyses were performed to investigate whether with CXCL14 fibroblasts exhibit enhanced EMT and invasion the CXCL14-dependent EMT phenotype also was associated with in vivo.

OF8 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

CXCL14 fibroblasts stimulate EMT in vitro and induce a Furthermore, invasion of MCF7 and SKBR3 cells through a layer mesenchymal morphology of breast cancer cells of Matrigel was enhanced by NIH-CXCL14 fibroblasts, as com- Next, the direct impact of CXCL14-expressing fibroblasts pared with control fibroblasts (Supplementary Fig. S7C). on the modulation of EMT markers in MCF7 breast cancer Together, these results demonstrate that the CXCL14 fibro- cells and MCF10-DCIS (ductal carcinoma in situ, DCIS) cells blast–induced changes in EMT markers are accompanied by an was investigated under in vitro coculture conditions. Western enhanced capacity of breast cancer cells to migrate and invade in blot analysis demonstrated a reduction of E-cadherin in both vitro. MCF7- and DCIS cells after direct coculture with CXCL14 fibroblasts that was not seen with control fibroblasts CXCL14 fibroblasts enhance lung colonization of MCF7 cells (Fig. 5A). To confirm that the downregulation occurred in the following tail-vein injection breast cancer cells and did not reflect changes in fibroblast The findings of CXCL14-induced effects on migration, inva- properties or abundance, immunofluorescence costaining for sion, and EMT, together with previous findings revealing a pro- E-cadherin or Cytokeratin 8/18 together with the fibroblast tumorigenic role of CXCL14, prompted in vivo studies to explore marker PDGFRb was performed on MCF7 fibroblast cocultures. the effects of CXCL14 fibroblasts. As shown in Fig. 5B, there was a specificlossofE-cadherin(left) Tail-vein experiments monitor the ability of cancer cells to and Cytokeratin 8/18 (right) in the breast cancer cells when survive in the circulation, extravasate, and colonize metastatic MCF7 cells were cocultured with CXCL14 fibroblasts, but not in sites. These abilities have previously been linked to EMT (37–39). the presence of control fibroblasts. Similar findings of altered Therefore, lung colonization of tail-vein–injected MCF7 cells, EMT markers in MCF7- and DCIS cells were detected after "primed" in vitro in a coculture format together with CXCL14 treatment with conditioned media (CM) from CXCL14 fibro- fibroblasts or control fibroblasts prior to injection, was studied blasts (Supplementary Fig. S6A and S6B). Another breast cancer (Fig. 2B). cell line, SKBR3, which has lost E-Cadherin, showed increased Abundance of breast cancer cells in the lungs, 4 weeks levels of the EMT transcription factor Snail when treated after injection, was determined by qRT-PCR analyses with with CM from NIH-CXCL14 compared with CM from NIH-ctr human-specific primers as described previously (29). As shown (Supplementary Fig. S6B). in Fig. 2C, a significantly higher number of MCF7 cells were Furthermore, treatment of MCF7, DCIS, and SKBR3 breast detected in the lungs of mice that had been injected with cancer cancer cells with CM from CXCL14 fibroblasts induced changes cells "primed" with CXCL14 fibroblasts, as compared with mice in cell morphology. The tumor cells formed filopodium-like injected with control fibroblast-primed cancer cells. protrusions and obtained a mesenchymal-like morphology when These findings were independently validated by IHC analyses cultured in CM from CXCL14 fibroblasts, but not in CM from of tissue sections from lungs of mice subjected to tail-vein injec- control fibroblasts or in standard DMEM (Fig. 5C; Supplementary tion of coculture primed cancer cells. As shown in Fig. 2D, these Fig. S6C). These phenotypes induced by CXCL14 fibroblasts analyses demonstrated a significantly higher number of breast were not seen in the mesenchymal metastatic breast cancer cell cancer cells in the lungs of mice that had been injected with line MDA-MB-231, a cell line that already has undergone EMT NIH-CXCL14–primed breast cancer cells. (Supplementary Fig. S6C). These experiments thus demonstrate that CXCL14 fibroblasts, These results demonstrate the ability of CXCL14 fibroblasts as compared with control fibroblasts, more potently stimulate to induce an EMT phenotype in certain breast cancer cells, in a lung colonization of blood-circulating breast cancer cells. paracrine manner independent of cell-to-cell contact. CXCL14-induced molecular signaling and cellular responses CXCL14 fibroblasts enhance migration and invasion of breast are mediated by the atypical G-protein–coupled receptor cancer cells ACKR2 The induction of EMT suggested functional effects of CXCL14 Next, we aimed at identifying key signaling components medi- fibroblasts on breast cancer cells. Thus, we compared the ability of ating the cellular and protumorigenic effects of the orphan che- NIH-ctr and NIH-CXCL14 fibroblasts to stimulate the migration mokine CXCL14. and invasion of MCF7, DCIS, and SKBR3 cells. Transwell migra- We have previously found that CXCL14 enhances MAPK sig- tion assays were performed to analyze whether the changes in naling in certain cancer cells and defined them as CXCL14- EMT markers and in morphology were accompanied by an responsive cell lines (22). Initial experiments, analyzing ERK increase in cell motility. CXCL14 fibroblasts displayed a stronger phosphorylation subsequent to stimulation with recombinant ability to stimulate the migration of MCF7, DCIS, and SKBR3 CXCL14, identified a panel of CXCL14-responsive cell lines cells, as compared with control fibroblasts (Fig. 2A). We also (MCF7, DCIS, SKBR3, NIH-3T3) and nonresponsive cell lines investigated whether CXCL14 fibroblast–induced EMT and (MDA-MB-231 and LNCaP) (Supplementary Fig. S8; ref. 22). migration could be observed in breast cancer cell lines represent- Treatment of the CXCL14-responsive cell lines MCF7 and ing the basal (triple negative) molecular subgroup of breast NIH-3T3 with pertussis toxin, which specifically inhibits the Gai cancer, including 4T1 cells and Hs578t cells (Supplementary subfamily of GPCRs, blocked CXCL14-induced ERK signaling in Fig. S7A and S7B). NIH-CXCL14 cells significantly enhanced both cell types (Supplementary Fig. S9). This finding suggests that the migration (Supplementary Fig. S7A) and stimulated EMT CXCL14, as other chemokines, signals through the Gai subfamily (Supplementary Fig. S7B) of 4T1 cells, as compared with NIH-ctr of GPCRs. cells. There was a trend toward enhanced migration of Hs578t Next, a sequence alignment approach was initiated to iden- cells, although not significant, possibly explained by the fact tify GPCRs that exhibit similarities with known chemokine that these cells already have undergone EMT (Supplementary receptors (see "Materials and Methods" for details; Supplemen- Fig. S7A). tary Fig. S1). CXCL14 is a highly evolutionary conserved

