Opposing Effects of Pigment Epithelium Derived Factor on ...BRins and 4T1-BR5 cells, by selection in...

Molecular and Cellular Pathobiology

Opposing Effects of Pigment Epithelium–Derived Factor onBreast Cancer Cell versus Neuronal Survival: Implication forBrain Metastasis and Metastasis-Induced Brain Damage

Daniel P. Fitzgerald1, Preeti Subramanian2, Monika Deshpande2, Christian Graves3, Ira Gordon3,Yongzhen Qian6, Yeva Snitkovsky5, David J. Liewehr4, Seth M. Steinberg4, Jos�e D. Palt�an-Ortiz5,Mary M. Herman5, Kevin Camphausen3, Diane Palmieri1, S. Patricia Becerra2, and Patricia S. Steeg1

AbstractBrain metastases are a significant cause of morbidity and mortality for patients with cancer, yet preventative

and therapeutic options remain an unmet need. The cytokine pigment epithelium–derived factor (PEDF) isdownregulated in resected human brain metastases of breast cancer compared with primary breast tumors,suggesting that restoring its expression might limit metastatic spread. Here, we show that outgrowth of largeexperimental brain metastases from human 231-BR or murine 4T1-BR breast cancer cells was suppressed byPEDF expression, as supported by in vitro analyses as well as direct intracranial implantation. Notably, thesuppressive effects of PEDF were not only rapid but independent of the effects of this factor on angiogenesis.Paralleling its cytotoxic effects on breast cancer cells, PEDF also exerted a prosurvival effect on neurons thatshielded the brain from tumor-induced damage, as indicated by a relative 3.5-fold reduction in the number ofdying neurons adjacent to tumors expressing PEDF. Our findings establish PEDF as both ametastatic suppressorand a neuroprotectant in the brain, highlighting its role as a double agent in limiting brainmetastasis and its localconsequences. Cancer Res; 72(1); 144–53. �2012 AACR.

Introduction

Brain metastasis, the most common cause of intracranialtumors in adults, occurs in approximately one quarter of adultpatients with cancer, and represents a significant cause ofmorbidity. The incidence of brain metastasis appears to beincreasing. One factor thought to be contributing to thisincrease is improvement in the systemic treatment of cancerleading to prolonged survival and thereby allowing the emer-gence of brain metastases (1–3). The brain is considered a"sanctuary site" because of restricted drug access caused by the

blood–brain barrier (4) and the inflamed brain microenviron-ment (5). The prognosis for patients diagnosed with brainmetastasis is poor. Palliative treatments include corticoster-oids, surgery, whole brain radiation therapy, and stereotacticradiosurgery (SRS; refs. 1, 3).

While a metastasis to any site is devastating, metastases tothe brain are particularly feared for their neural complications.The majority of patients with brain metastasis reportimpairment in neurocognitive function. Other symptomsinclude headache, focal weaknesses, seizures, ataxia, andspeech deficits (1, 2, 6). Permanent, progressive cognitivedeficits occur in some patients with brain metastasis, rangingfrom mild memory loss to severe dementia (2). These deficitsresult not only from progression of the metastatic disease butalso as side effects of treatment, including both systemicchemotherapy ("chemobrain"; ref. 7) andwhole brain radiationtherapy (8). Profiling of resected brain metastases and thedevelopment of mouse models have greatly expanded ourknowledge of acquisition of brain metastatic potential bycancer cells; however, almost nothing is known about theonset of neuronal damage that leads to neurologic symptoms.

A gene expression analysis of resected breast cancer speci-mens revealed the downregulation of pigment epithelium–derived factor (PEDF) expression in brain metastases byapproximately 14-fold relative to primary tumors (9). Normalbreast epithelial cells express high levels of PEDF protein. Lossof PEDF expression from these cells is correlated with cancerprogression, and PEDF expression is inversely related withsurvival outcome in patients with breast cancer (10). The gene

Authors' Affiliations: 1Women's Cancer's Section, Laboratory of Molec-ular Pharmacology, Center for Cancer Research, National Cancer Institute,2Section on Protein Structure and Function, Laboratory of Retinal Cell andMolecular Biology, National Eye Institute, 3RadiationOncology Branch and4Biostatistics andDataManagement Section, Center for Cancer Research,National Cancer Institute, 5Section on Neuropathology, Clinical BrainDisorders Branch, National Institute ofMental Health, Bethesda, Maryland;and 6Laboratory Animal Sciences Program, Science Applications Interna-tional Corporation, National Cancer Institute, Frederick, Maryland

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

Current address for D.P. Fitzgerald: Otsuka Pharmaceuticals, Rockville,Maryland.

Corresponding Author: Patricia S. Steeg, Laboratory of Molecular Phar-macology, NIH, Building 37, Room 1122, Bethesda, MD 20892. Phone:301-402-2732; Fax: 301-402-8910; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-11-1904

�2012 American Association for Cancer Research.

CancerResearch

Cancer Res; 72(1) January 1, 2012144

on November 26, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

expression trend suggested a further downregulation of PEDFin breast cancer cells associated with the acquisition of brainmetastatic potential.PEDF, a member of the serpin family of proteins, was

discovered as a secreted factor from human retinal pigmentepithelial cells with a potent neuronal differentiation activity(11). Numerous subsequent studies have shown a remarkableability of PEDF to protect neurons from a variety of insults.(12–16). PEDF shows a broad expression pattern, includinghigh levels in the brain and in blood (17). In addition to itsneuroprotective activity, PEDF has been implicated as aregulatory factor in numerous pathways, including lipidmetabolism (18), stem cell renewal (19), modulation of theinflammatory response (20), antiangiogenesis (21), and tumorsuppression (reviewed in refs. 22, 23). PEDF is under transla-tional investigation for use in a variety of diseases includingmacular degeneration (24) and neurodegeneration (16).Increasing lines of evidence point to multiple functionaldomains in the PEDF protein with either antiangiogenic orneuroprotective capacities (25–27), as well as multiple distinctcellular receptors (28–31).The joint tumor-suppressive and neuroprotective activities

of PEDF, combined with the observation that it is down-regulated in human brain metastases, suggested that it mighthave therapeutic value in the context of brain metastasis.Herein, we have used experimental brain metastasis andintracranial injection models to show that PEDF acts simul-taneously as a prodeath molecule for cancer cells and as aprosurvival molecule for peritumoral neurons.

