Blockade of Epha Receptor Tyrosine Kinase Activation Inhibits Vascular Endothelial Cell Growth Factor-Induced Angiogenesis

2 Vol. 1, 2–11, November 2002 Molecular Cancer Research

Blockade of EphA Receptor Tyrosine Kinase Activation Inhibits Vascular Endothelial Cell Growth Factor-Induced Angiogenesis

Nikki Cheng,1 Dana M. Brantley,2 Hua Liu,3,4 Qin Lin,2 Miriam Enriquez,4 Nick Gale,5 George Yancopoulos,5 Douglas Pat Cerretti,3 Thomas O. Daniel,3,4 and Jin Chen1,2,6

Departments of 1Cancer Biology, 2Medicine, Division of Rheumatology, 4Division of Nephrology, and 6Cell Biology, Vanderbilt University School of Medicine, Nashville, TN; 3Immunex Corporation, Seattle, WA; and 5Regeneron Inc., Tarrytown, NY

Abstract endothelial cell proliferation, migration, and assembly, as well Angiogenesis is a multistep process involving a diverse as recruitment of perivascular cells and extracellular matrix array of molecular signals. Ligands for receptor tyrosine remodeling. Three families of receptor tyrosine kinases (RTKs) kinases (RTKs) have emerged as critical mediators of have emerged as critical mediators of angiogenesis; these are the angiogenesis. Three families of ligands, vascular endo- vascular endothelial growth factor (VEGF), Tie, and Eph RTK thelial cell growth factors (VEGFs), angiopoietins, and families (1, 2). VEGF (VEGF-A/VEGF165) is a potent angio- ephrins, act via RTKs expressed in endothelial cells. genic factor in both embryonic development and in adult disease Recent evidence indicates that VEGF cooperates with states, such as cancer. VEGF and its RTKs, Flt-1/VEGFR1 and angiopoietins to regulate vascular remodeling and Flk-1/KDR/VEGFR2, are required for the development and angiogenesis in both embryogenesis and tumor remodeling of blood vessels during embryogenesis (3–8). neovascularization. However, the relationship between Moreover, VEGF signaling plays a crucial role in pathogenic VEGF and ephrins remains unclear. Here we show that angiogenesis, including the recruitment and maintenance of interaction between EphA RTKs and ephrinA ligands is tumor vasculature, and blocking VEGF function using soluble necessary for induction of maximal neovascularization receptors, neutralizing antibodies, or pharmacologic inhib- by VEGF. EphA2 RTK is activated by VEGF through itors significantly abrogates tumor angiogenesis and progres- induction of ephrinA1 ligand. A soluble EphA2-Fc sion (9–14). Recent evidence indicates that VEGF cooperates receptor inhibits VEGF-, but not basic fibroblast growth with other angiogenic factors, such as angiopoietins and their factor-induced endothelial cell survival, migration, RTK, Tie2, in regulating vascular remodeling and growth in sprouting, and corneal angiogenesis. As an indepen- tumors (15–17). However, the relationship between VEGF dent, but complementary approach, EphA2 antisense and ephrins remains unclear. oligonucleotides inhibited endothelial expression of The Eph family of RTKs and their ligands, originally EphA2 receptor and suppressed ephrinA1- and VEGF- identified as critical determinants of embryonic patterning and induced cell migration. Taken together, these data neuronal targeting (18), also regulates vascular development (1, indicate an essential role for EphA receptor activation in 2, 19). Targeted disruption of ephrinB2, EphB4,orEphB2/ VEGF-dependent angiogenesis and suggest a potential EphB3 results in embryonic lethality due to defects in primary new target for therapeutic intervention in pathogenic capillary network remodeling and subsequent patterning defects angiogenesis. in the embryonic vasculature (20–22), suggesting that Eph RTKs and their ligands are critical for vascular development during embryogenesis. The A class ligand, ephrinA1, has also Introduction been implicated in angiogenesis. EphrinA1 was originally Angiogenesis, the formation of new blood vessels from pre- identified as a tumor necrosis factor a (TNF-a)-inducible gene existing vasculature, is a multistep process involving a diverse in human umbilical vein endothelial cells (HUVECs) (23), and array of molecular signals. These include factors that stimulate is expressed in the developing vasculature during embryogenesis (24). Moreover, ephrinA1 induces endothelial cell migration and capillary assembly in vitro, and angiogenesis in a Received 3/6/02; revised 6/17/02; accepted 6/24/02. The costs of publication of this article were defrayed in part by the payment of corneal pocket assay in vivo (25, 26), suggesting a role in page charges. This article must therefore be hereby marked advertisement in neovascularization of adult tissues. Indeed, expression of accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ephrinA1 and its receptor EphA2 was observed in breast Grant support: NIH Grants HD36400 and DK47078; JDF grant I-2001-519; DOD grant BC010265; American Heart Association Grant 97300889N; ACS tumors and associated vasculature (27), and blocking EphA Institutional Research Grant IN-25-38 (to J. Chen); Vascular Biology Training receptor activation impaired tumor angiogenesis (28). These Grant T32-HL-07751-06 and American Heart Association Fellowship 0120147B studies indicate that Eph signaling is critical for normal blood (to D. Brantley); Cancer training Grant T-32 CA09592 (to N. Cheng); and a core facilities Grant 2P30CA68485 to the Vanderbilt-Ingram Cancer Center. vessel development as well as pathogenic angiogenesis. Requests for reprints: Jin Chen, Vanderbilt University School of Medicine, In this study, we provide evidence that Eph RTKs and their A-4323 MCN, 1161 21st Avenue South, Nashville, TN 37232. Phone: (615) 343-3819; Fax: (615) 343-7392. E-mail: [email protected] ligands are necessary for induction of maximal angiogenesis by Copyright D 2002 American Association for Cancer Research. VEGF. We show that ephrinA1 is a downstream target gene

product induced by VEGF. Blocking EphA receptor activation inhibits VEGF-dependent endothelial cell migration, sprouting, and survival in vitro and angiogenesis in vivo. These effects are specific to VEGF, as blocking EphA class signaling has no impact of fibroblast growth factor (FGF)-mediated angiogenesis. As VEGF is a critical mediator of angiogenesis in pathogenic events such as cancer, our data suggest a potential new target for anti-angiogenic therapy.