www.aacrjournals.org Clin Cancer Res; 2019 OF9

Sjoberg€ et al.

Figure 5. CXCL14 fibroblasts induce loss of epithelial marker in breast cancer cells. A, Protein levels of E-cadherin (E-cad) in MCF7 and DCIS cells upon coculture with NIH- ctr or NIH-CXCL14 fibroblasts were detected by Western blot analysis. Representative blots are shown in the top, and quantifications of three independent experiments are shown in the bottom. B, Immunofluorescence of MCF7 cells (green) cocultured with NIH-ctr or NIH-CXCL14 fibroblasts (red) for the markers depicted in the figure. Arrowheads mark sites of loss of E-cadherin (E-cad) or Cytokeratin 8/18. Scale bar, 50 mm. C, Light microscopy pictures of MCF7 cells exposed for 48 hours to conditioned medium (CM) collected from NIH-ctr or NIH-CXCL14 cells (10 magnification). The number of cells with protrusions was counted in five vision fields in three independent experiments. Results are shown as fold of untreated MCF7 cells. unstim, unstimulated. P values were derived from unpaired two-sided Student t tests. , P < 0.001; , P < 0.01; and , P < 0.05. Error bars, SD.

OF10 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

chemokine and the absence of chemokine receptor orthologs ACKR2 and CXCL14 expression correlates with an EMT gene in species expressing CXCL14 limited the number of potential expression signature and poor prognosis in clinical datasets of candidates (40). Furthermore, CXCL14 is a highly selective breast cancer chemokine for trophoblasts of the placenta, immature dendri- To explore the clinical relevance of the experimental findings tic cells, B cells, and NK cells, and does not induce chemotaxis of pro-EMT effects of CXCL14 fibroblasts, correlative analyses of T lymphocytes. Therefore, mediators of CXCL14 signaling were performed in publicly available gene expression datasets are likely present on trophoblasts, and on certain immune cells, of breast cancer to investigate potential associations between but not expressed on T cells (18–21, 41). These considerations CXCL14/ACKR2 expression and clinical features (35). led to the selection for continued studies of 11 candidate The analyses revealed significant positive correlations between proteins from the original list. Expression levels of these recep- expression of CXCL14 and the EMT gene expression signature in tors were tested in CXCL14-responsive and nonresponsive a meta-analysis of nine breast cancer cohorts (Supplementary cell lines (Supplementary Table S3). These results reduced Fig. S2A). Analysis of the CXCL14:EMT correlation in intrinsic the candidate set to six candidates, which were analyzed in subtypes of breast cancer across all cohorts revealed no major preliminary siRNA experiments that led to continued studies difference among the molecular subgroups of breast cancer, but a on ACKR2, CXCR4, GPR25, and GPR182. slightly stronger correlation in the Basal subgroup (Supplemen- Initial experiments, using CXCL14-induced ERK phosphoryla- tary Fig. S2B). Correlations between CXCL14 and EMT were tion as an endpoint, were performed in the CXCL14-responsive not affected by the amount of tumor stroma (Supplementary MCF7 cells. As shown in Supplementary Fig. S10A and S10B, Fig. S13). This indicates that the CXCL14:EMT correlation truly downregulation of ACKR2 significantly reduced CXCL14-induced is driven by cancer cell EMT rather than reflecting stroma ERK phosphorylation. In contrast, downregulation of CXCR4, abundance. GPR25, and GPR182 did not affect CXCL14-induced ERK phos- To extend these studies, additional analyses were performed phorylation (Supplementary Fig. S10A and S10C). ACKR2-down- that focused on relationships between CXCL14 expression and regulated cells maintained ERK responses after stimulation with individual EMT-related genes. As shown in Supplementary Fig. S3, CXCL12, indicating specificity of the effects of ACKR2 down- CXCL14-high breast cancer displayed, in general, an EMT profile regulation on CXCL14 signaling (Supplementary Fig. S10D). characterized by, for example, reduced expression of E-cadherin These initial findings were extended in two other CXCL14- and increased expression of EMT transcription factors including responsive cell lines with the same endpoint. As shown in SNAI2, TWIST1, and ZEB1 and mesenchymal markers including Supplementary Fig. S11A–S11E, CXCL14-induced ERK phos- VIM, ACTA2, FN1, and collagens. On the basis of earlier studies phorylation in NIH-3T3 and SKBR3 cells was also attenuated implying a tumor-promoting function of CXCL14 in prostate and after siRNA-mediated downregulation of ACKR2. Moreover, the ovarian cancer (22, 25, 42), analyses were also performed on the enhanced growth of CXCL14 fibroblasts was significantly reduced ovarian and prostate cancer TCGA datasets. These demonstrated after downregulation of ACKR2 (Supplementary Fig. S11F). In results similar to those seen in the breast cancer analyses (Sup- contrast, no effect on cell growth was observed after downregula- plementary Fig. S3). tion of CXCR4, GPR182, and GPR25 (Supplementary Fig. S11F). On the basis of the experimental studies, these associations These findings prompted generation of derivatives of NIH-ctr were also analyzed in subsets defined by their combined CXCL14 and NIH-CXCL14 cells with stable ACKR2 downregulation with and ACKR2 status. In agreement with a functional link between two different ACKR2 shRNAs (Fig. 1A). In agreement with find- CXCL14 and ACKR2, the association with the EMT profile was ings above, CXCL14-induced ERK phosphorylation was signifi- most prominent in the CXCL14high/ACKR2high subgroup cantly attenuated in NIH-3T3 cells with stable downregulation of (Fig. 3A). This pattern was also seen in prostate and ovarian ACKR2 (Fig. 1B and C). As an additional endpoint, CXCL14- cancer datasets (Supplementary Fig. S4). induced upregulation of NOS1 was studied. As shown in Fig. 1D Survival data in the TCGA datasets was also used to explore and E, stable ACKR2 downregulation reduced NOS1 protein and survival associations of the four CXCL14/ACKR2-defined sub- mRNA levels in NIH-3T3 cells, but not in control cells. Finally, groups. Initial analyses with all four groups in the breast cancer CXCL14-induced proliferation of NIH-3T3 cells was analyzed dataset indicated a particularly poor prognosis of the group with with regard to ACKR2 dependency. Notably, downregulation of high expression of CXCL14 and ACKR2 (Supplementary Fig. S14). ACKR2 reduced the growth capacity of CXCL14 fibroblasts, Notably, a significant poor survival association was seen for whereas no effect of ACKR2 downregulation was detected in the combined CXCL14high/ACKR2high group, when contrasted NIH-ctr cells (Fig. 1F). with the rest of the TCGA population (P ¼ 0.01; Fig. 3B). A Cox The above data prompted binding studies to analyze whether proportional hazard model revealed an increased risk CXCL14 directly interacts with ACKR2. However, contrarily to of death for patients in the CXCL14high/ACKR2high subgroup CCL5, a high-affinity ligand of ACKR2, binding of CXCL14 was (HR ¼ 2.494; 95% CI ¼ 1.218–5.104). This poor prognosis only weakly detectable at high concentrations on cells overex- association of the CXCL14high/ACKR2high subgroup in breast pressing ACKR2 and was not different compared with cells cancer also remained significant in multivariate analyses with that lack ACKR2 (Supplementary Fig. S12A). Furthermore, in a clinicopathologic characteristics, including breast cancer molec- b-arrestin1 recruitment assay, a dose-dependent recruitment of ular subsets (Supplementary Table S4). Survival correlations of b-arrestin1 toward ACKR2 could only be detected upon CCL5, but the CXCL14high/ACKR2high subgroup were also explored in not upon CXCL14 stimulation (Supplementary Fig. S12B). publicly available datasets, representing 12 other tumor types In summary, these results derived from analyses of different cell (Supplementary Table S5). Besides breast cancer, the CXCL14high/ types and using multiple endpoints identify ACKR2 as a critical ACKR2high subgroup was significantly correlated to worse survival mediator of CXCL14-induced signaling, although no direct inter- of low-grade glioma, prostate cancer, clear cell renal cancer, and action between CXCL14 and ACKR2 could be found. stomach cancer (Supplementary Table S5).

www.aacrjournals.org Clin Cancer Res; 2019 OF11

Sjoberg€ et al.

