vascular endothelial growth factor, interleukin 8, platelet-derived ... · vascular endothelial...

[CANCER RESEARCH 60, 4932–4938, September 1, 2000]

Vascular Endothelial Growth Factor, Interleukin 8, Platelet-derived EndothelialCell Growth Factor, and Basic Fibroblast Growth Factor PromoteAngiogenesis and Metastasis in Human Melanoma Xenografts1

Einar K. Rofstad2 and Ellen F. HalsørDepartment of Biophysics, Institute for Cancer Research, The Norwegian Radium Hospital, N-0310 Oslo, Norway

ABSTRACT

Angiogenesis is a significant prognostic factor in melanoma, but theangiogenic factors controlling the neovascularization are not well defined.The purpose of this study was to investigate whether the angiogenesis andmetastasis of melanoma are promoted by vascular endothelial growthfactor (VEGF), interleukin 8 (IL-8), platelet-derived endothelial cellgrowth factor (PD-ECGF), and/or basic fibroblast growth factor (bFGF).Cells from human melanoma lines (A-07, D-12, R-18, and U-25) trans-planted to BALB/c nu/nu mice were used as tumor models. Expression ofangiogenic factors was studied by ELISA, Western blotting, and immu-nohistochemistry. Angiogenesis was assessed by using an intradermalangiogenesis assay. Lung colonization and spontaneous lung metastasiswere determined after i.v. and intradermal inoculation of tumor cells,respectively. The specific roles of VEGF, IL-8, PD-ECGF, and bFGF intumor angiogenesis, lung colonization, and spontaneous metastasis wereassessed in mice treated with neutralizing antibody. The melanoma linesexpressed multiple angiogenic factors, and each line showed a uniqueexpression pattern. Multiple angiogenic factors promoted angiogenesis inthe most angiogenic melanoma lines, whereas angiogenesis in the leastangiogenic melanoma lines was possibly promoted solely by VEGF. Tu-mor growth, lung colonization, and spontaneous metastasis were con-trolled by the rate of angiogenesis and hence by the angiogenic factorspromoting the angiogenesis. Lung colonization and spontaneous metasta-sis in A-07 were inhibited by treatment with neutralizing antibody againstVEGF, IL-8, PD-ECGF, or bFGF. Each of these angiogenic factors maypromote metastasis in melanoma, because inhibition of one of them couldnot be compensated for by the others. Our observations suggest thatefficient antiangiogenic treatment of melanoma may require identificationand blocking of common functional features of several angiogenic factors.

INTRODUCTION

Malignant melanoma is a highly metastatic tumor type of increas-ing incidence, poor prognosis, and high resistance to treatment (1–3).The progression of melanoma is characterized by distinct sequentialsteps, from common acquired melanocytic nevus via dysplastic nevus,radial growth phase primary melanoma, and vertical growth phaseprimary melanoma to melanoma metastasis (4, 5). There is increasingevidence that melanoma progression requires induction of new cap-illary blood vessels, a process called angiogenesis (6–12). Thus, thetransition from the radial to the vertical growth phase, which repre-sents a significant worsening of the prognosis, has been shown to bedependent on neovascularization (7–9). Moreover, the probability ofmetastasis has been found to increase with increasing microvesseldensity of the primary tumor (10–12).

Melanoma cells have been shown to produce and secrete a wide

variety of angiogenic factors, including VEGF3 (13–16), IL-8 (17–19), PD-ECGF (20, 21), and bFGF (22–24). VEGF, also known asvascular permeability factor, is a strong specific mitogen for endo-thelial cells and may also stimulate endothelial cell migration andreorganization (25, 26). IL-8, which belongs to the superfamily ofCXC chemokines, is a multifunctional cytokine that exhibits potentangiogenic activities bothin vitro andin vivo (27, 28). PD-ECGF, alsoknown as thymidine phosphorylase and gliostatin, stimulates endo-thelial cell mitogenesis and chemotaxisin vitro and is strongly an-giogenicin vivo, possibly through modulation of nucleotide metabo-lism (29). bFGF, which belongs to the family of heparin-bindinggrowth factors, is a multifunctional protein having a well-establishedkey role in tumor angiogenesis (30–32).

There is significant experimental evidence that these angiogenicfactors are involved in the growth and metastasis of malignant mel-anoma. Inoculation of human melanoma cells transfected with thegene encoding VEGF into immune-deficient mice results in tumorswith increased vascularization (13–15), microvessel permeability (14,15, 33), and volumetric growth rate (14, 33). The constitutive expres-sion of IL-8 in human melanoma cells is correlated with their lungcolonization efficiency after i.v. inoculation in immune-deficient mice(17). Human melanoma cells show increased lung colonization effi-ciency after transfection with the gene encoding VEGF (15) or IL-8(18). Moreover, xenografted tumors initiated from human melanomacell lines transfected with antisense-VEGF (14, 15) or antisense-bFGF(24) show reduced vascularization and growth.

A panel of angiogenic factors may thus be involved in the angio-genesis, growth, and metastasis of melanomas, and the panel maydiffer among tumors differing in angiogenic activity (34–36). How-ever, angiogenic factors that show sufficiently high constitutive ex-pression to promote spontaneous metastasis have not been identifiedconclusively thus far. The work reported here was aimed at investi-gating whether the angiogenesis and metastasis of melanoma may bepromoted by VEGF, IL-8, PD-ECGF, and/or bFGF. Four humanmelanoma xenograft lines showing widely different constitutive ex-pression of these angiogenic factors were included in the study. Thespecific role of each of the factors was investigated by assessingtumor angiogenesis and metastasis in mice treated with neutralizingantibody.

MATERIALS AND METHODS

Mice. Adult (8–10 weeks of age) female BALB/cnu/numice were used toassess tumor angiogenesis, lung colonization, and spontaneous metastasis. Themice were bred at our institute and maintained under specific pathogen-freeconditions at constant temperature (24–26°C) and humidity (30–50%). Ster-ilized food and tap water were givenad libitum.

