tumor-associated macrophages promote ... · phyllodes tumor, and ccl18 is an independent prognostic...

Microenvironment and Immunology

Tumor-Associated Macrophages PromoteMalignantProgressionofBreast PhyllodesTumorsby Inducing Myofibroblast DifferentiationYan Nie1,2, Jianing Chen1,2, Di Huang1,2, Yandan Yao1,2, Jiewen Chen1,2,Lin Ding1,3, Jiayi Zeng4, Shicheng Su1,2, Xue Chao1,2, Fengxi Su1,2, Herui Yao1,3,Hai Hu1,3, and Erwei Song1,2

Abstract

Myofibroblast differentiation plays an important role in themalignant progression of phyllodes tumor, a fast-growing neo-plasm derived from periductal stromal cells of the breast. Macro-phages are frequently found in close proximity with myofibro-blasts, but it is uncertain whether they are involved in themyofibroblast differentiation during phyllodes tumor progres-sion. Here we show that increased density of tumor-associatedmacrophage (TAM) correlates with malignant progression ofphyllodes tumor. We found that TAMs stimulated myofibroblastdifferentiation and promoted the proliferation and invasion ofphyllodes tumor cells. Furthermore, we found that levels of thechemokine CCL18 in TAM was an independent prognostic factor

of phyllodes tumor. Mechanistic investigations showed thatCCL18 promoted expression of a-smooth muscle actin, a hall-mark of myofibroblast, along with the proliferation and invasionof phyllodes tumor cells, and that CCL18-driven myofibroblastdifferentiation was mediated by an NF-kB/miR-21/PTEN/AKTsignaling axis. In murine xenograft models of human phyllodestumor, CCL18 accelerated tumor growth, induced myofibroblastdifferentiation, and promoted metastasis. Taken together, ourfindings indicated that TAM drives myofibroblast differentiationandmalignant progression of phyllodes tumor through a CCL18-driven signaling cascade amenable to antibody disruption.Cancer Res; 77(13); 3605–18. �2017 AACR.

IntroductionBreast phyllodes tumor is a biphasic breast tumor composed of

cellular spindle stroma with epithelial elements. It constitutes0.3% to 1% of all breast tumors and 2.5% of fibroepitheliallesions of the breast (1). Currently, phyllodes tumors are histo-logically classified as benign, borderline, or malignant on thebasis of stromal cellularity, mitotic activity of stromal cells,stromal nuclear atypia, stromal overgrowth, and types of border(infiltrating or pushing). The clinical outcome of phyllodestumors is hard to predict, with frequent local relapse and some-times distant metastasis. Adjuvant chemotherapy or radiotherapyis not effective against phyllodes tumors (2). The potentiallyrecurring and metastatic behavior of phyllodes tumors is attrib-uted to the characteristics of stromal cells, mainly fibroblasts (3).

The normal fibroblast can acquire an "activated" phenotype,which expresses the a-smooth muscle actin (a-SMA) as a hall-mark and is so-called as myofibroblasts. Our previous studieshave indicated myofibroblasts were the major malignant com-ponent of phyllodes tumors. The increased myofibroblast pop-ulation drives the tumorigenicity of phyllodes tumors. In addi-tion, a-SMA can serve as an independent prognostic factor forphyllodes tumors with better predictive values than histologicclassification (4). We further demonstrated that the fibroblasts–myofibroblasts transition in phyllodes tumors is driven by theelevated miR-21 (4), whereas the mechanism of miR-21 upregu-lation and how it drives tumorigenicity of phyllodes tumorsremain unknown.

It is well established that tumor-associated macrophages(TAM) are one of the most abundant cell type in tumor micro-environment (5), which are involved in tumor metastasis andprogression (6). Clinical and epidemiologic studies have shown astrong correlation between the increased TAMs density and poorprognosis in several types of cancer (6, 7), including breast cancer(8, 9). However, how TAMs impact on the malignant progressionof phyllodes tumor andwhether TAMsdensity canbe aprognosticfactor for phyllodes tumors are still unknown.

Macrophages are usually found in close proximity with colla-gen-producing myofibroblasts (9). Macrophages produce profi-brotic mediators that directly activate fibroblasts, includingTGFb1 and platelet-derived growth factor (PDGF; ref. 10). Macro-phages alsoprovide insulin-like growth factor-1,which stimulatesthe proliferation and survival of myofibroblasts and promotescollagen synthesis by these cells (11). In this study, we found thatTAMs are essential for driving myofibroblast differentiation

1Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics andGene Regulation, Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University,Guangzhou, P.R. China. 2Breast Tumor Center, Sun Yat-Sen Memorial Hospital,Sun Yat-Sen University, Guangzhou, P.R.China. 3Department of Oncology, SunYat-Sen Memorial Hospital, Sun Yat-Sen University, Guangzhou, P.R. China.4Guangzhou Zhixin High School, Guangzhou, P.R. China.

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

Corresponding Authors: Erwei Song and Hai Hu, Sun Yat-Sen Memorial Hos-pital, Sun Yat-Sen University, 107 Yanjiang West Road, Guangzhou, Guangdong510120, P.R. China. Phone: 8620-8133-2603; Fax: 8620-8133-2853; E-mail:[email protected]; and Hai Hu, [email protected]

doi: 10.1158/0008-5472.CAN-16-2709

�2017 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 3605

on June 11, 2020. © 2017 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst May 16, 2017; DOI: 10.1158/0008-5472.CAN-16-2709

http://crossmark.crossref.org/dialog/?doi=10.1158/0008-5472.CAN-16-2709&domain=pdf&date_stamp=2017-6-17

http://cancerres.aacrjournals.org/

(fibroblasts–myofibroblasts transition) in the malignant progres-sion of phyllodes tumors via the CCL18/NF-kB/miR-21/PTEN/AKT axis and targeting CCL18 is a promising strategy for treatingphyllodes tumors.

Materials and MethodsPatients and tissue samples

Breast phyllodes tumor samples were obtained from 268female patientswith 167benign, 36borderline, and65malignantphyllodes tumors in the Breast Tumor Center, Sun Yat-SenMemorial Hospital, Sun Yat-Sen University, from January 2000to June 2011. The patients were followed up for 8–148 months(median follow-up is 112 months). Pathologic diagnosis, as wellas mitoses and stromal overgrowth status, was confirmed by twopathologists independently. All human samples were collectedwith informed consent from the donors according to the Inter-national Ethical Guidelines for Biomedical Research InvolvingHumanSubjects (CIOMS). The studywasperformed after approv-al by the institutional review board (IRB) of Sun Yat-Sen Memo-rial Hospital.

Separation and culture of primary cells from breast phyllodestumors

Phyllodes tumor cellswere isolated frombenign andmalignantphyllodes tumors as previously described (4). TAMswere isolatedfrom eight fresh breast malignant phyllodes tumor samples aspreviously described (12, 13), with slight modifications. Briefly,the tissues were minced into small (1–2 mm) pieces anddigested with 5% FBS DMEM containing 2 mg/mL collagenaseI and 2 mg/mL hyaluronidase (Sigma) at 37�C for 2 hours. Thecells were sequentially filtered through 500 mm mesh, 100 mm,and 70 mm cell strainer. The cells were then centrifuged in aBeckman Allegra X-15R centrifuge at 2,500 rpm for 20 minuteswith 1 mL cell suspension above 5 mL 45% Percoll (GE Health-care) in themiddle and5mL60%Percoll at thebottom in a15mLtube. Mononuclear cells were collected from the cell layer in theinterphase between45%and60%Percoll. CD14þmonocytes andmacrophages were isolated by a magnetic-activated cell sortingusing direct CD14 Isolation Kit (Miltenyi Biotec) according to themanufacturer's instructions.

