Dynamic Interactions between Cancer Cells and the...

Oncogenes and Tumor Suppressors

Dynamic Interactions between Cancer Cells and theEmbryonic Microenvironment Regulate Cell Invasion andReveal EphB6 as a Metastasis Suppressor

Caleb M. Bailey1 and Paul M. Kulesa1,2

AbstractMetastatic dissemination drives the highmortality associatedwithmelanoma.However, difficulties in visualizing

in vivo cell dynamics during metastatic invasion have limited our understanding of these cell behaviors. Recentevidence has revealed that melanoma cells exploit portions of their ancestral embryonic neural crest emigrationprogram to facilitate invasion. What remains to be determined is how embryonic microenvironmental signalsinfluence invasivemelanoma cell behavior, andwhether these signals are relevant to human disease. To address thesequestions, we interrogated the role of the neural crest microenvironment in dictating the spatiotemporal pattern ofmelanoma cell invasion in the chick embryo using 2-photon time-lapse microscopy. Results reveal that bothpermissive and inhibitory neural crest microenvironmental signals regulate the timing and direction of melanomainvasion to coincide with the neural crest migration pattern. These cues include bidirectional signaling mediatedthrough the ephrin family of receptor tyrosine kinases. We demonstrate that EphB6 reexpression forces metastaticmelanoma cells to deviate from the canonical migration pattern observed in the chick embryo transplant model.Furthermore, EphB6-expressing melanoma cells display significantly reduced metastatic potential in a chorioal-lantoic membrane (CAM) metastasis assay. These data on melanoma invasion in the embryonic neural crest andCAM microenvironments identify EphB6 as a metastasis suppressor in melanoma, likely acting at the stage ofintravasation.

Implications: This article links cellular metastasis to behaviors observed in the ancestrally related embryonicneural crest and demonstrates the powerful influence of the embryonic microenvironment in regulating cellmigratory behavior. Mol Cancer Res; 12(9); 1–11. �2014 AACR.

IntroductionThe vast majority of all cancer-related deaths can be

ascribed to metastasis. With tumor cell–microenvironmentinteractions at the forefront of metastatic disease progres-sion, insufficient attention has been given to the role of themicroenvironment in regulating cell migratory behaviors.This is primarily due to the inherent challenges associatedwith studying migratory behaviors throughout the meta-static cascade in vivo (1–3).The embryonic neural crest offers a unique model system

in which to study cell–microenvironment interactions in

vivo. The neural crest is a highly invasive, multipotent cellpopulation that displays a regulated spatiotemporal migra-tory pattern in the vertebrate embryo (4). Neural crest cellsemigrate from the dorsal neural tube in a rostral-to-caudalmanner all along the vertebrate axis, including from rhom-bomere segments (r1–r7) of the hindbrain. During emigra-tion, the neural crest follows stereotypical migratory path-ways that are thought to be sculpted by both intrinsic andextrinsic guidance cues found within the neural tube andsurrounding microenvironment (5, 6). Importantly, severaltumor types, including melanoma and neuroblastoma, orig-inate from neural crest–derived cells. This has led to thehypothesis that neural crest–related malignancies may beintrinsically predisposed to increased metastatic potentialdue to the inherent invasive abilities of the neural crest (7–9).Our laboratory has developed a model system using the

chick embryo to help overcome the significant roadblocks tostudying the cellular and molecular dynamics of melanomametastasis in vivo (10). Thismodel system takes advantage ofthe accessibility of the embryonic microenvironment to invivo imaging and molecular intervention, allowing us todirectly investigate how melanoma cells respond to micro-environmental signals. We and others have shown that

Authors' Affiliations: 1Stowers Institute for Medical Research, KansasCity, Missouri; and 2Department of Cell Biology and Anatomy, University ofKansas School of Medicine, Kansas City, Kansas

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

Corresponding Author: Paul M. Kulesa, Stowers Institute for MedicalResearch, 1000 East 50th Street, Kansas City, MO 64110. Phone: 816-926-4453; Fax: 816-926-2074; E-mail: [email protected]

doi: 10.1158/1541-7786.MCR-13-0673

�2014 American Association for Cancer Research.

MolecularCancer

Research

www.aacrjournals.org OF1

on August 9, 2019. © 2014 American Association for Cancer Research.mcr.aacrjournals.org Downloaded from

Published OnlineFirst May 16, 2014; DOI: 10.1158/1541-7786.MCR-13-0673

http://mcr.aacrjournals.org/

metastatic melanoma cells transplanted into the chick neuralcrest embryonic microenvironment migrate along stereo-typic neural crest migratory pathways (7, 10–12). However,the mechanisms guiding their migration are not known. Toaddress this, we recently performed a molecular analysiscomparing transplantedmelanoma cells and the neural crest,which revealed that metastatic melanoma cells revive por-tions of the embryonic neural crest emigration program (7).Thus, metastatic melanoma cells appear to hijack inherentneural crest–related developmental signaling pathways toenhance their metastatic potential. However, what remainsunclear is how the embryonic microenvironment dictatesmelanoma cell migratory behavior. Specifically, what are theembryonic signals that guide melanoma migration, and canperturbation of those signals significantly alter migratorybehavior?Here, we asked to what extent the chick embryonic neural

crest microenvironment regulates the timing and migratorypatterning of transplanted melanoma cells. We also asked towhat extent we could alter the migratory phenotype byperturbing cell–microenvironment interactions. We com-pared the invasion patterns of melanoma cells transplantedinto the chick hindbrain at various developmental stagesand axial positions. Single melanoma cell dynamics wereobserved in vivo using 2-photon microscopy. To perturbcell–microenvironment interactions, we examined howchanges in Eph expression in transplanted melanoma cellsaffected cell invasion patterns. Finally, to address the rele-vance of our studies to human disease, we assayed thetumorigenicity and metastatic potential of melanoma cellstransplanted onto the highly vascularized chick chorioallan-toic membrane (CAM). Our results highlight the impor-tance of tumor cell–microenvironment interactions in pro-moting, inhibiting, and guiding tumor cell movements andelucidate the antimetastatic properties of EphB6 in vivo.

Materials and MethodsCell cultureCell lines were kindly provided by Dr. Mary Hendrix,

Children's Memorial Research Center (Chicago, IL). Cellswere maintained in RPMI supplemented with 10% FBS.For drop culture conditions, cells were resuspended inRPMIplus 10% FBS at a concentration of 4e6 cells per mL. A 20mL drop was hung from the lid of a 35-mm culture dish andincubated for 40 hours under standard culture conditions.

