Molecular Cancer Therapeutics - Involvement of Prokineticin 2 ......Cancer Biology and Translational Studies Involvement of Prokineticin 2–expressing Neutrophil Inﬁltration in

Cancer Biology and Translational Studies

Involvement of Prokineticin 2–expressingNeutrophil Infiltration in 5-Fluorouracil–inducedAggravation of Breast Cancer Metastasis to LungSoichiro Sasaki1, Tomohisa Baba1, Hayato Muranaka2, Yamato Tanabe1,Chiaki Takahashi2, Seiichi Matsugo3, and Naofumi Mukaida1

Abstract

Adjuvant chemotherapy is used for human breast cancerpatients, even after curative surgery of primary tumor, to preventtumor recurrence primarily as a form of metastasis. However,anticancer drugs can acceleratemetastasis in severalmousemetas-tasis models. Hence, we examined the effects of postsurgicaladministration with 5-fluorouracil (5-FU), doxorubicin, andcyclophosphamide, on lung metastasis process, which developedafter the resection of the primary tumor arising from the ortho-topic injection of a mouse triple-negative breast cancer cell line,4T1.Only 5-FUmarkedly increased the numbers and sizes of lungmetastasis foci, with enhanced tumor cell proliferation andangiogenesis as evidenced by increases in Ki67-positive cell num-bers and CD31-positive areas, respectively. 5-FU–mediated aug-mented lung metastasis was associated with increases in intra-pulmonary neutrophil numbers and expression of neutrophilic

chemokines, Cxcl1 and Cxcl2 in tumor cells, with few effects onintrapulmonary T-cell or macrophage numbers. 5-FU enhancedCxcl1 and Cxcl2 expression in 4T1 cells in a NFkB-dependentmanner. Moreover, the administration of a neutrophil-depletingantibody or a Cxcr2 antagonist, SB225002, significantly attenu-ated 5-FU–mediated enhanced lung metastasis with depressedneutrophil infiltration. Furthermore, infiltrating neutrophils and4T1 cells abundantly expressed prokineticin-2 (Prok2) and itsreceptor, Prokr1, respectively. Finally, the administration of 5-FUafter the resection of the primary tumor failed to augment lungmetastasis in the mice receiving Prokr1-deleted 4T1 cells. Collec-tively, 5-FU can enhance lung metastasis by inducing tumor cellsto produce Cxcl1 and Cxcl2, which induced the migration ofneutrophils expressing Prok2 with a capacity to enhance 4T1 cellproliferation. Mol Cancer Ther; 17(7); 1515–25. �2018 AACR.

IntroductionBreast cancer is still the most commonly diagnosed cancer in

women, but can be frequently diagnosed at its early stage withimprovement in diagnostic measures such as mammogram andultrasonography (1, 2). As a consequence, breast cancer can oftenbe curatively resected with surgery and nearly 90% patients cansurvive for longer than 5 years after the surgery (3). However,even after curative surgery, breast cancer sometimes relapsesprimarily as a form of metastasis to distant organs includinglung, liver, and bone, probably originating frommicrometastasisfoci in these organs, which are undetected at surgery (4). Thus,to reduce the risk of recurrence as a form of metastasis, breastcancer patients generally receive various chemotherapeutic regi-

mens consistingmainly of cyclophosphamide, doxorubicin, or 5-fluorouracil (5-FU), as an adjuvant therapy after the surgery (5).

It is widely accepted that metastasis can proceed by the inter-action of tumor cells and local tissue microenvironments (6).Accumulating evidence indicates that chemotherapeutic drugscan have impacts on resident normal cells such as immune cellsbesides its direct antiproliferative effects on cancer cells (7),raising the possibility that the administration of chemotherapeu-tic drugs may modulate tumor metastasis. Consistently, severalindependent groups reported that chemotherapy exacerbatedrather than attenuated distant metastasis in several mouse breastcancer metastasis models and human patients (8–11). However,these models utilized either immunodeficient mice injected witha human breast cancer cell line, or mice intravenously injectedwith a cancer cell line, and as a consequence, cannot recapitulateaccurately the process of postsurgical adjuvant chemotherapy forbreast cancer in immunocompetent human patients. Hence, it isnecessary to examine the effects of chemotherapeutic drugs onspontaneous lung metastasis model in immunocompetent mice,the model which can reproduce more precisely the situationsobserved in human breast cancer patients undergoing postsurgi-cal adjuvant therapy.

Amouse triple-negative breast cancer (TNBC) cell line, 4T1, canmetastasize to lung after the complete resection of primary tumorarising from its orthotopic injection intomammary fat pad (MFP)at the time when cancer cells were hardly detected in lungs (12).This situation resembles lungmetastasis, which develops in breastcancer patients with latency after curative surgery of the primarytumors. Hence, we examined the effects of representative

1Division of Molecular Bioregulation, Cancer Research Institute, KanazawaUniversity, Kanazawa, Japan. 2Division of Oncology and Molecular Biology,Cancer Research Institute, Kanazawa University, Kanazawa, Japan. 3School ofNatural System, College of Science and Engineering, Kanazawa University,Kanazawa, Japan.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Corresponding Author: Naofumi Mukaida, Division of Molecular Bioregulation,Cancer Research Institute, Kanazawa University, Kakuma-machi, Kanazawa,Ishikawa 920-1192, Japan. Phone: 81-76-264-6735; Fax: 81-76-234-4520; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-17-0845

�2018 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org 1515

on October 27, 2020. © 2018 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst April 11, 2018; DOI: 10.1158/1535-7163.MCT-17-0845

http://crossmark.crossref.org/dialog/?doi=10.1158/1535-7163.MCT-17-0845&domain=pdf&date_stamp=2018-6-12

http://mct.aacrjournals.org/

chemotherapeutics used for postsurgery adjuvant therapy, cyclo-phosphamide, doxorubicin, and 5-FU, on this lung metastasisformation. We proved that 5-FU but neither cyclophosphamidenor doxorubicin accelerated lung metastasis development byinducing massive migration of neutrophils expressing a potentgrowth factor, prokineticin-2, in this model. Collectively, weunraveled hitherto unknown and paradoxical effects of 5-FU onbreast cancer metastasis to lungs.