Chromatin immunoprecipitation assayChromatin immunoprecipitation (ChIP) assay was performed

using Pierce Agarose ChIP Kit (26156, Thermo Scientific) accord-ing to the manufacturer's instructions. The specific sequencesfrom immunoprecipitated and input DNA were determined byPCR primers for miR-21 promoter upstream regions:

miR-21 promoter forward, 50 -TCCCCTCTGGGAAGTTTC-30,reverse, 50 -TTGGCTCTACCCTTGTTT-30.The negative control primer:miR-21-NC-1: forward, 50-TTCCTCATTTCCCTAAACAACAA-30

reverse, 50 -AATCTACCAGGGATAGCCATAGTC-30.miR-21-NC-2: forward, 50-GAGGACTTCCCCAACTTAACTATG -30

reverse, 50 -TTATTCTCAAGCAGCAGACCAG -30.

Animal experimentsAll animal work was conducted in accordance with a protocol

approved by the Institutional Animal Care and Use Committee atthe Medical School of Sun Yat-Sen University. Breast phyllodes

tumor cells (1 � 107) mixed with MDMs (2 � 106) were inoc-ulated into the mammary fat pads of 6-week-old female nudemice. Mice were from Beijing Vital River Laboratory AnimalTechnology Co., Ltd. For antibody treatment and the mice wereinjected with CCL18-specific neutralizing antibody (catalog no.ab9849, Abcam) or isotype-matched IgG (Abcam) via the tailvein at 1 mg/kg twice weekly after the xenografts becamepalpable (around 0.5 cm in diameter). In some groups, intra-tumoral injection of M-CSF at 0.2 mg/kg was performed twiceweekly after inoculating and 48 hours prior to antibody injec-tion. Tumor growth was evaluated by monitoring tumor vol-ume (TV¼ length�width2 � 0.5) every 3 days for 8 weeks. Theanimals were sacrificed when the xenografts reached 1.5 cm indiameter. Tumor xenografts and the livers and lungs of micewere harvested for further evaluation. Cryosections (4 mm) ofthe harvested organs were hematoxylin and eosin (H&E) forhistologic assessment, and total RNA was extracted for qRT-PCRanalysis of human hypoxanthine phosphoribosyltransferase(HPRT) mRNA expression.

Statistical analysisThe in vitro data were depicted as mean � SD of three inde-

pendent experiments performed in triplicate. All statistical anal-yses were performed using SPSS 16.0 statistical software package(SPSS). Student t test and one-way ANOVA was used to compareCCL18 expression levels between the phyllodes tumors withdifferent tumor grades, whereas c2 test was used to analyze therelationship between CCL18 expression and clinicopathologicstatus. Kaplan–Meier curves and log-rank test were used to com-pare overall survival (OS) and disease-free survival (DFS) indifferent patient groups. In all cases, P < 0.05 was consideredstatistically significant.

ResultsTAM density is associated with malignant progression ofphyllodes tumor, and CCL18 is an independent prognosticmarker for phyllodes tumor patients

Evidence from clinical and epidemiologic studies has shown astrong correlation between TAMs density and poor prognosis inseveral types of cancer (6, 7). To investigate whether malignantprogression of phyllodes tumor is correlated with TAMs density,we examined the presence of TAMs in 268 phyllodes tumorsamples, including 167 benign, 36 borderline, and 65 malignantphyllodes tumors. Normal breast tissues were used as control. Wetested the expression of CCL18, the hallmark of TAM in paraffin-embedded phyllodes tumor samples by IHC. The expression ofCCL18 level inmalignant phyllodes tumors is significantly higherthan that in benign and borderline phyllodes tumors (Fig. 1A).Previously, we have found PITPNM3 was the receptor of CCL18on the surface of breast tumor cells (12). CCR8 was also reportedas the CCL18 receptor (14). We then examined the level ofPITPNM3or CCR8 in phyllodes tumors and found that PIPTNM3was increased in malignant phyllodes tumors (Fig. 1A), whereasCCR8 was not (Supplementary Fig. S1A). CCL18 is expressed inboth TAMs and dendritic cells (DC; 14). To confirm whetherCCL18þ cells were macrophages, two macrophages markersCD68 and CD163 were used. CD68 is specially expressed bymacrophages and in tolerogenic DCs (15). CD163 is a marker ofM2 macrophages. Microscopy revealed that both CD68 (Supple-mentary Fig. S1B) and CD163 (Fig. 1B) immunostaining were

Nie et al.

Cancer Res; 77(13) July 1, 2017 Cancer Research3606




colocalized with CCL18þ cells in breast phyllodes tumor tissues,showing the CCL18þ cells were M2-like macrophages, which infact were TAMs in the phyllodes tumors.

Furthermore, we measured the CCL18 mRNA levels in freshfrozen phyllodes tumor tissues andprimary phyllodes tumor cellsisolated frombenign, borderline, ormalignant phyllodes tumors,

Figure 1.

TAMs density is associated with malignant progression of phyllodes tumors. A, IHC staining shows both CCL18 and its corresponding receptor PITPNM3 areelevated in the malignant breast phyllodes tumors. Scale bar, 20 mm. B, Microscopy for immunostaining of normal breast tissue and breast phyllodes tumortissues shows CCL18 is colocalizedwith CD163, themarker of M2macrophages. Scale bar, 20 mm.C,Real-time qRT-PCR demonstrates that the expression of CCL18 isincreased in phyllodes tumor tissues. RNAwere extracted fromnormal breast tissues or breast phyllodes tumors. Bars indicate RNAexpression normalized toGAPDH� SD of eight samples in each group (� , P < 0.05; �� , P < 0.01; and �� , P < 0.001 as compared with normal breast tissue). D, Serum CCL18 level is increased in thepatients with breast phyllodes tumor, as determined by ELISA. Bars correspond to mean � SD (� , P < 0.05; �� , P < 0.01, and �� , P < 0.001). E, Kaplan–Meier OScurve of patients with low and high CCL18, with a median follow-up period of 112 months. F, Kaplan–Meier DFS curve of patients with low and high CCL18,with a median follow-up period of 112 months.

TAMs Promote Malignancy of Phyllodes Tumors

www.aacrjournals.org Cancer Res; 77(13) July 1, 2017 3607




using normal breast tissues as a control. UsingqRT-PCR,we foundthat the mRNA levels of CCL18 from fresh frozen tissues wereprogressively increased from normal breast tissue to benign,borderline, andmalignant phyllodes tumors (Fig. 1C). ThemRNAlevels of CCL18 from primary phyllodes tumor cells did not showan increasing tendency from benign, borderline to malignantphyllodes tumor cells. Although TAMs isolated from the malig-nant phyllodes tumors expressed extremely high levels of CCL18mRNA (Supplementary Fig. S1C). ELISA assay was used to mea-sure the serum CCL18 level of phyllodes tumor patients or thesecreted CCL18 level in the culture suspension of primary cellsisolated from phyllodes tumor tumors. We found that the CCL18protein levels in the serum of patients with malignant phyllodestumors were significantly higher than in those with benign andborderline phyllodes tumors (Fig. 1D). However, primary phyl-lodes tumor cells secreted few CCL18 into the culture medium.Similar to themRNA level, the secreted CCL18 level in the culturesupernatant of the primary TAMs isolated from malignant phyl-lodes tumor tumors was much higher than in those from phyl-lodes tumor cells (Supplementary Fig. S1D). Together, theseresults indicate that TAM density is associated with the malignantprogression of phyllodes tumors and CCL18 is mostly producedby TAMs.

Next, we tested whether CCL18 had clinical prognostic valuefor phyllodes tumor patients. The 268 phyllodes tumor patientswere followed up for 8–148 months (median follow-up time is112months). During the follow-up, 49 caseswere diagnosedwithrecurrence, including 18 in the benign group, 9 in the borderlinegroup, and 22 in the malignant group. In addition, 31 cases werediagnosed with metastasis, with 3 in the borderline group and 28in the malignant group.

