|
|
||||||||

* Department of Pharmacology, Rush Medical College, Chicago, Illinois 60613;
Centocor, Malvern, Pennsylvania 19355
| Abstract |
|---|
|
|
|---|
Key Words: angiogenesis endothelial cell PECAM-1 (CD31) re-endothelialization tensegrity TGF-ß1
| Introduction |
|---|
|
|
|---|
Pronounced changes in cell-cell contact occur both during cell proliferation and cell migration. As cell density increases, so does cell-cell adhesion, ultimately leading to contact inhibition of growth. Cell migration requires migrating cells to sever connections with neighboring cells. Because of the association between cell-cell adhesion and cell growth and migration, we studied the expression and localization of the EC-EC adhesion molecule platelet-endothelial cell adhesion molecule-1 (PECAM-1; also called CD31), which is associated with the regulation of cell migration and angiogenesis (16), in sparse and confluent cultured EC, as well as in migrating EC.
In an in vitro model of reendothelialization (7), an injury is made in the middle of a confluent EC monolayer, inducing EC from intact areas of the monolayer to migrate into the denuded region, proliferate, and eventually repair the injury. These migrating cells sever connections with cells of the intact area of the monolayer, remain solitary for sometime, undergo morphologic changes, and form new cell-cell contacts. Our results indicate that profound changes in PECAM-1 expression takes place in EC at different stages of growth and movement. These changes occur at the protein and the steady-state mRNA level. The alterations in PECAM-1 expression seem unrelated to the proliferative ability of EC. However, the fact that cell spreading and cell-matrix interactions correlate with PECAM-1 expression prompts us to hypothesize that structural integrity of the cell, termed cellular tensegrity (8, 9), regulates PECAM-1 expression. Because PECAM-1 is involved not only in EC-EC adhesion, but also in EC-leukocyte binding, our findings may be significant in deciphering the molecular mechanisms underlying leukocyte transmigration during inflammation.
| Materials and Methods |
|---|
|
|
|---|
Generation of PECAM-1 Antibody.
Monoclonal antibody to human PECAM-1 was generated as recently described (10). Total RNA was isolated from low passage HUVEC and was used to prepare a full-length human PECAM-1 cDNA. This cDNA was used as a template in PCR amplifications to assemble a PECAM domain 1 and 2-immunoglobulin fusion protein. A PECAM DNA fragment was ligated into human immunoglobulin
4 constant region-containing expression vector such that the first two PECAM immunoglobulin-like domains were joined to an immunoglobulin
4 heavy chain in which the VH and CH1 domains had been deleted (1-2PECAM-IgG4). This construct was expressed in SP2/0 murine myeloma cells and was purified to homogeneity from culture supernatants by protein A chromatography. Mice were immunized and boosted with 1-2PECAM-IgG4 fusion protein. Spleen cells were isolated and fused with murine myeloma F/0 cells (ATCC). Clones producing anti-PECAM antibodies were selected by ELISA for their reactivity with immobilized 1-2PECAM-IgG4 and their lack of reactivity with a control IgG4 fusion protein. These clones were subcloned and single clones were isolated and purified antibodies were characterized. An antibody recognizing both human and bovine PECAM-1 was used for these studies.
Other Anti-PECAM-1 Antibodies.
In addition to the antibody generated by us, two commercially available monoclonal anti-human PECAM antibodies were usedone from Sigma (St. Louis) and one from Chemicon (Temecula, CA). Neither of them recognized bovine PECAM-1, so these were only used with HUVEC.
In Vitro Reendothelialization Model.
The in vitro reendothelialization model was done essentially by the method of Herman (7). EC were grown to confluence on two-chamber slides (2.1 cm2 growth area), and a 2-mm-wide strip of cells in the middle of the monolayer was removed by scraping with a pipet tip. Since measurement of cell migration rate was not an objective of this study, no attempt was made to keep the strip width precisely equal in each experiment. The culture was washed to remove floating cells and was returned to the incubator. Cells in the intact areas of the monolayer migrated into the denuded area from both sides and completely repaired the injury in 2 to 3 days. In general, immunostaining was performed after 24 hr of migration to determine PECAM-1 expression in migrating cells.
Immunocytochemistry.
