Cancer Research AACR Membership Frontiers in Basic Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 68, 5167, July 1, 2008. doi: 10.1158/0008-5472.CAN-08-1229
© 2008 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tremblay, P.-L.
Right arrow Articles by Auger, F. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tremblay, P.-L.
Right arrow Articles by Auger, F. A.

Cell, Tumor, and Stem Cell Biology

Mechanisms by which E-Selectin Regulates Diapedesis of Colon Cancer Cells under Flow Conditions

Pierre-Luc Tremblay1,2, Jacques Huot2 and François A. Auger1

1 Laboratoire d'Organogenèse Expérimentale, Centre Hospitalier Affilié Universitaire de Québec; and 2 Le Centre de Recherche en Cancérologie de l'Université Laval, Québec, Canada

Requests for reprints: François A. Auger, Department of Surgery, Université Laval/Laboratoire d'Organogenèse Expérimentale, Centre Hospitalier Affilié Universitaire de Québec, 1050 Chemin St-Foy, Québec, G1S 4L8, Canada. Phone: 418-682-7663; Fax: 418-682-8000; E-mail: Francois.Auger{at}chg.ulaval.ca.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 
Diapedesis, the passage of circulating tumor cells across the endothelium, is a critical determinant in most cases of metastasis. Using a laminar flow chamber and a tissue-engineered blood vessel, we found that E-selectin is required not only for the initial adhesion and rolling of circulating HT-29 colon cancer cells on the endothelium but also for their subsequent diapedesis. These processes require both the intracellular and extracellular domains of E-selectin. We also identified three distinct mechanisms by which circulating cancer cells interact with E-selectin to initiate their diapedesis: formation of a mosaic between cancer cells and endothelial cells, paracellular diapedesis at the junction of three endothelial cells, and transcellular diapedesis. We also obtained evidence indicating that E-selectin–dependent paracellular extravasation is independent of intercellular adhesion molecule and vascular cell adhesion molecule and that it requires the activation of extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase downstream of E-selectin. This is supported by the observation that the adenoviral-mediated expression of the E-selectin mutant Y603F is associated with both an inhibition of ERK and paracellular extravasation. Our study is the first to clearly establish, under dynamic and shear stress conditions, how E-selectin regulates diapedesis of circulating cancer cells. These results provide new insights in understanding the metastatic process. [Cancer Res 2008;68(13):5167–76]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 
The formation of metastasis requires the successful completion of several sequential interrelated steps by the cancer cells (1). These include detachment of cancer cells from the primary tumor and their entry into the blood circulation. Subsequently, it follows the aggregation or not of cancer cells with leukocytes or platelets, their survival in the circulation, and their arrest within the capillaries of a distant organ (2). In certain cases, cancer cells will grow locally within the vessels, but more frequently, they will cross the vessel wall in a process called diapedesis to invade the surrounding tissue. This series of events destroys most cancer cells, and the formation of metastases is an intrinsically inefficient process (3–5). Nevertheless, a small number of cancer cells (~0.01%) withstand all these steps and form metastases. The metastatic inefficiency is principally determined by the failure of solitary cells to initiate growth and also of early micrometastases to grow into macroscopic tumors (6, 7).

Interestingly, several types of cancers show an organ preference for metastasis formation. This "homing concept of metastasis" requires specific interactions between adhesion receptors on cancer cells and their counter-receptors on vascular endothelial cells (4, 8). Many studies point to the key role played by endothelial E-selectin in being involved in the adhesion and homing of colon cancer cells to the liver (9–11). E-selectin is expressed at the surface of endothelial cells after activation by inflammatory cytokines, such as interleukin-1β (IL-1β) and tumor necrosis factor-{alpha} (12). The physiologic role of E-selectin is to mediate the rolling of neutrophils on the endothelium, ultimately enabling their diapedesis at sites of inflammation (13). Several elements of evidence indicate that colon cancer cells and other types of cancer cells exploit the inflammatory system and interact with E-selectin to facilitate their adhesion and extravasation, thus promoting invasiveness (14–17). In particular, Lewis lung carcinoma cells trigger the expression of E-selectin by liver sinusoidal endothelium, increasing their metastatic potential and suggesting that E-selectin contributes to the organ specificity of metastatic colonization (18, 19).

Until now, most of the studies aimed at understanding the mechanism by which E-selectin regulates the adhesion and diapedesis of cancer cells were done in static models (10, 20). In the present study, we have used a laminar flow chamber and a tissue-engineered human blood vessel to investigate the mechanism by which E-selectin regulates the diapedesis of HT-29 cells in dynamic and shear stress conditions that mimic the in vivo situation. By using adenoviral constructs expressing wild-type (WT) and mutant forms of E-selectin in endothelial cells, we identified three distinct mechanisms by which E-selectin interact with the endothelium to regulate diapedesis of circulating cancer cells: (a) formation of mosaic between cancer cells and endothelial cells, (b) extracellular signal-regulated kinase (ERK)–dependent paracellular diapedesis at the junction of three endothelial cells, and (c) transcellular diapedesis.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosure of Potential...
 References
 
Reagents. Recombinant human IL-1β, PD98059, and SB203580 were purchased from Calbiochem. Calcein AM and Vybrant Dil cell-labeling solutions were obtained from Invitrogen-Molecular Probes. AdEasy Adenoviral Vector System was purchased from Stratagene. pLNX plasmids expressing WT E-selectin was graciously obtained from Dr. Reiner Zeisig (Max Delbrück Center, Berlin, Germany). Adenoviral vectors carrying GFP were generously given by Dr. Josée N. Lavoie (Laval University, Québec, Canada). The adenovirus expressing a dominant-negative mutant of p38{alpha} (p38AGF) was a gift from Dr. Kris Valerie (Virginia Commonwealth University), and the dominant-negative mutant of Erk was graciously given by Dr. Marie T. Filbin (Hunter College).