Together, these correlative analyses support the notion that the key player in promoting tumor cell invasion and development of previously observed poor prognosis association of stroma- metastasis in SCID mice (13). In another study, paracrine cross- derived CXCL14 in breast cancer is related to a molecular pathway talk between tumor cells, myeloid cells, and endothelial cells, that also involves ACKR2. involving CXCL1 and CXCL2 signaling, was shown to drive metastasis and chemoresistance in the MMTV-PyMT mouse mod- Paracrine effects of CXCL14 fibroblasts depend on autocrine el of breast cancer (14). In addition, in experimental breast CXCL14 signaling tumors, CXCL12 secreted from CAFs was shown to select for Data presented above do not clearly resolve whether the poor clones of cancer cells with a high Src activity, and the ability to prognosis–associated CXCL14/ACKR2 pathway reflects autocrine specifically form metastasis in bone with a CXCL12-rich CXCL14/ACKR2 signaling, supporting EMT and metastasis in a microenvironment (15). paracrine manner, or rather reflects paracrine actions of CXCL14- Earlier studies have also linked chemokines and their receptors activated fibroblasts that involves ACKR2 in breast cancer cells. specifically to EMT. A constitutively active form of CXCR4, the As shown in Fig. 6A and B, CXCL14 fibroblast–induced cancer receptor for CXCL12, has been shown to be involved in modu- cell migration and E-Cadherin downregulation was significantly lation of breast tumor cell EMT markers and to enhance formation inhibited by knockdown of ACKR2 in fibroblasts. Of note, ACKR2 of lymph node metastasis in mice (43). CCR7, the receptor for downregulation in control fibroblasts did not affect migration or CCL21, and CXCR5 and its ligand CXCL13 have been shown to E-Cadherin levels of cocultured MCF7 cells. significantly correlate with EMT markers and enhanced lymph Further evidence supporting autocrine CXCL14 signaling as the node metastasis of human breast tumors (44, 45). Moreover, a driver of the prometastatic effects was provided by analyses of the GM-CSF-CCL18–positive feedback loop have been implied in effects of downregulation of NOS1; an earlier identified down- EMT and breast tumor metastasis in mice and associated with stream component of CXCL14 fibroblast signaling (25). As shown worse outcome in patients with breast cancer (16). in Fig. 6C, downregulation of NOS1 significantly reduced the The comparison of proteins secreted by NIH-CXCL14 and ability of CXCL14 fibroblasts to stimulate migration of MCF7 NIH-ctr cells provided new insight in CXCL14 signaling in cells. Notably, NOS1 downregulation did not affect cancer cell fibroblasts and revealed a set of candidates that mediate EMT migration induced by control fibroblasts (Fig. 6C). The reduction stimulated by CXCL14-expressing fibroblasts either individually of NOS1 signaling also attenuated the CXCL14 fibroblast– and/ or in combination. For example, CXCL1, CX3CL1, HGF, and induced downregulation of E-cadherin and upregulation of Snail TIMP-1 and tissue factor have previously been demonstrated to in CM-treated MCF7 cells (Fig. 6D and E). affect EMT and metastasis of breast cancer cells (39, 46–48). The reduced paracrine effects of CXCL14 fibroblasts following Recently, Wang and colleagues identified CCL17 derived from downregulation of ACKR2 or NOS1 suggest that CXCL14 itself is CXCL14-activated fibroblasts as another mediator of CXCL14- not promoting EMT, but rather stimulates the expression of EMT stimulated breast cancer EMT and metastasis (49). Furthermore, regulators in fibroblasts in an ACKR2-/ NOS1-dependent manner. CXCL14 was shown to act as a chemoattractant for M2 macro- To identify such putative EMT-regulating soluble factors derived phages that are known to promote EMT (16, 50, 51). Together, from CXCL14 fibroblasts, we used a protein profiler and com- these findings suggest that CXCL14 operates different axis of pared the secretome of NIH-CXCL14 and NIH-ctr fibroblasts. stromal signaling shifting the phenotype of stromal cells toward Among the factors that are more abundantly expressed by tumor progression and metastasis. CXCL14 fibroblasts (Supplementary Fig. S15) are proangiogenic A key finding of this study is the demonstration that CXCL14- factors (e.g., FGF-2, Angiogenin, VEGF-A), supporting the previ- induced molecular signaling and cellular responses depend on ous notion that NIH-CXCL14 cells stimulate angiogenesis (22), ACKR2, classified as an atypical chemokine receptor. ACKR2- molecules involved in matrix remodeling such as Adamts1, dependent CXCL14 signaling was shown in cell types of different MMP8, TIMP-1, as well as inducers and effectors of EMT including origin (breast cancer cells and fibroblasts) and the downregula- CXCL1, CX3CL1, TIMP-1, FGF2, HGF, and tissue factor. tion of ACKR2 almost completely abolished CXCL14-induced These data, together with the findings of Fig. 4, which effects. These data suggest that ACKR2 is a required component of show reduced EMT in the tumors formed after coinjection with CXCL14 signaling. Findings of ACKR2 expression on some breast NOS1-downregulated CXCL14 fibroblasts, indicate that the pro- cancer cells (Supplementary Table S3) also suggest the possibility migratory and EMT-modulatory effects of CXCL14 fibroblasts that CXCL14, in certain settings, might be prometastatic or depend on autocrine CXCL14/ACKR2/NOS1 signaling. EMT-stimulatory through direct effects on malignant cells. This topic should be further explored in future studies. Continued mechanistic analyses are also warranted regarding the roles of Discussion CXCL14/ACKR2 on other steps of the metastatic process than This study extends earlier findings that have identified stroma- those covered by the analyses of the invasive border (Fig. 4) and derived CXCL14 as a poor prognosis factor in breast cancer. the "tail-vein" experiment (Fig. 2). Mechanistic and correlative studies together suggest a novel Studies have suggested the chemokine receptors CXCR4 and prometastatic pathway composed of autocrine CXCL14/ GPR85 to directly bind CXCL14 and modulating signaling ACKR2/NOS1 signaling in fibroblasts that generates a fibroblast (49, 52). However, Otte and colleagues demonstrated that phenotype that supports cancer cell migration, invasion, EMT, CXCL14 does not bind and impact on CXCR4 signaling (53). In and metastasis (Supplementary Fig. S16). line with these data, we found that siRNA-mediated downregula- The finding of prometastatic effects of CXCL14 adds to earlier tion of CXCR4 did not affect CXCL14-induced MAPK signaling literature indicating the involvement of chemokines in metastasis in MCF7 breast cancer cells (Supplementary Fig. S10C) suggesting development. CCL5 secreted from bone marrow–derived mesen- that the functions of CXCL14 might be more complex than for chymal cells, recruited to the breast tumor stroma, was identified a other chemokines and might require formation of chemokine or

OF12 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

A B

1.5 tr) 4 *** ** * ** ** 3 1

2 NIH-ctr f NIH-ctr/shc E-Cad/Actin 0.5 NIH-CXCL14 o 1 (fold (fold of untreated MCF7) 0 Migration of MCF7 cells tr - r r :A :A :A :B :B C 2 Ct Ct 2 2 2 2 sh R h h R R R R K /s /s r 4 K K K K C t 1 C C C C A L A A A A h H-c h h h h s I C s /s /s N X tr/ tr 4 C c 14/s c 1 - - L IH IH- CL H C N N X NI X C -C IH- IH N N