Results A Soluble EphA2-Fc Chimeric Receptor Inhibits VEGF-, but not FGF-induced Angiogenesis Given that both VEGF and ephrins regulate angiogenesis, and that VEGF is known to cooperate with other angiogenic factors to modulate blood vessel formation, we investigated whether EphA receptor activation is required for VEGF- mediated angiogenesis. To achieve this goal, we have utilized a soluble chimeric protein, EphA2-Fc, as a blocking reagent. This approach was taken because mutant mice deficient for individual Eph family members are either embryonic lethal or do not display overt phenotypes (20–22, 29). The EphA2-Fc soluble receptor fusion proteins consist of extracellular domain of the EphA2 receptor and the Fc portion of the human immunoglobulin IgG1. Because there is promiscuous binding between Eph receptors and ligands within the same subclass, this soluble receptor variant prevents multiple ephrinA class ligand interactions with endogenous receptors, effectively blocking signaling through class A Eph receptors. As shown in Fig. 1A, endothelial cells treated with ephrinA1 in the presence of excess EphA2-Fc showed a dramatic reduction in endogenous EphA2 receptor phosphorylation as compared to cells treated with ephrinA1 alone. However, EphA2-Fc does not affect the phosphorylation of VEGFR2 (Flk-1) (Fig. 1B), demonstrating the specificity of EphA2-Fc to class A ephrins. In the mouse cornea, soluble EphA2-Fc also markedly inhibited the angiogenic response induced by ephrinA1 (Fig. 1C), showing the efficacy of this reagent in blocking A class Eph receptor activation and neovascularization in vivo. We next investigated whether ephrin/Eph A class signals are required for VEGF-dependent angiogenesis. To do this, we used a well-established assay in which hydron pellets impregnated with either test or control proteins are implanted into mouse corneal pockets (30). Angiogenesis induced by exogenous factors in the normally avascular cornea was then documented and quantified. Consistent with previous observations, FIGURE 1. A soluble EphA2 receptor blocks endogenous EphA2 ephrinA1, VEGF, and basic fibroblast growth factor (bFGF) receptor signaling in vitro and ephrinA1-induced angiogenesis in vivo. A. EA926 endothelial cells were incubated with ephrinA1 in the presence or each induced corneal neovascularization (Figs. 1C and 2, A and absence of EphA2-Fc, and endogenous EphA2 was immunoprecipitated B) (16, 25, 31). To determine if VEGF function is dependent from cell lysates. EphrinA1-induced tyrosine phosphorylation of endogenous EphA2 receptor was inhibited by excess soluble EphA2-Fc. Uniform upon signaling by ephrinA class ligands, we implanted hydron loading was confirmed by immunoblotting with an anti-EphA2 antibody. B. pellets impregnated with VEGF or bFGF in the presence or HUVECs were stimulated with VEGF in the presence or absence of absence of EphA2-Fc. Soluble EphA2-Fc receptor itself does EphA2-Fc. Endogenous VEGF receptor was immunoprecipitated from cell lysates. VEGF-induced tyrosine phosphorylation of VEGR2 is not affected not induce angiogenesis in the cornea. Rather, the addition of by excess soluble EphA2-Fc. C. EphA2-Fc inhibits ephrinA1-induced EphA2-Fc in the hydron pellet markedly inhibited the corneal angiogenesis. Hydron pellets impregnated with vehicle, bFGF, angiogenic response induced by VEGF (Fig. 2A). This effect ephrinA1, or ephrinA1 plus EphA2-Fc were implanted into mouse corneas and photographed at 5 days postimplantation. Images were digitized and is specific, as EphA2-Fc did not affect angiogenesis induced analyzed using Bioquant software. Columns, means of values for the by bFGF, another potent inducer of angiogenesis (Fig. 2B) fractional areas vascularized (VA), the vascular density within that region (RVD), and the vessel density as a fraction of the total corneal image area (32). These data suggest that EphA receptor activation is (TVD); bars, SE. EphA2-Fc inhibited ephrinA1-induced corneal neovas- specifically required for VEGF-induced angiogenesis in vivo. cularization (two-tailed Student’s t test, P < 0.01).

Soluble EphA2-Fc Receptors Inhibit VEGF-induced Endothelial Sprouting Sprouting of new capillaries from pre-existing blood vessels is a hallmark of angiogenesis, and VEGF has been shown to induce sprouting activity in an in vitro capillary sprouting assay (33). To determine whether VEGF-induced sprouting activity is affected by blocking EphA receptor activation, we cultured endothelial cells on collagen-coated beads, and assessed sprouting in fibrin gels in response to ephrinA1-Fc, VEGF, bFGF, or VEGF or bFGF in the presence of EphA2-Fc or control Fc proteins, or medium alone. As shown in Fig. 3, endothelial cells extended from collagen-coated beads in response to ephrinA1, VEGF, bFGF, or VEGF or bFGF plus control Fc proteins. In contrast, EphA2-Fc inhibited sprouting activity induced by VEGF, but not FGF, suggesting that VEGF- induced endothelial cell sprouting requires EphA class receptor function. Taken together, our findings revealed that activation of EphA receptor is required for VEGF-mediated endothelial cell sprouting, an essential step of angiogenic process.

Soluble EphA2-Fc Receptors Inhibit VEGF-Induced Cell Survival but Do Not Affect Cell Proliferation To determine which steps of VEGF-induced angiogenesis require EphA receptor activation, we performed a series of in vitro angiogenesis assays. First, we tested whether blocking EphA receptor function affects VEGF-induced cell proliferation. Endothelial cells were treated with VEGF or bFGF in the presence or absence of EphA2-Fc, and cell growth was measured by [3H]thymidine incorporation. Consistent with the observation that ephrin signaling does not affect endothelial cell growth, ephrinA1-Fc did not affect cell proliferation, nor did EphA2-Fc impair VEGF- or bFGF-induced cell proliferation significantly (Fig. 4A) (1, 18). To determine the effect of EphA signaling in VEGF-induced cell survival, endothelial cells were growth factor/serum starved to induce apoptosis. At the time of growth factor and serum withdrawal, endothelial cells were treated with VEGF, bFGF, or VEGF or FGF in the presence of EphA2-Fc or control Fc proteins. Compared to cells treated with VEGF and VEGF plus Fc control, cells treated with VEGF plus EphA2-Fc showed a significant reduction in cell survival, as determined by TUNEL assay (Fig. 4B) and trypan blue exclusion (data not shown). However, bFGF-induced cell survival was not affected by EphA2-Fc soluble receptor treatment. Taken together, these data suggest that EphA class signaling is not required for the mitogenic activity of VEGF, but is required for VEGF-mediated survival of endothelial cells.