Antibodies. Mouse IgG{kappa} (MOPC21) and mouse IgG1 monoclonal anti-GFP were purchased from Sigma. Mouse monoclonal anti–phosphorylated ERK mitogen-activated protein kinase (MAPK) and rabbit polyclonal anti–phosphorylated p38 MAPK were purchased from Cell Signaling Technology. Rabbit polyclonal anti-ERK MAPK and rabbit polyclonal p38 MAPK were kind gifts from Drs. John Groce and Jacques Landry (Laval University, Québec, Canada), respectively. Goat anti-mouse IgG (H + L) and goat anti-rabbit IgG (H + L) both conjugated with horseradish peroxidase were purchased from Jackson ImmunoResearch. H18/7 was obtained from American Type Culture Collection. Mouse IgG1 monoclonal anti–platelet endothelial cell adhesion molecule 1 (PECAM-1) was purchased from Chemicon. Rabbit IgG polyclonal anti-mouse coupled to Alexa-488 and rabbit IgG polyclonal anti-mouse coupled to Alexa-594 were obtained from Invitrogen-Molecular Probes. Mouse IgG1 monoclonal anti–intercellular adhesion molecule-1 (ICAM-1) and mouse IgG1 monoclonal anti–vascular cell adhesion molecule-1 (VCAM-1) were purchased from R&D Systems.

Cells. Human umbilical vascular endothelial cells (HUVEC) were obtained from healthy newborns and were isolated by enzymatic digestion with 0.250 mg/mL thermolysin (Sigma) and cultivated in M199 medium supplemented with 20% FetalClone II serum (21, 22). Replicated cultures were obtained by trypsinization and were used at passages ≤5. HT-29 human colon carcinoma cells were cultivated in McCoy 5A medium supplemented with 10% FCS (HyClone) and antibiotics. Human fibroblasts were isolated from breast reduction surgery by using 0.2 IU/mL collagenase H (Roche Diagnostics). These cells were grown in DMEM supplemented with 10% FCS and antibiotics and were used between passages 3 and 7. All cultures were kept at 37°C in a humidified atmosphere containing 5% CO2.

Adenoviral vectors and gene transfer. Wild-type (WT) E-selectin was introduced in the XhoI site within the multiple cloning site region of p-Shuttle-CMV vector, whereas intracellular deletant form of E-selectin (ES-{Delta}ICD) and E-selectin mutant (ES-Y603F) were inserted within the XhoI-XbaI sites. The introduction in adenoviral vectors was done according to the Stratagene's protocol. For gene transfer, subconfluent HUVEC cultures were infected for 24 h with adenoviral vectors expressing the different forms of E-selectin or GFP at a multiplicity of infection of 25. In some conditions, adenoviruses expressing a dominant-negative mutant form of p38{alpha} or ERK were used simultaneously with the E-selectin adenoviruses. Then 6 h later, infection media were changed for fresh media. After an additional 16 h, the cells were trypsinized, plated on new gelatinized culture surface, and used in the next 24 to 72 h, depending on the experiment. More than 95% of the cells expressed the selectin or GFP proteins when they were used for treatments after transduction with those vectors.

Adhesion assays in a laminar flow chamber. HUVECs were infected with virus containing DNA for WT E-selectin, ES-Y603F, or ES-{Delta}ICD. HUVECs were also infected with GFP-containing virus as a negative control. One day postinfection, HUVECs were trypsinized and grown for an additional 24 h on gelatin-coated slides. HUVECs treated or not with 20 ng/mL IL-1β for 4 h to induce the expression of E-selectin were used as controls. The cultures were then placed in the laminar flow chamber GlycoTech under a shear stress of 1 dyne/cm2. In certain experiments, H18/7 or MOPC21 antibodies were added in the culture medium 30 min before injection of cancer cells. Meanwhile, tumor cells in suspension were labeled for 30 min with Calcein AM and washed twice with M199 medium. Then, HT-29 cells were added in the flow medium. Videos were taken directly from a TE2000 fluorescence microscope using magnification of 20x (Nikon).

Flow cytometry analysis. Cytofluorometric determination of E-selectin ICAM-1 or VCAM-1 was evaluated in HUVECs fixed in 3.7% formaldehyde. After fixation, cells were labeled at room temperature in PBS–1% bovine serum albumin (pH 7.4) with antibody against E-selectin, ICAM-1, or VCAM-1 and with a rabbit IgG polyclonal anti-mouse coupled to Alexa-594. The analysis was performed with a Becton Dickinson FACSCalibur flow cytometer.

MAPK activation. The activation of ERK and p38 MAPKs was assayed in Western blots using phosphorylated-specific antibodies.

Western blot analysis. Western blot analysis of proteins was done as previously reported (10). Quantification of the immunoreactive bands was done by densitometric scanning using NIH Image software.

Transendothelial migration assays in modified Boyden chamber. HUVECs were infected with adenoviral vectors carrying DNA for a WT E-selectin, ES-Y603F, or ES-{Delta}ICD together with or without a dominant-negative mutant form of p38{alpha} (p38AGF) or a dominant-negative form of ERK. One day later, HUVECs were trypsinized and grown until they reached confluence on a 8.0-µm pore size gelatinized polycarbonate membranes separating the two compartments of Fluoroblok (BD Bioscience) of modified Boyden chambers (10). Meanwhile, tumor cells in suspension were labeled for 30 min with Calcein-AM (1 µmol/L) and washed twice with M199 medium. Then, 150,000 HT-29 cells were added in the medium on the upper face of the membrane. After 16 h, the number of HT-29 cells that had migrated to the lower face of the filter was counted with a TE2000 fluorescence microscope using a magnification of 20x.

Transendothelial cell migration assays in a laminar flow chamber. HUVECs were infected with adenoviral vectors carrying DNA WT E-selectin, ES-Y603F, or ES-{Delta}ICD. One day later, HUVECs were trypsinized and grown for an additional 24 h on gelatinized glass slides. HUVECs treated or not with 20 ng/mL IL-1β for 4 h to induce the expression of E-selectin were used as controls. The cultures were inserted in a laminar flow chamber under a shear stress of 1 dyne/cm2. Meanwhile, HT-29 cells in suspension were labeled for 30 min with Vybrant Dil cell-labeling solution and washed twice with M199 medium before being added in the flow medium. The cultures were then submitted to time-lapse imaging, taking pictures every 2 min during 24 h using a TE2000 fluorescence microscope using 40x magnification.