C D

2 * ** hctr) s

CF7 cells 1.5

of NIH-ctr/ 0.5 (fold

Migration of M 0 tr tr C :A :A :B :B h 1 1 1 /shC S S tr 4/s OS O O -c 1 N N N H L hNOS1 h h I C /s /sh /s /s N X r 4 tr 4 t 1 c -C -c L - L1 IH IH C IH C N N X N X C -C H- IH NI N

E ) ) tr tr c - H H-c I 1.5 ** ** ** 2.5 *** ** ** 2 in ted NI ted N a a

e 1 1.5 tre tr l/Act r NIH-ctr 1 NIH-ctr Ct Ctr

0.5 h h Snai E-Cad/Actin

fs NIH-CXCL14 0.5 NIH-CXCL14 o d

l 0 0 r r fo t A B t A B (fold of s ( C : : C : : S 1 h S 1 sh O S s O S N O N O h N N s h sh h s s

Figure 6. Paracrine effects of fibroblast-derived CXCL14 depend on NOS1 and ACKR2. A, Migration of MCF7 cells for 24 hours in response to NIH-ctr or NIH-CXCL14 derivatives without (shCtr) or with stable knockdown of ACKR2 (shACKR2:A and shACKR2:B). B, Western blot analysis and quantification of E-cadherin levels in MCF7 cells subsequent to coculture with control (shCtr) or ACKR2-targeting (shACKR2:A) NIH-ctr or NIH-CXCL14 fibroblasts for 48 hours. C, MCF7 cells were allowed to migrate for 24 hours toward NIH-ctr or NIH-CXCL14 derivatives without (shCtr) or with stable knockdown of NOS1 (shNOS1:A and shNOS1:B). D, Western blot analysis of E-cadherin and Snail levels in MCF7 cells following coculture with NIH-ctr or NIH-CXCL14 fibroblasts with or without stable suppression of NOS1 expression. E, Quantification of Western blots as shown in B for E-cadherin (E-cad; left) and Snail (right) expression from three independent experiments. Representative blots are shown, and quantifications are based on three independent experiments. P values were derived from unpaired two-sided Student t tests. , P < 0.001; , P < 0.01 and , P < 0.05. Error bars, SD or SEM.

www.aacrjournals.org Clin Cancer Res; 2019 OF13

Sjoberg€ et al.

receptor heterocomplexes, which may explain the difficulty to studies of the current report suggest that autocrine CXCL14/ precisely define its signaling components (54). ACKR2 signaling in the stroma contributes to the survival and These atypical chemokine receptors have earlier been defined as EMT associations. Compartment-specific analyses of ACKR2 are scavenging receptors that bind chemokines with high affinity but prompted by the findings of the current study. are unable to induce signaling and cell migration (11). Absence of In summary, this study thus identifies a novel potentially the well-conserved DRYLAIVHA motif (DKYLEIVHA in ACKR2) druggable CXCL14/ACKR2 pathway involved in breast cancer has been assumed to explain the inability of atypical chemokine EMT and metastasis. Important tasks for future studies include receptors to induce downstream receptor signaling subsequent to development of inhibitory agents for initial testing in experimen- ligand binding. tal breast cancer models and further exploration of relevance of In this study, we observe an impact of ACKR2 on the molecular this pathway in other tumor types. Furthermore, these results also signaling, including MAPK activation (Fig. 1A–C; Supplementary encourage continued studies exploring biological activity of Figs. S10B and S11A–S11E) and cellular functions, including ACKRs and to decipher the exact interplay between CXCL14 with enhanced fibroblast proliferation (Fig. 1F; Supplementary classical and atypical chemokine receptors. Fig. S11F) in response to CXCL14. Independent recent evidence indeed does support signaling functions for some of the atypical Disclosure of Potential Conflicts of Interest chemokine receptors. These signaling properties were proposed to No potential conflicts of interest were disclosed. be possibly cell type dependent as in one study, experiments demonstrated coupling of ACKR3 to Gai proteins and induction Authors' Contributions of CXCL12-dependent conformational changes, but no activation Conception and design: E. Sjoberg,€ M. Augsten, A. Ostman€ of calcium signaling (55). Another study confirmed the binding Development of methodology: E. Sjoberg,€ L. Milde, D. H€agerstrand, of ACKR3 to PTX-sensitive Ga proteins and revealed activation of A. Chevigne i Acquisition of data (provided animals, acquired and managed patients, calcium mobilization, ERK signaling, and AKT signaling and provided facilities, etc.): E. Sjoberg,€ M. Meyrath, L. Milde, M. Herrera, enhanced migration and proliferation, subsequent to CXCL12 J. Lovrot,€ D. H€agerstrand, M. Bartish, C. Rolny, A. Chevigne, M. Augsten binding in rodent astrocytes and human glioma cell lines (56). Analysis and interpretation of data (e.g., statistical analysis, biostatistics, Earlier overexpression studies have also demonstrated ACKR2- computational analysis): E. Sjoberg,€ M. Meyrath, L. Milde, M. Herrera, J. Lovrot,€ € induced calcium mobilization by murine ACKR2 (57). A recent O. Frings, M. Bartish, E. Sonnhammer, A. Chevigne, M. Augsten, A. Ostman € study also proposed G-protein–independent, b-arrestin–depen- Writing, review, and/or revision of the manuscript: E. Sjoberg, M. Meyrath, € € dent, activation of the cofilin pathway [Rac1-p21–activated kinase J. Lovrot, D. Hagerstrand, C. Rolny, E. Sonnhammer, A. Chevigne, M. Augsten, A. Ostman€ 1 (PAK1)-LIM kinase 1 (LIMK1) cascade] following ACKR2 Administrative, technical, or material support (i.e., reporting or organizing stimulation suggesting that ACKR2 is not a totally silent data, constructing databases): A. Ostman€ fi receptor (58). Taken together, these ndings challenge the Study supervision: A. Ostman€ definition of ACKRs as exclusive nonsignaling, chemokine scav- enger receptors. fi Acknowledgments Earlier studies have identi ed stromal, but not epithelial, € CXCL14 as a bad prognosis marker in breast cancer (1). The Members of A. Ostman's group are acknowledged for support throughout the studies. Studies were supported by grants from the Swedish Cancer Society, correlative data of this study support the notion of functional BRECT, the Linne STARGET grant from Swedish Research Council and the clinical relevant interaction between CXCL14 and ACKR2. As KI/AZ-collaborative initiative, the Luxembourg Institute of Health (LIH) MESR shown in Fig. 3, the association between CXCL14 and an "EMT (grants 20160116 and 20170113), and the Luxembourg National Research profile" is enhanced when ACKR2 status is integrated in patient Fund PhD fellows (grants AFR-3004509 and INTER/FWO "Nanokine" - grant classification (Fig. 3A). Similarly, the survival association of 15/10358798). Technical support was provided by the histo-pathology unit of fi CXCL14 in the breast cancer TCGA gene expression dataset is Cancer Centrum Karolinska. Animal experiments bene ted from the expertise of the MTC animal facility. only detected in analyses that also consider ACKR2 status (Fig. 3B; Supplementary Fig. S14). Importantly, the combined CXCL14/ fi The costs of publication of this article were defrayed in part by the payment of ACKR2 metric is also a signi cant marker in multivariate analyses page charges. This article must therefore be hereby marked advertisement in including molecular breast cancer subtypes (Supplementary accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Table S4). It is recognized that the TCGA-based analyses fail to assign the prognostically relevant ACKR2 expression to the Received April 26, 2018; revised December 30, 2018; accepted March 4, 2019; stromal or epithelial compartment. However, the mechanistic published first March 8, 2019.