Cell Lines. Four human melanoma cell lines (A-07, D-12, R-18, and U-25)were included in the study (37). They were maintained in monolayer culture inRPMI 1640 (25 mM HEPES andL-glutamine) supplemented with 13% bovinecalf serum, 250 mg/l penicillin, and 50 mg/l streptomycin. The cultures were

Received 12/29/99; accepted 7/6/00.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby markedadvertisementin accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 Supported by grants from The Norwegian Cancer Society.2 To whom requests for reprints should be addressed, at Department of Biophysics,

Institute for Cancer Research, The Norwegian Radium Hospital, Montebello, N-0310Oslo, Norway. Phone: 47-2293-4279; Fax: 47-2293-4270; E-mail: [email protected].

3 The abbreviations used are: VEGF, vascular endothelial growth factor; bFGF, basicfibroblast growth factor; IL, interleukin; PD-ECGF, platelet-derived endothelial cellgrowth factor.

4932

on June 3, 2020. © 2000 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

incubated at 37°C in a humidified atmosphere of 5% CO2 in air and subcul-tured twice a week. The cell lines were verified to be free fromMycoplasmacontamination.

ELISA. Commercial ELISA kits (Quantikine; R&D Systems, Abingdon,United Kingdom) were used according to the instructions of the manufacturerto measure VEGF, IL-8, and bFGF concentrations in culture medium. Mediumsamples from cultures in exponential growth were collected 24 h after changeof medium. Cell numbers at the time of medium change and at the time ofmedium collection were determined by using a hemocytometer and a phasecontrast microscope. The medium samples were centrifuged to remove parti-cles, diluted to appropriate concentrations, and analyzed in duplicate at severaldilutions. Absorbances were determined at 450 nm. Readings at 570 nm weresubtracted from the readings at 450 nm to correct for optical imperfections. Astandard curve was obtained by linear regression analysis of protein concen-tration versusabsorbance in a double logarithmic plot. The rate of proteinsecretion (Rsec) was calculated as:

Rsec5VDC

NiDt3

ln~Nf/Ni!

~Nf/Ni 2 1)

whereDC is the increase in protein concentration during the time intervalDt.Here,Dt was 24 h.Ni andNf are the initial and final cell numbers, andV is thevolume of medium. Replicate cultures were used to determineNi. The secondfactor of the product is based on the assumption that the cell number increasedexponentially with time duringDt, an assumption that was verified to be valid.

Western Blotting. Cells from cultures in exponential growth were washedin PBS and boiled in Laemmli lysis buffer for 5 min (38). Proteins wereseparated by SDS-PAGE and transferred to a polyvinylidene fluoride mem-brane. Membranes were incubated with antihuman VEGF rabbit polyclonalantibody (Santa Cruz Biotechnology, Santa Cruz, CA), antihuman IL-8 mousemonoclonal antibody (R&D Systems), antihuman PD-ECGF goat polyclonalantibody (R&D Systems), or antihuman bFGF rabbit polyclonal antibody(Oncogene Science, Cambridge, MA) for 60 min. Bound antibody was de-tected by using a biotin-streptavidin alkaline phosphatase staining procedure.Recombinant human VEGF, IL-8, PD-ECGF, or bFGF (R&D Systems) wasused as positive control. The specificity of the antibody-antigen interactionswas confirmed by peptide competition studies and by incubation of membranesin solutions without primary antibody. Protein molecular weights were esti-mated by using broad range or high range prestained standards according to theinstructions of the manufacturer (SDS-PAGE standards; Bio-Rad Laboratories,Hercules, CA).

Immunohistochemistry. Immunohistochemical staining of tumor tissuewas performed by using an indirect immunoperoxidase method. Tumors werefixed in phosphate-buffered 4% paraformaldehyde or snap-frozen in liquidnitrogen. Antihuman VEGF rabbit polyclonal antibody (Santa Cruz Biotech-nology), antihuman IL-8 rabbit polyclonal antibody (Endogen, Woburn, MA),antihuman PD-ECGF goat polyclonal antibody (R&D Systems), or antihumanbFGF rabbit polyclonal antibody (Santa Cruz Biotechnology) was used asprimary antibody. Controls included omission of the primary antibody, incu-bation with normal rabbit immunoglobulin or normal rabbit serum, and incu-bation with blocking peptides before staining. The sections were counter-stained with hematoxylin.

Angiogenesis Assay.Tumor angiogenesis was assessed by using an intra-dermal angiogenesis assay (39). A 100-ml Hamilton syringe was used toinoculate aliquots of 10ml of cell suspension into the flanks of mice. Theinoculation point lay above the s.c. muscle tissue in the deeper part of thedermis. The number of cells per inoculum was 1.03 106. The mice were killedon day 7 after the inoculation. Small vascularized tumors had developed in theinoculation sites at that time. The skin around the inoculation sites wasremoved, and the tumors were located with a dissecting microscope. Thecapillaries in the dermis oriented toward the tumors were counted, and the di-ameters of the tumors were measured, using an ocular with a scale. Thenumber of capillaries was corrected for the background, determined afterthe injection of 10ml of HBSS. Angiogenesis was quantified as number ofcapillaries per tumor or number of capillaries per mm of tumor circumference.

Lung Colonization Assay. Aliquots of 3.03 106 A-07 cells or 3.03 105

D-12 cells suspended in 0.2 ml of HBSS were inoculated into the lateral tailvein of mice by using a tuberculin syringe with a 26-gauge needle. The micewere killed and autopsied 5 weeks after the inoculation. The lungs were

removed, rinsed in HBSS, and fixed in Bouin’s solution for 24 h to facilitatethe scoring of colonies. The number of surface colonies was determined byusing a stereomicroscope.

Spontaneous Metastasis Assay.Aliquots of 3.53 105 A-07 or D-12 cellssuspended in 10ml of HBSS were inoculated intradermally in the left flank ofmice, using the same procedure as used in the angiogenesis assay describedabove. The inoculations were performed 24 h after the mice had been immu-nosuppressed by 5.0 Gy of whole-body irradiation. The whole-body irradiationwas performed by using a Siemens Stabilipan X-ray unit, operated at 220 kV,19–20 mA, and with 0.5-mm copper filtration (40). The tumors were removedsurgically when the largest diameter had attained;10 mm, and the woundswere closed with surgical clips. The mice were examined for clinical signs ofmetastases twice a week. They were killed and autopsied 3 months after theprimary tumor was removed or when they were moribund. The lungs wereexamined for the presence of macroscopic metastases as described above.

