ﬁltrating Macrophages Promote Prostate Tumorigenesis via...

Microenvironment and Immunology

Infiltrating Macrophages Promote Prostate Tumorigenesisvia Modulating Androgen Receptor-Mediated CCL4–STAT3Signaling

Lei-Ya Fang1, Kouji Izumi1, Kuo-Pao Lai1, Liang Liang1, Lei Li1, Hiroshi Miyamoto1, Wen-Jye Lin1,2, andChawnshang Chang1,3

AbstractInfiltrating macrophages are a key component of inflammation during tumorigenesis, but the direct

evidence of such linkage remains unclear. We report here that persistent coculturing of immortalizedprostate epithelial cells with macrophages, without adding any carcinogens, induces prostate tumorigenesisand that induction involves the alteration of signaling of macrophage androgen receptor (AR)-inflammatorychemokine CCL4–STAT3 activation as well as epithelial-to-mesenchymal transition and downregulation ofp53/PTEN tumor suppressors. In vivo studies further showed that PTENþ/� mice lacking macrophage ARdeveloped far fewer prostatic intraepithelial neoplasia (PIN) lesions, supporting an in vivo role formacrophage AR during prostate tumorigenesis. CCL4-neutralizing antibody effectively blocked macro-phage-induced prostate tumorigenic signaling and targeting AR via an AR-degradation enhancer, ASC-J9,reduced CCL4 expression, and xenografted tumor growth in vivo. Importantly, CCL4 upregulation wasassociated with increased Snail expression and downregulation of p53/PTEN in high-grade PIN and prostatecancer. Together, our results identify the AR-CCL4-STAT3 axis as key regulators during prostate tumorinitiation and highlight the important roles of infiltrating macrophages and inflammatory cytokines for theprostate tumorigenesis. Cancer Res; 73(18); 5633–46. �2013 AACR.

IntroductionProstate cancer is the most frequently diagnosed cancer

in men and considered the second leading cause of cancerdeath in the United States (1). The inflammatory microen-vironment has been reported to play an important role intumor initiation, but a direct causal interaction betweenwhich specific-type of inflammatory infiltrates and epithelialcells results in tumor initiation is still not clear (2). Agedprostate lesions have revealed that focal prostate inflam-matory atrophy (PIA) is often associated with infiltratingimmune cells (3) and transition from PIA to high-gradeprostatic intraepithelial neoplasia (HGPIN) or prostate can-cer, may be also linked to suppression of some selective

tumor suppressor genes (4). These results have implicatedthat inflammatory events in PIA may potentially contributeto protein downregulation of tumor suppressor genes inprostate and later lead to the development of HGPIN andprostate cancer. Importantly, one previous study showedthat pharmacologic depletion of macrophages in differentmouse tumor models significantly reduced tumor angiogen-esis and progression, suggesting that macrophages are crit-ical components in the tumor microenvironment for tumorprogression (5). Therefore, research focusing on infiltratingmacrophages and their inflammatory activities for promot-ing early tumor development will lead to discovering newtherapeutic agents to block tumor formation.

Interestingly, our previous study identified a novel role ofandrogen receptor (AR) in a mouse-wound healing model viacontrolling macrophage migration and TNFa production dur-ing the wound healing process (6). This study has providedcompelling evidence that AR could direct the function ofmacrophages and established a new regulation of AR fromthe endocrinal regulation to inflammatory response in thewound healing microenvironment. Given the protumor func-tion ofmacrophages as described above and the important roleof AR in prostate cancer development, further studies areneeded to investigate whether AR would modulate macro-phage function in the process of prostate cancer initiationthrough mediating induction of cytokines/chemokines.

Considering that there are no direct experimental modelsavailable for addressing the impact of macrophages todirectly induce prostate tumorigensis, here we used in vitro

Authors' Affiliations: 1GeorgeWhipple Lab for Cancer Research, Depart-ments of Pathology, Urology, and Radiation Oncology, TheWilmot CancerCenter, University of Rochester Medical Center, Rochester, New York;2Immunology Research Center, National Health Research Institutes, Zhu-nan, Miaoli County, and 3Sex Hormone Research Center, China MedicalUniversity and Hospital, Taichung, Taiwan

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

L.-Y. Fang and K. Izumi contributed equally to this work.

Corresponding Authors: Chawnshang Chang, University of RochesterMedical Center; 601 Elmwood Avenue, Box 626, Rochester, NY 14642.Phone: 585-275-9994; Fax: 585-756-4133; E-mail:[email protected]; and Wen-Jye Lin, Phone: 037-246166 ext.37605; Fax: 886-037-586642; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-12-3228

�2013 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 5633

on July 4, 2018. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 22, 2013; DOI: 10.1158/0008-5472.CAN-12-3228

http://cancerres.aacrjournals.org/

coculture/three-dimensional models that recapitulated aninteraction between immortalized prostate epithelial cells(RWPE-1 cells) and macrophages, and found for the first timethat infiltrating macrophages alone, without adding any car-cinogens, could induce prostate tumorigenesis via a novelpathway involving AR-inflammatory cytokine CCL4–STAT3activation, downregulation of p53/PTEN tumor suppressors,and promotion of epithelial-to-mesenchymal transition (EMT)signaling pathways. These findingsmay help us to develop newpotential therapeutic approaches to battle prostate cancer atthe early PIN development stages.

Materials and MethodsAntibodies and reagents

ASC-J9 (5-hydroxy-1,7-bis(3,4-dimethoxyphenyl)-1,4,6-hep-tatrien-3-1) was a gift from AndroScience (7). Antibody infor-mation is provided in the Supplementary Methods andMaterials.

Cell cultureRWPE-1 and BPH-1 cells (the nonneoplastic, immortalized

human prostatic epithelial cell lines) and THP-1 cells (thehuman acute monocytic leukemia cell line) were obtainedfrom the American Type Culture collection. For other celllines, coculture, and three-dimensional culture experiments,see Supplementary Methods and Materials.

Human cytokine antibody array and ELISAThe conditioned media was collected from 48 hours mono-

cultures of RWPE-1, THP-1, or 48 hours cocultures of RWPE-1/THP-1 cells. The conditioned media collected from monocul-tures or cocultures were used to determine relative amounts ofcytokine levels using Human Cytokine Array Kit (R&D Sys-tems) and for detection of CCL4 using human CCL4 ELISAkits (R&D Systems) according to the manufacturer'sinstructions.