Tissue-engineered human blood vessel. Tissue-engineered vascular substitutes were made by using the tissue-engineering procedures previously described (22, 23). Briefly, 104 cells/cm2 human fibroblasts were cultured in DMEM-Ham, 10% bovine FetalClone II serum (HyClone), and antibiotics. The medium was supplemented with 50 µg/mL of sodium ascorbate (Sigma) to stimulate the extracellular matrix (ECM) synthesis. After 4 wk of culture, cells formed a thick living tissue sheet, comprising cells embedded in their own ECM. The tissue sheet was then peeled off from the culture flask and wrapped around a mandrel tubular support until the formation of 12 layers of surrounding fibroblast sheets. The tubular constructs were cultivated for an additional 4-wk period, during which the medium was changed thrice a week. After this maturation period, the layers adhered firmly to each other, forming a cohesive tubular tissue. At this stage, the mandrel was removed and control HUEC endothelial cells treated with IL-1β or infected with adenoviral vectors carrying DNA for WT E-selectin, ES-Y603F, or ES-{Delta}ICD, were seeded in the lumen of the tubular structure. The vessel construct was placed 24 h in a bioreactor designed to provide both a luminal flow of culture medium and a mechanical support.

Adhesion and diapedesis of HT-29 cells in a tissue-engineered human blood vessel. The vessel constructs expressing the various forms of E-selectin were used as a model of live blood vessels, in which the endothelial cell layer was exposed or not to 20 ng/mL IL-1β for 4 h to induce the expression of E-selectin. HT-29 cells in suspension were labeled for 30 min with Vybrant Dil cell-labeling solution and washed twice with M199 medium. Then, the shear stress within the vessel was increased at 1 dyne/cm2 and HT-29 cells were added in the flow medium for 24 h. After the experiment, the vascular substitute was opened, fixed with formaldehyde at 3.7%, and examined with a Nikon Diaphot-TDM (Eclipse 2000) confocal microscope equipped with a 40x objective and the Bio-Rad MRC-1024 system (Nikon). The adhesion and diapedesis of HT-29 cells were visualized by means of Vybrant Dil cell labeling and the interendothelial junctions by using a mouse anti–PECAM-1 antibody and a rabbit IgG anti-mouse coupled to Alexa-488.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosure of Potential...
 References
 
E-selectin mediates the adhesion of colon cancer cells to endothelial cells under flow conditions. The central objective of our investigation was to uncover the mechanisms by which E-selectin regulates the diapedesis of colon cancer cells under dynamic conditions that mimic the in vivo situation. Considering that the adhesion of cancer cells to the endothelium is a prerequisite to diapedesis, we first ascertained the role of E-selectin in mediating adhesion of colon cancer cells under flow conditions using a laminar flow chamber and a tissue-engineered human blood vessel.

E-selectin mediates the adhesion of colon cancer cells to endothelial cells in a laminar flow chamber. HUVECs forming a tight monolayer on glass slides were treated or not for 4 hours with IL-1β to induce the expression of E-selectin. Then, the cultures were placed in a laminar chamber in which M199 medium circulated under a flow that gave a physiologic shear stress of 1 dyne/cm2 (24). Then, live HT-29 cells were injected in the flow system and video sequences were taken at 10 and 30 minutes. The cells attached to the endothelium were counted in >10 fields per condition. Results showed that no HT-29 cancer cells adhered to endothelial cells that did not express E-selectin, whereas they adhered in a time-dependent manner to HUVECs expressing E-selectin (Fig. 1A and Supplementary Movie S1). The adhesion was E-selectin–dependent because an inhibition was observed when the endothelial cells were treated with the anti–E-selectin antibody H18/7 that binds to the extracellular domain (ECD) of E-selectin.


Figure 1
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. E-selectin mediates the adhesion of colon cancer cells to endothelial cells in a laminar flow chamber. A, HUVECs were stimulated or not (Control) with 20 ng/mL IL-1β for 4 h. Thereafter, they were treated or not with E-selectin antibody (H18/7) or with a control irrelevant antibody (MOPC21) for 30 min. B, in some conditions, HUVECs were infected only with adenoviral vectors containing GFP as controls or a WT form human E-selectin (ES-WT) or a truncated form of E-selectin deleted from its cytoplasmic domain (ES-{Delta}ICD). After treatment, the cultures were placed in a laminar chamber under a shear stress of 1 dyne/cm2. Then, live HT-29 cells were injected in the flow system and video sequences were taken at 10 and 30 min. The cells attached to the endothelium were counted in >10 fields per condition. C and D, the endothelial cells were trypsinized and fixed in 3.7% paraformaldehyde and labeled with antibody against ICAM-1 (C) or VCAM-1 (D). Representative FACS analysis. *, P < 0.01 (Student's t test using the IL-1β condition as reference).

 
To further highlight the role of E-selectin and to eliminate the potential influence played by other adhesion receptors, whose expression is coinduced with E-selectin by IL-1β, we performed experiments using adenoviral-mediated expression of E-selectin (Fig. 1B). More specifically, adenoviral vectors containing GFP as controls, a WT form of human E-selectin, or a truncated form of E-selectin deleted from its cytoplasmic domain (ES-{Delta}ICD) were introduced in HUVEC. After infection, endothelial cells were cultivated as monolayers, before being inserted in the flow chamber. Then, HT-29 cells were added in the flow medium. After 10 minutes, the adhesion of HT-29 cells to HUVECs expressing WT E-selectin and ES-{Delta}ICD was the same and comparable with that observed in cells treated with IL-1β. After 30 minutes, the adhesion of HT-29 cells to HUVECs expressing WT E-selectin or treated with IL-1β further increased but not in cells that expressed the ES-{Delta}ICD. Interestingly, this was associated with a very low level of expression of ICAM-1 or VCAM-1, as observed by fluorescence-activated cell sorting (FACS) analysis (Fig. 1C and D). This indicates that the adhesion of colon cancer cells in circulation relies mainly on E-selectin, involves its cytoplasmic domain, and can be independent of ICAM-1 and VCAM-1.