Treatment with Neutralizing Antibody in Vivo. The specific roles ofVEGF, IL-8, PD-ECGF, and bFGF in tumor angiogenesis, lung colonization,and spontaneous metastasis were investigated by treating host mice withneutralizing antibody against these angiogenic factors. The antibodies used fortreatment were antihuman VEGF mouse monoclonal antibody (R&D Sys-tems), antihuman IL-8 mouse monoclonal antibody (R&D Systems), anti-human PD-ECGF goat polyclonal antibody (R&D Systems), and antihumanbFGF goat polyclonal antibody (R&D Systems). The antibody solutions werediluted in PBS and administered to the mice in volumes of 0.25 ml by i.p.injection. In the angiogenesis and lung colonization experiments, the treat-ments consisted of four doses of 25mg (VEGF and bFGF) or 100mg (IL-8 andPD-ECGF) of antibody given in intervals of 24 h. The first dose was given 1 hbefore the tumor cell inoculation. In the spontaneous metastasis experiments,the treatments consisted of eight doses of antibody given daily the last 8 daysbefore the primary tumor was removed.

Treatment with Neutralizing Antibody in Vitro. Possible cytotoxic orantiproliferative effects of the neutralizing antibodies described above wereinvestigatedin vitro. A-07, D-12, R-18, or U-25 cells were cultured in RPMI1640 (25 mM HEPES andL-glutamine) supplemented with 13% bovine calfserum, 250 mg/l penicillin, and 50 mg/l streptomycin in the absence orpresence of 5mg/ml of antibody for up to 8 days. The number of cells in thecultures was determined 2, 4, 6, or 8 days after the cultures were initiated bycounting cells in a hemocytometer.

Statistical Analysis. Results are presented as arithmetic mean6 SE.Statistical comparisons of data were performed by one-way ANOVA. Whensignificant differences were found, the Dunnett’s method was used to identifythe data sets that differed from the control data. Probability values ofP , 0.05were considered significant. AllPs were determined from two-sided tests. Thestatistical analysis was performed by using SigmaStat statistical software(Jandel Scientific GmbH, Erkrath, Germany).

RESULTS

Significant secretion of VEGF and IL-8 was seen in all cell lines,whereas A-07 was the only line that showed detectable secretion ofbFGF (Table 1). A-07 showed higher VEGF secretion than U-25(P , 0.05), which in turn showed higher VEGF secretion than D-12and R-18 (P, 0.05). The IL-8 secretion was also higher in A-07 thanin the other lines (P, 0.05). Moreover, D-12 showed higher IL-8secretion than U-25 (P, 0.05), which in turn showed higher IL-8secretion than R-18 (P, 0.05).

All cell lines showed significant expression of VEGF, IL-8, andbFGF, whereas A-07 was the only line that showed significant ex-

Table 1 Rate of secretion of angiogenesis factors of human melanoma cell lines

Melanoma

Secretion rate (pg/106 cells•h)a

VEGF IL-8 bFGF

A-07 7146 140 5716 153 1.96 0.8D-12 186 1 2916 86 0R-18 166 2 46 1 0U-25 856 8 1096 27 0

a Mean6 SE of five independent experiments.

4933

MELANOMA ANGIOGENESIS AND METASTASIS



pression of PD-ECGF, as revealed by Western blot analysis (Fig. 1).A-07 showed substantially higher VEGF, IL-8, and bFGF expressionthan the other lines. Moreover, six specific bFGF bands were seen inA-07, whereas D-12, R-18, and U-25 showed only 2–4 specific bFGFbands.

A-07 tumors showed significant expression of VEGF, IL-8, PD-ECGF, and bFGF, as revealed by immunohistochemistry (Fig. 2).Also D-12, R-18, and U-25 stained positive for VEGF, IL-8, andbFGFin vivo, but the staining was weaker than that in A-07 tumors forall proteins. Significant expression of PD-ECGF could not be detectedin D-12, R-18, or U-25 tumors.

Intradermal inoculation of A-07, D-12, R-18, or U-25 cells in miceevoked a strong angiogenic response (Fig. 3). The angiogenic re-sponse, measured as number of tumor-oriented capillaries per tumor atday 7 after the inoculation of 1.03 106 cells, differed substantiallyamong the lines. A-07 evoked the strongest angiogenic response,followed by D-12, R-18, and U-25. The differences in number oftumor-oriented capillaries between A-07 and D-12, D-12 and R-18,and R-18 and U-25 were all statistically significant (P , 0.05). Theangiogenic response evoked by A-07 was inhibited in mice treatedwith neutralizing antibody against VEGF, IL-8, PD-ECGF, or bFGF(P , 0.05). Treatment with neutralizing antibody against VEGF orIL-8 also reduced the angiogenic response evoked by D-12 (P , 0.05)but not treatment with neutralizing antibody against PD-ECGF orbFGF (P. 0.05). The angiogenic response evoked by R-18 and U-25was inhibited in mice treated with neutralizing antibody againstVEGF (P , 0.05) but not in mice treated with neutralizing antibodyagainst IL-8, PD-ECGF, or bFGF (P . 0.05). Similar results andidentical conclusions were achieved by expressing angiogenic re-sponse as number of tumor-oriented capillaries per mm of tumorcircumference (data not shown).

i.v. inoculation of A-07 or D-12 cells in mice resulted in significantlung colonization (Fig. 4). The lung colonization efficiency of A-07was reduced in mice treated with neutralizing antibody against VEGF,

IL-8, PD-ECGF, or bFGF (P, 0.05). Treatment with neutralizingantibody against VEGF or IL-8 also reduced the lung colonizationefficiency of D-12 (P, 0.05) but not treatment with neutralizingantibody against PD-ECGF or bFGF (P. 0.05). R-18 and U-25 cellsdo not develop lung colonies after i.v. inoculation in mice (41).

Intradermal A-07 and D-12 tumors developed spontaneous lungmetastases in;50% of the mice (Fig. 5). The metastatic efficiency ofA-07 tumors was reduced in mice treated with neutralizing antibodyagainst VEGF, IL-8, PD-ECGF, or bFGF (P , 0.05). Treatment withneutralizing antibody against VEGF or IL-8 also reduced the meta-static efficiency of D-12 tumors (P, 0.05) but not treatment withneutralizing antibody against PD-ECGF or bFGF (P . 0.05).