Histology/H&E staining/immunohistochemistryXenograft tumors and prostate tissues were harvested for

histologic examination as described in the SupplementaryMethods andMaterials. Hematoxylin and eosin staining (H&E)and immunohistochemistry (IHC) staining was conducted asdescribed previously (7).

Coculture and three-dimensional culture, cell proliferation/migration assay, colony formation assay,Western blot analysis,quantitative real-time PCR (qRT-PCR), AR silencing in THP-1cells by lentiviral siRNA, orthotopic implantation, ASC-J9treatment, generation of MARKO/PTENþ/� (macrophage ARknockout mice), and human prostate tissue microarray (TMA)analysis were conducted as described in the SupplementaryMethods and Materials.

Statistical analysisThe data values were presented as the mean � SD. P values

were determined by unpaired Student t test. Differences inCCL4 expression in prostate TMA were analyzed by Fisher'sexact test or c2 test. P � 0.05 was considered statisticallysignificant.

ResultsIncreased macrophage infiltration in HGPIN andprostate cancer lesions

To investigate the potential linkage of infiltrating macro-phages in prostate tumorigenesis, we first conducted IHC onhuman prostate TMAs containing benign, HGPIN, and pros-tate cancer lesions using an anti-CD68 antibody. Consistentwith the findings in an early study (8), the number of CD68-positive macrophages was significantly increased in HGPIN(P ¼ 0.0004) or prostate cancer (P < 0.0001) lesions comparedwith that in benign prostate (Fig. 1A). There was no significantdifference in the number of CD68-positive cells betweenHGPIN and prostate cancer (P¼ 0.8518). On the basis of theseresults, we hypothesized that infiltrating macrophages inthe prostate could contribute to the promotion of prostatetumor development.

Coculture of immortalized-human prostate epithelialcells with human macrophage THP-1 cells inducesprostate tumorigenesis

To recapitulate the interaction of macrophages with pros-tate epithelial cells during prostate tumorigenesis, we cocul-tured THP-1 cells with RWPE-1 cells in Transwell plates. Weobserved a 50% increase in cell proliferation at 48 hours (Fig.1B). RWPE-1 cells cultured in the coculturemedium resulted ina 3-fold increase in the migration ability (Fig. 1C). Next, wedeveloped a three-dimensional coculture Matrigel model tomimic an inflamed microenviroment that would allow inter-action between macrophages and prostate epithelial cells, asimmortalized prostate epithelial cells, cultured in purifiedextracellular matrix, can differentiate into well organizedspheroids of glandular prostate epithelial cells, so-called pros-taspheres, acinar-like spheroid structures with lumens (9, 10).We found that RWPE-1 cells alone were able to developprostaspheres after 24 days (Fig. 1D). In contrast, RWPE-1cells cultured with the coculture-conditioned media differen-tiated into a disorganized aggregate structure (Fig. 1D), sug-gesting that soluble factors derived from the coculture maydisrupt the normal differentiation process of RWPE-1 cells viainitiating prostate tumorigenic events (11). Similar resultswerealso obtained when we replaced RWPE-1 cells with anotherhuman epithelial cell-line, BPH-1 (Supplementary Fig. S1A andS1B). Our results support that soluble factors in the coculture-conditioned media may contribute to transformation of non-tumorigenic RWPE-1 cells and disrupt their differentiation inthree-dimensional Matrigel culture.

To confirm that RWPE-1 cells were indeed undergoing atumorigenic process after coculture, RWPE-1 cells (� THP-1cells) were plated on soft-agar to determine their anchorage-independent growth. As expected, nontumorigenic RWPE-1cells were able to form colonies on soft-agar after coculturewith THP-1 cells (Fig. 1E). Next, RWPE-1 and THP-1 cells wereeither monocultured or cocultured for 5 days and then weresubcutaneously injected into the flanks of athymic nude mice.The injection of RWPE-1/THP-1 cells resulted in tumor devel-opment in 7 out of 7 nude mice. In contrast, none of the miceinjected with monocultured RWPE-1 or THP-1 cells developedtumors (Fig. 1F and Supplementary Fig. S2). Taken together,

Fang et al.

Cancer Res; 73(18) September 15, 2013 Cancer Research5634




Figure 1. The coculture of macrophage-prostate epithelial cells induces prostate tumorigenesis. A, increased macrophage infiltration was noted inHGPIN and prostate cancer. IHC staining of human prostate tissue array was conducted using anti-CD68 antibody. CD68-positive cells werecounted and their mean numbers per high-power field are shown at right. Arrowheads indicate CD68-positive macrophages. B, RWPE-1 cells werecocultured with or without THP-1 macrophages for 24 and 48 hours. Cell proliferation assay was conducted using MTT. �, P < 0.05. C, RWPE-1 cellswere seeded in the top chamber of 8 mm pore Transwell plates with conditioned media from coculture of THP-1/RWPE1 cells in the bottom chamber.Cells were incubated for 20 hours. Cells migrating through pores were stained with toluidine blue and counted in 6 random fields. Results areexpressed as the average number of cells per field and are mean � SD. ��, P < 0.01. D, microscopic analysis of acinar morphogenesis andglandular differentiation of RWPE-1 cells in a three-dimensional condition in the presence or absence of the coculture-conditioned media for 8, 15, or24 days. Magnification, �10; �20; �40. E, colony formation in soft agar assay of RWPE-1 cells (alone) and RWPE-1 cells cocultured with THP-1cells. F, left, athymic nude mice that received RWPE-1 cells (alone) or RWPE-1/THP-1 cells. Specific tumor growth of xenografts was visibleafter 10 weeks of subcutaneous injection (indicated by the black arrows). Middle, top, tumors from mice. Top right, H&E staining of paraffin sectionsof xenograft growth after injection of RWPE-1/THP-1 cells. Scale bar, 100 mm. Bottom right, the incidence of tumorigenesis of RWPE-1 andRWPE-1/THP-1 cells in nude mice.

CCL4 Promotes Prostate Tumorigenesis

www.aacrjournals.org Cancer Res; 73(18) September 15, 2013 5635




our findings show, both in vitro and in vivo that induction ofprostate tumorigenesis can be achieved in nontumorigenic-prostate RWPE-1 cells by coculture with THP-1 macrophages.