E-selectin mediates the adhesion of colon cancer cells to endothelial cells in engineered human blood vessels. One of the best vascular substitute created to date to mimic the in vivo situation is the model produced by self-assembly (22, 25, 26). It displays excellent mechanical and contractile properties and a well-defined three-dimensional organization with an abundant ECM (22, 23).

We thus decided to use such a vessel to further study diapedesis of cancer cells under more physiologic conditions, including a physiologic shear stress of 1 dyne/cm2. HT-29 cells, previously marked with Vybrant red, were injected in the medium that circulated inside engineered vessels made with endothelial cells expressing or not WT E-selectin and ES-{Delta}ICD or treated with IL-1β. We found that the cancer cells bound maximally to endothelium stimulated by IL-1β or that expressed WT E-selectin (Fig. 2A and B ). No cells adhered to the endothelium that did not express E-selectin, and the adhesion was reduced by 3-fold in endothelium expressing ES-{Delta}ICD.


Figure 2
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. E-selectin mediates the adhesion of colon cancer cells to endothelial cells in engineered human blood vessels. The vascular substitutes were made by using the tissue-engineering procedures described in Materials and Methods. Engineering blood vessels expressing various forms of E-selectin (ES-WT, ES-{Delta}ICD) were used, as well as vessels in which the endothelial cell layer was stimulated or not (Control) with 20 ng/mL IL-1β for 4 h. Then, live HT-29 cells labeled with Vybrant Dil cell-labeling solution (red) were injected in the flow medium and shear stress was increased at 1 dyne/cm2 for 24 h. After the experiment, the vascular substitutes were opened and fixed with formaldehyde at 3.7%. A, pictures were taken, and HT-29 cells attached to the endothelium were counted in >5 fields per condition. B, HUVECs were also detected by using a mouse anti–PECAM-1 antibody and a rabbit IgG anti-mouse coupled to Alexa-488 (green). Pictures were taken with a confocal microscope equipped with a 40x objective. Representative fields are shown. Box #1, cancer cell that has fully crossed the endothelium (arrow); box #2, squeezed cancer cell between endothelial cells (arrow). *, P < 0.01 (Student's t test using the ES-WT condition as reference).

 
Overall, these findings are the first firm support indicating that E-selectin mediates the adhesion of cancer cells in flow conditions and that the process requires both the extracellular and cytoplasmic domains of E-selectin.

Activation of E-selectin by the adhesion of circulating colon cancer cells triggers the activation of ERK and p38 MAPKs in endothelial cells. In a previous study, we reported that activation of ERK and p38 MAPKs were important regulators of both transendothelial permeability and cancer cell migration in static conditions (10). We thus investigated next whether the ERK and p38 pathways are activated in response to the adhesion of HT-29 cells that circulate over a layer of endothelial cells that express E-selectin. Naive monolayers of HUVECs treated with IL-1β to induce maximal expression of E-selectin were exposed to a shear stress of 1 dyne/cm2 in a laminar flow chamber. Thereafter, to induce the activation of the MAPKs, HT-29 cells previously fixed with 2% paraformaldehyde were injected in the flow laminar chamber. Then, the activation of endothelial p38 and ERK was determined after increasing time intervals after the addition of HT-29. Results showed that both kinases were activated by HT-29 cells and that the kinetics of activation followed a typical bell-shaped curve with a peak of activation of 4-fold at 10 minutes of exposure (Fig. 3A and B ). Intriguingly, at this time interval, the flow itself induces 1.3-fold and 1.6-fold increases in the activity of ERK and p38, respectively. We have brought the flow-mediated increase in activation back to 1 to evaluate the ratio of ERK and p38 activation by the cancer cells (Control).


Figure 3
View larger version (43K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Activation of E-selectin by the adhesion of circulating colon cancer cells triggers the activation of ERK and p38 MAPKs in endothelial cells. A and B, HUVECs were stimulated or not (Control) with 20 ng/mL IL-1β for 4 h. C and D, HUVECs were untreated (Control) or stimulated with 20 ng/mL IL-1β for 4 h or were infected with adenoviral vectors containing GFP as controls or a WT form of human E-selectin (ES-WT) or an E-selectin mutant (ES-Y603F) or a truncated form of E-selectin deleted from its cytoplasmic domain (ES-{Delta}ICD). After treatment, all the cultures were placed in a laminar chamber under a shear stress of 1 dyne/cm2. Then, HT-29 cells fixed for 20 min with paraformaldehyde 2% were injected in the flow system for increasing periods of time (A and B) or for 10 min (C and D). After treatments, ERK (A and C) and p38 (B and D) activation was determined in Western blot using phosphorylated-specific antibodies. In each condition, control corresponds to a 10-min exposure to flow only. The activation of ERK and p38 by flow was maximally calculated as 1.3 and 1.6, respectively, and brought back to 1 to evaluate the ratio of ERK and p38 activation by the cancer cells. Columns, mean of three separate experiments; bars, SD. Representative autoradiograms are shown. *, P < 0.01 (Student's t test using the ES-WT condition as reference).

 
We next verified the direct involvement of E-selectin in the activation of ERK and p38. As above, we eliminated the contribution of other adhesion receptors induced by IL-1β using adenoviral-mediated expression of E-selectin. Endothelial cells expressing WT E-selectin, ES-{Delta}ICD, or ES-Y603F mutant were cultivated as monolayers before being inserted in the flow chamber and exposed to circulating HT-29 cells. Endothelial cells stimulated with IL-1β were used as positive controls. Results showed that the addition of HT-29 cells activated, at 10 min, ERK and p38 pathways in endothelial cells expressing WT E-selectin or that were treated by IL-1β. In contrast, the activation of both p38 and ERK was abolished in cells that expressed the ES-{Delta}ICD. Moreover, consistent with the fact that phosphorylation of Y603 within E-selectin is required to activate ERK (27), we found that the activation of ERK, but not that of p38, was hampered in cells expressing ES-Y603F (Fig. 3C and D). As expected, the activation/phosphorylation of ERK and p38 by adhering cancer cells was inhibited by PD98059 and SB203580, respectively, or with appropriate dominant-negative forms of p38{alpha} (p38AGF) or ERK (Fig. 4A and B ). Overall, this set of results are the first to directly suggest that activation of E-selectin by cancer cells circulating in the blood is signaling to p38 and ERK MAPKs and that activation of the latter required the phosphorylation of Y603 within E-selectin. They further suggest that the activation of E-selectin alone was sufficient to activate ERK and p38 and that it did not require the coexpression of other inflammation-sensitive adhesion receptors.