Treatment with neutralizing antibody against VEGF, IL-8, PD-ECGF, or bFGFin vitro had no cytotoxic or antiproliferative effectson A-07, D-12, R-18, or U-25 cells; the number of cells in antibody-treated cultures was not significantly different from that in controlcultures (P. 0.05), irrespective of cell line, treatment time, andneutralizing antibody (data not shown).

DISCUSSION

A-07, D-12, R-18, and U-25 melanoma cells expressed multipleangiogenic factorsin vitro, and each cell line showed a uniqueexpression pattern. Thus, A-07 showed high expression and secretionof VEGF, IL-8, and bFGF and significant expression of PD-ECGF.D-12, R-18, and U-25 showed significant expression of VEGF, IL-8,and bFGF, but the expression of each of these factors was lower thanthat in A-07. D-12, R-18, and U-25 differed from A-07 also byshowing no expression of PD-ECGF and no secretion of bFGF. D-12differed from R-18 and U-25 by showing substantially higher secre-tion of IL-8. Finally, U-25 differed from R-18 by showing signifi-cantly higher secretion of VEGF and IL-8. These observations areconsistent with previous observations suggesting that melanoma celllines may differ substantially in the expression of angiogenic factors(34–36).

The differences among A-07, D-12, R-18, and U-25 cells in theexpression of VEGF and IL-8 were more pronounced when theexpression was measured by ELISA than when measured by Westernblotting. Western blot analysis gives information on intracellularprotein concentrations, whereas ELISA analysis provides informationon the rate of protein secretion. There is noa priori reason to believethat the rate of protein secretion should be correlated to the intracel-lular concentration of protein. In fact, the present study suggests thatthe rate of protein secretion cannot be predicted from the intracellularprotein concentration as determined by Western blot analysis. The rateof protein secretion is probably more relevant for the rate of tumorneovascularization than is the intracellular concentration of protein,because angiogenesis is induced by binding of protein to endothelialcell receptors.

A-07 cells, which evoked the strongest angiogenic responsein vivo,showed substantially higher expression of bFGF than did D-12, R-18,and U-25 cells. bFGF may promote tumor angiogenesis directly byacting synergistically with VEGF (42–44) and indirectly by up-regulating the synthesis and secretion of VEGF (45, 46).

The expression of VEGF, IL-8, PD-ECGF, and bFGF of A-07,D-12, R-18, and U-25 tumors, determined by immunohistochemicalanalysis, was in agreement with that of the corresponding cell linesinvitro, i.e., A-07 tumors showed substantially stronger staining forVEGF, IL-8, and bFGF than did D-12, R-18, and U-25 tumors andwere the only tumors that stained positive for PD-ECGF, consistentwith the in vitro data in Table 1 and Fig. 1. This observation was notexpected, because A-07, D-12, R-18, and U-25 tumors differ substan-tially in the fraction of hypoxic cells (47), and hypoxia may

Fig. 1. Western blots showing the expression of VEGF, IL-8, PD-ECGF, or bFGF inexponentially growing A-07, D-12, R-18, and U-25 monolayer cell cultures. Proteins from5.03 105 cells were loaded into each lane. Specific bands are indicated by thearrows tothe left. The other bands were attributable to nonspecific staining, as shown by peptidecompetition studies and in blots where incubation with primary antibody was omitted. Thearrows to the right indicate the positions of standards with known molecular weights.

4934




up-regulate the expression of VEGF (48, 49), IL-8 (19, 50), andPD-ECGF (51). The expression of VEGF has been shown to beup-regulated in A-07, D-12, R-18, and U-25 cells exposed to hypoxiain vitro (16, 52). However, hypoxia-induced up-regulation of VEGF,IL-8, or PD-ECGF was not detected in the immunohistochemicalanalysis. All tissue sections stained homogeneously for VEGF, IL-8,and PD-ECGF,i.e., the staining was not more intense in regionsadjacent to necrosis than in regions close to functional capillaries,possibly because VEGF, IL-8, and PD-ECGF were secreted anddiffused from hypoxic tissue into surrounding normoxic tissue. Thestaining patterns of VEGF, IL-8, and PD-ECGF differed substantiallyfrom the staining pattern of the hypoxia marker pimonidazole in A-07,D-12, R-18, and U-25 tumors (53).

The angiogenic factors expressedin vivo were produced primarilyby the parenchymal melanoma cells and were thus of human origin.The immunohistochemical analysis showed that the angiogenic fac-tors were mainly localized intracellularly in melanoma cells and not instromal host cells. Moreover, host cells,i.e., macrophages, leukocytes,and fibroblasts, have been shown to constitute,1% of the totalnumber of cells in A-07, D-12, R-18, and U-25 tumors (37). Ourconclusion is also supported by the observation that the expression ofangiogenic factorsin vivo was similar to thatin vitro.

The angiogenic response after intradermal cell inoculation differedsignificantly among the melanoma lines. The sequence of the lines

from high to low angiogenic response, measured as number of tumor-oriented capillaries per tumor or number of tumor-oriented capillariesper mm of tumor circumference, was A-07. D-12 . R-18. U-25.The tumor growth rate has also been shown to differ significantlyamong the melanoma lines (39). The sequence of the lines from highto low volumetric growth rate is identical to that for angiogenicresponse,i.e., A-07 . D-12 . R-18 . U-25. The close associationbetween volumetric growth rate and angiogenic response suggests thatthe growth of A-07, D-12, R-18, and U-25 tumors is governedprimarily by the rate of angiogenesis.