Materials and Methods

Cell linesThe humanMDA-MB-231 brain-tropic cell line (231-BR) and

murine 4T1-BR5 brain-tropic line were described previously(32, 33). The 231-BR line was obtained from Toshiyuki Yoneda,University of Texas, San Antonio, TX, and was authenticatedby our laboratory to the American Type Culture CollectionMDA-MB-231 line by short-tandem repeat profiling. The4T1-BR5 line was obtained from Suyun Huang, MD AndersonCancer Center, Houston, TX. R28 retinal progenitor cellswere obtained from Gail Seigel, University of Buffalo, NY (34).

Subcloning and production of stable cell linesThe coding region for the human PEDF gene was a gift from

the laboratory of Marta Garcia-Diez, CIEMAT, Madrid, Spain.The coding region, with two tandem FLAG tags added to its C-terminus, was subcloned into the lentiviral expression vectorspCDH-CMV-MCS-EF1-Puro and pCDH-EF1-MCS-T2A-Puro(SBI). PEDF was also subcloned into the pJC13-1 vector (35),a gift from the laboratory of Gary Felsenfeld, NIDDK, Bethesda,MD. Additional details of DNA subcloning can be found inSupplementary Methods. Stable cell lines were generated in231-BR cells by selection in 500mg/mLof neomycin, and in 231-BRins and 4T1-BR5 cells, by selection in 2 to 30 mg/mL ofpuromycin, for at least 2 weeks. Cells were subsequentlymaintained in the absence of selective agent and showed noloss of transgene expression.

Purification of PEDF, real-time label-free cellularanalysis, and Western blot analysis

Recombinant human PEDF (or PEDF-FLAG) was purified aspreviously published (36). Cell impedance wasmeasured usinga real-time microelectronic sensor system (ACEA/Roche) asdescribed previously (37). Cell viability was measured using aCellTiter-Glo viability assay kit (Promega) following the man-ufacturer's instructions. Data from two ormore replicateswereaveraged. Western blot analyses were conducted according tostandard protocols. The following antibodies were used: anti-FLAG (Sigma), anti-GAPDH (Cell Signaling Technology),mouse anti-human PEDF (Millipore), rabbit anti-PEDF (Bio-Products MD), and anti-a-tubulin (Calbiochem), diluted1:1,000 to 1:5,000. Secondary antibodies (Santa Cruz) werediluted 1:2,000 to 1:5,000.

Hematogenous mouse models of brain metastasisExperiments were carried out under an approved Animal Use

Agreement with the NCI. Under isoflurane/O2 anesthesia, 5- to7-week-old female NCr nu/numice (Charles River Laboratories)were inoculated in the left cardiac ventricle with 1.75 � 105

human 231-BR cells and immunocompetent BALB/c mice wereinjected with 5 � 104 murine 4T1-BR5 cells. Mice were eutha-nized under CO2 asphyxiation at time points correlated with theonset of weight loss, ataxia, and/or paralysis. Brains, bisectedalong the sagittal plane, were processed for cryosectioning aspreviously described (38). Ten-micrometer cryosections wereprepared and processed for histology and immunofluorescence.To quantify the number of metastases, 10 sagittal sections,spaced every 300 mm, were cut through one hemisphere andhematoxylin and eosin (H&E) stained.Metastaseswere countedat 50� magnification using an ocular grid. Any metastasis thatmeasured more than 300 mm in any direction was considered alarge metastasis, as previously published (38).

Intracranial model of brain metastasisSix- to eight-week old female, NCr nu/numice (Charles River

Laboratories) were used. Experiments were carried out underanapprovedAnimalUseAgreementwith theNCI.Mice, dividedinto 3 groups to receive intracranial injections of PBS, vectorcontrol 231-BRins cells, or PEDF-231-BRins cells, were anes-thetized with ketamine (83 mg/kg) and xylazine (8.3 mg/kg).Cells were injected into the right caudate nucleus as previouslydescribed (39). Mice, anesthetized as above, were perfusionfixed with approximately 10 mL of PBS followed by approxi-mately 10 mL of 4% paraformaldehyde prior to extirpation ofthe brain 0, 1, or 3 days postimplantation. Whole brain wasimmediately frozen in optimum cutting temperature. Serial10-mm cryosections were prepared along the horizontal planestarting from the dorsal surface of the cerebral cortex andprogressing to a depth of 3 mm or above. H&E sections wereprepared every 100 mm.

Histologic and immunofluorescence analysisFor Bielschowsky silver impregnation, 10-mm frozen sec-

tions were fixed in 10% phosphate-buffered formalin for 24hours at 37�C or 48 hours at room temperature and stainedusing Sigma reagents. Fluorojade-B (Histochem) staining

PEDF Suppresses Brain Metastasis and Neuronal Damage

www.aacrjournals.org Cancer Res; 72(1) January 1, 2012 145



was conducted similar to the published method (40), omittingsodium hydroxide treatment and staining in a final dye con-centration of 0.001%. Immunofluorescence analyses were con-ducted as previously described (41). The following antibodieswere used: anti-CD31 (BDPharmingen), anti-cleaved caspase-3(Cell Signaling Technology), anti-FLAG (Sigma), anti-humanmitochondria (Thermo Scientific), anti-Ki67 (Vector Labora-tories), and anti-human PEDF (Millipore). Secondary antibo-dies were conjugated to Alexa Fluor 488 and 568 (MolecularProbes). Nuclei were visualized with 200 mg/mL DAPI (40,6-diamidino-2-phenylindole, dilactate) stain (Molecular Probes).

Numbers of cleaved caspase-3, FLAG, Fluorojade-B, Ki67, orPEDF-FLAG–positive cells or CD31-positive blood vessels werecounted in images using DAPI to counterstain all nuclei andanti-human mitochondria to label human breast cancer cells.Ten sections, approximately 100mmapart, were collected fromeach mouse for Fluorojade analysis. Five sections, approxi-mately 200 mm apart, were collected for analysis of Ki67 andcleaved caspase-3 in the intracranial model.

Statistical analysisA variety of ANOVAs were conducted on the raw or trans-

formed data (as indicated by the Box–Cox transformation).Samples from experimental units were averaged and thesquare root of the sample size was used as a weight insubsequent analyses. One-way ANOVAs were used to analyzesimple fixed-effect models. Two- or three-factor factorialANOVAs were used to analyze more complex fixed-effectmodels. Mixed models were used as necessary to modelrandom variation due to repeated experiments or other ran-dom effects. Analysis of covariance was used to account forimportant covariates. Linear regression was used to estimategrowth rates of individual mice. Finally, for all ANOVAs,residuals were examined for normality and homogeneity andresiduals were partitioned if found to be heterogeneous. TheWilcoxon rank-sum test was used to compare distributions ifthe data were not normally distributed and/or the sample sizeswere very small. For pairwise comparisons, Holm method wasused to adjust pairwise P values.