EphA2 Antisense Oligonucleotides or Soluble EphA2- Fc Receptors Inhibit VEGF-Induced Endothelial Cell Migration To test the effect of EphA2-Fc on VEGF-mediated FIGURE 2. EphA2-Fc inhibits VEGF-, but not FGF-induced corneal endothelial cell migration, we used both ‘‘wound closure’’ angiogenesis. A. Hydron pellets harboring VEGF, EphA2-Fc, or VEGF plus EphA2-Fc were implanted into mouse corneas and analyzed as and modified Boyden chamber assays. As shown in Fig. 5A, described in Fig. 1. EphA2-Fc inhibited VEGF-induced angiogenesis in the the rate of endothelial migration to close a circular wound in mouse cornea (Student’s t test, P < 0.01). B. EphA2-Fc did not inhibit a confluent endothelial cell monolayer was increased in bFGF-induced angiogenesis in the mouse cornea. Hydron pellets containing bFGF, EphA2-Fc, or bFGF plus EphA2-Fc were implanted into serum-free medium supplemented with ephrinA1-Fc, VEGF, mouse corneas and analyzed as described. bFGF, or VEGF or bFGF plus Fc control protein. In contrast,

induced angiogenesis, we utilized a published antisense oligonucleotide to specifically inhibit EphA2 receptor expression (34). Endothelial cells transfected with antisense oligonucleotides show approximately a 3-fold reduction in endothelial endogenous EphA2 receptor expression, whereas EphA2 levels in cells transfected with control inverted antisense oligonucleotides are not significantly affected (Fig. 5C). Inhibition of EphA2 receptor expression results in a significant reduction in ephrinA1- and VEGF-induced migration in the wound closure assay (Fig. 5D) and in the modified Boyden chamber assay (Fig. 5E), indicating that EphA2 receptor is required for VEGF- induced endothelial cell migration.

VEGF Induces Endothelial Expression of EphrinA1 Ligand and Phosphorylation of EphA2 Receptor Given that EphA receptor activation is required for several steps in VEGF-dependent angiogenesis, we investigated whether ephrinA1, the ligand for EphA2 receptor, is a downstream target gene induced by VEGF signaling. EphrinA1 was previously shown to be induced by TNF-a, and to mediate TNF-a-induced corneal angiogenesis (25). As shown in Fig. 6, like TNF-a, VEGF induced ephrinA1 expression in both HMEC-1 and HUVEC endothelial cells, as judged by Northern blot (Fig. 6A) and Western blot analyses (Fig. 6B). Because ephrinA1 is angiogenic, this result suggests that the induction of ephrinA1 and subsequent activation of the EphA receptors could be partially responsible for the angiogenic effects of VEGF. To address this possibility directly, we asked whether VEGF could induce EphA2 phosphorylation. As shown in Fig. 6C, exposure of endothelial cells to VEGF resulted in phosphorylation of EphA2 receptor. This activation of receptor was apparently due to the induction of ephrinA1 because treatment of cells with VEGF in the presence of soluble EphA2-Fc receptor resulted in inhibition of endogenous EphA2 receptor phosphorylation (Fig. 6C). Addition of a control Fc protein had no effect on EphA2 phosphorylation or ephrinA1 expression (data not shown). Thus, the level of expression of ephrinA1 correlated directly with the extent of phosphorylation of endogenous EphA2 receptor, whereas the absolute amount of EphA2 remained unchanged. Taken together, these data demonstrate that VEGF induces ephrinA1 expression in endothelial cells, suggesting a juxtacrine mechanism for activation of EphA receptor signaling. FIGURE 3. Eph-A2 inhibits VEGF-induced endothelial cell sprouting. HMEC-1 cells on collagen-coated beads in 3D fibrin gels were incubated with test or control proteins as indicated. Addition of EphA2-Fc to VEGF- Discussion treated cells caused an approximate 2-fold reduction in cell sprouting as The Eph family of RTKs is one of three major families of quantified by the number of resulting endothelial cell sprouts exceeding RTKs, which also include the Tie2 and VEGF receptor families, bead length. *, P < 0.05 compared with VEGF or VEGF/Fc treatments (two-tailed Student’s t test). that regulate blood vessel formation. Recent evidence indicates that VEGF cooperates with angiopoietins, the ligands for Tie2 receptors, in regulating angiogenesis, vascular remodeling, and EphA2-Fc significantly reduced the migration of endothelial growth in tumors (15, 17). Our data demonstrate that ephrins cells into the ‘‘wounded’’ area in response to VEGF, but not also act in concert with VEGF in promoting angiogenesis. bFGF. Consistent with ‘‘wound closure’’ data, excess EphA2- Several lines of evidence support this conclusion. First, VEGF Fc also significantly inhibited migration of endothelial cells induces endothelial expression of ephrinA1 and phosphoryla- in response to VEGF in a modified Boyden chamber assay tion of endogenous EphA2 receptor. Second, VEGF-induced (Fig. 5B). endothelial cell migration, survival, and sprouting are inhibited As an independent, but complementary, approach to by a soluble EphA2-Fc receptor in vitro. Third, blocking determine the role of EphA receptor activation in VEGF- EphA2 receptor expression by antisense oligonucleotides also

inhibits VEGF-induced endothelial cell migration, implying the because treatment cells with VEGF in the presence of soluble role of EphA2 receptor in VEGF-induced angiogenesis. Fourth, EphA2-Fc receptor resulted in inhibition of endogenous EphA2 blocking EphA receptor activation inhibits VEGF-, but not phosphorylation. Thus, these data support a hypothesis that FGF-induced angiogenesis in corneal assays. Taken together, VEGF induces ephrinA1 expression in endothelial cells and these results provide evidence for a specific link between the subsequent activation of EphA receptor signaling to promote ephrinA1/EphA2 and the VEGF pathways in formation of new angiogenesis through a juxtacrine mechanism. blood vessels. A second hypothesis is that VEGF-induced signaling and How does inhibition of EphA RTK activation lead to ephrinA1-induced signaling are two separate pathways, and suppression of VEGF-induced angiogenesis? One hypothesis is activation of EphA RTK signaling positively regulates VEGF that EphA signaling may be a part of VEGF signaling cascade signaling pathways to promote angiogenesis. Thus, soluble to regulate angiogenesis. In support of this hypothesis, we EphA2-Fc receptor could affect VEGF signaling at several showed that VEGF induced endothelial ephrinA1 expression steps. First, EphA2-Fc could affect VEGF receptor phospho- and phosphorylation of EphA2 receptor (Fig. 6). This activation rylation, which subsequently affects several VEGF-induced of receptor was apparently due to the induction of ephrinA1 cellular responses. However, as shown in Fig. 1B, we found no