CCL4 andSTAT3 as potentialmediators formacrophage-mediated prostate tumorigenesis

As an earlier study showed that constitutively-active formsof STAT3 promotes epithelial-to-mesenchymal transition(EMT) and tumorigenesis of RWPE-1 cells (12), we examinedthe STAT3 signaling pathway and found an increase of pSTAT3and its downstream genes (COX-2 and c-Myc) in RWPE-1 cellsafter coculture with THP-1 cells (Fig. 2A). Consistently, we alsoobserved STAT3 activation in an immortalizedmouse prostateepithelial cell-line, mPrE, during coculture with a murine

macrophage cell line, RAW264.7 or bone marrow-derivedmacrophages. (Supplementary Fig. S3A and S3B; refs. 13,14). These data suggest that the crosstalk between macro-phages and prostate epithelial cells was able to enhance STAT3activation in prostate epithelial cells, regardless of whether themacrophages originated from an established cell line ormurine bone marrow. In addition, several EMT-associatedgenes, such as Snail, MMP9, and N-cadherin, were signficantlyincreased in the cocultured RWPE-1 cells (Fig. 2B). Simlilarupregulation of EMT-associated geneswas found in coculturedBPH-1 andmPrE cells (Supplementary Fig. S1C andS3B). Theseresults suggest that STAT3 activation with induction of EMTgenes might play important roles in mediating macrophage-induced prostate tumorigeneis (12, 15).

Figure 2. Coculture with THP-1cells induces variouscytokines anddownstream signaling in RWPE-1cells. A, whole protein extractsisolated from RWPE-1 cells andRWPE-1/THP-1 cells wereanalyzed for the protein levels ofCOX-2, pSTAT3, AR, and c-myc.STAT3 and GAPDH proteins wereused as loading controls. B, qPCRanalysis of EMT-related genes inRWPE-1 cells and RWPE-1/THP-1cells. C, cytokine array analysis ofconditioned media isolated fromRWPE-1 or RWPE-1/THP-1 cellsfor 48 hours, and the expression ofsoluble mediators was determinedby Human Cytokine Array (R&DSystems). D, qPCR analysis ofcytokine expression levels inRWPE-1 cells 48 hours aftercoculture with THP-1 cells. CCL3and CCL4 were more highlyexpressed after coculture. E, theamount of CCL4 was determinedby ELISA in the media of RWPE-1/THP-1 or RWPE-1 cells alone. F–I,neutralization ofCCL4by a specificantibody attenuates THP-1–induced cytokine expression, cellmigration, EMT-related genes, anddownstream signaling mediators.F, qPCR analysis of the expressionof various cytokines in RWPE-1 �THP-1 cells cultured in thepresence of anti-CCL4-neutralizing antibody (aCCL4) orisotype control antibody (IgG). G,RWPE-1� THP-1 cells were platedin a Transwell plate as described inF for cell migration assay. H, qPCRanalysis of Snail, MMP-9, and E-cadherin in RWPE-1 cells � THP-1cells as described in F. I, Westernblot analysis of COX-2 andpSTAT3in RWPE-1 cells that were culturedas described in F. STAT3 andGAPDH served as protein loadingcontrols. All data shown are meanSD. �, P < 0.05, ��, P < 0.01.

Fang et al.





We then appliedWestern blot-based cytokine array analysisto globally identify inflammatory mediators in the coculture-conditioned media. The most abundant cytokines/chemo-kines were CCL4, CCL5, interleukin (IL)-1b, IL-1ra, G-CSF,and IL-8 (Fig. 2C and Supplementary Fig. S4). Interestingly,we found consistent upregulation of CCL4 and CCL5 expres-sion in cocultured BPH-1 cells (Supplementary Fig. S1D). Wefocused on CCL4, as our concurrent study has identifiedCCL4 as an AR downstream gene linked to prostate tumorinitiation (16). The qRT-PCR analysis confirmed that cocul-ture led to the greatest increase in mRNA and proteinexpression in CCL4 (Fig. 2D and E). To validate the role ofCCL4 in mediating EMT and prostate tumorigenesis viaSTAT3 activation, we used a CCL4-neutralizing antibody todetermine whether suppression of CCL4 activity mightinhibit the crosstalk between THP-1 and RWPE-1 cells. Wefound significant downregulation of mRNA expression ofCCL3, CCL5, and IL-6 (Fig. 2F), suggesting that CCL4 induc-tion could possibly be an early and vital event during thecrosstalk. Consistently, suppressing CCL4 activity led to asignficant reduction in THP-1–mediated cell migration and

EMT-related gene induction (Fig. 2G and H), with decreasesin STAT3 activation and COX-2 induction (Fig. 2I). Takentogether (Fig. 2F–I), our results suggest that CCL4 plays anearly and crucial role in macrophage-mediated tumorigenicsignaling.

CCL4 is a critical mediator for suppression of p53and PTEN tumor suppressors

To rule out the possiblity that xenografted tumors ofRWPE-1/THP-1 cells may be derived from THP-1 cells (Fig.1F), we used the alternative approach for long-term culture ofRWPE-1 cells with the coculture conditionedmedia for 60 days,and then implanted these cells into nude mice. Interestingly,we found distinctive morphologic changes of RWPE-1 cells,especially in their mesenchymal shape (spindle-like), com-pared with control cells (Fig. 3A). We also found that thelong-term culture of RWPE-1 cells with the coculture-condi-tioned media resulted in increased colony formation (Fig. 3B).More importantly, following orthotopic injection of long-termcultured RWPE-1 cells into the anterior prostate, 8 out of 12mice developed tumors (Fig. 3C), confirming that RWPE-1 cells

Figure 3. Characterization ofmacrophage-mediated RWPE-1cell transformation after long-termculture with the coculture-conditioned media. A, image ofrepresentative fields of parentaland transformed RWPE-1 cells. B,increased colony formation ofRWPE-1 cells with coculture-conditioned media was observedafter 2 weeks. C, H&E staining andIHC (top left and right) analysis ofcross-sections through theanterior prostate of athymic nudemice after 10 weeks of orthotopicinjection of parental RWPE-1 orlong-term–cultured RWPE-1 cells.Gross observation and histologicanalysis of the anterior prostate ofathymic nudemice after orthotopicinjection of long-term–culturedRWPE-1 cells (middle). Anantibody against E6/E7 was usedas amarker for detectingHPV18E6and E7 oncoproteins in RWPE-1cells (bottom right). D, p53 andPTENprotein expression in RWPE-1 cells with the coculture-conditioned media for 20, 40, and60 days. E, neutralizing CCL4activity by an anti-CCL4 antibodyinhibits THP-1–mediated PTEN/P53 protein downregulation andinduction of EMT-related genes.






can become tumorigenic after long-term culture in the cocul-ture conditioned media. Importantly, IHC staining of E6/E7proteins further revealed that these tumors were originatedfrom injected RWPE-1 cells (Fig. 3C).