The angiogenic factors promoting angiogenesis in A-07, D-12,R-18, and U-25 tumors, identified by measuring tumor-induced an-giogenic response in mice treated with neutralizing antibody, differedamong the lines. The angiogenesis of A-07 tumors, which showed thehighest rate of angiogenesis, was promoted by VEGF, IL-8, andPD-ECGF, as well as bFGF. VEGF and IL-8, but probably notPD-ECGF or bFGF, promoted angiogenesis in D-12 tumors, whichshowed the second highest rate of angiogenesis. The angiogenesis ofR-18 and U-25 tumors was promoted by VEGF but probably not byIL-8, PD-ECGF, or bFGF. All lines expressed bFGF, whereas A-07was the only line secreting bFGF, suggesting that promotion of tumorangiogenesis by bFGF may require bFGF secretion. A-07 and D-12showed significantly higher secretion of IL-8 than did R-18 and U-25,indicating that promotion of tumor angiogenesis by IL-8 may require

Fig. 2. Immunohistochemical preparations of A-07 tumors stained with anti-VEGF, anti-IL-8, anti-PD-ECGF, and anti-bFGF by using the avidin-biotin immunoperoxidase method.The sections were counterstained with hematoxylin.Bars,50 mm.

4935




high IL-8 concentrations. Our observations suggest that multipleangiogenic factors are involved in the angiogenesis of highly angio-genic melanomas, whereas the angiogenesis in poorly angiogenicmelanomas may be promoted solely by VEGF. Angiogenic factorsother than those studied here may also be involved in the angiogenesisof melanoma, including angiogenin (54) and angiopoietin-1 (55).

The angiogenic factors that were found to promote tumor angio-genesis were also found to be essential for tumor metastasis. Lungcolonization and spontaneous metastasis in A-07 were inhibited bytreatment with neutralizing antibody against VEGF, IL-8, PD-ECGF,or bFGF, as was A-07-induced angiogenesis. Treatment with neutral-izing antibody against VEGF or IL-8 inhibited lung colonization andspontaneous metastasis as well as angiogenesis in D-12, whereas noneof these biological phenomena was inhibited by treatment with neu-tralizing antibody against PD-ECGF or bFGF. These observationssuggest that spontaneous lung metastasis in A-07 and D-12 tumors isgoverned by the angiogenic potential of the tumor cells and hencecontrolled by the angiogenic factors controlling the angiogenesis.

The metastatic process is composed of a cascade of linked, sequen-tial, and highly selective steps involving multiple host-tumor interac-tions. These steps include invasion of tumor cells into blood vessels,survival in the blood circulation, arrest in the capillary bed of asecondary organ, extravasation into the secondary organ interstitiumand parenchyma, and tumor cell proliferation and angiogenesis in thesecondary organ. The metastatic propensity of a tumor may be influ-enced by the angiogenic potential of the tumor cells by two indepen-dent mechanisms: high microvessel density in the primary tumor mayenhance the opportunity of tumor cells to gain access to the bloodcirculation; and elevated capacity to induce neovascularization mayincrease the probability of tumor cells trapped in secondary organcapillary beds to give rise to macroscopic tumor growth (56). Theprobability of spontaneous lung metastasis in A-07 and D-12 tumorswas probably limited by the capacity of the tumor cells to induce

Fig. 3. A-07, D-12, R-18, and U-25 cells were harvested fromexponentially growing monolayer cultures and inoculated intrader-mally in BALB/c nu/numice for evocation of angiogenesis and tumorformation. Angiogenesis was quantified by counting the capillaries inthe dermis oriented toward the tumors at day 7 after the inoculationof 1.0 3 106 cells. The plots show the number of tumor-orientedcapillaries/tumor in control mice and in mice treated with neutralizingantibody against VEGF, IL-8, PD-ECGF, or bFGF. Mean values(columns) and SEs (bars) of 20 tumors are shown.

Fig. 4. A-07 and D-12 cells were harvested from exponentially growing monolayercultures and inoculated i.v. in BALB/cnu/numice for formation of lung colonies. Lungcolonization was quantified by counting the surface colonies at day 35 after the inocula-tion of 3.03 106 A-07 cells or 3.03 105 D-12 cells. The plots show the number of lungcolonies in control mice and in mice treated with neutralizing antibody against VEGF,IL-8, PD-ECGF, or bFGF. Mean values (columns) and SEs (bars) of 20 mice are shown.

4936




angiogenesis in the lungs, because spontaneous lung metastasis andlung colonization were inhibited by treatment with neutralizing anti-body against the same angiogenic factors. However, it cannot beexcluded that the probability of spontaneous lung metastasis wasinfluenced by the angiogenic potential of the tumor cells also via themicrovessel density of the primary tumor. Interestingly, the microves-sel density is;2.5-fold higher in A-07 tumors than in D-12 tumors(57).

The antibody treatments that inhibited lung colonization and spon-taneous metastasis in A-07 and D-12 melanomas lasted 4 and 8 days,respectively. The reduced lung colonization and spontaneous metas-tasis in antibody-treated mice were most likely a consequence ofinhibited angiogenesis rather than of cytotoxic or antiproliferativeeffects of the antibodies, because the growth of melanoma cellsinvitro was not inhibited by antibody treatment. This interpretation isbased on the assumption that neovascularization and macroscopicgrowth of lung colonies can be inhibited by treatment with neutral-izing antibodies against angiogenic factors also when the antibodytreatment is given to lung colonies consisting of less than;50 cells,which is the maximum size of 4-day-old A-07 and D-12 lung colonies.This assumption is not consistent with the general current thinkingthat angiogenesis is initiated when the tumor mass reaches a diameterof ;1 mm. However, the validity of our assumption is stronglysupported by recent observations by Liet al. (58) studying the initialphases of tumor cell-induced angiogenesis in skin window chambers.They observed modifications of the host vasculature when the tumormass reached 60–80 cells and saw functional new blood vessels whenthe tumor mass reached 100–300 cells. Moreover, they found thatex-flk 1, a truncated soluble VEGF receptor protein known to be

antiangiogenic and nontoxic, prevented angiogenesis and tumorgrowth in four of six window chambers when administered as a singledose at the same time as 40–50 tumor cells were inoculated into thewindow chambers. This observation led the authors to conclude thatindividual tumor cells are dependent on early angiogenic activities forboth survival and proliferationin vivo (58).