We then analyzed the association of CCL18 expressionwith theclinicopathologic status of phyllodes tumors (SupplementaryTable S1). The expression of CCL18 increased with higher tumorgrade, mitotic activity, and stromal overgrowth (P < 0.001), butwas not associatedwith age and the size of tumor (SupplementaryTable S1). The expression of CCL18 was also more abundant inthe phyllodes tumors with local recurrence and distal metastasis(P<0.001; Supplementary Table S1). Furthermore, Kaplan–Meiersurvival curve demonstrated that patients with lowCCL18 expres-sion (staining index, SI� 4, refers to the Supplementary Methodsand ref. 4) have a longer overall survival (OS) and disease-freesurvival (DFS) than those with high CCL18 expression (P <0.001, Fig. 1E and F). We also used ROC curve to evaluate theefficacy of serum CCL18 in phyllodes tumor patients as a diag-nostic marker. It showed that CCL18 serum level could distin-guish aggressive cases from benign phyllodes tumors (Supple-mentary Fig. S1E and S1F). In addition, multivariate Cox regres-sion analyses demonstrated that CCL18 (P < 0.001), stromalovergrowth (P < 0.001), and tumor grading (P ¼ 0.006) wereindependent prognostic predictors for local recurrence-free sur-vival (LRFS; Supplementary Table S2). These results suggest thatthe TAMs density is associated with malignant progression ofphyllodes tumor, and the levels of CCL18 can be used to predictthe outcome of phyllodes tumor patients.

TAMs induce myofibroblast differentiation and promote theproliferation and invasion of the phyllodes tumor cells

Macrophages are frequently found in close proximity withcollagen-producing myofibroblasts (9), and there is strong evi-dence that this interaction is reciprocal (16). Macrophages pro-

duce profibrotic mediators that directly activate fibroblasts,including TGFb1, platelet-derived growth factor (PDGF; ref. 10).Macrophages also produce insulin-like growth factor-I, whichstimulates the proliferation and survival of myofibroblasts andpromotes collagen synthesis by these cells (11). As we have foundTAMdensity is correlatedwithmalignant progression of phyllodestumors and our previous studies have reported that myofibroblastdifferentiation is associated with the malignant progression ofphyllodes tumors (4), we speculated that TAMs may play animportant role in the myofibroblast differentiation of breast phyl-lodes tumors. Todemonstrate this hypothesis, we culturedprimarybenign phyllodes tumor cells with primary TAMs isolated frommalignant phyllodes tumor patients. TAMs in breast cancers areprimarily M2 cells (17). Because M2 macrophages, which can beinduced from monocyte-derived macrophages (MDM) by M-CSF(18), have been shown to induce myofibroblast transition (9), wealso cultured benign phyllodes tumor cells with M-CSF pretreatedMDMs by transwell assay. Benign phyllodes tumor cells wereseeded into the bottom chamber. MDMs or TAMs were seededinto the top chamber of a six-well transwell apparatus. The mRNAand protein level of a-SMA, the hallmark of myofibroblasts, wasdetected to evaluate the phenotype change of phyllodes tumorcells. Compared with untreated phyllodes tumor cells, cells cocul-tured with primary TAMs or M-CSF–activated MDMs had a sig-nificant higher level ofa-SMA (Fig. 2A and B). The expression levelof a-SMA in phyllodes tumor cells treated with M-CSF aloneor cocultured with untreated MDMs had no obviously changes(Fig. 2AandB). Immunofluorescent staining (Fig. 2C) also showedthe increased levels of a-SMA and fibroblast activation protein(FAP), which is also a marker of myofibroblast.

Because myofibroblasts are known to have an increased abilityto induce collagen gel contraction (19), collagen contraction assaywas used to test whether TAMs-treated phyllodes tumor cellsfunction as myofibroblast. We observed that benign phyllodestumor cells cocultured with TAMs or M-CSF–pretreated MDMscontracted collagen gels to a much greater extent than phyllodestumor cells treated with or without M-CSF or phyllodes tumorcells cocultured with untreated MDMs (Fig. 2D). Together, thesefindings suggest that TAMs induce myofibroblast function in thephyllodes tumor cells.

Previous studies have reported thatmyofibroblasts in epithelialtumors have an increased proliferative activity and can promotecancer invasion and metastasis (20, 21). We then examined theeffects of TAMs on the proliferation, migration, and invasion ofprimary phyllodes tumor cells. Cell viability and clonogenicassays showed that TAMs or M-CSF–pretreated MDMs increasedthe foci formation of phyllodes tumor cells (Fig. 2E), indicatingthat TAMs promoted the growth of phyllodes tumor cells. Boydenchamber assays also showed that TAMs or M-CSF–pretreatedMDMs significantly increased the number of migrated and invad-ed phyllodes tumor cells (Fig. 2F and G; P < 0.01). These datasuggest that TAMs not only induces the myofibroblast differen-tiation, but also promotes their malignancy.

CCL18 is responsible for TAM-induced myofibroblastdifferentiation, proliferation, and invasion

Previously, we have reported that CCL18 released by TAMspromotes breast cancermetastasis, causing poor survival of breastcancer patients (22). It has been shownabove that elevatedCCL18expression is also associated with poor outcome of breast phyl-lodes tumor patients, and that myofibroblast differentiation is

Nie et al.





Figure 2.

TAMs induce the myofibroblast differentiation and promote the growth and invasion of phyllodes tumor cells. A, The RNA level of a-SMA in benign phyllodestumor cells (5� 105/well) increased after the cellswere coculturedwithM-CSF (50 ng/mL), pretreatedMDMs (5� 105/well), or primary TAMs (5� 105/well) for 3 days.Peripheral blood monocytes (PBM) from healthy donors were isolated and then induced to M2 phonotype by M-CSF for 6 days. Eight cases of benign phyllodestumor cells were isolated from eight benign phyllodes tumor tissues. TAMs were isolated from human malignant phyllodes tumor tissue. Bars correspond tomeans � SD of eight samples, with experimental triplicates for each sample. B, Western blot analysis shows the protein level of a-SMA in benign phyllodestumor cells isolated and treated as inA.C, Immunofluorescent staining ofa-SMAandFAP in benign phyllodes tumor cells isolated and treated as inA. Scale bar, 20mm.D, Collagen gel contraction was measured in phyllodes tumor cells isolated and treated as in A. Bars correspond to means� SD of eight samples, with experimentaltriplicates for each sample. E, Colony formation assays in phyllodes tumor cells isolated and treated as in A. Bars correspond to means � SD of eight samples, withexperimental triplicates for each sample. F andG, Migration and invasion assays of phyllodes tumor cells isolated and treated as inA. Scale bar, 50 mm. Bars correspondto means � SD of experimental triplicates from one case of primary cells (�� , P < 0.01 as compared with the untreated benign phyllodes tumor cells).






Figure 3.

TAMs induce myofibroblast differentiation and promote the growth and invasion of phyllodes tumor cells via CCL18. A, The RNA level of a-SMA in phyllodestumor cells. The primary benign phyllodes tumor cells were isolated from benign phyllodes tumor (5 � 105/well) and were cultured alone, or treated with20 ng/mL CCL18 for 24 hours, or cocultured with primary TAMs (5 � 105/well) for 3 days. Primary TAMs were isolated from malignant phyllodes tumor. Indicatedcontrol IgG (10 mg/mL) or anti-CCL18–neutralizing antibody (10 mg/mL) was added into the experiment simultaneously as the cells were cocultured withprimary TAMs. The primary malignant phyllodes tumor cells (5� 105/well) were used as the positive control. B, The protein level of a-SMA in malignant and benignphyllodes tumor cells isolated and treated as in A. C, Immunofluorescent staining of a-SMA and FAP in malignant and benign phyllodes tumor cells isolated andtreated as in A. Scale bar, 20 mm. D, Representative images of collagen gel contraction and colony formation assay for benign and malignant phyllodes tumor cellsisolated and treated as in A. E, Collagen gel contraction and colony formation was measured in phyllodes tumor cells isolated and treated as in A. F, Representativeimages of Boyden chamber assay for migrated and invaded phyllodes tumor cells isolated and treated as in A. Scale bar, 50 mm. Bars correspond to means� SD ofthree samples with experimental triplicates.