Figure 4
View larger version (33K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Activation of E-selectin by the adhesion of circulating colon cancer cells is impaired by the specific inhibition of ERK and p38 MAPKs in endothelial cells. A and B, HUVECs were left untreated (Control), were stimulated with 20 ng/mL IL-1β for 4 h, or were infected with adenoviruses containing WT E-selectin (ES-WT). To inhibit ERK (A) or p38 (B), HUVEC were coinfected with adenoviruses containing a mutant form of ERK (ERK–/–) or of p38{alpha} (p38–/–) or were treated 45 min with PD98059 (PD; 50 µmol/L), SB203580 (SB; 5 µmol/L), or DMSO. Thereafter, all the cultures were placed in a laminar chamber under a shear stress of 1 dyne/cm2. Then, HT-29 cells fixed for 20 min with paraformaldehyde 2% were injected in the flow system for 10 min. After treatments, ERK and p38 activation was determined in Western blot using phosphorylated-specific antibodies. In each condition, Control corresponds to a 10-min exposure to flow only. Columns, mean of three separate experiments; bars, SD. Representative autoradiograms are shown. *, P < 0.01 (Student's t test using the ES-WT condition as reference).

 
E-selectin mediates diapedesis of HT-29 colon cancer cells by different mechanisms. In Boyden chambers, the transendothelial migration of HT-29 cells was increased by 2.8-fold when HT-29 cells were added to a layer of HUVEC that expressed E-selectin after IL-1β stimulation or that were transduced with the adenoviral vector expressing WT E-selectin (Fig. 5A ). This represents only a minor percentage (~0.1%) of cancer cells that successfully complete diapedesis in response to E-selectin. Interestingly, this increase was abrogated in cells that expressed ES-{Delta}ICD and was reduced in cells that expressed the Y603F mutant, being in agreement with our previous findings that p38 and ERK are required for E-selectin–mediated transendothelial migration of colon cancer cells (10). Along these lines, we further found that the inhibition of ERK and p38 with PD098059 or SB203580, respectively, or with appropriate dominant-negative forms of these kinases was associated with an inhibition of E-selectin–mediated transendothelial migration (Fig. 5B).


Figure 5
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. E-selectin mediates diapedesis of HT-29 colon cancer cells via ERK and p38 in static condition. A and B, HUVECs were grown to confluence on a 8.0-µm pore size gelatinized polycarbonate membrane separating the two compartments of a 6.5-mm Fluoroblok chamber. HUVECs were treated or not (Control) with 20 ng/mL IL-1β for 4 h. In some conditions, HUVECs were infected with adenoviral vectors containing GFP or E-selectin (ES-WT, ES-Y603F, ES-{Delta}ICD) together with or without adenoviruses containing a mutant form of p38{alpha} (p38–/–) or ERK (ERK–/–) before experiment. In other conditions, HUVECs were treated with or without PD09859 (50 µmol/L), SB203580 (5 µmol/L), or DMSO for 45 min. In each case, the culture media in the migration chamber were changed for fresh media before the addition of HT-29 cells labeled with Calcein-AM on the upper surface of the membrane for an additional 16 h. Cells that have crossed the opaque membranes were counted in a fluorescence microscope. *, P < 0.01 (Student's t test using the ES-WT condition as reference).

 
Given that the use of Boyden chambers allowed to evaluate migration in static condition, we attempted next to characterize the mechanisms by which E-selectin mediates diapedesis in flow conditions using time-lapse microscopy. Endothelial cells expressing the various E-selectin constructs were cultivated on gelatinized supports and placed in the flow laminar chamber. Then, live HT-29 cells marked with Vybrant were introduced into the circulating medium and images were taken every 2 minutes during 24 hours. As reported in Figs. 1 and 2, we found that E-selectin mediates the initial adhesion and rolling of circulating cancer cells. Moreover, our results revealed that most of the cells that adhered to the endothelium penetrated within the endothelial layer to initiate diapedesis. Intriguingly, ~75% of these cancer cells remained inserted within the layer, thus forming a chimeric mosaic layer that incorporated both cell types (Fig. 6A and B and Supplementary Movie S2A). This is especially noticeable after staining of endothelial cell junction with PECAM-1 in the tissue-engineered blood vessel (Fig. 2B). These "squeezed" cancer cells did not complete diapedesis even after 24 hours. The number of cells that formed such a mosaic was increased in HUVECs expressing ES-{Delta}ICD, which suggested that the intracellular domain (ICD) of E-selectin was required to complete diapedesis.


Figure 6
View larger version (50K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. E-selectin mediates diapedesis of HT-29 colon cancer cells in dynamic condition. HUVECs were infected with adenoviral vectors carrying DNA WT E-selectin (ES-WT), ES-Y603F, or ES-{Delta}ICD. HUVECs treated with 20 ng/mL IL-1β for 4 h were used as positive control. Then, the cultures were placed in a laminar flow chamber under a shear stress of 1 dyne/cm2. Meanwhile, HT-29 cells in suspension were labeled with Vybrant Dil cell-labeling solution (red) and added in the flow medium. A, pictures were taken every 2 min during 24 h to examine the three processes associated with diapedesis: mosaic, paracellular migration (PM), and transcellular (TM). B, a graph shows the percentage of cancer cells undergoing the three different diapedesis mechanisms for each condition. In this graph, the sum of all diapedesis processes for each condition is 100%. C, fused cancer cells/endothelial cells that detached and reattached after transcellular penetration of cancer cells in endothelial cell. Arrows, point out the cells of interest. Scale bars, 40 µm (A and C). *, P < 0.01 (Student's t test using the ES-WT condition as reference).