References 1. Sjoberg E, Augsten M, Bergh J, Jirstrom K, Ostman A. Expression of the 5. Del Pozo Martin Y, Park D, Ramachandran A, Ombrato L, Calvo F, chemokine CXCL14 in the tumour stroma is an independent marker of Chakravarty P, et al. Mesenchymal cancer cell-stroma crosstalk promotes survival in breast cancer. Br J Cancer 2016;114:1117–24. niche activation, epithelial reversion, and metastatic colonization. Cell Rep 2. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transition. 2015;13:2456–69. J Clin Invest 2009;119:1420–8. 6. Beerling E, Seinstra D, de Wit E, Kester L, van der Velden D, Maynard C, et al. 3. De Craene B, Berx G. Regulatory networks defining EMT during cancer Plasticity between epithelial and mesenchymal states unlinks EMT from initiation and progression. Nat Rev Cancer 2013;13:97–110. metastasis-enhancing stem cell capacity. Cell Rep 2016;14:2281–8. 4.ZhengX,CarstensJL,KimJ,ScheibleM,KayeJ,SugimotoH,etal. 7. Ostman A, Augsten M. Cancer-associated fibroblasts and tumor growth– Epithelial-to-mesenchymal transition is dispensable for metastasis but bystanders turning into key players. Curr Opin Genet Dev 2009;19:67–73. induces chemoresistance in pancreatic cancer. Nature 2015;527: 8. Kalluri R. The biology and function of fibroblasts in cancer. Nat Rev Cancer 525–30. 2016;16:582–98.