EphB6 cloning, lentiviral infection, and WesternblottingThe vector containing wild-type human EphB6 was

kindly provided by Dr. Andrew Freywald, University ofSaskatchewan (Saskatoon, SK, Canada). From this vector,EphB6 was PCR amplified and inserted into the Nhe1 andBamH1 sites of the lentiviral vector pCDH-CMV-MCS-EF1-RFP (System Biosciences) using the following PCRamplification primers: forward, 50-AGTCGCTAGCATG-GTGTGTAGCCTATGGGTGC-30; reverse, 50-GACTG-GATCCTCAGACCTCCACTGAGCCCT-30. Lentiviral

particles were prepared using the ViraPower LentiviralPackaging mix, Lipofectamine 2000, and the 293FT cellline according to the manufacturer's directions (Invitrogen).Supernatant was collected at 48 and 72 hours posttransfec-tion and was concentrated using PEG precipitation. Briefly,the supernatant was collected, spun at 3,000 rpm for 5minutes, and then filtered using a 0.45-mm filter. For theprecipitation, 3 mL of filter-sterilized 40% PEG-8000 inPBS and 180mLFBSwas added to each 9mLof supernatant.The virus was allowed to precipitate for at least 72 hoursbefore pelleting at 1,500 � g for 10 minutes. The superna-tant was removed and the tubewas spun down again at 1,500� g for 1 minute to remove residual PEG solution. The viralpellet was resuspended in culture media by gently pipettingup and down, before being stored at �80�C before use.Supernatant was concentrated 1:500 to 1:1,000, resulting inviral titers as high as 108 titerable units (TU)/mL. Forinfection, C8161 cells were seeded into 24-well plates at1� 104 cells per well and allowed to adhere overnight. Thenext day, the cells were infected at various multiplicity ofinfection (MOI) in culturemedia in the presence of 6mg/mLpolybrene (Sigma-Aldrich) overnight, before undergoing amedia change. The cells were then expanded before trans-plant experiments. EphB6 protein expression was verified byWestern blotting. Equal amounts of cellular protein weresubjected to SDS-PAGE andWestern blot analysis using anantibody specific for EphB6 (EphB6 monoclonal antibodyM03, clone 5D8, Abnova H00002051-M03).

Chick embryo transplantsFertile white leghorn chick eggs were acquired from

Centurion Poultry. Eggs were incubated at 38�C for approx-imately 39 hours until the 9-somite stage of development.Eggs were windowed and embryos were visualized followingthe injection of a solution of 10% India ink in HowardRinger's solution below the blastodisc. Embryos were stagedaccording to the criteria of Hamburger and Hamilton(Hamburger and Hamilton, 1951), denoted as stage 10(HH10), for example. Transplantation of the embryos wasperformed as previously described (10). Briefly, C8161hanging drop cultures were cut with a finely pulled glassneedle to approximately 50 mm3 (corresponding roughly to300 cells). Tumor cell transplants were grafted into thelumen of the chick embryo neural tube through a slit alongthe dorsal midline. Eggs were sealed with cellophane tapeand incubated for 24 hours.

Static imagingIndividual embryos were removed from eggs with paper

rings, rinsedwith Ringer's solution, and placed dorsal side upwithin a thin ring of high-vacuum grease (Dow Corning) on22 � 75 mm2 microslides. Embryos were imaged using aZeiss 710 multi-photon upright microscope. 2-photon z-stacks of dual labeled C8161 Gap43:CFP::H2B:YFP wereacquired at a wavelength of 850 nm. RFP-labeled EphB6þC8161 cells were imaged with an excitation of 561 nm.Images using aWPlan-Apochromat 20�/1.0DIC objective(Zeiss).

Bailey and Kulesa

Mol Cancer Res; 12(9) September 2014 Molecular Cancer ResearchOF2




PhotoconversionC8161 cells were dual labeled with H2B:mCherry and

H2B:PSCFP2 (photo-switchable cyan fluorescent protein)by lentiviral infection. Cells were transplanted into the chickembryo neural tube as described above. A Teflon membranewindow was installed in the egg as previously described (13).H2B:PSCFP2-labeled cells were photoconverted within 1hour following transplantation. Photoconversion was per-formed on a Zeiss 710 multiphoton upright microscopeusing 2-photon excitation at 780 nm and a W Plan-Apoc-hromat 20�/1.0 DIC objective (Zeiss).

Time-lapse imagingThe time-lapse imaging platform, including sample prep-

aration, is detailed in a previously published protocol (13).Briefly, Teflon windows were placed over the developingchick embryo and sealed with beeswax. The egg was main-tained in an environmental box placed around a Zeiss 710multiphoton upright microscope stage and heated to 38�C.The chamber was humidified with a sponge placed in a dishof water. Nonevaporating immersion liquid (Immersol W,Carl Zeiss) was used to bridge the W Plan-Apochromat20�/1.0 DIC objective (Zeiss) with the Teflon membrane.Simultaneous 2-photon excitation of the dual-labeledC8161 Gap43:CFP::H2B:YFP cells was achieved at a wave-length of 850 nm. Z-stacks with a 10-mm z-slice wereacquired every 7 minutes for 18 hours.

CAM metastasis assayFertile white leghorn chick eggs were incubated for 7

hours, following which an artificial air sac was created in theegg by removing 4-mL albumin from the egg with an 18-guage needle. The puncture hole was sealed with cellophanetape and the egg was further incubated for 7 days. At the timeof transplantation, a small section (2 � 2 mm2) of the sur-face periderm of the CAM was removed with filter paper(Whatman) in a manner that did not compromise theintegrity of the basal layer. Disruption of the periderm isrequired for tumor cells to gain access to the underlyingmesoderm (14). Following periderm removal, eggs wereallowed to recover for 2 hours before placing cells on theCAM. For transplantation, cells were suspended at a con-centration of 1e5 cells/mL. A total of 1e6 cells (10 mL) wasplaced on the CAM. The egg was resealed with cellophanetape and incubated for 48 hours. Following this incubation,the egg was bisected along the longitudinal axis, and theCAM from the lower half of the egg was placed in a 15-mLconical tube and flash frozen in liquid nitrogen. The frozentissue was then ground with a chilled mortar and pestle inthe presence of liquid nitrogen. gDNA was extracted fromthe ground tissue using the Wizard Genomic DNA Puri-fication kit (Promega) according to the manufacturer'sprotocol. Resulting DNA was quantified using a Nanodropspectrophotometer.