 
In fact, only 25% of the cancer cells that penetrated within the endothelial layer completed diapedesis and extravasated. This occurred either through paracellular or transcellular ways. The paracellular diapedesis took place at the connecting points of three endothelial cells, and it ended with the full migration of the cancer cells under the endothelial layer (Fig. 6A and B). This is dynamically shown in Supplementary Movie S2B. Here, again, this process was particularly evident in endothelial cells stained with PECAM-1 in the tissue-engineered blood vessel (Fig. 2B). Moreover, the fact that cancer cells go underneath the endothelium is still further highlighted by confocal microscopic analysis performed on a Calcein AM–labeled endothelial cell layer expressing E-selectin. This is illustrated in Supplementary Fig. S3 that shows that Vybrant Dil–labeled HT-29 cells penetrated, but not crossed, the layer (Supplementary Fig. S3D–F), whereas other HT-29 cells appeared underneath the endothelial cells being stained yellow-orange in merged staining (Supplementary Fig. S3F). From this figure, one can also observe that the endothelial cell layer remains sufficiently tight during the course of the flow experiment (Supplementary Fig. S3B and C).

Interestingly, the process of diapedesis was abrogated in cells expressing ES-Y603F mutant, suggesting that it requires signaling from E-selectin to ERK. On the other hand, few cancer cells penetrated inside individual endothelial cells (Fig. 6A and B and Supplementary Movie S2C). This transcellular process began at least 14 hours after cancer cell adhesion, and it did not occur at endothelial cell junctions. The cancer cells were firmly attached to endothelial cells, and they induced endothelial cell retraction and blebbing. They were then engulfed in large vacuoles and transported within and through the endothelial cells that concomitantly lost their contact with the ECM. Intriguingly, most of the time, the fused cancer cells/endothelial cells detached from the matrix being released into the flow medium (Fig. 6C and Supplementary Movie S2D). Only a small number of them crossed the layer at the site of endothelial cell/ECM loosening (data not shown). Of note, some circulating fused cancer/endothelial cells could bind again to the endothelium after their detachment (Fig. 6C and Supplementary Movie S2E).

The sum of these results highlights for the first time that the mechanisms by which E-selectin may mediate diapedesis of cancer cells present in circulation and that ERK activation in endothelial cells is required for the paracellular type of diapedesis.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosure of Potential...
 References
 
Adhesion of colon cancer to E-selectin is a key determinant of metastasis (9, 28). This process relies on the induction of a bidirectional signaling by which the binding of E-selectin to its counter-receptor, DR3, on cancer cells induces reverse signals in the cancer cells that increase their intrinsic motile and survival potentials and a forward signaling in endothelial cells that enhances interendothelial permeability (8–10). In our present work, we have used a laminar flow chamber and a tissue-engineered blood vessel to further assay the mechanisms by which E-selectin regulates the diapedesis of colon cancer cells under shear stress conditions that mimic the in vivo situation. Using these two dynamic models, we showed for the first time that three different mechanisms are associated with E-selectin–mediated diapedesis: (a) formation of a mosaic chimeric tissue between cancer cells and endothelial cells, (b) paracellular diapedesis at the junctions of three endothelial cells, and (c) transcellular diapedesis. By means of adenoviral constructs expressing WT and mutant forms of E-selectin in endothelial cells, we showed that paracellular emigration required activation of ERK and that E-selectin may mediate diapedesis independently of ICAM-1 and VCAM-1. However, this does not exclude that other adhesion receptors are also involved, as it was shown for neutrophils (29).

It has been previously reported that extravasation of circulating cancer cells to endothelial cells involves two mechanisms characterized by radial spreading that follows diapedesis and axial spreading where cancer cells form a chimeric mosaic-like structure with endothelial cells (30). One of the major contributions of our study was to highlight for the first time the key role played by E-selectin as a regulator of both processes. This is supported by our findings that E-selectin mediated the initial adhesion and rolling of cancer cells on the endothelium in both a laminar flow chamber and a tissue-engineered blood vessel. Indeed, no cancer cells adhered to endothelial cells that did not express E-selectin, and the adhesion of cancer cells to endothelial cells expressing E-selectin was markedly reduced after its exposure to an E-selectin neutralizing antibody. Moreover, we found that once HT-29 cells have adhered to the endothelium via E-selectin, then most of them initiated the process of diapedesis. However, the majority of them (~75%) did not complete the diapedesis process even after 24 hours. They remained axially squeezed between two endothelial cells forming a mosaic. The inability of cancer cells to complete diapedesis can be explained by the presence of a rudimentary form of tight junctions that act as a barrier (31, 32). However, given that many cells complete the processes, it is unlikely that HT-29 cells that remain between endothelial cells do not traverse the layer because they merely have no space under it. Then, our results suggest that a subpopulation of cancer cells can initiate diapedesis but cannot complete the process. It is possible that axially spreading cancer cells may proliferate locally and partially account for metastases that develop inside a vessel without extravasating (2). In fact, only a small percentage (~15%) of the cancer cells undergo radial spreading and cross the endothelium to reach a subendothelial location. These cells cross the endothelium predominantly at the junction of three adjacent endothelial cells via paracellular extravasation, as initially reported for neutrophil emigration (31). Incidentally, it is probable that the cancer cells that manage to cross the endothelial layer in Boyden chambers use this type of paracellular mechanism. This is supported by the fact that the proportion of cancer cells that fully migrated through the endothelium under flow was similar to that of cancer cells counted under the endothelial monolayer in the Boyden chamber assay. These results are consistent with previous findings showing that, after cell adhesion, shear stress does not have any significant effect on radial spreading (30).

Interestingly, we obtained the first evidence indicating that cancer cells, after their adhesion to E-selectin, may further migrate across the endothelium via their passage through individual endothelial cells (~10%). This process has already been described for neutrophils, but not for cancer cells (33). By analogy with neutrophils, one can speculate that this transcellular type of diapedesis may occur via specialized structures called vesiculo-vacuolar organelles (VVO). The VVOs are a large collection of vesicules and vacuoles unequally distributed in endothelial cells (34, 35). They can fuse to form larger VVOs that will allow extravasation of cells (36). Along these lines, we found endothelial cell vesicules containing HT-29 cells that have presumably been phagocyted by the endothelial cells, enabling their transcellular migration. Interestingly, it seemed that transcellular diapedesis was fatal for several endothelial cells. This might be caused by cancer cell–induced anoikis of endothelial cells after disruption of the interendothelial cell junctions and the endothelial cell interactions with the ECM (37, 38). Of note, >70% of the cancer cells phagocyted by endothelial cells were released into the circulating media. These cancer cell–endothelial cell complexes may possibly contribute to the pool of mature endothelial cells that are found in the blood of cancer patients (39). More interestingly, some fused cells could bind again to endothelial cells, downstream of the detachment area. This result may be a new process for cancer cells spreading.

