Signal Transduction Pathways Mediated by PECAM-1
New Roles for an Old Molecule in Platelet and Vascular Cell Biology
Recent studies of platelet endothelial cell adhesion molecule-1 (PECAM-1 [CD31])-deficient mice have revealed that this molecule plays an important role in controlling the activation and survival of cells on which it is expressed. In this review, we focus on the complex cytoplasmic domain of PECAM-1 and describe what is presently known about its structure, posttranslational modifications, and binding partners. In addition, we summarize findings that implicate PECAM-1 as an inhibitor of cellular activation via protein tyrosine kinase–dependent signaling pathways, an activator of integrins, and a suppressor of cell death via pathways that depend on damage to the mitochondria. The challenge of future research will be to bridge our understanding of the functional and biochemical properties of PECAM-1 by establishing mechanistic links between signals transduced by the PECAM-1 cytoplasmic domain and discrete cellular responses.
Series Editor: Lawrence Brass
ATVB In Focus Platelet Activation and the Formation of the Platelet Plug
Previous Brief Reviews in this Series:
•Tsai H-M. Deficiency of ADAMTS13 causes thrombotic thrombocytopenic purpura. 2003;23:388–396.
•Quinn MJ, Byzova TV, Qin J, Topol EJ, Plow EF. Integrin αIIbβ3 and its antagonism. 2003;23:945–952.
Platelet endothelial cell adhesion molecule (PECAM)-1 is a 130-kDa type I transmembrane glycoprotein (GP) that was originally described as the endothelial cell equivalent of platelet membrane GPIIa (the integrin β1 subunit),1 a myeloid differentiation antigen,2,3⇓ and the CD31 antigen present on the surface of monocytes, granulocytes, platelets, and endothelial cells.4,5⇓ The common identity of these previously disparate entities as PECAM-1 (CD31) was established in 1990 on its cloning by 3 different groups.6–8⇓⇓ Fifteen years later, more than 1500 articles list PECAM-1 or CD31 in their title or abstract, although many of these can be attributed to the pragmatic use of anti–PECAM-1 antibodies to immunochemically identify endothelial cells in histological sections or to mark angiogenic blood vessels.9 Nonetheless, a great deal has been learned about the cell and molecular biology of PECAM-1 in the blood and vascular cells in which it is expressed, including its chromosomal location, the structure of its rather complex gene, and its adhesive and signaling properties. Because all but the latter have been previously reviewed in detail,10,11⇓ this review will focus on recent studies that have shed light on the ability of PECAM-1 to transmit signals that alter cell adhesion, activation, and survival.
Structure of PECAM-1 and Its Cytoplasmic Domain
Mature PECAM-1 consists of a 574-amino acid extracellular domain comprised of 6 immunoglobulin (Ig)-like homology domains, a 19-residue transmembrane domain, and a 118-amino acid cytoplasmic tail. Extracellular Ig domain 1 contains specialized sites that mediate trans-homophilic interactions between PECAM-1 molecules on adjacent cells12–14⇓⇓ and antibodies whose epitope maps to this region have been shown to be effective in blocking transendothelial migration of leukocytes15–18⇓⇓⇓ and hematopoietic progenitor cells,19 malarial parasite binding,20 and angiogenesis in vivo.21
The cytoplasmic domain of PECAM-1 is complex22 (see Figure 1) and is encoded by 8 short exons that are differentially susceptible to alternative splicing, resulting in generation of mRNA species that encode distinct PECAM-1 isoforms (see Table 1). These species include a transmembraneless, soluble form of PECAM-1 lacking exon 9 that is produced by human umbilical vein endothelial cells and secreted into plasma23 as well as numerous other variants that lack 1 or more cytoplasmic domain exons.22,24–31⇓⇓⇓⇓⇓⇓⇓⇓ Although the existence of few of these PECAM-1 species have been demonstrated at the protein level, several of them (indicated with an asterisk in Table 1) represent relatively abundant mRNA species. Interestingly, whereas exons 1 to 9, 11 to 14, and 16 are all phase 1 exons,22 cytoplasmic domain–encoding exons 10 and 15 are phase 2 and phase 0, respectively, and their removal results in PECAM-1 mRNA species encoding isoforms with a C-terminus that is shorter and has a different amino acid sequence (○, Figure 1). Although no variants lacking exon 10 have been found to date, there are at least 3 different PECAM-1 mRNAs missing exon 15, including 2 very abundant species (Δ15 and Δ14,15) present, for example, in the inner cell mass of mouse blastulas.29 As shown, these 2 PECAM-1 isoforms end with a different sequence, E-N-G-R-L-P, than does full-length PECAM-1. The biologic consequence of this variation is not known.
Tyrosine Phosphorylation of the PECAM-1 Cytoplasmic Domain
Regulated phosphorylation of both serine and tyrosine residues occurs within the PECAM-1 cytoplasmic domain in response to numerous forms of cellular stimulation. As is the case with cytoplasmic domains of other plasma membrane receptors, PECAM-1 cytoplasmic domain phosphorylation regulates assembly of signaling complexes and, in some cases, interactions with various elements of the cytoskeleton.