OF14 Clin Cancer Res; 2019 Clinical Cancer Research

Fibroblast CXCL14/ACKR2/NOS1 Signaling in Breast Cancer

9. Mishra P, Banerjee D, Ben-Baruch A. Chemokines at the crossroads of 31. Pawitan Y, Bjohle J, Amler L, Borg AL, Egyhazi S, Hall P, et al. Gene tumor-fibroblast interactions that promote malignancy. J Leukoc Biol expression profiling spares early breast cancer patients from adjuvant 2011;89:31–9. therapy: derived and validated in two population-based cohorts. 10. Thelen M. Dancing to the tune of chemokines. Nat Immunol 2001;2: Breast Cancer Res 2005;7:R953–64. 129–34. 32. Wang Y, Klijn JG, Zhang Y, Sieuwerts AM, Look MP, Yang F, et al. Gene- 11. Bonecchi R, Savino B, Borroni EM, Mantovani A, Locati M. Chemokine expression profiles to predict distant metastasis of lymph-node-negative decoy receptors: structure-function and biological properties. Curr Top primary breast cancer. Lancet 2005;365:671–9. Microbiol Immunol 2010;341:15–36. 33. Curtis C, Shah SP, Chin SF, Turashvili G, Rueda OM, Dunning MJ, et al. The 12. Balkwill F. Cancer and the chemokine network. Nat Rev Cancer 2004;4: genomic and transcriptomic architecture of 2,000 breast tumours reveals 540–50. novel subgroups. Nature 2012;486:346–52. 13. Karnoub AE, Dash AB, Vo AP, Sullivan A, Brooks MW, Bell GW, et al. 34. Cancer Genome Atlas Network. Comprehensive molecular portraits of Mesenchymal stem cells within tumour stroma promote breast cancer human breast tumours. Nature 2012;490:61–70. metastasis. Nature 2007;449:557–63. 35. Taube JH, Herschkowitz JI, Komurov K, Zhou AY, Gupta S, Yang J, et al. 14. Acharyya S, Oskarsson T, Vanharanta S, Malladi S, Kim J, Morris PG, et al. A Core epithelial-to-mesenchymal transition interactome gene-expression CXCL1 paracrine network links cancer chemoresistance and metastasis. signature is associated with claudin-low and metaplastic breast cancer Cell 2012;150:165–78. subtypes. Proc Natl Acad Sci U S A 2010;107:15449–54. 15. Zhang XH, Jin X, Malladi S, Zou Y, Wen YH, Brogi E, et al. Selection of bone 36. Parker JS, Mullins M, Cheang MC, Leung S, Voduc D, Vickery T, et al. metastasis seeds by mesenchymal signals in the primary tumor stroma. Cell Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin 2013;154:1060–73. Oncol 2009;27:1160–7. 16. Su S, Liu Q, Chen J, Chen J, Chen F, He C, et al. A positive feedback loop 37. Guarino M, Rubino B, Ballabio G. The role of epithelial-mesenchymal between mesenchymal-like cancer cells and macrophages is essential to transition in cancer pathology. Pathology 2007;39:305–18. breast cancer metastasis. Cancer Cell 2014;25:605–20. 38. Tsai JH, Yang J. Epithelial-mesenchymal plasticity in carcinoma metastasis. 17. Zlotnik A, Burkhardt AM, Homey B. Homeostatic chemokine receptors and Genes Dev 2013;27:2192–206. organ-specific metastasis. Nat Rev Immunol 2011;11:597–606. 39. Bourcy M, Suarez-Carmona M, Lambert J, Francart ME, Schroeder H, 18. Sleeman MA, Fraser JK, Murison JG, Kelly SL, Prestidge RL, Palmer DJ, et al. Delierneux C, et al. Tissue factor induced by epithelial-mesenchymal B cell- and monocyte-activating chemokine (BMAC), a novel non-ELR transition triggers a procoagulant state that drives metastasis of circulating alpha-chemokine. Int Immunol 2000;12:677–89. tumor cells. Cancer Res 2016;76:4270–82. 19. Kurth I, Willimann K, Schaerli P, Hunziker T, Clark-Lewis I, Moser B. 40. DeVries ME, Kelvin AA, Xu L, Ran L, Robinson J, Kelvin DJ. Defining the Monocyte selectivity and tissue localization suggests a role for breast and origins and evolution of the chemokine/chemokine receptor system. kidney-expressed chemokine (BRAK) in macrophage development. J Exp J Immunol 2006;176:401–15. Med 2001;194:855–61. 41. Kuang H, Chen Q, Fan X, Zhang Y, Zhang L, Peng H, et al. CXCL14 inhibits 20. Shellenberger TD, Wang M, Gujrati M, Jayakumar A, Strieter RM, Burdick MD, trophoblast outgrowth via a paracrine/autocrine manner during early et al. BRAK/CXCL14 is a potent inhibitor of angiogenesis and a chemotactic pregnancy in mice. J Cell Physiol 2009;221:448–57. factor for immature dendritic cells. Cancer Res 2004;64:8262–70. 