To explore the molecular basis for THP-1–mediated tumor-igenesis of RWPE-1 cells, we examined 2 important tumorsuppressor pathways, p53 and PTEN, whose suppression playsessential roles in prostate tumorigenesis (17, 18). Although theexpression of p53 and PTEN at RNA levels was comparable

between parental and long-term cultured RWPE-1 cells after 5days (Supplementary Fig. S5), their protein levels were signif-icantly decreased after 20 days and almost disappeared after 60days in the presence of the coculture-conditioned media (Fig.3D). Next, we further investigated the potential mechanismsunderlying increased CCL4 induction for prostate tumorigen-esis via downregulation of p53/PTEN proteins. With CCL4-neutralizing antibody, we observed a severe impairment inTHP-1–induced downregulation of PTEN and p53 expression

Figure 4. Effects of macrophage ARsilencing on cell proliferation,migration, acinar morphogenesis,and colony formation. A, RWPE-1cells were cocultured with THP-1scramble or ARsi macrophagesand assayed with MTT at 24 and48 hours. B, RWPE-1 cells(1 � 105/well) were incubated withcoculture-conditioned media ofTHP-1 scramble or ARsimacrophages for 20 hours inTranswell plates (8 mm). Migratedcells were stained with toluidineblue and counted in 6 randomfields. Results are expressed as theaverage number of cells per fieldand aremean� SD. C, reduced ARexpression in THP-1 cellsattenuated the inhibitory effects ofTHP-1 cells on the acinarmorphogenesis and glandulardifferentiation of RWPE-1 cells inculture at day 24. Magnification,�10; �20; and �40. Results areexpressed as the mean number ofthe intact acinar cells (right). D,colony formation in soft agar ofRWPE-1 cells that were platedwitheither THP-1 scramble or THP-1ARsi macrophages and thenumbers of colonies shown (right).��, P < 0.01. E, silencing ARexpression in THP-1 cells inhibitsinduction of COX-2, pSTAT3, AR,and c-myc proteins in RWPE-1cells during coculture. F, reducedEMT-related gene expressionin RWPE-1/THP-1 ARsi cells.�, P < 0.01. (Continued on thefollowing page.)

Fang et al.





(Fig. 3E). Similarly, the anti-CCL4 antibody reduced the induc-tion of EMT markers, N-cadherin and Snail (Fig. 3E). Theseresults suggested that CCL4 plays a consistent and essentialrole in the entire process of THP-1–induced prostate tumor-igenesis in RWPE-1 cells. Together, our study supports amechanism by which THP-1 cells induced prostate tumori-genensis involves the sequential activation of CCL4- andSTAT3-mediated pathways, leading to downregulation ofp53/PTEN and induction of EMT for prostate tumorigenesis.To determine whether CCL4 directly functions as a tumor

promoter for prostate epithelial cells, we found that treatingRWPE-1 cells with CCL4 failed to induce the expression ofEMT markers (MMP9 and Snail) that were upregulated inthe coculture-conditioned media (Supplementary Fig. S6A).We postulated that the reason that naive RWPE-1 cells wereunable to respond to CCL4 could be the chemokine receptorlevels. To test this hypothesis, we used qRT-PCR analysis toexamine the expression levels of 2 CCL4 receptors, CCR1 andCCR5, in RWPE-1 cells. CCR1 expression was low in mono-cultured RWPE-1 cells, but was significantly induced aftercoculture with THP-1 cells (Supplementary Fig. S6B), sug-gesting the cooperative role of such receptors with CCL4during the interaction between macrophages and prostateepithelial cells.

Macrophage AR promotes macrophage-inducedprostate tumorigenesis via CCL4/STAT3-dependentpathwaysInterestingly, the results of our earlier report identified a

regulatory role of AR inmacrophages during the wound healingprocess via modulation of chemokine receptors, macrophagemigration, and selective proinflammatory cytokines (6). How-ever, the precise nature and function of AR in directing macro-

phages during prostate tumorigenesis is not clear. This is animportant yet unanswered question because most AR-pros-tate cancer research has been focused on AR's roles withinprostate epithelial and stromal cells during prostate cancerdevelopment (19, 20). Therefore, we hypothesized that AR inmacrophages is also a key player in the crosstalk betweenmacrophages and prostate epithelial cells for macrophage-mediated prostate tumorigenesis shown in Fig. 1. Impor-tantly, we found that AR silencing by lentiviral AR-siRNA inTHP-1 cells (THP-1 ARsi), suppressed their ability to pro-mote cell proliferation (Fig. 4A) and migration (Fig. 4B) ofRWPE-1 cells during coculture. Consistently, the in vitrotransformation capacity of THP-1 ARsi cells on RWPE-1 cellswas reduced in both three-dimensional culture and colonyformation assays (Fig. 4C and D). Similarly, reduced EMT-related gene and inflammatory cytokine expression wasoberved in BPH-1 cells that were cocultured with THP-1ARsi cells (Supplementary Fig. S7A–S7C), supporting a rolefor AR in macrophage-induced prostate tumorigenesis.

We also found that silencing macrophage AR expression inTHP-1 cells reduced expression of downstream oncogenicmediators, such as pSTAT3, AR, and c-Myc in RWPE-1 cells(Fig. 4E). THP-1 ARsi cells failed to significantly upregulateEMT markers, Snail, Vimentin, and N-cadherin expression inRWPE-1 cells (Fig. 4F). Several cytokines mRNA expressionlevels were reduced by AR silencing (Fig. 4G) with CCL4showing the most significant change comparing RWPE-1/THP-1sc (scramble) with RWPE-1/THP-1ARsi cells (Fig. 4H),consistent with our recent study identifying CCL4 as an ARtarget gene (16). Reduced CCL4 protein levels in the coculture-conditioned media of RWPE-1/THP-1ARsi cells were con-firmed by ELISA (Fig. 4I). Interestingly, AR silencing in THP-1cells also resulted in the reduction of the CCR1 expression

Figure 4. (Continued. ) G,quantification of cytokine arrayanalysis of the coculture-conditioned media from RWPE-1/THP-1 scramble or RWPE-1/THP-1 ARsi cells. H, qPCRanalysis of CCL3, CCL4, CCL5,and IL-8 mRNA levels in RWPE-1/THP-1 scramble and RWPE-1/THP-1ARsi cells. I, the amount ofCCL4 was determined by ELISA inthe media of RWPE-1/THP-1scramble and RWPE-1/THP-1ARsi cells cocultured for 48 hours.All data shown are mean � SD.�, P < 0.05, ��, P < 0.01.






levels (Supplementary Fig. S6B), indicating a similar regu-latory mechanism of CCL4/CCR1 upregulation by macro-phage AR during coculture. These results suggest that infil-trated macrophages may simultaneously trigger CCL4 andits cognate receptor expression in RWPE-1 cells and enableRWPE-1 cells to respond to CCL4 stimulation. Collectively,these results suggest that macrophage AR plays a role inmediating CCL4 induction, STAT3, chemokine receptorCCR1 expression, and EMT to promote the macrophage-induced prostate tumorigenesis.

Macrophage AR ablation inhibits PIN formation inPTENþ/� mice

To further investigate the in vivo role of macrophage AR inprostate tumorigenesis, we generated a bigenic mouse PTENmutant line with the genetic background of macrophage ARknockout (MARKO: ARfl/y: lyzM-creþ/PTENþ/�, see Supple-mentary Fig. S8 for details onmating strategy and confirmationof the genotypes), taking advantage of thewell known lysozymepromoter-driven Cre enzyme to delete the target gene inmature macrophages at 83% to 98% efficiency (6, 21, 22). Using

Figure 5. Characterization ofMARKO-PTENþ/� prostate.A, gross observation of thedorsolateral prostate, anteriorprostate, and ventral prostate fromwild-type, PTENþ/� and MARKO/PTENþ/� mice. Reduced size ofanterior prostate in MARKO/PTENþ/� mice. B1–B9, IHCanalysis of F4/80, CCL4, p53,pAKT, pSTAT3, E-cadherin, andN-cadherin expression in anteriorprostate of wild-type, PTENþ/�,and MARKO/PTENþ/� mice. TheAR ablation in macrophagereduced expression levels ofF4/80, CCL4, pAkt, pSTAT3, p53,and N-cadherin in PTENþ/�

prostate. Magnification, �10. Insetmagnification, �40. Arrowheadsindicate p53-positive cells.(Continued on the following page.)

Fang et al.





this MARKO model we can observe the in vivo prostate tumordevelopment, whereasAR is ablated in infiltratedmacrophagesby the Cre–loxP system to mimic our in vitro cocultureexperiments.Mice were sacrificed at 6 months, when PIN formation

can be detected in PTENþ/� mice. We observed increasedanterior prostate size in PTENþ/�mice when compared withthe controls. Deleting AR in macrophages led to reducedprostate size in MARKO/PTENþ/� mice when comparedwith PTENþ/� mice (Fig. 5A), along with minimal effectson serum testosterone (Supplementary Fig. S8C). Strikingly,we found reduced PIN formation in MARKO/PTENþ/� pros-tate when compared with abundant moderate/low-gradePIN in PTENþ/� prostates (Fig. 5B1), suggesting that defectsin AR-deficient macrophages can suppress haplodeficiencyof PTEN-induced prostate tumorigenesis.We then conducted F4/80 immunostaining and found that

F4/80þmacrophages were reduced inMARKO/PTENþ/� pros-tates when compared with PTENþ/� prostates (Fig. 5B2).These findings suggest a positive correlation ofmouse prostatetumorigenicity with the degree of macrophage infiltrationmodulated by macrophage AR. We also examined the keymediators previously identified in our in vitro coculture mod-els. Our results, consistent with our in vitro and in vivo findings(Figs. 2–4), show the consequences of deletingmacrophage AR:decreased macrophage infiltration into the prostate, reducedexpression of CCL4, pAKT (a marker of PTEN function),pSTAT3, p53, and EMT genes (N-cadherin and Snail) andincreased E-cadherin expression in the anterior prostate ofMARKO/PTENþ/�mice compared with those found in controlPTENþ/� prostate (Fig. 5B3–B9). Taken together, our bigenicmouse model shows that macrophage AR ablation markedlyattenuates PIN formation in PTENþ/� mice. These resultsprovide the first in vivo evidence showing essential roles of

infiltrating macrophages for the induction of prostate tumor-igenesis and support a role for macrophage AR or its down-stream genes (such as CCL4) in PIN formation of PTENþ/�

mice.

Targeting AR with an AR-degradation enhancer, ASC-J9,to suppress macrophage-induced prostatetumorigenesis

We have shown that the AR-CCL4 axis is important formacrophage-induced prostate tumorigenesis. Therefore, wetested a potential therapeutic approach by targeting theupstream AR with an AR-degradation enhancer, ASC-J9,which could selectively degrade AR proteins in certain cells,including prostate cells and macrophages (6, 7, 23).

We first treated the cocultured RWPE-1 cells with 5 mmol/LASC-J9 or vehicle control and found that ASC-J9–suppressedTHP-1–mediated cell proliferation (Fig. 6A) and migration(Fig. 6B). ASC-J9 also suppressed the THP-1–mediated dis-ruption of the development of acinar-like spheroids ofRWPE-1 cells in a three-dimensional culture condition (Fig.6C). More importantly, as shown in Fig. 6D, ASC-J9 reducedthe growth of in vivo orthotopic xenografted tumors devel-oped from long-term cultured RWPE-1 cells (left and middlepanels) with induced AR degradation in RWPE-1 tumor cellsand prevented p53/PTEN downregulation of xenograftedtumors (right panel). When we examined CCL4 expression,we found that ASC-J9 and Casodex (an antiandrogen cur-rently used to treat prostate cancer) effectively suppressedinduction of CCL4 expression (Fig. 6E), further confirming anovel role of AR in controlling CCL4 expression as these twodrugs target AR in various cell types in the xenograftedtumor microenviroment. Consistently, ASC-J9 preventedp53/PTEN downregulation in cocultured RWPE-1 cellsand inhibited expression of EMT-related genes (Snail and

Figure 5. (Continued. )






N-cadherin) in cocultured RWPE-1 cells in a three-dimen-sional culture condition (Fig. 6F and G). In addition to ARprotein downregulation, one striking feature of this datais the apparent downregulation of pSTAT3 by ASC-J9 at 5mmol/L (Fig. 6F), suggesting that this small molecule can

simultaneously target AR and STAT3 function in trans-formed RWPE-1 cells and inhibit downstream tumorigenicevents. Taken together, results shown in Fig. 6A–G supportour working model that therapeutic targeting of prostatic/macrophage AR and STAT3 by ASC-J9 might provide a new

Figure 6. Targeting macrophage AR and its downstream consequences. A and B, ASC-J9 effects on cell growth (A) and migration (B). C, ASC-J9 effecton acinar morphogenesis by three-dimensional culture. ASC-J9 degraded AR in THP-1 can reverse the macrophages caused by glandular structurechange. D, gross observation, H&E, and weight analysis of xenografted tumors (left and middle) and Western blot analysis of in vivo orthotopic xenograftedtumors (right). E, effects of ASC-J9 or casodex on CCL4 mRNA level using qPCR. F, effects of ASC-J9 on PTEN and p53 protein downregulation incocultured RWPE-1 cells. G, ASC-J9 suppressed macrophages-induced EMT marker expression (N-cadherin and Snail) in RWPE-1 cells. Differentiatedmonoculture or coculture RWPE-1 spheroids were photographed by confocal microscopy in a three-dimensional culture condition. Data shown are meanSD. �, P < 0.05, ��, P < 0.01.

Fang et al.





approach for treating patients during the early-stage ofprostate cancer development.

Expression of CCL4, p53, and EMT-related proteins inhuman prostate tissuesTo determine the clinical significance of our findings about

the keymediator, CCL4, in prostate cancer, we examined CCL4expression in prostatectomy specimens (Fig. 7A). CCL4 waspositive in 21/72 (29%) benign, 46/62 (74%) HGPIN, and 50/75

(67%) carcinoma tissues (Supplemetary Table S1). Thus, CCL4levels were significantly higher in HGPIN and prostate cancerthan in nonneoplastic prostate, whereas there was no signif-icant difference in CCL4 expression between HGPIN andprostate cancer, suggesting its involvement in early prostatecancer development. We also found no strong associations ofCCL4 expression with clinicopathologic features of prostatecancer, including Gleason score, pathologic stage, lymphnode metastasis, and biochemical recurrence, except higher

Figure 7. IHC analysis of CCL4 (A)and p53/pAkt/E-cadherin/Snail (B)in prostate TMAs. Arrowheadsindicate benign glands adjacent tocancer glands (A) or p53/pAKTcells (B).






expression score in GS � 6 tumors compared with GS7 (P ¼0.0088) or GS� 7 (P¼ 0.0090) tumors (Supplementary Table S2and Fig. S9). Next, we examined the expression of p53, pAKT,E-cadherin, and Snail in these prostate tissues (Fig. 7B).p53 (Supplementary Table S3) and E-cadherin (SupplementaryTable S5) were significantly reduced in HGPIN, compared withbenign prostate. In contrast, pAKT (Supplementary Table S4)and Snail (Supplementary Table S6) were significantly in-creased in HGPIN or prostate cancer, compared with benignprostate. pAKT and E-cadherin were also found to be higherin prostate cancer than in HGPIN. Collectively, resultsfrom human prostate tissue array analysis (Fig. 7A and B)confirmed our findings showing that these mediators identi-fied in our in vitro coculture models are important in the earlyprostate cancer development and could be of prognostic valuesfor prostate cancer development.

DiscussionOne previous study showed that E. coli infections in the

prostate induce PIN formation in mice via persistent secretionof free radicals and/or inflammatory cytokines by infiltratingleukocytes during recurrent bacterial infections (24). Theseresults suggest that infiltrating leukocytes are key players inprostate cancer development. Among infiltrating leukocytes,macrophages are generally regarded as the key players in thetumor microenvironment that supports tumor progression(25). Macrophages are often viewed as double agents in themicroenvironment because their functional plasticity enablesthem to switch to a phenotype that is either for or againsttumor development and progression (26). Importantly, theTHP-1 cells used here were reported to possess characteristicsof M2-like TAMs (tumor-associated macrophages) with pro-tumor and immunosuppressive activity in the preexistingtumor microenvironment (26, 27). However, whether the pre-existing proinflammatory environment in prostate glands,before tumorigenesis, would favor M1 or M2 polarization ofmacrophages remains unclear. It is possible that our in vitrococulture model may provide a preneoplastic environment forprostate epithelial cells and allow THP-1 cells to expand theirM2-like roles to promote prostate tumorigenesis (27). There-fore, our findings suggest that macrophages are capable ofstimulating prostate tumorigenesis and this is a new discoverydistinct from previous studies focusing on macrophage func-tion on tumor angiogenesis/metastasis (28, 29).

Importantly, we identified CCL4 as a novel downstreammediator of macrophage AR, and showed that neutralizing theCCL4 activity can effectively block macrophage-mediatedinduction of cytokine expression, STAT3 activation, EMTgenes, p53/PTEN downregulation, and cell migration (Figs.2F–I and 3E and F). These findings suggest a new role for CCL4and its downstream signaling in early prostate tumorigenesis.As CCL4 was mainly induced in cocultured RWPE-1 cells in anAR-dependent fashion, this data raises the intriguing possibil-ity thatmacrophage AR could be functioning to elicit unknownsignals for autocrine action of CCL4 induction in RWPE-1 cellsduring coculture. More importantly, additional perplexingdata shows that direct CCL4 treatments failed to elicit anydownstream signaling pathways in RWPE-1 cells, suggesting

that additional signals generated by the macrophages forupregulation of CCL4 receptors may underlie the requirementfor RWPE-1 cells to respond to CCL4. It will be fascinating todetermine precisely how macrophages induce upregulation ofCCL4 receptors in RWPE-1 cells.

An early study showed that STAT3 activation might antag-onize the p53/PTEN function in promoting tumor migrationand invasion (30) and transformation of nontumorigenic pros-tate epithelial cells to prostate cancer cells may require sup-pression of tumor suppressor genes, such as PTEN and p53 (31,32). Our demonstration that the long-term coculture of non-tumorigenic prostate cells in the coculture medium resulted inCCL4-dependent suppression of p53/PTEN function, suggest-ing that this key step may be required for prostate tumori-genesis. A delicate regulatory relationship exists between p53and PTEN in which inactivation of one of the genes results inreduction of the protein levels in the other (33). It will beinteresting to dissect how CCL4 downstream signaling canlead to STAT3 activation and downregulation of both p53 andPTEN in a sequential manner during long-term coculture-induced transformation.