Materials and MethodsCell lines

A mouse mammary carcinoma cell line, BALB/c-derived 4T1(CRL-2539; ref. 13), and human breast cancer cell line, BT-20(HTB-19; ref. 14), were purchased in 2007 from ATCC. TS/A,another mouse mammary carcinoma cell established fromBALB/c mammary adenocarcinoma (15), was obtained in2010 from Dr. Nanni (University of Bologna). All these celllines were cultured at 37�C under 5% CO2 in a RPMI1640medium supplemented with 10% FBS and used within 15passages of receiving from the source. Mycoplasma and authen-tication tests of all cell lines were not performed before use.

Lung metastasis modelsParental or prokineticin receptor (Prokr) 1–deleted 4T1 cells, or

TS/A cells were suspended at a cell density of 1.5� 106 cells/mL inHank balanced salt solution and 100-mL suspensions wereinjected into the secondary MFP. Tumor growth was monitoreddaily and the resultant primary tumor was surgically removed 12days after the tumor injection, when it reached a diameter of 8.0–10.0 mm. Two days after the surgery, the mice received intraper-itoneal injection of 5-FU (50 mg/kg) or cyclophosphamide(150 mg/kg), or intravenous administration of doxorubicin(5 mg/kg). Mice were sacrificed at the indicated time intervalsfor macroscopical inspection, histologic analysis, and total RNAextraction. Todeplete neutrophils,mice received intraperitoneally500 mg of a rat anti-Ly6GmAb (clone 1A8) in 500 mL or PBS onceevery 4 days after 5-FU injection. In some experiments, micereceived intraperitoneally 200 mg of a rat anti-Gr-1 mAb (cloneRB6-8C5) in 200-mL PBS or vehicle once every 3 days after 5-FUinjection. In another series of experiments, SB225002 (16) wasadministered at a dose of 0.5 mg/kg body weight once every dayafter 5-FU treatment.

Statistical analysisThemeansþ SDwere calculated for all parameters determined.

Statistical significance was evaluated using one-way ANOVA,followed by Tukey–Kramer post hoc test or Mann–Whitney U test.P values less than 0.05 were considered statistically significant.

Supplementary informationThe SupplementaryMethods provides detailed information on

mice, reagents and antibodies, generation of Prokr1-deleted 4T1cells, IHC analyses of lung tissues, flow cytometric analysis ofsingle-cell suspensions obtained from lung or bone marrow,in vitro cell proliferation assay and cell culture of 4T1, TS/A, orBT-20 cells, quantitative (q) RT-PCR analysis, chromatin immu-noprecipitation (ChIP) assay, determination of intracellular reac-tive oxygen species (ROS) generation, isolation of mouseCD11bþLy6Gþ myeloid cells and tumor cells from lung tissue,cell-cycle analysis, coculture of tumor cells and CD11bþLy6Gþ

myeloid cells using nonmigration transwell system, transwell

migration assay, preparation of H2O2 and H2O2 treatment,in vitro treatment of mouse CD11bþLy6Gþ neutrophils, senes-cence-associated b-galactosidase (SA-b-gal) staining, and clinicaldatabase analysis.

Results5-FU treatment enhances lung metastasis of mouse breastcancer cell lines

We administered chemotherapeutic drugs used for postsurgicaladjuvant therapy, doxorubicin, cyclophosphamide, and 5-FU, tothe mice after the surgical removal of the primary tumors arisingfrom the injection of a mouse breast cancer cell line, 4T1, intoMFP, to investigate their effects on lung metastasis development(Fig. 1A). We observed that a single injection of 5-FU had adverseeffects on survival, compared with that of cyclophosphamide ordoxorubicin (Fig. 1B). Moreover, at 28 days after tumor cellinjection, only a small number of tumor foci were detected inlungs of untreated mice. In contrast, 5-FU but neither cyclophos-phamide nor doxorubicin, progressively increased the numbers ofmetastatic tumor foci in lungs (Fig. 1C–E). Moreover, 5-FUtreatment increased Ki67-positive proliferating tumor cell num-bers (Fig. 1F) and CD31-positive vascular areas inside metastaticfoci (Fig. 1G) with few effects on ssDNA-positive apoptotic cellnumbers (Fig. 1H). Likewise, 5-FU increased lungmetastasis focusnumbers even when another mouse breast cancer cell line, TS/A,was used (Supplementary Fig. S1). Thus, 5-FU treatment canaccelerate lung metastasis when administered after the resectionof the primary breast cancer.

5-FU–mediated enhanced lung metastasis is associated withmassive neutrophil infiltration

We next evaluated the effects of 5-FU on the cells infiltratinginto lungs. IHC analysis identified Ly6Gþ myeloid cells butneither F4/80þ macrophages nor CD3þ T lymphocytes as apredominant type of cells present in lungs in mice treated with5-FU (Fig. 2A). Consistently, among the cell types that we exam-ined, a flow cytometric analysis revealed the increases onlyin CD11bþLy6Gþ myeloid cells in lungs from mice treated with5-FU, at 26 and 28 days after the tumor injection (Fig. 2B;Supplementary Fig. S2A). These CD11bþLy6Gþ myeloid cellsexhibited typical neutrophilic morphologic features on Giemsastaining (Supplementary Fig. S3A) and did not express granulo-cyte-myeloid–derived suppressor cell (G-MDSC)markers, CD115and CD244 (Supplementary Fig. S3B). These observations iden-tified CD11bþLy6Gþ cells as neutrophils but not G-MDSC. Incontrast, CD11bþLy6Gþ neutrophils were not increased in bonemarrow in 5-FU–treated tumor-bearing mice or in lung of 5-FU–

treatedmicewithout tumor injection (Fig. 2B; Supplementary Fig.S2B). Todefine the contribution of neutrophils to lungmetastasis,we next treated mice with anti-Gr-1 antibody. The antibodytreatment reduced the numbers of intrapulmonary Ly6Gþ neu-trophils (Supplementary Fig. S2C) and metastatic foci, particu-larly those with a diameter larger than 2.0 mm (Fig. 3A and B),together with reductions in Ki67-positive proliferating tumor cellnumbers and CD31-positive neovascular areas, but not ssDNA-positive apoptotic cell numbers in metastatic foci (Fig. 3C–E).Similarly, the treatment of a neutrophil-specific antibody, anti-Ly6G, reduced lungmetastasis formation (Fig. 3F andG).Massiveneutrophil infiltration further prompted us to examine the expres-sion of chemokines with a potent neutrophil chemotactic activity,Cxcl1 and Cxcl2, in lungs. Indeed, Cxcl1 and Cxcl2 mRNA

Sasaki et al.