Quantitative PCRHuman DNA was detected and quantified using Alu

(YB8 subfamily) element-based PCR (15). Quantitative

PCR (qPCR) was performed on an ABI7500 (AppliedBiosystems Inc.) in optical 96-well reaction plates usingthe following 3-stage cycling protocol: 50�C for 2 min-utes, 95�C for 10 minutes and 40 cycles of 95�C for15 seconds and 68�C for 1 minute. The following Alu-specific primers and probe were used: forward, 50-GTC-AGGAGATCGAGACCATCCT-30; reverse, 50-AGT-GGCGCAATCTCGGC-30 and the following probe: 50-6-FAM-AGCTACTCGGGAGGCTGAGGCAGGATA-MRA-30 as previously reported (16). Each 50 mL qPCRreaction included 2 mg genomic DNA, 25 mL 2X GeneExpression Master Mix (Applied Biosystems Inc.), 25pmol forward primer, 25 pmol reverse primer, and12.5 pmol probe. Resulting cycle threshold values (Ct)were compared with a serial dilution of human gDNAfrom C8161 cells, from which quantitative DNA amountswere extrapolated.

Data analysis and statistics, image processingThree-dimensional images acquired from 2-photon static

and time-lapse imaging were analyzed with Imaris software(Bitplane Scientific Software). Analysis included spots detec-tion cell tracking, tracks translation, displacement vectoranalysis, and movie creation and editing. For the cylindricalcoordinate conversion, Cartesian coordinates from Imariswere converted to cylindrical coordinates using standardmathematical methods. Densitometric scatterplots werecreated in MatLab. Histograms and curve-fitting analyseswere created and performed using OriginPro software.Curve fitting was based on a Gaussian peak function, whereXc is the center of the peak andW equals 2 times the SD ofthe Gaussian distribution (�0.849 the width of the peak athalf height). Statistical analysis was performed using Micro-soft Excel and the Data Analysis Tools pack. For migratorydistance comparisons, a single-factor ANOVA was used tocalculate the P value. For the CAM metastasis assay, a 2-sample equal variance t test with 2-tailed distribution wasused. Statistically significant P values are <0.05. Figureprocessing was performed with Adobe Photoshop CS3.

ResultsMelanoma cells transplanted into the chick embryonic

neural crest microenvironment sense and respond to micro-environmental cues by following host cranial neural crest cellmigratory pathways (7, 10). However, it remained unclearwhether there was a temporal and/or spatial restriction tomelanoma migration that corresponded to the developmen-tal pattern of the host cranial neural crest.To examine this question, we transplanted human C8161

melanoma cells into different rhombomere segments at peakand off-peak times of cranial neural crest cell migration(Fig. 1). Cranial neural crest cells begin to migrate from r1 atHamburger andHamilton stage 8þ (HH8þ) andmigrationtypically ceases (from r7) by HH11þ (17). We observedmaximal invasion of C8161 cells when transplanted into r4at HH10. Under these conditions, C8161 cells invaded thepermissive area adjacent to r4, travelling as a discrete mul-ticellular stream while respecting inhibitory cues present in

EphB6 Is a Metastasis Suppressor in Melanoma

www.aacrjournals.org Mol Cancer Res; 12(9) September 2014 OF3




and adjacent to r3 and r5 that are thought to help sculpt theneural crest migration pathway (Fig. 1B). To analyze migra-tory behaviors, positions of invasive cells from 9 differenttransplants were quantified using a cylindrical coordinatesystem (Fig. 1A and B). The embryonic dorsal midline andthe r4–r5 rhombomere boundary were used as fiduciallandmarks (Fig. 1A and B). The cell invasion patternrevealed a tight cluster of endpoint cell positions locatedalong the r4 stream approximately 25mm rostral to the r4–r5boundary, demonstrating reproducible, guided directionalmigration (Fig. 1B). Following 24 hours, the average migra-tory distance was 200 mm, with some cells migrating nearly600 mm (Fig. 1C).Importantly, guided directionalmigrationwas observed for

other embryonic and neural crest–related cell types, includingneuroblastoma, but not for the invasive breast cancer cell lineMDA-MB-231 (Supplementary Fig. S1). This suggests thatembryonic neural crest signals are nonpermissive to cellslacking the appropriate signaling mechanism.

Cranial neural crest streams are sculpted by both per-missive and inhibitory signals present within the micro-environment, with inhibitory neural crest cell–free zoneslocated adjacent to neural crest migratory streams (5, 18).As such, we investigated the extent to which inhibitorycues associated with r3 and r5 also regulate melanoma cellmigration. For this, we transplanted C8161 cells into r3and r5 and quantified their migratory behaviors. Cellstransplanted into either r3 or r5 displayed a significantreduction in migration compared with cells transplantedinto r4 (Fig. 1B and C). C8161 cells transplanted into r3showed an average migratory distance of 104 mm, repre-senting a 50% reduction in migration compared with cellstransplanted into r4 (Fig. 1C). Cells transplanted into r5displayed even greater inhibition, with an average migra-tory distance of only 66 mm. Those cells showing apropensity to migrate appeared as multicellular streamsoriented rostrally or caudally toward adjacent neural crestmigratory streams. This suggests that the microenvironments

Figure 1. Cytometric quantification of migratory behaviors of melanoma cells transplanted into varying regions and developmental stages of the chickembryonic hindbrain. A, cartoon depicting the experimental methods. Tumor cells are transplanted into various regions of the embryonic chick hindbrain andallowed tomigrate for 24 hours. Cells are then imaged in high resolution, and cell positions are quantified using a cylindrical coordinate systemwhere length (L)represents the cell position along the rostral–caudal axis and theta (q) represents the angle deviant from the vertical (dorsal–ventral) axis of the embryo and is adirect representation of lateral migratory distance. B, cytometric quantification of transplanted cells in different regions of the hindbrain (r3–r5) duringactive neural crestmigration (HH10) and following the cessation of neural crestmigration (HH12). The representativemicrograph shows typicalmelanoma cellmigration relative to the location and timing of the transplant. Meter bar, 100 mm. The 30 most invasive cells from 9 transplants were used to populate thedensitometric scatterplot that defines the cell invasion pattern. The scatterplot uses embryonic landmarks as fiducial points representing the embryonicmidline (y-axis) and the boundary between r4 and r5 (red line at axis position 0). C, comparison of migratory ability (total cell displacement) is shown bythe histogram and accompanying Gaussian fit. Xc is the center of the peak and represents the average x-value. W equals 2 times the SD of the Gaussiandistribution or approximately 0.849 the width of the peak at half height. The dashed line in the histograms represents the Xc value of the HH10/r4 transplant.n ¼ 270 cells for each histogram. �, P < 0.05.

Bailey and Kulesa





within and adjacent to r3 and r5 induce potent antimigratoryresponses.Cranial neural crest cells at the r4 axial level cease to

emerge from the dorsal neural tube at approximatelyHH11þ (17). We asked whether age-related changes inthe neural crest microenvironment would affect tumor cellbehaviors in later staged transplants. To test this, we trans-planted C8161 cells into r4 at HH12 (Fig. 1B). We foundthat melanoma cells transplanted into r4 at later develop-mental stages spread radially from the transplant site, with-out respecting the neural crest cell–free zones (Fig. 1B).Furthermore, cells migrated independent of one another anddid not appear to form discrete multicellular streams likethose observed in cells transplanted at HH10 (Fig. 1B).Interestingly, with an average migratory distance of 172 um,these cells also migrated less (based on total displacement)than cells transplanted at stage HH10 (Fig. 1C). From this,we conclude that guidance factors present in the embryonic

microenvironment at HH10 do not appear to guide migra-tion at later developmental stages.

The ability to migrate is cell-autonomousThese results suggest that the embryonic neural crest

microenvironment regulates melanoma migratory behaviorswhen transplantation is correlated with the developmentaltiming of cranial neural crest migration. We next askedwhether the proximity of a melanoma cell to the microenvi-ronment (neural tube lumen) influences melanoma cell inva-sion. To test this, we used an in vivo selective cell labelingtechnique in which H2B-PSCFP2 (photoswitchable cyanfluorescent protein)-labeled C8161 cell clusters were trans-planted into the embryo (r4,HH10).Using 2-photonmicros-copy,we photoconverted a small number of cells located eitherwithin the core or at the periphery of the transplant (Fig. 2Aand B; Supplementary Video S1). Following 24 hours, photo-converted cells were scored as migratory or nonmigratory.We

Figure 2. Migratory ability is cell-autonomous. A, cartoon depicting the 2-photon in ovo photoconversion of H2B:PSCFP2-labeled C8161 melanoma cells.Following transplantation into the chick embryo (r4,HH10), subpopulations of tumor cells in either the core or the perimeter of the graft were photoconvertedwith a 2-photon microscope using a wavelength of 780 nm. B, representative micrographs showing the photoconversion of tumor cells at t¼ 0 hour in eitherthe core or the perimeter of the graft. XY and XZ views are provided. C, representative micrographs showing the locations of migratory photoconvertedtumor cells at t ¼ 24 hours. The white circles outline the initial transplant graft location and show the nonmigrating cells remaining in the neural tube.D, pie charts depicting the percentage of photoconvertedmigratory verses nonmigratory cells. Cells were scored asmigrating or nonmigrating based on theirlocation within or without this circle. E, in vitro time-lapse analysis of a C8161 melanoma cell cluster placed on a basement membrane matrix. Cells werelabeled based on position (red ¼ periphery, green ¼ core). Migratory tracks are shown at 5 hours. F, rose plot comparing cell directionality in vitro withdirectionality observed in the embryo. In the embryo, 90 degrees represents the anterior–posterior embryo axis. Cell positionswere calculated for all migratingcells from 9 different transplants (r4, HH10, >1,000 cells). Angles between cell trajectory and the horizontal r4 migratory stream were then determined.The size of each bar depicts the number of binned cells for a given angle. The colored segments depict the distance migrated by cells within the bin.






found thatmigratory cells derived from either the periphery orcore were similarly likely to be located in the leading popu-lation of migratory cells as those that started at the periphery(Fig. 2C and D). This contrasted with melanoma cell clustersplaced on a basement membrane matrix in culture. In vitro,cells spread radially and without direction from the microen-vironment, with the cells originating at the outer rim of thecluster remaining on the migratory front (Fig. 2E and F).Thus, the ability to sense and respond to migratory anddirectional cues within the embryonic neural crest microen-vironment appears to be cell autonomous.

The embryonic neural crest microenvironment dictatesmigratory behaviorsWe next verified by in ovo 2-photon time-lapse micros-

copy that migratory cells unpredictably arose from any-where within the transplant cluster, with no observed biasfor any given region (Fig. 3A; Supplementary Video S2).Yet it remained unclear whether cells migrated indepen-dently or in association with neighboring cells. Thus, foreach of the 10 most invasive cells identified in thetransplant, we analyzed directional migration for thehighly invasive cell's 6 closest neighbors (Fig. 3B–D,Supplementary Video S3). This 3D analysis revealed thatdisplacement vectors for neighboring cells rarely pointedin the same direction as the most invasive cell (Fig. 3C).We calculated the average deviation between the displace-ment vector of the invasive cell and those of its neighborsto be 32 degrees, indicating that neighboring cells do nottend to follow the direction of the highly migratory cells(Fig. 3D). We conclude that highly invasive tumor cellsmigrate independently of neighboring cells and do notadopt a follow-the-leader pattern of migration.We then investigated whether the location of a cell within

the cluster at the time of transplantation could predict thetiming and trajectory of its invasion pattern (Fig. 4A and B;Supplementary Video S4). We pseudocolored invasive mel-anoma cells at endpoint times based on whether the cell hadmigrated to the left or to the right of the embryo anterior–posterior axis along the dorsalmidline.We then rewound thetime-lapse to reveal the cell's initial position (Fig. 4A and B).We observed that cells that had migrated to the right side ofthe embryo originated from the right half of the transplantcluster and vice versa, mimicking the host embryonic neuralcrest dispersion pattern. This suggested that transplantedmelanoma cells were sorted by microenvironmental signalsat the dorsal midline.

EphB6 reexpression in melanoma cells alters migratorybehaviorWhat embryonic neural crest microenvironmental factors

are responsible for directing melanoma cell migratorybehaviors? As Eph/ephrin signaling is known to directembryonic neural crest migration (19–21), we sought toidentify novel migratory phenotypes resulting from alteredEph expression. To accomplish this, we reexpressed EphB6,a kinase-defective Eph receptor, in C8161 melanoma cells.EphB6 protein expression was confirmed by Western blot-

ting (Supplemental Fig. S2). We previously demonstratedthat EphB6 gene expression was silenced in C8161 cells butpresent in both primary melanocytes and poorly invasiveC81-61 melanoma cells (7). Furthermore, EphB6 has beenshown to dimerize with, and modulate the activities of,several other Eph receptors. These include EphB1, EphB2,EphB4, and EphA2, all of which are highly expressed inC8161 cells (7, 22–24). As such, we hypothesized thatEphB6 reexpression would have significant effects on theinteraction between transplanted C8161 cells and the chickmicroenvironment. To test this, we transplanted EphB6þ

C8161 cells into the embryo (r4, HH10) and quantified theresulting cell invasion pattern (Fig. 5). EphB6þ C8161 cellsinitially exited the neural tube at the proper location rostralto the r4–r5 boundary. However, after encountering themicroenvironment adjacent to r4, EphB6þ cells altered their

Figure 3. Invasive melanoma do not display follow-the-leader migratorybehaviors. A, colored spheres represent the initial positions of the 10most invasive cells (based on total displacement). z depth can be gaugedby the diameter of the sphere. Both the raw data image and a referencecartoon are provided. B, the migratory direction (displacement) of arepresentative highly invasive cell (red) was comparedwith the directionsof its 6 closest neighboring cells (yellow). Initial positions are viewed at t¼0 hour. Final positions, including displacement vectors, are shown at t¼18 hours. The dashed circles represent the initial locations of each cell.C, displacement vectors for all of the cells shown in section "D" weretranslated to a common origin and compared for changes in direction.The red þ marks the origin. D, the deviation angles between thedisplacement vectors of the 10 most invasive cells and each of their6 closest neighborswascalculated andgraphedona roseplot histogram.The red arrow indicates the direction of the highly invasive cells.The average deviation was �32 degrees.

Bailey and Kulesa





trajectories away from the area adjacent to r4 and toward thearea adjacent r5 (Fig. 5A–F). This resulted in a caudal shift inthe mean position of migrating cells (�75 mm) along therostral–caudal axis (Fig. 5F and G). Importantly, theEphB6þ cells maintained the discrete multicellular stream-ing behavior observed in transplanted parental C8161 cells,suggesting that cell–cell interactions among the melanomacells was not altered. Also, the overall distance that EphB6þ

C8161 cells invaded into the embryonic neural crest micro-environment was unaffected (Fig. 5H).

EphB6 is a metastasis suppressor in melanomaFollowing this observation, we investigated whether the

altered cell directionality observed in the embryo reflects achange in tumorigenicity or metastatic potential (Fig. 6).Parental C8161 cells or EphB6þ C8161 cells were placedonto the highly vascularized CAM of an E10 chick embryoand incubated for 48 hours. We observed that EphB6þ

C8161 cells maintained their ability to form tumors on theCAM in a similar manner to that observed with parentalC8161 cells (Fig. 6A). The isogenic but poorly aggressivemelanoma cell line C81-61 did not form tumors whenplaced on the CAM but rather remained as a monolayer ofdispersed cells (Fig. 6A).To test whether EphB6 reexpression directly affects

metastasis, we examined the ability of tumor cells to invadeinto the host vasculature. To assay intravasation, the bottomhalf of the CAM was removed 48 hours posttransplantation

and assayed for the presence of human DNA. Because ofphysical and temporal constraints, tumor cells must intra-vasate into the blood or lymph circulation to populate thebottom half of the CAM within 48 hours (25, 26). PCRamplification of a human-specific Alu repeat sequenceallowed us to quantify the presence of metastatic humancells (15, 27).We evaluated 16 replicates from both parentalC8161 cells and EphB6þ C8161 cells. We observed that50% of the parental C8161 samples showed high metastaticability, with maximal rates observed at more than 900 pghuman DNA per 2 mg chick DNA (Fig. 6C). ParentalC8161 cells showed an average of 225 pg human DNA persample over 48 hours (corresponding roughly to 45 cells persample, based on 1 cell ¼ 5 pg DNA, Fig. 6C). In contrast,EphB6þ C8161 cells displayed a marked reduction inmetastatic ability (17 pg per sample or 3.4 cells). Theseresults support the hypothesis that EphB6 acts as ametastasissuppressor inmelanoma cells and suggest intravasation as thestep likely blocked by EphB6.

DiscussionIn the present study, we sought to use an embryonicmodel

system to examine the role of the microenvironment inpromoting and directing metastatic invasion in vivo. Using adynamic in vivo imaging platform to assess single-cell beha-viors, we discovered 3 main results. First, embryonic neuralcrest microenvironmental signals regulate the timing, tra-jectory, and order of metastatic melanoma dissemination

Figure 4. Invasive melanoma cellsare guided by the embryonic neuralcrest microenvironment. A, 2-photon time-lapse cell trackingwas used to determine the initiallocations of migratory tumor cells.3D time-lapse images wereacquired in ovo using 2-photonmicroscopy. Images wereanalyzed using Imaris software totrack cells over time. B, cells werelabeled at t ¼ 18 hours based onwhether they had migrated to theleft or right of the embryonic dorsalmidline. Cells on the left werelabeled green and cells on the rightwere labeled red. Cell positionswere then tracked backwards toidentify the initial position at t ¼ 0hour. Both XY and XZ views areprovided.






after transplantation into the chick hindbrain. Second, cueswithin the chick neural tube and surrounding microenvi-ronment regulated the migratory behaviors of invasive mel-anoma cells, but we could not predict, based on location,which cells would respond in an invasive manner to thosecues. Third, reexpression of a kinase-defective Eph receptor,EphB6, in melanoma cells induced aberrant directionaltumor cell migration from the hindbrain and loss of met-astatic potential within the chick CAM vascular bed.Our results are the first comprehensive details of in vivo

human melanoma cell dynamics at the single-cell levelwithin the embryonic neural crest and vascularized CAMmicroenvironments.Our work approaches discerning just how susceptible

tumor cells are to microenvironmental migration cuesand whether an understanding of those cues might helppredict metastatic behaviors. The early embryonic hind-brain uses a complex set of signaling mechanisms tocontrol and guide the invasion of the neural crest (5).These same signals appear to be sufficient to regulatemelanoma invasion in both time and space, includingdirectional and nonpermissive cues. Melanoma cells trans-planted at a developmental stage after neural crest cellshave emigrated from the hindbrain, or into odd-numberedrhombomere hindbrain segments (that typically sup-port fewer migrating neural crest cells) fail to replicatethe behaviors of cells transplanted into r4 at stage HH10(Fig. 1). These observations demonstrate that confin-ed microenvironments can directly regulate tumor cell

Figure 6. Reexpression of EphB6 in C8161 melanoma cells causes asignificant loss of metastatic potential but does not affect tumorigenicityin a CAMmetastasis assay. A, images of a native chick embryo CAM andrepresentative tumors from parental C8161 cells (green) and EphB6þC8161 cells (red) grown on the CAM. 1e6 cells in suspension (10 mlvolume) were dropped onto the CAM at day E10. Tumor formationoccurred over 48 hours. NonmetastaticC81-61 cells did not form tumors.The black meter bar represents 1 cm and the white meter bars represent2.5 mm. B, cartoon depicting the CAMmetastasis assay. Tumor cells areplaced onto the upper CAM through awindow in the egg shell. The egg isresealed for 48 hours. The lower CAM is removed and genomic DNA isharvested from the tissue. Metastatic human cells are detected by qPCRusing primers that amplify a human-specific Alu element. C, scatterplotshowing detected amounts of human DNA per 2 mg chick CAM DNA.Each X represents one biologic replicate. A t test was used to calculatestatistical significance.

Figure 5. Reexpression of EphB6 in C8161 melanoma cells results inaberrant directional migration without affecting migratory ability. A,cartoon depicting the transplant of EphB6þ C8161 melanoma cells intor4 atHH10. B, representativemicrograph showingmigratory behaviors oftransplanted EphB6þ C8161 cells. The neural tube boundary is outlinedin yellowand thepositionof theotic vesicle (ov) is shownby thedotted yellline. Individual rhombomeres are also labeled. C, densitometricscatterplot of the cell invasion pattern of the positions of the 30 mostinvasive cells from 10 transplant experiments. The y-axis represents theembryo midline. The x-axis represents lateral migration and is given indegrees (cylindrical coordinates). The r4–r5 boundary is depicted by thered line. D and E, cartoons depicting the invasion patterns of parentalC8161 cells (fromFig. 1) andEphB6þC8161 cells. F, an overlay of the cellinvasion pattern from parental C8161 cells (green) and EphB6þ C8161cells (red), highlighting the caudal shift in cell positions at t¼ 24 hours. G,Gaussian fit to a histogram comparing cell positions of parental C8161cells and EphB6þ C8161 cells along the rostral–caudal axis at t ¼ 24hours. H, Gaussian fit to a histogram comparing total distance migrated(total displacement) between parental C8161 cells and EphB6þ C8161cells. Xc is the center of the peak and represents the average X-value.Wequals 2 times the SDof theGaussian distribution or approximately 0.849the width of the peak at half height. �, P < 0.05.

Bailey and Kulesa





migratory behaviors such as migratory capacity and direc-tionality and suggest that an understanding of these sig-nals may provide insight into the metastatic abilities oftumor cells.The establishment of nonpermissive tissue boundaries is

critical for maintaining discrete neural crest migratoryroutes. What are these cues and can we use them to obstructthe metastatic process? The neural crest provides a superbmodel system in which to study cell–microenvironmentexchanges during cell invasion.One key question was whether the microenvironment

dictated which cells would migrate, as the majority oftransplanted tumor cells remain in the neural tube 24 hoursposttransplantation. If the microenvironment was respon-sible for inducing themigratory phenotype, we would expectto observe a reduction in migratory efficiency in cells at thecore of the transplant that are insulated from direct contactwith the embryo. Likewise, if all cells responded to themicroenvironment equally, we would expect cells at theperimeter to migrate farther simply because they could morerapidly delaminate from the tumor graft (as was observed invitro). However, we were able to show that migratory cellsoriginated from the core of the transplant at nearly the samerate as cells located on the periphery and that these core cellscould be found at the leading front of invasion (Fig. 2). Thissuggests that heterogeneities among the transplanted cellsenabled some cells to recognize the proper migratory cuespresent in the microenvironment, independent of cellposition.Dynamic in vivo 2-photon time-lapse microscopy pro-

vided further insights into how the embryonic microen-vironment regulates tumor cell migration. First, weobserved that transplanted cells are compliant to signalsestablishing a dorsal midline boundary (Fig. 4). This samemidline segregation is observed during neural crest migra-tion and is thought to be induced by direct cell guidancemechanisms, including Slit-Robo signaling (28). We pos-tulate that a similar guidance mechanism segregatesmigrating tumor cells. Also, cells do not maintain strictneighbor relationships but act independently in respond-ing to microenvironmental signaling cues (Fig. 3). Weconclude that the ability to migrate is cell-autonomous,whereas specific migratory behaviors are dictated by themicroenvironment.Several guidance mechanisms are used by the embryo to

sculpt discrete neural crest migratory streams and direct cellsto precise destinations (13). Among these mechanisms, Eph/ephrin signaling has been well described in guiding neuralcrest migration in both the hindbrain and the trunk (19–21, 29). In the hindbrain, distinct combinations of Ephs andephrins are thought to help establish boundaries that sculptneural crest streams en route to their respective branchialarches.Aberrant Eph/ephrin signaling has also been reported to

mediate cancer metastasis (30). In the current study, weshow that EphB6 reexpression altered tumor cell invasiondirectionality and significantly reduced metastatic potential(Fig. 6). EphB6 has been identified as reduced or absent in

several metastatic solid tumor types, including breast cancer,lung cancer, and melanoma (31–33). As Eph/ephrin signal-ing is commonly associated with cell–cell repulsion, it istempting to speculate that the ephrin signature expressed bythe host neural crest or the underlying mesenchymal cellsinduces an inhibitory response in EphB6þC8161 cells. Thiswould act to divert the melanoma cells away from the r4migratory route. This is supported by work from Matsuokaand colleagues which demonstrates that EphB6 is repelled byhigh concentrations of the ligand ephrin-B2, which isthought to be expressed by migrating cranial neural crestcells (21, 34). However, it has also been suggested thatvarying concentrations of EphB6 may induce a switch fromrepulsion to attraction (34). Indeed, ephrin-B1 has beenshown to be expressed by non–neural crest cells lining neuralcrest migratory boundaries in the chick hindbrain, includingareas around the otic vesicle (21). Thus, it is possible thattypically repulsive ephrin-B1 signals have become attractiveto EphB6þ C8161 cells.Importantly, the reduction in metastatic potential