PECAM-1 can become tyrosine phosphorylated in nearly all vascular cells in which it is expressed. Although low levels of constitutive PECAM-1 tyrosine phosphorylation have been reported in platelets32 and endothelial cells,33 PECAM-1 is generally not found to be tyrosine phosphorylated in cells maintained in a resting state. PECAM-1 tyrosine phosphorylation has been detected in stirred platelets,34 in platelets exposed to pervanadate32 or wheat germ agglutinin,35 and in platelets induced to aggregate via receptors for collagen34–36⇓⇓ or thrombin.32,34,35,37,38⇓⇓⇓⇓ Cross-linking of platelet PECAM-1 itself also results in its tyrosine phosphorylation.36,39,40⇓⇓ In endothelial cells, PECAM-1 tyrosine phosphorylation can be induced by mechanical force applied directly to PECAM-1,41 on adhesion to immobilized PECAM-1,42 fibronectin,42 or collagen,33 or on exposure to wheat germ agglutinin,35 fluid shear stress, or osmotic shock.41,43,44⇓⇓ When it has been compared, the level of PECAM-1 tyrosine phosphorylation seems to be decreased in migrating (or nonconfluent) endothelial cells.33,41⇓ PECAM-1 tyrosine phosphorylation has also been observed in T lymphocytes on cross-linking of PECAM-145 or the T cell antigen receptor45,46⇓ and in RBL-2H3 mast-like cells on adherence to fibronectin46 or on cross-linking of the IgE receptor, FcεRI.47 However, not all stimuli induce tyrosine phosphorylation of PECAM-1, because exposure of endothelial cells to agents that induce membrane lipid hydrolysis or calcium mobilization does not result in tyrosine phosphorylation of PECAM-1,43 nor does exposure of RBL-2H3 cells to the protein kinase C (PKC) agonist phorbol myristate acetate.47
Because PECAM-1 does not possess intrinsic kinase activity, the identity of tyrosine kinases able to phosphorylate PECAM-1 has been the subject of extensive investigation. A fairly large body of evidence obtained from coprecipitation,34,35⇓ in vitro kinase,48,49⇓ and overexpression50 studies supports a role for Src family kinases in PECAM-1 tyrosine phosphorylation. The extent to which individual Src family kinases are redundant in their ability to phosphorylate PECAM-1 may, however, depend on the array of Src kinases that a given cell expresses and the activation conditions to which that cell is exposed.34,51⇓ Inhibition of PECAM-1 tyrosine phosphorylation by Src family kinase inhibitors is reported to be more or less complete, suggesting that a non-Src family tyrosine kinase may also be able to phosphorylate PECAM-1.34,35,41⇓⇓ Candidates include members of the Csk50 and Syk47 families of protein tyrosine kinases, although the latter remains controversial.50,51⇓ It remains to be determined whether, under physiological conditions, PECAM-1 is phosphorylated by both Src and non-Src family tyrosine kinases, how the cell type and its environment influence the activation states of enzymes able to phosphorylate PECAM-1, and whether the tyrosine kinases able to phosphorylate PECAM-1 are dependent on each other’s activity.
The sequence of PECAM-1 cDNA has been determined from 6 species, including human,6 mouse,52 cow,43,53⇓ pig (Genbank accession No. X98505), rat, and dog (unpublished data, 1996). Tyrosine residues are perfectly conserved at positions corresponding to human residues 596, 636, 663, and 686 (yellow-filled circles in Figure 1). Incompletely conserved tyrosine residues include those at positions 638 (histidine in humans, tyrosine in all others), 674 (histidine in all species except dogs), and 701 (phenylalanine in rats). A mutant form of human PECAM-1 in which the tyrosine residues at positions 663 and 686 were replaced with phenylalanine (Y663,686F) failed to become tyrosine phosphorylated in transfected HEK293 cells exposed to pervanadate, revealing that, at least under these conditions, Y663 and Y686 are the sole tyrosine phosphorylation sites in human PECAM-1.54 However, a homologous mutant form of murine PECAM-1 could become tyrosine phosphorylated on overexpression of the Src family kinase p56lck in COS-1 cells.50 Although the tyrosine phosphorylation of Y663,686F PECAM-1 in those cells may have resulted from overexpression of p56lck, it is also possible that COS-1, but not HEK293, cells contain a kinase that can phosphorylate PECAM-1 on tyrosines other than Y663 and Y686, or that mouse PECAM-1 becomes phosphorylated at a site (possibly Tyr638) not present in human PECAM-1.50 Interestingly, Y663 is much less efficiently phosphorylated by either Src or Csk kinases than is Y686,43,50⇓ suggesting that its phosphorylation may be a rate-limiting step in PECAM-1–mediated signal transduction. The identity of the kinases or conditions required for efficient phosphorylation of Y663, therefore, remain important areas of investigation.