Nie et al.





Figure 4.

CCL18 upregulates miR-21 expression via activating NF-kB. A, Percentage of miR-21þ cells is correlated with the percentage of CCL18þ cells in breast phyllodestumor samples. A total of 268 phyllodes tumor samples, including 167 benign, 36 borderline, and 65 malignant phyllodes tumor were used. Percentage ofmiR-21þ cells in each sample was detected by in situ hybridization. Percentage of CCL18þ cells in each sample is determined by IHC. B, Real-time qRT-PCR revealedthat the expression of miR-21 was upregulated by CCL18 treatment in phyllodes tumor cells. The primary phyllodes tumor cells were isolated from benign phyllodestumor and then treatedwith CCL18 for 24 hours. ThemiR-21 levels are indicated as RNA expression normalized toU6. Bars correspond tomean� SDof eight samples,with three experimental triplicates in each group (�� , P < 0.01 as compared with PBS group). C, NF-kB activity increases upon the treatment ofCCL18 and TNFa. NF-kB activity was exhibited by luciferase reporter assay (mean � SD, n ¼ 8; �� , P < 0.01; �� , P < 0.001 vs. PBS). D, Real-time qRT-PCRshows the expression of NF-kB target genes are upregulated by CCL18. Cells were isolated and treated as described in B. Bars correspond to mean � SD of threesamples with three experimental triplicates in each group (��, P < 0.01, �� , P < 0.001 vs. PBS). E, Real-time qRT-PCR reveals that CCL18-induced miR-21expression can be reversed byNF-kB inhibitor. Primary cells isolated from benign phyllodes tumorwere pretreatedwith DMSO, BAY-117082, or JSH-23 for 1 hour andthen CCL18 for 24 hours before harvest. Bars correspond to means � SD of three samples, with three experimental triplicates in each group (�� , P < 0.001compared with cells treated with PBS; ##, P < 0.01 compared with cells treated with CCL18). F, CCL18 induces p65 to bind to miR-21 promoter. Primary benignphyllodes tumor cells were treated with PBS or CCL18 for 30 minutes. The ChIP assay was performed by using anti-p65 antibody to pull down p65 bondDNA fragments. RT-PCR was used to determine amount of the precipitated DNA. Bar represents mean � SD of experimental triplicates (�� , P < 0.001 vs. PBS).G, CCL18 promotes miR21 transcription activity. Luciferase reporter assays were performed in benign phyllodes tumor cells transfected with reporter plasmidscontaining miR-21 promoter and treated with PBS or CCL18 for 24 hours (bar represents mean � SD of experimental triplicates; P < 0.001 vs. PBS).






Figure 5.

CCL18 upregulates miR-21 expression and thus induces myofibroblast differentiation via activating NF-kB. A and B, a-SMA in benign phyllodes tumor cellscan be induced by CCL18 and reversed by NF-kB inhibitor or miR21 ASO. Benign phyllodes tumor cells were isolated as in Fig. 3A and then pretreated withDMSO, BAY-117082, or JSH-23 for 1 hour or transfected with lin4 ASO (NC), Lipofectamine alone (mock), or miR-21 ASO for 24 hours before treatment with20 ng/mL CCL18 for 24 hours. (Continued on the following page.)

Nie et al.





progressively increased during the malignant progression ofphyllodes tumor in breast. Therefore, we examined whetherCCL18 induced the myofibroblast differentiation. We found thatbreast phyllodes tumor cells treated with 20 ng/mL CCL18 had asignificant increase in the mRNA and the protein levels of a-SMA(Fig. 3A–C) and enhanced activity of collagen gel contraction,proliferation (Fig. 3DandE),migration, and invasiveness (Fig. 3F;Supplementary Fig. S2A and S2B), suggesting that CCL18 inducedmyofibroblast differentiation andpromoted the proliferation andinvasion of the phyllodes tumor cells. Consistent with the resultsdescribed above,when coculturedwith primary TAMs and controlIgG, the phenotype and function changes of phyllodes tumor cellswere similar to CCL18 treated alone. However, anti-CCL18–neutralizing antibody inhibited myofibroblast differentiation,proliferation, and invasion induced by TAMs (Fig. 3A–F), indi-cating that the TAM-released CCL18 is responsible for the myofi-broblast differentiation of phyllodes tumor cells.

CCL18 upregulates miR-21 expression by activating NF-kBOur previous study has shown thatmiR-21 expression levels are

upregulated in the malignant phyllodes tumors. miR-21 not onlyinduces themyofibroblast differentiation of phyllodes tumor cells,but also promotes their malignant properties, including prolifer-ation and invasion (4). Thus, we speculate that CCL18 mayregulate the myofibroblast differentiation through upregulatingmiR-21. To further demonstrate this hypothesis, we first analyzedthe expression correlation between miR-21 (detected by ISH as inref. 4) and CCL18, and revealed that the percentage of miR-21þ

cells was positively correlated with that of CCL18-producing cellsin the268phyllodes tumor samples (Fig. 4A, r¼ 0.752,P < 0.001).Thenwe used qRT-PCR to test themiR-21 level of phyllodes tumorcells in response to CCL18 treatment. Compared with the untreat-ed group, the RNA level of miR-21 of CCL18-treated phyllodestumor cells increased by 3.9-fold (Fig. 4B).

A recent study showed NF-kB could bind to the promoterof miR-21 and regulate pancreatic b cell death through miR-21(23, 24). As we have shown that CCL18 activated NF-kB pathwayto induce epithelial–mesenchymal transition in breast cancer (25),we proposed that CCL18 might upregulate miR-21 expression viaactivating NF-kB. To clarify the roles of NF-kB in CCL18-inducedmiR-21 expression, the luciferase reporter assay was applied inphyllodes tumor cells. We found that NF-kB activity was escalatedto a 6.2-fold increase in CCL18-treated cells. The TNFa-treatedphyllodes tumor cells were used as positive control (Fig. 4C).Constantly, we revealed that CCL18-induced p65 nuclear translo-cation could be reversed by both NF-kB pathway inhibitors andsiPITPNM3, suggesting CCL18 induced activation of NF-kB path-way through membrane-associated phosphatidylinositol transferprotein 3 (PITPNM3; Supplementary Fig. S3A and S3B). ThePITPNM3 RNAi efficiency was detected by Western blot analysis(Supplementary Fig. S3C). Again, we confirmed that CCL18 pro-

motedNF-kBpathwayviaPITPNM3,as siPITPNM3eliminated thephosphorylation of IKK and IKB in the presence of CCL18 (Sup-plementary Fig. S3D). In addition, CCL18 upregulated the mRNAlevels of several NF-kB target genes, including IL8, IL4, Twist, etc.(Fig. 4D). Collectively, these data suggest that CCL18 can enhanceNF-kB transcriptional activity in phyllodes tumor cells.

To investigate whether NF-kB can directly regulate miR-21expression, inhibitors of NF-kB pathway [BAY-117082, an IKKinhibitor that also interferes with the ubiquitin conjugatingenzymes (26), and JSH-23, an inhibitor of NF-kB nuclear trans-location] were used. Both BAY-117082 and JSH-23 reversedCCL18-induced miR-21 upregulation in phyllodes tumor cells(Fig. 4E), indicating that NF-kB activation is involved in CCL18-mediated upregulation of miR-21. A ChIP-qPCR assay with anti-bodies against NF-kB p65was used to determine whethermiR-21is the direct target of NF-kB p65. CCL18 increased the binding ofNF-kB p65 to the promoters of miR-21 by 12.7-fold (Fig. 4F). Wenext examined the functional relationships among CCL18, NF-kB, and miR-21 in breast phyllodes tumor cells. When phyllodestumor cells were treated byCCL18 for 24 hours,miR-21 promoteractivity was increased by 8.6-fold (Fig. 4G). Taken together, ourresults indicate that CCL18 upregulates miR-21 expression viaactivating NF-kB.