Cells were washed twice with phosphate-buffered saline (PBS), fixed with 4% paraformaldehyde for 15 min at room temperature, and washed again with PBS. They were then treated with 0.2% Triton X-100 for permeabilization, washed again, and incubated with anti-PECAM antibody (10 µg/ml in 3% BSA-containing PBS) for 45 min at 37°C. (Permeabilization was performed to detect possible cytosolic pools of PECAM-1). Cells were then washed with PBS three times and then incubated with a FITC-labeled goat anti-mouse IgG (ICN; diluted 1:75 in 3% BSA-containing PBS) for 45 min at 37°C. After washing three times with PBS, cells were coverslipped with antifade mounting media (90% glycerin, 0.1% p-phenylenediamine, pH 9) and viewed under an Optiphot microscope.
Northern Blotting.
Total RNA was prepared from confluent (>100,000 cells/cm2) and sparse (30,00050,000 cells/cm2) EC 48 hr after plating using RNAStat (Tel-Test, Friendsville, TX) and following the manufacturer's protocol. RNA was resolved on formaldehyde-agarose gels, transferred to nitrocellulose, and hybridized to 32P-labeled bovine PECAM-1 cRNA probe prepared using a Riboprobe kit (Promega, Madison, WI) at 56° to 60°C overnight. The blot was then washed twice for 15 min in 0.1x SSC, 0.1x SDS at 68°C, and exposed 4 hr with an X-ray film with two intensifying screens.
Cell Proliferation Assay.
Cells were plated on 2.1-cm2 wells of 24-well dishes subconfluent density (about 25,00030,000 cells/cm2). The next day, unattached cells were washed off with PBS and the cell number was determined from one well (T0). Then the test substances (mitomycin C, TGFß, and SNAP) were added individually in triplicate wells per substance in regular media. After 3 days, cells were washed with PBS, trypsinized, and counted electronically in a Coulter Counter. Graphs were generated by Microsoft Excel and statistics were done with InStat (GraphPad, San Diego, CA).
Reagents and Chemicals.
The bovine PECAM-1 cDNA was a generous gift from Dr. P.A. Marsden (Toronto, Canada). All chemicals were purchased from Sigma.
| Results |
|---|
|
|
|---|
25,000 cells/cm2) BPAEC, except areas of cell-cell contact (Fig. 1B
|
Upon closer examination, it was seen that cell-cell contact was not an absolute determinant of PECAM-1 expression. At the injured edge of the monolayer where cells that have just separated from the intact area, most cells are making partial contact or no contact with other cells. Immunostaining for PECAM-1 showed that whereas a subpopulation of migrating cells lacked any PECAM-1 immunoreactivity even though they had some residual cell-cell contacts, other cells expressed PECAM-1 even at border areas where no cell-cell contact occurred (Fig. 1E
, some indicated by arrows). Thus, cell-cell contact alone was not the determinant of PECAM-1 expression. The pattern of PECAM-1 immunostaining in HUVEC by our antibody was totally identical to that generated by two other commercially available antibodies as described in ``Materials and Methods,'' indicating these results were not due to any abnormalities in the antibody.
PECAM-1 Expression Is Not Dependent on Proliferative Ability of EC.
In addition to migrating ability and degree of cell-cell contact, another difference between confluent EC and sparse or migrating EC is the ability to proliferate. Whereas confluent cells are prevented from proliferation by contact inhibition, sparse and migrating cells proliferate. To determine whether PECAM-1 expression is dependent the proliferative ability of EC, BPAEC were treated with TGFß1 (2 ng/ml, 72 hr), a well-characterized EC growth inhibitor, in another experiment, they were treated with SNAP (200 µM, 72 hr), a nitric oxide (NO) donor that inhibits BPAEC growth (11), and in another experiment, they were treated with mitomycin C (25 µg/ml for 30 min), which inhibits cell growth in general by cross-linking DNA. Cells were incubated for an additional 72 hr. Parallel proliferation assays were conducted to ensure these agents were indeed inhibiting BPAEC growth (Fig. 2A
). Immunostaining with antiPECAM-1 antibody revealed that there was very low staining in SNAP- or mitomycin-treated growth-arrested BPAEC at cell-cell borders (Fig 2B, 1![]()
and 2
) compared with growth-arrested untreated confluent cells, as shown in Figure 1, A and D
. This indicated that abrogation of proliferative ability was not inducing PECAM-1 expression. Some punctate staining was observed in both cases in the cytosol, suggesting low amounts of PECAM-1 protein is synthesized, but not transported to the cell membrane. However, TGFß1-treated EC had higher staining than SNAP- or mitomycin C-treated cells and in some areas comparable with confluent cell staining (Fig. 2
). This observation will be commented on in the ``Discussion'' section.