FIGURE 4. EphA2-Fc inhibits VEGF-dependent cell survival. A. EphA2-Fc does not affect VEGF-induced cell proliferation. HUVECs were serum starved in medium alone, ephrinA1-Fc, VEGF, bFGF, and VEGF or bFGF plus Fc or EphA2-Fc, and thymidine incorporation was measured using a beta scintillation counter. B. HUVECs were serum starved as described above, treated with control or test proteins, and subjected to terminal deoxynucleotidyl transferase-mediated nick end labeling (TUNEL) assay. Arrows indicate apoptotic cells (stained red). The addition of EphA2-Fc leads to a reduction in VEGF-induced cell survival as seen by an increase of rhodamine- stained nuclei. Percentage of cell apoptosis was quantified as the number of rhodamine-stained nuclei over total number of DAPI-stained nuclei (stained blue). *, P < 0.01 compared with VEGF or VEGF plus Fc treatment (two-tailed Student’s t test).

FIGURE 5. EphA2-Fc and antisense oligonucleotides inhibit VEGF-induced cell migration. A. Confluent HMEC-1 cells were serum starved before mechanical ‘‘wounding.’’ Medium was supplemented with test and control proteins as indicated. Fractions of the areas remaining in triplicate wounds were determined by analysis of serial digital images obtained at 0, 2, 4, 8, and 10 h. B. Migration response of HUVECs to test and control proteins (as indicated) was assayed using modified Boyden chambers. Data are presented as the number of cells that migrated through the filter in 10 high-power fields after 5 h of incubation. *, P < 0.05 or **, P < 0.01, compared to VEGF/Fc treatment (two-tailed Student’s t test). C. Western blot analysis of HUVECs transfected with either antisense-EphA2 (AS) or control inverted antisense oligonucleotides (IAS). D and E. Migration response of HUVECs transfected with EphA2-AS or IAS, in the presence of VEGF or ephrinA1-Fc, was measured by wound closure (D) and Boyden chamber assays (E). *, P < 0.05 or **, P < 0.01, compared to either ephrinA1- or VEGF-treated HUVECs transfected with control IAS oligonucleotides (two-tailed Student’s t test).

effects of EphA2-Fc on VEGFR-2 phosphorylation in response receptor expression is not expected to affect VEGF-induced to VEGF stimulation. Second, VEGF is known to inhibit responses greatly. Because the EphA2 antisense oligonucleo- endothelial cell apoptosis by activation of Akt via a PI3 kinase- tides and soluble EphA2-Fc receptor have similar effects on dependent pathway (35). Blockade of EphA receptor binding to VEGF-induced cell migration (Fig. 5), the effect of EphA2-Fc ephrinA ligand could thus affect VEGF-induced PI3 kinase/Akt is probably due to the inhibition of ephrinA-EphA ligand activation, leading to increased apoptosis. Third, activation of receptor interaction. FAK or p38MAP kinase is shown to be required for VEGF- Recent studies from our laboratory and others suggest that induced cell migration (36). Inhibition of EphA receptor EphA receptor activation plays a critical role in tumor activation either by soluble EphA2-Fc or antisense oligonu- angiogenesis. Elevated expressions of ephrinA1 and EphA2 cleotide treatment could affect FAK/p38 MAP kinase activity, were observed in various tumors and associated vasculature resulting in inhibition of cell migration. Further experiments are including colon, breast, and pancreatic carcinoma (27). required to determine whether blocking EphA receptor Furthermore, in a separate manuscript, we reported that EphA activation affects VEGF-induced activation of PI3 kinase/Akt receptor activation is required for tumor neovascularization in and/or FAK/p38 MAP kinase pathways. two tumor models (28). These data suggest that ephrinA1 could Although there is a bidirectional signaling between both function to induce tumor angiogenesis through both juxtacrine class A and class B Eph receptors and ligands (37–40), it and paracrine mechanisms. As VEGF has been shown to be remains unclear whether soluble EphA2-Fc could signal to a key regulator in tumor angiogenesis, EphA signaling may ephrinA1 ligand on the surface of endothelial cells. Based on be an important mediator of VEGF-dependent tumor an- our results, EphA2-Fc neither elicit endothelial cell responses giogenesis. Thus, it is possible that high levels of VEGF above the controls in vitro, nor induce corneal angiogenesis in expression in tumor cells could induce ephrinA1 expression in vivo. However, in principle, soluble EphA2-Fc could inhibit adjacent endothelial cells. Engagement of ephrinA1 to EphA2 VEGF-induced angiogenesis through ephrinA1-mediated receptor may then activate juxtacrine signaling to regulate reverse signaling. If this is the case, then reduction of EphA changes in endothelial cell survival and migration to promote