Previous studies have shown that lung colonization of humanmelanoma cells in immune-deficient mice can be inhibited by treat-ment with neutralizing antibody against VEGF (52) or by treatmentwith agents blocking the receptor of IL-8 (59). The present studyconfirms these observations and shows further for the first time thattreatment with neutralizing antibody against VEGF or IL-8 can inhibitspontaneous lung metastasis of human melanoma xenografts. It hasbeen shown by others that treatment with neutralizing antibodyagainst VEGF can inhibit spontaneous metastasis of human epider-moid carcinoma xenografts (60), human prostate carcinoma xe-nografts (61), and human colon and gastric carcinoma xenografts (62),and treatment with neutralizing antibody against IL-8 can inhibitspontaneous metastasis of non-small cell lung carcinoma xenografts(63). Moreover, the study reported here also shows for the first timethat spontaneous lung metastasis of human melanoma xenografts canbe inhibited by treatment with neutralizing antibody against PD-ECGF or bFGF.

The most important finding reported here is that the angiogenesisand spontaneous lung metastasis of A-07 melanoma xenografts can beinhibited by treatment with neutralizing antibody against VEGF, IL-8,PD-ECGF, or bFGF. Each of these angiogenic factors was probablyessential for the angiogenesis and metastasis of A-07 tumors, becauseinhibition of one of them could not be compensated for by the others.This observation suggests that inhibition of a single angiogenic factorpathway may be beneficial in the treatment of malignant melanoma.However, it is unlikely that complete inhibition of tumor growth andmetastasis of highly angiogenic melanomas would be possible by thisstrategy, because multiple angiogenic factors are involved. Efficientantiangiogenic treatment of malignant melanoma may require identi-fication and blocking of common functional features of several an-giogenic factors. Treatment strategies that target the common signal-ing cascades in the end organ of angiogenesis,i.e., the endothelial cell,may be more universally effective than those targeting specific an-giogenic factors or their receptors.

ACKNOWLEDGMENTS

We thank Berit Mathiesen and Kanthi Galappathi for excellent technicalassistance.

REFERENCES

1. Koh, H. K. Cutaneous melanoma. N. Engl. J. Med.,325: 171–182, 1991.2. Newton-Bishop, J. A. Molecular pathology of melanoma. Cancer Metastasis Rev.,16:

141–154, 1997.3. Johnson, T. M., Yahanda, A. M., Chang, A. E., Fader, D. J., and Sondak, V. K.

Advances in melanoma therapy. J. Am. Acad. Dermatol.,38: 731–741, 1998.4. Herlyn, M., and Koprowski, H. Melanoma antigens: immunological and biological

characterization and clinical significance. Ann. Rev. Immunol.,6: 283–308, 1988.5. Herlyn, M. Human melanoma: development and progression. Cancer Metastasis Rev.,

9: 101–112, 1990.6. Srivastava, A., Laidler, P., Hughes, L. E., Woodcock, J., and Shedden, E. J. Neovas-

cularization in human cutaneous melanoma: a quantitative morphological and Dopp-ler ultrasound study. Eur. J. Cancer Clin. Oncol.,22: 1205–1209, 1986.

7. Marcoval, J., Moreno, A., Graells, J., Vidal, A., Escriba, J. M., Garcia-Ramirez, M.,and Fabra, A. Angiogenesis and malignant melanoma. Angiogenesis is related to thedevelopment of vertical (tumorigenic) growth phase. J. Cutan. Pathol.,24: 212–218,1997.

8. Straume, O., Salvesen, H. B., and Akslen, L. A. Angiogenesis is prognosticallyimportant in vertical growth phase melanomas. Int. J. Oncol.,15: 595–599, 1999.

9. Erhard, H., Rietveld, F. J., van Altena, M. C., Brocker, E. B., Ruiter, D. J., and deWaal, R. M. Transition of horizontal to vertical growth phase melanoma is accom-

Fig. 5. The percentage of BALB/cnu/nu mice that developed spontaneous lungmetastases after intradermal inoculation of 3.53 105 A-07 or D-12 cells. The metastaticefficiency was assessed in control mice and in mice treated with neutralizing antibodyagainst VEGF, IL-8, PD-ECGF, or bFGF.Points, single experiments involving 10 miceeach.

4937




panied by induction of vascular endothelial growth factor expression and angiogen-esis. Melanoma Res.,7 (Suppl. 2):s19–s26, 1997.

10. Srivastava, A., Laidler, P., Davies, R. P., Horgan, K., and Hughes, L. E. Theprognostic significance of tumor vascularity in intermediate-thickness (0.76–4.0 mmthick) skin melanoma: a quantitative histologic study. Am. J. Pathol.,133: 419–423,1988.

11. Denijn, M., and Ruiter, D. J. The possible role of angiogenesis in the metastaticpotential of human melanoma. Clinicopathological aspects. Melanoma Res.,3: 5–14,1993.

12. Graham, C. H., Rivers, J., Kerbel, R. S., Stankiewicz, K. S., and White, W. L. Extentof vascularization as a prognostic indicator in thin (,0.76 mm) malignant melano-mas. Am. J. Pathol.,145: 510–514, 1994.

13. Potgens, A. J. G., Lubsen, N. H., van Altena, M. C., Schoenmakers, J. G. G., Ruiter,D. J., and de Waal, R. M. W. Vascular permeability factor expression influencestumor angiogenesis in human melanoma lines xenografted to nude mice. Am.J. Pathol.,146: 197–209, 1995.

14. Oku, T., Tjuvajev, J. G., Miyagawa, T., Sasajima, T., Joshi, A., Joshi, R., Finn, R.,Claffey, K. P., and Blasberg, R. G. Tumor growth modulation by sense and antisensevascular endothelial growth factor gene expression: effects on angiogenesis, vascularpermeability, blood volume, blood flow, fluorodeoxyglucose uptake, and prolifera-tion of human melanoma intracerebral xenografts. Cancer Res.,58: 4185–4192,1998.

15. Claffey, K. P., Brown, L. F., del Aguila, L. F., Tognazzi, K., Yeo, K-T., Manseau,E. J., and Dvorak, H. F. Expression of vascular permeability factor/vascular endo-thelial growth factor by melanoma cells increases tumor growth, angiogenesis, andexperimental metastases. Cancer Res.,56: 172–181, 1996.

16. Rofstad, E. K., and Danielsen, T. Hypoxia-induced angiogenesis and vascular endo-thelial growth factor secretion in human melanoma. Br. J. Cancer,77: 897–902, 1998.