Results

Effects of PEDF on cultured breast cancer cells andneural retinal precursors

The effects of PEDF on two "triple-negative" breast cancercell lines were examined in vitro. Cells were analyzed in real-time by measuring cell impedance and at endpoint by mea-suring biomarkers of live cells to confirm cell viabilitymonitoring. At high cell density, 4T1-BR5 cells, a brain-tropicderivative of the 4T1 murine cell line, remained viable forseveral days in the absence of serum, but addition of increasingconcentrations of purified recombinant human PEDF progres-sively decreased 4T1-BR5 cell viability (Fig. 1A; SupplementaryFig. S1A and S1B). Comparison of the 0, 10, and 100 nmol/Ldoses using the exact Jonckheere–Terpstra test showed asignificant diminution of viability by PEDF (P ¼ 0.022; Fig.1B); comparison of vehicle versus PEDF for 4T1 cells (P¼ 0.022by the same test; Supplementary Fig. S1A). Similar trends were

also observed using 231-BR cells, a brain-tropic derivative ofthe MDA-MB-231 cell line (Fig. 1C and D).

We also tested the prosurvival activity of PEDF on neuralprogenitor cells. As illustrated in Fig. 1E, low-density rat retinalprogenitor (R28) cells progressively died within a day of serumdeprivation, but PEDF significantly increased R28 cell viabilityin a dose-dependent manner. Comparison of the 0, 10, and 100nmol/L doses showed a significant protective effect of PEDF(P ¼ 0.022; Fig. 1F). PEDF protein altered to include two-tandem FLAG-epitope tags on the C-terminus of the proteinmaintained its biologic activity, showing a statistical trend inreduced viability using the exact Jonckheere–Terpstra test(Supplementary Fig. S1C and S1D; P ¼ 0.33 for PEDF, 0.067for PEDF-FLAG). Thus, PEDF appears to have direct butopposing effects on the viability of breast cancer cells andnormal neural retinal progenitor cells.

Transient PEDF expression suppressed proliferation ofbrain metastases in two breast carcinoma models

In agreement with the observation that PEDF expression isdownregulated in human breast cancer brain metastases,neither of the brain metastatic breast cancer cell lines haddetectable expression of PEDFprotein (Fig. 2A, left column anddata not shown). Human 231-BR cells were transduced with ahuman PEDF transgene to produce stable clones. The FLAG-tagged PEDF protein was efficiently secreted from the cancercells and was detectable as a single band of 50 kD by Westernblot analysis with both anti-PEDF (Fig. 2A) and anti-FLAGantibodies (not shown).

While PEDF exhibited negative effects on breast cancercell viability in the absence of serum (Fig. 1), PEDF expres-sion appeared well tolerated when cells were cultured incomplete media, showing no decrease in anchorage-depen-dent growth when compared with vector control clones oruntransfected 231-BR cells (Supplementary Fig. S2A). Nordid the PEDF-expressing cells show a decrease in migrationacross a collagen-coated membrane in a Boyden chamberassay (Supplementary Fig. S2B). PEDF protein expressionappeared to be relatively uniform across cell populations asvisualized by immunofluorescence and could be maintainedin the absence of selection agent for multiple passageswithout noticeable changes in PEDF protein expression(data not shown).

To analyze the effect of PEDF on the brain metastaticpotential of breast cancer cells, 3 vector control 231-BR clonesand 3 PEDF-231-BR clones were injected into the left cardiacventricle of immunocompromised mice. The number of brainmetastases was quantified by histology of step sections throughone brain hemisphere of eachmouse 26 to 28 days later (Fig. 2B).PEDF expression had no significant effect on the number ofmicrometastases (<300 mm in a single direction) or leptome-ningeal metastases. However, the number of large (>300 mm)parenchymalmetastases formed fromPEDF-231-BR cells (mean¼ 11.9; n ¼ 60 mice for the three combined PEDF-expressingclones) was decreased to 66% of controls (mean ¼ 18.1; n ¼ 48mice for the three combined control clones; P ¼ 0.001).

Staining of tissue sections with anti-FLAG revealed thatonly 1% to 2% of tumor cells (352 of 22,092) retained PEDF

Fitzgerald et al.

Cancer Res; 72(1) January 1, 2012 Cancer Research146



expression after 26 days in vivo. Metastases were costainedfor PEDF-FLAG and Ki67 (Fig. 2C), and the numbers ofPEDF-FLAG, Ki67, and double-positive cells were quantified(Fig. 2D). As previously published (41), Ki67 staining wasconfined to metastatic tumor cells and absent in the sur-rounding brain cells (Fig. 2E). The majority (59%) of cells(7,812 of 13,317) were proliferating in control metastases(Fig. 2D, left column). In the mice inoculated with PEDF-231-BR cells, the proliferation rate among the subset of metas-tases in which no PEDF-expressing cells were visible was notsignificantly different from controls (57%; 8,387 of 14,732cells). In contrast, the rate of proliferation (42%; 3,133 of7,360 cells) was significantly reduced in the subset of metas-tases with mosaic PEDF expression (Fig. 2D, middle col-umn). There was an even greater decrease in proliferation(12%; 42 of 352 cells) among the individual cells whichmaintained PEDF expression (PEDF-FLAG and Ki67 dou-ble-labeled cells; Fig. 2D, right column). The reductions inproliferation, both in PEDF-expressing cells themselves and

in mosaic metastases containing a minority population ofPEDF cells, were highly significant (P ¼ 0.002 vs. control).

Vascular density was examined in control and PEDF brainmetastases. Blood vessels were stainedwith anti-CD31, and theblood vessel density was counted both in brainmetastases andin an adjacent region of brain devoid of tumor cells (Fig. 2F). Inagreement with previously published results (42), brain metas-tases in the 231-BR model had a reduced vascular densityrelative to the surrounding normal brain (P ¼ 0.0005). PEDFexpression did not significantly alter the number of bloodvessels in brain metastases (P ¼ 0.37, n ¼ 6 mice per group).This suggests that the observed reduction in proliferation andformation of large brain metastases is not due to an antiangio-genic effect of PEDF.