observed in EphB6þ cells did not likely result from amigratory phenotype, as EphB6þ C8161 cells traveled asimilar distance as parental cells in our transplant model(Fig. 5). As such, we hypothesize that EphB6 may func-tion to perturb the process of intravasation, as our CAMassay uniquely assesses intravasation potential by specif-ically identifying and quantifying tumor cells that wereable to intravasate into the CAM vasculature (Figs. 6and Fig. 7, model). Specifically, we postulate that EphB6may function to regulate the tumor cell–endothelial cellinteraction, possibly receiving a repulsive signal followingengagement with ephrins expressed on endothelial cells.Endothelial cells have been reported to express multipleEphs and ephrins, including ephrin-B2, which has beenshown to have a repulsive effect on EphB6 (34, 35). Eph/ephrin signaling has also recently been reported to play arole in the transendothelial migration of some immunecells (35, 36). Thus, we hypothesize that the endotheliumbecomes an impenetrable barrier to cells expressing aspecific Eph/ephrin signature that includes EphB6, where-as the absence of EphB6 may facilitate passage across theendothelium (Fig. 7, model).In summary, the focus of this study was to understand

how metastatic tumor cells respond to and interact withtheir microenvironment at the level of a single cell. Weshow that the embryonic neural crest microenvironmentexerts a powerful, controlling influence on the migratorybehaviors of an aggressive metastatic melanoma (C8161).This regulation is enacted both temporally and spatially bythe embryo and was most evident by the induction of aspatially controlled symmetric exit pattern of melanomacells from the neural tube. These data highlight the ideathat successful invasion (and metastasis) of tumor cellsrequires that a cell be able to interpret complex micro-environmental cues encountered along the route of inva-sion and either be guided or adapt to overcome those cuesthat are meant to disrupt or block migration. As anexample, we show that EphB6 acts specifically as a






metastasis suppressor. Because EphB6þ cells maintaintheir invasiveness and tumorigenicity, we postulate thatmetastasis suppression occurs as EphB6 disrupts intrava-sation by forcing the tumor cell to recognize a nonper-missive endothelial boundary. These findings were madepossible by studying melanoma metastasis within theaccessible embryonic neural crest microenvironment.

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

Authors' ContributionsConception and design: C.M. Bailey, P.M. KulesaDevelopment of methodology: C.M. Bailey, P.M. KulesaAcquisition of data (provided animals, acquired and managed patients, providedfacilities, etc.): C.M. Bailey, P.M. KulesaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics, compu-tational analysis): C.M. Bailey, P.M. KulesaWriting, review, and/or revision of the manuscript: C.M. Bailey, P.M. Kulesa

Administrative, technical, or material support (i.e., reporting or organizing data,constructing databases): P.M. KulesaStudy supervision: P.M. Kulesa

AcknowledgmentsThe authors thank the laboratory of Dr. Mary Hendrix for providing the cell lines

used in these experiments and for thoughtful and constructive discussion. They alsothank the laboratory of Dr. Andrew Freywald for providing the EphB6 expressionvector used for these studies.

Grant SupportC.M. Bailey thanks the NIH Ruth L. Kirschtein Postoctoral Fellowship

Program (award 5F32CA144297) for funding and partial support from theStowers Institute for Medical Research. P.M. Kulesa thanks the kind generosityof the Stowers Institute for Medical Research for support.

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

Received December 19, 2013; revised April 8, 2014; accepted May 1, 2014;published OnlineFirst May 16, 2014.

Figure 7. A model comparison of metastatic melanoma invasion patterns in the embryonic hindbrain and on the vascularized embryonic CAM. A, Threedifferent cartoons comparing metastatic melanoma invasion patterns from different rhombomeres in the hindbrain both during and after neural crestmigration. The embryonic hindbrain directs the formation of discrete neural crest migratory pathways emerging from specific hindbrain segments calledrhombomeres, listed as r1 through r6. Neural crest migration proceeds in a rostral-to-caudal progression. Neural crest invasion patterns are depicted by redarrows. Hashed red lines highlight neural crest cell–free zones that are thought to be established by the presence of inhibitory cues within themicroenvironment. Whenmetastatic melanoma cells are transplanted into the dorsal neural tube during active neural crest migration, melanoma cells invadealong neural crestmigratory pathways.Melanoma invasion patterns are regulated spatially and temporally by the developing embryo. This pattern is disruptedinmelanomacells expressingEphB6.Melanoma invasionpatterns are shown for bothparentalC8161cells (green) andEphB6þC8161cells (yellow).B,C8161melanoma cells in suspension are dropped onto the CAM of an E10 chick embryo. Cells aggregate over 48 hours to form a tumor. At this time, cells that arecapable of intravasation can be detected in the lower CAM tissue. The reexpression of EphB6 on C8161 cells abrogates metastasis, possibly by inhibitingC8161 cells from entering the CAM vasculature.


Bailey and Kulesa




References1. Spano D, Heck C, De Antonellis P, Christofori G, Zollo M. Molecular

networks that regulate cancer metastasis. Semin Cancer Biol 2012;22:234–49.

2. Condeelis J,Weissleder R. In vivo imaging in cancer. Cold Spring HarbPerspect Biol 2010;2:a003848.

3. Polacheck WJ, Zervantonakis IK, Kamm RD. Tumor cell migration incomplex microenvironments. Cell Mol Life Sci 2013;70:1335–56.

4. Le Douarin NM, Kalcheim C. The neural crest. 2nd ed. Cambridge,England: Cambridge University Press; 1999.

5. Trainor PA, Krumlauf R. Patterning the cranial neural crest: hindbrainsegmentation and Hox gene plasticity. Nat Rev Neurosci 2000;1:116–24.

6. Trainor PA, Sobieszczuk D, Wilkinson D, Krumlauf R. Signallingbetween the hindbrain and paraxial tissues dictates neural crestmigration pathways. Development (Cambridge, England) 2002;129:433–42.

7. Bailey CM, Morrison JA, Kulesa PM. Melanoma revives an embryonicmigration program to promote plasticity and invasion. Pigment CellMelanoma Res 2012;25:573–83.

8. Gupta PB, Kuperwasser C, Brunet JP, Ramaswamy S, Kuo WL, GrayJW, et al. The melanocyte differentiation program predisposes tometastasis after neoplastic transformation. Nat Genet 2005;37:1047–54.

9. NesbitM, Setaluri V, HerlynM. Biology ofmelanocytes andmelanoma.In: Balch CM, editor. Cutaneous melanoma. 3rd ed. St. Louis, MO:Quality Medical Publishing; 1998. p. xxii, 596.

10. Kulesa PM, Kasemeier-Kulesa JC, Teddy JM, Margaryan NV, SeftorEA, Seftor RE, et al. Reprogramming metastatic melanoma cells toassume a neural crest cell-like phenotype in an embryonic micro-environment. Proc Natl Acad Sci U S A 2006;103:3752–7.