angiogenesis. Alternatively, tumor cells may also recruit new Immunoprecipitation and Western Blot Analysis blood vessels directly through paracrine signaling between EA926 endothelial cells were serum starved overnight and ephrinA1 expressed on tumor cells and EphA2 expressed on stimulated for 20 min with 0.5 Ag/ml ephrinA1-Fc in the surrounding endothelial cells. presence or absence of 0.5–5 Ag/ml EphA2-Fc in serum-free In summary, we provide the first evidence that class A Eph media. Serum-starved HUVECs were stimulated for 15 min RTKs play a critical role in VEGF-dependent angiogenesis. with 20 ng/ml VEGF in the presence or absence of 5 Ag/ml Thus, EphA class receptors and ephrinA ligands may provide EphA2-Fc. Cells were lysed in a buffer containing: 1Â PBS, potential novel targets for therapeutic intervention in diseases 1% Ipegal CA-630 (Sigma, St. Louis, MO), 0.5% sodium associated with pathogenic angiogenesis. deoxycholate, 0.1% SDS with 100 AM of protease and phosphatase inhibitors: aprotinin, phenylmethylsulfonyl fluo- Materials and Methods ride, sodium orthovanadate, and leupeptin. Cell lysates were Reagents cleared by sonication and centrifugation. Endogenous EphA2 The EphA2-Fc soluble receptor cDNA construct was receptors were immunoprecipitated using either an ephrinA1-Fc provided by Regeneron Inc. (Tarrytown, NY) and subcloned protein or an anti-EphA2 monoclonal antibody clone D7 into episomal expression vector pCEP4 (Invitrogen, San Diego, (Upstate Biotechnology, Inc.). Immunoprecipitated proteins CA). pCEP4/EphrinA1-Fc expression vector was provided by were fractionated on 8% SDS-PAGE, transferred to nitro- Dr. A. Pandy (University of Michigan, Ann Arbor, MI). cellulose membrane, and blotted with anti-phosphotyrosine Recombinant EphA2-Fc and ephrinA1-Fc proteins were either antibodies 4G10, according to manufacturer’s instructions. The purified from culture supernatant of stable 293T clones blots were stripped and reprobed with anti-EphA2 monoclonal expressing these factors using protein A-Sepharose column, antibodies as loading control. or purchased from R&D Systems, Inc. (Minneapolis, MN). The rabbit polyclonal antibody against the nonconserved spacer Mouse Corneal Angiogenesis Assay region of ephrinA1 was provided by Immunex Inc. (Seattle, Mouse corneal angiogenesis assays were performed as WA). Anti-EphA2 monoclonal antibody clone D7 and anti- described previously (30). Briefly, hydron pellets containing phosphotyrosine monoclonal antibody 4G10 were purchased sucralfate with either vehicle alone (PBS or IgG), bFGF, from Upstate Biotechnology, Inc. (Lake Placid, NY). A VEGF-A, ephrinA1-Fc, or EphA2-Fc were prepared. Pellets monoclonal anti-VEGFR-2 antibody (1A8) was kindly were surgically implanted into corneal micropockets created 1 provided by Dr. R. Brekken (The Hope Heart Institute, Seattle, mm to the lateral corneal limbus of C57/BL6 mice. At day 5 WA). Primary endothelial cells HUVEC and HMEC were postimplantation, corneas were photographed at an incipient purchased from Clonetics (Walkersville, MD). angle of 35–50j from the polar axis in the meridian containing

FIGURE 6. VEGF induction of ephrinA1 expression and EphA2 phosphorylation in endothelial cells. A. HMEC-1 or HUVECs were pretreated with cycloheximide and stimulated with VEGF. Total cellular RNA was subjected to Northern blot analysis for ephrinA1 expression. B. Protein expression was assessed by Western blot analysis using an anti-ephrinA1 antibody. Uniform loading was confirmed by immunoblotting with an anti-h-tubulin antibody. C. VEGF induces endogenous EphA2 phosphorylation in endothelial cells. Quiescent HMEC-1 and HUVECs were untreated, treated with VEGF, ephrinA1, or VEGF in the presence or absence of soluble EphA2-Fc receptors for the indicated time. Cell lysates were immunoprecipitated with anti-EphA2 followed by Western blotting with anti-phosphotyrosine antibody 4G10. VEGF-induced endogenous EphA2 phosphorylation is inhibited by soluble EphA2-Fc (left panel).

the pellet, using a Zeiss split lamp. The fraction of the total VWR Scientific, West Chester, PA) were coated with 50 Ag/ml corneal image that was vascularized (VA), and the ratio of fibronectin (Sigma-Aldrich), 0.1% gelatin in PBS for 30 min at pixels marking neovascular capillaries both within the vascu- room temperature, followed by equilibration into DMEM/0.1% larized region (RVD) and within the total corneal image (TVD) BSA at 37jC for 1 h. Growth factor-deprived HUVECs were calculated using Bioquant software (Vanderbilt University, (Clonetics, passages 4–6, 1 Â 105) were plated in the upper Nashville, TN). Statistical analysis was performed by two- chamber of the filter well and allowed to migrate toward the tailed, paired Student’s t test. undersides of the filters in the bottom chamber containing serum-free media supplemented with 5–10 Ag/ml ephrinA1-Fc, Cell Proliferation VEGF (20 ng/ml), bFGF (20 ng/ml), or VEGF or bFGF in the Cell proliferation was measured by [3H]thymidine incorpo- presence of EphA2-Fc (5 or 10 Ag/ml) or control Fc