In an attempt to increase the fidelity of PEDF expression, asecond expression construct was engineered. Transcription ofPEDF was placed under control of the human EF1a promoter,and the transgene was flanked with duplicate transcriptionalinsulator elements derived from the chicken heat shock 4 locus

Figure 1. Opposing effects of PEDFon cultured breast cancer cells andneural retinal progenitor cells in vitro.Real-time cell impedance andendpoint cell viability assays wereconducted on 4T1-BR5 and 231-BRbreast cancer cells and R28 neuronalcells. After attachment in completemedia for 4 to 20 hours, cells weredepleted of serum for 4 hours for4T1-BR5, 8 hours for 231-BR, and0 hour for R28 cells before addinghuman recombinant PEDF protein(concentrations indicated). Cellimpedance measured the cell indexas a function of time (A, C, E; errorbars indicate SD between replicatewells); two experiments were carriedout for each cell line. Cell viability wasmeasured at the endpoint (120, 80,and 76 hours, respectively) usingintracellular ATP as biomarker of livecells (B, D, F). Dot plots from 2experiments quantify relative cellviability at the 0, 10, and 100 nmol/LPEDF doses. PEDF significantlyaltered cell viability in each test,P ¼ 0.022 by exact Jonckheere–Terpstra trend test. A and B,4T1-BR5 murine mammarycarcinoma cells. C and D, 231-BRhuman breast cancer cells. E and F,R28 neural retinal precursors.

4

3

2

1

0

120

100

80

60

40

20

0

120

100

80

60

40

20

0

5,000

4,000

3,000500

300

100

3

2

1

0

5

4

3

2

1

0

0 24 48 72

Time (h)

No

rma

lize

ed

ce

ll in

dex

No

rma

lize

ed

ce

ll in

dex

No

rma

lize

ed

ce

ll in

dex

96

0 24 48 72

Time (h)

0 24 48 72

Time (h)

96

1200 10 100

PEDF

A B

C D

E F

PEDF (nmol/L)

0 10 100

PEDF (nmol/L)

0 10 100

PEDF (nmol/L)

Re

lative

ce

ll via

bili

ty (

%)

Re

lative

ce

ll via

bili

ty (

%)

Re

lative

ce

ll via

bili

ty (

%)

0

10 nmol/L

100 nmol/L

PEDF

10 nmol/L

0

100 nmol/L

PEDF

10 nmol/L

0

100 nmol/L





(35). This construct was used to express PEDF in the murine4T1-BR5 brain metastatic cell line (Fig. 3A). PEDF expressionappeared to be uniform and stable among clonal populationsovermany generations in vitro and had no effect on anchorage-

dependent growth of these cells (Supplementary Fig. S2C). Theexpression construct is illustrated in Supplementary Fig. S2D.

Control and PEDF-expressing clones of 4T1-BR5 cellswere inoculated into the left ventricle of syngeneic,

A

C

B

D

E F

1 2 3 4

PEDF

Secreted

PEDF

α-Tubulin

PEDFControl

PEDFControl

Controlmetastases(no PEDF)

PEDF±mosaic

metastases

PEDF+metastatic

cells

Brain metastases Normal brain

Larg

e p

are

nchym

al bra

in

meta

sta

ses p

er

mouse ±

SE

M

% K

i67-p

ositiv

e c

ells

Mean n

um

ber

of blo

od

vessels

± S

EM

20

15

10

5

0

70

60

50

40

30

20

10

0

80

70

60

50

40

30

20

10

0

P = 0.001

P = 0.002

NS

P = 0.002

FLAG (PEDF)DAPI Ki67

Ki67DAPI Human-mito

Figure 2. Transient PEDF expression reduced the proliferation of 231-BR breast cancer cells colonizing the brain. A, Western blot analysis of PEDF proteinexpression of a vector control clone (lane 1) and stable clones expressing a humanPEDF transgene (lanes 2–4) in 231-BR cells;a-tubulin is shownas a loadingreference. PEDF secreted into culture medium conditioned for 24 hours is shown. B, number of large metastases (�SEM) formed in the brain parenchyma26 to 28dayspost–intracardiac injection of breast cancer cells. The combined results of 2 experimentswith a total of 48mice injectedwith 3 control clones and60 mice with 3 PEDF-231-BR clones are shown. C, immunofluorescence for PEDF-FLAG (anti-FLAG, red) and proliferation marker (anti-Ki67, green)in a representative section ofmousebrain (counterstainedwithDAPI, blue) 28 days postinjection of PEDF-231-BRcells. D, percentageof Ki67-positive cells inbrain metastases 28 days postinjection was quantified. Left column, control 231-BR metastases; middle column, metastases from PEDF-231-BR cellscontaining at least one PEDF-expressing cell (mosaic); right column, PEDF-231-BR cells positive for both PEDF and Ki67. Results from 6 mice injected withcontrol 231-BR cells and 6 mice injected with PEDF-231-BR cells pooled from 3 clones of each and from 2 experiments. E, immunofluorescence for 231-BRcells (anti-human mitochondria, green) and anti-Ki67 (red) in a representative mouse brain with control metastases. F, numbers of CD31-positiveblood vessels were counted in brain metastases (gray bars) and in adjacent normal brain (black bars) in 6 mice from each group, with either control orPEDF-231-BR cells.

Fitzgerald et al.




immunocompetent BALB/c mice. The number of brainmetastases was quantified after 9 days (Fig. 3B). The insu-lator construct did not improve transgene stability in vivo.Similar to the results with 231-BR cells, PEDF expression wasrapidly silenced in the syngeneic mouse model, and noPEDF-positive cells were visible at the endpoint (data notshown). However, transient PEDF expression in 4T1-BR5cells lead to a significant reduction in micrometastases(median ¼ 73) as compared with controls (median ¼ 432,P ¼ 0.018) and a significant reduction in large brain metas-tases (median ¼ 0 vs. 8, P ¼ 0.005).To determine whether the tumor microenvironment influ-

enced either the stability of PEDF expression or its effects onbreast cancer cell proliferation, we implanted breast cancercells into the mammary fat pad. The insulator construct wasused to generate stable polyclonal (231-BRins) control andPEDF-expressing populations in 231-BR cells (SupplementaryFig. S2E). PEDF expression was relatively uniform across thepopulations in vitro (Supplementary Fig. S2F). PEDF expres-sion significantly slowed mammary tumor growth rate(P¼ 0.005, two-sample t test; Supplementary Fig. S3A). Similarto the experimental brain metastasis models, very few cells atthe endpoint maintained PEDF expression. However, the rare

PEDF-expressing cells were Ki67 negative (Supplementary Fig.S3B). Blood vessel density was not significantly different (P ¼0.55) between the two groups, suggesting that the slowergrowth of PEDF tumors was not due to anti-angiogenesis(Supplementary Fig. S3C). Thus, using two models of breastcancer metastasis, and in two microenvironments, despiteachieving only transient PEDF expression, a growth-suppres-sive effect of PEDF was repeatedly observed.