Cytoplasmic Proteins That Bind Tyrosine Phosphorylated PECAM-1
The best-characterized structural feature of the PECAM-1 cytoplasmic domain is the presence of 2 distinct immunoreceptor tyrosine-based inhibitory motifs (ITIMs) centered around Y663 and Y686, respectively, and it is the presence of the paired ITIM within the PECAM-1 cytoplasmic domain that led several years ago to its assignment to the Ig-ITIM family of inhibitory receptors.55 Like other members of the Ig-ITIM family, PECAM-1, when tyrosine phosphorylated, is able to recruit Src homology 2 (SH2) domain–containing signaling proteins, which then can initiate signaling pathways, many of which remain to be defined. Proteins reported to be able to associate with the PECAM-1 cytoplasmic domain are summarized in Table 2; however, it is important to recognize that the extent to which these potential binding partners interact with PECAM-1 is likely to be influenced by their relative local concentrations, their relative binding affinities, and the effect of additional posttranslational modifications of the PECAM-1 cytoplasmic domain.56
The protein most commonly reported to interact with the PECAM-1 cytoplasmic domain is the SH2 domain–containing protein-tyrosine phosphatase, SHP-2. It is widely accepted that phosphorylation of the PECAM-1 ITIM tyrosine residues results in both recruitment and activation of SHP-2.37,41,45,46,48,50,54,57–59⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓ Interestingly, in a completely unbiased approach, a bait GST fusion protein containing the tandem SH2 domains of SHP-2 exclusively bound a peptide comprised of PECAM-1 residues 617 to 711 (ie, containing its dual ITIM), when tyrosine-phosphorylated, out of an entire phage display cDNA library.60 The PECAM-1/SHP-2 interaction in cells requires PECAM-1 tyrosine phosphorylation at both positions 663 and 686.50,54⇓ However, in vitro binding studies using the individual SH2 domains of SHP-2 have shown that the N-terminal SH2 domain interacts with high affinity with the N-terminal ITIM of PECAM-1,54,58⇓ whereas the C-terminal SH2 domain of SHP-2 preferentially interacts, albeit with lower affinity, with pY686-containing PECAM-1 phosphopeptide.54 It is possible that the high affinity interaction between the N-terminal SH2 domain of SHP-2 and the pY663-containing ITIM of PECAM-1 may compensate for the inefficiency with which Y663 is phosphorylated under physiological conditions (see above). Furthermore, the level of PECAM-1 tyrosine phosphorylation seems to be enhanced when PECAM-1 is expressed along with a dominant-negative form of SHP-2, suggesting that PECAM-1 is an SHP-2 substrate and that the extent of PECAM-1/SHP-2 complex formation may be regulated by PECAM-1 dephosphorylation mediated by bound SHP-2.50
In contrast with numerous reports documenting PECAM-1/SHP-2 interactions, there are conflicting data concerning interactions between PECAM-1 and the SHP-2–related phosphatase, SHP-1, which has also been shown to associate with PECAM-1, albeit in much lesser amounts than does SHP-2.46,50,57–59⇓⇓⇓⇓⇓ As with SHP-2, both PECAM-1 ITIMs seem to be required to support SHP-1 binding50; however, unlike SHP-2, the affinity between SHP-1 and pY663 is identical to that of pY686, and the affinities of both interactions are significantly lower than those reported for SHP-2.57 In fact, one study was unable to detect an interaction of the individual SH2 domains of SHP-1 with either PECAM-1 phosphopeptide.58 Thus, the functional relevance of the PECAM-1/SHP-1 interaction remains controversial.
Several other SH2 domain–containing proteins have been evaluated for their ability to associate with PECAM-1, including selected members of the Src family,43,48,49⇓⇓ the 5′-inositol phosphatase, SHIP,58 and PLCγ1,58,59⇓ although the latter association is controversial.48 Interestingly, the SH2 domain of SHIP is similar to that present in an adaptor protein known as SAP (SLAM-associated protein), both of which seem to interact preferentially with a unique type of ITIM that has been termed an ITSM.61 SAP binding sites are characterized by the presence of a T or S residue at position −2 relative to the ITIM phosphotyrosine.62 Because the C-terminal ITIM of PECAM-1 contains a T at the −2 position of Y686 (Figure 1), it may be able to bind SAP (Kim Nichols, University of Pennsylvania, unpublished observations, 2002), although the biological conditions under which PECAM-1 and SAP might interact in intact cells are presently not known.
PECAM-1 immunoprecipitates have been shown to contain several phosphoproteins of unknown identity,34,42,49,54,57⇓⇓⇓⇓ and it is likely that some of these associate with the PECAM-1 cytoplasmic domain indirectly or in an ITIM-independent manner. Signal transducers and activators of transcription 3 and 5 likely represent the best known examples of ITIM-independent binding, because their interaction with PECAM-1 seems to require PECAM-1 Y701 but not ITIM residues Y663 or Y686.63 Phosphatidylinositol-3 kinase, on the other hand, may be an example of a signaling molecule that interacts indirectly with PECAM-1, because it could be detected in anti–PECAM-1 mAb-coated microwells incubated with whole neutrophil lysates64 but was not found associated with PECAM-1 in direct binding studies.48,58⇓ The adaptor molecules Grb2 and Gab1 also may interact with tyrosine-phosphorylated PECAM-1 in an indirect manner. Although evidence for an association between PECAM-1 and Grb2 is conflicting,46,58⇓ Gab1 has been shown to colocalize with PECAM-1 and SHP-2 at endothelial cell borders after exposure to fluid shear stress.41 Immunoprecipitation experiments, however, were unable to demonstrate direct binding of Gab1 to PECAM-1.41 Because both Grb265,66⇓ and Gab167 can bind directly to tyrosine-phosphorylated SHP-2, it is possible that PECAM-1/SHP-2/Gab1 and PECAM-1/SHP-2/Grb2 ternary signaling complexes may form in response to certain forms of cellular stimulation. The relevance of these interactions in regulating PECAM-1–mediated cell adhesion, activation, and survival represent an important area of future investigation.
