FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by ILAN, N.
Right arrow Articles by MADRI, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by ILAN, N.
Right arrow Articles by MADRI, J. A.
(The FASEB Journal. 2001;15:362-372.)
© 2001 FASEB

PECAM-1 shedding during apoptosis generates a membrane-anchored truncated molecule with unique signaling characteristics

NETA ILAN, AMIR MOHSENIN, LAWRENCE CHEUNG and JOSEPH A. MADRI1

Department of Pathology, Yale University School of Medicine, New Haven Connecticut 06510, USA

1Correspondence: Department of Pathology, Yale University School of Medicine, 310 Cedar St., P.O. Box 208023, New Haven, CT 06520-8023, USA. E-mail: joseph.madri{at}yale.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Shedding of cell surface molecules, including growth factor receptors, provides a mechanism by which cells regulate signal transduction events. Here we show that platelet-endothelial cell adhesion molecule (PECAM)-1 is shed from the endothelial cell surface during apoptosis and accumulates in the culture medium as a ~100 kDa soluble protein. The cleavage mediating the shedding is matrix metalloproteinase (MMP) dependent, as GM6001, a broad-spectrum MMP inhibitor, inhibits PECAM-1 accumulation in the culture medium in a dose-responsive manner. In addition to the 100 kDa soluble fragment, PECAM-1 cleavage generates the formation of a truncated (Tr.) ~28 kDa molecule, composed of the transmembrane and the cytoplasmic PECAM-1 domains. Transfections of the full-length (Fl) and the Tr. PECAM-1 gene constructs into endothelial and nonendothelial cells were performed. We found 1) significantly more {gamma}-catenin and SHP-2 bound to the truncated than to the full-length PECAM-1; 2) stable expression of the truncated PECAM-1 in SW480 colon carcinoma cells resulted in a dramatic decrease in cell proliferation, whereas expression of comparable levels of the full-length PECAM-1 had no effect; 3) the decrease observed in cell proliferation is due, in part, to an increase in programmed cell death (apoptosis) and correlated with continuous caspase 8 cleavage and p38/JNK phosphorylation. These results support the intimate involvement of PECAM-1 in signal transduction cascades and also suggest that caspase substrates (e.g., PECAM-1) may possess distinct and unique functions on cleavage.—Ilan, N., Mohsenin, A., Cheung, L., Madri, J. A. PECAM-1 shedding during apoptosis generates a membrane-anchored truncated molecule with unique signaling characteristics.


Key Words: platelet-endothelial cell adhesion molecule (CD31) • endothelium • apoptosis • shedding • cleavage


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE VASCULAR NETWORK of arteries, veins, arterioles, venules, and capillary blood vessels nourishes and protects the body’s tissues. Once the vascular network is in place in the adult, the vascular endothelial cells are quiescent and the cellular turnover is measured in terms of years. Only on the right angiogenic signal would these cells start to proliferate and initiate the formation of new blood vessels. Although extensive efforts have been dedicated to elucidate the mechanisms that switch on angiogenesis (1 2 3) , little is known about the mechanisms responsible for blood vessel regression. Apoptosis of endothelial cells has been shown to contribute to vascular maturation and pruning during development (4) and in response to VEGF withdrawal in vitro (5 , 6) and in vivo (7) . Similar to growth factors and hormones, cell–cell and cell–matrix interactions play important roles in the regulation of cell growth, differentiation, and survival of many cell types including endothelial cells (8 9 10 11 12) . Whereas integrin-mediated cell–matrix interactions initiate a cascade of cytoplasmic protein phosphorylation, cadherin-mediated cell–cell adhesion generates a physical link, via plaque proteins, with the cytoskeleton. Actin itself as well as molecules involved in cytoskeleton regulation (i.e., {alpha}-fodrin, Gas2, gelsolin) have been shown to be substrates for proteolytic cleavage by caspases during apoptosis (13) , suggesting that the cytoskeleton integrity is a critical determinant of cell death (14) . Cleavage of plaque proteins, including ß- (15) and {gamma}-catenin (16) , as well as cleavage and shedding of vascular endothelial (VE) -cadherin (16) have also been observed during endothelial cell apoptosis. These data suggest an active disruption of the endothelial adherens junctions, followed by cell detachment and accumulation of dying cells floating in the culture medium.

Platelet-endothelial cell adhesion molecule (PECAM)-1 is a 130 kDa glycoprotein member of the immunoglobulin (Ig) superfamily of cell adhesion molecules. PECAM-1 expression is restricted to cells of the vascular system: platelets, monocytes, neutrophil selected T cells, and endothelial cells. In the latter, PECAM-1 is constitutively expressed in all vessel types, localizing to areas of cell–cell junctions in addition to lumen-facing areas of blood vessels. PECAM-1 has been shown to play an important role in the adhesion cascades leading to extravasation of leukocytes during inflammation in vitro (17) and in vivo (18 19 20) . Extravasation, however, is expected to occur predominantly at capillaries and postcapillary venules under most circumstances, suggesting that PECAM-1 may have other functions common to the vascular system as a whole.

Given the observed cleavage of VE-cadherin (16) , we were interested in studying PECAM-1’s fate during endothelial cells apoptosis. Here we show that PECAM-1 is cleaved early during endothelial cell apoptosis and prior to cell detachment. PECAM-1 cleavage results in a secreted, shed protein (~100 kDa) and a truncated fragment composed of the transmembrane and cytoplasmic domains (~28 kDa). {gamma}-Catenin and SHP-2, proteins previously reported to be recruited by PECAM-1 (21 , 22) , were preferentially associated with the truncated PECAM-1 fragment in transient transfection and immunoprecipitation experiments. Moreover, stable expression of the truncated PECAM-1 gene construct in SW480 colon carcinoma cells resulted a significant attenuation of cell growth due, at least in part, to an increase in apoptosis. In contrast, the full-length PECAM-1 protein was noted to inhibit cell death. We suggest that PECAM-1 cleavage during endothelial cell apoptosis would not only contribute to the loss of cell–cell adhesion, but would generate a truncated fragment functioning to accelerate cell death.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cells and cell cultures
Human umbilical vein endothelial cells (HUVEC) were obtained from Jordan Pober (Yale Medical School) and cultured in gelatin-coated flasks as described (6 , 22 , 23) . Hemangioendothelioma (EOMA) cells were obtained from Robert Auerbach (University of Wisconsin, Madison, Wis.) and grown in complete Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and antibiotics (24) . Bovine aortic endothelial cells (BAEC) were isolated and cultured as described (25) . 293T cells were purchased from the American Type Culture Association (Gaithersburg, Md.) and grown in DMEM supplemented with 10% FBS. Scp2 cells, a mouse mammary epithelial cell line (26) , were kindly provided by Mina Bissell (University of California, Berkeley, Calif.) and maintained in DMEM/F12 medium (1:1) supplemented with 2% FBS, 50 µg/ml gentamicin, and insulin (5 µg/ml). SW480 human colon carcinoma cells stably expressing the full-length (22 , 23) and the truncated PECAM-1 cDNAs were generated as described (23) and grown in DMEM supplemented with 10% FBS.

Embedding and culturing of HUVEC in 3-dimensional (3D) type I collagen gels were performed as described (6) .

For serum-free cultures, EOMA or SW480 cells were washed twice with and cultured in the presence of the very same medium, only lacking FBS. After 36 h, the medium was collected and floating cells were either counted (see below) or harvested by centrifugation (1000 g for 5 min) and lysed. The medium was then centrifuged at high speed (100,000 g for 20 min) to remove any residual membrane fractions, concentrated ~fourfold with centriprep-10 concentrator (Amicon, Beverly, Mass.), and saved in -70°C until analyzed. Adhering cells were washed and lysed in parallel with control cells.