Mol Cancer Ther; 17(7) July 2018 Molecular Cancer Therapeutics1516




Figure 1.

5-FU–induced increase in lung metastasis of a mouse breast cancer cell line, 4T1. A, Schematic representation of experimental procedure of orthotopic lungmetastasis model. Mice were injected with cyclophosphamide (CTX; light gray circles), doxorubicin (DOX; dark gray circles), or 5-FU (closed circles) along withA. B, Kaplan–Meier survival curves for mice that were injected with cyclophosphamide, doxorubicin, or 5-FU as chemotherapeutic drugs (n ¼ 10). � , P < 0.05;n.s., not significant. The lungs were removed 28 days after 4T1 cell injection to inspect the macroscopic appearance of the lung (C), and the tumor focusnumbers were determined (D). Each symbol represents the metastatic tumor focus numbers of an individual mouse (n ¼ 5). The bars represent the averageof each group. �� , P < 0.01; n.s., not significant. E, Animals were sacrificed at each time point until day 28 after 4T1 cell injection with or without 5-FU treatment,to determine tumor focus numbers of lung. The data represent the mean þ SD (n ¼ 3). �, P < 0.05; �� , P < 0.01. Twenty-eight days after 4T1 cell injection, lungwas removed from mice to determine the numbers of Ki67-positive cells (F), CD31-positive areas (G), and ssDNA-positive cells (H). All values represent themean þ SD (n ¼ 4). � , P < 0.05; ��, P < 0.01; n.s., not significant.

Promotion of Lung Metastasis by 5-FU Treatment

www.aacrjournals.org Mol Cancer Ther; 17(7) July 2018 1517




expression was enhanced marginally in lungs of mice after theresection of the primary tumor, but subsequent 5-FU adminis-tration augmented their mRNA expression in whole lung tissues(Fig. 4A) and tumor cell fraction purified from lung 28 days aftertumor injection (Supplementary Fig. S4A). IHC analysis furtherdetected Cxcl1 and Cxcl2 proteins in intrapulmonary tumor cellsin 5-FU–treated mice (Fig. 4B). The treatment with a Cxcr2antagonist, SB225002, reduced the numbers of intrapulmonaryLy6Gþ neutrophils (Supplementary Fig. S4B) and the numbersand sizes of metastatic foci (Fig. 4C and D), together with reduc-tions in Ki67-expressing proliferating tumor cell numbers andCD31-positive neovascular areas, but not ssDNA-positive apo-ptotic cell numbers in metastatic foci (Fig. 4E–G), similarly asanti-Gr-1 antibody did. These observations implicated the asso-ciation of 5-FU–mediated enhanced lung metastasis with 5-FU–

induced neutrophilic chemokine expression in tumor cells andsubsequent Ly6Gþ myeloid cell infiltration.

5-FU induces Cxcl1 and Cxcl2 expression in breast cancer cellsthrough NF-kB signaling via ROS elevation

We next examined the effects of 5-FU, cyclophosphamide, ordoxorubicin onCxcl1 andCxcl2 expression inmouse breast cancercell lines, 4T1 and TS/A. 5-FU enhanced mRNA expression ofCxcl1 andCxcl2 in 4T1 and TS/A cells in a dose-dependentmanner(Fig. 5A; Supplementary Fig. S5A), while cyclophosphamideor doxorubicin failed to do (Supplementary Fig. S5B and S5C).In line with the previous report to indicate the crucial role ofNF-kB activation in inducible Cxcl1 and Cxcl2 mRNA expression(17), IKK inhibitor, CID-2858522 (18), and NF-kB inhibitor,bortezomib, inhibited 5-FU–induced enhancement in Cxcl1 and

Figure 2.

Effects of 5-FU treatment on cells infiltrating into lungs.A, Lung tissue was obtained from mice 28 days after4T1 cell injection and was subjected to H&E staining orIHC analysis using anti-Ly6G, anti-F4/80, or anti-CD3antibodies. Representative results from five independentexperiments are shown here. Insets indicate enlargedarea of lungs. Original magnification, �100; scale bars,100 mm. B, CD11bþLy6Gþ myeloid cell numbersof lung or bonemarrowwere counted byflowcytometry.Animals were sacrificed at each time point until28 days after 4T1 cell injection, to determine theCD11bþLy6Gþ myeloid cell numbers. Data represent themean þ SD (n ¼ 4). � , P < 0.05; �� , P < 0.01.

Sasaki et al.





Cxcl2 mRNA expression (Fig. 5B; Supplementary Fig. S5D). Sim-ilarly, 5-FUenhanced themRNAexpressionofCXCL1 andCXCL2,and their related molecule, CXCL8, in a human breast cancercell line, BT-20, and their mRNA expression was reduced bybortezomib (Supplementary Fig. S6A and S6B). Moreover, ChIPassay revealed that 5-FU increased the binding of a component ofNF-kB, p65, to Cxcl1 and Cxcl2 promoters in 4T1 cells (Fig. 5C).The capacity of ROS to activate NF-kB incited us to examine theeffects of 5-FU on ROS generation in 4T1 cells. Consistently, 5-FUincreased intracellular ROS concentration in 4T1 cells (Fig. 5D).An antioxidant, N-acetyl cysteine (NAC), reduced 5-FU–induced

p65 binding to Cxcl1 and Cxcl2 promoters (Fig. 5C) and atten-uated 5-FU–induced Cxcl1 and Cxcl2 mRNA expression in 4T1cells (Fig. 5E) together with reduced ROS levels (Fig. 5D). Theseobservations would indicate that 5-FU increased intracellularROS levels, thereby activating NF-kB and subsequent Cxcl1 andCxcl2 expression.

5-FU–mediated metastasis depends on the Prok2-expressingneutrophils

Increases in intrapulmonary tumor cells can be induced bytheir enhanced invasion into lungs and/or their augmented

Figure 3.

Effects of neutrophil depletion on5-FU–induced increase in lungmetastasis. A–E, Mice were injectedintraperitoneally with anti-Gr-1antibody (200 mg/head) or PBS onceevery 3 days, starting 14 days after 4T1cell injection. The lungs were removed28 days after 4T1 cell injection todetermine the tumor numbers (A) andsizes (B). Each symbol represents thetumor numbers per mouse, and barsrepresent the average for each group(n¼ 6). � , P <0.05; �� , P <0.01; n.s., notsignificant. Ki67-positive cell numbers(C), CD31-positive areas (D), andssDNA-positive cell numbers (E) weredetermined with IHC analysis. Allvalues represent the mean þ SD (n ¼5). � , P < 0.05; n.s., not significant. Micewere injected intraperitoneally withanti-Ly6G antibody (500 mg/ head) orPBS once every 4 days. The lungswereremoved 28 days after 4T1 cellinjection to determine the tumornumbers (F) and sizes (G). Eachsymbol represents the tumor numbersper mouse, and bars represent theaverage for each group (n ¼ 10).� , P < 0.05; �� , P < 0.01.






proliferation, and infiltrating neutrophils may be able to providecues to induce tumor cells to invade or proliferate. However,neutrophil-derived supernatants failed to induce 4T1 cells toinvade (Supplementary Fig. S7A). Thus, we next investigated theexpression of various growth factors in lungs. Among the protu-morigenic growth factors that we examined (19), only Prok2expression was enhanced in lungs of mice who received 5-FUafter the resection of the primary tumor (Fig. 6A). Of interest isthat its enhanced expression was observed only at 26 and 28 daysafter the tumor injection (Supplementary Fig. S7B) when neu-trophil infiltration into lungs became evident (Fig. 1E). Prok2mRNAwas expressed selectively in CD11bþLy6Gþ neutrophils of

lung tissues derived from 5-FU–treated mice (Fig. 6B). Moreover,Prok2 protein was detected in Ly6G-positive cells obtained from5-FU–treated, but not untreated animals (Fig. 6C; SupplementaryFig. S7C). Furthermore, anti-Gr-1 antibody or SB225002 signif-icantly reduced 5-FU–mediated enhancement in Prok2 mRNAexpression in lungs (Fig. 6D). These observations prompted us toexplore the involvement of Prok2 in 5-FU–induced and neutro-phil-mediated enhancement in lung metastasis. 4T1 cellsexpressed Prokr1 but not Prokr2 among the receptors for Prok2(Fig. 6E). Prok2 enhanced the in vitro proliferation and cell-cycleprogressionof parental cells, butProkr1deletion abrogatedProk2-induced enhancement in in vitro cell proliferation (Fig. 6F;

Figure 4.

Involvement of the Cxcl1 and Cxcl2 in5-FU–mediated lung metastasisenhancement. A, Cxcl1 and Cxcl2mRNA expression in the lungs of 5-FUuntreated (open squares) or treatedmice (closed squares). Expressionlevels were normalized to Hprt mRNAlevels. Data represent the mean þ SD(n ¼ 4). � , P < 0.05; �� , P < 0.01.B, Lungs were removed from 5-FU–treated mice 28 days after the tumorinjection. The obtained tissues werefixed and immunostained with Cxcl1(top) or Cxcl2 (bottom) antibody.Representative results from threeindependent experiments are shownhere. Original magnification, �100;scale bar, 100 mm. C–G, Mice wereinjected intraperitoneally withSB225002 (0.5 mg/kg) or vehicle,once every day, starting 14 days after4T1 cell injection. The lungs wereremoved 28 days after 4T1 cellinjection to determine the tumornumbers (C) and sizes (D). Eachsymbol represents the tumor numbersper mouse, and bars represent theaverage for each group (n ¼ 10).Ki67-positive cell numbers (E), CD31-positive areas (F), and ssDNA-positivecell numbers (G) were determinedwith IHC analysis. All values representthe mean þ SD (n ¼ 5). � , P < 0.05;�� , P < 0.01; n.s., not significant.

Sasaki et al.





Supplementary Fig. S8A). Moreover, a transwell assay demon-strated that in vitro 4T1 cell proliferation was enhanced in thepresence of the CD11bþLy6Gþ neutrophils obtained from thelungs from 5-FU–treated tumor bearing mice, but not those from5-FU–untreated tumor bearing mice (Fig. 6G, left). The enhance-ment was abrogated when Prokr1-deleted 4T1 cells were usedinstead of the parental cells (Fig. 6G, right). Furthermore, theadministration of 5-FU after the resection of the primary tumor

failed to augment lung metastasis when mice were injected withProkr1-deleted 4T1 cells (Fig. 6H; Supplementary Fig. S8B) withfew effects on Ki67-positive cell numbers, CD31-positive areas,and ssDNA-positive cell numbers (Supplementary Fig. S8C–S8E).Thus, 5-FU induced the infiltration of Prok2-expressing neutro-phils through the action of Cxcl1 and Cxcl2, thereby promotingthe intrapulmonary growth of breast cancer cells expressingProkr1 (Supplementary Fig. S9).

Figure 5.

Molecular mechanisms underlying 5-FU-induced Cxcl1 and Cxcl2 expression. A, 4T1 cells were treated with the indicated concentrations of 5-FU for 24 hours. Thelevels of mRNA were quantified by qRT-PCR. All values represent the mean þ SD (n ¼ 3). � , P < 0.05; �� , P < 0.01. B, 4T1 cells were pretreated for 1 hourwith the indicated concentrations of CID-2858522 andwere further incubatedwith 5-FU (5 mg/mL) for 24 hours. The levels ofmRNAwere quantified by qRT-PCR. Allvalues represent the mean þ SD (n ¼ 3). �� , P < 0.01. C, NFkB p65 binding to the promoter region of the mouse Cxcl1 and Cxcl2 was assessed by ChIP assayin untreated or 5-FU–treated 4T1 cells with or without pretreatment with NAC. Input or eluted chromatin was subjected to qRT-PCR analysis using promoter-specificprimers. Data are represented as the percentage input of the immunoprecipitated chromatin for each gene from three separate chromatin preparations. Datarepresent the meanþSD (n ¼ 3). � , P < 0.05; n.s., not significant. D, Intracellular ROS levels were determined. 4T1 cells were incubated in the absence orthe presence of 5-FU (5 mg/mL) for 24 hours together with 1 hour pretreatment with NAC. Representative results from three independent experiments areshown here. E, 4T1 cells were incubated in the absence or the presence of 5-FU (5 mg/mL) for 24 hours after 1 hour pretreatment with indicated concentrationsof NAC. The levels of mRNA were quantified by qRT-PCR. All values represent the mean þ SD (n ¼ 3). � , P < 0.05.






Figure 6.

Involvement of Prok2-expressing CD11bþLy6Gþ neutrophils in 5-FU–induced increase in lung metastasis. A, Intrapulmonary growth factor expression wasdetermined inmice administered with or without 5-FU after the resection of the primary tumor. Twenty-eight days after 4T1 cell injection, total RNAswere extractedfrom lung and subjected to qRT-PCR to determine the mRNA expression levels of the indicated growth factors. All values represent mean þ SD (n ¼ 4).� , P < 0.05; �� , P < 0.01. B, Prok2 mRNA expression in intrapulmonary neutrophils. Twenty-eight days after 4T1 cell injection, CD11bþLy6Gþ neutrophils andnon-neutrophil populations were purified from whole lung cells by a flow cytometry. All values represent mean þ SD (n ¼ 3). � , P < 0.05. C, Double-colorimmunofluorescence analysis of lungs was conducted with the anti-Ly6G and anti-Prok2 antibodies. Representative results are shown from three independentexperiments. Original magnification, �200; scale bar, 50 mm. D, Prok2 mRNA expression was determined in mice injected intraperitoneally with anti-Ly6Gantibody (500 mg/head) or SB225002 (0.5mg/kg) starting 14 days after 4T1 cell injection. Twenty-eight days after 4T1 cell injection, total RNAswere extracted fromlung. All values represent mean þ SD (n ¼ 3). �� , P < 0.01. E, Prokr1 and Prokr2 mRNA expression in 4T1 cells. The cells were untreated (open squares) ortreated (closed squares) with 5-FU (5 mg/mL) for 24 hours. Data represent the meanþ SD (n¼ 3). � , P < 0.05. F, In vitro proliferation rates of control vector–treated4T1 (left) or Prokr1-deleted 4T1 cells (right). The cells were cultured with the indicated concentrations of Prok2 for 48 hours, to determine cell proliferationrates using the Cell Counting Kit-8. The ratios of cell numbers were determined by comparing the OD value of untreated cells. All values represent themeanþSD (n¼ 5). � ,P<0.05; �� ,P<0.01; n.s., not significant.G, TheCD11bþLy6Gþneutrophilswere purified fromuntreated or 5-FU–treatedmice andwere placed inthe top well of 0.4 mm pore transwells, while parental or Prokr1-deleted 4T1 cells were placed in the bottom wells of 96-well plate. After 48 hours, the topwell was removed to determine cell proliferation rates of control vector 4T1 (left) or Prokr1-deleted 4T1 cells (right) using the Cell Counting Kit-8. The ratios of cellnumbers were determined by comparing the OD value of untreated cells. All values represent the mean þ SD (n ¼ 5). � , P < 0.05; n.s., not significant. H,Micewere injected intraperitoneallywith 5-FU 14days after control vector-treated 4T1 (Control) orProkr1-deleted 4T1 cells (Prokr1D) injection to determine the tumornumbers. Each symbol represents the tumor numbers per mouse, and bars represent the average for each group (n ¼ 5). �, P < 0.05; n.s., not significant.

Sasaki et al.





DiscussionBreast cancer can be frequently diagnosed at its early stage,

leading to curative resection with surgery (20). However, breastcancer can often recur with a long latency after the surgeryprimarily as a form of metastasis, which is presumed to originatefrom aminute number of dormant cancer cells in the organ (21).Thus, postsurgical adjuvant chemotherapy is commonlyemployed to reduce the risk of recurrence, but biological hetero-geneities in breast cancer can cause difficulties in evaluating theefficacy of the adjuvant therapy as a whole. Moreover, somechemotherapeutics can rather aggravate metastasis in murineexperimental breast cancer models as well as human patients(8–11). Hence, we evaluated the effects of chemotherapeutics onlung metastasis model arising from orthotopic injection of breastcancer cells. We proved that 5-FU, but neither doxorubicin norcyclophosphamide, accelerated lung metastasis together withenhanced neutrophil infiltration. Moreover, neutrophil infiltra-tion was mediated by 5-FU–induced NF-kB–dependent produc-tion of neutrophilic chemokines, Cxcl1 and Cxcl2, in cancer cellsand was associated with enhanced expression of a growth factor,Prok2, with a capacity to augment tumor cell growth. Thus, undersome circumstances, 5-FU can accelerate lung metastasis in micewhen administered after the resection of primary breast cancer.Furthermore, a similar mechanism may work in some cases ofpostsurgical adjuvant therapy for breast cancer, as 5-FU enhancedthe expression of neutrophilic chemokines in a human breastcancer cell line, BT-20, in a NFkB-dependent manner.

5-FU, doxorubicin, and cyclophosphamide exhibit cytotoxicactivities against proliferating cells in distinct manners. 5-FU, ananalogue of uracil (22), promptly enters the cells and is intracel-lularly converted to its active metabolites, several fluoronucleo-tides, which are incorporated into DNA and RNA to inhibit thesynthesis (22). Doxorubicin and cyclophosphamide interferewith DNA synthesis by intercalating with (23) and alkylatingDNA (24), respectively. In addition to the inhibition of DNAsynthesis, 5-FU–derived fluoronucleotides can inhibit the nucle-otide synthetic enzyme, thymidine synthetase (TS). The inhibi-tion can prevent the conversion of dUDP to dTTP, resulting in theimbalance of dNTP amounts and the accumulation of dUTP, bothof which cause DNA damage and eventually p53 activation (22).Through activation of p53, 5-FU can augment the activities ofAtaxia telangiectasia mutant (Atm) kinase to repair DNA damage(25), induce senescence to counteract cell damages (26), andenhance ROS generation (27).