|
|
| Discussion |
|---|
|
|
|---|
Several differences between sparse and confluent EC have previously been reported (1416). These differences suggest that cell-cell contact, as well as proliferative ability of EC, govern many of the biochemical and functional characteristics of EC. Our findings indicate that neither cell-cell contact nor proliferative ability appears to be the sole determinant of PECAM-1 regulation, even though some contribution cannot be ruled out. In addition to cell contact and proliferative ability, however, another difference between sparse or migrating and confluent EC is cell shape and cell spreadingsparse cells are generally more spread out and migrating cells are more elongated than the polygonal cells in a confluent monolayer. The architectural integrity of the cell, termed tensegrity, controls cell shape and spreading and is maintained by mechanical forces associated with the cell such as cell-matrix adhesion, cell-cell adhesion, and the cytoskeleton. Strong evidence suggests changes in cellular tensegrity are mechanochemical regulators of cell function (8, 9). Thus, sparse and migrating EC have a different, more relaxed, tensegrity from confluent EC. In the case of TGFß1, however, two opposing forces may be regulating PECAM-1 expression; on the one hand, this cytokine inhibits cell growth, thus reducing cell-cell contact forces, but TGF-ß1 significantly increases matrix deposition by EC and thereby may strengthen cell-matrix contact forces. This may explain the observation that growth-arrested TGFß1-treated cells have higher PECAM-1 levels than growth-arrested SNAP- or mitomycin C-treated cells. Taken together, these observations suggest that PECAM-1 expression in EC is at least partially under mechanochemical regulation, with stronger cell-associated mechanical forces favoring PECAM-1 upregulation, and weaker or more relaxed forces favoring suppression. Indeed, an inverse relation has been reported between matrix regulation by TGF-ß1 and cell migration (17). Of course, additional studies are required to verify this concept.
Expression of PECAM-1 is altered during angiogenesis (3, 5), a process that requires EC migration. PECAM-1 has been suggested to be required for cell elongation, migration, and/or invasion during angiogenesis (6). Romero et al. (18) showed that PECAM-1 downregulation occurs during in vitro angiogenesis. Sheibani and Frazier (19) reported that downregulation of PECAM-1 alters EC phenotype, and Dosanjh et al. (20) have shown that reduced expression of PECAM-1 occurs in morphologically altered endothelial cells isolated from hemangiomas. Several regulators of angiogenesis such as TNF
and oxygen alter PECAM-1 levels in EC (12, 21). In addition to its adhesive role, PECAM-1 has been reported to have additional regulatory roles through acting as a reservoir for and regulator of tyrosine-phosphorylated ß-catenin (22) and inhibition of TCR-mediated signal transduction (23). Thus, the downregulation of PECAM-1 observed by us may be physiologically significant during angiogenesis and reendothelialization. PECAM-1 also mediates EC-leukocyte adhesion during leukocyte extravasation (24), an event critical to inflammation. Blocking antibodies to PECAM-1 has been shown to inhibit acute inflammation in vivo (25). Thus, our observations may facilitate understanding of certain steps in the inflammatory process as well, especially in comparing the responses of intact versus denuded endothelium.
| Acknowledgments |
|---|
| Footnotes |
|---|
1 To whom requests for reprints should be addressed at 4170 North Marine Drive, No. 5M, Chicago, IL 60613. E-mail: arcdanku{at}earthlink.net ![]()
| References |
|---|
|
|
|---|
and IFN-
. J Immunol 156:12211228, 1996.[Abstract]
This article has been cited by other articles:
![]() |
C. S. Lim, F. Alexander-Sefre, M. Allam, N. Singh, J. C. Aleong, H. Al-Rawi, and I. J. Jacobs Clinical Value of Immunohistochemically Detected Lymphovascular Space Invasion in Early Stage Cervical Carcinoma Ann. Surg. Oncol., September 1, 2008; 15(9): 2581 - 2588. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Joner, G. Nakazawa, A. V. Finn, S. C. Quee, L. Coleman, E. Acampado, P. S. Wilson, K. Skorija, Q. Cheng, X. Xu, et al. Endothelial Cell Recovery Between Comparator Polymer-Based Drug-Eluting Stents J. Am. Coll. Cardiol., July 29, 2008; 52(5): 333 - 342. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |