VDR Status Arbitrates the Prometastatic Effects of Tumor ...Signal Transduction VDR Status...

Signal Transduction

VDR Status Arbitrates the Prometastatic Effects of Tumor-Associated Macrophages

Yan Zhang1, Quanjun Guo1, Zhujun Zhang2, Nan Bai1, Ze Liu1, Min Xiong2, Yuquan Wei3,Rong Xiang1, and Xiaoyue Tan2

AbstractThe relationship between tumor-associated macrophages (TAM) and epithelial-to-mesenchymal transition

(EMT) during the initiation and progression of metastasis is still unclear. Here, a role for the vitamin D receptor(VDR) in metastasis was identified, as well as a role in the relationship between TAMs and EMT. First, theexpression level of VDRwas examined in clinical tissue fromhuman patients with breast cancer or amousemodel ofbreast cancer with differential metastasis. These results revealed that VDR expression negatively correlates withmetastasis in breast cancer. Second, coculture of VDR-overexpressing breast cancer cells with amacrophage cell linedemonstrated that overexpression of VDR alleviated the prometastatic effect of cocultured macrophages on breastcancer cells. Furthermore, VDRoverexpression abrogated the induction of EMT in breast cancer cells by coculturedmacrophage cells, as measured by a loss of E-cadherin (CDH1) and induction ofa-smooth muscle actin (a-SMA).TNFa in macrophage conditioned media inhibited VDR expression, whereas downregulation of VDR furthermediated the promotion of TGFb-induced EMT by TNFa. In addition, b-catenin expression was inhibited inVDR-overexpressing breast cancer cells and tumor xenografts. Finally, administration of calcitriol [1,25-(OH)2D3], an active vitaminDmetabolite, exerted similar antimetastatic effects in breast cancer cells in vitro and amousemodel of breast cancer in vivo with preservation of VDR and suppression of b-catenin.

Implications:VDR suppression byTNFamediates the prometastatic effect of TAMs through enhancement of theb-catenin pathway. Mol Cancer Res; 12(8); 1181–91. �2014 AACR.

IntroductionMetastasis has emerged as the primary cause of poor

prognosis for patients with breast cancer, in part, as a resultof significant progress in the early diagnosis and therapyduring recent years. Accumulating evidence suggests that thederangement of the tumor microenvironment is one of thecritical factors in the malignant progression of tumor. Thetumormicroenvironment includes a wide variety of cells thatare involved in the acquisition of malignant tumor hallmarktraits (1). It is currently believed that macrophages are themost abundant cells in the tumor microenvironment, play-ing active roles in almost all aspects of tumor growth anddevelopment (2, 3). Antitumor strategies targeting tumor-

associated macrophages (TAM) have achieved encouragingresults in impairing the metastasis of solid tumors (4, 5).Several factors have been found to be involved in macro-phage-stimulated invasiveness, such as an EGF—colony-stimulating factor 1 (CSF1) paracrine interaction, theWnt5a noncanonical pathway, and the induction of TNFaby theNF-kBpathway (6–8).However, the precisemechan-isms underlying the prometastatic role of macrophagesremain to be fully elucidated.The vitamin D receptor (VDR) belongs to the nuclear

hormone receptor superfamily and mediates the majorbiologic effects of vitamin D. Upon ligand binding, VDRrecruits and forms complexes with cofactors such as theretinoid X receptor. The complex then binds to the VDRelement in the promoter region of target genes to regulategene transcription. Previous studies have shown that VDRgene polymorphism alters the risk of breast cancer (9–11).Comparative genome analysis identified VDR as a directtranscriptional target of p53 and that VDR plays a role inp53-mediated suppression of tumor growth (12). Morerecently, a positive association between VDR expressionlevel and a prolonged progression-free and overall survivalof patients with breast cancer have been reported (13).However, the mechanisms behind the loss of VDR and itssubsequent influence on tumor metastasis remain poorlyunderstood.

Authors' Affiliations: Departments of 1Immunology and 2Pathology,Medical School of Nankai University, Tianjin; and 3State Key Laboratoryof Biotherapy, West China Hospital, Sichuan University, Chengdu, China

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

Corresponding Authors: Xiaoyue Tan, Department of Pathology, MedicalSchool of Nankai University, 94 Weijin Road, Tianjin 300071, China.Phone: 86-22-23504447; Fax: 86-22-23502554; E-mail:[email protected]; and Rong Xiang. Phone: 86-22-23509505;Fax: 86-22-23502554; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-14-0036

�2014 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org 1181

on August 5, 2020. © 2014 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst May 12, 2014; DOI: 10.1158/1541-7786.MCR-14-0036

http://mcr.aacrjournals.org/

In a previous study, we reported that TNFa in theextracellular matrix inhibited VDR expression in renalepithelial cells, potentially mediating the interactionbetween inflammation and fibrosis (14). Here, we examinewhether the tumor microenvironment, especially TAMs,exerts a similar effect on tumor cells. We also examine thecytokines or chemokines responsible for this effect. Inaddition, we investigate the involvement of VDR, TAMs,and metastasis in breast cancer.

Materials and MethodsTissue microarrayVDRexpressionwas detected in high-density tissuemicro-

arrays on samples from a cohort of 80 patients with breastcancer (catalog no. BR801, Alenabio). For quantification,VDR expression in breast cancer tissues was assessed accord-ing to Remmele and Stegner as previously described (15). Inbrief, category A documented the intensity of immunostain-ing as 0 (no immunostaining), 1 (weak), 2 (moderate), and 3(strong); category B documented the number of immuno-reactive cells as 1 (0%–25%), 2 (26%–50%), 3 (51%–75%),and 4 (76%–100%). Values for category A and B weremultiplied to construct an immunoreactivity score (IRS)ranging from 0 to 12.

AnimalsFemale BALB/c mice aged 6 to 8 weeks were purchased

from the Laboratory Animal Center, the Academy of Mil-itary Medical Sciences (Beijing, China), and housed understandard laboratory conditions. All animal experiments wereperformed according to health guidelines of the NankaiUniversity Institutional Animal Use and Care Committee.For establishing the syngeneic, orthotopic mouse models ofbreast cancer, mice were injected once with either 1 � 105

wild-type (WT), vector control, or VDR-overexpressing 4T1cells into the fourth mammary fat pads (16). For the activevitamin D administration experiments, 1,25-dihydroxyvita-minD3 [1,25(OH)2D3] or vehicle controlwas administeredfrom the day before tumor cells injection by intraperitonealinjection at a dose of 0.3 mg/kg body weight once every otherday, respectively. Tumor volumes were measured daily fromthe tenth day after injection. Mice were sacrificed at theeighth week after injection. Lung and tumor tissues wereisolated under anesthesia for further analysis. Metastaticnodule count per lung was used as the quantitative indicatorfor the lung metastasis as described previously (17).

Cell lines and reagents4T1 andRAW264.7 cells were obtained from the ATCC.

Cells were grown in RPMI-1640 (GIBCO) supplementedwith 10%FBS and grown in a 5%CO2 atmosphere at 37�C.Recombinant human TGFb1 and TNFa were purchasedfrom R&D Systems. SPD304 and 1,25(OH)2D3 werepurchased from Sigma and Bio Basic Inc.

Vector constructionTo establish stable VDR-overexpressing cell line, a VDR

expression plasmid was generated by inserting the mouse

VDR gene into the pLV-EF1-MCS-IRES-Bsd (Biosettia).Lentivirus production and infection were performed accord-ing to the manufacturer's protocol, and positive cells wereselected by blasticidin S at a concentration of 4 mg/mL. Forknockdown analysis, 2 shRNAs targeting mouse VDR weredesigned and chemically synthesized as shRNA-mVDR anda scrambled sequence was used as control. The palindromicDNA oligos were annealed to each other and ligated to thelinearized pLV-H1-EF1a-puro vector (Biosettia). 4T1 cellswere transfected with pLV-H1-EF1a- shRNA-sc-puro orpLV-H1-EF1a-shRNA-mVDR-puro expression plasmids.In brief, 1� 106 4T1 cells were plated in the 6-well plate andthen transfected with Lipofectamine 2000 (Invitrogen)according to the manufacturer's instructions.

IHC and immunofluorescence stainingFor IHC,paraffin sectionswere incubated at 4�Covernight

with primary antibody after dewaxing and hydration. Then,slides were incubated with a biotinylated secondary antibodyfor 1.5 hours and then with avidin–peroxidase complex for0.5 hour. The slides were visualized with 3,30-diaminoben-zidene (DAB) and counterstained with hematoxylin. Forimmunofluorescence staining, cells were fixed with coldmethanol at�20�C for 20 minutes and blocked in 2% BSAfor 1 hour. Cells were incubated with primary antibodiesovernight at 4�C, followed by incubation with FITC-labeledsecondary antibody for 1 hour. For nuclear staining, cells werestained with 40,6-diamidino-2-phenylindole (DAPI; LifeTechnologies). Antibodies for E-cadherin (Cell Signaling),VDR (Santa Cruz Biotechnology), and b-catenin (Cell Sig-naling) were used at a 1:100 dilution, whereas the anti-a-SMA (Sigma-Aldrich) antibody was used at 1:50.

Coculture assay4T1 cells were cocultured with RAW 264.7 macrophage

cells without cell–cell contact. RAW 264.7 cells (4 � 105)suspended in 1-mL RPMI-1640 were added to the hanginginserts of a 6-well Boyden chamber with a 0.4-mm poremembrane (Millipore). 4T1 cells (1.6 � 106) in 2-mLRPMI-1640 were seeded on the bottom of each well. Cellswere cocultured for 48 hours and cell lysates were collectedseparately. For longer cocultures, the cell suspensions werefurther diluted with fresh media. Suspensions were dilutedby 4 times for 3-day cocultures, 16 times for 5 days, and 64times for 7 days.

Cell migration and invasion assays4T1 cells (1 � 105) cocultured with or without RAW

264.7 cells were added to the hanging insert of a Boydenchamber with an 8-mm pore membrane and 1-mL RPMI-1640 medium supplemented with 10% FBS in the bottomwell. After 8 hours of incubation at 37�C, cells on the upperside of the insert were removed with a cotton swab. Thebottom side was then fixed and stained with DAPI. Viablecells were counted under amicroscope (Olympus Co.). Eachassay was done in triplicate. For invasion assays, Matrigel(BD Biosciences) was diluted to 1 mg/mL with serum-freeculturemedium and immediately applied to eachmembrane

Zhang et al.

Mol Cancer Res; 12(8) August 2014 Molecular Cancer Research1182




insert to form the upper chambers of the multiwell invasionassay plate. About 1 � 105 4T1 cells were seeded into theupper chambers and incubated for 24 hours at 37�C beforeanalysis. The wound-healing assay was carried out accordingto the established protocols (18).

Western blot analysisDetection of protein expression by Western blotting was

carried out according to established protocols describedpreviously (19). Anti-VDR (1:1,000), a-smooth muscleactin (a-SMA; 1:500), E-cadherin (1:1,000), b-catenin(1:1,000), and b-actin (1:10,000) primary antibodies wereused. Secondary horseradish peroxidase–conjugated goatanti-rabbit or anti-mouse antibodies (Bio-Rad) were usedat a 1:10,000 dilution and detected using enhanced chemi-luminescence (ECL, Millipore).

Immunoprecipitation assayCell lysates were immunoprecipitated with 4 mg of anti-

VDR antibody conjugated to Dynabeads for 1.5 hours atroom temperature. Beads were then washed and the targetantigen eluted and boiled in SDS-PAGE sample buffer.Immunoblotting with anti-b-catenin antibody was thenperformed. The proteins were detected using ECL reagent(Millipore).

Dual luciferase assayDual-Luciferase Reporter Assay System (Promega) was

used to test the relative activity of firefly luciferase (FL) versusthat of Renilla luciferase (RL). Briefly, vector control orVDR-overexpressing 4T1 cells were cultured in 24-wellplates at 2 � 105 per well and transfected with a DNA mixof 700 ng pGM-Luciferase vector containing the TCF/LEF1response element sequence and 70 ng pRL-TK plasmid. Forvitamin D–treated group, 1,25(OH)2D3 was added to thegrowth medium 6 hours after transfection. Cells were har-vested after 48 hours of transfection, and the activationof TCF/LEF1 response element was quantified as a ratio ofFL/RL activity in each well following the manufacturer'sinstructions.

Preparation of conditioned medium and treatment withcytokines and chemical inhibitorsFollowing coculture with 4T1 cells for 48 hours, the RAW

264.7 cell culture medium was changed to serum-free medi-um. RAW 264.7 conditioned medium (CM) was thenharvested after 24 hours. 4T1 cells were seeded at approxi-mately 50% confluence and cultured in completemedium for24 hours. The culturemediumwas then replacedwith a 2-mLmixture of theRAW264.7CMandbasicmediumat differentratios with or without 50 mmol/L SPD 304. For the experi-ments of cytokine treatment, 4T1 cells were treated with 1 or2 ng/mL TGFb1 in the absence or presence of 10 ng/mLTNFa and various concentration of 1,25(OH)2D3 for 48hours.