Evidence is accumulating to indicate that 5-FU can activate atranscription factor, NF-kB (28). We also observed that 5-FUinduced mouse and human breast cancer cell lines to expressneutrophilic chemokines, whose enhancer regions possess canon-ical NF-kB–binding sites (29, 30). Moreover, 5-FU enhanced thebinding of NF-kB p65 to these NF-kB–binding sites, indicatingthat 5-FU–induced chemokine expression was mediated byNF-kB activation. NF-kB activation can be observed in cellularsenescence provoked by p53 activation(31), but 5-FU failed toincrease senescence-associated b-galactosidase staining, a charac-teristic feature of cellular senescence, in 4T1 and TS/A cells(Supplementary Fig. S10). Activated p53 can also activate Atmkinase, which can sequentially activate NF-kB essential modifier(NEMO), a component of the IKK complex with a capacity toinduce nuclear translocation and subsequent activation of NF-kBp65/p50 complex (32, 33). The involvement of this mechanism,

however, was negated by our observation that a specific Atminhibitor, KU-60019 (34), failed to abrogate 5-FU–inducedexpression of Cxcl1 and Cxcl2 in mouse breast cancer cell lines,4T1 and TS/A cells (Supplementary Fig. S11). In contrast, 5-FUincreased intracellular ROS amounts and NAC-mediated reduc-tion inROS significantly reduced 5-FU–inducedNF-kBbinding toits cis-elements in mouse Cxcl1 and Cxcl2 genes, and their sub-sequent mRNA expression. Consistently, H2O2 enhanced Cxcl1and Cxcl2 mRNA expression in mouse breast cancer cell lines(Supplementary Fig. S12). Thus, 5-FU can activate NF-kB via ROSelevation and eventually augment the expression of NF-kB targetgenes Cxcl1 and Cxcl2.

Accumulating evidence indicates that as metastasis proceeds,lungs are infiltrated by inflammatory cells, which can contributeto metastasis progression by providing growth factors and sup-pressing immune responses (35). Among inflammatory cells, theroles of macrophages in lung metastasis have been extensivelyexamined. Intrapulmonary macrophages can be classified intointerstitial and alveolar macrophages, which reside in interstitialand alveolar spaces, respectively (36). Metastasis-associatedmacrophages (MAM) are defined as those which migrate fromthe circulation into interstitial space of metastatic lungs (37–39)and can promote lung metastasis. Another intrapulmonary mac-rophage subpopulation, alveolar macrophages, can also fosterlung metastasis through their immunosuppressive actions (40).However,we couldnot detect a significant increase inmacrophagenumbers in the present metastatic lungs, making it a remotepossibility of the contribution of macrophages to this acceleratedlung metastasis process.

Other types of infiltrating inflammatory cells include neutro-phils (41) and myeloid-derived suppressor cells (MDSC; ref. 42).MDSCs are immature myeloid cells in contrast to terminallydifferentiated myeloid cells such as neutrophils. Mouse MDSCsexpress simultaneously CD11b and Gr-1 (also known as Ly6G/Ly6C), and function as immunosuppressive cells. Inmice, surfacemarkers are shared by neutrophils andMDSCs, butMDSCs can bediscriminated from neutrophils by using the combination ofseveral markers (43). It is presumed that MDSCs are recruitedfrom bonemarrow to a target organ to provide a niche to supportcancer cell growth before the arrival of cancer cells (44). Incontrast, in the present lung metastasis model, we showed thatCD11bþLy6Gþ myeloid cells were increased in lungs only at thelate phase of metastasis progression but lacked characteristic G-MDSC features. Thus, it is more likely that the increasedCD11bþLy6Gþ cells represented neutrophils, but not G-MDSCs.

Several lines of evidence indicate that infiltrating neutrophilscan promotemetastasis by producingmetalloproteinase (45) andgrowth factors such as Prok2 (46). Indeed, our unbiased analysisrevealed a selectively enhanced expression of Prok2 by neutro-phils infiltrating intometastatic lungs. Prok1 andProk2 are highlyconserved small peptides (47) andmammalian homologs of Bv8,whichwasoriginally identified fromamphibian skin (48). Bothofthem were identified as a potent contractor of gastrointestinalsmooth muscle (47), and subsequent studies revealed that Prok1and Prok2 have diverse effects on feeding, drinking, neuronmigration and survival, angiogenesis, hematopoiesis, and inflam-mation by binding their specific G protein–coupled receptors,Prokr1 and Prokr2 (49). Accumulating evidence further indicatesthe crucial involvement of Prok1 and Prok2 in neovascularizationobserved in various types of cancers (50, 51). Prok2 or Prok2-expressing neutrophils in vitro augmented, in a dose-dependent






manner, the proliferation of 4T1 cells, which expressed Prokr1.Moreover, the proliferative responsewas abrogated byProkr1 genedeletion in 4T1 cells. Furthermore, 4T1 cells deficient in Prokr1gene did not exhibit 5-FU–mediated enhanced lung metastasisformation. Thus, the Prok2–Prokr1 axis can directly regulatetumor cell proliferation, thereby contributing to tumor progres-sion. This assumption may be supported by the analysis using apublic database, PrognoScan, which demonstrates that highPROKR1 and PROK2 gene expression is associated with shorteroverall survival and relapse-free survival, respectively (Supple-mentary Fig. S13A and S13B).

Of interest is that enhancement of Prok2 expression wasdetected in neither neutrophils isolated from the lungs of untreat-ed mice (Supplementary Fig. S2B) nor neutrophils stimulatedin vitrowithCxcl1 orCxcl2 (Supplementary Fig. S14). Thus, 5-FU–

induced production of these chemokines may not be sufficient tofully activate neutrophils to express Prok2. We observed thatgranulocyte colony-stimulating factor (G-CSF), a representativeNFkB targetmolecule (52),was abundantly detected in lungs after5-FU administration (Supplementary Fig. S15), and that G-CSFcan in vitro induce Prok2 expression inCD11bþLy6Gþneutrophils(Supplementary Fig. S14), consistent with the previous reports(50, 53). Thus, 5-FU–induced enhanced lung metastasis can bemediated by the cooperative actions of neutrophilic chemokinesand G-CSF, both of which are NFkB target and, therefore, NFkBmay be a candidate target molecule to improve the efficacy ofpostsurgical neoadjuvant chemotherapy for breast cancer.

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

Authors' ContributionsConception and design: S. Sasaki, T. Baba, N. MukaidaDevelopment of methodology: S. Sasaki, T. BabaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S. Sasaki, T. Baba, H. Muranaka, Y. Tanabe,C. Takahashi, N. MukaidaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Sasaki, T. Baba, H. Muranaka, N. MukaidaWriting, review, and/or revision of the manuscript: S. Sasaki, N. MukaidaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): S. Matsugo, N. MukaidaStudy supervision: N. Mukaida

AcknowledgmentsWe would like to express our sincere appreciation to Professors Noriko

Gotoh (Kanazawa University) and Tsuneyasu Kaisho (Wakayama MedicalUniversity) for providing us with material support and advising us onCRIPSR-Cas 9 methodology, respectively. This work was supported partly bythe Grant-in-Aid for Scientific Research (C) from the Japan Society for thePromotion of Science (JSPS) KAKENHI grant number 17K07159.