42. Zhao L, Ji G, Le X, Wang C, Xu L, Feng M, et al. Long noncoding 21. Starnes T, Rasila KK, Robertson MJ, Brahmi Z, Dahl R, Christopherson K, RNA LINC00092 acts in cancer-associated fibroblaststodrive et al. The chemokine CXCL14 (BRAK) stimulates activated NK cell migra- glycolysis and progression of ovarian cancer. Cancer Res 2017;77: tion: implications for the downregulation of CXCL14 in malignancy. 1369–82. Exp Hematol 2006;34:1101–5. 43. Sobolik T, Su YJ, Wells S, Ayers GD, Cook RS, Richmond A. CXCR4 22. Augsten M, Hagglof C, Olsson E, Stolz C, Tsagozis P, Levchenko T, et al. drives the metastatic phenotype in breast cancer through induction of CXCL14 is an autocrine growth factor for fibroblasts and acts as a multi- CXCR2 and activation of MEK and PI3K pathways. Mol Biol Cell 2014; modal stimulator of prostate tumor growth. Proc Natl Acad Sci U S A 2009; 25:566–82. 106:3414–9. 44. Li F, Zou Z, Suo N, Zhang Z, Wan F, Zhong G, et al. CCL21/CCR7 axis 23. Allinen M, Beroukhim R, Cai L, Brennan C, Lahti-Domenici J, Huang H, activating chemotaxis accompanied with epithelial-mesenchymal transi- et al. Molecular characterization of the tumor microenvironment in breast tion in human breast carcinoma. Med Oncol 2014;31:180. cancer. Cancer Cell 2004;6:17–32. 45. Biswas S, Sengupta S, Roy Chowdhury S, Jana S, Mandal G, Mandal PK, 24. Gu XL, Ou ZL, Lin FJ, Yang XL, Luo JM, Shen ZZ, et al. Expression of CXCL14 et al. CXCL13-CXCR5 co-expression regulates epithelial to mesenchymal and its anticancer role in breast cancer. Breast Cancer Res Treat 2012;135: transition of breast cancer cells during lymph node metastasis. 725–35. Breast Cancer Res Treat 2014;143:265–76. 25. Augsten M, Sjoberg E, Frings O, Vorrink SU, Frijhoff J, Olsson E, et al. 46. Wang N, Liu W, Zheng Y, Wang S, Yang B, Li M, et al. CXCL1 derived Cancer-associated fibroblasts expressing CXCL14 rely upon NOS1-derived from tumor-associated macrophages promotes breast cancer metasta- nitric oxide signaling for their tumor-supporting properties. Cancer Res sis via activating NF-kappaB/SOX4 signaling. Cell Death Dis 2018;9: 2014;74:2999–3010. 880. 26. Szpakowska M, Nevins AM, Meyrath M, Rhainds D, D'Huys T, Guite-Vinet 47. D'Angelo RC, Liu XW, Najy AJ, Jung YS, Won J, Chai KX, et al. TIMP-1 via F, et al. Different contributions of chemokine N-terminal features attest to a TWIST1 induces EMT phenotypes in human breast epithelial cells. different ligand binding mode and a bias towards activation of ACKR3/ Mol Cancer Res 2014;12:1324–33. CXCR7 compared with CXCR4 and CXCR3. Br J Pharmacol 2018;175: 48. Tardaguila M, Mira E, Garcia-Cabezas MA, Feijoo AM, Quintela-Fandino 1419–38. M, Azcoitia I, et al. CX3CL1 promotes breast cancer via transactivation of 27. Szpakowska M, Meyrath M, Reynders N, Counson M, Hanson J, Steyaert J, the EGF pathway. Cancer Res 2013;73:4461–73. et al. Mutational analysis of the extracellular disulphide bridges of the 49. Wang Y, Weng X, Wang L, Hao M, Li Y, Hou L, et al. HIC1 deletion promotes atypical chemokine receptor ACKR3/CXCR7 uncovers multiple binding breast cancer progression by activating tumor cell/fibroblast crosstalk. and activation modes for its chemokine and endogenous non-chemokine J Clin Invest 2018;128:5235–50. agonists. Biochem Pharmacol 2018;153:299–309. 50. Cereijo R, Gavalda-Navarro A, Cairo M, Quesada-Lopez T, Villarroya J, 28. Sonnhammer EL, Hollich V. Scoredist: a simple and robust protein Moron-Ros S, et al. CXCL14, a brown adipokine that mediates brown-fat- sequence distance estimator. BMC Bioinformatics 2005;6:108. to-macrophage communication in thermogenic adaptation. Cell Metab 29. Malek A, Catapano CV, Czubayko F, Aigner A. A sensitive polymerase chain 2018;28:750–63. reaction-based method for detection and quantification of metastasis in 51. Linde N, Casanova-Acebes M, Sosa MS, Mortha A, Rahman A, Farias E, et al. human xenograft mouse models. Clin Exp Metastasis 2010;27:261–71. Macrophages orchestrate breast cancer early dissemination and metastasis. 30. Miller LD, Smeds J, George J, Vega VB, Vergara L, Ploner A, et al. An Nat Commun 2018;9:21. expression signature for p53 status in human breast cancer predicts 52. Tanegashima K, Suzuki K, Nakayama Y, Tsuji K, Shigenaga A, Otaka A, et al. mutation status, transcriptional effects, and patient survival. Proc Natl CXCL14 is a natural inhibitor of the CXCL12-CXCR4 signaling axis. Acad Sci U S A 2005;102:13550–5. FEBS Lett 2013;587:1731–5.