11. Hendrix MJ, Seftor EA, Seftor RE, Kasemeier-Kulesa J, Kulesa PM,Postovit LM. Reprogramming metastatic tumour cells with embryonicmicroenvironments. Nat Rev 2007;7:246–55.

12. Busch C, Krochmann J, Drews U. The chick embryo as an experimen-tal system for melanoma cell invasion. PLoS ONE 2013;8:e53970.

13. Kulesa PM, Bailey CM, Cooper C, Fraser SE. In ovo live imaging ofavian embryos. Cold Spring Harb Protoc 2010. Epub June 3, 2010.

14. ArmstrongPB,Quigley JP, SidebottomE . Transepithelial invasion andintramesenchymal infiltration of the chick embryo chorioallantois bytumor cell lines. Cancer Res 1982;42:1826–37.

15. Walker JA, Kilroy GE, Xing J, Shewale J, Sinha SK, Batzer MA. HumanDNAquantitation using Alu element-based polymerase chain reaction.Anal Biochem 2003;315:122–8.

16. van der Horst EH, Leupold JH, Schubbert R, Ullrich A, Allgayer H.TaqMan-based quantification of invasive cells in the chick embryometastasis assay. Biotechniques 2004;37:940–2, 944, 946.

17. LumsdenA,SprawsonN,GrahamA.Segmental origin andmigration ofneural crest cells in the hindbrain region of the chick embryo. Devel-opment (Cambridge, England) 1991;113:1281–91.

18. Kulesa P, Ellies DL, Trainor PA. Comparative analysis of neural crestcell death, migration, and function during vertebrate embryogenesis.Dev Dyn 2004;229:14–29.

19. Krull CE, Lansford R, Gale NW, Collazo A, Marcelle C, YancopoulosGD, et al. Interactions of Eph-related receptors and ligands conferrostrocaudal pattern to trunk neural crest migration. Curr Biol 1997;7:571–80.

20. Kasemeier-Kulesa JC, Bradley R, Pasquale EB, Lefcort F, Kulesa PM.Eph/ephrins and N-cadherin coordinate to control the pattern ofsympathetic ganglia. Development (Cambridge, England) 2006;133:4839–47.

21. Mellott DO, Burke RD. Divergent roles for Eph and ephrin in aviancranial neural crest. BMC Dev Biol 2008;8:56.

22. Freywald A, Sharfe N, Roifman CM. The kinase-null EphB6 receptorundergoes transphosphorylation in a complex with EphB1. J BiolChem 2002;277:3823–8.

23. Fox BP, Kandpal RP. A paradigm shift in EPH receptor interaction:biological relevance of EPHB6 interaction with EPHA2 and EPHB2 inbreast carcinoma cell lines. Cancer Genomics Proteomics 2011;8:185–93.

24. Truitt L, Freywald T, DeCoteau J, Sharfe N, Freywald A. The EphB6receptor cooperates with c-Cbl to regulate the behavior of breastcancer cells. Cancer Res 2010;70:1141–53.

25. Kim J, Yu W, Kovalski K, Ossowski L. Requirement for specificproteases in cancer cell intravasation as revealed by a novel semi-quantitative PCR-based assay. Cell 1998;94:353–62.

26. Deryugina EI, Quigley JP. Chick embryo chorioallantoic membranemodel systems to study and visualize human tumor cell metastasis.Histochem Cell Biol 2008;130:1119–30.

27. Zijlstra A, Mellor R, Panzarella G, Aimes RT, Hooper JD, MarchenkoND, et al. A quantitative analysis of rate-limiting steps in themetastaticcascade using human-specific real-time polymerase chain reaction.Cancer Res 2002;62:7083–92.

28. Ypsilanti AR, Zagar Y, Chedotal A. Moving away from the midline: newdevelopments for Slit and Robo. Development (Cambridge, England)2010;137:1939–52.

29. Smith A, Robinson V, Patel K, Wilkinson DG. The EphA4 and EphB1receptor tyrosine kinases and ephrin-B2 ligand regulate targetedmigration of branchial neural crest cells. Curr Biol 1997;7:561–70.

30. Astin JW, Batson J, Kadir S, Charlet J, Persad RA, Gillatt D, et al.Competition amongst Eph receptors regulates contact inhibition oflocomotion and invasiveness in prostate cancer cells. Nat Cell Biol2010;12:1194–204.

31. Hafner C, Bataille F, Meyer S, Becker B, Roesch A, Landthaler M, et al.Loss of EphB6 expression in metastatic melanoma. Int J Oncol2003;23:1553–9.

32. Fox BP, Kandpal RP. Transcriptional silencing of EphB6 receptortyrosine kinase in invasive breast carcinoma cells and detection ofmethylated promoter by methylation specific PCR. Biochem BiophysRes Commun 2006;340:268–76.

33. Yu J, Bulk E, Ji P, Hascher A, Tang M, Metzger R, et al. The EPHB6receptor tyrosine kinase is a metastasis suppressor that is frequentlysilenced by promoter DNA hypermethylation in non-small cell lungcancer. Clin Cancer Res 2010;16:2275–83.

34. Matsuoka H, Obama H, Kelly ML, Matsui T, Nakamoto M. Biphasicfunctions of the kinase-defective Ephb6 receptor in cell adhesion andmigration. J Biol Chem 2005;280:29355–63.

35. Ivanov AI, Romanovsky AA. Putative dual role of ephrin-Eph receptorinteractions in inflammation. IUBMB Life 2006;58:389–94.

36. Trinidad EM, Ballesteros M, Zuloaga J, Zapata A, Alonso-ColmenarLM. An impaired transendothelial migration potential of chronic lym-phocytic leukemia (CLL) cells can be linked to ephrin-A4 expression.Blood 2009;114:5081–90.






Published OnlineFirst May 16, 2014.Mol Cancer Res Caleb M. Bailey and Paul M. Kulesa Reveal EphB6 as a Metastasis SuppressorEmbryonic Microenvironment Regulate Cell Invasion and Dynamic Interactions between Cancer Cells and the

Updated version

10.1158/1541-7786.MCR-13-0673doi:

Access the most recent version of this article at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://mcr.aacrjournals.org/content/early/2014/09/03/1541-7786.MCR-13-0673To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-13-0673

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/early/2014/09/03/1541-7786.MCR-13-0673


Dynamic Interactions between Cancer Cells and the...

Documents

Transcript of Dynamic Interactions between Cancer Cells and the...