regulation of vascular endothelial growth factor expression in

(CANCER RESEARCH 58. 4008-4014, September I, 1998]

Regulation of Vascular Endothelial Growth Factor Expression in Human ColonCancer by Insulin-like Growth Factor-I1

Yoshito Akagi, Wenbiao Liu, Brian Zebrowski, Keping Xie, and Lee M. Ellis2

Departments of Cell Biologv [Y. A., W. L., K. X., L.M.E.] and Surgical Oncologv Â¡B.Z., L. M. E.I, The Universitv of Texas M. D. Anderson Cancer Center, Houston,Texas 77030

ABSTRACT

We investigated the role of insulin-like growth factor (IGF)-I andIGF-binding proteins (IGFBPs) in the regulation of vascular endothelial

growth factor (VE(Ã¬F)expression in colon cancer cells and the mechanismby which this regulation occurs. HT29 human colon cancer cells weretreated with K.I -I for various time periods. VEGF mRNA expression

increased within 2 h and peaked at 24 h. SYV620 colon cancer cellsexhibited a peak induction of VEGF mRNA K h after IGF-I treatment.IGF-I induction of VEGF was confirmed at the protein level. In experiments using transient transfection of VEGF promoter-reporter constructsinto HT29 cells, IGF-I increased the activity of the VEGF promoter, and

pretreatment of HT29 cells with dactinomycin abrogated the induction ofVEGF mRNA by IGF-I. The half-life of VEGF mRNA was not prolongedby treatment with IGF-I. Blocking the activity of IGFBP-4 did not significantly modulate the effect of IGF-I induction of VEGF mRNA in HT29cells. Treating cells with des-( 1-31-IGF-I (an active derivative of IGF-I that

does not hind to binding proteins) had effects on VEGF mRNA expressionthat were similar to those of IGF-I. These findings suggest that IGF-Iregulates Y'EGF expression in human colon cancer cells by induction of

transcription of the VEGF gene. IGFBPs do not significantly affect IGF-I

induction of VEGF.

INTRODUCTION

Neovasculari/.ation is essential tor tumor growth and metastasisformation (1. 2). The development of new blood vessels in tumorsdepends on the production of angiogenic factors released from thetumor cells and/or cells inhabiting the tumor microenvironment.VEGF' is a potent and unique angiogenic protein that stimulates

capillary formation and has specific mitogenic and chcmotactic activity for vascular endothelial cells (3). VEGF has been shown toinduce angiogenesis in numerous in vivo and in vitro assays (4, 5).However, antibodies to VEGF do not inhibit the growth of tumor cellsin vitro, confirming the factor's selective activity on endothelial cells

(6). The critical role played by VEGF in developmental angiogenesisand vasculogenesis is demonstrated by the fact that knockout (ordisruption) of the VEGF gene in mice results in embryo death due toabnormal blood vessel development (7). VEGF is expressed in nearlyall cell types, but many malignant tumor cells demonstrate overex-pression of VEGF (8-16). Studies from our laboratory as well as

others suggest that VEGF expression is intimately associated with

Received 3/20/98: accepted 7/3/98.The costs of publication of this article were defrayed in part hy the payment of page

charges. This article must Iherefore he hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported in part by American Cancer Society Career Development

Award 94-21 (to L. M. E.), the Gillson Longenbaugh Foundation (to L. M. E.), and IheCharlotte Geyer Foundation (lo L. M. E.).

2 To whom requests for reprints should be addressed, at Department of Surgical

Oncology. Box 106. The University of Texas M. D. Anderson Cancer Center. 1515Holcombe Boulevard. Houston. TX 77030. Phone: (713) 792-6926; Fax: (713) 792-0722:E-mail: leihst notes.mdacc.tmc.edu.

1The abbreviations used are: VEGF, vascular endothelial growth factor; IGF, insulin-

like growth factor: IGFBP. IGF-hinding protein; IL. interleukin: PDGF. platelet-derivedgrowth factor; TGF. transforming growth factor; EGF. epidermal growth factor: HGF.hepaUKyte growth factor: PD-ECGF. platelet-derived endothelial cell growth factor:ActD. daclinomvcin (aclinomycin D): GAPDH. glyceraldehyde 3-phosphate dehydrogen-

inducing and maintaining the neovasculature of human colon cancers(11, 12, 17, 18).

VEGF expression is regulated by numerous physiological andpathological processes. The best-characterized process that regulates

VEGF is hypoxia (19, 20). Hypoxia increases VEGF expressionacutely through both an increase in transcription of the gene and aprolongation of the mRNA half-life (21, 22). Recently, IGF-I, a

protein that induces proliferation of certain malignant cells (23, 24),has been shown to increase VEGF mRNA and protein expression inboth malignant and nonmalignant cell lines (25-27). This observation

is particularly relevant to colon cancer in that the primary organ ofmetastasis, the liver, synthesizes levels of IGF-I that are 30-fold

higher than those of any other organ (28). The proliferative activitiesof IGFs are regulated by signaling through the type 1 IGF receptors onvarious cell types, including human colon cancer cells (29, 30). Inaddition, the biological actions of IGFs are modulated by a family ofIGFBPs, which may either increase or decrease IGF availability toreceptors (31).

VEGF expression is also regulated by numerous cytokines andgrowth factors that are highly dependent on the tumor system understudy. Factors that have been demonstrated to increase VEGF expression include IL-lÃŸ(32), IL-6 (33), PDGF-BB (34), TGF-ÃŸ(35), basic

fibroblasl growth factor (36), EGF (36), and HGF (37). Previously, wedemonstrated that Src kinase activity regulates VEGF expression incolon cancer cells, and it is possible that factors that increase Srcactivity also increase VEGF expression (38, 39). Growth factorsindigenous to the colon, e.g., gastrin and bombesin, represent suchfactors.

Here, we investigated the effects of various cytokines and growthfactors on the induction of VEGF in human colon cancer cells. Thefactors chosen for study were those that are present in the colon cancermicroenvironment, those that activate Src, or both. In our initialstudies, IGF-I caused the greatest increase in VEGF mRNA expression. Therefore, we chose to further investigate the role of IGF-I in the

induction of VEGF in human colon cancer and the mechanism bywhich this induction occurs.

MATERIALS AND METHODS

Materials. Recombinant human IGF-I and TGF-a were purchased fromIntergen Company (Purchase. NY). Recombinant human EGF, HGF, IL-lÃŸ.IL-6, PDGF-BB, PD-ECGF, and TGF-ÃŸwere purchased from R&D Systems

(Minneapolis. MN). Bombesin and gastrin were purchased from BachernBioscience (Philadelphia. PA). ActD was purchased from Calbiochem-Nova-biochem (La Jolla. CA). Antiserum to IGFBP-4 was purchased from UpstateBiotechnology. Inc. (Lake Placid, NY). Des-(l-3)-IGF-I was purchased from

GroPep Pty. Ltd. (Adelaide. Australia). CoCI, was purchased from SigmaChemical Co. (St. Louis, MO).

Cell Lines and Culture Conditions. The human colon cancer cell linesHT29 and SW620 were obtained from the American Type Culture Collection(Manassas, VA). These cells were cultured and maintained in MEM sup'ple-

mented with 10% FBS. 2 units/ml penicillin-streptomycin, vitamins, 1 mMsodium pyruvate. 2 mM L-glutamine. and nonessential amino acids at 37Â°Cin

5% CO2 and 95% air. All experiments were performed after the cells weregrown to 90-100% confluence to avoid the effects of cell density on VEGF

expression (40).

4008

Research. on January 3, 2019. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

[GF-I REGULATION OF VEGF

Cytokine/Growth Factor Treatment of Cells. Initially, we sought todetermine which cytokines or growth factors increase VEGF mRNA inductionin HT29 cells. HT29 cells were incubated in serum-free medium overnight andwere then incubated in the presence of EOF (50 ng/ml), HGF (50 ng/ml), IGF-I(100 ng/ml), PD-ECGF (50 ng/ml), PDGF-BB (10 ng/ml), IL-lÃŸ(10 ng/ml),IL-6 (80 ng/ml), TGF-a (25 ng/ml), TGF-/3 (10 ng/ml), bombesin (1.5 /j.g/ml),

or gastrin (20 ng/ml) for 4 or 24 h. time points at which other investigatorshave shown VEGF induction in other systems. Total RNA was extracted fromthe cells as described below, and Northern blot analyses were performed.These studies were performed in triplicate.

mRNA Extraction and Northern Blot Analysis. Total RNA was extracted from cells using Tri Reagent (Molecular Research Center, Cincinnati.OH). Northern blot hybridization was performed as described previously (40).Briefly, total RNA (25 fig) was separated by electrophoresis in 1% denaturingformaldehyde-agarose gels. The RNA was transferred to the Hybond-N +

positively charged nylon membrane (Amersham Life Science, ArlingtonHeights, IL) overnight by capillary elution and UV cross-linked at 120.000/xJ/cm2 with a UV Stratalinker 1800 (Stratagene, La Jolla, CA). After prehy-bridization of blots for 3-4 h at 65Â°Cin Rapid hybridization buffer (Amersham), the membranes were hybridized overnight at 65Â°Cwith the cDNA

probe for VEGF or GAPDH. The probed nylon membranes were then washedand exposed to radiographie film.

cDNA Probes. The cDNA probes used in this analysis were a humanVEGF-specific 204-bp probe (a gift of Dr. Brygida Berse. Harvard Medical

School, Boston, MA; Ref. 41) and a GAPDH probe, purchased from theAmerican Type Culture Collection. The VEGF probe identifies all alternatively spliced forms of its mRNA transcripts. Probes were purified by agarose

gel electrophoresis using the QIAEX Gel Extraction kit (Qiagen, Chatworth.CA). Each cDNA probe was radiolabeled with [<*-32P]dNTP by the random-

priming technique using the Rediprime labeling system (Amersham).Determination of VEGF Protein Levels. HT29 cells were incubated in

serum-free medium overnight and then in the presence or absence of IGF-I

(100 ng/ml). Cell lysates and supernatants were harvested at 8 and 24 h. Thesupernatant was centrifuged to remove debris and stored at -70Â°C until

analysis. Protein was extracted from the cell lysates with protein lysis buffer[20 mM Tris-HCl (pH 8.0), 137 mM NaCl, 10% glycerol, 1% Triton X-100, 1

mM Na,VO4, 2 mM EDTA, 100 mg/ml phenylmethylsulfonyl fluoride, 5mg/ml leupeptin. and 10 mg/ml aprotinin]. The total protein concentration wasquantitated spectrophotometrically. The VEGF protein level in the supernatantand cell lysates was determined using an ELISA kit according to the manufacturer's instructions (R&D Systems).

Time Course of VEGF mRNA Induction by IGF-I. HT29 and SW620cells were incubated in serum-free medium overnight and were then incubatedin the presence or absence of IGF-I (100 ng/ml) for 2, 4, 8, 24. or 48 h. Total

RNA was extracted from the cells as described below. VEGF mRNA expression was determined by Northern blot analysis. The time point of peakexpression of VEGF was used in subsequent studies. We examined the effectof IGF-I on VEGF expression in SW620 colon cancer cells to assure that theeffect of IGF-I on VEGF induction was not due to a phenomenon that was

unique to the HT29 cell line.VEGF Promoter-Reporter Studies in Response to IGF-I. The role of

transcriptional regulation of VEGF by IGF-I was examined using transienttransfection with a VEGF promoter-luciferase reporter construct. The following plasmids were used: pGL3-VEGF (containing the human VEGF promoter

linked to the firefly luciferase reporter gene), pRLTK (an internal controlplasmid containing the herpes simplex thymidine kinase promoter linked to aconstitutively active Renilla luciferase reporter gene), and pGL3 (plasmidvector alone as a negative control). HT29 cells (5 X IO6) were seeded and

grown to 60-70% confluence, and then pRLTK and pGL3-VEGF (5 mg:20

mg) were cotransfected into cells using the Lipofectin method (Life Technologies, Inc., Grand Island. NY) as outlined by the manufacturers. pRLTK andpGL3 were cotransfected as a negative control. After cells were incubated inthe transfection medium for 20 h, the medium was changed to standardmedium, and cells were incubated for 12 h. Cells were then incubated in thepresence or absence of IGF-I or CoCU (200 /Â¿M)for 24 h (the period

determined to afford the greatest VEGF mRNA expression |. Cells were harvested with passive lysis buffer (Dual-Luciferase Reporter Assay System;Promega, Madison. WI), and luciferase activity was determined using a single-sample luminometer, as outlined in the manufacturer's protocol. CoCU was

used as a positive control because of its ability to mimic the hypoxic response(42).

Effect of IGF-I on VEGF Transcriptional Activity. To confirm that the

increase in VEGF mRNA in colon cancer cells was due to an increase intranscription, transcriptional activity was blocked with ActD prior to treatmentof cells with IGF-I. HT29 cells, incubated in serum-free medium overnight,

were incubated in the presence or absence of ActD (1 /Mg/ml) 2 h prior toexposure to IGF-I (optimal dosing was determined in preliminary experi

ments). Total RNA was extracted from the cells after 24 h. Control cells weretreated with ActD without IGF-I.

VEGF mRNA Half-Life Studies. To determine the effect of IGF-I on

VEGF mRNA stability. HT29 cells were incubated in the presence or absenceof IGF-I for 24 h. Further transcription in cells was then blocked by the

addition of ActD (final concentration of 5 /ng/ml). Total RNA was extractedfrom the cells 0, 0.5, 1, 2, 4, or 6 h after the addition of ActD, and Northernblot analysis was performed for VEGF mRNA expression. VEGF mRNAexpression of each time point was compared to the control value (total RNAextracted from cells prior to ActD treatment was arbitrarily defined as 100%).The half-life of VEGF mRNA was determined by plotting relative VEGF

mRNA expression levels on a semilogarithmic axis versus time (CricketSoftware, Malvern, PA).

Effect of Antiserum to IGFBP-4 and of Des-(l-3)-IGF-I on IGF-IInduction of VEGF. HT29 cells primarily secrete IGFBP-4. which may bindIGF-I and diminish its biological activity (28. 43). Therefore, it was necessaryto determine the role of IGFBP-4 in VEGF mRNA induction by IGF-I. HT29cells were incubated in the presence or absence of IGF-I. antiserum toIGFBP-4 (1:100 dilution), and/or nonspecific control antibodies (rabbit anti-

IgG antibodies, 1:100 dilution) for 24 h. Northern blot analyses were performed for VEGF expression after extraction of total RNA from the cells.

In a separate experiment. HT29 cells were incubated in the presence orabsence of IGF-I and/or des-(l-3)-IGF-I (50 ng/ml) for 8 or 24 h. Des-(l-3)-IGF-I is an IGF-I derivative that binds poorly to IGFBPs but can effectivelyactivate IGF-I receptors (44). Northern blot analyses were performed for

VEGF expression after extraction of total RNA from the cells.Densitometric Quantitation. The software program Image Quant (Molec

ular Dynamics. Sunnyvale. CA) was used to quantitate the VEGF mRNAexpression and GAPDH mRNA transcripts in the linear range of the film.GAPDH mRNA was used as an internal control for possible loading differences.

RESULTS

Effect of IGF-I on Induction of VEGF inRNA in HT29 Cells.VEGF mRNA expression was increased 5- and 3-fold by treatmentwith IGF-I and IL-lÃŸ,respectively. EOF, HGF, PD-ECGF, PDGF-BB. IL-6, TGF-a, TGF-/3, bombesin, and gastrin did not increase

VEGF mRNA expression (Fig. 1).In both HT29 and SW620 cells, IGF-I-induced VEGF mRNA

expression was increased by 8 h and remained increased at 24 h (Fig.2A). However, we noted that control cells that were grown in serum-free medium without the addition of IGF-I also exhibited an increase

in VEGF mRNA expression (Fig. 2A). This observation has beennoted in other experiments performed in our laboratory.4 We hypoth

esized that serum starvation for greater than 24 h induces a stressresponse in the cells under study. To determine the effect of serumstarvation on VEGF mRNA expression, HT29 cells were incubated inserum-free medium for 0, 2, 4, 8, 24, or 48 h, and VEGF mRNAexpression was determined. Serum starvation caused a 4.5-fold in

crease in VEGF mRNA after 48 h compared to time 0 (data notshown). Therefore, following densitometric analysis, the increase inVEGF expression due to serum starvation was corrected for bysubtracting it from the increase in VEGF mRNA expression in thecells treated with IGF-I. [Other investigators (25, 26, 32) have com

pared treated cells at later time points to controls at time 0, without

J Unpublished observations.

4009



lOF-l REGULATION OF VEGF

Fig. 1. VEGF mRNA expression in cells treatedwith cytokines or growth factors. Cells weretreated with EOF (50 ng/ml). HOP (50 ng/ml).IGF-I (100 ng/ml), PD-ECGF (50 ng/ml).PDGF-BB ( 10 ng/ml). IL-lÃŸ(10 ng/ml), 1L-6 (80ng/ml). TGF-a (25 ng/ml), TGF-/3 (10 ng/ml).homhesin (1.5 /ig/ml). or gastrin (20 ng/ml) inserum-free medium for 24 h. Control cells weregrown under identical conditions but without cytokines or growth factors. Studies were performedin triplicate. Total RNA was extracted from cells,and Northern blot analyses were performed to determine VEGF mRNA expression. GAPDHmRNA transcripts were used as an internal controlto correct for loading differences. A. representativeNorthern blot. IGF-I increased VEGF mRNA expression in HT29 cells >5-fold at 24 h. B. densi-

tometric values. Columns, mean corrected intensity of VEGF expression; bars, SE.

VEGF

GAPDH

B

4.4kb3.7kb

1.3kb

7-0

6 -LUc

5-COO

3-_C"O

2-Ou_nSo

OnLLo111InnLL

â€”¿�LLOli.C31 s aQ

O.innnm

oÃ toaBV Â¿illi. ^ ~ (3

0 HP0.nCO.!,'oC

CÂ«

iOJoÃ

-Q <8E (3offl

corrcclion for stress induction of VEGF.] In HT29 cells, expression ofVEGF mRNA in cells treated with IGF-I peaked at 24 h, with a7.3-fold greater increase than in control cells. In SW620 cells, thepeak was at 8 h and was 4.7-fold greater than that in controls (Fig.

2B).IGF-I Increased VEGF Protein in HT29 and SW620 Cells. To

determine whether the increase in VEGF mRNA expression inducedby IGF-I was associated with an increase in VEGF protein, VEGF

protein levels in cell lysates and supernatants were determined byELISA (Fig. 3). IGF-I-treated cells demonstrated an increase in VEGF

protein levels in the cell lysates and supernatants of both cell lines. InHT29 cell lysate, VEGF protein increased >4-fold at 24 h. In SW620cell lysate, VEGF protein increased nearly 3-fold at 24 h.

IGF-I Increased VEGF mRNA Expression by an Increase inTranscriptional Activity. To determine the mechanism for the IGF-I

induction of VEGF, transient transfections were performed with promoter-reporter constructs in HT29 cells. Control cells transfected with

pGL3 (plasmid vector alone) and pRLTK demonstrated no increase inpromoter activity (Fig. 4/4). Cells transfected with pGL3-VEGF(promoter-reporter construct) and pRLTK and treated with CoCl2(positive control) demonstrated a 9-fold increase in activity. Cellstransfected with pGL3-VEGF and pRLTK and treated with IGF-Isimilarly demonstrated a 9-fold increase in activity.

To determine that the mechanism by which IGF-I induced VEGF

mRNA expression occurred at a transcriptional level, transcriptionwas blocked with ActD in HT29 cells prior to the addition of IGF-I.IGF-I induction of VEGF mRNA expression was completely abol

ished by the blockade of transcription (Fig. 4B). These data suggestthat IGF-I induction of VEGF is regulated by an increase in transcrip

tion of the gene.IGF-I Did Not Alter the Stability of VEGF mRNA. To further

explore the mechanism by which IGF-I enhanced the expression of

VEGF mRNA, the stability of VEGF mRNA was investigated by

examining its half-life. The half-life of VEGF mRNA treated withIGF-I was similar to that of cells not exposed to IGF-I (Fig. 5). Thus,the half-life of VEGF mRNA was not prolonged by treatment withIGF-I.

Effect of Antiserum to IGFBP-4 and of Des-(l-3)-IGF-I onVEGF mRNA Induction. To determine whether IGFBP-4 contributes to the regulation of IGF-I induction of VEGF, expression of

VEGF was examined by incubating HT29 cells in the presence orabsence of antiserum to IGFBP-4. VEGF mRNA expression in cellstreated with antiserum to IGFBP-4 alone was not different from that

in control cells. VEGF mRNA expression in cells treated with bothIGF-I and control antibody (IgG) or with both IGF-I and antiserum toIGFBP-4 was similarly increased (Fig. 6).

To further determine the effect of IGFBPs on VEGF mRNA expression induced by IGF-I, HT29 cells were incubated with IGF-I ordes-(l-3)-IGF-I, which does not bind to IGFBPs. There was no

difference in VEGF expression at 8 and 24 h in cells treated withIGF-I or des-(l-3)-IGF-I. These results suggest that IGFBPs do notsignificantly affect IGF-I induction of VEGF mRNA expression.

DISCUSSION

Angiogenesis is an essential step in tumor growth and metastasis,and this process is driven by the balance of positive and negativeeffector molecules (45). Several laboratories, including our own, havedemonstrated that microvessel counts are strong prognostic factors inhuman colorectal cancer (11, 46, 47). Previous studies demonstratingthat VEGF expression correlates with microvessel count have suggested that VEGF is involved in regulating human colon cancerangiogenesis (11, 46). Other studies have demonstrated the importance of VEGF in the growth and metastasis of human colon cancer;for example, neutralizing VEGF antibodies given to mice bearinghuman colon cancer xenografts decrease tumor growth and inhibit

4010



IGF-I REGULATION OF VEGF

HT29 SW620

2 4 8 24 48Hours after

incubation with IGF-I 2 4 8 24 48â€¢¿�*-IGF-Iâ€”>â€¢"^ "^ ~r~+ ^~+~^~

Â«â€¢Â«â€¢â€¢MVEGF

GAPDH1.3kb

â€¢¿�â€¢â€¢¿�*lil-lB

HT29 SW620

24 8 24 48

Hours after addition of IGF-I

-8

-6

Fold increase" 4 in VEGF

-2

0

6 -

4 .

2 .

0~~1

2 4 8 24 48

Hours after addition of IGF-I

Fig. 2. Time course of VEGF mRNA induction by IGF-1. HT29 and SW620 cells were treated with or without IGF-I (100 ng/ml) for the indicated times. Northern blots were

performed after extraction of total RNA from cells (A). For each time point, untreated cells were used as a control. Relative VEGF mRNA expression was determined by Northernblot analysis as described in "Materials and Methods." To correct for increases in VEGF expression in "control cells" due to stress (serum-free condition), the increased signal intensity

for VEGF mRNA expression in control cells was subtracted from the signal intensity observed in cells treated with IGF-I. The signal intensity for VEGF after this correction wascompared to VEGF mRNA expression in untreated cells at the earliest time point. The final signal intensity of VEGF was expressed as a relative value (cÃÃ//Ãmn.v.ÃŸ).

metastasis formation (17). Thus, it appears that VEGF is a criticalfactor in determining the angiogenic phenotype in a majority ofhuman colon cancers.

Factors that regulate VEGF expression in tumor and nontumor cellsare now being elucidated. The best-characterized mediator that induces an increase in VEGF expression is hypoxia (19-21, 48). Hy-poxia increases VEGF expression within 3-6 h, whereas normaliza

tion of oxygen tension causes cellular VEGF mRNA to return tobaseline levels (48). Hypoxie induction of VEGF may be regulated byan increase in transcription and/or stabilization of the mRNA (21, 22).

PDGF-BB, EOF, basic fibroblast growth factor, TGF-a, TGF-ÃŸ,HGF, IL-IÃŸ,IL-6, and other cytokines and growth factors are known

to affect VEGF expression. However, not all of these factors increaseVEGF expression in all tumor systems; i.e., the factors that areinvolved in the regulation of VEGF may be dependent upon the tumorsystem under study. In addition, it is not clear whether these factorsaffect common signal transduction pathways or multiple pathways inthe regulation of VEGF expression. Here, we investigated the effect ofIGF-I on VEGF expression in human colon cancer cell lines. Al

though VEGF is constitutively expressed in nearly all cells, its induction is regulated by specific factors that may be intracellular, extracellular, or both. The sites of growth of colon cancer are sites whereIGF-I is indigenous to the microenvironment. The liver, the major

organ of metastasis in patients with colon cancer, harbors levels of

HT29

Fig. 3. VEGF protein level in supematants andcell lysates of cells incubated with IGF-I. Cells weretreated with or without IGF-I (100 ng/ml) in serum-

free medium for 8 or 24 h after incubation overnightin serum-free medium. Supematants and cell lysateswere examined for VEGF protein levels by ELISA.Columns, increase in VEGF protein induced byIGF-I, represented as a percentage of the VEGFprotein level in untreated cells; bars, SE.

SW620

8 24

Hours after treatment with IGF-I Hours after treatment with IGF-I

Supernatant of cells treated with IGF-I

4011

Cell lysates of cells treated with IGF-I



lOF-I REGULATION OF VEOF

Control CoCI2 IGF-I

BActD (1 ug/ml)

IGF-1 (100ng/ml)

VEGF 4.4kb3.7kb

GAPDH 1.3kb ^

Fig. 4. IGF-I induction of VEGF transcriplion. A. HT29 cells were cotransfected withpGL.I-VEGF (â€¢.VEGF promoler-lucifera.se reporter construct) and pRLTK (D. tocontrol for transfection efficiency). Transfection with pGL3 and pRLTK was performedas a negative control. Twenty-four h after transient transfection. cells were treated withIGF-I ( IOOng/ml), CoCU (200 UM: positive control), or medium alone (negative control)for 24 h. and luciferase activity was determined. Columns, relative promoter activity. B.HT29 cells were treated with ActD (1 ng/ml) for 2 h to block further transcription. Cellswere then treated with IGF-I for 24 h, and RNA was extracted and Northern blots wereperformed. Cells treated with IGF-I but not pretreated with ActD were used as positive

controls.

100

Half-life

Hours after administration of actinomycin D

Fig. 5. VEGF mRNA half-life studies. HT29 cells were incubated in the presence(â€”â€¢or absence (- - Q) of IGF-I for 24 h prior to exposure to ActD (5 u.g/ml). Total

RNA was extracted from cells at 0, 0.5, I, 2, 4, or 6 h after the addition of ActD. andNorthern blots were performed to determine VEGF mRNA expression. Relative VEGFmRNA expression was calculated, and the half-life was determined by plotting represent

ative relative VEGF expression values on a semilogarithmic scale.

VEGF that are 30-fold higher than those of any other organ (28). Thecolon mucosa itself synthesizes significant amounts of IGF-I, andcolon cancers also synthesize IGF-I (49, 50).

The role of IGF-I in malignant progression is not clear, but evidence suggests that IGF-I plays a role in preventing apoptosis innumerous tumor cell systems (reviewed in Ref. 51 ). In addition, IGF-I

may play a role in cell proliferation, although this is a subject ofdebate (26, 52). The role of IGF-I in malignant growth is, at times,

difficult to define because IGFBPs may modulate the biological

effects of IGF-I. These factors can bind IGF-I and either enhance or

inhibit its ability to bind to its receptor, thus affecting its biologicalactivity (this is dependent upon the specific IGFBP). IGFBP-4 has

been identified in human colon cancer cell lines, including the HT29cell line, and may regulate, at least in part, the biological function ofIGF-I (43. 53).

Other investigators have shown that IGF-I increases VEGF expression in a human osteoblast-like cell line (SaOS-2) and in murine

osteoblasts (25). Cells were grown in the presence of various concentrations of IGF-I for several time periods, and VEGF mRNA andprotein expression was found to increase in a dose-dependent fashion,

with peak expression at 2 h. Further studies demonstrated that thisincrease is due to an increase in transcription (25). A more recentpublication has demonstrated that IGF-I increases VEGF expression

in colon carcinoma cell lines (26). This occurs independently of theability of IGF-I to increase cellular proliferation. Furthermore,

Punglia and associates (27) have demonstrated that VEGF protein isincreased in retinal pigment epithelial cells after treatment with IGF-I

and that this induction of VEGF occurs primarily through enhancedtranscription of the VEGF gene.

We have shown here that IGF-I increases VEGF mRNA expression

in human colon carcinoma cell lines. Initial studies were performed intwo representative cell lines, HT29 (no ras mutation) and SW620(ras-mutated cell line), to demonstrate that our findings were not

limited to a single cell line. In addition, we also demonstrated thatVEGF protein levels are increased by IGF-I treatment to show that

changes at the mRNA level translate into changes at the protein level.Once the above observations were confirmed, we investigated mechanisms for this induction of VEGF.

In the two human colon cancer cell lines studied, the peak timepoint for VEGF mRNA induction by IGF-I varied, with SW620 cells

showing the highest VEGF levels at 8 h and HT29 cells showing the

4.4 kb3.7 kb

1.3kb

B24 h

4.4kb3.7kb

1.3kb

Fig. 6. Effect ofIGFBP-4 antibody (IGFBP-4 Ab) and des-(l-3)-IGF-l (des-IGF-1) onVEGF mRNA expression by IGF-I. A, HT29 cells were incubated in the presence orabsence of IGF-I, IGFBP-4 antibody (IGFBP-4 Ab}, or control rabbit antibody (IgG Ab)for 24 h. Total RNA was extracted from the cells, and VEGF mRNA expression wasdetermined by Northern blot analysis. B. HT29 cells were treated with IGF-1 or des-(l-3)-IGF-I for 8 or 24 h. Total RNA was extracted from the cells, and VEGF mRNAexpression was determined by Northern blot analysis.

4012



IGF-] REGULATION OF VEGF

highest expression at 24 h. Investigators examining the role of IGF-I

induction of VEGF in other cell systems have shown levels of VEGFpeaking at time points ranging from as early as 2 h in SaOS-2, retinal

pigment epithelial, and bovine smooth muscle cells to 8 h in anothercolorectal cancer cell line (COLO 205; Refs. 25-27). This again

demonstrates that the effect of certain stimuli may affect different celllines in distinct ways.

We designed our studies to eliminate the effect of other factorsknown to induce VEGF expression. For example, all of our studieswere performed in cells grown to near 100% confluence because wehave previously shown that cell density increases VEGF expression(40). In addition, we treated cells with IGF-I in serum-free medium to

avoid the effects of other growth factors known to be present in serum.However, in subsequent studies,4 we have found that IGF-I in serum-

free medium causes a similar increase in VEGF expression. Consistently in these studies, we have found that 36 h of growth in serum-

free conditions leads to an increase in VEGF expression. In examiningthe effects of various factors on VEGF expression, other investigatorshave compared VEGF expression in treated cells to VEGF expressionin cells untreated at time 0 (25, 26). Because of our previous observation that cell density and serum starvation increase VEGF expression, we included, as a control, untreated cells grown for identicaltime periods. When comparing cells treated with IGF-I to untreatedcells at the same time period, we observed a 3-5-fbld increase in

VEGF expression. However, when the increase in VEGF expressiondue to serum starvation was taken into account, we observed an evengreater increase in VEGF expression, with an expected return tonear-baseline levels after prolonged incubation (Fig. 1).

VEGF induction by IGF-I appears to be regulated by an increase inthe transcription of the VEGF gene. Using VEGF promoter-reporter

constructs, we demonstrated that VEGF promoter activity increasesafter treatment with IGF-I. CoCl->, a chemical mimic of hypoxia, was

used as a positive control and induced a similar increase in VEGFpromoter activity. Furthermore, when transcription was blocked withActD prior to the addition of IGF-I. the induction of VEGF was

likewise blocked. Very little is known about the transcription factorsthat are induced by treatment with IGF-I. It appears that Spl and

another unknown transcription factor, denoted P2. are both necessaryfor the increase in transcription due to activation of the IGF responseelement (54). Other studies have identified a factor designated YY1 asa possible transcription factor induced by IGF-I. The binding se

quence for this protein is unknown at present (55). Factors thatdown-regulate transcription factor expression through IGF-I signalinghave also been identified. IGF-I regulates c-myc mRNA by causing a3-5-fold decrease in c-myc expression in human neuroblastoma cells.

Therefore, transcriptional activation of the VEGF gene may includeboth positive and negative regulators (56).

The half-life of VEGF mRNA was not significantly affected byIGF-I treatment. Data on the role of IGF-I in VEGF mRNA stabilityare conflicting. Warren et al. (26) demonstrated a 3-fold increase in

VEGF mRNA stability when colon cancer cells were treated withIGF-I. In human SaOS-2 osteoblast-like cells, however. Goad et al.

(25) found no increase in VEGF mRNA stability after treatment withIGF-I. Again, these apparently conflicting data may be due to char

acteristics of the specific cell types.IGFBPs are regulators of IGF-I biological activity. They may act by

sequestering IGF-I or by increasing the availability of IGF-I. There isalso evidence to suggest that IGFBPs function independently of IGF-I.

Therefore, it was necessary for us to determine the role of IGFBPs inIGF-I induction of VEGF. Because HT29 cells synthesize IGFBP-4 asthe predominant binding protein (43), we examined the effect of IGF-Ion cells treated with or without neutralizing antibodies to IGFBP-4. Inaddition, we also used a derivative of IGF-I that is devoid of the first

4013

three NH2-terminal amino acids that bind to IGFBPs but have similaraffinity for the IGF-I receptor (57). In both of these studies, it did notappear that IGFBPs affected the induction of VEGF by IGF-I.

Our studies demonstrate that IGF-I induces VEGF mRNA and

protein expression in human colon carcinoma cell lines. This increasein VEGF expression is due to an increase in gene transcription withouta significant effect on mRNA half-life. IGFBPs do not regulate IGF-Iinduction of VEGF. IGF-I may play a role in the site-specific regu

lation of angiogenic factors in colon cancer angiogenesis and growth.

REFERENCES

1. Folkman. J. How is blood vessel growth regulated in normal and neoplaslic tissue?G. H. A. Clowes Memorial Award Lecture. Cancer Res.. 46: 467-473. 1986.

2. Folkman. J. What is the evidence that tumors are angiogenesis dependent? J. Nati.Cancer Inst. (Bethesda), 82: 4-6. 1989.

3. Ferrara. N., Houch. K. A., Jackeman. L. B.. Winer. J.. and Leung. D. W. The vascularendothclial growth factor family of polypeptides. J. Cell. Biochem.. 47: 211-218.

1991.4. Connolly, D. T.. Heuvelman. D. M. Nelson. R.. Olander. J. V.. Eppley. B. L..

Delfino. J. J.. Siegel. N. R.. Leimgruber, R. M., and Feder. J. Tumor vascularpermeability factor stimulates cndolhelial cell growth and angiogenesis. J. Clin.Invest.. 84: 1470-1478. 1989.

5. Wilting, J.. Christ. B.. Bokeloh. M.. and Weich. H. A. In vivo effects of vascularendothelial growth factor on the chick chorioallanloic membrane. Cell Tissue Res..274.- 163-172. 1993.

6. Kim. K. J.. Li. B.. Winer. J.. Armanini. M.. Gillett. N.. Phillips. H. S.. and Ferrara. N.Inhibition of vascular endothetial growth factor-induced angiogenesis suppressestumour growth in vivo. Nature (Lond.l, 362: 841-844. 1993.

7. Ferrara, N., Carver. M. K.. Chen. H., Dowd, M.. Lu. L.. O'Shea. K. S.. Powell. B. L..

Ihll.in. K. J.. and Moore. M. W. Hetcro/ygous embryonic lethality induced bytargeted inactivation of the VEGF gene. Nature (Lond.). 380: 439-42. 1996.

8. Brown. L. F.. Berse. B.. Jackman. R. W.. Tognaz/.i. K.. Manseau. E. J.. Senger. D. R..and Dvorak. H. F. Expression of vascular permeability factor (vascular cndothclialgrowth factor) and its receptors in adenocareinomas of the gastrointestinal tract.Cancer Res.. 5.?: 4727-4735. 1993.

9. Brown. L. F.. Berse. B., Jackman. R. W.. Tognaz/.i. K.. Manseau. E. J.. Dvorak. H. F..and Senger. D. R. Increased expression of vascular permeability factor (vascularendothelial growth factor) and its receptors in kidney and bladder carcinomas. Am. J.Pathol.. 143: 1255-1262. 1993.

10. Berkman. R. A.. Merrill. M. J.. Reinhold. W. C.. Monacci. W. T., Saxena. A.. Clark.W. C.. Robertson. J. T.. Ali. 1. U.. and Oldfield. E. H. Expression of (he vascularpermeability factor/vascular endothelial growth factor gene in central nervous systemneoplasms. J. Clin. Invest.. 91: 153-159. 1993.

11. Takahashi. Y.. Kitadai, Y.. Bucana. C. D.. Cleary. K. R.. and Ellis. L. M. Expressionof vascular endothelial growth factor and its receptor. KDR. correlates with vascu-larily. metastasis, and proliferation of human colon cancer. Cancer Res.. 55: 3964-

3968. 1995.12. Takahashi. Y.. Cleary. K. R.. Mai. M.. Kitadai. Y.. Bucana, C. D.. and Ellis, L. M.

Significance of vessel count and vascular cndolhclial growth factor and its receptor(KDR) in intestinal-type gastric cancer. Clin. Cancer Res.. 2: 1679-1684. 1996.

13. Toi. M.. Kashitani. J.. and Tominaga. T. Tumor angiogenesis is an independentprognostic indicator in primary breast carcinoma. Ini. J. Cancer. 55: 371-374. 1993.

14. Maliern. J.. and Volm. M. Microvessel density and vascular endothelial growth factorexpression in human tumors of different localisation. Oncol. Rep., 3: 465-468. 1996.

15. Ilakura. J.. Ishikawa. T.. Friess, H.. Fujii. H.. and Matsumoto. Y. Enhanced expressionof vascular endothelial growth factor in human pancreatic cancer correlates with localdisease progression. Clin. Cancer Res.. 3: 1309-1316. 1997.

16. Tanigawa. N.. Amaya. H.. Matsumura. M.. and Shimomatsuya. T. Correlation between expression of vascular endothelial growth factor and tumor vascularily, andpatient outcome in human gastric carcinoma. J. Clin. Oncol.. /5: 826-832. 1997.

17. Warren. R. S.. Yuan, H.. Malli. M. R.. Gillett. N. A., and Ferrara, N. Regulation byvascular endothelial growth factor of human colon cancer tumorigenesis in a mousemodel of experimental liver metastasis. J. Clin. Invest.. 95: 1789-1797. 1995.

18. Ellis. L. M.. Liu. W., and Wilson. M. Downregulation of vascular endothelial growthfactor in human colon carcinoma cell lines by antisense transfection decreasesendothelial cell proliferation. Surgery (St. Louis). 120: 871-878. 1996.

19. Shima. D. T., Deutsch. U., and D'Amore. P. A. Hypoxie induction of vascular

endothelial growth factor (VEGF) in human epithelial cells is mediated by increasesin mRNA stability. FEBS Lett.. 370: 203-208. 1995.

20. Shweiki. D.. Kin. A.. Stoffer. D., and Keshet. E. Vascular endothelial growth factorinduced by hypoxia may mediate hypoxia-iniliatcd angiogenesis. Nature (Lond.).359: 843-845. 1992.

21. Levy, A. P.. Levy. N. S.. Wegner. S., and Goldberg. M. A. Transcriptional regulationof the rat vascular endothelial growth factor gene by hypoxia. J. Biol. Chem., 270:13333-13340, 1995.

22. Levy. A. P.. Levy. N. S.. and Goldberg. M. A. Post-transcripiional regulation ofvascular endolhelial growth factor by hypoxia. J. Biol. Chem.. 277: 2746-2753.

1996.23. Baserga. R. The insulin-like growth factor I receptor: a key to tumor growth? Cancer

Res., 55: 249-252, 1995.



IGF-I REGULATION OF VEGF

24. Daughaday, W. H. The possible autocrine/paracrine and endocrine roles of insulin-like growth factors of human tumors. Endocrinology. 127: 1-4, 1990.

25. Goad, D. L.. Rubin, J., Wang, H.. Tashjian. A. H.. and Patterson. C. Enhancedexpression of vascular endolhelia! growth factor in human SaOS-2 osteoblast-likecells and murine osteoblasts induced by insulin-like growth factor I. Endocrinology.137: 2262-2268. 19%.

26. Warren. R. S.. Yuan. H., Malli. M. R.. Ferrara. N., and Donner, D. B. Induction ofvascular endothclial growth factor by insulin-like growth factor 1 in colorectal cancer.J. Biol. Chem.. 271: 29483-29488. 1996.

27. Punglia. R. S.. Lu. M.. Hsu, J., Kuroki. M.. Tolentino, M. J., Keough, K.. Levy, A. P..Levy. N. S.. Goldberg. M. A.. D'Amato. R. J.. and Adamis, A. P. Regulation of

vascular endothelial growth factor expression by insulin-like growth factor I. Diabetes, 46: 1619-1626, 1997.

28. Singh, P.. and Rubin. N. Insulinlike growth factors and binding proteins in coloncancer. Gastroenterology, 105: 1218-1237. 1993.

29. Czech, M. P. Signal transmission by the insulin-like growth factors. Cell. 59:235-238, 1989.

30. Janicot, M.. Flores-Riveros, J. R., and Lane, M. D. The insulin-like growth factor I(IGF-1 ) receptor is responsible for mediating the effect of insulin. IGF-1. and IGF-2in Xenopus laevis oocytes. J. Biol. Chem.. 266. 9382-9391, 1991.

31. Busby. W. J., Klapper. D. G.. and Clemmons. D. R. Purification of a 31,000-daltoninsulin-like growth factor binding protein from human amniotic fluid. Isolation of twoforms with different biologic actions. J. Biol. Chem., 263: 14203-14210. 1988.

32. Li, J.. Perrella, M. A., Tsai. J. C., Yet. S. F.. Hsieh, C. M., Yoshizumi. M.. Patterson.C.. Endege. W. O.. Zhou, F.. and Lee. M. E. Induction of vascular endothclial growthfactor gene expression by inlerleukin-lÃŸ in rat aortic smooth muscle cells. J. Biol.Chem.. 270: 308-312. 1995.

33. Cohen, T.. Nahari. D.. Cerem. L. W., Neufeld, G., and Levi, B. Z. Inlerleukin 6induces the expression of vascular endothelial growth factor. J. Biol. Chem., 271:736-741, 1996.

34. Slavri, G. T.. Hong, Y., Zachary, 1. C.. Breier. G., Baskerville. P. A.. Yla-Herttuala.S.. Risau. W.. Martin. J. F.. and Erusalimsky, J. D. Hypoxia and platelet-derivedgrowth factor-BB synergistically upregulate the expression of vascular endothelialgrowth factor in vascular smooth muscle cells. FEBS Lett., 358: 311-315, 1995.

35. Dolecki, G. J., and Connolly, D. T. Effects of a variety of cytokines and inducingagents on vascular permeability factor mRNA levels in U937 cells. Biochem. Bio-phys. Res. Commun., 180: 572-578, 1991.

36. Koochekpour, S.. Merzak, A., and Pilkington, G. J. Vascular endothelial growthfactor production is stimulated in response to growth factors in human glioma cells.Oncol. Rep., 2: 1059-1061. 1995.

37. Silvagno. F.. Follenzi, A.. Arese. M.. Prat. M.. Giraudo. E.. Caudino. G.. Camussi, G..Comoglio. P. M., and Bussolino. F. In vivo activation of met lyrosine kinase byheterodimeric hepatocyte growth molecule promotes angiogenesis. Arterioscler.Thromb. Vase. Biol.. 15: 1857-1865, 1995.

38. Ellis, L. M., Slaley, C. A., Liu. W., Fleming, R. Y. D.. Parikh, N.. Bucana. C. D.. and(iallick. G. E. Down-regulation of vascular endothelial growth factor in a humancolon carcinoma cell line Iransfected with an anlisense expression vector specific forc-src. J. Biol. Chem., 273: 1052-1057, 1998.

39. Fleming. R. Y. D.. Ellis, L. M.. Parikh. N. U.. Liu. W.. Staley, C. A., and Gallick,G. E. Regulation of vascular endothelial growth factor expression in human coloncarcinoma cells by activity of Src kinase. Surgery (St. Louis), 122: 501-507. 1997.

40. Koura. A. N., Radinsky. R.. Kiladai. Y.. Takahashi. Y.. Liu. W.. Singh, R.. and Ellis.L. M. Regulation of vascular endothelial growth factor expression in human coloncarcinoma cells hy cell density. Cancer Res., 5ft: 3891-3894, 1996.

41. Berse, B.. Brown, L. F.. vanDeWater. L., Dvorak, H. A., and Senger. D. R. Vascularpermeability factor (vascular endothelial growth factor) is expressed differentially innormal tissues, macrophages, and tumors. Mol. Biol. Cell. 3: 211-220. 1992.

42. Minchenko. A.. Salceda. S., Bauer, T.. and Caro. J. Hypoxia regulatory elements ofthe human vascular endothelial growth factor gene. Cell. Mol. Biol. Res., 40: 35-39,

1994.43. Culouscou, J. M., and Shoyab. M. Purification of a colon cancer cell growth inhibitor

and its identification as an insulin-like growth factor binding protein. Cancer Res.. 51:2813-2819. 1991.

44. Sara. V. R., and Hall. K. Insulin-like growth factors and their binding proteins.Physiol. Rev., 70: 591-614, 1990.

45. Fidler, 1. J.. and Ellis. L. M. The implications of angiogenesis to the biology andtherapy of cancer metastasis. Cell. 79: 185-188. 1994.

46. Takahashi. Y.. Tucker. S. L.. Kiladai, Y.. Koura, A. N.. Bucana. C. D.. Cleary, K. R.,and Ellis, L. M. Vessel counts and VEGF expression as prognostic factors innode-negative colon cancer. Arch. Surg.. 132: 541-546, 1997.

47. Frank. R. E.. Saclarides, T. J., Leurgans, S., Speziale, N. J., Drab, E. A., and Rubin.D. B. Tumor angiogenesis as a predictor of recurrence and survival in palienls withnode-negative colon cancer. Ann. Surg.. 222: 695-699. 1995.

48. Ikeda. E.. Achen. M. G., Breier. G., and Risau, W. Hypoxia-induced Iranscriptional

activation and increased mRNA stability of vascular endothelial growth factor in C6glioma cells. J. Biol. Chem., 34: 19761-19768, 1995.

49. Tricoli. J. V., Rail, L. B., Karakousis. C. P.. Herrera, L.. Petrelli. N. J., Bell, G. I., andShows. T. B. Enhanced levels of insulin-like growth faclor messenger RNA in humancolon carcinomas and liposarcomas. Cancer Res.. 46: 6169-6173, 1986.

50. Michel!. N. P., Langman. M. J., and Eggo, M. C. Insulin-like growth faclors andIheir binding proteins in human colonocyles: preferential degradation of insulin-like growlh faclor binding prolein 2 in colonie cancers. Br. J. Cancer, 76: 60-66.

1997.51. Baserga. R.. Hongo. A.. Rubini. M.. Prisco, M., and Valenlinis. B. The IGF-I receplor

in cell growlh. Iransformalion and apoptosis. Biochim. Biophys. Acta. 1332: F105-

FI26, 1997.52. Quo, Y. S.. Narayan. S.. Yallampalli. C.. and Singh. P. Characterization of insulinlike

growth factor 1 receptors in human colon cancer. Gastroenterology. 102: 1101-1108,

1992.53. Culouscou. J. M., Remacle-Bonnel, M., Garrouste. F., Fantini. J.. Marvaldi, J., and

Pommier. G. Production of insulin-like growth factor II (IGF-I1) and different formsof IGF-binding proteins by HT-29 human colon carcinoma cell line. J. Cell. Physiol..143: 405-415, 1990.

54. Urban. R. J., and Bodcnburg. Y. Transcriptional activalion of the porcine P450 11Ainsulin-like growlh faclor response element in MCF-7 breasl cancer cells. J. Biol.Chem.. 271: 31695-31698, 1996.

55. Flanagan. J. R. Aulologous slimulation of YY1 Iranscriplion faclor expression: role ofan insulin-like growlh faclor. Cell Growlh Differ., 6: 185-190. 1995.

56. Simulili,HI. V. N., and Feldman, E. L. Insulin-like growlh faclor I regÃºlalesc-myc andGAP-43 messenger ribonucleic acid expression in SH-SY5Y human neuroblaslomacells. Endocrinology, 132: 2017-2023, 1993.

57. Ballard, F. J., Francis, G. L., Ross, M., Bagley. C. J.. May, B.. and Wallace. J. C.Nalural and synthelic forms of insulin-like growth factor-1 (IGF-1) and the potentderivative, destripeptide IGF-1: biological activities and receptor binding. Biochem.Biophys. Res. Commun.. I4V: 398-404, 1987.

4014



1998;58:4008-4014. Cancer Res Yoshito Akagi, Wenbiao Liu, Brian Zebrowski, et al. Human Colon Cancer by Insulin-like Growth Factor-IRegulation of Vascular Endothelial Growth Factor Expression in

Updated version

http://cancerres.aacrjournals.org/content/58/17/4008

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/58/17/4008To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



regulation of vascular endothelial growth factor expression in

Documents