On the other hand, the results of our study indicated that both ECD and ICD of E-selectin were required for the adhesion and diapedesis of the cancer cells on and across the endothelium. The requirement for the ECD is supported by the fact that both processes are impaired by a neutralizing antibody directed against the ECD of E-selectin. The requirement for the ICD is supported by the observations that the increase in adhesion of HT-29 cells at 30 minutes is impaired in HUVECs expressing ES-{Delta}ICD. This requirement probably resulted from impairment in the formation of the E-selectin clusters appropriately connected with the cytoskeleton. Intriguingly, we found that the mosaic type of diapedesis was increased by the expression of the ES-{Delta}ICD, which may indicate that the ICD is required for complete diapedesis because mosaic formation is associated with the inability of the cancer cells to complete this step. Accordingly, the two other steps of full diapedesis were decreased in the cells that express ES-{Delta}ICD. Thus, the absence of the intercellular portion could reduce the capacity of cancer cells to bind firmly to the endothelium and to complete diapedesis.

In a previous study, we clearly showed in static conditions and Boyden chambers that activation of ERK and p38 MAPKs, downstream of E-selectin, are required for transendothelial cell migration. From the present study, one may deduce that the ICD of E-selectin is also required for the activation of MAPK and for diapedesis in dynamic conditions (10). This is supported by the observation that the activation of both kinases is totally abrogated in cells that express the ES-{Delta}ICD. Moreover, transendothelial migration of HT-29 cells is impaired after inhibition of ERK and p38 by chemical and genetic inhibition. The involvement of ERK is further supported by the observation that the ES-Y603F mutant inhibits the activation of ERK, as previously reported (27). In contrast, the activation of p38 does not rely on Y603 and thereby is connected to another tyrosine within E-selectin. These findings are important because they constitute the first direct evidence, suggesting that E-selectin mediates the activation of ERK and p38 in flow conditions and that their activation is required for diapedesis of cancer cells.

In summary, using flow and shear stress conditions that mimic the in vivo situation encountered by cancer cells in blood circulation, we have identified three distinct mechanisms by which E-selectin regulates diapedesis of circulating cancer cells: ERK-dependent paracellular diapedesis at the junction of three endothelial cells, transcellular diapedesis, and incomplete diapedesis, characterized by the formation of a mosaic between cancer cells and endothelial cells.


   Disclosure of Potential Conflicts of Interest
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
References
 
The authors declare no competing financial interests.


   Acknowledgments
 
Grant support: Canadian Research Society, Inc., Cancer Research Society, and Canadian Institutes of Health Research. P-L. Tremblay owns a studentship from Fonds de Recherche en Santé du Québec.

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

We thank Andrée Poirier, Kathleen Baker, Jean-François Oligny, Cindy Perron, Julie Guérard, Vicky Gagnon, Alexandre Deschambault, Mickael Morin, and Marie-Christine Fiola for the experimental support and François Houle and Dr. Dan Lacroix for reading the manuscript.


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

Received 8/15/07. Revised 4/ 1/08. Accepted 4/ 5/08.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosure of Potential...
 References
 