CCL18 upregulates miR-21 expression, thus inducingmyofibroblast differentiation via activating NF-kB

To investigate whether CCL18 upregulates miR-21 expres-sion and thus induces myofibroblasts' differentiation via acti-vating NF-kB, phyllodes tumor cells were pretreated with BAY-117082 or JSH-23 for 1 hour or transfected with miR-21antisense oligos (ASO) for 48 hours and then exposed to 20ng/mL CCL18 or cocultured with TAMs for 48 hours. We foundthat CCL18 treatment increased the mRNA (Fig. 5A and C) andprotein level of a-SMA (Fig. 5B and D). However, the NF-kBinhibitor BAY-117082 or JSH-23 or the miR-21 inhibitorblocked the CCL18 or TAM-induced increase of a-SMA in thephyllodes tumor cells, indicating that myofibroblast differen-tiation was prevented under these conditions. Similarly, colla-gen contraction assay showed that CCL18 contracted collagengels to a much greater extent than the untreated cells (P < 0.01),whereas BAY, JSH, or miR-21 ASO suppressed the contractileability (P < 0.01; Fig. 5E and Supplementary Fig. S4A). We thenevaluated the effect of NF-kB and miR-21 on the invasive abilityof phyllodes tumor cells treated with CCL18 or cocultured withTAMs. Consistent with previous results, BAY, JSH, and miR-21ASO also abrogated the CCL18 or TAM-promoted migration,invasion (Fig. 5E; Supplementary Fig. S4B) and proliferation(Fig. 5F; Supplementary Fig. S4C) of phyllodes tumor cells.These data suggest that CCL18 upregulates miR-21 expressionand thus induces myofibroblast differentiation and promotestheir tumorigenicity via activating NF-kB.

(Continued.) The RNA and protein level of a-SMA in benign phyllodes tumor cells were determined by RT-PCR (A) or Western blot analysis (B). Barscorrespond tomeans� SDof experimental triplicates (�� ,P<0.001 vs. untreated phyllodes tumor cells; ###,P <0.001 vs. without pretreatedphyllodes tumor cells).C and D, a-SMA in benign phyllodes tumor cells can be induced by coculture with TAMs and reversed by NF-kB inhibitor or miR-21 ASO. Benign phyllodestumor cells were isolated as in Fig. 3A and then pretreated with DMSO, BAY-117082, or JSH-23 for 1 hour or transfected with lin4 ASO (NC), lipofectamine alone(mock), or miR-21 ASO for 24 hours before coculture with TAMs (5 � 105/well) for 3 days. The RNA and protein level of a-SMA in benign phyllodes tumorcells were determined by RT-PCR (C) or Western blot analysis (D). Bars correspond to means � SD of experimental triplicates (###, P < 0.001 as compared withuntreated phyllodes tumor cells). E, Representative images of collagen gel contraction and Boyden chamber assay for benign phyllodes tumor cells treated as in Aand C. CCL18 or coculture with TAMs promotes the indicated ability of phyllodes tumor cells, whereas NF-kB inhibitor or miR-21 ASO can reverse theenhanced malignant behaviors. Scale bar, 50 mm. F, Colony numbers of benign phyllodes tumor cells treated as in A (left) and C (right).






CCL18 promotes AKT activation in myofibroblasts throughNF-kB/miR-21/PTEN axis

Our previous work has revealed that mRNA of PTEN can betargeted by miR-21 in phyllodes tumor cells (4). As PTEN is themajor negative regulator of PI3K–AKT pathway, which is mostfrequently mutated and inactivated in malignant tumors, wehypothesize CCL18 drives tumorigenicity of phyllodes tumorthrough activating PI3K–AKT pathway. Therefore, the phosphor-ylation of AKT was examined to explore whether this pathway isactivated. We found that CCL18 treatment or TAM coculturedecreased PTEN level and increased AKT phosphorylation fol-lowed by the elevated phosphorylated p65 (Fig. 6A), suggestingthe activation of PI3K–AKT pathway. When CCL18 antibody wasadded, the PI3K–AKTpathway kept inactive. Consistently, theNF-kB inhibitor (BAY-117082 or JSH-23) or the miR-21 ASO couldcompletely reverse the CCL18-enhanced AKT phosphorylationand increase PTEN level in phyllodes tumor cells (Fig. 6B). Theseresults suggested that the CCL18/NF-kB/miR-21 axis was neces-sary and sufficient for TAMs to activate PI3K–AKT pathway inphyllodes tumor cells. Then, we detected PTEN and p-AKT inhumanphyllodes tumor tumors by IHCand confirmed that PTENlevel decreased and p-AKT increased from benign to malignantphyllodes tumors (Fig. 6C). Together, these data suggested thatTAMs induced myofibroblast differentiation and promoted phyl-lodes tumor tumorigenicity through secreting CCL18 to activatePI3K–AKT pathway via NF-kB/miR-21/PTEN axis (Fig. 6D). Wealso explored which factor in the microenvironment promotedmacrophages producing CCL18. We compared the cytokineexpressed by the phyllodes tumor cells and found that hepatocytegrowth factor (HGF) was highly secreted by malignant phyllodestumor cells comparing with benign phyllodes tumor cells (Sup-plementary Fig. S5A and S5B). Then a HGF-neutralized antibodywas used to block the phyllodes tumor cell–secreted HGF in atranswell culture system with malignant phyllodes tumor cellsand corresponding TAMs. We found that TAMs produced lessCCL18 upon HGF-neutralized antibody treatment (Supplemen-tary Fig. S5C).

M2 macrophage–secreted CCL18 accelerates tumor growth,induces myofibroblast differentiation, and promotesmetastasis of breast phyllodes tumor xenografts

To investigate the role of CCL18 on tumor formation andprogression in vivo, we inoculated benign phyllodes tumor cellsandMDMs into themammary fat pads of athymic nudemice (n¼10 per group), and evaluated tumor metastasis to the lungs andlivers. When the xenografts became palpable, CCL18-specificneutralizing antibody (CCL18-Ab, 1 mg/kg twice weekly) orisotype IgG (1 mg/kg twice weekly) were injected via the tail veintwice weekly. Intratumoral injection of M-CSF (0.2 mg/kg twiceweekly) was performed twice weekly after inoculating and 48hours prior to antibody injection. We found that compared withMDMs inoculated alone, additional injection of M-CSF signifi-cantly increased the tumor formation efficiency (P < 0.01, Fig. 7A)and accelerated tumor growth of xenografts (P < 0.01, Fig. 7B).CCL18-specific neutralizing antibody markedly inhibited thetumor formation and lowered tumor growth of xenografts (P <0.01, Fig. 7A and B). These data suggest that CCL18 plays animportant role in the malignant transformation of breast phyl-lodes tumors.

To further evaluate whether CCL18 regulates myofibroblastdifferentiation in vivo, we examined the protein levels of a-SMA,

PTEN, p-Akt, and Ki67 in the xenografts using IHC (Fig. 7C).Similar to the results obtained in vitro, injection of M-CSFenhanced the levels of a-SMA, p-Akt, and Ki67 in the xenografts,but decreased the expression of PTEN, whereas CCL18-specificneutralizing antibody attenuated the upregulation of a-SMA, p-Akt, or Ki67 in the xenografts, but increased the expression ofPTEN (Fig. 7C). CD163þ cells were significantly increased in M-CSF–treatedMDMs group, indicating that M-CSF inducedMDMstoM2macrophages in the xenografts. As increasedmigration andinvasion of tumor cells are linked with metastasis, we evaluatedwhether CCL18 promoted the metastasis of breast phyllodestumor xenografts. Consistent with the findings in vitro, injectionof M-CSF into the fat pads of the athymic nude mice activatedMDMs to M2 macrophages, induced myofibroblast differenti-ation, promoted migration and invasion of breast phyllodestumor, and significantly increased the number of metastaticnodules and human hypoxanthine-guanine-phos-phoribosyl-transferase (HPRT) mRNA in the livers (Fig. 7D, E, and G).However, there was no significant lung metastasis observed inthe mice according to histologic examination (data not shown),wet weight of lung (Fig. 7F), and human HPRT mRNA level(Fig. 7G). In addition to the M2 polarization of macrophages inthe tumors, we also found that the peritoneal macrophage ofthe M-CSF–treated mice, especially of those from the groupwith the fastest tumor growth, showed M2 polarization (Sup-plementary Fig. S6).