Serine Phosphorylation of the PECAM-1 Cytoplasmic Domain
PECAM-1 serine, but not threonine, residues have been shown to be phosphorylated in resting platelets and endothelial cells, and the level of serine phosphorylation increases 2- to 3-fold on cellular activation.32,68–72⇓⇓⇓⇓⇓ PKC is thought to play a primary role in PECAM-1 serine phosphorylation68,71,73⇓⇓; however, a role for other kinases has been proposed because of the observation that the level of platelet PECAM-1 serine phosphorylation induced by phorbol myristate acetate, which activates only PKC, is lower than that induced by thrombin, which activates PKC and possibly other kinases as well.69 Proposed serine phosphorylation sites in PECAM-1 include S673,73 S620, S670, and S687.43 Although 3 of these (S620, S670, and S687; see orange-filled circles in Figure 1) are well conserved, the sites and consequences of serine phosphorylation within the PECAM-1 cytoplasmic domain remain to be determined.
There is growing evidence that PECAM-1 associates both physically and functionally with the underlying cytoskeleton. Using the operational definition of detergent insolubility as a measure of cytoskeletal connectedness, it was found more than 10 years ago that, in resting platelets, only ≈10% of PECAM-1 partitions with the Triton-insoluble cytoskeleton, which is largely composed of F-actin, filamen (ABP-280), and spectrin.69 During thrombin-induced platelet aggregation, however, more than 60% of total cellular PECAM-1 becomes detergent insoluble, indicating that the degree of PECAM-1 association with the Triton-insoluble cytoskeleton depends on the activation state of the cell.69 In support of this notion, ≈20% to 30% of PECAM-1 was found associated with the cytoskeleton in confluent endothelial cells,73,74⇓ increasing to ≈65% during cell migration.73
Poggi et al,75 using an NK cell model system, were the first to demonstrate a functional relationship between PECAM-1 and the cytoskeleton when they showed that mAb-induced cross-linking of PECAM-1 on the cell surface could induce cell spreading and cytoskeletal rearrangement. Shortly thereafter, Matsumura et al76 showed that addition to endothelial cells of a combination of VE-cadherin–specific and PECAM-1–specific antibodies resulted in reorganization of F-actin filaments into discrete foci. Although less specific than using PECAM-1–specific reagents, addition of wheat germ agglutinin, a multivalent lectin that binds to and cross-links PECAM-1 (as well as a host of other cell surface receptors), has also been shown to increase actin assembly at the cell periphery.35 Taken together, it would seem that PECAM-1 is able to bind to and coordinate the assembly of F-actin filaments, especially in association with changes in cell shape or during cell migration.
Although the molecules linking PECAM-1 and the actin cytoskeleton have not been established with certainty, β-catenin and γ-catenin (also known as plakoglobin) seem to be likely candidates. The catenins are scaffolding proteins normally involved in anchoring a class of adherens junctional proteins known as cadherins to the cortical actin cytoskeleton. Although early studies did not find an association between α-, β-, or γ-catenin and PECAM-1,77 Matsumura et al76 found that β-catenin could be coimmunoprecipitated with PECAM-1 using special cell homogenization conditions that improved the solubility of the underlying membrane skeleton. The authors proposed that a functional adherens junctional complex of PECAM-1/β-catenin/F-actin, perhaps in conjunction with VE-cadherin, might regulate processes such as endothelial cell tube formation. The finding that anti–PECAM-1 antibodies markedly inhibit the ability of endothelial cells to organize and form 3-dimensional networks in Matrigel78 also support this concept. Ilan et al79 went on to characterize the molecular requirements for PECAM-1/β-catenin complex formation, which include tyrosine phosphorylation of β-catenin. This interaction seems to be ITIM-independent, because tyrosine-phosphorylated β-catenin bound similarly to wild-type, Y663F, and Y686F forms of PECAM-1. Because SHP-2 and β-catenin seemed to bind different sites on the PECAM-1 cytoplasmic domain, the authors suggested that SHP-2, when recruited to PECAM-1/β-catenin complexes, regulates the tyrosine phosphorylation state of β-catenin. The physiological importance of β-catenin tyrosine phosphorylation, however, remains to be established. In addition to β-catenin, PECAM-1 also has been reported to associate with γ-catenin.73 The molecular requirements for γ-catenin association with PECAM-1 differ from those of β-catenin, however, because γ-catenin does not need to be tyrosine phosphorylated to bind. Interestingly, PKC-mediated serine phosphorylation of the PECAM-1 cytoplasmic domain, as might occur early in the platelet activation process or during endothelial cell migration, was found to inhibit PECAM-1/γ-catenin interactions. Taken together, these data suggest that dynamic interactions between PECAM-1 and the catenins may be responsible for anchoring PECAM-1 to the cytoskeleton. The relationship between such cytoskeletal connections and the ability of PECAM-1 to regulate integrin activation, endothelial responses to fluid shear stress, protection from phagocytosis, or cell survival (all discussed below) are important issues that remain to be addressed.