Statistical analysisAll data are presented as the mean � SEM. Statistical

analysis of the data was performed using theGraphPad Prism

software (GraphPad Software). Differences between indi-vidual groups were analyzed by paired t test or c2, asappropriate. P < 0.05 was considered statistically significant.

ResultsExpression level of VDR negatively correlated with themetastatic progress of breast cancerTo reveal the correlation between the VDR level and the

metastatic status in breast cancer, we first performed immu-nohistologic staining on the samples from human patientswith breast cancer as well as samples from the orthotopicmouse model of breast cancer with 4T1 cells. In the humansamples, our results identified a stratified expression patternfor VDR in tissue sections (Fig. 1A). VDR level in the tumorsite was significantly lower than that in the normal orparacarcinoma tissue. Moreover, the expression of VDRwas negatively correlated with tumor grade (Table 1).Similarly, immunohistologic analysis of samples from themouse model of breast cancer with 4T1 cells demonstratedthat the suppression of VDR in the tumor correlated withdisease progression (Fig. 1B).

VDR overexpression suppressed metastasis in a mousemodel of breast cancerTo further explore the role of VDR in breast cancer

metastasis, we established VDR-overexpressing and vectorcontrol 4T1 stable cell lines. Mouse models of breast cancerwere generated by injecting VDR-overexpressing, vectorcontrol, or WT 4T1 cells into the mammary fat padseparately. Immunohistologic staining and real-time PCRconfirmed VDR overexpression at the tumor site in theVDR-overexpressing group (Fig. 2A and B). In accordancewith the results described above, VDR expression in tumorsappeared almost undetectable by the eighthweek after tumorcells implantation in the WT and vector control group.Notably, there was no significant difference in the tumor sizeamong the 3 groups (P¼ 0.27, Fig. 2C). Evaluation of lungmetastasis showed that the number of metastatic lungnodules in the VDR-overexpressing group was significantlyless than that in the other 2 control groups (Fig. 2D and E).Taken together, these results suggested that while VDRoverexpression did not affect primary tumor size, it didsuppress metastasis.

VDR overexpression reduced the migration and invasionability of breast cancer cells induced by macrophagecocultureTranswell assays were performed using cultured VDR-

overexpressing, vector control, or WT 4T1 cells to explorethe mechanism underlying the effect of VDR on breastcancer metastasis. No significant difference could bedetected in migration and invasion among the 3 groups(Fig. 3A and B). Given the critical role of TAMs and thetumor microenvironment, we next performed coculture ofbreast cancer cells with macrophages. Cancer cell migra-tion increased with increased coculture time. Moreover,this increased migration ability conferred by coculture

VDR Suppression Mediates the Prometastatic Effect of TAMs

www.aacrjournals.org Mol Cancer Res; 12(8) August 2014 1183




with macrophages was suppressed in VDR-overexpressing4T1 cells when compared with vector control (Fig. 3C).Similar results were obtained from an invasion assay;macrophage coculture increased 4T1 cell invasive-ness, which was also suppressed by VDR overexpression(Fig. 3D).

VDR overexpression suppressed the macrophage-induced inhibition of E-cadherin and induction ofa-SMA in breast cancer cellsEpithelial-to-mesenchymal transition (EMT) is essential

in breast cancer cell metastasis. Therefore, we measuredexpression of the epithelial marker E-cadherin and themesenchymal marker a-SMA in 4T1 cell lines cocultured

with RAW 264.7 cells by Western blotting. Macrophagecoculture inhibited E-cadherin expression and induceda-SMA expression in control but not in VDR-overexpres-sing 4T1 cells (Fig. 3E). There are 2 major characteristics ofthe EMT process: loss of the epithelial phenotype andacquisition of the mesenchymal phenotype. Previous stud-ies have suggested that the loss of E-cadherin is the initialand essential step for EMT (20). Here, immunofluores-cence staining showed a prominent downregulation of E-cadherin in vector control 4T1 cells from the third day ofcoculture with RAW 264.7 cells, becoming almost unde-tectable by the seventh day. However, E-cadherin levelsremained stable in VDR-overexpressing cells even after 7days of coculture (Fig. 3F).

Table 1. The correlation of VDR with clinical status of patients with breast cancer

VDR IRS

Tissue classification 0–1 2–4 6–12 Total P

Tissue source <0.001Normal/paracarcinoma 0 1 (10%) 9 (90%) 10Tumor 12 (17.1%) 31 (44.3%) 27 (38.6%) 70

Histologic grade 0.2Grade 1–2 8 (15.4%) 23 (44.2%) 21 (40.4%) 52Grade 3 4 (22.2%) 8 (44.5%) 6 (33.3%) 18

TNM stage <0.05I 3 (12.5%) 10 (41.7%) 11 (45.8%) 24II 2 (16.6%) 5 (41.7%) 5 (41.7%) 12III–IV 7 (20.6%) 16 (47.1%) 11 (32.3%) 34

aA

B Week 3 Week 5 Week 7

1.0

3 5 7 (week)

Fol

d ch

ange

0.5

0.0

b c d e

Figure 1. Expression level of VDR correlates negatively with the metastatic status of breast cancer. A, high-density tissue microarrays of breast cancercontaining samples from 80 patients were stained with antibody against VDR. Representative pictures show the stratified expression of VDR in tissuesamples: (a) normal breast tissue, (b) malignant breast tumor (grade 1, well-differentiated), (c) malignant breast tumor (grade 2, moderately differentiated), (d)malignant breast tumor (grade 3, poorly differentiated), and (e) malignant breast tumor (grade 4, undifferentiated). Scale bar, 100 mm. B, representativeimages showVDR expression in the tumor tissues of an orthotopicmousemodel of breast cancer with 4T1 cells at 3, 5, and 7weeks after injection. Scale bar,100 mm. Right, the quantification of the VDR expression shown in B. The data are shown as mean � SEM. ��, P < 0.001.

Zhang et al.