protein. ration. Briefly, HUVECs were grown to 80% confluency in 48- After 5 h, cells adhering to the top of the transwell were well plates (Nalge NUNC International, Rochester, NY) and removed with a cotton swab, and the cells that had migrated to starved for 16 h in growth factor-deprived media. Cells were the underside of the filter were fixed and stained. For either incubated in medium alone, or stimulated for 24 h with quantification, 10 high-power fields were counted on each 5–10 Ag/ml ephrinA1-Fc, 20 ng/ml VEGF, or 20 ng/ml bFGF filter and triplicate filters were counted per experiment. in the presence of 10 Ag/ml EphA2-Fc or Fc control protein, in Experiments were repeated twice and data were pooled. growth factor-deprived media. After 24 h, cells were treated Statistical analysis was performed by two-tailed, paired with 1 ACi [3H]dTh for an additional 24 h, harvested by Student’s t test. trypsinization and transferred to 3 mm Whatmann filter paper. 3 [ H]dTh incorporation per well was determined using a beta Endothelial Cell Sprouting Assay scintillation counter. Briefly, HMEC-1 (passages 9–11) were grown to confluency on collagen-coated Cytodex 3 beads (Amersham-Pharma- Apoptosis Assay cia, Piscataway, NJ) for 5–7 days using EBM-supplemented HUVECs were cultured to near confluency on silanized media (Clonetics). The beads were plated in a gel matrix slides (Nalge NUNC, Rochester, NY) and starved for 16 h containing 5.46 mg/ml fibrinogen (Sigma-Aldrich), 2 units/ml in growth factor-deprived media. Cells were subsequently thrombin (Sigma-Aldrich), DMEM/2% fetal bovine serum and incubated with 5–10 Ag/ml ephrinA1-Fc, 20 ng/ml VEGF, the following test and control proteins: 5–10 Ag/ml ephrinA1- or 20 ng/ml bFGF in the presence of 10 Ag/ml EphA2-Fc or Fc, 20 ng/ml VEGF, VEGF plus 10 Ag/ml Fc, and VEGF plus Fc control protein, in growth factor-deprived media for 48 h. 10 Ag/ml EphA2-Fc. Serum-free media supplemented with test Cells were then fixed, subjected to TUNEL assay using an and control proteins were added to the gel matrix every 2 days. ApopTag Red in situ apoptosis detection kit (Intergen Co., Photographs were taken at days 3 and 4 using a Nikon 35-mm Purchase, NY), and counterstained with DAPI (0.05 Ag/ml camera and Ectachrome 100 film. The number of endothelial in PBS, Sigma-Aldrich, St. Louis, MO). Apoptotic cells cell sprouts exceeding the diameter of the bead was determined were quantified by numerating rhodamine-stained nuclei in for every 40 beads counted per experiment, as described (33). six random 20Â fields per sample using Scion Image ana- Experiments were repeated three times, and results were pooled lysis software. Two samples per experiment were analyzed. and analyzed for statistical significance using two-tailed, paired Experiments were repeated three times, and results were Student’s t test. pooled. Percentage of apoptosis was calculated as the number of rhodamine-stained nuclei/total number of DAPI- Northern Blot and Western Blot Analyses stained nuclei. Statistical analysis was performed using the HMEC-1 (42) or HUVECs (Clonetics) were grown to two-tailed, paired Student’s t test. near confluency and starved in endothelial cell basal medium (Clonetics) with 2% fetal bovine serum overnight. Cells were Migration Assays preincubated with 10 Ag/ml cycloheximide for 30 min and Endothelial cell migration was determined by two inde- stimulated with 20 ng/ml VEGF for 0, 2, 5, and 16 h. RNA pendent assays: a modified Boyden chamber assay (9) and a from stimulated cells were isolated using Trizol reagent (Life ‘‘wound closure’’ assay (41). For ‘‘wound closure’’ assays, Technologies, Inc., Rockville, MD). Ten micrograms of total replicate circular ‘‘wounds’’ were generated in confluent human RNA were resolved on a 1% agarose gel containing 2.2 M dermal microvascular endothelial cell (HMEC-1, passages 10– formaldehyde, transferred to a nylon membrane, and probed 12) monolayers using a silicon-tipped drill press. Serum-free with an ephrinA1 cDNA fragment. Membranes were washed media were then supplemented with the indicated factors at the once with 2Â SSC at room temperature, followed by two time of wounding. Residual fractional ‘‘wound’’ areas were stringent washes (1Â SSC at 65jC for 10 min and 0.5Â measured at 0, 2, 4, 8, and 10 h using Bioquant software SSC at 65jC for 10 min). The RNA blots were stripped and (Vanderbilt University). Mean fractional residual areas of three reprobed with a GAPDH cDNA probe as loading control. wounds were calculated, reflecting migration rates. Experi- HMEC-1 and HUVECs were maintained in 10-cm dishes ments were repeated three times and results were pooled. (unless otherwise indicated) in endothelial cell basal medium Statistical analysis was performed by two-tailed, paired with growth factors (Clonetics). HMEC-1 cells were used to Student’s t test. Briefly, for modified Boyden chamber, passage 13, and HUVECs to passage 5. Endothelial cells were polycarbonate filter wells (Transwell, Costar, 8 Am pore, cultured to 80% confluency, starved for 16 h in medium