ITIM-Mediated Inhibitory Function
By virtue of their ability to recruit and activate protein-tyrosine or inositol phosphatases, ITIM-containing inhibitory receptors are thought to function primarily by counteracting signal transduction pathways initiated by activating receptors that recruit, via their immunoreceptor tyrosine-based activating motifs (ITAMs), protein tyrosine kinases.80 Inhibitory receptors, therefore, raise the threshold for cellular activation and regulate ITAM receptor–mediated cellular activation events. PECAM-1 is the only known ITIM-containing receptor on human or murine platelets, and, because the most prominent PECAM-1–binding protein, SHP-2, is capable of transmitting both stimulatory and inhibitory signals into the cell,81 determining whether PECAM-1 exerts positive or negative regulatory effects has been the subject of much investigation.
The first evidence that PECAM-1 could function as an inhibitory receptor came from studies in which PECAM-1 was artificially forced into proximity with selected activating receptors. This was done by cross-linking antibodies bound to PECAM-1 with antibodies bound to ITAM-containing antigen receptors on lymphocytes. On coligation with the T cell45 or B cell51,59⇓ receptor, PECAM-1 was found to attenuate antigen receptor–induced signaling and cellular responses in a manner that depended on the integrity of the PECAM-1 ITIMs and that either required SHP-251 or preferred SHP-2 over SHP-1.59
Evidence, albeit less direct, for the inhibitory function of PECAM-1 has also been obtained by examining the effects of PECAM-1 cross-linking on collagen-induced platelet activation. Collagen binds to platelets via the integrin α2β1, but subsequent signal transduction occurs primarily via GPVI, which exists noncovalently associated within the plane of the plasma membrane with the ITAM-bearing FcRγ-chain.82 PECAM-1/SHP-2 complexes formed as a result of PECAM-1 cross-linking36,39,40⇓⇓ should, in theory, be able to inhibit signaling by nearby GPVI/FcRγ chain complexes. Indeed, several in vitro studies have demonstrated that activation of PECAM-1 via mAb-induced PECAM-1 cross-linking inhibits Ca2+ mobilization,40 granule secretion,40 aggregation,36,40⇓ and thrombus formation36 in collagen-activated platelets. The extent to which PECAM-1 inhibits responses of platelets to agonists that bind to non-ITAM–bearing receptors is, however, as yet unclear. One study found that mAb-induced PECAM-1 cross-linking inhibited platelet activation induced by binding of thrombin to the G-protein–coupled receptor, PAR1, which is not thought to be regulated by ITIMs; however, this study did not rule out nonspecific heterologous desensitization as the mechanism of inhibition.40 In another study, the anti–PECAM-1 mAb, AAP2, inhibited platelet aggregation induced by all agonists examined, including collagen, ADP, epinephrine, and thrombin; however, this study did not discriminate between inhibition of PECAM-1–mediated signaling versus adhesion.83
The notion that PECAM-1 inhibits specifically the action of ITAM-bearing agonist receptors has recently gained support from studies that have had the opportunity to examine blood cell function in PECAM-1–deficient mice. Compared with wild-type cells, PECAM-1–deficient B cells84 and mast cells85 fail to regulate signaling through the ITAM-bearing B cell receptor and FcεRI, respectively. Similar hyperactive responses have been observed in PECAM-1–deficient platelets, which exhibit enhanced aggregation and granule secretion responses to GPVI/FcRγ chain–specific agonists,36,86⇓ whereas PECAM-1 deficiency has no measurable effect on in vitro platelet responses to agonists that bind to G protein–coupled receptors for ADP36,87⇓ or thrombin.36,86⇓ These data suggest that, in vivo, platelet PECAM-1 plays a major role in regulating signaling pathways of ITAM-containing, but not G protein–coupled, receptors. Relative to its regulation of ITAM-mediated cellular activation, however, it is important to note that PECAM-1 inhibitory signaling in platelets can be easily overcome by strong stimulation of the GPVI/FcRγ chain collagen receptor.86 Thus, in situations where endothelial damage is severe enough to expose a high concentration of underlying extracellular matrix, PECAM-1 will likely not be able to inhibit platelet thrombus formation. This is a good thing. This concept may explain the discrepancy between the findings of Vollmar et al,88 who found that PECAM-1–deficient and wild-type mice did not differ in either the rate or extent of thrombus formation in blood vessels damaged by exposure to laser light–induced injury to the endothelium, versus those of Rosenblum and colleagues,89,90⇓ who reported that anti–PECAM-1 mAbs could inhibit platelet adhesion and aggregation over mildly injured endothelial cell beds. Studies using mild forms of vascular injury in PECAM-1–deficient mice might be expected to yield similar results.