17. Singh, R. K., Gutman, M., Radinsky, R., Bucana, C. D., and Fidler, I. J. Expressionof interleukin 8 correlates with the metastatic potential of human melanoma cells innude mice. Cancer Res.,54: 3242–3247, 1994.

18. Luca, M., Huang, S., Gershenwald, J. E., Sing, R. K., Reich, R., and Bar-Eli, M.Expression of interleukin-8 by human melanoma cells up-regulates MMP-2 activityand increases tumor growth and metastases. Am. J. Pathol.,151: 1105–1113, 1997.

19. Kunz, M., Hartmann, A., Flory, E., Toksoy, A., Koczan, D., Thiesen, H-J., Mukaida,N., Neumann, M., Rapp, U. R., Brocker, E-B., and Gillitzer, R. Anoxia-inducedup-regulation of interleukin-8 in human malignant melanoma. Am. J. Pathol.,155:753–763, 1999.

20. Leyva, A., Appel, H., Kraal, I., and Pinedo, H. M. Differential metabolism ofthymidine in human lymphoid and melanoma cellsin vitro. Anticancer Res.,4:173–178, 1984.

21. Asgari, M. M., Haggerty, J. G., McNiff, J. M., Milstone, L. M., and Schwartz, P. M.Expression and localization of thymidine phosphorylase/platelet-derived endothelialcell growth factor in skin and cutaneous tumors. J. Cutan. Pathol.,26: 287–294, 1999.

22. Halaban, R., Kwon, B. S., Ghosh, S., Delli Bovi, P., and Baird, A. bFGF as anautocrine growth factor for human melanoma. Oncogene Res.,3: 177–186, 1988.

23. Becker, D., Meier, C. B., and Herlyn, M. Proliferation of human malignant melano-mas is inhibited by antisense oligodeoxynucleotides targeted against basic fibroblastgrowth factor. EMBO J.,8: 3685–3691, 1989.

24. Wang, Y., and Becker, D. Antisense targeting of basic fibroblast growth factor andfibroblast growth factor receptor-1 in human melanomas blocks intratumoral angio-genesis and tumor growth. Nat. Med.,3: 887–893, 1997.

25. Dvorak, H. F., Brown, L. F., Detmar, M., and Dvorak, A. M. Vascular permeabilityfactor/vascular endothelial growth factor, microvascular hyperpermeability, and an-giogenesis. Am. J. Pathol.,146: 1029–1039, 1995.

26. Ferrara, N. Vascular endothelial growth factor. Eur. J. Cancer,32A: 2413–2422,1996.

27. Koch, A. E., Polverini, P. J., Kunkel, S. L., Harlow, L. A., DiPietro, L. A., Elner,V. M., Elner, S. G., and Strieter, R. M. Interleukin-8 as a macrophage derivedmediator of angiogenesis. Science (Washington DC),258: 1798–1801, 1992.

28. Bar-Eli, M. Role of interleukin-8 in tumor growth and metastasis of human mela-noma. Pathobiology,67: 12–18, 1999.

29. Griffiths, L., and Stratford, I. J. Platelet-derived endothelial cell growth factorthymidine phosphorylase in tumour growth and response to therapy. Br. J. Cancer,76:689–693, 1997.

30. Slavin, J. Fibroblast growth factors: at the heart of angiogenesis. Cell Biol. Int.,19:431–444, 1995.

31. Ellis, L. M., and Fidler, I. J. Angiogenesis and metastasis. Eur J. Cancer,32A:2451–2460, 1996.

32. Bikfalvi, A., Klein, S., Pintucci, G., and Rifkin, D. B. Biological roles of fibroblastgrowth factor-2. Endocr. Rev.,18: 26–45, 1997.

33. Potgens, A. J. G., van Altena, M. C., Lubsen, N. H., Ruiter, D. J., and de Waal,R. M. W. Analysis of the tumor vasculature and metastatic behavior of xenografts ofhuman melanoma cell lines transfected with vascular permeability factor. Am.J. Pathol.,148: 1203–1217, 1996.

34. Shih, I-M., and Herlyn, M. Autocrine and paracrine roles for growth factors inmelanoma. In Vivo,8: 113–123, 1994.

35. Halaban, R. Growth factors and melanomas. Semin. Oncol.,23: 673–681, 1996.36. Ijland, S. A. J., Jager, M. J., Heijdra, B. M., Westphal, J. R., and Peek, R. Expression

of angiogenic and immunosuppressive factors by uveal melanoma cell lines. Mela-noma Res.,9: 445–450, 1999.

37. Rofstad, E. K. Orthotopic human melanoma xenograft model systems for studies oftumour angiogenesis, pathophysiology, treatment sensitivity and metastatic pattern.Br. J. Cancer,70: 804–812, 1994.

38. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head ofbacteriophage T4. Nature (Lond.),227: 680–685, 1970.

39. Danielsen, T., and Rofstad, E. K. VEGF, bFGF and EGF in the angiogenesis ofhuman melanoma xenografts. Int. J. Cancer,76: 836–841, 1998.

40. Rofstad, E. K. Local tumor control following single dose irradiation of humanmelanoma xenografts: relationship to cellular radiosensitivity and influence of animmune response by the athymic mouse. Cancer Res.,49: 3163–3167, 1989.

41. Rofstad, E. K. Metastatic behavior of human tumors in congenitally athymic nudemice: intrinsic properties of the tumor cells and host immune reactivity. Int. J. Cancer,63: 744–749, 1995.

42. Pepper, M. S., Ferrara, N., Orci, L., and Montesano, R. Potent synergism betweenvascular endothelial growth factor and basic fibroblast growth factor in the inductionof angiogenesisin vitro. Biochem. Biophys. Res. Commun.,189: 824–831, 1992.

43. Goto, F., Goto, K., Weindel, K., and Folkman, J. Synergistic effects of vascularendothelial growth factor and basic fibroblast growth factor on the proliferation andcord formation of bovine capillary endothelial cells within collagen gels. Lab.Investig.,69: 508–517, 1993.

44. Asahara, T., Bauters, C., Zheng, L. P., Takeshita, S., Bunting, S., Ferrara, N., Symes,J. S., and Isner, J. M. Synergistic effect of vascular endothelial growth factor andbasic fibroblast growth factor on angiogenesisin vivo. Circulation,92: 365–371,1995.