www.aacrjournals.org Clin Cancer Res; 2019 OF15

Sjoberg€ et al.

53. Otte M, Kliewer A, Schutz D, Reimann C, Schulz S, Stumm R. CXCL14 is no proteins in primary rodent astrocytes and human glioma cells. Glia direct modulator of CXCR4. FEBS Lett 2014;588:4769–75. 2012;60:372–81. 54. Kleist AB, Getschman AE, Ziarek JJ, Nevins AM, Gauthier PA, Chevigne A, 57. Nibbs RJ, Wylie SM, Pragnell IB, Graham GJ. Cloning and characterization et al. New paradigms in chemokine receptor signal transduction: Moving of a novel murine beta chemokine receptor, D6. Comparison to three other beyond the two-site model. Biochem Pharmacol 2016;114:53–68. related macrophage inﬂammatory protein-1alpha receptors, CCR-1, CCR- 55. Levoye A, Balabanian K, Baleux F, Bachelerie F, Lagane B. CXCR7 hetero- 3, and CCR-5. J Biol Chem 1997;272:12495–504. dimerizes with CXCR4 and regulates CXCL12-mediated G protein signal- 58. Borroni EM, Cancellieri C, Vacchini A, Benureau Y, Lagane B, Bachelerie F, ing. Blood 2009;113:6085–93. et al. beta-arrestin-dependent activation of the coﬁlin pathway is required 56. Odemis V, Lipfert J, Kraft R, Hajek P, Abraham G, Hattermann K, et al. The for the scavenging activity of the atypical chemokine receptor D6. Sci Signal presumed atypical chemokine receptor CXCR7 signals through G(i/o) 2013;6:ra30.

OF16 Clin Cancer Res; 2019 Clinical Cancer Research

A Novel ACKR2-Dependent Role of Fibroblast-Derived CXCL14 in Epithelial-to-Mesenchymal Transition and Metastasis of Breast Cancer

Elin Sjöberg, Max Meyrath, Laura Milde, et al.

Clin Cancer Res Published OnlineFirst March 8, 2019.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-1294

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/04/19/1078-0432.CCR-18-1294.DC2

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2019/04/17/1078-0432.CCR-18-1294. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.