PECAM-1 gene constructs and transfection
Full-length (27) or truncated (28) myc-tagged human PECAM-1 cDNAs in the expression vector pcDNA3 were used. The truncated gene construct was composed of a short, 40 amino acid segment of the ectodomain, the transmembrane, and the cytoplasmic domain. 10 µg DNA was transfected into subconfluent 293T, BAEC, and Scp-2 cells using lipofectamine (293T, Scp-2), or lipofectamine 2000 reagents (BAEC, Gibco BRL, Grand Island, N.Y.). Cells were lysed and used for biochemical analysis 24 h after transfection.

Cell lysate preparation, immunoprecipitation, and protein blotting
Cell cultures were pretreated with 1 mM orthovanadate for 15 min at 37°C, washed twice with ice-cold phosphate-buffered saline (PBS) containing 1 mM orthovanadate, and scraped into lysis buffer [50 mM Tris, pH 7.4, 150 mM NaCl, 1% Triton-X100, 1% Nonidet P-40, 0.5% deoxycholate, 1 mM orthovanadate, 1 mM PMSF, and a mixture of proteinase inhibitors (Boehringer Mannheim, GmbH, Germany)]. Total cellular protein concentration was determined by the BCA assay (Pierce, Rockford, Ill.), according to the manufacturer’s instructions. 20 µg of cellular protein were fractionated on sodium dodecyl sulfate (SDS) -polyacrylamide gels and protein immunoblotting was performed as described (6) . For immunoprecipitation, 100 µg of cellular protein were brought to volume of 1.0 ml in buffer containing 50 mM Tris, pH 7.5, 0.4M NaCl, 5 mM EDTA, and 0.5% Nonidet P-40, preabsorbed with normal rabbit serum, followed by protein A/G-Sepharose (Santa Cruz Biotechnology, Santa Cruz, Calif.) precipitation. The cleared supernatant was incubated with the appropriate antibody for 2 h on ice, followed by protein A/G-Sepharose immunoprecipitation. Beads were washed three times with the same buffer supplemented with 5% sucrose and once with the same buffer without sucrose and reduced salt concentration (50 mM NaCl). Sample buffer was then added; after boiling, samples were subjected to gel electrophoresis and immunodetection as described.

Antibodies and reagents
Rabbit polyclonal antibodies to human (BooBoo, raised against the cytoplasmic tail) and mouse (Sleet, raised against the ectodomain) PECAM-1 have been described (27 , 29) . Anti-phosphotyrosine (PY99), anti-human (C-20), and mouse (M-20) PECAM-1 (both raised against cytoplasmic epitopes), anti-myc epitope tag (9E10), anti-caspase 8 (H-134), anti-cyclin D1 (A-12), anti-p38 (N-20), and anti-JNK2 (D-2) were purchased from Santa Cruz Biotechnology. Rabbit polyclonal antibodies specific for the p85 fragment of PARP were purchased from Promega (Madison, Wis.). Monoclonal antibodies to ß-catenin, SHP-2 (also known as PTP1D), and casein kinase I{varepsilon} were purchased from Transduction Laboratories (Lexington, Ky.). Other monoclonal antibodies included anti-{gamma}-catenin (clone 15F11) and anti-ß-actin (both from Sigma). Anti-phospho-AKT, phospho-p38, and phospho-JNK/SAPK were purchased from New England BioLabs (Beverly, Mass.).

ZVAD-FMK was purchased from Calbiochem (La Jolla, Calif.) and dissolved in DMSO to make a stock concentration of 20 mM. GM6001 was purchased from AMS Scientific (Concord, Calif.) and dissolved in DMSO to a stock concentration of 1 mM. Matching volumes of DMSO were added to control cells.

Cell proliferation and apoptosis scores
Cells were trypsinized, counted and plated at 1.5 x 106 per flask. Cell number was counted 3, 6, and 10 days after plating with hemacytometer. For each time point, the average of three flasks was calculated and the experiment was repeated 4 times. The medium from each flask was saved and cell number was similarly counted in medium aliquots for each time point, cell type, and according to the experimental procedure. The total calculated cell number in the culture medium divided by the number of adhering cells was used as an apoptosis index.

Flow cytometry
A PECAM-1 expression profile was determined by indirect immunofluorescence staining, followed by flow cytometry. Cells were trypsinized, washed, and fixed with 4% paraformaldehyde for 30 min. The cells were then washed and permeabilized (0.1% saponin in PBS) for 30 min; after blocking (10% NGS, 0.1% saponin, 0.1% sodium azide in PBS, 30 min), cells were exposed to anti-PECAM-1 antibodies (raised against the cytoplasmic tail) for 60 min on ice. Cells were then washed and treated with goat anti-rabbit IgG-FITC-conjugated secondary antibodies for 30 min on ice. Immunofluorescent analysis was performed using a FACSCalibur fluorescent activated cell sorter and CellQuest software (Becton Dickinson, Mountain View, Calif.).

All experiments were repeated at least twice with similar results.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Endothelial cells have been reported to undergo apoptosis in response to growth factor withdrawal (30) . The first morphological changes observed after serum removal are cell retraction and membrane blebbing, with subsequent loss of cell–cell and cell–matrix contacts, resulting in cell detachment (30 , 31) . In endothelial cells, cell–cell contact is mediated by two distinct adhesion compartments: VE-cadherin and PECAM-1. To study PECAM-1’s fate during endothelial cell apoptosis, hemangioma-derived EOMA cells were grown for 36 h under serum-free conditions. As reported previously, these conditions effectively induce apoptosis, resulting in EOMA cell detachment and accumulation of cells floating in the culture medium. Lysate samples from control cells grown in complete medium (Con), cells grown under serum-free conditions but remained attached to the culture dish (At), cells that were floating in the culture medium (Fl), as well medium samples (Med) were analyzed for PECAM-1 expression by immunoblotting (Fig. 1A ). A significant loss of PECAM-1 was noted in apoptotic cells found floating in the culture medium, more than 80% as judged by densitometric analysis (Fig. 1A , upper panel). In agreement with this observation, a low molecular mass form of PECAM-1 (~100 kDa) was detected in the medium sample (Fig. 1A , upper panel). Reprobing the blot with antibodies specific for the cytoplasmic domain of PECAM-1 failed to detect the ~100 kDa form (Fig. 1A , lower panel), suggesting that the accumulation of the ~100 kDa PECAM-1 form in the culture medium is due to PECAM-1 shedding from the cell surface. If true, this would result the generation of an additional, truncated fragment composed of the transmembrane and the cytoplasmic domains of PECAM-1. Indeed, immunoblot analysis of the same samples from 13% SDS-PAGE gels revealed the presence of a truncated, ~28 kDa PECAM-1 fragment (Fig. 1B , upper panel). Cleavage of multiple caspase substrates, including ß-, and {gamma}-catenins, p21, p27, and pp125FAK, has been observed during endothelial cells apoptosis (16 , 30 , 31) . However, cleavage products of these proteins were detected only in cells that were floating in the culture medium and already committed to cell death. This was demonstrated by stripping and reblotting the same membrane with anti-ß-catenin antibodies (Fig. 1B , lower panel). The truncated fragment of PECAM-1 was already detected in the cells that remained attached to the culture dish under serum-free, but not control, conditions (Fig. 1B , upper panel) and prior to cell death commitment.



View larger version (41K):
[in this window]
[in a new window]
  		Figure 1. PECAM-1 shedding and appearance of a truncated ~28 kDa cytoplasmic fragment during endothelial cell apoptosis. Hemangioma-derived EOMA cells were left untreated (Con.) or grown for 36 h under serum-free (SF) conditions. Lysate samples from control cells (Con), cells that remain attached (At), or cells that were floating in the culture medium (Fl) under SF conditions, as well as medium samples (Med) were analyzed by 8% (A) and 13% (B) SDS-PAGE. PECAM-1 was detected by antibodies directed against the extracellular (ecto) domain (A, upper panel) or antibodies specific for the cytoplasmic (cyto) domain (A, lower panel; B, upper panel). Note the accumulation of ~100 kDa PECAM-1 molecules in the culture medium (A) and the appearance of a truncated ~28 kDa PECAM-1 fragment prior to cell detachment (B). In contrast, ß-catenin cleavage was detected only in the floating, apoptotic cells (B, lower panel). C) HUVEC were embedded in 3D collagen gels and grown for 1 day in the absence (-) or presence (+) of PMA (16 nM). Lysate samples were analyzed for PECAM-1 cleavage (*) as in panel B. Some nonspecific (ns) bands appeared in this assay, most probably due to serum components trapped in the collagen gels. Note the significant inhibition of PECAM-1 cleavage in the presence of PMA.

We next wanted to confirm PECAM-1 cleavage during apoptosis induced by means other than serum withdrawal. We recently reported that the apoptosis machinery is rapidly induced once HUVEC are embedded and grown in 3D collagen gels and that phorbol esters (PMA) act as a survival factor and as a morphogen (6) . Indeed, the 28 kDa fragment of PECAM-1 was detected in lysates made from HUVEC grown for 1 day in 3D collagen gels, but not in PMA-treated cultures (Fig. 1C ). This confirms PECAM-1 cleavage under two different apoptosis systems and two different cell types.

To better understand the mechanisms that lead to PECAM-1 shedding, EOMA cells were grown under serum-free conditions in the absence (0) or presence of increasing concentrations of GM6001, a broad spectrum inhibitor of metalloproteinases, and medium samples were analyzed for PECAM-1 shedding by immunoblotting (Fig. 2A ). In agreement with our previous results, the ~100 kDa PECAM-1 form was readily detected in the culture medium of control (0) and in the presence of low GM6001 concentrations (10 nM). However, treatment with 100 nM GM6001 or higher concentrations significantly inhibited PECAM-1 shedding (Fig. 2A , upper panel). Shedding was again confirmed by the failure of an antibodies directed to PECAM-1’s cytoplasmic tail to recognize the ~100 kDa form of PECAM-1 found in the culture medium (Fig. 2A , 2nd panel). Whereas GM6001 effectively inhibited PECAM-1 shedding, it did not affect the actual PECAM-1 cleavage (Fig. 2A , lower panel). As expected, PECAM-1 cleavage was effectively inhibited by ZVAD, a caspase inhibitor (Fig. 2B ). This may suggest that PECAM-1 shedding involves two major steps: the first being the actual cleavage mediated by caspase activity, followed by release of the ectodomain, which is mediated by metalloproteinase activity (Fig. 7) .



View larger version (24K):
[in this window]
[in a new window]
  		Figure 2. PECAM-1 shedding, but not cleavage, is mediated by MMP activity, whereas PECAM-1 cleavage is mediated by caspase activity. A) EOMA cells were left untreated or were grown under SF conditions without (0) or with increasing concentrations of GM6001, an inhibitor of MMP activity. Medium (1st and 2nd panels) and lysate (3rd panel) samples were analyzed with anti-PECAM-1 antibodies by immunoblotting. Note the dose-dependent inhibition of PECAM-1 accumulation in the culture medium, but not PECAM-1 cleavage, upon GM6001 treatment. B) EOMA cells were grown under SF conditions without (0) or with increasing concentrations of ZVAD, a caspase inhibitor. Lysate samples were analyzed with anti-PECAM-1 (cyto) antibodies by immunoblotting. Note the dose-dependent inhibition of PECAM-1 cleavage.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 7. Working model for PECAM-1 shedding and function of the truncated cleaved fragment. Apoptosis signals (serum withdrawal, dispersion in 3D collagen gels, [1]) lead to caspase activation [2] and cleavage of PECAM-1. However, PECAM-1’s ectodomain remains tethered to the cell surface [3] and PECAM-1’s shedding depends on MP activity [4]. PECAM-1’s cytoplasmic domain (which remains tethered to the plasma membrane), preferentially recruits adapter and signaling molecules [5], resulting in caspase 8 cleavage and p38/JNK activation, leading to the induction of apoptosis [6].

The appearance of the ~28 kDa truncated PECAM-1 fragment prior to cell detachment (Fig. 1B ) raised the possibility that such a fragment is not inert but may possess some function(s). To test this hypothesis, we transiently transfected 293T cells with control DNA (Vo), the full-length (Fl), or a truncated (Tr.) Myc-tagged PECAM-1 gene construct (Fig. 3A ). The truncated gene construct was composed of a short, 40 amino acid segment from the ectodomain, the transmembrane, and the whole cytoplasmic domain and appears as a ~32 kDa band in immunoblot analysis (see below). Immunoprecipitation (IP) for the Myc tag, followed by immunoblotting, revealed similar PECAM-1 expression (Fig. 3A , upper panel) and tyrosine phosphorylation (Fig. 3A , 2nd panel) levels of both PECAM-1 gene constructs. However, reprobing the same membrane with anti-{gamma}-catenin, a recently identified PECAM-1 partner (22) , clearly indicated preferential {gamma}-catenin association with the truncated PECAM-1 fragment (Fig. 3A , 3rd panel). In contrast, no association of PECAM-1 with ß-catenin was detected (Fig. 3A , lower panel), suggesting specificity in the recruitment of selected proteins. This was further confirmed by similar transfection/IP experiments using BAEC (Fig. 3B ), suggesting that such preferential recruitment also occurs in the context of endothelial cells. In contrast, a different recruitment pattern was observed on transfection of Scp2, a mouse mammary epithelial cell line. In this case, SHP2 rather than {gamma}-catenin was preferentially associated with the truncated PECAM-1 fragment (Fig. 3C ). The amount of full-length and truncated PECAM-1 proteins pulled down, as well as tyrosine phosphorylation levels, were similar in all of our experiments (Fig. 3A and data not shown). Together, these results suggest to us that indeed the truncated PECAM-1 fragment recruits selected PECAM-1 partners with higher affinity and that the exact set of proteins recruited is also cell type dependent. Thus, the generation of a truncated PECAM-1 fragment during apoptosis may create a unique PECAM-1 domain(s) capable of participating in and enhancing signaling events.



View larger version (48K):
[in this window]
[in a new window]
  		Figure 3. Preferential recruitment of adapter proteins to PECAM-1’s truncated fragment. A) 293T cells were transiently transfected with control DNA (Vo), full-length (Fl) or truncated (Tr) PECAM-1 gene constructs containing the Myc epitope tag, and lysate samples were immunoprecipitated (IP) with anti-myc epitope tag, followed by immunoblotting with anti-PECAM-1 (1st panel), anti-phosphotyrosine (P-Y, 2nd panel), anti-{gamma}-catenin (3rd panel), and anti-ß-catenin (4th panel) antibodies. Note preferential recruitment of {gamma}- but not ß-catenin to PECAM-1’s truncated fragment. Bovine aortic endothelial cells (BAEC, B) and Scp2 mouse mammary epithelial cells (Scp, C) were similarly transfected with the full-length (Fl) or truncated (Tr) PECAM-1/myc-tag gene constructs. Lysate samples were analyzed for PECAM-1 (first panels) and actin (second panels) expression levels by immunoblotting. The same lysate samples were IP for Myc, followed by {gamma}-catenin (B, 3rd panel) and SHP-2 (C, 3rd panel) immunoblotting, confirming preferential recruitment of the adapter and signaling molecules {gamma}-catenin (BAEC) and SHP2 (Scp2 cells) to PECAM-1’s truncated fragment in endothelial and mammary epithelial cells, respectively.

To take this concept a step further, we stably transfected our full-length and truncated PECAM-1 gene constructs into human colon carcinoma SW480 cells. Flow cytometry analysis revealed similar expression levels for both forms of PECAM-1 (Fig. 4A ). Moreover, the FACS staining pattern suggests that both cell populations are heterogeneous and are composed of a pool of clones expressing PECAM-1 at a wide range of levels. Immunostaining revealed that the truncated construct localized to the cell surface, although not exclusively to areas of cell–cell contact. In contrast, the full-length construct localized mainly to cell–cell contacts (22 , 23 ; data not shown). SW480 cells expressing the truncated PECAM-1 fragment were noted to grow at a much slower rate than the Vo control cells or ones that express the full-length PECAM-1 (Fig. 4B ). Plating cells at 1.5 x 106/flask yielded ~46 x 106 Vo control or full-length PECAM-1-expressing cell 10 days later. In contrast, only 3.5 x 106 of the truncated PECAM-1-expressing cells were counted 10 days later, and no obvious increase in cell number was noted before day 6 (Fig. 4B ). No differences in initial cell adhesion were noted. However, the truncated PECAM-1-expressing cells observed to be adhering were noted to detach from the culture dish at a relatively high rate compared to the Vo and full-length PECAM-1-expressing cells. The ratio between the floating cells and the cells that remained attached was used as an estimation for apoptosis (Fig. 5A ). A low basal level of 3.5–4.5% cell death was calculated for the Vo control and for the full-length PECAM-1-expressing cells (Fig. 5A ). In contrast, a high level (over 20%) of cell death was calculated for the truncated PECAM-1-expressing cells (Fig. 5A ). This high score of apoptosis was noticed in cells grown in complete medium containing 10% serum and mimic our initial studies in which apoptosis was induced by serum withdrawal (Figs. 1 , 2) . To further define apoptosis as the reason for cell detachment and floating, the truncated PECAM-1-expressing cells were either left untreated (con) or incubated with ZVAD, a caspase inhibitor, and apoptosis scores were calculated (Fig. 5A , right). A high apoptotic score was noted in the control cells, as previously observed (Fig. 5A , left). However, ZVAD (at 2 µM) significantly reduced the apoptosis level to its basal ~5% level (Fig. 5A , right). Figure 5B summarizes cell number and apoptosis data shown in Fig. 4B and Fig. 5A . Next, we confirmed these morphological assessments of apoptosis with selected biochemical indications of apoptosis. Caspase 8, a caspase that is believed to lie upstream in the caspase cascade (32 , 33) , was found to be cleaved into cleavage products typical for caspase 8 activation (34) in lysates made from the truncated, but not the full-length PECAM-1-expressing or control cells (Fig. 5C , upper panel). In addition, the p85 PARP cleavage product was detected only in the truncated PECAM-1-expressing cells (Fig. 5C , 2nd panel), in agreement with caspase activation in these cells. Culturing the full-length PECAM-1-expressing cells under serum-free conditions resulted in the appearance of the apoptotic cleaved fragment (Fig. 5D ), as was noted for endothelial cells (Figs. 1 , 2) , although the SW480 cells appear to be less sensitive to growth factor removal. However, a robust cleavage of the truncated PECAM-1 into its apoptotic 28 kDa fragment was observed in the truncated PECAM-1-expressing cells under normal conditions and without any stimulation (Fig. 5D ). This is another indication of the high apoptotic index in these cells in addition to caspase activation and PARP cleavage (Fig. 5C ) and, moreover, suggests that the cleavage site lies within the membrane proximal 40 amino acids.



View larger version (25K):
[in this window]
[in a new window]
  		Figure 4. Inhibition of SW480 cell growth on stable expression of PECAM-1’s truncated fragment. SW480 colon carcinoma cells were transfected with vector only (Vo), full-length (Fl) or truncated (Tr) PECAM-1 gene constructs, and selected with G418 (400 µg/ml). Surviving cells were pooled and expanded. A) FACS analysis of the full-length (Fl) and truncated (Tr) PECAM-1-expressing SW480 cells. Note the heterogeneous nature of both cell populations. B) Vector-only (Vo), full-length (Fl), and truncated (Tr) cells were plated at 1.5 x 106 per flask and cell number was counted after 3, 6, and 10 days with hemacytometer. Numbers represent an average of 4 different experiments each performed in triplicate. PECAM-1 expression was further confirmed by immunoblotting (insert). Note the dramatic inhibition of cell proliferation on the expression of the PECAM-1 truncated, but not the full-length, PECAM-1 gene construct.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 5. SW480 cells stably expressing the truncated PECAM-1 gene construct exhibit a high apoptosis index. High numbers of the SW480 cells stably expressing the truncated PECAM-1 gene construct were noted to be floating in the culture medium. The ratio between floating and adhering cells was calculated as measure of the apoptosis index (A, left). Note the high apoptosis index in SW480 cells expressing the PECAM-1 truncated gene construct. SW480 cells stably expressing the PECAM-1 truncated gene construct were left untreated (Con.) or treated with 2 or 5 µM ZVAD, a caspase inhibitor, and the apoptosis index was calculated as above (A). Note the inhibition of the Tr cell detachment (apoptosis) on treatment with low concentrations of ZVAD. Data summarizing Vo, Fl, and Tr SW480 cell proliferation and apoptosis behaviors are shown in panel B. C) Lysate samples from Vo, Fl and Tr SW480 cells were analyzed for caspase 8 (upper panel), p85 PARP (middle panel) and actin (lower panel) by immunoblotting, confirming high apoptosis levels in the Tr. PECAM-1-expressing cells. D) Full-length PECAM-1-expressing SW480 cells were grown in complete medium (Con) or cultured under serum-free conditions and cells that remained attached (A) or cells that were floating in the culture medium (F), as well as Tr. PECAM-1-expressing cells grown in complete medium, were collected, lysed and analyzed as described in Fig. 1B . Note the robust cleavage of the truncated PECAM-1 gene product into the 28 kDa apoptotic fragment (*) under control, serum-containing conditions.

We recently reported that stable PECAM-1 expression in SW480 cells results in the recruitment of ß- (23) and {gamma}-catenin (22) into areas of cell–cell contacts, preventing their nuclear translocation. Since the truncated PECAM-1 gene product was noted to preferentially associate with {gamma}-catenin (Fig. 3) , we hypothesized that the decrease in catenin-regulated transcription is one possible explanation for the differences observed in cell growth (Figs. 4 , 5) . Cyclin D1 expression was recently reported to lie downstream of, and to be induced by ß-catenin (35 , 36) . We therefore compared cyclin D1 levels in the Vo control and in the full-length and truncated PECAM-1-expressing cells. A significant threefold decrease in cyclin D1 expression was noted on PECAM-1 expression in SW480 cells (Fig. 6A , 4th panel), as previously reported (22) . However, a similar decrease in cyclin D1 expression was noted for both the full-length and the truncated PECAM-1-expressing cells. Besides no change in casein kinase I (CKI, Fig. 4A , lower panel), a recently identified Wnt component upstream of ß- and {gamma}-catenin (37 , 38 ; Fig. 4A , 2nd and 3rd panels, respectively) expression levels was noted, suggesting a specific inhibition of components downstream the catenin pathway. Nevertheless, the decrease in cyclin D1 levels cannot account for the changes observed in cell number. We therefore raised the possibility of differences in expression and/or activation of a survival pathway such as PKB/Akt, which has been implicated as an apoptosis inhibitor (39) . No major differences in Akt phosphorylation were noted by immunoblotting cell lysates with anti-phospho-Akt antibodies (Fig. 6B , upper panel). In sharp contrast, a robust phosphorylation, and hence activation, of p38 and JNK/SAPK, MAPK-family members generally associated with the promotion of apoptosis (40 41 42 43) were observed in the truncated, but not the full-length PECAM-1-expressing or control cells (Fig. 6B , 2nd and fourth panels). Thus, the generation of the truncated fragment of PECAM-1 during apoptosis may further enhance the apoptosis machinery—for example, by activating the p38 and JNK pathways. This activity is the exact opposite of the anti-apoptotic effect recently attributed to the full-length PECAM-1 molecule (44) . To test the possible anti-apoptotic effect of the full-length PECAM-1 protein, Vo control cells, full-length, and truncated PECAM-1-expressing SW480 cells were subjected to serum-free conditions and apoptosis was scored as described (Fig. 6C ). A twofold increase in apoptosis was noted for the Vo control cells (4.5% vs. 9%). However, no significant changes in the apoptosis scores were observed in the full-length PECAM-1-expressing cells (3.5% vs. 4%), in agreement with the lack of PECAM-1 cleavage in adherent cells (Fig. 5D ). Moreover, subjecting the truncated PECAM-1-expressing cells to serum free conditions not only did not yield further induction, but also actually reduced the apoptosis levels by 30% (19% vs. 13%). These results may suggest that the full-length and the truncated PECAM-1 proteins indeed have an opposite function in the context of SW480 cell survival and, moreover, that in the presence of the truncated PECAM-1 fragment, SW480 cells become more sensitive to serum-containing factor(s).



View larger version (28K):
[in this window]
[in a new window]
  		Figure 6. Apoptosis of the truncated PECAM-1-expressing SW480 cells correlates with p38/JNK activation. A) Vo control cells (Vo), full-length, and truncated PECAM-1-expressing cell lysates were analyzed for PECAM-1 (1st panel). ß- and {gamma}-catenin (2nd and 3rd panels, respectively), cyclin D 1 (CD1: 4th panel), and casein kinase I (CKI, 5th panel) expression by immunoblotting. B) Cell lysates were similarly analyzed with anti-phospho-AKT/PKB antibodies (p-AKT, 1st panel), anti-phospho-, or total p38 antibodies (2nd and 3rd panels) and anti-phospho-JNK/SAPK (p-JNK/SAPK, fourth panel) or anti-JNK2 antibodies (5th panel). C) Control (Vo), full-length (Fl), and truncated (Tr) PECAM-1-expressing SW480 cells were grown under control, serum-containing conditions (open bars) or were cultured under serum-free conditions (filled bars) for 2 days and apoptosis was scored as described. Note the decrease in the apoptotic scores of the truncated expressing cells under serum-free conditions.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Solubilization, or shedding, of transmembrane proteins by specific proteolytic cleavage in the juxtamembrane region is a well-recognized phenomenon that plays an important regulatory role in a wide variety of physiological and developmental events. Numerous membrane proteins that are members of diverse functional classes are susceptible to this type of processing. These include growth factor (45 46 47) , growth factor receptors (48 49 50 51) , cell adhesion molecules (52) , extracellular matrix proteins (53) and other membrane proteins (54 , 55) . The majority of transmembrane proteins nonetheless are resistant to proteolytic release, suggesting that this is a specific regulated process catalyzed by specialized proteases. Stimulation by phorbol esters (i.e., PMA, TPA) and inhibition by metalloproteinases inhibitors (i.e., TAPI) are universal characteristics of regulated ectodomain shedding (55) . Less is known, however, about the role and regulation of protein shedding in the context of apoptosis. PECAM-1 shedding during endothelial cell apoptosis ultimately results in the generation of two fragments: a secreted ectodomain (~100 kDa) and a truncated molecule comprised of the transmembrane and the cytoplasmic domains (~28 kDa; Figs. 1 , 2 ). Soluble recombinant PECAM-IgG fusion protein has been shown to block transendothelial cell migration of neutrophils, monocytes, and natural killer cells in vitro and to inhibit acute inflammation in vivo (56) . More recently, constitutive secretion of soluble PECAM-1 into the plasma of transgenic mice has been demonstrated to severely blunt the acute inflammatory response (20) . These data suggest that the secreted PECAM-1 molecules may exhibit an important function on their own. Moreover, an increase in the serum levels of soluble PECAM-1 have been observed in multiple sclerosis (MS) patients with active gadolinium-enhancing lesions (57) . This may reflect an in vivo PECAM-1 shedding and may link endothelial cell apoptosis to the pathogenicity of MS.

We wanted to study the possible function of the remaining truncated PECAM-1 cytoplasmic fragment. The truncated PECAM-1 fragment appears relatively early in the onset of apoptosis and prior to cell detachment (Fig. 1B ). Exposure of cells to serum-free conditions results in viable, adhering, and apoptotic floating cells. Whereas viable adhering cells will survive and proliferate once resupplemented with growth factors (30) , floating cells are at the point of no return and will never regrow. Thus, the loss of cell–cell and cell–matrix contacts are crucial events that will ultimately lead to apoptosis commitment. Several proteins involved in adherens complexes, including ß- and {gamma}-catenin, have been implicated as caspase substrates. However, catenin cleavage was detected only in the floating, apoptotic-committed endothelial (Fig. 1B ; 16 ) and nonendothelial (15) cells and thus may only represent a by-product of the apoptotic machinery. In contrast, shedding of VE-cadherin (16) and PECAM-1 (Figs. 1 , 2) may play an important role in the disassembly of junctional complexes, leading to cell detachment. Processing of VE-cadherin was detected mainly in floating cells (16) , and the expected VE-cadherin truncated fragment has not been detected and characterized. In contrast, the PECAM-1 cytoplasmic apoptotic fragment was detected in adherent cells (Fig. 1B ) and therefore may represent the first (or an early) endothelial adhesion component to be subjected to caspase-mediated cleavage. The ability of low concentration of ZVAD (2 µM) to effectively inhibit PECAM-1 cleavage (Fig. 2B ) clearly implicates PECAM-1 as a caspase target. Complete PECAM-1 shedding depends, however, on the activity of another component—a metalloproteinase (Fig. 2A ). This suggests that the cleaved PECAM-1 ectodomain remains tethered to the cell surface by an as yet unidentified mechanism, the release of which depends on metalloproteinase activity (Fig. 7 ). Attempts directed at identifying the cleavage site(s) and the tethering mechanism(s) are in progress.

Proteins targeted for cleavage by caspase during apoptosis are involved in diverse functions such as RNA splicing, DNA repair, and scaffolding of the cytosol and nucleus (13 , 58) , resulting in inactivating attempts of repair mechanisms. In contrast with the loss-of-function phenotype associated with such dismantling, a gain-of-function phenotype was also suggested for caspase substrates. For example, caspase 3-mediated cleavage of the DNA fragmentation factor generates an active factor that produces DNA fragmentation (59) -the hallmark of apoptosis; gelsolin cleavage generates a constitutively active fragment that acts to depolymerize F-actin (60) . PECAM-1, however, lacks enzymatic activity; thus, its ability to participate and enhance signal transduction events depends on the recruitment of selected adapter proteins. Indeed, preferential recruitment of {gamma}-catenin (Fig. 3A , B ) and SHP-2 (Fig. 3C ) to the truncated PECAM-1 cytoplasmic fragment have been noted on transient transfections of endothelial and nonendothelial cells. Whereas {gamma}-catenin recruitment is regulated by PECAM-1 serine/threonine phosphorylation (22) , the recruitment of SHP-2 depends on PECAM-1 tyrosine phosphorylation (21) . No changes in the tyrosine phosphorylation levels have been noted when comparing the truncated and full-length PECAM-1 proteins (Fig. 3A and data not shown). This may suggest that once cleaved, the remaining cytoplasmic PECAM-1 tail assumes a different conformation that mimics the phosphorylated conformation (Fig. 7) . Such a conformation may also be more susceptible to a phosphatase, resulting in serine/threonine dephosphorylation and hence an increase in {gamma}-catenin recruited (22) . Our interpretation of a gain-of-function phenotype associated with the truncated PECAM-1 fragment was further confirmed by stably expressing the truncated and the full-length PECAM-1 gene constructs in SW480 cells. Cells selected for cellular and molecular analysis were first characterized as a pool of clones (Fig. 4A ). Thus, the observed phenotype is not due to a specific clone but rather to a sum of many different ones. A significant decrease in adhering cell numbers was noted for the truncated, but not the full-length PECAM-1-expressing or control cells (Fig. 4B ). This decrease is due at least in part to a significant increase of apoptosis observed in the truncated PECAM-1-expressing cells (Fig. 5A , B ).

Several lines of evidence support this observation. First, low concentrations (2 mM) of ZVAD were able to prevent the high apoptotic rate and to restore it to the basal level (Fig. 5A , left). In addition, caspase 8 cleavage products were detected only in the lysate samples of the truncated, but not the full-length PECAM-1-expressing or the control cells lysates (Fig. 5C , upper panel). Similarly, PARP cleavage into its signature 85 kDa product was detected only in the truncated PECAM-1-expressing cells (Fig. 5C , middle panel). Last, a robust and constitutive cleavage of the truncated PECAM-1 gene product into the 28 kDa apoptotic fragment was observed under normal growth medium conditions whereas cleavage of the full-length PECAM-1 was inefficient even after serum withdrawal (Fig. 5D ). Together, cellular and biochemical criteria point to the induction of SW480 cell apoptosis on the expression of the truncated PECAM-1 gene product. Thus, the early cleavage of PECAM-1 during endothelial cell apoptosis (Fig. 1B ) would result in not only the initiation of junctional complex disassembly, but would also generate a truncated fragment that may participate in the enhancement of the apoptotic process. Transient transfection of SW480 cells with the secreted form of the interferon (IFN) ß gene similarly induced growth arrest, apoptosis, and caspase 8 activation (61) . This is most likely due to the IFN ß receptor signaling, since the nonsecreted form had no effect. Therefore, the truncated PECAM-1 fragment may mimic signaling pathways normally induced by growth factor receptors.

The mechanism underlying this PECAM-1 function is not yet clear. Recently, we demonstrated that stable PECAM-1 expression in SW480 cells results in the recruitment of both ß- (23) and {gamma}-catenin (22) to areas of cell–cell contact, preventing their nuclear translocation. The observed decrease in cyclin D1 expression (Fig. 6 , 4th panel) further supports the inhibition of catenin nuclear translocation and, as a result, of catenin-regulated-transcription noted on PECAM-1 expression. Whereas this result confirms the notion that PECAM-1 functions as a catenin modulator, it cannot account for the dramatic effect noted on the truncated PECAM-1 expression (Figs. 4 , 5) , as both the full-length and the truncated fragment yielded similar decrease in cyclin D1 expression. Similarly, no major changes in Akt phosphorylation levels, a mechanism that has been strongly implicated in cell survival (39) , were observed (Fig. 6B , upper panel). In contrast, a robust phosphorylation of p38 and JNK, MAPK family members that have been shown to be activated by environmental stress such as ultraviolet light, osmotic stress, or heat shock, was specific for the truncated PECAM-1-expressing cells. Activation of the p38/JNK pathway was observed in PC-12 cells, followed by nerve growth factor withdrawal (40) , in Fas-induced apoptosis of Jurkat T cells (62 , 63) and as part of the TNFR signaling pathway (64) . In the PC-12 cell system, p38/JNK activation was observed to be an early event and preceded the induction of apoptosis (40) . This agrees with the relatively early cleavage of PECAM-1 in endothelial cells after serum withdrawal (Fig. 1B ). Whereas SW480 cells do express the Fas ligand (Fas-L), they do not express any detectable levels of Fas (data not shown), eliminating the Fas/Fas-L system as a possible reason for SW480 cell apoptosis. Growth factor withdrawal would enhance apoptosis presumably by decreasing protective and survival pathways rather than inducing promoting ones. Similar phosphorylation levels of Akt (Fig. 6B , upper panel) may suggest that at least this particular protective and survival pathway cannot account for the robust apoptosis in the truncated PECAM-1-expressing cells. The finding that serum-free conditions actually protect these cells from dying (Fig. 6C ) suggests that the presence of the truncated PECAM-1 protein enhances signaling cascades initiated by serum-containing factors that lead to p38/JNK activation. We recently reported that by serving as a scaffolding, PECAM-1 is involved in regulating the phosphorylation levels of selected proteins: PECAM-1 can mediate ß-catenin dephosphorylation (23) and an increase in STAT5 phosphorylation levels (N. Ilan et al., unpublished results) by bringing an enzyme (a phosphatase or a kinase) and its substrate to close proximity. We therefore argue for a similar scenario in the case of our truncated PECAM-1-expressing cells. More specifically, we suggest that components of the TNFR signaling pathway are preferentially associated with the truncated PECAM-1 protein and that such an interaction leads to their activation, either by assuming plasma membrane localization or by PECAM-1-bound kinase/phosphatase. Reasonable candidates include the TNF receptor-associated factor 2, which has been shown to mediate activation of JNK (65) and the receptor-interacting protein. This would help explain the constitutive activation of caspase 8 (Fig. 5D ), which has been shown to be recruited to and activated by the TNFR complex (32) . Studies examining these possibilities are currently under way.

In contrast to the death-promoting effect of the truncated PECAM-1 fragment, the full-length PECAM-1 protein seems to inhibit cell death (Fig. 6C ), in agreement with previous reports (44) . This is most likely due to its function as an adhesion molecule, further supporting our notion that PECAM-1 functions to stabilize the vascular endothelium (23) , whereas PECAM-1 cleavage would enhance its disruption.

Thus, PECAM-1 appears to be a molecule capable of exhibiting diverse functions: mediating adhesion between neighboring endothelial cells and the interaction between endothelial cells and cells of the immune system; serving as a dynamic scaffolding for ß- and {gamma}-catenin, SHP-2 and STAT family members, leading to intimate involvement in signaling cascades. Such interactions would affect diverse cellular and biochemical functions such as cell death, cell migration, calcium flux, and integrin affinity. Further elucidation of PECAM-1 structure/function relationships will result in a more complete appreciation of this dynamic, multi-functional protein.


   ACKNOWLEDGMENTS
 
Supported in part by USPHS grants R37-HL28373 and PO1-DK38979 to J.A.M. and a Reed Foundation Fellowship to N.I.

Received for publication June 1, 2000. Revision received August 8, 2000.
   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Folkman, J., D‘Amore, P. A. (1996) Blood vessel formation: what is its molecular basis. Cell 87,1153-1155[Medline]
  2. Hanahan, D., Folkman, J. (1996) Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell 86,353-364[Medline]
  3. Ferrara, N., Davis-Smyth, T. (1997) The biology of vascular endothelial growth factor. Endocr. Rev. 18,4-25[Abstract/Free Full Text]
  4. Alon, T., Hemo, I., Itin, A., Peer, J., Stone, J., Keshet, E. (1995) Vascular endothelial growth factor acts as a survival factor for newly formed retinal vessels and has implications for retinopathy of prematurity. Nature Med 1,1024-1028[Medline]
  5. Greber, H.-P., McMurtrey, A., Kowalski, J., Yan, M., Keyt, B. A., Dixit, V., Ferrara, N. (1998) Vascular endothelial growth factor regulates endothelial cell survival through the phosphatidylinositol 3'-kinase/Akt signal transduction pathway. J. Biol. Chem. 273,30336-30343[Abstract/Free Full Text]
  6. Ilan, N., Mahooti, S., Madri, J. A. (1998) Distinct signal transduction pathways are utilized during the tube formation and survival phases of in vitro angiogenesis. J. Cell Sci. 111,3621-3631[Abstract]
  7. Benjamin, L. E., Golijanin, D., Itin, A., Pode, D., Keshet, E. (1999) Selective ablation of immature blood vessels in established human tumors follows vascular endothelial growth factor withdrawal. J. Clin. Invest. 103,159-158[Medline]
  8. Meredith, J. E., Fazeli, B., Schwartz, M. A. (1993) The extracellular matrix as a cell survival factor. Mol. Biol. Cell 4,953-961[Abstract]
  9. Frisch, S. M., Francis, H. (1994) Disruption of epithelial cell–matrix interactions induces apoptosis. J. Cell Biol. 124,619-626[Abstract/Free Full Text]
  10. Hermiston, M. L., Gordon, J. I. (1995) In vivo analysis of cadherin function in the mouse intestinal epithelium: essential roles in adhesion, maintenance of differentiation and regulation of programmed cell death. J. Cell Biol. 129,489-506[Abstract/Free Full Text]
  11. Meredith, J. E., Schwartz, M. A. (1997) Integrins, adhesion and apoptosis. Trends Cell Biol 7,146-150[Medline]
  12. Tlsty, T. D. (1998) Cell-adhesion-dependent influences on genomic instability and carcinogenesis. Curr. Opin. Cell Biol. 10,647-653[Medline]
  13. Tan, X., Wang, J. Y. J. (1998) The caspase-RB connection in cell death. Trends Cell Biol 8,116-120[Medline]
  14. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M., Ingber, D. E. (1997) Geometric control of cell life and death. Science 276,1425-1428[Abstract/Free Full Text]
  15. Brancolini, C., Lazarevic, D., Rodriguez, J., Schneider, C. (1997) Dismantling cell–cell contacts during apoptosis is coupled to a caspase-mediated proteolytic cleavage of ß-catenin. J. Cell Biol. 139,759-771[Abstract/Free Full Text]
  16. Herren, B., Levkau, B., Rains, E. W., Ross, R. (1998) Cleavage of ß-catenin and plakoglobin and shedding of VE-cadherin during endothelial apoptosis: evidence for a role for caspases and metalloproteinases. Mol. Biol. Cell 9,1589-1601[Abstract/Free Full Text]
  17. Muller, W. A., Weigel, S. A., Deng, X., Phillips, D. M. (1993) PECAM-1 is required for transendothelial migration of leukocytes. J. Exp. Med. 178,449-460[Abstract/Free Full Text]
  18. Vaporciyan, A. A., DeLisser, H. M., Yan, H., Mendiguren, I. I., Thom, S. R., Jones, M. L., Ward, P. A., Albelda, S. M. (1993) Involvement of platelet endothelial cell adhesion molecule-1 in neutrophil recruitment in vivo. Science 262,1580-1582[Abstract/Free Full Text]
  19. Duncan, G. S., Andrew, D. P., Takimoto, H., Kaufman, S. A., Yoshida, H., Spellberg, J., de la Pompa, J. L., Elia, A., Wakeham, A., Karan-Tamir, B., Muller, W. A., Senaldi, G., Zukowski, M. M., Mak, T. W. (1999) Genetic evidence for functional redundancy of platelet/endothelial cell adhesion molecule-1 (PECAM-1): CD31-deficient mice reveal PECAM-1-dependent and PECAM-1-independent functions. J. Immunol. 162,3022-3030[Abstract/Free Full Text]
  20. Liao, F., Schenkel, A. R., Muller, W. A. (2000) Transgenic mice expressing different levels of soluble platelet/endothelial cell adhesion molecule-IgG display distinct inflammatory phenotype. J. Immunol. 163,5640-5648[Abstract/Free Full Text]
  21. Jackson, D. E., Ward, C. M., Wang, R., Newman, P. J. (1997) The protein-tyrosine phosphatase SHP-2 binds platelet/endothelial cell adhesion molecule-1 (PECAM-1) and forms a distinct signaling complex during platelet aggregation. J. Biol. Chem. 272,6986-6993[Abstract/Free Full Text]
  22. Ilan, N., Cheung, L., Pinter, E., Madri, J. A. (2000) PECAM-1 (CD31): A scaffolding molecule for selected catenin family members whose binding is mediated by different tyrosine and serine/threonine phosphorylation. J. Biol. Chem. 275,21435-21443[Abstract/Free Full Text]
  23. Ilan, N., Mahooti, S., Rimm, D. L., Madri, J. A. (1999) PECAM-1 (CD31) functions as a reservoir for and a modulator of tyrosine-phosphorylated ß-catenin. J. Cell Sci. 112,3005-3014[Abstract]
  24. Obeso, J., Weber, J., Auerbach, R. (1990) A hemangioendothelioma-derived cell line: Its use as a model for the study of endothelial cell biology. Lab. Invest. 63,259-269[Medline]
  25. Lu, T. T., Barreuther, M., Davis, S., Madri, J. A. (1997) Platelet endothelial cell adhesion molecule-1 is phosphorylatable by c-Src, binds Src-Src homology 2 domain, and exhibits immunoreceptor tyrosine-based activation motif-like properties. J. Biol. Chem. 272,14442-14446[Abstract/Free Full Text]
  26. Roskelley, C. D., Desprez, P. Y., M. J. Bissell, M. J. (1994) Extracellular matrix-dependent tissue-specific gene expression in mammary epithelial cells requires both physical and biochemical signal transduction. Proc. Natl. Acad. Sci. USA 91,12378-12382[Abstract/Free Full Text]
  27. Lu, T. T., Yan, L. G., Madri, J. A. (1996) Integrin engagement mediates tyrosine dephosphorylation on platelet-endothelial cell adhesion molecule 1. Proc. Natl. Acad. Sci. USA 93,11808-11813[Abstract/Free Full Text]