45. Tsai, J. C., Goldman, C. K., and Gillespie, G. Y. Vascular endothelial growth factorin human glioma cell lines: induced secretion by EGF, PDGF-BB, and bFGF.J. Neurosurg.,82: 864–873, 1995.

46. Stavri, G. T., Zachary, I. C., Baskerville, P. A., Martin, J. F., and Erusalimsky, J. D.Basic fibroblast growth factor upregulates the expression of vascular endothelialgrowth factor in vascular smooth muscle cells. Synergistic interaction with hypoxia.Circulation,92: 11–14, 1995.

47. Rofstad, E. K., and Måseide, K. Fraction of radiobiologically hypoxic cells in humanmelanoma xenografts measured by using single-cell survival, tumour growth delayand local tumour control as end points. Br. J. Cancer,78: 893–898, 1998.

48. Shweiki, D., Neeman, M., Itin, A., and Keshet, E. Induction of vascular endothelialgrowth factor expression by hypoxia and by glucose deficiency in multicell spheroids:implications for tumor angiogenesis. Proc. Natl. Acad. Sci. USA,92: 768–772, 1995.

49. Waleh, N. S., Brody, M. D., Kapp, M. A., Mendonca, H. L., Lord, E. M., Koch, C. J.,Laderoute, K. R., and Sutherland, R. M. Mapping of the vascular endothelial growthfactor-producing hypoxic cells in multicellular tumor spheroids using a hypoxia-specific marker. Cancer Res.,55: 6222–6226, 1995.

50. Desbaillets, I., Diserens, A-C., de Tribolet, N., Hamou, M-F., and van Meir, E. G.Regulation of interleukin-8 expression by reduced oxygen pressure in human glio-blastoma. Oncogene,18: 1447–1456, 1999.

51. Griffiths, L., Dachs, G. U., Bicknell, R., Harris, A. L., and Stratford, I. J. Theinfluence of oxygen tension and pH on the expression of platelet-derived endothelialcell growth factor/thymidine phosphorylase in human breast tumor cells growninvitro and in vivo. Cancer Res.,57: 570–572, 1997.

52. Rofstad, E. K., and Danielsen, T. Hypoxia-induced metastasis of human melanomacells: involvement of vascular endothelial growth factor-mediated angiogenesis. Br.J. Cancer,80: 1697–1707, 1999.

53. Rofstad, E. K., and Måseide, K. Radiobiological and immunohistochemical assess-ment of hypoxia in human melanoma xenografts: acute and chronic hypoxia inindividual tumours. Int. J. Radiat. Biol.,75: 1377–1393, 1999.

54. Hartmann, A., Kunz, M., Kostlin, S., Gillitzer, R., Toksoy, A., Brocker, E-B., andKlein, C. E. Hypoxia-induced up-regulation of angiogenin in human malignantmelanoma. Cancer Res.,59: 1578–1583, 1999.

55. Siemeister, G., Schirner, M., Weindel, K., Reusch, P., Menrad, A., Marme, D., andMartiny-Baron, G. Two independent mechanisms essential for tumor angiogenesis:inhibition of human melanoma xenograft growth by interfering with either thevascular endothelial growth factor receptor pathway or the Tie-2 pathway. CancerRes.,59: 3185–3191, 1999.

56. Fidler, I. J., and Ellis, L. M. The implications of angiogenesis for the biology andtherapy of cancer metastasis. Cell,79: 185–188, 1994.

57. Tufto, I., Lyng, H., and Rofstad, E. K. Vascular density in human melanomaxenografts: relationship to angiogenesis, perfusion and necrosis. Cancer Lett.,123:159–165, 1998.

58. Li, C-J., Shan, S., Huang, Q., Braun, R. D., Lanzen, J., Hu, K., Lin, P., and Dewhirst,M. W. Initial stages of tumor cell-induced angiogenesis: evaluation via skin windowchambers in rodent models. J. Natl. Cancer Inst.,92: 143–147, 2000.

59. Fujisawa, N., Hayashi, S., and Miller, E. J. A synthetic peptide inhibitor fora-chemokines inhibits the tumour growth and pulmonary metastasis of human mela-noma cells in nude mice. Melanoma Res.,9: 105–114, 1999.

60. Melnyk, O., Shuman, M. A., and Kim, K. J. Vascular endothelial growth factorpromotes tumor dissemination by a mechanism distinct from its effect on primarytumor growth. Cancer Res.,56: 921–924, 1996.

61. Melnyk, O., Zimmerman, M., Kim, K. J., and Shuman, M. Neutralizing antivascularendothelial growth factor antibody inhibits further growth of established prostatecancer and metastases in a preclinical model. J. Urol.,161: 960–963, 1999.

62. Kanai, T., Konno, H., Tanaka, T., Baba, M., Matsumoto, K., Nakamura, S., Yukita,A., Asano, M., Suzuki, H., and Baba, S. Antitumor and antimetastatic effects ofhuman-vascular-endothelial-growth-factor-neutralizing antibody on human colon andgastric carcinoma xenotransplanted orthotopically into nude mice. Int. J. Cancer,77:933–936, 1998.

63. Arenberg, D. A., Kunkel, S. L., Polverini, P. J., Glass, M., Burdick, M. D., andStrieter, R. M. Inhibition of interleukin-8 reduces tumorigenesis of human non-smallcell lung cancer in SCID mice. J. Clin. Investig.,97: 2792–2802, 1996.

4938




2000;60:4932-4938. Cancer Res Einar K. Rofstad and Ellen F. Halsør Metastasis in Human Melanoma XenograftsFibroblast Growth Factor Promote Angiogenesis andPlatelet-derived Endothelial Cell Growth Factor, and Basic Vascular Endothelial Growth Factor, Interleukin 8,

Updated version

http://cancerres.aacrjournals.org/content/60/17/4932

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/60/17/4932.full#ref-list-1

This article cites 60 articles, 13 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/60/17/4932.full#related-urls

This article has been cited by 31 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] 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/60/17/4932To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/60/17/4932.full#ref-list-1

http://cancerres.aacrjournals.org/content/60/17/4932.full#related-urls

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

mailto:[email protected]



vascular endothelial growth factor, interleukin 8, platelet-derived ... · vascular endothelial...

Documents