TNFa secreted bymacrophages inhibits VDR expressionin breast cancer cellsWe next explored the functional role of macrophage

coculture on the metastatic ability and EMT potential ofbreast cancer cells. CM from a 48-hour coculture of RAW264.7 and 4T1 cells were collected and mixed with freshbasal media (BM) at several ratios. CM/BM was then addedto 4T1 cells. VDR expression was suppressed in these cells ina dose-dependent manner (Fig. 4A). Furthermore, elevatedTNFa mRNA level in RAW264.7 cells could be detectedafter coculture of 4T1 cells (Fig. 4B); at the same time, theinhibitory effect onVDR could be rescued by administrationof SPD304, a specific TNFa inhibitor (Fig. 4C), whichsuggests a role for TNFa in the inhibition of VDR expres-sion. This effect does not appear to be cell line- or species-specific as we found thatTNFa inhibitedVDR expression ineither 4T1 cells (Fig. 4D and E) or MCF7 cells (Supple-mentary Fig. S1A and S1B) in both time- and dose-depen-dent manners.

VDR downregulation mediates the effect of TNFa onpromoting the EMT potential in breast cancer cellsWe next evaluated whether VDR inhibition by TNFa

increased the EMT potential in breast cancer cells. Pre-

treatment with TNFa significantly enhanced the sensi-tivity of 4T1 cells to TGFb1-induced EMT (Fig. 4F).Consistent with our observations from above, knockdownof VDR expression by shRNA transfection in conjunctionwith a low dose of TGFb1 decreased E-cadherin andincreased a-SMA (Fig. 4G). Taken together, these dataindicated that the loss of VDR triggered by TNFasensitized 4T1 cells to EMT. Similar results were obtainedin the experiments with MCF7 cells (Supplementary Fig.S1C and S1D).

VDR overexpression suppressed b-catenin in breastcancer cells in vitro and in vivoAccumulating evidence supports an important role for the

b-catenin pathway in the EMT process. Therefore, weinvestigated whether b-catenin pathway was involved inmediating the effects of VDR on EMT. We detected asignificant downregulation ofb-catenin inVDR-overexpres-sing 4T1 cells when compared with vector control (Fig. 5Aand B). Immunofluorescence staining of b-catenin in thesecells supported these results.Moreover, b-catenin expressionwas suppressed in both the nuclei and cytoplasm of VDR-overexpressing 4T1 cells (Fig. 5C). Correspondingly,

A

C

D

E

BControl

NSTu

mor

vol

ume

(mm

3 )

1,000

500

0

Tum

or n

odul

es (

per

lung

)

40

30

20

10

0

Con

trol

4T1-vectorcontrol

4T1-

vect

orco

ntro

l

4T1-VDR

4T1-

VD

R

Contro

l

4T1-

vecto

r

cont

rol4T

1-VDR

Contro

l

4T1-

vecto

r

cont

rol4T

1-VDR

Contro

l

4T1-

vecto

r

cont

rol4T

1-VDR

15

10

5

0Fol

d ch

ange

Figure 2. VDR upregulationsuppresses lung colonization in theorthotopic mouse model of breastcancer. Female BALB/c mice wereinjected into the fourth fat pad withstable VDR-overexpressing 4T1cells. WT or stable control vector4T1 cells served as control (n ¼ 5).Mice were sacrificed at the eighthweeks after injection and thenisolated primary tumor and lung foranalysis. A, representative imagesshow VDR expression in primarytumors. Scale bar, 100 mm. B, real-time PCR shows themRNA level ofVDR expression in primary tumors.C and D, bar plots show thestatistical results of the tumor sizesand metastatic nodules counts oflung in different groups. E, left, therepresentative pictures of themetastatic lung nodules in differentgroups (black arrows). Right,representative pictures ofhematoxylin and eosin (H&E)staining of malignant lung tissueunder 100� or 200�. Scale bar, 1mm/200 mm. The data are shownas mean � SEM. �, P < 0.05;��, P < 0.001.






suppression of b-catenin was detected at primary tumor sitesin the orthotopic mouse model of breast cancer generated bythe VDR-overexpressing 4T1 cells (Fig. 5D). We alsodemonstrated interaction between VDR and b-catenin inVDR-overexpressing 4T1 cells using immunoprecipitation(Fig. 5E). Furthermore, luciferase reporter assay data sug-gested that the TCF/LEF1 transcriptional activity wasrepressed in VDR-overexpressing 4T1 cells (Fig. 5F).

1,25(OH)2D3 protected against the loss of VDR andincreased sensitivity to pro-EMT stimuli induced byTGFb1 in 4T1 cells in vitro; 1,25(OH)2D3administration suppressed EMT and metastasis in amouse model of breast cancerIn view of the critical role of vitamin D, ligand of VDR, in

mediating the various effects of VDR,we further checked theeffect of 1,25(OH)2D3 on themigration in 4T1 cells in vitro

4T1-vector control

A B

C

E F

D

Control4T1-vector

control

Coculture days

Coculture days

VDR(long exposure) (short exposure)

E-cadherin

α-SMA

β-Actin

0 3 7

Coculture days

0 3 7

Coculture days0 3 7

Coculture days0 3 7

4T1-VDR

4T1-

vect

orco

ntro

l4T

1-V

DR

4T1-VDR

4T1-vector control0 1 3 5 7 0 1 3 5 7

4T1-VDR

VDR/E-cadherin/DAPI

4T1-vector control4T1-VDR

4T1-

vect

orco

ntro

l4T

1-V

DR

Contro

l

4T1-

vecto

r

cont

rol

4T1-

VDR

Contro

l

4T1-

vecto

r

cont

rol

4T1-

VDR

100

50

0

15

10

5

0

Mig

rate

d ce

lls p

er fi

eld

1,000

500

0

60

40

20

0Mig

rate

d ce

lls p

er fi

eld

Inva

sive

cel

ls p

er fi

eld

Inva

sive

cel

ls p

er fi

eld

Figure 3. Ectopic overexpression of VDR suppressesmetastatic ability and attenuates inhibition of E-cadherin induced by coculture of 4T1 breast cancer cellswith macrophages. We established stable control vector or VDR-overexpressing 4T1 cell lines, respectively. Transwell assay were performed to detectmetastatic ability of different cells. A, left, representative images of fields in the Transwell migration assay of WT, stable vector control, and stableVDR-overexpressing 4T1 cells, respectively. Scale bar, 100 mm. Bar plot in the right shows the quantification of the migrated cells per field shown in A. B,quantification of the amount of the cells invaded through theMatrigel in the Transwell invasion assay. C, stable vector control or VDR-overexpressing 4T1 cellswere cocultured with RAW 264.7 cells. Then, 1 � 105 4T1 cells were transferred to the upper chamber of the plates for Transwell assay. Eight hours later,the bottom chambers were fixed with methanol and stained with DAPI. Left, the representative images of the fields (scale bar, 100 mm) and the rightindicates the statistical results. D, quantification of results of Transwell invasion assay performed after the coculture of 4T1 and RAW 264.7 cells. E, stablevector control or VDR-overexpressing 4T1 cells were cocultured with RAW 264.7 cells for different time periods as indicated. Results ofWestern blot analysisshow the expression of VDR, E-cadherin, and a-SMA in 4T1 whole-cell lysate after coculture of different periods. F, immunofluorescence stainingusing primary antibody against E-cadherin and VDR in different 4T1 cells coculturedwith RAW264.7. Three independent experimentswere performed and 10fields per well were counted for each experiment. Scale bar, 50 mm. The data are shown as mean � SEM. �, P < 0.05; ��, P < 0.01; ��, P < 0.001.

Zhang et al.





as well as the EMT and metastasis in mouse model of breastcancer. Our data showed that 1,25(OH)2D3 prevented theinhibition of VDR induced by TNF-a in 4T1 cells as well asloss of VDR in the tumor tissue of mouse model of breastcancer (Fig. 6A and B). Migration assay revealed that 1,25(OH)2D3 rescued the increased migration ability of 4T1 onthe stimulation of TNFa and TGFb1 (Fig. 6C). Similarresults were obtained from the experiments with MCF-7cells (Supplementary Fig. S2). In vivo experiment of 1,25(OH)2D3 administration showed suppression of a-SMAand restore of E-cadherin in the group of 1,25(OH)2D3(Fig. 6D and E), as well as the suppression of the metastaticlung nodules when compared with vehicle control (Fig. 6F),with no influences on the tumor volume (Fig. 6G). Con-sistent with the results from the in vitro experiment, VDRexpression was protected and b-catenin was inhibited in the1,25(OH)2D3 administration group (Fig. 6H and I).

DiscussionA correlation between VDR polymorphism, breast cancer

susceptibility, and tumor angiogenesis has been previouslyidentified (10, 21). However, data investigating the rela-tionship between VDR and tumor metastasis remain quiterare.Ditsch and colleagues previously reported that theVDRexpression level correlates closely with the survival of patientswith breast cancer (13). In this study, we demonstrated thatdecreased VDR expression correlated with increased levels oftumor metastasis in both a tissue microarray of breast cancer

patient samples and a mouse model of breast cancer. Whilefurther evidence is required before the use of VDR as aprognostic indicator in breast cancer can be reliably adopted,our current data clearly demonstrate that VDR expression iscorrelated with the breast cancer metastatic potential.VDR has a well-recognized role in impairing proliferation

and inducing apoptosis of tumor cells (22). Unexpectedly,while metastasis of VDR-overexpressing tumors was inhib-ited in our mouse model, there was no reduction in primarytumor size. This might be explained by use of a routine dietwithout additional vitamin D supplementation or the sat-uration of VDR at the tumor initiation stage. In vitro, nodifferences between the migration and invasive capacities ofVDR-overexpressing and control cell lines were detected.However, the increased metastatic capacity conferred uponcells following coculture with macrophages was significantlyinhibited by VDR overexpression. Furthermore, our find-ings suggested that overexpression of VDR impaired thedecrease of E-cadherin and the increase of a-SMA normallyobserved in 4T1 cells following coculture withmacrophages.This loss of epithelial markers with a concurrent increase inmesenchymalmarkers is the key indicator for EMT (23, 24),which is regarded as a critical pathologic event in theinitiation and promotion of metastasis (1). Therefore, ourfindings suggest that the loss of VDR is likely required for theprometastatic effect of TAMs.The contribution of TAMs to various aspects of tumor

behavior has been extensively studied (25, 26). Therefore,we hypothesize that the inhibitory effect of VDR on tumor

A

E F G

B C D

CM/BM

TNFα (ng/mL)

TNFα (10 ng/mL)

TNFα (ng/mL)TGFβ (ng/mL)

VDR shRNATGFβ (ng/mL)

α-SMA

TNFα (ng/mL) 0 1 2 5 10

VDR

β-Actin

CM/BM 0 1/2

00

100

01

02

101

102

–0

+0

+1

+2

1/2

0 h 12 h 24 h 48 h

0 h 12 h 24 h 48 h

SPD304

VDR

β-Actin

VDR

E-cadherin

β-Actin

α-SMA

VDR

E-cadherin

β-Actin

VDR

β-Actin

Fol

d ch

ange

VDR

β-Actin

0

0 1 2 5 10

1/8 1/4 1/2

Control

Cocultu

re

with 4T1

4

2

0

1.0

0.5

0.0

Fol

d ch

ange

1.0

0.5

0.0Rel

ativ

e m

RN

A le

vel

Figure 4. TNFa inhibits VDR expression and promotes EMT initiated by TGFb1 stimulation. CM of RAW 264.7 cells after 48 hours of coculturing with 4T1 cellswere collected and mixed with elementary media with variable ratios. A, Western blot analysis reveals the downregulation of VDR expression with CM fromRAW 264.7 cells in a dose-dependent pattern. B, real-time PCR analysis reveals the upregulation of TNFa mRNA expression in RAW264.7 cells aftercocultured with 4T1 cells for 48 hours. C, 4T1 cells were cultured in CM from RAW 264.7 with pre-cocultured of 4T1 cells with or without SPD304, a specificinhibitor of TNFa. Western blot analysis shows that SPD304 rescues the VDR expression suppressed by CM from RAW 264.7 cells. D, 4T1 cells weretreatedwith 10 ng/mLof TNF-a for different timecourse. VDRexpressionwasdetected byWestern blot analysis.Quantification of results is shownat the right.Data presented are shown as mean � SEM and collected from 3 independent experiments. E, 4T1 cells were treated with 1, 2, 5, or 10 ng/mL ofTNFa for 48 hours. Western blot analysis was used to detect VDR expression. Quantification of results is shown at the right. The data are shown as mean� SEM. ��, P < 0.01; ��, P < 0.001. F, 4T1 cells were treated with 10 ng/mL TNF-a, with or without 1 or 2 ng/mL TGF-b1 for 48 hours. Western blotanalysis was performed using anti-VDR, E-cadherin, and a-SMA. G, 4T1 cells were treated with or without 1 or 2 ng/mL TGF-b1 for 48 hours after beingtransfected with VDR shRNA. Western blotting was used to detect expression of VDR, E-cadherin, and a-SMA. Our results shows that downregulationof VDR induced by TNFa or VDR shRNA has similar effect on promoting the inhibition of E-cadherin and induction of a-SMA initiated by TGFb1 in breastcancer cells.






metastasis might depend on the interruption of the criticalfeedback loop between TAMs and tumor cells. Given thatthe coculture system used in this study does not involvedirect cell–cell contact, we speculated that macrophage-derived cytokines in the CM are involved in the effect onVDR expression. Our data showed that treatment withspecific inhibitor of TNFa, an important proinflammatorycytokine, suppressed the inhibition of VDR expressioninduced by macrophage CM. Moreover, TNFa can inhibitVDR expression while enhancing the ability of TGFb1 tostimulate EMT in breast cancer cells, as determined by theinduction of a-SMA and the loss of E-cadherin. This is inaccord with previous research from our group and otherswhich has identified both the inhibitory effect of TNFa onVDR expression in different cell lines and the synergisticeffect of TNFa andTGFb1 onEMT induction (14, 27, 28).

Our current findings suggest that inhibition of VDR med-iates TNFa-mediated EMT and that this may be a keyunderlying mechanism in the interaction between TAMsand tumor cells.It is generally accepted that activation of VDR after

binding with the ligand, active vitamin D, although a fewresearchers have indicated the non–ligand-dependent effectof VDR (29, 30). Ellison and Engelhard and their colleagueshave reported the ligand-independent regulatory effect ofVDR on the vitamin D–responsive 24-hydroxylase promot-er and the direct transcriptional regulation of hairless byVDR, respectively. Whether the ligand-independent trans-activation also contributes to the effect of VDR on EMT orthe trace amount of active vitaminDpresents in the standardmedium intrigues the initial step of cascade activation ofVDR is still an unsolved puzzle in the study. However,

A C D

E F

B

Con

trol

4T1-

vect

orco

ntro

l

Control

Inpu

t

Rab

bit I

gGAn

ti-VD

R IP

Inpu

tR

abbi

t IgG

Anti-

VDR

IPIn

put

Rab

bit I

gGAn

ti-VD

R IP

Contro

l

VD trea

tmen

t

4T1-vectorcontrol 4T1-VDR 4T1-vector control

4T1-VDR

Rel

ativ

e lu

cife

rase

act

ivity

4T1-

VD

R

Con

trol

4T1-

vect

orco

ntro

l4T

1-V

DR

Contro

l

4T1-

vecto

r

cont

rol4T

1-VDR

Contro

l

4T1-

vecto

r

cont

rol

4T1-

VDR

1.0

0.5

0.0

Fol

d ch

ange

Contro

l

4T1-

vecto

r

cont

rol

4T1-

VDR

1.0

0.5

0.0

0.8

0.4

0.0

Fol

d ch

ange

VDR (shortexposure)

(longexposure)

E-catenin

VDR β-catenin

WB:β-catenin

β-Actin

VDR

β-Actin

Figure 5. Upregulation of VDR suppresses b-catenin expression both in vitro and in vivo. mRNA and protein samples were isolated separately fromWT, vectorcontrol, or VDR- overexpressing stable cells. A, real-time PCR assay for the inhibition of b-catenin mRNA expression in the VDR-overexpression cellscompared with vector control or WT 4T1 cells. B, Western blot analysis using the antibody against VDR and b-catenin shows the suppression of b-catenin inthe VDR-overexpressing cells compared with vector control or WT 4T1 cells. For VDR expression, data of different exposure time were presented. C,results of immunofluorescence staining of WT, vector control, and VDR- overexpressing stable cells using antibody against VDR or b-catenin. Scale bar, 50mm. D, orthotopic mouse model was established according to method described in Fig. 2. Left, the representative images of immunohistologic staining ofb-catenin expression in the tumor tissues. Scale bar, 100 mm. Right, the quantification results of b-catenin expression. E, detection of interaction betweenVDR and b-catenin in WT, vector control, and VDR-overexpressing 4T1 cells. Cell extracts were immunoprecipitated with VDR antibody–conjugatedDynabeads, followed by immunoblotting with antibody against anti-b-catenin. Immunoblotting of whole-cell lysates without immunoprecipitation was usedto detect protein expression throughout experiments. F, dual luciferase method was used to analyze the TCF/LEF1 transcriptional activity in vectorcontrol or VDR-overexpressing 4T1 cells treated with or without 10�7 mol/L 1,25(OH)2D3. The data are shown as mean � SEM. �, P < 0.05; ��, P < 0.01;��, P < 0.001.

Zhang et al.





further evidence is undoubtedly worthy especially consid-ering the "dependence receptors theory" which postulatesthat some receptors are active in the absence of their ligands(31). To further evaluate the role of active vitamin D in theprometastatic effect of inflammation on the metastasis ofbreast cancer cells, we performed a set of separated experi-ments. In vitro, treatment of 1,25(OH)2D3 inhibited themigration induced by TNFa combined with TGFb1 and

alleviated EMT as well as metastasis in the mouse model ofbreast cancer in vivo. Therefore, it is not hard to speculatethat the protective effect on VDR contributes to the inhib-itory role of 1,25(OH)2D3 in metastasis, at least partially.Activation of VDR was dependent on several signaling

pathways, which have been implicated in inflammation-related activation of EMT during tumor metastasis, includ-ing TGFb, Wnt, Notch, and Hedgehog. These pathways

A B

D

F G H I

E

C

Con

trol

VD tr

eatm

ent

Con

trol

VD

trea

tmen

t

Con

trol

VD

trea

tmen

tDAPI α-SMA Merge DAPI E-cadherin Merge

0.9

0.6

0.3

0.0

Fol

d ch

ange

3

2

1

0

0.4

0.0

Inte

rval

dist

ance

(cm

)

Fol

d ch

ange

0.8

0.4

0.0Fol

d ch

ange

TNFα (ng/mL) 0 0 10 10 10 10 10

0 10–6 0 10–9 10–8 10–7 10–6Vitamin D (mol/L)

TNFα (ng/mL) 0 0 10 10 10 10 100 10–6 0 10–9 10–8 10–7 10–6Vitamin D (mol/L)

TNFα (ng/mL)

0 h

48 h

24 h0 h

TGFβ (ng/mL)Vitamin D (mol/L)

0

00

10

00

10

20

10

210–7

TNFα (ng/mL)

TGFβ (ng/mL)Vitamin D (mol/L)

0

00

10

00

10

20

10

210–7

VDR

β-Actin

ControlVD treatment

ControlVD treatment

20

10

0

1,000

500

0Tum

or n

odul

es (

per

lung

) ControlVD treatment

Control

NS

Tum

or v

olum

e (m

m3 )

VD treatment ControlVD treatment

1.2

0.8

0.4

0.0Fol

d ch

ange

1.0

0.5

0.0Fol

d ch

ange

ControlVD treatment

Control VD treatment

VDR

β-Actin

Figure 6. Administration of vitamin D decreasesmetastasis of breast cancer cells both in vitro and in vivo.WT 4T1 cells were treatedwith 10 ng/mL TNFa aloneor TNFa plus pretreatment of 10�9, 10�8, 10�7, or 10�6 mol/L 1,25(OH)2D3. A, Western blot analysis was performed using antibody against VDR. Bottom,the quantification result of VDR expression. B, female Balb/c mice were injected into the fourth fat pad with 4T1 cells to establish the orthotopic mousemodel. 1,25(OH)2D3or vehicle controlwas given via intraperitoneal injection at a doseof 0.3mg/kg bodyweight per day (n¼ 5).Western blot analysis results ofVDR expression in the homogenate of tumor tissue. C, 4T1 cells were treated with 10 ng/mL TNFa plus 2 ng/mL TGFb1 with or without pretreatment of 10�7

mol/L 1,25(OH)2D3. Wound-healing assay compared the migration ability of 4T1 cells with different treatment. Bottom, quantification result of theinterval distance. D andE, representative images andquantification of immunofluorescence staining using antibody againsta-SMAandE-cadherin in the 1,25(OH)2D3 treatment and control groups. Scale bar, 100 mm. F, bar plots show the statistical results of the metastatic nodules counts of lung in differentgroups. G, bar plots show the statistical results of the tumor volume in different groups. H, representative images of immunohistologic staining using antibodyagainst b-catenin in the orthotopic mouse model treated with 1,25(OH)2D3 or vehicle control. Scale bar, 100 mm. I, quantification result of b-cateninexpression. Data are shown as mean � SEM. �, P < 0.05; ��, P < 0.01; ��, P < 0.001.






converge on a common set of transcription factors, includingSnail, Slug, Twist, ZEB1/2, and the SMADs, therebyfacilitating EMT (32, 33). b-Catenin is a downstreameffector of the Wnt pathway. Upon activation of Wntsignaling, b-catenin accumulates in the cytoplasm andtranslocates to the nucleus, influencing gene transcription(34). A previous study demonstrated that VDR knockdownenhanced b-catenin activation in colon cancer cells (35). Inthis study, we found that b-catenin expression was decreasedin both cultured 4T1 cells stably overexpressing VDR and intumor tissues from our VDR-overexpressing mouse model.Our data further showed that ectopic expression of VDRpromotes the formation of a complex with b-catenin andfurther inhibits the transcriptional activity of TCF/LEF1.This suggests that inhibition of the Wnt/b-catenin pathwaymay be the mechanism by which VDR suppresses tumormetastasis. However, the exact nature of the interactionbetweenVDR andb-catenin remains unclear. A recent studyhas also revealed that VDR binds SMAD3 target sites andreduces SMAD3 occupancy, thus affecting activation of theTGFb/SMAD pathway (23). This suggests that severalmechanisms may mediate the effect of VDR on tumormetastasis. Clearly, further study on the broader range ofpathways contributing to this effect is warranted.In summary, our findings demonstrate that VDR sup-

pression by TNFa may mediate the promotion of breastcancer metastasis by TAMs. This effect is likely related to therelief of Wnt/b-catenin pathway inhibition, thus facilitating

the EMT process. These results provide a new angle bywhich to view the relationship between EMT and the tumormicroenvironment. Furthermore, these findings suggest thatVDR ligands may be a potential therapeutic target for breastcancer metastasis.

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

Authors' ContributionsConception and design: X. TanDevelopment of methodology: Z. ZhangAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): Y. Zhang, Q. Guo, Z. Zhang, N. Bai, Z. Liu, M. Xiong, R. Xiang,X. TanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): Y. Zhang, X. TanWriting, review, and/or revision of the manuscript: Y. Zhang, Y. Wei, X. TanAdministrative, technical, or material support (i.e., reporting or organizing data,constructing databases): Y. Zhang, X. TanStudy supervision: R. Xiang

Grant SupportThe study was supported by a grant from the National Basic Research Program of

China 2011CB944003 (R. Xiang) and the National Natural Science Foundation ofChina 30900540 (X. Tan) and 81273331 (R. Xiang).

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be herebymarked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Received January 20, 2014; revised April 28, 2014; accepted April 29, 2014;published OnlineFirst May 12, 2014.

References1. Mallini P, Lennard T, Kirby J, Meeson A. Epithelial-to-mesenchymal

transition: what is the impact on breast cancer stem cells and drugresistance. Cancer Treat Rev 2014;40:341–8.

2. Gabrilovich DI, Ostrand-Rosenberg S, Bronte V. Coordinated regula-tion of myeloid cells by tumours. Nat Rev Immunol 2012;12:253–68.

3. Qian BZ, Pollard JW. Macrophage diversity enhances tumor progres-sion and metastasis. Cell 2010;141:39–51.

4. Tang X, Mo C, Wang Y, Wei D, Xiao H. Anti-tumour strategies aimingto target tumour-associated macrophages. Immunology 2013;138:93–104.

5. LuoY, ZhouH, Krueger J, KaplanC, LeeSH,DolmanC, et al. Targetingtumor-associated macrophages as a novel strategy against breastcancer. J Clin Invest 2006;116:2132–41.

6. Pardoll D. Metastasis-promoting immunity: when T cells turn to thedark side. Cancer Cell 2009;16:81–2.

7. Green CE, Liu T, Montel V, Hsiao G, Lester RD, Subramaniam S, et al.Chemoattractant signaling between tumor cells and macrophagesregulates cancer cell migration, metastasis and neovascularization.PLoS One 2009;4:e6713.

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

9. Trabert B,MaloneKE, Daling JR, DoodyDR, Bernstein L, UrsinG, et al.Vitamin D receptor polymorphisms and breast cancer risk in a largepopulation-based case-control study of Caucasian andAfrican-Amer-ican women. Breast Cancer Res 2007;9:R84.

10. Raimondi S, Johansson H, Maisonneuve P, Gandini S. Review andmeta-analysis on vitamin D receptor polymorphisms and cancer risk.Carcinogenesis 2009;30:1170–80.

11. MehtaRG,PengX, Alimirah F,MurilloG,MehtaR. VitaminDandbreastcancer: emerging concepts. Cancer Lett. 2012 Nov 8. pii: S0304-3835(12)00639-8.

12. Maruyama R, Aoki F, Toyota M, Sasaki Y, Akashi H, Mita H, et al.Comparative genome analysis identifies the vitamin D receptor geneas a direct target of p53-mediated transcriptional activation. CancerRes 2006;66:4574–83.

13. Ditsch N, Toth B, Mayr D, Lenhard M, Gallwas J, Weissenbacher T,et al. The association between vitamin D receptor expression andprolonged overall survival in breast cancer. J Histochem Cytochem2012;60:121–9.

14. Xiong M, Gong J, Liu Y, Xiang R, Tan X. Loss of vitamin D receptor inchronic kidney disease: a potential mechanism linking inflammation toepithelial-to-mesenchymal transition. Am J Physiol Renal Physiol2012;303:F1107–15.

15. Remmele W, Stegner HE. [Recommendation for uniform definition ofan immunoreactive score (IRS) for immunohistochemical estrogenreceptor detection (ER-ICA) in breast cancer tissue]. Pathologe 1987;8:138–40.

16. Stuelten CH, Cervoni-Curet FN, Busch JI, Sutton E, Webster JD,Kavalukas SL, et al. SDF-1alpha mediates wound-promoted tumorgrowth in a syngeneic orthotopicmousemodel of breast cancer. PLoSOne 2013;8:e60919.

17. Duan X, Xiao J, Yin Q, Zhang Z, Yu H, Mao S, et al. Multi-targetedinhibition of tumor growth and lungmetastasis by redox-sensitive shellcrosslinked micelles loading disulfiram. Nanotechnology 2014;25:125102.

18. Li X, Xu Y, Chen Y, Chen S, Jia X, Sun T, et al. SOX2 promotes tumormetastasis by stimulating epithelial-to-mesenchymal transition viaregulation ofWNT/beta-catenin signal network. Cancer Lett 2013;336:379–89.

19. Lu D, Chen S, Tan X, Li N, Liu C, Li Z, et al. Fra-1 promotes breastcancer chemosensitivity by driving cancer stem cells from dormancy.Cancer Res 2012;72:3451–6.

Zhang et al.





20. DeCraeneB,BerxG. Regulatory networks defining EMTduring cancerinitiation and progression. Nat Rev Cancer 2013;13:97–110.

21. Garcia-Quiroz J, Rivas-Suarez M, Garcia-Becerra R, Barrera D, Marti-nez-Reza I, Ordaz-Rosado D, et al. Calcitriol reduces thrombospon-din-1 and increases vascular endothelial growth factor in breast cancercells: implications for tumor angiogenesis. J Steroid BiochemMol Biol.2013 Oct 9. pii: S0960-0760(13)00193-3.

22. MoukayedM, Grant WB. Molecular link between vitamin D and cancerprevention. Nutrients 2013;5:3993–4021.

23. DingN, YuRT, SubramaniamN, ShermanMH,WilsonC, RaoR, et al. Avitamin D receptor/SMAD genomic circuit gates hepatic fibroticresponse. Cell 2013;153:601–13.

24. Wells A, Yates C, Shepard CR. E-cadherin as an indicator of mesen-chymal to epithelial reverting transitions during themetastatic seedingof disseminated carcinomas. Clin Exp Metastasis 2008;25:621–8.

25. Tang X. Tumor-associated macrophages as potential diagnostic andprognostic biomarkers in breast cancer. Cancer Lett 2013;332:3–10.

26. Balkwill FR, Mantovani A. Cancer-related inflammation: commonthemes and therapeutic opportunities. Semin Cancer Biol 2012;22:33–40.

27. MeldrumKK,Misseri R,Metcalfe P,DinarelloCA,Hile KL,MeldrumDR.TNF-alpha neutralization ameliorates obstruction-induced renal fibro-sis and dysfunction. Am J Physiol Regul Integr Comp Physiol 2007;292:R1456–64.

28. Bates RC, Mercurio AM. Tumor necrosis factor-alpha stimulates theepithelial-to-mesenchymal transition of human colonic organoids.MolBiol Cell 2003;14:1790–800.

29. Ellison TI, Eckert RL, MacDonald PN. Evidence for 1,25-dihydroxyvi-tamin D3-independent transactivation by the vitamin D receptor:uncoupling the receptor and ligand in keratinocytes. J Biol Chem2007;282:10953–62.

30. Engelhard A, Bauer RC, Casta A, Djabali K, Christiano AM. Ligand-independent regulation of the hairless promoter by vitamin D receptor.Photochem Photobiol 2008;84:515–21.

31. Mehlen P. Dependence receptors: the trophic theory revisited. SciSignal 2010;3:pe47.

32. Kalluri R, Weinberg RA. The basics of epithelial-mesenchymal transi-tion. J Clin Invest 2009;119:1420–8.

33. Talbot LJ, Bhattacharya SD, Kuo PC. Epithelial-mesenchymal transi-tion, the tumormicroenvironment, andmetastatic behavior of epithelialmalignancies. Int J Biochem Mol Biol 2012;3:117–36.

34. Arend RC, Londono-Joshi AI, Straughn JM Jr, Buchsbaum DJ. TheWnt/b-catenin pathway in ovarian cancer: a review. Gynecol Oncol2013;131:772–9.

35. Larriba MJ, Ordonez-Moran P, Chicote I, Martin-Fernandez G, Puig I,Munoz A, et al. Vitamin D receptor deficiency enhances Wnt/beta-catenin signaling and tumor burden in colon cancer. PLoSOne 2011;6:e23524.






2014;12:1181-1191. Published OnlineFirst May 12, 2014.Mol Cancer Res Yan Zhang, Quanjun Guo, Zhujun Zhang, et al. Tumor-Associated MacrophagesVDR Status Arbitrates the Prometastatic Effects of

Updated version

10.1158/1541-7786.MCR-14-0036doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2014/05/12/1541-7786.MCR-14-0036.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/12/8/1181.full#ref-list-1

This article cites 33 articles, 5 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://mcr.aacrjournals.org/content/12/8/1181To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-14-0036

http://mcr.aacrjournals.org/content/suppl/2014/05/12/1541-7786.MCR-14-0036.DC1

http://mcr.aacrjournals.org/content/12/8/1181.full#ref-list-1

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/12/8/1181


VDR Status Arbitrates the Prometastatic Effects of Tumor ...Signal Transduction VDR Status...

Documents

Transcript of VDR Status Arbitrates the Prometastatic Effects of Tumor ...Signal Transduction VDR Status...