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

Received August 31, 2017; revised January 23, 2018; accepted April 5, 2018;published first April 11, 2018.

References1. Duijm LE, Guit GL, Zaat JO, Koomen AR, Willebrand D. Sensitivity,

specificity and predictive values of breast imaging in the detection ofcancer. Br J Cancer 1997;76:377–81.

2. Morrow M, Schnitt SJ, Norton L. Current management of lesionsassociated with an increased risk of breast cancer. Nat Rev Clin Oncol2015;12:227–38.

3. Chen L, Linden HM, Anderson BO, Li CI. Trends in 5-year survival ratesamong breast cancer patients by hormone receptor status and stage. BreastCancer Res Treat 2014;147:609–16.

4. Zhang XH, Giuliano M, Trivedi MV, Schiff R, Osborne CK. Metastasisdormancy in estrogen receptor-positive breast cancer. Clin Cancer Res2013;19:6389–97.

5. Anampa J, Makower D, Sparano JA. Progress in adjuvant chemotherapy forbreast cancer: an overview. BMC Med 2015;13:195.

6. Joyce JA, Pollard JW.Microenvironmental regulation ofmetastasis. Nat RevCancer 2009;9:239–52.

7. Adjei IM, Blanka S. Modulation of the tumormicroenvironment for cancertreatment: a biomaterials approach. J Funct Biomater 2015;6:81–103.

8. Gingis-Velitski S, LovenD, Benayoun L,MunsterM, Bril R, Voloshin T, et al.Host response to short-term, single-agent chemotherapy induces matrixmetalloproteinase-9 expression and accelerates metastasis in mice. CancerRes 2011;71:6986–96.

9. KaragiannisGS, Pastoriza JM,WangY,HarneyAS, EntenbergD, Pignatelli J,et al. Neoadjuvant chemotherapy induces breast cancermetastasis througha TMEM-mediated mechanism. Sci Transl Med 2017;9. pii: eaan0026.

10. Volk-Draper L, Hall K, Griggs C, Rajput S, Kohio P, DeNardo D, et al.Paclitaxel therapy promotes breast cancer metastasis in a TLR4-dependentmanner. Cancer Res 2014;74:5421–34.

11. Pierga JY, Asselain B, JouveM,Dieras V,CartonM, LaurenceV, et al. Effect ofadjuvant chemotherapy on outcome in patients with metastatic breastcarcinoma treated with first-line doxorubicin-containing chemotherapy.Cancer 2001;91:1079–89.

12. Wong CW, Song C, Grimes MM, Fu W, Dewhirst MW, Muschel RJ, et al.Intravascular location of breast cancer cells after spontaneousmetastasis tothe lung. Am J Pathol 2002;161:749–53.

13. DuPre SA, Redelman D, Hunter KW Jr. The mouse mammary carcinoma4T1: characterization of the cellular landscape of primary tumours andmetastatic tumour foci. Int J Exp Pathol 2007;88:351–60.

14. Lasfargues EY, Ozzello L. Cultivation of human breast carcinomas. J NatlCancer Inst 1958;21:1131–47.

15. Lollini PL, de Giovanni C, Eusebi V, Nicoletti G, Prodi G, Nanni P.High-metastatic clones selected in vitro from a recent spontaneousBALB/c mammary adenocarcinoma cell line. Clin Exp Metastasis 1984;2:251–9.

16. White JR, Lee JM, Young PR, Hertzberg RP, Jurewicz AJ, Chaikin MA, et al.Identification of a potent, selective non-peptide CXCR2 antagonist thatinhibits interleukin-8-induced neutrophil migration. J Biol Chem 1998;273:10095–8.

17. Richmond A. Nf-kappa B, chemokine gene transcription and tumourgrowth. Nat Rev Immunol 2002;2:664–74.

18. Shi R, Re D, Dudl E, Cuddy M, Okolotowicz KJ, Dahl R, et al. Chemicalbiology strategy reveals pathway-selective inhibitor of NF-kappaB activa-tion induced by protein kinase C. ACS Chem Biol 2010;5:287–99.

19. Granot Z, Jablonska J. Distinct functions of neutrophil in cancer and itsregulation. Mediators Inflamm 2015;2015:701067.

20. Rostas JW, Dyess DL. Current operative management of breast cancer: anage of smaller resections and bigger cures. Int J Breast Cancer 2012;2012:516417.

21. Jin X, Mu P. Targeting breast cancer metastasis. Breast Cancer 2015;9:23–34.

22. Longley DB, Harkin DP, Johnston PG. 5-Fluorouracil: mechanisms ofaction and clinical strategies. Nat Rev Cancer 2003;3:330–8.

23. Thorn CF, Oshiro C, Marsh S, Hernandez-Boussard T, McLeod H, Klein TE,et al. Doxorubicin pathways: pharmacodynamics and adverse effects.Pharmacogenet Genomics 2011;21:440–6.

24. Fleming RA. An overview of cyclophosphamide and ifosfamide pharma-cology. Pharmacotherapy 1997;17:146s–54s.

25. Elia�s J, Dimitrio L, Clairambault J, Natalini R. The p53 protein and itsmolecular network: modelling a missing link between DNA damage andcell fate. Biochim Biophys Acta 2014;1844:232–47.


Sasaki et al.




26. Tonnessen-Murray CA, Lozano G, Jackson JG. The regulation of cellularfunctions by the p53 protein: cellular senescence. Cold Spring HarborPerspect Med 2017;7. pii: a026112.

27. Vurusaner B, Poli G, Basaga H. Tumor suppressor genes and ROS: complexnetworks of interactions. Free Radical Biol Med 2012;52:7–18.

28. Vinod BS, Antony J, Nair HH, Puliyappadamba VT, Saikia M, NarayananSS, et al. Mechanistic evaluation of the signaling events regulating curcu-min-mediated chemosensitization of breast cancer cells to 5-fluorouracil.Cell Death Dis 2013;4:e505.

29. Singha B, Gatla HR, Vancurova I. Transcriptional regulation of chemokineexpression in ovarian cancer. Biomolecules 2015;5:223–43.