  28. Kim, C. S., Wang, T., Madri, J. A. (1998) Platelet endothelial cell adhesion molecule-1 expression modulates endothelial cell migration in vitro. Lab. Invest. 78,583-590[Medline]
  29. Pinter, E., Barreuther, M., Lu, T., Imhof, B. A., Madri, J. A. (1997) Platelet-endothelial cell adhesion molecule-1 (PECAM-1/CD31) tyrosine phosphorylation state changes during vasculogenesis in the murine conceptus. Am. J. Pathol. 150,1523-1530[Abstract]
  30. Levkau, B., Koyama, H., Raines, E. W., Clurman, B. E., Herren, B., Orth, K., Robert, J. M., Ross, R. (1998) Cleavage of p21Cip1/Waf1 and p27 Kip1 mediates apoptosis in endothelial cells through activation of Ckd2: role of caspase cascade. Mol. Cell 1,553-563[Medline]
  31. Levkau, B., Herren, B., Koyama, H., Ross, R., Rains, E. W. (1998) Caspase-mediated cleavage of focal adhesion kinase pp125FAK and disassembly of focal adhesion in human endothelial cells apoptosis. J. Exp. Med. 187,579-586[Abstract/Free Full Text]
  32. Chaudhary, P. M., Eby, M. T., Jasmin, A., Hood, L. (1999) Activation of the c-Jun N-terminal kinase/stress-activated protein kinase pathway by overexpression of caspase-8 and its homologs. J. Biol. Chem. 274,19211-19219[Abstract/Free Full Text]
  33. Bratton, S. B., MacFarlane, M., Cain, K., Cohen, G. M. (2000) Protein complexes activate distinct caspase cascades in death receptor and stress-induced apoptosis. Exp. Cell Res. 256,27-33[Medline]
  34. Muzio, M., Chinnaiyan, A. M., Kischkel, F. C., O’Rourke, K., Shevchenko, A., Ni, J., Scaffidi, C., Bretz, J. D., Zhang, M., Gentz, R., Mann, M., Krammer, P. H., Peter, M. E., Dixit, V. M. (1996) FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95 (Fas/APO-1) death-inducing signaling complex. Cell 85,817-827[Medline]
  35. Tetsu, O., McCormic, F. (1999) ß-Catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature (London) 398,422-426[Medline]
  36. Shtutman, M., Zhurinsky, J., Simcha, I., Alberts, C., D’Amico, M., Pestell, R., Ben-Ze’ev, A. (1999) The cyclin D1 gene is a target of the ß-catenin/LEF-1 pathway. Proc. Natl. Acad. Sci. USA 96,5522-5527[Abstract/Free Full Text]
  37. Peters, J. M., McKay, R. M., Graff, J. M. (1999) Casein kinase I transduces Wnt signals. Nature (London) 401,345-350[Medline]
  38. Sakanaka, C., Leong, P., Xu, L., Harrison, S. D., Williams, L. T. (1999) Casein kinase I{varepsilon} in the Wnt pathway: regulation of ß-catenin function. Proc. Natl. Acad. Sci. USA 96,12548-12552[Abstract/Free Full Text]
  39. Datta, S. R., Brunet, A., Greenberg, M. E. (1999) Cellular survival: a play in three Akts. Gene Dev 13,2905-2927[Free Full Text]
  40. Xia, Z., Dickens, M., Raingeaud, J., Davis, R. J., Greenberg, M. E. (1995) Opposing effects of ERK and JNK-p38 MAP kinases on apoptosis. Science 270,1326-1331[Abstract/Free Full Text]
  41. Cross, T. G., Scheel-Toellner, D., Henriquez, N. V., Deacon, E., Salmon, M., Lord, J. (2000) Serine/threonine protein kinases and apoptosis. Exp. Cell Res. 256,34-41[Medline]
  42. Galan, A., Garcia-Bermejo, M. L., Troyano, A., Vilaboa, N. E., de Blas, E., Kazanietz, M. G., Aller, P. (2000) Stimulation of p38 mitogen-activated protein kinase is an early regulatory event for the cadmium-induced apoptosis in human promonocytic cells. J. Biol. Chem. 275,11418-11424[Abstract/Free Full Text]
  43. Tournier, C., Hess, P., Yang, D. D., Xu, J., Turner, T. K., Nimnual, A., Bar-Sagi, D., Jones, S. N., Flavell, R. A., Davis, R. J. (2000) Requirement of JNK for stress-induced activation of the cytochrome c-mediated death pathway. Science 288,870-874[Abstract/Free Full Text]
  44. Bird, I. N., Taylor, V., Newman, J. P., Spragg, J. H., Simmons, D. L., Salomon, M., Buckley, C. D. (1999) Homophilic PECAM-1(CD31) interactions prevent endothelial cell apoptosis but do not support cell spreading or migration. J. Cell Sci. 112,1989-1997[Abstract]
  45. Massague, J., Pandiella, A. (1993) Membrane-anchored growth factors. Annu. Rev. Biochem. 62,515-541[Medline]
  46. Mohler, K. M., Sleath, P. R., Fitzner, J. N., Cerretti, D. P., Alderson, M., Kerwar, S. S., Torrance, D. S., Otten-Evans, C., Greenstreet, T., Weerawarna, K., Kronheim, S. R., Petersen, M., Gerhart, M., Kozlosky, C. J., Mark, C. J., Black, R. A. (1994) Protection against a lethal dose of endotoxin by an inhibitor of tumor necrosis factor processing. Nature (London) 370,218-220[Medline]
  47. Izumi, Y., Hirata, M., Hasuwa, H., Iwamoto, R., Umata, T., Miyado, K., Tamai, Y., Kurisaki, T., Sehara-Fujisawa, A., Ohno, S., Mekada, E. (1998) A metalloproteinase-disintegrin, MDC9/meltrin-{gamma}/ADAM9 and PKC {delta} are involved in TPA-induced ectodomain shedding of membrane-anchored heparin-binding EGF-like growth factor. EMBO J 17,7260-7272[Medline]
  48. Cabrera, N., Diaz-Rodriguez, E., Becker, E., Martin-Zanca, D., Pandiella, A. (1996) TrkA receptor ectodomain cleavage generates a tyrosine-phosphorylated cell-associated fragment. J. Cell Biol. 132,427-436[Abstract/Free Full Text]
  49. Vecchi, M., Carpenter, G. (1997) Constitutive proteolysis of the ErbB-4 receptor tyrosine kinase by unique, sequential mechanism. J. Cell Biol. 139,995-1003[Abstract/Free Full Text]
  50. McCarthy, M. J., Burrows, R., Bell, S. C., Christie, G., Bell, P. R. F., Brindle, N. P. J. (1999) Potential roles of metalloprotease mediated ectodomain cleavage in signaling by the endothelial receptor tyrosine kinase Tie-1. Lab. Invest. 79,889-895[Medline]
  51. Yabkowitz, R., Meyer, S., Black, T., Elliott, G., Merewether, L. A., Yamane, H. K. (1999) Inflammatory cytokines and vascular endothelial growth factor stimulate the release of soluble Tie receptor from human endothelial cells via metalloprotease activation. Blood 93,1969-1979[Abstract/Free Full Text]
  52. Feehan, C., Darlak, K., Kahn, J., Walcheck, B., Spatola, A. F., Kishimoto, T. K. (1996) Shedding of the T lymphocyte L-selectin adhesion molecule is inhibited by a hydroxamic acid-based protease inhibitor. Identification with an L-selectin-alkaline phosphatase reporter. J. Biol. Chem. 271,7019-70224[Abstract/Free Full Text]
  53. Werb, Z. (1997) ECM and cell surface proteolysis: regulating cellular ecology. Cell 91,439-442[Medline]
  54. Hooper, N. M., Karran, E. H., Turner, A. J. (1997) Membrane protein s