PEDF expression rapidly inhibits proliferation andactivates apoptosis of breast cancer cells in the mousebrain

To study both the cancer-suppressive and the neuroprotec-tive effects of PEDF simultaneously, metastatic breast cancercells were directly implanted into mouse brains. A time coursestudy was conducted to determine the stability of PEDFprotein expression in vivo. Five hours postimplantation,PEDF-FLAG expression was visible in the majority of PEDF-231-BRins cells but was lost in many cells 24 hours postim-plantation andwas present in less than 1% of the cells by 3 dayspostimplantation (Fig. 4A).

To determine the effect of PEDF on the viability of the brainmetastatic tumor cells, mice were necropsied 24 hours post-implantation of control or PEDF-expressing tumor cells. Athird group of mice received an intracranial injection of salineto determine the effects of the stab wound on the mouse brain.Serial sections were prepared, starting from the needle entrysite along the dorsal surface of the brain and progressing to thebottom of the lesion, at a depth of approximately 3 mm.Representative H&E-stained sections are shown in Fig. 4B.Five step sections, evenly spaced over a length of approxi-mately 1 mm (which spanned most of the depth of theimplanted tumor cells), were stained for either Ki67 (Fig. 4Cand D) or cleaved caspase-3 (Fig. 4E and F). The majority (64%,17,658 of 27,625 cells) of the implanted 231-BRins cells incontrol lesions were Ki67 positive (proliferating; Fig. 4C, left).In PEDF-expressing lesions (Fig. 4C, right), proliferation wasreduced to aminority population (37%, 5,278 of 14,186 cells, P <0.0001). In control 231-BRins tumors (Fig. 4E, left), apoptoticcells expressing cleaved caspase-3 were rare (4%, 1,286 of28,763 cells). In contrast, more than a third (39%) of the cellsin PEDF-231-BRins lesions (5,596 of 14,461 cells; P < 0.0001 vs.control) were undergoing apoptosis (Fig. 4E, right). Thus, whilePEDF expression was well tolerated by breast cancer cells invitro, it has a rapid and profound suppressive effect on tumorviability in the brain. Both control (Supplementary Fig. S4A andS4B) and PEDF tumors (Supplementary Fig. S4C and S4D)showed no alterations in their vasculature, suggesting thatthese effects are independent of the antiangiogenic effects ofPEDF.

PEDF protects neurons in the brain from damageinduced by tumor cells

In addition to the direct effect of PEDF on tumor cells, theintracranial model permitted an examination of the effectsof secreted PEDF on the brain. Silver stain was used tovisualize the morphology of brain neurites and was evalua-ted independently by two neuropathologists. Adjacent to

A

B

1 2 3 4

PEDF

Secreted

PEDF

PEDFControl

P = 0.005

Larg

e b

rain

meta

sta

ses

per

mouse

20

15

10

5

0

α-Tubulin

Figure 3. PEDF reduced the number of large 4T1-BR5 metastases in asyngeneic model of experimental brain metastasis. A, Western blotanalysis of PEDF protein expression in a control clone ofmurine 4T1-BR5cells (lane 1) and in stable clones expressing various levels of PEDF(lanes 2–4); a-tubulin was used as a loading control. Secreted PEDF inconditionedmedium from the 4T1-BR5 cells is shown. B, number of largemetastases formed in the brains of mice 9 days after injection of control4T1-BR5 cells (n ¼ 19 mice) or PEDF-4T1-BR5 cells (n ¼ 10 mice).Each dot represents one mouse.





control 231-BRins lesions, the bundles of nerve fibers werefrequently disorganized, with obvious thickened and thinnedneurites indicative of neuronal damage (Fig. 5A, middle).A saline-only injection confirmed that neuronal damageoutside of the lesion was due primarily to the tumor cellsrather than the injection process (Fig. 5A, left). Far lessdamage was apparent adjacent to PEDF-231-BRins tumors,as compared with control tumors (Fig. 5A, right).

As the silver stains were only qualitative, a second stainfor neuronal damage was used, Fluorojade-B (Fig. 5B).Fluorojade-B has been shown to label damaged brain neu-rons specifically. This dye appears to label neurons under-going either necrotic or apoptotic death and has been usedto analyze brain damage resulting from a variety of insults(40, 43–47). Fluorojade-B staining was not visible aroundbreast cancer micrometastases but became apparent in

advanced disease in the hematogenous brain metastasismodel (Supplementary Fig. S5).

To examine the extent of damage in the intracranial model,10 step sections spanning approximately 1 mm of the lesionwere stained and counted to determine the number of injuredneurons surrounding the lesions (Fig. 5C). In agreement withthe silver stains, a small number of damaged neurons (200 perbrain) were evident surrounding the stab wound in salineinjection controls (Fig. 5B, left). There was a 15-fold increasein damaged neurons (2,950 per brain; P < 0.0001 vs. salinecontrols) when control 231-BR cells were implanted (Fig. 5B,middle). The majority of this damage was suppressed by PEDFexpression in the implanted cells (840 per brain; P ¼ 0.008 vs.control tumor; Fig. 5B, right). Moreover, PEDF appeared todecrease the area of the brain impacted by the lesion (Fig. 5D).The farthest Fluorojade-positive neurons were only a few cell

A

B

C D

E F

PEDF tumor day 0

Control tumor PEDF tumor


% K

i67-p

ositiv

e c

ells

± S

EM

% C

leave

d-c

aspase-3

–

positiv

e c

ells

± S

EM


Saline injection

PEDF-FLAG DAPI

Human-mitochondriaDAPIKi67

Human-mitochondriaDAPICleaved caspase-3

Day 1 Day 3

60

40

20

0

45

30

15

0

Control PEDF

Control PEDF

P < 0.0001

P < 0.0001

Figure 4. PEDF rapidly impactedthe proliferation and viability ofbreast cancer cells implantedin the mouse brain. A,immunofluorescence for PEDF-FLAG protein (anti-FLAG, red) inPEDF-231-BRins tumors 5 hours(left), 24 hours (middle), and 72hours postimplantation (right).Nuclei are counterstained withDAPI (blue). B, H&E-stainedsections of mouse brains withcontrol 231-BRins tumor cells(left), PEDF-231-BRins tumorcells (middle), and a salinecontrol injection (right) 24hours postimplantation. C,immunofluorescence for aproliferation marker (anti-Ki67,green) in control lesions (left) and inPEDF lesions (right) 24 hourspostimplantation. Xenograft breastcancer cells were colabeled forhuman-specific mitochondria (red)and nuclei counterstained withDAPI (blue). D, percentage of Ki67-positive cells in control and PEDFlesions after 24 hours. E, apoptoticcells expressing cleaved caspase-3 (green) in control (left) and inPEDF-231-BRins lesions (right) 24hours postimplantation. F,percentage of cleaved caspase-3–positive cells in control and PEDFlesions after 24 hours. Five miceinjected with control cells and 5mice injected with PEDF-231-BRins cells were averaged from 2experiments for these analyses.