30. Orlichenko LS, Behari J, Yeh TH, Liu S, Stolz DB, Saluja AK, et al. Transcrip-tional regulation of CXC-ELR chemokines KC and MIP-2 in mouse pancre-atic acini. Am J Physiol Gastrointest Liver Physiol 2010;299:G867–76.

31. Coppe JP, Patil CK, Rodier F, Sun Y, Munoz DP, Goldstein J, et al.Senescence-associated secretory phenotypes reveal cell-nonautonomousfunctions of oncogenic RAS and the p53 tumor suppressor. PLoS Biol2008;6:2853–68.

32. McCool KW, Miyamoto S. DNA damage-dependent NF-kappaB activa-tion: NEMO turns nuclear signaling inside out. Immunol Rev 2012;246:311–26.

33. Huang TT, Wuerzberger-Davis SM, Wu ZH, Miyamoto S. Sequentialmodification of NEMO/IKKgamma by SUMO-1 and ubiquitin mediatesNF-kappaB activation by genotoxic stress. Cell 2003;115:565–76.

34. Golding SE, Rosenberg E, Valerie N, Hussaini I, Frigerio M, Cockcroft XF,et al. Improved ATM kinase inhibitor KU-60019 radiosensitizes gliomacells, compromises insulin, AKT and ERK prosurvival signaling, andinhibits migration and invasion. Mol Cancer Ther 2009;8:2894–902.

35. Kitamura T, Qian BZ, Pollard JW. Immune cell promotion of metastasis.Nat Rev Immunol 2015;15:73–86.

36. LumengCN. Lungmacrophage diversity and asthma. AnnAmThoracic Soc2016;13:S31–4.

37. Qian BZ, Li J, Zhang H, Kitamura T, Zhang J, Campion LR, et al. CCL2recruits inflammatory monocytes to facilitate breast-tumour metastasis.Nature 2011;475:222–5.

38. Kitamura T, Qian BZ, Soong D, Cassetta L, Noy R, Sugano G, et al. CCL2-induced chemokine cascade promotes breast cancer metastasis by enhanc-ing retention of metastasis-associated macrophages. J Exp Med2015;212:1043–59.

39. Qian BZ, Zhang H, Li J, He T, Yeo EJ, Soong DY, et al. FLT1 signaling inmetastasis-associated macrophages activates an inflammatory signaturethat promotes breast cancer metastasis. J Exp Med 2015;212:1433–48.

40. Sharma SK, ChintalaNK, Vadrevu SK, Patel J, KarbowniczekM,MarkiewskiMM. Pulmonary alveolar macrophages contribute to the premetastaticniche by suppressing antitumor T cell responses in the lungs. J Immunol2015;194:5529–38.

41. Singel KL, Segal BH.Neutrophils in the tumormicroenvironment: trying toheal the wound that cannot heal. Immunol Rev 2016;273:329–43.

42. Marvel D, Gabrilovich DI. Myeloid-derived suppressor cells in thetumor microenvironment: expect the unexpected. J Clin Invest 2015;125:3356–64.

43. Youn JI, Collazo M, Shalova IN, Biswas SK, Gabrilovich DI. Characteriza-tion of the nature of granulocytic myeloid-derived suppressor cells intumor-bearing mice. J Leukoc Biol 2012;91:167–81.

44. Keskinov AA, Shurin MR. Myeloid regulatory cells in tumor spreading andmetastasis. Immunobiology 2015;220:236–42.

45. Dumitru CA, GholamanH, Trellakis S, Bruderek K, Dominas N, Gu X, et al.Tumor-derived macrophage migration inhibitory factor modulates thebiology of head and neck cancer cells via neutrophil activation. Int JCancer 2011;129:859–69.

46. Kowanetz M, Wu X, Lee J, Tan M, Hagenbeek T, Qu X, et al. Granulocyte-colony stimulating factor promotes lung metastasis through mobilizationof Ly6GþLy6Cþ granulocytes. PNAS 2010;107:21248–55.

47. Li M, Bullock CM, Knauer DJ, Ehlert FJ, Zhou QY. Identification of twoprokineticin cDNAs: recombinant proteins potently contract gastrointes-tinal smooth muscle. Mol Pharmacol 2001;59:692–8.

48. Mollay C, Wechselberger C, Mignogna G, Negri L, Melchiorri P, BarraD, et al. Bv8, a small protein from frog skin and its homologuefrom snake venom induce hyperalgesia in rats. Eur J Pharmacol 1999;374:189–96.

49. Monnier J, Samson M. Prokineticins in angiogenesis and cancer. CancerLett 2010;296:144–9.

50. Shojaei F, Wu X, Qu X, Kowanetz M, Yu L, Tan M, et al. G-CSF-initiatedmyeloid cell mobilization and angiogenesis mediate tumor refractorinessto anti-VEGF therapy in mouse models. PNAS 2009;106:6742–7.

51. Goi T, Nakazawa T, Hirono Y, Yamaguchi A. The anti-tumor effect isenhanced by simultaneously targeting VEGF and PROK1 in colorectalcancer. Oncotarget 2015;6:6053–61.

52. Nishizawa M, Nagata S. Regulatory elements responsible for inducibleexpression of the granulocyte colony-stimulating factor gene in macro-phages. Mol Cell Biol 1990;10:2002–11.

53. Zhong C, Qu X, Tan M, Meng YG, Ferrara N. Characterization andregulation of bv8 in human blood cells. Clin Cancer Res 2009;15:2675–84.






2018;17:1515-1525. Published OnlineFirst April 11, 2018.Mol Cancer Ther Soichiro Sasaki, Tomohisa Baba, Hayato Muranaka, et al. Lung

induced Aggravation of Breast Cancer Metastasis to−5-Fluorouracil expressing Neutrophil Infiltration in−Involvement of Prokineticin 2

Updated version

10.1158/1535-7163.MCT-17-0845doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2018/04/11/1535-7163.MCT-17-0845.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/17/7/1515.full#ref-list-1

This article cites 51 articles, 13 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mct.aacrjournals.org/content/17/7/1515To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-17-0845

http://mct.aacrjournals.org/content/suppl/2018/04/11/1535-7163.MCT-17-0845.DC1

http://mct.aacrjournals.org/content/17/7/1515.full#ref-list-1

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/17/7/1515


Documents

Molecular Cancer Therapeutics - Involvement of Prokineticin 2 ......Cancer Biology and Translational Studies Involvement of Prokineticin 2–expressing Neutrophil Inﬁltration in