  1. Ribatti D, Mangialardi G, Vacca A. Stephen Paget and the "seed and soil" theory of metastatic dissemination. Clin Exp Med 2006;6:145–9.[CrossRef][Medline]
  2. Chambers AF, Naumov GN, Varghese HJ, et al. Critical steps in hematogenous metastasis: an overview. Surg Oncol Clin N Am 2001;10:243–55.[Medline]
  3. Mehlen P, Puisieux A. Metastasis: a question of life or death. Nat Rev Cancer 2006;6:449–58.[CrossRef][Medline]
  4. Haier J, Nicolson GL. Tumor cell adhesion under hydrodynamic conditions of fluid flow. APMIS 2001;109:241–62.[CrossRef][Medline]
  5. Weiss L. Metastatic inefficiency. Adv Cancer Res 1990;54:159–211.[Medline]
  6. Cameron MD, Schmidt EE, Kerkvliet N, et al. Temporal progression of metastasis in lung: cell survival, dormancy, and location dependence of metastatic inefficiency. Cancer Res 2000;60:2541–6.[Abstract/Free Full Text]
  7. Luzzi KJ, MacDonald IC, Schmidt EE, et al. Multistep nature of metastatic inefficiency: dormancy of solitary cells after successful extravasation and limited survival of early micrometastases. Am J Pathol 1998;153:865–73.[Abstract/Free Full Text]
  8. Gout S, Morin C, Houle F, Huot J. Death receptor-3, a new E-selectin counter-receptor that confers migration and survival advantages to colon carcinoma cells by triggering p38 and ERK MAPK activation. Cancer Res 2006;66:9117–24.[Abstract/Free Full Text]
  9. Laferriere J, Houle F, Taher MM, Valerie K, Huot J. Transendothelial migration of colon carcinoma cells requires expression of E-selectin by endothelial cells and activation of stress-activated protein kinase-2 (SAPK2/p38) in the tumor cells. J Biol Chem 2001;276:33762–72.[Abstract/Free Full Text]
  10. Tremblay PL, Auger FA, Huot J. Regulation of transendothelial migration of colon cancer cells by E-selectin-mediated activation of p38 and ERK MAP kinases. Oncogene 2006;25:6563–73.[CrossRef][Medline]
  11. Auguste P, Fallavollita L, Wang N, et al. The host inflammatory response promotes liver metastasis by increasing tumor cell arrest and extravasation. Am J Pathol 2007;170:1781–92.[Abstract/Free Full Text]
  12. Bevilacqua MP, Pober JS, Mendrick DL, Cotran RS, Gimbrone MA, Jr. Identification of an inducible endothelial-leukocyte adhesion molecule. Proc Natl Acad Sci U S A 1987;84:9238–42.[Abstract/Free Full Text]
  13. Kansas GS. Selectins and their ligands: current concepts and controversies. Blood 1996;88:3259–87.[Free Full Text]
  14. Brodt P, Fallavollita L, Bresalier RS, et al. Liver endothelial E-selectin mediates carcinoma cell adhesion and promotes liver metastasis. Int J Cancer 1997;71:612–9.[CrossRef][Medline]
  15. Dejana E, Martin-Padura I, Lauri D, et al. Endothelial leukocyte adhesion molecule-1-dependent adhesion of colon carcinoma cells to vascular endothelium is inhibited by an antibody to Lewis fucosylated type I carbohydrate chain. Lab Invest 1992;66:324–30.[Medline]
  16. Krause T, Turner GA. Are selectins involved in metastasis? Clin Exp Metastasis 1999;17:183–92.[CrossRef][Medline]
  17. Yamada N, Chung YS, Sawada T, Okuno M,Sowa M. Role of SPan-1 antigen in adhesion of human colon cancer cells to vascular endothelium. Dig Dis Sci 1995;40:1005–12.[CrossRef][Medline]
  18. Khatib AM, Kontogiannea M, Fallavollita L, et al. Rapid induction of cytokine and E-selectin expression in the liver in response to metastatic tumor cells. Cancer Res 1999;59:1356–61.[Abstract/Free Full Text]
  19. Khatib AM, Auguste P, Fallavollita L, et al. Characterization of the host proinflammatory response to tumor cells during the initial stages of liver metastasis. Am J Pathol 2005;167:749–59.[Abstract/Free Full Text]
  20. Hu Y, Kiely JM, Szente BE, Rosenzweig A, Gimbrone MA, Jr. E-selectin-dependent signaling via the mitogen-activated protein kinase pathway in vascular endothelial cells. J Immunol 2000;165:2142–8.[Abstract/Free Full Text]
  21. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest 1973;52:2745–56.[Medline]
  22. L'Heureux N, Paquet S, Labbe R, Germain L, Auger FA. A completely biological tissue-engineered human blood vessel. FASEB J 1998;12:47–56.[Abstract/Free Full Text]
  23. Laflamme K, Roberge CJ, Labonte J, et al. Tissue-engineered human vascular media with a functional endothelin system. Circulation 2005;111:459–64.[Abstract/Free Full Text]
  24. Jones DA, Smith CW, McIntire LV. Leucocyte adhesion under flow conditions: principles important in tissue engineering. Biomaterials 1996;17:337–47.[CrossRef][Medline]
  25. Isenberg BC, Williams C, Tranquillo RT. Small-diameter artificial arteries engineered in vitro. Circ Res 2006;98:25–35.[Abstract/Free Full Text]
  26. Remy-Zolghadri M, Laganiere J, Oligny JF, Germain L, Auger FA. Endothelium properties of a tissue-engineered blood vessel for small-diameter vascular reconstruction. J Vasc Surg 2004;39:613–20.[CrossRef][Medline]
  27. Hu Y, Szente B, Kiely JM, Gimbrone MA, Jr. Molecular events in transmembrane signaling via E-selectin. SHP2 association, adaptor protein complex formation and ERK1/2 activation. J Biol Chem 2001;276:48549–53.[Abstract/Free Full Text]
  28. Khatib AM, Fallavollita L, Wancewicz EV, Monia BP, Brodt P. Inhibition of hepatic endothelial E-selectin expression by C-raf antisense oligonucleotides blocks colorectal carcinoma liver metastasis. Cancer Res 2002;62:5393–8.[Abstract/Free Full Text]
  29. Dejana E. The transcellular railway: insights into leukocyte diapedesis. Nat Cell Biol 2006;8:105–7.[CrossRef][Medline]
  30. Chotard-Ghodsnia R, Haddad O, Leyrat A, et al. Morphological analysis of tumor cell/endothelial cell interactions under shear flow. J Biomech 2006;21:21.[CrossRef]
  31. Burns AR, Walker DC, Brown ES, et al. Neutrophil transendothelial migration is independent of tight junctions and occurs preferentially at tricellular corners. J Immunol 1997;159:2893–903.[Abstract]
  32. Burns AR, Bowden RA, MacDonell SD, et al. Analysis of tight junctions during neutrophil transendothelial migration. J Cell Sci 2000;113:45–57.[Abstract]
  33. Feng D, Nagy JA, Dvorak HF, Dvorak AM. Ultrastructural studies define soluble macromolecular, particulate, and cellular transendothelial cell pathways in venules, lymphatic vessels, and tumor-associated microvessels in man and animals. Microsc Res Tech 2002;57:289–326.[CrossRef][Medline]
  34. Kohn S, Nagy JA, Dvorak HF, Dvorak AM. Pathways of macromolecular tracer transport across venules and small veins. Structural basis for the hyperpermeability of tumor blood vessels. Lab Invest 1992;67:596–607.[Medline]
  35. Feng D, Nagy JA, Pyne K, et al. Pathways of macromolecular extravasation across microvascular endothelium in response to VPF/VEGF and other vasoactive mediators. Microcirculation 1999;6:23–44.[Medline]
  36. Feng D, Nagy JA, Pyne K, Dvorak HF, Dvorak AM. Neutrophils emigrate from venules by a transendothelial cell pathway in response to FMLP. J Exp Med 1998;187:903–15.[Abstract/Free Full Text]
  37. Heyder C, Gloria-Maercker E, Entschladen F, et al. Realtime visualization of tumor cell/endothelial cell interactions during transmigration across the endothelial barrier. J Cancer Res Clin Oncol 2002;128:533–8.[CrossRef][Medline]
  38. Kebers F, Lewalle JM, Desreux J, et al. Induction of endothelial cell apoptosis by solid tumor cells. Exp Cell Res 1998;240:197–205.[CrossRef][Medline]
  39. Bertolini F, Shaked Y, Mancuso P, Kerbel RS. The multifaceted circulating endothelial cell in cancer: towards marker and target identification. Nat Rev Cancer 2006;6:835–45.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tremblay, P.-L.
Right arrow Articles by Auger, F. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tremblay, P.-L.
Right arrow Articles by Auger, F. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online