Fitzgerald et al.




lengths (mean¼ 86mm) from the stabwound in saline-injectedbrain. Damage around 231-BRins tumors extended muchfurther (mean ¼ 269 mm; P ¼ 0.0003 vs. saline controls). Theradius of damage around PEDF-231-BRins tumors was signif-icantly smaller than the damage around control tumors (mean¼ 145 mm; P ¼ 0.009). These results indicate that PEDF cansimultaneously act to promote survival in neurons whileactivating cell death in cancer cells in the brain.

Discussion

The effects of PEDF on breast cancer cells in the brain wereanalyzed using two brain-tropic cell lines, two expressionconstructs, and several routes of tumor cell inoculation. Theoutgrowthofbreast cancer cells into largebrainmetastaseswassuppressed in both a xenograftmodel and in a syngeneicmouse

model of experimental brain metastasis. PEDF also slowed thegrowthof tumors implanteddirectly into themammary fat pad.These observations stand in agreement with the wide recog-nition of PEDF as being tumor suppressive (22, 23) and extendsits activity to include breast cancer and brain metastasis.

Most of the previous studies which showed a tumor-sup-pressive activity of PEDF focused on its potent antiangiogenicproperties. Our in vitro results show that PEDF was directlyinhibitory to breast cancer cells. Importantly, this effect wasreproduced in vivo. Transient expression of PEDF in brainmetastases resulted in a reduction in proliferation withoutaffecting the tumor vasculature. Moreover, when breast cancercells were inoculated directly into the mouse brain, PEDFrapidly reduced cancer cell proliferation and also activatedapoptosis. These effects occurred without an apparent changein tumor vasculature.

Figure 5. PEDF protects neuronsfrom tumor-induced damage in vivo.A, Bielschowsky silver stains ofmouse brains 24 hourspostimplantation. Left, a saline-injected control brain. Examples ofneurite bundles adjacent to the lesionwith normal morphology areindicated with red arrows. The lesionis located to the left of the dashedline. Middle, silver stain of neuritessurrounding implanted control 231-BRins cells. Examples of damagedbundles of neurites with thickenedand thinned fibers are designated bywhite arrows. The implanted lesion islocated to the left of the dashed line.Right, silver stain of neurites adjacentto implanted PEDF-231-BRins cells.B, Fluorojade-B stains of dyingneurons in mouse brains 24 hourspostimplantation. Left, the stabwound in a saline-injected controlbrain. Examples of Fluorojade-positive neurons (green) areindicated by white arrows and thelesion encircled by the dashed line.Middle, mouse brain adjacent to acontrol 231-BR tumor. Right, mousebrain around a PEDF-231-BRinstumor (left of dashed line). C, numberof Fluorojade-B–positive neuronsper mouse brain surrounding salineinjections, control 231-BRinstumors, and PEDF-231-BRinstumors. Results were averaged from2 experiments and 5 mice from eachgroup. D, the radius of damage(measured in mm) in mouse brainsurrounding stab wounds, controltumors, and PEDF-231-BRinstumors 24 hours postimplantation.

A

B

C D

Control tumor PEDF tumorSaline injection

Flu

oro

jade-p

ositiv

e c

ells

per

mouse

Dis

tance o

f fluoro

jade-p

ositiv

e

cells

fro

m lesio

n ±

SE

MP = 0.008

P = 0.0003P = 0.009

Saline Control

5,000

4,000

3,000

2,000

1,000

0

300

200

100

0PEDF Saline Control PEDF

P < 0.0001





While PEDF transgene expression was stable in culture,expression was lost within hours in vivo. Loss of proteinexpression was observed with two different promoters and inboth the presence and absence of transcriptional insulators.Because translational studies using adenoviral-driven PEDFexpression are ongoing (24), it would be of interest to deter-mine whether and how posttranscriptional regulation of PEDFprotein expression occurs in vivo.

Cognitive decline is frequently observed in patients withbrain metastases. There are at least three factors which con-tribute: progression of brain metastatic disease, "chemobrain"resulting from rounds of systemic chemotherapy, and radia-tion therapy used specifically to treat the brainmetastases. It iscurrently difficult to determine the relative contributions ofthese factors to the observed cognitive decline. Newmodels areneeded to measure and differentiate between the negativeeffects of disease progression versus the negative effects ofcancer therapies, on the brain. Our data show the use ofFluorojade-B staining as a quantitative measure of neuronaldamage induced by brainmetastases. The intracranial implan-tation model revealed that neuronal damage results not justfrom direct contact with tumor cells but extends to a concen-tric area around the brain metastasis. It is unknown whetherthis "collateral damage" results from the pressure of theexpanding mass, edema, and/or toxic byproducts of tumormetabolism. Regardless, PEDF expression resulted in a 3.5-folddiminution in neuronal damage. Previous studies have estab-lished the neuroprotective activity of PEDF in vitro and in vivo(reviewed in ref. 48). This report, to the best of our knowledge,provides the first evidence that PEDF can protect brain neu-rons from cancer-induced damage.

The data prompt the exciting hypothesis that PEDF canprevent some of the neuronal and cognitive sequelae associ-atedwith the development of brainmetastases, both by tumor-suppressive and neuroprotective effects. Further investigationwill require that neurocognitive endpoints be validated inmouse experimental brain metastasis models. Moreover, the

delivery of PEDF to the brain remains a significant obstacle.Functional subdomains of PEDF which recapitulate the neu-rotropic and antiangiogenic activity have been described (25–27). It will be important to determine which regions of PEDFare necessary for its direct effects on cancer cells. Delivery ofthe identified active peptide(s) could be attempted via anOmaya reservoir, but active blood–brain barrier transport maybe needed such as Angiopep-2 conjugation (49).

There is evidence for multiple PEDF receptors at the cellsurface (28–31). Differential expression of these receptors onneuronal, endothelial, and cancer cells may provide a partialexplanation for the differential effects on these cell popula-tions. Identification of which of these PEDF receptors arepresent on cancer cells, as well as further elucidation ofsignaling downstream of PEDF, could lead to the identificationof new pharmacologic targets for both anti-cancer and neu-ronal survival therapies.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

The authors thank Dr. Lin Tang and the laboratory of Dr. Howard Fine(National Cancer Institute) for their assistance. The authors also thank Drs.Glenn Merlino and Lalage Wakefield for input on strategies to overcometranscriptional silencing.