As these and other studies in PECAM-1–deficient mice begin to emerge, opportunities to resolve several important issues are likely to be realized. First, it is important to determine the extent to which formation of PECAM-1/SHP-2 complexes is required to modulate platelet, B cell, and mast cell activation in vivo. Expression of an ITIM-less form of PECAM-1 in vivo would go a long way toward answering this question. If SHP-2 binding to the PECAM-1 ITIMs is important for the inhibitory activity of PECAM-1 in vivo, specific roles for SHP-2 will have to be resolved. One possibility is that PECAM-1 normally recruits SHP-2 and triggers its phosphatase activity, resulting in dephosphorylation of nearby components of ITAM-dependent signal transduction pathways and diminished cellular responsiveness. In cells expressing an ITIM-less form of PECAM-1, dephosphorylation of signal transduction pathway components would be delayed, resulting in augmented cellular responsiveness. An alternative possibility is that SHP-2 normally functions as a positive adapter or activating protein in growth factor receptor signaling pathways91 from which PECAM-1 sequesters SHP-2 away. ITIM-less PECAM-1 would fail to sequester SHP-2, again resulting in augmented cellular responses. Once the mechanism by which PECAM-1 exerts its inhibitory function is more precisely understood, it may be possible to exploit it to block pathologic cellular activation (eg, to inhibit thrombosis, autoimmunity, or hypersensitivity) or constrain it to enhance desirable cell activation (eg, to control bleeding or overcome immunosuppression).
More than 10 years ago, Tanaka et al92 noticed that addition of anti–PECAM-1 mAbs to human T lymphocytes enhanced their ability to bind to immobilized β1 integrin substrates, such as fibronectin and VCAM-1, and proposed that PECAM-1 could act as an integrin-function modulator. Since that time, there have been numerous studies demonstrating that addition of anti–PECAM-1 mAbs to the cell surface results in upregulation of β1,93 β2,94–96⇓⇓ and β339,97⇓ integrin function, leading to speculation that PECAM-1 engagement might influence diverse integrin-mediated processes, such as thrombosis, cell migration, or transendothelial leukocyte extravasation. Although the precise mechanism by which PECAM-1 modulates integrin function is not known, several clues have begun to emerge. First, PECAM-1–mediated integrin activation seems to require an intact PECAM-1 cytoplasmic domain and its 2 ITIMs98 and that PECAM-1 be induced to form lateral oligomers within the plane of the plasma membrane.99 What subsequently occurs to effect integrin activation is not clear, but recent studies by Reedquist et al98 suggest that small GTPases may be involved. They found that addition of anti–PECAM-1 mAbs to Jurkat T cells selectively activated Rap1, which, together with its exchange factor CalDAG-GEFI, has been increasingly implicated in integrin activation,100,101⇓ perhaps by inducing integrin clustering (avidity modulation) or by inducing conformational changes in the integrin itself (affinity modulation). Support for both of these mechanisms of action, in fact, exists, because Poggi et al75 found that addition of anti–PECAM-1 mAbs to human NK cells resulted in actin rearrangements and recruitment of talin, a known integrin-binding102 and -activating103 protein, to membrane ruffles, whereas Varon et al39 found that certain anti-PECAM-1 mAbs, when bound to platelets, induce exposure of ligand-induced binding sites (so-called LIBS epitopes) on the integrin αIIbβ3. Future studies using recently developed Rap1-deficient mice may help to clarify the relationships between Rap1 activation and cytoskeletal rearrangements, integrin clustering, and integrin conformational changes in PECAM-1–mediated modulation of integrin function.
In addition to affecting integrin activation, PECAM-1 may also regulate integrin trafficking. A recent report by Dangerfield et al104 provides compelling evidence that PECAM-1/PECAM-1 homophilic interactions may be required in vivo for redistributing α6β1 from neutrophil intracellular granules to the plasma membrane during the process of transendothelial migration,105 thereby facilitating α6β1-mediated neutrophil migration through the laminin- and collagen-rich perivascular basement membrane. Interestingly, endothelial cell surface PECAM-1 seems to serve as a passive ligand for neutrophil PECAM-1, which after engagement leads to neutrophil signal transduction and subsequent integrin activation.106 The signaling pathway that regulates PECAM-1–mediated integrin redistribution remains to be elucidated; however, given the reported ability of PECAM-1 to activate Rap1, it is tempting to speculate that PECAM-1 might similarly be activating 1 or more Rab proteins107—related members of the Ras superfamily—to effect integrin-containing granule trafficking and membrane fusion during neutrophil extravasation. A summary of molecular mechanisms by which PECAM-1 might amplify integrin-mediated cell adhesion is shown in Figure 2.
In addition to its role in vascular cell adhesion and signaling, there is growing evidence that PECAM-1 may be able to transduce signals that suppress programmed cell death. In 1999, Noble et al108 reported that PECAM-1/PECAM-1 homophilic interactions between monocytes and endothelial cells reduced apoptotic endothelial cell death after serum deprivation and speculated that PECAM-1 homophilic interactions might lead to the transmission of prosurvival signals. Similar findings have been reported by Bird et al42 and by Evans et al,109 each of whom found that endothelial cells could be protected from serum deprivation–induced cell death if first bound to PECAM-1/IgG or treated with an anti–PECAM-1 monoclonal antibody. Although the molecular mechanisms by which PECAM-1 exerts its cytoprotective effects are not yet known, Gao et al110 recently reported preliminary findings showing that PECAM-1 can inhibit cytochrome c release from mitochondria after exposure of cells to a wide range of cytotoxic stimuli that activate Bax, a proapoptotic pore-forming member of the Bcl-2 family that plays a central role in mitochondria-dependent apoptosis.111 Interestingly, both extracellular homophilic binding function and intact cytoplasmic ITIMs seem to be required for PECAM-1 to suppress programmed cell death, because neither homophilic binding-crippled K89A14 nor ITIM-less Y663,686F54 mutant forms of PECAM-1 were able to protect cells from Bax-overexpression–induced apoptosis. Taken together with previous reports, these data suggest that signals emanating from the PECAM-1 extracellular domain initiate association of 1 or more cytosolic signaling molecules with the PECAM-1 cytoplasmic tail, resulting in the transmission of prosurvival signals that suppress the mitochondria-dependent, Bax-mediated intrinsic apoptotic pathway. Candidate proteins/pathways that might be deserving of additional investigation for their potential to support PECAM-1–mediated cell survival are illustrated in Figure 3.
Other Functions Mediated by the PECAM-1 Cytoplasmic Domain
As noted above, the cytoplasmic ITIMs of PECAM-1 become tyrosine phosphorylated in response to numerous forms of cellular activation, including thrombin- or collagen-induced platelet aggregation, exposure to oxidative injury, and as a consequence of T or B cell receptor and mast cell Fcε receptor cross-linking. In each case, tyrosine phosphorylation of the PECAM-1 cytoplasmic domain results in recruitment and activation of the protein-tyrosine phosphatase, SHP-2. The downstream effectors of the PECAM-1/SHP-2 signaling complex, however, have not yet been described in any of these experimental systems. Progress in this area has recently been made by Osawa et al,41 who discovered that formation of the PECAM-1/SHP-2 signaling complex is required for extracellular signal-regulated kinase (Erk) to become activated when endothelial cells are subjected to fluid shear stress. Erk is an early shear stress response element that regulates transcription of multiple genes, some of which are thought to be involved in atherogenesis, and an earlier study had shown the potential for certain PECAM-1 isoforms to activate Erk.112 Interestingly, the signaling pathway from PECAM-1/SHP-2 to Erk activation seems to involve several steps, including (1) shear-induced perturbation of the plasma membrane, causing (2) activation of one or more tyrosine kinases, which (3) phosphorylate PECAM-1 cytoplasmic ITIM tyrosine residues, resulting in (4) recruitment of SHP-2 and Gab1, a multisite PH-domain–containing adaptor protein involved in growth factor–induced ERK activation,113 initiating (5) SHP-2/Gab-mediated activation of Erk, probably via the ability of Gab to associate with Shc,114 which in turn recruits the Grb2-Sos Ras activation complex. Taken together, these data implicate PECAM-1 as a mechanoresponsive cell-surface receptor that, together with SHP-2 and Gab1, regulates Erk-mediated endothelial responses to fluid shear stress.
One of the more intriguing and novel functions of PECAM-1–mediated signal transduction relates to the ability of PECAM-1 to send “leave me alone” signals from healthy cells to potentially hostile cellular adversaries. Thus, Brown et al115 recently reported that homophilic interactions between PECAM-1–positive macrophages and PECAM-1-positive leukocytes results in the transmission of signals that facilitate either active detachment if the leukocyte is viable or phagocytic ingestion if the leukocyte is apoptotic. The nature of the PECAM-1–mediated release versus tethering signal has not been fully explored, but it is known to require PECAM-1 cytoplasmic ITIMs, as neither a Y663,686F- nor a glycophosphatidylinositol-linked form of PECAM-1 was able to mediate attachment of apoptotic cells or detachment of viable ones. SHP-2 binding to PECAM-1 was also associated with release signaling, although a causative relationship was not established. It will be interesting in the future to determine the relationships, if any, between PECAM-1–mediated detachment signals, PECAM-1–mediated integrin activation, and PECAM-1–mediated inhibitory signaling.
Contributions from many laboratories over the past 15 years have revealed that PECAM-1 serves several distinct roles in the biology of blood and vascular cells, some no doubt related to its adhesive properties, and others as a result of it ability to transduce signals into cells. The biochemical pathways through which PECAM-1 inhibits cellular activation mediated by ITAM-bearing agonist receptors, modulates integrin function, regulates vascular integrity, and controls cell survival, however, are just now beginning to be defined. Studies in mice and other well-defined experimental systems using selected PECAM-1 variants lacking either adhesive or signaling capacity, or both, will be required to elucidate the manner by which PECAM-1 functions in thrombosis, inflammation, cell survival, and the immune response.
- Received February 25, 2003.
- Accepted March 24, 2003.
- ↵van Mourik JA, Leeksma OC, Reinders JH, de Groot PG, Zandbergen-Spaargaren J. Vascular endothelial cells synthesize a plasma membrane protein indistinguishable from platelet membrane glycoprotein IIa. J Biol Chem. 1985; 260: 11300–11306.
- ↵Ohto H, Maeda H, Shibata Y, Chen R-F, Qzaki Y, Higashihara M, Takeuchi A, Tohyama H. A novel leukocyte differentiation antigen: two monoclonal antibodies TM2 and TM3 define a 120-kd molecule present on neutrophils, monocytes, platelets, and activated lymphoblasts. Blood. 1985; 66: 873–881.
- ↵Goyert SM, Ferrero EM, Seremetis SV, Winchester RJ, Silver J, Mattison AC. Biochemistry and expression of myelomonocytic antigens. J Immunol. 1986; 137: 3909–3914.
- ↵McMichael AJ. White Cell Differentiation Antigens. Oxford: Oxford University Press; 1987.