deficient in growth factors and stimulated with 20 ng/ml VEGF inactivation of VEGF-A in areas of collagen2a1 expression results in embryonic lethality in the heterozygous state. Development, 127: 1445 – 1453, 2000. for 0, 2, 5, and 16 h. Cells were lysed in SDS sample buffer, 8. Miquerol, L., Langille, B. L., and Nagy, A. Embryonic development is and cell extracts were cleared by sonication and centrifugation. disrupted by modest increases in vascular endothelial growth factor gene Fifty micrograms of protein from cell lysates were fractionated expression. Development, 127: 3941 – 3946, 2000. on 12% SDS-PAGE, and transferred to a nitrocellulose 9. Lin, P., Sankar, S., Shan, S., Dewhirst, M. W., Polverini, P. J., Quinn, T. Q., membrane (ECL+, Amersham Pharmacia Biotech, Piscataway, and Peters, K. G. Inhibition of tumor growth by targeting tumor endothelium using a soluble vascular endothelial growth factor receptor. Cell Growth & Differ., NJ). Membranes were blocked at room temperature for 1 h in 9: 49 – 58, 1998. Tris-buffered saline containing 0.1% Tween 20 and 5% 10. Brekken, R. A., Overholser, J. P., Stastny, V. A., Waltenberger, J., Minna, powdered dry milk, and then incubated with ephrinA1-specific J. D., and Thorpe, P. E. Selective inhibition of vascular endothelial growth factor (VEGF) receptor 2 (KDR/Flk-1) activity by a monoclonal anti-VEGF antibody rabbit polyclonal antibodies (P2, specifically against the spacer blocks tumor growth in mice. Cancer Res., 60: 5117 – 5124, 2000. region of ephrinA1, Immunex) for 1 h. Immunoreactive 11. Drevs, J., Hofmann, I., Hugenschmidt, H., Wittig, C., Madjar, H., Muller, M., proteins were detected with anti-rabbit secondary antibodies Wood, J., Martiny-Baron, G., Unger, C., and Marme, D. Effects of PTK787/ZK (Amersham Pharmacia) conjugated to horseradish peroxidase, 222584, a specific inhibitor of vascular endothelial growth factor receptor tyrosine kinases, on primary tumor, metastasis, vessel density, and blood flow in a murine using enhanced chemiluminescence (Amersham Pharmacia). renal cell carcinoma model. Cancer Res., 60: 4819 – 4824, 2000. 12. Vitaliti, A., Wittmer, M., Steiner, R., Wyder, L., Neri, D., and Klemenz, R. Inhibition of tumor angiogenesis by a single-chain antibody directed against Oligonucleotide Transfection of HUVEC Cells vascular endothelial growth factor. Cancer Res., 60: 4311 – 4314, 2000. To inhibit the expression of EphA2, HUVECs were 13. Wood, J. M., Bold, G., Buchdunger, E., Cozens, R., Ferrari, S., Frei, J., transiently transfected with phosphorothioate-modified anti- Hofmann, F., Mestan, J., Mett, H., O’Reilly, T., Persohn, E., Rosel, J., Schnell, sense oligonucleotides (5V-CCAGCAGTACCACTTCCTT C., Stover, D., Theuer, A., Towbin, H., Wenger, F., Woods-Cook, K., Menrad, A., Siemeister, G., Schirner, M., Thierauch, K. H., Schneider, M. R., Drevs, J., GCCCTGCGCCG-3V), or control inverted antisense oligonu- Martiny-Baron, G., and Totzke, F. PTK787/ZK 222584, a novel and potent cleotides (5V-GCCGCGTCCCGTTCCTTCACCATGA- inhibitor of vascular endothelial growth factor receptor tyrosine kinases, impairs CGACC-3V) as described (33). Briefly, HUVECs (passages vascular endothelial growth factor-induced responses and tumor growth after oral administration. Cancer Res., 60: 2178 – 2189, 2000. 5–6) were grown at 50% confluency, and transfected with 100 14. Vajkoczy, P., Menger, M. D., Vollmar, B., Schilling, L., Schmiedek, P., Hirth, nM of oligonucleotide using oligofectamine reagent (Life K. P., Ullrich, A., and Fong, T. A. Inhibition of tumor growth, angiogenesis, and Technologies, Inc.) for 4 h at 37jC. Forty-eight hours after microcirculation by the novel Flk-1 inhibitor SU5416 as assessed by intravital transfection, cells were analyzed for effects on VEGF-induced multi-fluorescence videomicroscopy. Neoplasia, 1: 31 – 41, 1999. migration by ‘‘wound closure’’ and Boyden chamber assays, as 15. Maisonpierre, P. C., Suri, C., Jones, P. F., Bartunkova, S., Wiegand, S. J., Radziejewski, C., Compton, D., McClain, J., Aldrich, T. H., Papadopoulos, N., described above. Transfected cells were also analyzed for Daly, T. J., Davis, S., Sato, T. N., and Yancopoulos, G. D. Angiopoietin-2, a expression of EphA2 by Western blot analysis using anti- natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science, 277: 55 – EphA2 monoclonal antibody (D7, Upstate Biotechnology, Inc.). 60, 1997. 16. Asahara, T., Chen, D., Takahashi, T., Fujikawa, K., Kearney, M., Magner, M., Yancopoulos, G. D., and Isner, J. M. Tie2 receptor ligands, angiopoietin-1 and angiopoietin-2, modulate VEGF-induced postnatal neovascularization. Circ. Res., Acknowledgments 83: 233 – 240, 1998. We thank Erin Thompson for excellent technical assistance, and B. L. M. Hogan, 17. Holash, J., Maisonpierre, P. C., Compton, D., Boland, P., Alexander, C. R., L. M. Matrisian, R. DuBois, M. Boothby, J. W. Thomas, G. Miller, and R. S. Zagzag, D., Yancopoulos, G. D., and Wiegand, S. J. Vessel cooption, regression, Muraoka for helpful discussions and comments on the manuscript. Special thanks and growth in tumors mediated by angiopoietins and VEGF. Science, 284: to A. Pandey for providing pCEP-4/ephrinA1-Fc construct, R. Brekken for 1994 – 1998, 1999. generous gift of anti-VEGFR-2 antibodies, N. Boudreau for providing detailed protocol of endothelial sprouting assay, and A. Pozzi for helping with Scion 18. Holder, N. and Klein, R. Eph receptors and ephrins: effectors of morpho- image analysis. genesis. Development, 126: 2033 – 2044, 1999. 19. Cheng, N., Brantley, D., and Chen, J. The ephrins and Eph receptors in angiogenesis. Cytokine Growth Factor Rev., 13: 75 – 85, 2002. References 20. Wang, H. U., Chen, Z. F., and Anderson, D. J. Molecular distinction and 1. Gale, N. W. and Yancopoulos, G. D. Growth factors acting via endothelial angiogenic interaction between embryonic arteries and veins revealed by ephrin- cell-specific receptor tyrosine kinases: VEGFs, angiopoietins, and ephrins in B2 and its receptor Epph-B4. Cell, 93: 741 – 753, 1998. vascular development. Genes Dev., 13: 1055 – 1066, 1999. 21. Adams, R. H., Wilkinson, G. A., Weiss, C., Diella, F., Gale, N. W., Deutsch, 2. Yancopoulos, G. D., Davis, S., Gale, N. W., Rudge, J. S., Wiegand, S. J., and U., Risau, W., and Klein, R. Roles of ephrinB ligands and EphB receptors in Holash, J. Vascular-specific growth factors and blood vessel formation. Nature, cardiovascular development: demarcation of arterial/venous domains, vascular 407: 242 – 248, 2000. morphogenesis, and sprouting angiogenesis. Genes Dev., 3: 295 – 306, 1999. 3. Shalaby, F., Rossant, J., Yamaguchi, T. P., Gertsenstein, M., Wu, X. F., 22. Gerety, S. S., Wang, H. U., Chen, Z. F., and Anderson, D. J. Symmetrical Breitman, M. L., and Schuh, A. C. Failure of blood-island formation and mutant phenotypes of the receptor EphB4 and its specific transmembrane ligand vasculogenesis in Flk-1-deficient mice. Nature, 376: 62 – 66, 1995. ephrin-B2 in cardiovascular development. Mol. Cell, 4: 403 – 414, 1999. 4. Fong, G. H., Rossant, J., Gertsenstein, M., and Breitman, M. L. Role of the 23. Holzman, L. B., Marks, R. M., and Dixit, V. M. A novel immediate-early Flt-1 receptor tyrosine kinase in regulating the assembly of vascular endothelium. response gene of endothelium is induced by cytokines and encodes a secreted Nature, 376: 66 – 70, 1995. protein. Mol. Cell. Biol., 10: 5830 – 5838, 1990. 5. Carmeliet, P., Ferreira, V., Breier, G., Pollefeyt, S., Kieckens, L., Gertsenstein, 24. McBride, J. L. and Ruiz, J. C. ephrinA1 is expressed at sites of vascular M., Fahrig, M., Vandenhoeck, A., Harpal, K., Eberhardt, C., Declercq, C., development in the mouse. Mech. Dev., 77: 201 – 204, 1998. Pawling, J., Moons, L., Collen, D., Risau, W., and Nagy, A. Abnormal blood vessel development and lethality in embryos lacking a single VEGF allele. 25. Pandey, A., Shao, H., Marks, R. M., Polverini, P. J., and Dixit, V. M. Role of Nature, 380: 435 – 439, 1996. B61, the ligand for the Eck receptor tyrosine kinase, in TNF-a-induced angiogenesis. Science, 268: 567 – 569, 1995. 6. Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O’Shea, K. S., Powell-Braxton, L., Hillan, K. J., and Moore, M. W. Heterozygous embryonic 26. Daniel, T. O., Stein, E., Cerretti, D. P., St. John, P. L., Robert, B., and lethality induced by targeted inactivation of the VEGF gene. Nature, 380: 439 – Abrahamson, D. R. ELK and LERK-2 in developing kidney and microvascular 442, 1996. endothelial assembly. Kidney Int. Suppl., 57: S73 – S81, 1996. 7. Haigh, J. J., Gerber, H. P., Ferrara, N., and Wagner, E. F. Conditional 27. Ogawa, K., Pasqualini, R., Lindberg, R. A., Kain, R., Freeman, A. L., and

Pasquale, E. B. The ephrin-A1 ligand and its receptor, EphA2, are expressed 35. Gerber, H. P., McMurtrey, A., Kowalski, J., Keyt, B. A., Dixit, V., and during tumor neovascularization. Oncogene, 19: 6043 – 6052, 2000. Ferrara, N. Vascular endothelial growth factor regulates endothelial cell survival 28. Brantley, D. M., Cheng, N., Thompson, E. J., Lin, Q., Brekken, R. A., through the phophatidylinositol 3V-kinases/Akt signal transduction pathway. Thorpe, P. E., Muraoka, R. S., Cerretti, D. P., Pozzi, A., Jackson, D., Lin, C., and J. Biol. Chem., 273: 30336 – 30434, 1998. Chen, J. Soluble Eph A receptors inhibit tumor angiogenesis and progression in 36. Rousseau, S., Houle, F., Landry, J., and Huot, J. p38 MAP kinase activation vivo. Oncogene, 21: 7011 – 7026, 2002. by vascular endothelial growth factor mediates actin reorganization and cell 29. Chen, J., Nachebeh, A., Scherer, C., Ganju, P., Reith, A., Bronson, R., and migration in human endothelial cells. Oncogene, 15: 2169 – 2177, 1997. Ruley, H. E. Germline inactivation of the murine Eck receptor tyrosine kinase by 37. Holland, S. J., Gale, N. W., Mbamalu, G., Yancopoulos, G. D., Henkemeyer, retroviral insertion. Oncogene, 12: 979 – 988, 1996. M., and Pawson, T. Bidirectional signaling through the EPH-family receptor Nuk 30. Kenyon, B. M., Voest, E. E., Chen, C. C., Flynn, E., Folkman, J., and and its transmembrane ligands. Nature, 383: 722 – 725, 1996. D’Amato, R. J. A model of angiogenesis in the mouse cornea. Invest. 38. Bruckner, K., Pasquale, E. B., and Klein, R. Tyrosine phosphorylation of Ophthalmol. Vis. Sci., 37: 1625 – 1632, 1996. transmembrane ligands for Eph receptors. Science, 275: 1640 – 1643, 1997. 31. Phillips, G. D., Stone, A. M., Jones, B. D., Schultz, J. C., Whitehead, R. A., 39. Davy, A., Nicholas, W. G., Murray, E. W., Klinghoffer, R. A., Soriano, P., and Knighton, D. R. Vascular endothelial growth factor (rhVEGF165) stimulates Feuerstein, C., and Robbins, S. M. Compartmentalized signaling by GPI- direct angiogenesis in the rabbit cornea. In Vivo, 8: 961 – 965, 1994. anchored ephrin-A5 requires the Fyn tyrosine kinase to regulate cellular adhesion. 32. Griffioen, A. W. and Molema, G. Angiogenesis: potentials for pharmacologic Genes Dev., 13: 3125 – 3135, 1999. intervention in the treatment of cancer, cardiovascular diseases, and chronic 40. Huai, J. and Drescher, U. An ephrinA dependent signaling pathway controls inflammation. Pharmacol. Rev., 52: 237 – 268, 2000. integrin function and is linked to the tyrosine phosphorylation of a 120 kDa 33. Kolizek, T. I., Weiss, C., Yancopoulos, G. D., Deutsch, U., and Risau, W. protein. J. Biol. Chem., 276: 6689 – 6694, 2001. Angiopoietin-1 induces sprouting angiogenesis in vitro. Curr. Biol., 8: 529 – 532, 41. Daniel, T. O., Liu, H., Morrow, J. D., Crews, B. C., and Marnett, L. J. 1998. Thromboxane A2 is a mediator of cyclooxygenase-2-dependent endothelial 34. Hess, A. R., Seftor, E. A., Gardner, L. M., Carles-Kinch, K., Schneider, G. B., migration and angiogenesis. Cancer Res., 59: 4574 – 4577, 1999. Seftor, R. E., Kinch, M. S., and Hendrix, M. J. Molecular regulation of tumor cell 42. Ades, E. W., Candal, F. J., Swerlick, R. A., George, V. G., Summers, S., vasculogenic mimicry by tyrosine phosphorylation: role of epithelial cell kinase Bosse, D. C., and Lawley, T. J. HMEC-1: establishment of an immortalized human (Eck/EphA2). Cancer Res., 61: 3250 – 3255, 2001. microvascular endothelial cell line. J. Invest. Dermatol., 99: 683 – 690, 1992.

Downloaded from mcr.aacrjournals.org on September 30, 2021. © 2002 American Association for Cancer Research. Blockade of EphA Receptor Tyrosine Kinase Activation Inhibits Vascular Endothelial Cell Growth Factor-Induced Angiogenesis11 NIH Grants HD36400 and DK47078; JDF grant I-2001-519; DOD grant BC010265; American Heart Association Grant 97300889N; ACS Institutional Research Grant IN-25-38 (to J. Chen); Vascular Biology Training Grant T32-HL-07751-06 and American Heart Association Fellowship 0120147B (to D. Brantley); Cancer training Grant T-32 CA09592 (to N. Cheng); and a core facilities Grant 2P30CA68485 to the Vanderbilt-Ingram Cancer Center.

Nikki Cheng, Dana M. Brantley, Hua Liu, et al.

Mol Cancer Res 2002;1:2-11.

Updated version Access the most recent version of this article at: http://mcr.aacrjournals.org/content/1/1/2

Cited articles This article cites 42 articles, 22 of which you can access for free at: http://mcr.aacrjournals.org/content/1/1/2.full#ref-list-1

Citing articles This article has been cited by 36 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/1/1/2.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/1/1/2. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.