It is generally believed that EMT is the major focus as theconvergence point between inflammation and cancer progres-sion, such as in metastasis (34). Interestingly, our data suggestthat the induction of EMT could be an important processinvolved in early prostate tumorigenesis. It has been shownthat expression of constitutively-active STAT3 induces pros-tate tumorigenesis of immortalized-prostate epithelial cellsand the induction of EMT has been identified in this study (12),suggesting that the EMT could be one of the key componentsduring prostate tumor initiation. To obtain the in vivo evidenceof EMT during prostate tumor initiation, we examined Snailexpression in the PIN lesions of prostate of PTENf/f:probasin-creþmice and found that notable Snail expression is associatedwith PIN lesions (Supplementary Fig. S10), similar to onerecent study using qPCR analysis showing that the EMTinducer, Snail expression, is upregulated in PTEN-null pros-tates (35). Altogether, our studies support that CCL4-inducedEMT in RWPE-1 cells could be a novel and important processinvolved in prostate tumor initiation.

Using ASC-J9, an AR protein-degradation enhancer, weprovide evidence that ASC-J9 can effectively reduce xeno-grafted RWPE-1 tumor size via promoting AR degradationand preventing p53/PTEN downregulation. Consistent withour recent study (19), our data showed that ASC-J9 suppressesPIN formation and reduces infiltration of inflammatorymacro-phages in PTENþ/� mice that can spontaneously develop PIN,suggesting that targeting AR activity by ASC-J9 in the tumormicroenvironment may not only inhibit prostatic AR functionin prostate epithelial cells (36–38), but also prevent macro-phages from exerting their protumor function during prostatetumorigenesis. These findings may provide a new therapeuticapproach using ASC-J9 for inhibiting prostate cancer at thePIN development stage.

In summary, our study establishes an in vitromodel in whichthe paracrine action was initiated by macrophages-inducedCCL4 production during coculture with immortalized-pros-tate epithelial cells. Then, the induction of CCL4 promotes

Fang et al.





prostate tumorigenesis through STAT3 activation, EMT, anddownregulation of p53/PTEN proteins in RWPE-1 cells (Figs.2H and I and 3E). Targeting AR with ASC-J9 in xenograftedRWPE-1 tumors inhibits tumor growth and reduces CCL4expression via induction of AR protein degradation and inhib-iting STAT3 activation (Fig. 6D andE), suggesting AR andCCL4in the tumor microenvironment as potential therapeutic tar-gets to effectively block inflammation-associated prostatetumor initiation.

Disclosure of Potential Conflicts of InterestChawnshang Chang has ownership interest (including patents) in

AndroScience Corporation (ASC). No potential conflicts of interest were dis-closed by the other authors.

Authors' ContributionsConception and design: L.-Y. Fang, L. Li, W.-J. Lin, C. ChangDevelopment of methodology: L.-Y. Fang, W.-J. Lin, C. ChangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K. Izumi, K.-P. Lai, L. Liang, H. Miyamoto, W.-J. Lin

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): L.-Y. Fang, K.-P. Lai, H.Miyamoto,W.-J. Lin, C. ChangWriting, review, and/or revisionof themanuscript: L. Li, H.Miyamoto,W.-J.Lin, C. ChangAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): W.-J. LinStudy supervision: W.-J. Lin, C. Chang

AcknowledgmentsThe authors thank K. Wolf for help in editing the manuscript.

Grant SupportThis work was supported by NIH Grants (CA127300 and CA156700), DOD

Grant (W81XWH-10-0300), Taiwan Department of Health Clinical Trial andResearch Center of Excellence Grant DOH99-TD-B-111-004 (China MedicalUniversity, Taichung, Taiwan), and National Basic Research Program of China2012CB518305.

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

Received August 16, 2012; revised June 17, 2013; accepted July 3, 2013;published OnlineFirst July 22, 2013.

References1. JemalA,BrayF,CenterMM,Ferlay J,WardE, FormanD.Global cancer

statistics. CA Cancer J Clin 2011;61:69–90.2. Karin M, Lawrence T, Nizet V. Innate immunity gone awry: linking

microbial infections to chronic inflammation and cancer. Cell 2006;124:823–35.

3. Perletti G, Montanari E, Vral A, Gazzano G, Marras E, Mione S, et al.Inflammation, prostatitis, proliferative inflammatory atrophy: 'fertileground' for prostate cancer development? Mol Med Report 2010;3:3–12.

4. DeMarzo AM, Platz EA, Sutcliffe S, Xu J, Gronberg H, Drake CG, et al.Inflammation in prostate carcinogenesis. Nat Rev Cancer 2007;7:256–69.

5. Sica A, Larghi P, Mancino A, Rubino L, Porta C, Totaro MG, et al.Macrophage polarization in tumour progression. Semin Cancer Biol2008;18:349–55.

6. Lai JJ, Lai KP, Chuang KH, Chang P, Yu IC, Lin WJ, et al. Monocyte/macrophage androgen receptor suppresses cutaneous wound heal-ing in mice by enhancing local TNF-alpha expression. J Clin Invest2009;119:3739–51.

7. Yang Z, Chang YJ, Yu IC, Yeh S, Wu CC, Miyamoto H, et al.ASC-J9 ameliorates spinal and bulbar muscular atrophy pheno-type via degradation of androgen receptor. Nat Med 2007;13:348–53.

8. Zhu P, Baek SH, Bourk EM, Ohgi KA, Garcia-Bassets I, Sanjo H,et al. Macrophage/cancer cell interactions mediate hormone resis-tance by a nuclear receptor derepression pathway. Cell 2006;124:615–29.

9. Kim JB. Three-dimensional tissue culture models in cancer biology.Semin Cancer Biol 2005;15:365–77.

10. Hu WY, Shi GB, Lam HM, Hu DP, Ho SM, Madueke IC, et al.Estrogen-initiated transformation of prostate epithelium derivedfrom normal human prostate stem-progenitor cells. Endocrinology2011;152:2150–63.

11. Chu JH, Yu S, Hayward SW, Chan FL. Development of a three-dimensional culture model of prostatic epithelial cells and its use forthe study of epithelial–mesenchymal transition and inhibition of PI3Kpathway in prostate cancer. Prostate 2009;69:428–42.

12. Azare J, Leslie K, Al-Ahmadie H, Gerald W, Weinreb PH, Violette SM,et al. Constitutively activated Stat3 induces tumorigenesis andenhances cell motility of prostate epithelial cells through integrin beta6. Mol Cell Biol 2007;27:4444–53.

13. Jiang M, Jerome WG, Hayward SW. Autophagy in nuclear receptorPPARgamma-deficient mouse prostatic carcinogenesis. Autophagy2010;6:175–6.

14. Jiang M, Fernandez S, JeromeWG, He Y, Yu X, Cai H, et al. Disruptionof PPARgamma signaling results in mouse prostatic intraepithelialneoplasia involving active autophagy. Cell Death Differ 2010;17:469–81.

15. Chen T, Wang LH, Farrar WL. Interleukin 6 activates androgen recep-tor-mediated gene expression through a signal transducer and acti-vator of transcription 3-dependent pathway in LNCaP prostate cancercells. Cancer Res 2000;60:2132–5.

16. Lai KP, Yamashita S, Huang CK, Yeh S, Chang C. Loss of stromalandrogen receptor leads to suppressed prostate tumourigenesis viamodulation of proinflammatory cytokines/chemokines. EMBO MolMed 2012;4:791–807.

17. Dong JT. Prevalent mutations in prostate cancer. J Cell Biochem2006;97:433–47.

18. Chen Z, Trotman LC, Shaffer D, Lin HK, Dotan ZA, Niki M, et al. Crucialrole of p53-dependent cellular senescence in suppression of Pten-deficient tumorigenesis. Nature 2005;436:725–30.

19. Lai KP, Yamashita S, Vitkus S, Shyr CR, Yeh S, Chang C. Suppressedprostate epithelial development with impaired branching morphogen-esis in mice lacking stromal fibromuscular androgen receptor. MolEndocrinol 2012;26:52–66.

20. Niu Y, Chang TM, Yeh S, Ma WL, Wang YZ, Chang C. Differentialandrogen receptor signals in different cells explain why androgen-deprivation therapy of prostate cancer fails. Oncogene 2010;29:3593–604.

21. Di Cristofano A, Pesce B, Cordon-Cardo C, Pandolfi PP. Pten isessential for embryonic development and tumour suppression. NatGenet 1998;19:348–55.

22. Clausen BE, Burkhardt C, Reith W, Renkawitz R, Forster I. Conditionalgene targeting inmacrophages and granulocytes using LysMcremice.Transgenic Res 1999;8:265–77.

23. Miyamoto H, Yang Z, Chen YT, Ishiguro H, Uemura H, Kubota Y, et al.Promotion of bladder cancer development and progression by andro-gen receptor signals. J Natl Cancer Inst 2007;99:558–68.

24. Elkahwaji JE, Hauke RJ, Brawner CM. Chronic bacterial inflammationinduces prostatic intraepithelial neoplasia in mouse prostate. Br JCancer 2009;101:1740–8.

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

26. Sica A. Role of tumour-associated macrophages in cancer-relatedinflammation. Exp Oncol 2010;32:153–8.

27. Kaler P, Augenlicht L, Klampfer L. Macrophage-derived IL-1betastimulates Wnt signaling and growth of colon cancer cells: a crosstalkinterrupted by vitamin D3. Oncogene 2009;28:3892–902.






28. Qian B, Deng Y, Im JH, Muschel RJ, Zou Y, Li J, et al. A distinctmacrophage population mediates metastatic breast cancer cellextravasation, establishment and growth. PLoS ONE 2009;4:e6562.

29. Joyce JA, Pollard JW. Microenvironmental regulation of metastasis.Nat Rev Cancer 2009;9:239–52.

30. Mukhopadhyay UK, Mooney P, Jia L, Eves R, Raptis L, Mak AS.Doubles game: Src-Stat3 versus p53-PTEN in cellular migration andinvasion. Mol Cell Biol 2010;30:4980–95.

31. Assinder SJ, Dong Q, Kovacevic Z, Richardson DR. The TGF-beta,PI3K/Akt, and PTEN pathways: established and proposedbiochemical integration in prostate cancer. Biochem J 2009;417:411–21.

32. Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, et al. PTEN,a putative protein tyrosine phosphatase gene mutated inhuman brain, breast, and prostate cancer. Science 1997;275:1943–7.

33. Freeman DJ, Li AG, Wei G, Li HH, Kertesz N, Lesche R, et al. PTENtumor suppressor regulates p53 protein levels and activity throughphosphatase-dependent and -independent mechanisms. Cancer Cell2003;3:117–30.

34. Lopez-Novoa JM, Nieto MA. Inflammation and EMT: an alliancetowards organ fibrosis and cancer progression. EMBO Mol Med2009;1:303–14.

35. MulhollandDJ,KobayashiN,RuscettiM, Zhi A, Tran LM,Huang J, et al.Pten loss and RAS/MAPK activation cooperate to promote EMT andmetastasis initiated from prostate cancer stem/progenitor cells. Can-cer Res 2012;72:1878–89.

36. Yamashita S, Lai KP, Chuang KL, Miyamoto H, Tochigi T, Pang ST,et al. ASC-J9 suppresses castration-resistant prostate cancer growthvia degradation of full-length and splice variant androgen receptors.Neoplasia 2012;14:74–83.

37. Lin TH, Niu Y, Lee SO, Xu D, Liang L, Li L, et al. Differential androgendeprivation therapies with anti-androgens asodex/bicalutamide orMDV3100/enzalutamide versus anti-androgen receptor ASC-J9� leadto promotion versus suppression of prostate cancer metastasis. JBC2013;288:19359–69.

38. Lin TH, Izumi K, Lee SO, Lin WJ, Yeh S, Chang C, et al. Anti-androgenreceptor of ASC-J9 vs anti-androgens casodex or MDV3100 leadsto opposite effects on prostate cancer metastasis via differentialmodulation of macrophage infiltration and STAT3-CCL2 signaling.Cell Death and Disease 2013; in press.


Fang et al.




2013;73:5633-5646. Published OnlineFirst July 22, 2013.Cancer Res Lei-Ya Fang, Kouji Izumi, Kuo-Pao Lai, et al.

STAT3 Signaling−Modulating Androgen Receptor-Mediated CCL4 Infiltrating Macrophages Promote Prostate Tumorigenesis via

Updated version

10.1158/0008-5472.CAN-12-3228doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/07/22/0008-5472.CAN-12-3228.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/73/18/5633.full#ref-list-1

This article cites 37 articles, 9 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/73/18/5633.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/73/18/5633To 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-12-3228

http://cancerres.aacrjournals.org/content/suppl/2013/07/22/0008-5472.CAN-12-3228.DC1

http://cancerres.aacrjournals.org/content/73/18/5633.full#ref-list-1

http://cancerres.aacrjournals.org/content/73/18/5633.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/73/18/5633


ﬁltrating Macrophages Promote Prostate Tumorigenesis via...

Documents

Transcript of ﬁltrating Macrophages Promote Prostate Tumorigenesis via...