DiscussionMyofibroblasts are activated mesenchymal stromal cells that

are differentiated from fibroblasts upon tissue injury. They con-tribute to tissue repair during wound healing by their strongcontractility and extracellular matrix (ECM) secretory function.At the end stage of tissue repairing, myofibroblasts disappear bymassive apoptosis (27). However, the excessive contraction andECM protein secretion by myofibroblasts severely impair tissuerepairing, such as in hypertrophic scars, scleroderma, andDupuyt-ren's disease (28). In addition, continuous stimulation by toxic,infectious, andmetabolic agents or chronic inflammation leads tocontinuous myofibroblast differentiation and induces fibrosis inliver, kidney, and heart (29, 30). Myofibroblasts were also foundto promote invasion and metastasis of breast, pancreatic, andcolorectal cancer (31). In phyllodes tumors, myofibroblastsderive from stromal fibroblasts and constitute the major malig-nant component of tumor mass. Transition of the mesenchymalfibroblasts to myofibroblasts (FMT) has been suggested as a keyprocess in the tumorigenesis of phyllodes tumors (4). TGFb1was shown to be able to induce FMT in fibrosis and thus playedan important role in cancer progression (32, 33). In this study,we found that the infiltration of TAMs increased dramaticallyalong with the transformation from benign to malignant phyl-lodes tumors. TAMs in breast cancer are skewed to M2 pheno-type by secreting high levels of IL10, CCL18, and CCL22 (25),and our previous studies have shown that the chemokineCCL18 released by TAMs induces epithelial–mesenchymaltransition and promotes metastasis of breast cancer (12). Herewe went one step forward by showing that CCL18 from theTAMs of phyllodes tumors is also a key regulator of FMT inmyofibroblast differentiation. Blocking CCL18 by anti-CCL18–neutralizing antibody dramatically reverses the myofibroblastdifferentiation and thus inhibits phyllodes tumor growth


Nie et al.




in vivo. According to our results, the FMT in phyllodes tumors isdriven by CCL18 from TAMs, suggesting an essential role ofTAMs in the process of FMT of phyllodes tumor malignantdevelopment.

It has been well documented that TAMs secrete CCL18 topromote cancer progression and metastasis, such as breast cancer(34), pancreatic cancer (35), and lung cancer (36), by activatingPI3K/AKT (37), Lin 28 (38), andNir1-ELMO1/DOC180 signaling

Figure 6.

CCL18 promotes AKT activation through NF-kB/miR-21/PTEN axis. A, PTEN is decreased and p-AKT is elevated at protein level upon CCL18 treatment.Benign phyllodes tumor cells were cultured alone, or treatedwith 20 ng/mL CCL18 for 24 hours, whereas control IgG (10 mg/mL) or anti-CCL18–neutralizing antibody(10 mg/mL) was introduced into the experiments simultaneously as indicated. Isolated primary malignant phyllodes tumor cells were used as positive control.B,Western blotting reveals that CCL18-induced changes of PTEN and p-AKT levels can be reversed by NF-kB inhibitor or miR-21 ASO. Benign phyllodes tumor cellswere pretreated with BAY-117082 or JSH-23 for 1 hour or transfected with lin4 ASO (NC), Lipofectamine alone (mock), or miR-21 ASO for 24 hours and then treatedwith 20 ng/mL CCL18 for 24 hours. C, IHC staining in paraffin-embedded normal breast tissue and breast phyllodes tumor reveals that PTEN is downregulatedand p-AKT is unregulated along with increased malignancy in phyllodes tumor. Scale bar, 20 mm. D, The working model of how macrophages releaseCCL18 to promote phyllodes tumor tumorigenesis via NF-kB/miR-21/PTEN/AKT axis.






Figure 7.

CCL18 accelerate tumor growth, induce myofibroblast differentiation and promote metastasis of breast phyllodes tumor xenografts. A, Xenografts of indicatedgroups in nudemice, harvested 60days postinjection. Benignphyllodes tumor cells (B) andMDMswere injected into themammary fat pads of athymic nudemice. Micewere injectedwith control IgG (IgG; 1mg/kg twiceweekly), or anti-CCL18–neutralizingantibody (CCL18-Ab, 1mg/kg twiceweekly), via the tail vein as indicated after thexenografts became palpable. Intratumoral injection of M-CSF (0.2 mg/kg twice weekly) was performed twice weekly after inoculation and 48 hours prior to antibodyinjection (n¼ 10 per group). When cocultured with M-CSF–activated MDMs, the tumors from benign phyllodes tumor grow much faster and show a more aggressivephenotype than the other group. The CCL18-neutralizing antibody can prevent the activated MDMs enhanced aggressivemanner of benign phyllodes tumor.B, Tumorsizeduring thecourseofeach indicated treatment.Errorbars show�SEM(n¼ 10pergroup; �� ,P<0.01 vs. benignphyllodes tumorcells inoculatedalone;##,P<0.01 vs.control IgG group, with M-CSF treated). C, Representative images of H&E staining for xenografts and IHC staining for a-SMA, PTEN, p-Akt, Ki67, and CD163 asdescribed above. Scale bar, 50 mm. D, Representative images of H&E staining for xenografts as described above, showing the presence or absence of liver metastasis.Scale bar, 50 mm. E, The number of metastatic liver nodules inmice xenografts as indicated. The bars correspond to themean� SD (�� , P < 0.001 vs. benign phyllodestumor cells inoculated alone; ##, P < 0.01 vs. control IgG group, with M-CSF treated). F,Wet lung weights in tumor-bearing mice. Bars correspond to mean� SD. G,Expression of human HPRTmRNA relative to mouse 18S rRNA in the lungs and livers of mouse xenografts as described above (�� , P < 0.01 vs. benign phyllodes tumorcells inoculated alone; ##, P < 0.01 ves control IgG group, with M-CSF treated).

Nie et al.





pathways (39, 40). Our previous work revealed that CCL18 couldserve as an NF-kB activator to promote breast cancer progression(25). In this study, we showed that TAMs induced myofibroblastdifferentiation and promoted themalignant progression of breastphyllodes tumors via secreting CCL18. Furthermore, CCL18 levelcan serve as an independent prognostic factor of phyllodestumors. The NF-kB pathway has been identified as a key regulatorof fibroblast function and matrix remodeling (41). Activated NF-kB pathway is involved in maintaining myofibroblast phonotype(42, 43). However, whether NF-kB pathway is involved in theFMT process and promotes tumor progression, especially inphyllodes tumors, remains obscure. In this study, we haverevealed that the NF-kB pathway was activated by CCL18 inmyofibroblast. Subsequently, the nuclear translocated p65 bindsto the promoter of miR-21 gene and increases its transcription inthe myofibroblast. We have previously illustrated that miR-21induces myofibroblast differentiation and promotes the malig-nant progression of phyllodes tumors (4). Thus, miR-21 is a keydownstream mediator of p65 in TAM-induced FMT of myofibro-blast differentiation.

It is well established thatmiR-21 is involved in cellular survival,invasiveness, and apoptosis by suppressing specific target genes asPTEN, PDCD4, RECK, TIMP3, Smad3, etc (44–46). Our previousdata suggested that Smad7 and PTEN are targets of miR-21 inphyllodes tumors (4). PI3K–AKT pathway is most frequentlyactivated in breast cancer. A case report based on whole genomesequencing revealed N-RAS mutation with concomitant activa-tion of PI3K/Akt/mTOR in phyllodes tumor (47), suggesting thePI3K–AKT pathwaymay contribute tomyofibroblast transition inphyllodes tumors. Indeed, CCL18/NF-kB/miR-21 axis betweenTAMs and phyllodes tumor cells decreases PTEN level and furtheractivates AKT. Therefore, an intercellular communication, whereTAMs secreted CCL18 to activate the NF-kB/miR-21/PTEN–AKTpathway inmyofibroblasts, is established (Fig. 6D). Our previousdata and this study have demonstrated that CCL18 and miR-21playmajor roles in myofibroblast transition of phyllodes tumors,suggesting that the above intercellular communication pathwayfrom TAMs to myofibroblasts may serve as a major driver oftumorigenicity in phyllodes tumors.

Even with surgical resection, local recurrence rate of phyllodestumors is still as high as 8% to 36% (48). Moreover, recurrentphyllodes tumors are frequently more aggressive in phenotype(49). Our findings demonstrate that CCL18 drives phyllodestumor progression by inducing FMT of the stromal cells. HigherCCL18 level dramatically correlates with local or distal recurrenceof the malignancy and is a better prediction factor for recurrencecomparedwithother clinical andpathologicmarkers of phyllodestumors (Supplementary Tables S1 and S2). Given that chemo-therapy or radiotherapy is not effective against phyllodes tumors(2), there is a pressing requirement to develop new therapeutic

strategy. In this study, our in vivo findings show that blockingCCL18 with neutralizing antibody effectively shrinks phyllodestumors in mouse xenograft models, which suggests that antago-nizing the CCL18 signalingmay emerge as a promising strategy totreat phyllodes tumors. Together, our data suggest that the inter-cellular communication between TAMs and myofibroblasts viaCCL18/NF-kB/miR-21/PTEN/AKT axis plays a central role in thetumorigenesis of phyllodes tumors. Monitoring CCL18 level andtargeting this pathway raise the possibility of precision diagnosisand treatment for breast phyllodes tumors.

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

Authors' ContributionsConception and design: E. SongDevelopment of methodology: Y. Nie, J. ChenAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y. Nie, J. Chen, D. Huang, Y. Yao, J. Zeng, X. Chao,H. YaoAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Y. Nie, Y. Yao, X. ChaoWriting, review, and/or revision of the manuscript: Y. Nie, S. Su, F. Su, H. Hu,E. SongAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): Y. Nie, Y. Yao, J. Chen, L. Ding, E. SongStudy supervision: H. Hu, E. Song

Grant SupportE.W. Song is supported by grants from the National Key R&D Program

(2016YFC1302301), Natural Science Foundation of China (81490750,81230060, 81442009), Science Foundation of Guangdong Province(S2012030006287), Guangzhou Science Technology and Innovation Commis-sion (201508020008, 201508020249), Guangdong Science and TechnologyDepartment (2015B050501004), Translational Medicine Public Platform ofGuangdong Province (4202037), and Guangdong Department of Science &Technology Translational Medicine Center grant (2011A080300002), grantKLB09001 from the Key Laboratory of Malignant Tumor Gene Regulationand Target Therapy of Guangdong Higher Education Institutes, Sun Yat-SenUniversity and grant [2013]163 from the Key Laboratory of Malignant TumorMolecular Mechanism and Translational Medicine of Guangzhou Bureau ofScience and Information Technology. H. Hu is supported by Natural ScienceFoundation of China (81672738). Y. Nie is supported by the Natural ScienceFoundation of China (81502301). H.R. Yao is supported by theNatural ScienceFoundation of China (81372819, 81572596, U1601223), Specialized ResearchFund for the Doctoral Program of Higher Education (20120171110075),and received funding from Guangzhou Science and Technology Bureau(2014J4100170).

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 October 5, 2016; revised March 1, 2017; accepted May 10, 2017;published OnlineFirst May 16, 2017.

References1. Kucuk U, Bayol U, Pala EE, Cumurcu S. Importance of P53, Ki-67 expres-

sion in the differential diagnosis of benign/malignant phyllodes tumors ofthe breast. Indian J Pathol Microbiol 2013;56:129–34.

2. Chaney AW, Pollack A, McNeese MD, Zagars GK, Pisters PW, Pollock RE,et al. Primary treatment of cystosarcoma phyllodes of the breast. Cancer2000;89:1502–11.

3. Parker SJ, Harries SA. Phyllodes tumours. PostgradMed J 2001;77:428–35.4. Gong C, Nie Y, Qu S, Liao JY, Cui X, Yao H, et al. miR-21

induces myofibroblast differentiation and promotes the malignant

progression of breast phyllodes tumors. Cancer Res 2014;74:4341–52.

5. Allavena P, Sica A, Solinas G, Porta C, Mantovani A. The inflammatorymicro-environment in tumor progression: the role of tumor-associatedmacrophages. Crit Rev Oncol Hematol 2008;66:1–9.

6. Condeelis J, Pollard JW. Macrophages: obligate partners for tumor cellmigration, invasion, and metastasis. Cell 2006;124:263–6.

7. Pollard JW.Tumour-educated macrophages promote tumour progressionand metastasis. Nat Rev Cancer 2004;4:71–8.






8. Mantovani A, Marchesi F, Porta C, Sica A, Allavena P. Inflammation andcancer: breast cancer as a prototype. Breast 2007;16:S27–33.

9. Thompson RW, Pesce JT, Ramalingam T,WilsonMS,White S, Cheever AW,et al. Cationic amino acid transporter-2 regulates immunity bymodulatingarginase activity. PLoS Pathog 2008;4:e1000023.

10. Wynn TA, Barron L. Macrophages: master regulators of inflammation andfibrosis. Semin Liver Dis 2010;30:245–57.

11. Wynes MW, Riches DW. Induction of macrophage insulin-like growthfactor-I expression by the Th2 cytokines IL-4 and IL-13. J Immunol2003;171:3550–9.

12. Chen J, Yao Y,GongC, Yu F, Su S, Liu B, et al. CCL18 from tumor-associatedmacrophages promotes breast cancer metastasis via PITPNM3. Cancer Cell2011;19:541–55.

13. Dangaj D, Abbott KL, Mookerjee A, Zhao A, Kirby PS, Sandaltzopoulos R,et al. Mannose receptor (MR) engagement by mesothelin GPI anchorpolarizes tumor-associated macrophages and is blocked by anti-MRhuman recombinant antibody. PLoS One 2011;6:e28386.

14. Islam SA, Ling MF, Leung J, Shreffler WG, Luster AD. Identification ofhuman CCR8 as a CCL18 receptor. J Exp Med 2013;210:1889–98.

15. Ferenbach D, Hughes J. Macrophages and dendritic cells: what is thedifference? Kidney Int 2008;74:5–7.

16. Friedman SL.Hepatic stellate cells: protean,multifunctional, and enigmaticcells of the liver. Physiol Rev 2008;88:125–72.

17. DeNardoDG, Barreto JB, Andreu P, Vasquez L, TawfikD, Kolhatkar N, et al.CD4(þ) T cells regulate pulmonarymetastasis ofmammary carcinomas byenhancing protumor properties of macrophages. Cancer Cell 2009;16:91–102.

18. Verreck FA, de Boer T, Langenberg DM, Hoeve MA, Kramer M, Vaisberg E,et al. Human IL-23-producing type 1 macrophages promote but IL-10-producing type 2 macrophages subvert immunity to (myco)bacteria. ProcNatl Acad Sci U S A 2004;101:4560–5.

19. Liu G, Friggeri A, Yang Y, Milosevic J, Ding Q, Thannickal VJ, et al. miR-21mediates fibrogenic activation of pulmonary fibroblasts and lung fibrosis.J Exp Med 2010;207:1589–97.

20. Aranda FI, Laforga JB, Lopez JI. Phyllodes tumor of the breast. An immu-nohistochemical study of 28 cases with special attention to the role ofmyofibroblasts. Pathol Res Pract 1994;190:474–81.

21. KalluriR,ZeisbergM.Fibroblasts in cancer.NatRevCancer2006;6:392–401.22. Lin X, Chen L, Yao Y, Zhao R, Cui X, Chen J, et al. CCL18-mediated down-

regulation of miR98 and miR27b promotes breast cancer metastasis.Oncotarget 2015;6:20485–99.

23. LingM, Li Y, Xu Y, Pang Y, Shen L, Jiang R, et al. Regulation ofmiRNA-21 byreactive oxygen species-activated ERK/NF-kB in arsenite-induced cell trans-formation. Free Radic Biol Med 2012;52:1508–18.

24. Ruan Q, Wang T, Kameswaran V, Wei Q, Johnson DS, Matschinsky F, et al.The microRNA-21-PDCD4 axis prevents type 1 diabetes by blockingpancreatic b cell death. Proc Natl Acad Sci U S A 2011;108:12030–5.

25. Su S, Liu Q, Chen J, Chen F, He C, Huang D, et al. A positive feedback loopbetween mesenchymal-like cancer cells and macrophages is essential tobreast cancer metastasis. Cancer Cell 2014;25:605–20.

26. Strickson S, Campbell DG, Emmerich CH, Knebel A, Plater L, Ritorto MS,et al. The anti-inflammatory drug BAY 11–7082 suppresses the MyD88-dependent signalling network by targeting the ubiquitin system. Biochem J2013;451:427–37.

27. Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA.Myofibroblastsand mechano-regulation of connective tissue remodelling. Nat Rev MolCell Biol 2002;3:349–63.

28. Hinz B.Formation and function of the myofibroblast during tissue repair.J Invest Dermatol 2007;127:526–37.

29. Desmouliere A, Darby IA, Gabbiani G. Normal and pathologic soft tissueremodeling: role of the myofibroblast, with special emphasis on liver andkidney fibrosis. Lab Invest 2003;83:1689–707.

30. Brown RD, Ambler SK, Mitchell MD, Long CS. The cardiac fibroblast:therapeutic target in myocardial remodeling and failure. Annu Rev Phar-macol Toxicol 2005;45:657–87.

31. De Wever O, Demetter P, Mareel M, Bracke M. Stromal myofibroblasts aredrivers of invasive cancer growth. Int J Cancer 2008;123:2229–38.

32. Hinz B, Phan SH, Thannickal VJ, Galli A, Bochaton-PiallatML, Gabbiani G.The myofibroblast: one function, multiple origins. Am J Pathol 2007;170:1807–16.

33. OtrantoM, SarrazyV, Bonte F,HinzB,GabbianiG,DesmouliereA. The roleof the myofibroblast in tumor stroma remodeling. Cell Adh Migr2012;6:203–19.

34. Lin L, Chen YS, Yao YD, Chen JQ, Chen JN, Huang SY, et al. CCL18 fromtumor-associated macrophages promotes angiogenesis in breast cancer.Oncotarget 2015;6:34758–73.

35. Meng F, Li W, Li C, Gao Z, Guo K, Song S. CCL18 promotes epithelial-mesenchymal transition, invasion and migration of pancreatic cancercells in pancreatic ductal adenocarcinoma. Int J Oncol 2015;46:1109–20.

36. Shi L, Zhang B, Sun X, Zhang X, Lv S, Li H, et al. CC chemokine ligand 18(CCL18) promotes migration and invasion of lung cancer cells by bindingto Nir1 through Nir1-ELMO1/DOC180 signaling pathway. Mol Carcinog2016;55:2051–62.

37. Zhang B, Yin C, Li H, Shi L, Liu N, Sun Y, et al. Nir1 promotes invasion ofbreast cancer cells by binding to chemokine (C-Cmotif) ligand 18 throughthe PI3K/Akt/GSK3b/Snail signalling pathway. Eur J Cancer 2013;49:3900–13.

38. Wang L, Wang YX, Zhang DZ, Fang XJ, Sun PS, Xue HC. Let-7a mimicattenuates CCL18 induced breast cancer cell metastasis through Lin 28pathway. Biomed Pharmacother 2016;78:301–7.

39. Barat S, Bozko P,ChenX, Scholta T,Hanert F, Gotze J, et al. Targeting c-METby LY2801653 for treatment of cholangiocarcinoma. Mol Carcinog2016;55:2037–50.

40. Pivarcsi A, Gombert M, Dieu-NosjeanMC, Lauerma A, Kubitza R, Meller S,et al. CC chemokine ligand 18, an atopic dermatitis-associated anddendritic cell-derived chemokine, is regulated by staphylococcal productsand allergen exposure. J Immunol 2004;173:5810–7.

41. Zhang C, Ferrari R, Beezhold K, Stearns-Reider K, D'Amore A, Haschak M,et al. Arsenic promotes NF-kB-mediated fibroblast dysfunction and matrixremodeling to impair muscle stem cell function. Stem Cells 2016;34:732–42.

42. Pan HW, Cui YH, Zeng JW. NF-kB mediates the survival of cornealmyofibroblast induced by angiotensin II. Invest Ophthalmol Vis Sci2014;55:4220–8.

43. Tian K, Liu Z, Wang J, Xu S, You T, Liu P. Sirtuin-6 inhibits cardiacfibroblasts differentiation into myofibroblasts via inactivation of NF-kBsignaling. Transl Res 2015;165:374–86.

44. Gabriely G, Wurdinger T, Kesari S, Esau CC, Burchard J, Linsley PS, et al.MicroRNA 21 promotes glioma invasion by targeting matrix metallopro-teinase regulators. Mol Cell Biol 2008;28:5369–80.

45. Shi Z, Zhang J, Qian X, Han L, Zhang K, Chen L, et al. AC1MMYR2, aninhibitor of dicer-mediated biogenesis of Oncomir miR-21, reverses epi-thelial-mesenchymal transition and suppresses tumor growth and pro-gression. Cancer Res 2013;73:5519–31.

46. Kim YJ, Hwang SJ, Bae YC, Jung JS. MiR-21 regulates adipogenicdifferentiation through the modulation of TGF-b signaling in mesen-chymal stem cells derived from human adipose tissue. Stem Cells 2009;27:3093–102.

47. Jardim DL, Conley A, Subbiah V. Comprehensive characterization ofmalignant phyllodes tumor by whole genomic and proteomic analysis:biological implications for targeted therapy opportunities. Orphanet J RareDis 2013;8:112.

48. Barth RJ Jr. Histologic features predict local recurrence after breast con-serving therapy of phyllodes tumors. Breast Cancer Res Treat 1999;57:291–5.

49. Shpitz B, Bomstein Y, Sternberg A, Klein E, Tiomkin V, Kaufman A, et al.Immunoreactivity of p53, Ki-67, and c-erbB-2 in phyllodes tumors of thebreast in correlation with clinical and morphologic features. J Surg Oncol2002;79:86–92.


Nie et al.




2017;77:3605-3618. Published OnlineFirst May 16, 2017.Cancer Res Yan Nie, Jianing Chen, Di Huang, et al. Breast Phyllodes Tumors by Inducing Myofibroblast DifferentiationTumor-Associated Macrophages Promote Malignant Progression of

Updated version

10.1158/0008-5472.CAN-16-2709doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2017/05/16/0008-5472.CAN-16-2709.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/77/13/3605.full#ref-list-1

This article cites 49 articles, 12 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/77/13/3605.full#related-urls

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

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/77/13/3605To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-16-2709

http://cancerres.aacrjournals.org/content/suppl/2017/05/16/0008-5472.CAN-16-2709.DC1

http://cancerres.aacrjournals.org/content/77/13/3605.full#ref-list-1

http://cancerres.aacrjournals.org/content/77/13/3605.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/77/13/3605


tumor-associated macrophages promote ... · phyllodes tumor, and ccl18 is an independent prognostic...

Documents