Grant Support

Thisworkwas supported by the intramural research programs of theNationalCancer Institute and the National Eye Institute and by the U.S. Department ofDefense Breast Cancer Research Program, grant number: W81 XWH-062-0033.Dr. Patricia Steeg's laboratory lab receives research support from GlaxoSmithKline and Millenium Pharmaceuticals.

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

Received June 3, 2011; revised October 24, 2011; accepted November 1, 2011;published online January 3, 2012.

References1. Kienast Y, Winkler F. Therapy and prophylaxis of brain metastases.

Expert Rev Anticancer Ther 2010;10:1763–77.2. Platta CS, Khuntia D, Mehta MP, Suh JH. Current treatment strat-

egies for brain metastasis and complications from therapeutictechniques: a review of current literature. Am J Clin Oncol 2009;33:398–407.

3. SteegPS,CamphausenKA,SmithQR.Brainmetastases aspreventiveand therapeutic targets. Nat Rev Cancer 2011;11:352–63.

4. deVries NA, Beijnen JH, Boogerd W, van Tellingen O. Blood-brainbarrier and chemotherapeutic treatment of brain tumors. Expert RevNeurother 2006;6:1199–209.

5. Lin Q, Balasubramanian K, Fan D, Kim SJ, Guo L, Wang H, et al.Reactive astrocytes protect melanoma cells from chemotherapy bysequestering intracellular calcium through gap junction communica-tion channels. Neoplasia 2010;12:748–54.

6. Eichler AF, Loeffler JS. Multidisciplinary management of brain metas-tases. Oncologist 2007;12:884–98.

7. Argyriou AA, Assimakopoulos K, Iconomou G, Giannakopoulou F,Kalofonos HP. Either called "Chemobrain" or "Chemofog," the long-term chemotherapy-induced cognitive decline in cancer survivors isreal. J Pain Symptom Manage 2010;41:126–39.

8. Welzel G, Fleckenstein K, Mai SK, Hermann B, Kraus-Tiefenbacher U,Wenz F. Acute neurocognitive impairment during cranial radiationtherapy in patients with intracranial tumors. Strahlenther Onkol2008;184:647–54.

9. Palmieri D, Fitzgerald D, Shreeve SM,Hua E, Bronder JL,Weil RJ, et al.Analyses of resected human brain metastases of breast cancer revealthe association between up-regulation of hexokinase 2 and poorprognosis. Mol Cancer Res 2009;7:1438–45.

10. Cai J, Parr C, Watkins G, Jiang WG, Boulton M. Decreased pigmentepithelium-derived factor expression in human breast cancer progres-sion. Clin Cancer Res 2006;12:3510–7.

11. Steele FR, Chader GJ, Johnson LV, Tombran-Tink J. Pigment epithe-lium-derived factor: neurotrophic activity and identification as a mem-ber of the serine protease inhibitor gene family. Proc Natl Acad SciU S A 1993;90:1526–30.

12. Cao W, Tombran-Tink J, Elias R, Sezate S, Mrazek D, McGinnis JF. Invivo protection of photoreceptors from light damage by pigmentepithelium-derived factor. InvestOphthalmolVisSci 2001;42:1646–52.

13. Cayouette M, Smith SB, Becerra SP, Gravel C. Pigment epithelium-derived factor delays the death of photoreceptors in mouse models ofinherited retinal degenerations. Neurobiol Dis 1999;6:523–32.

Fitzgerald et al.




14. Jablonski MM, Tombran-Tink J, Mrazek DA, Iannaccone A. Pigmentepithelium-derived factor supports normal development of photore-ceptor neurons and opsin expression after retinal pigment epitheliumremoval. J Neurosci 2000;20:7149–57.

15. Stellmach V, Crawford SE, Zhou W, Bouck N. Prevention of ischemia-induced retinopathy by the natural ocular antiangiogenic agent pig-ment epithelium-derived factor. Proc Natl Acad Sci U S A 2001;98:2593–7.

16. Yabe T, Sanagi T, Yamada H. The neuroprotective role of PEDF:implication for the therapy of neurological disorders. Curr Mol Med2010;10:259–66.

17. Tombran-Tink J, Mazuruk K, Rodriguez IR, Chung D, Linker T, Eng-lander E, et al. Organization, evolutionary conservation, expressionand unusual Alu density of the human gene for pigment epithelium-derived factor, a unique neurotrophic serpin. Mol Vis 1996;2:11.

18. Chung C, Doll JA, Gattu AK, Shugrue C, Cornwell M, Fitchev P, et al.Anti-angiogenic pigment epithelium-derived factor regulates hepato-cyte triglyceride content through adipose triglyceride lipase (ATGL). JHepatol 2008;48:471–8.

19. Ramirez-Castillejo C, Sanchez-Sanchez F, Andreu-Agullo C, FerronSR, Aroca-Aguilar JD, Sanchez P, et al. Pigment epithelium-derivedfactor is a niche signal for neural stem cell renewal. Nat Neurosci2006;9:331–9.

20. Wang JJ, Zhang SX, Mott R, Chen Y, Knapp RR, Cao W, et al. Anti-inflammatory effects of pigment epithelium-derived factor in diabeticnephropathy. Am J Physiol Renal Physiol 2008;294:F1166–73.

21. DawsonDW, Volpert OV, Gillis P, Crawford SE, XuH, BenedictW, et al.Pigment epithelium-derived factor: a potent inhibitor of angiogenesis.Science 1999;285:245–8.

22. Fernandez-Garcia NI, Volpert OV, Jimenez B. Pigment epithelium-derived factor as a multifunctional antitumor factor. J Mol Med2007;85:15–22.

23. Hoshina D, Abe R, Yamagishi SI, Shimizu H. The role of PEDF in tumorgrowth and metastasis. Curr Mol Med 2010;10:292–5.

24. Campochiaro PA, Nguyen QD, Shah SM, Klein ML, Holz E, Frank RN,et al. Adenoviral vector-delivered pigment epithelium-derived factorfor neovascular age-related macular degeneration: results of a phase Iclinical trial. Hum Gene Ther 2006;17:167–76.

25. BilakMM,BecerraSP,VincentAM,MossBH,AymerichMS,KunclRW.Identification of the neuroprotective molecular region of pigmentepithelium-derived factor and its binding sites on motor neurons. JNeurosci 2002;22:9378–86.

26. Filleur S, Volz K, Nelius T,Mirochnik Y, HuangH, Zaichuk TA, et al. Twofunctional epitopes of pigment epithelial-derived factor block angio-genesis and induce differentiation in prostate cancer. Cancer Res2005;65:5144–52.

27. Mirochnik Y, Aurora A, Schulze-Hoepfner FT, Deabes A, Shifrin V,Beckmann R, et al. Short pigment epithelial-derived factor-derivedpeptide inhibits angiogenesis and tumor growth. Clin Cancer Res.2009;15:1655–63.

28. Bernard A, Gao-Li J, Franco CA, Bouceba T, Huet A, Li Z. Lamininreceptor involvement in the anti-angiogenic activity of pigment epi-thelium-derived factor. J Biol Chem 2009;284:10480–90.

29. Notari L, Arakaki N, Mueller D, Meier S, Amaral J, Becerra SP. Pigmentepithelium-derived factor binds to cell-surface F(1)-ATP synthase.FEBS J 2010;277:2192–205.

30. Notari L, Baladron V, Aroca-Aguilar JD, Balko N, Heredia R, Meyer C,et al. Identification of a lipase-linked cell membrane receptor forpigment epithelium-derived factor. J Biol Chem 2006;281:38022–37.

31. Park K, Lee K, Zhang B, Zhou T, He X, Gao G, et al. Identification of anovel inhibitor of the canonical Wnt pathway. Mol Cell Biol 2011;31:3038–51.

32. Palmieri D, Bronder JL, Herring JM, Yoneda T,Weil RJ, Stark AM, et al.Her-2 overexpression increases the metastatic outgrowth of breastcancer cells in the brain. Cancer Res 2007;67:4190–8.

33. Lockman PR, Mittapalli RK, Taskar KS, Rudraraju V, Gril B, Bohn KA,et al. Heterogeneous blood-tumor barrier permeability determinesdrug efficacy in experimental brain metastases of breast cancer. ClinCancer Res 2010;16:5664–78.

34. Seigel GM, Mutchler AL, Imperato EL. Expression of glial markers in aretinal precursor cell line. Mol Vis 1996;2:2.

35. Chung JH, Whiteley M, Felsenfeld G. A 50 element of the chickenbeta-globin domain serves as an insulator in human erythroid cellsand protects against position effect in Drosophila. Cell 1993;74:505–14.

36. Stratikos E, Alberdi E, Gettins PG, Becerra SP. Recombinant humanpigment epithelium-derived factor (PEDF): characterization of PEDFoverexpressed and secreted by eukaryotic cells. Protein Sci 1996;5:2575–82.

37. Solly K, Wang X, Xu X, Strulovici B, Zheng W. Application of real-timecell electronic sensing (RT-CES) technology to cell-based assays.Assay Drug Dev Technol 2004;2:363–72.

38. Palmieri D, Lockman PR, Thomas FC, Hua E, Herring J, Hargrave E,et al. Vorinostat inhibits brain metastatic colonization in a model oftriple-negative breast cancer and induces DNA double-strand breaks.Clin Cancer Res 2009;15:6148–57.

39. Baschnagel A, Russo A, Burgan WE, Carter D, Beam K, Palmieri D,et al. Vorinostat enhances the radiosensitivity of a breast cancer brainmetastatic cell line grown in vitro and as intracranial xenografts. MolCancer Ther 2009;8:1589–95.

40. Schmued LC, Hopkins KJ. Fluoro-Jade B: a high affinity fluorescentmarker for the localization of neuronal degeneration. Brain Res2000;874:123–30.

41. Fitzgerald DP, Palmieri D, Hua E, Hargrave E, Herring JM, Qian Y, et al.Reactive glia are recruited by highly proliferative brain metastases ofbreast cancer and promote tumor cell colonization. Clin Exp Metas-tasis 2008;25:799–810.

42. Gril B, Palmieri D, Qian Y, Smart D, Ileva L, Liewehr DJ, et al.Pazopanib reveals a role for tumor cell B-Raf in the prevention ofHER2þ breast cancer brain metastasis. Clin Cancer Res 2011;17:142–53.

43. AndersonKJ,Miller KM,Fugaccia I, ScheffSW.Regional distributionoffluoro-jade B staining in the hippocampus following traumatic braininjury. Exp Neurol 2005;193:125–30.

44. Frank TC, Nunley MC, Sons HD, Ramon R, Abbott LC. Fluoro-jadeidentification of cerebellar granule cell and purkinje cell death in thealpha1A calcium ion channel mutant mouse, leaner. Neuroscience2003;118:667–80.

45. Krinke GJ, Classen W, Vidotto N, Suter E, Wurmlin CH. Detectingnecrotic neurons with fluoro-jade stain. Exp Toxicol Pathol 2001;53:365–72.

46. Kundrotiene J, Wagner A, Liljequist S. Fluoro-Jade and TUNEL stain-ing as useful tools to identify ischemic brain damage following mod-erate extradural compression of sensorimotor cortex. Acta NeurobiolExp (Wars) 2004;64:153–62.

47. Mitruskova B, Orendacova J, Racekova E. Fluoro Jade-B detection ofdying cells in the SVZ and RMS of adult rats after bilateral olfactorybulbectomy. Cell Mol Neurobiol 2005;25:1255–64.

48. Tombran-Tink J. The neuroprotective and angiogenesis inhibitoryserpin, PEDF: new insights into phylogeny, function, and signaling.Front Biosci 2005;10:2131–49.

49. Demeule M, Regina A, Che C, Poirier J, Nguyen T, Gabathuler R, et al.Identification and design of peptides as a new drug delivery system forthe brain. J Pharmacol Exp Ther 2008;324:1064–72.





2012;72:144-153. Cancer Res Daniel P. Fitzgerald, Preeti Subramanian, Monika Deshpande, et al. Metastasis and Metastasis-Induced Brain DamageCancer Cell versus Neuronal Survival: Implication for Brain

Derived Factor on Breast−Opposing Effects of Pigment Epithelium

Updated version

http://cancerres.aacrjournals.org/content/72/1/144

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2011/12/28/72.1.144.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/72/1/144.full#ref-list-1

This article cites 49 articles, 20 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/72/1/144.full#related-urls

This article has been cited by 3 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/72/1/144To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/suppl/2011/12/28/72.1.144.DC1

http://cancerres.aacrjournals.org/content/72/1/144.full#ref-list-1

http://cancerres.aacrjournals.org/content/72/1/144.full#related-urls

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

mailto:[email protected]



Opposing Effects of Pigment Epithelium Derived Factor on ...BRins and 4T1-BR5 cells, by selection in...

Documents

Transcript of Opposing Effects of Pigment Epithelium Derived Factor on ...BRins and 4T1-BR5 cells, by selection in...