- ↵Newman PJ, Berndt MC, Gorski J, White GC, Lyman S, Paddock C, Muller WA. PECAM-1 (CD31) cloning and relation to adhesion molecules of the immunoglobulin gene superfamily. Science. 1990; 247: 1219–1222.
- ↵Simmons DL, Walker C, Power C, Pigott R. Molecular cloning of CD31, a putative intercellular adhesion molecule closely related to carcinoembryonic antigen. J Exp Med. 1990; 171: 2147–2152.
- ↵Stockinger H, Gadd SJ, Eher R, Majdic O, Schreiber W, Kasinrerk W, Strass B, Schnabl E, Knapp W. Molecular characterization and functional analysis of the leukocyte surface protein CD31. J Immunol. 1990; 145: 3889–3897.
- ↵Sun Q-H, DeLisser HM, Zukowski MM, Paddock C, Albelda SM, Newman PJ. Individually distinct Ig homology domains in PECAM-1 regulate homophilic binding and modulate receptor affinity. J Biol Chem. 1996; 271: 11090–11098.
- ↵Sun J, Williams J, Yan H-C, Amin KM, Albelda SM, DeLisser HM. Platelet endothelial cell adhesion molecule-1 (PECAM-1) homophilic adhesion is mediated by immunoglobulin-like domains 1 and 2 and depends on the cytoplasmic domain and the level of surface expression. J Biol Chem. 1996; 271: 18561–18570.
- ↵Newton JP, Buckley CD, Jones EY, Simmons DL. Residues on both faces of the first immunoglobulin fold contribute to homophilic binding sites on PECAM-1/CD31. J Biol Chem. 1997; 272: 20555–20563.
- ↵Vaporciyan AA, DeLisser HM, Yan H-C, Mendiguren II, Thom SR, Jones ML, Ward PA, Albelda SM. Involvement of platelet endothelial cell adhesion molecule-1 in neutrophil recruitment in vivo. Science. 1993; 262: 1580–1582.
- ↵Muller WA, Weigl SA, Deng X, Phillips DM. PECAM-1 is required for transendothelial migration of leukocytes. J Exp Med. 1993; 178: 449–460.
- ↵Bogen S, Pak J, Garifallou M, Deng X, Muller WA. Monoclonal antibody to murine PECAM-1 (CD31) blocks acute inflammation in vivo. J Exp Med. 1994; 179: 1059–1064.
- ↵Liao F, Huynh HK, Eiroa A, Greene T, Polizzi E, Muller WA. Migration of monocytes across endothelium and passage through extracellular matrix involve separate molecular domains of PECAM-1. J Exp Med. 1995; 182: 1337–1343.
- ↵Yong KL, Watts M, Shaun TN, Sullivan A, Ings S, Linch DC. Transmigration of CD34+ cells across specialized and nonspecialized endothelium requires prior activation by growth factors and is mediated by PECAM-1 (CD31). Blood. 1998; 91: 1196–1205.
- ↵Kirschbaum NE, Gumina RJ, Newman PJ. Organization of the gene for human platelet/endothelial cell adhesion molecule-1 (PECAM-1) reveals alternatively spliced isoforms and a functionally complex cytoplasmic domain. Blood. 1994; 84: 4028–4037.
- ↵Goldberger A, Middleton KA, Oliver JA, Paddock C, Yan H-C, DeLisser HM, Albelda SM, Newman PJ. Biosynthesis and processing of the cell adhesion molecule PECAM-1 includes production of a soluble form. J Biol Chem. 1994; 269: 17183–17191.
- ↵Newman PJ, Doers MP, Gorski J. Molecular cloning of a 130 kD membrane glycoprotein expressed on human platelets, umbilical vein endothelial cells, and human erythroleukemia (HEL) cells. J Cell Biol. 105;53a: 1987.
- ↵Baldwin HS, Shen HM, Yan H-C, DeLisser HM, Chung A, Mickanin C, Trask T, Kirschbaum N, Newman PJ, Albelda SM, Buck CA. Platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31): alternatively spliced, functionally distinct isoforms expressed during early mammalian cardiovascular development. Development. 1994; 120: 2539–2553.
- ↵Yan H-C, Baldwin HS, Sun J, Buck CA, Albelda SM, DeLisser HM. Alternative splicing of a specific cytoplasmic exon alters the binding characteristics of murine platelet/endothelial cell adhesion molecule-1 (PECAM-1). J Biol Chem. 1995; 270: 23672–23680.
- ↵Wang Y, Su X, Sorenson CM, Sheibani N. The tissue specific distribution of alternatively spliced human PECAM-1 Isoforms. Am J Physiol Heart Circ Physiol. 2003; 284: H1008–H1017.
- ↵Modderman PW, von dem Borne AEGKr, Sonnenberg A. Tyrosine phosphorylation of P-selectin in intact platelets and in a disulfide-linked complex with immunoprecipitated pp60c-src. Biochem J. 1994; 299: 613–621.
- ↵Lu TT, Yan LG, Madri JA. Integrin engagement mediates tyrosine dephosphorylation on platelet-endothelial cell adhesion molecule 1. Proc Natl Acad Sci U S A. 1996; 93: 11808–11813.
- ↵Cicmil M, Thomas JM, Sage T, Barry FA, Leduc M, Bon C, Gibbins JM. Collagen, convulxin, and thrombin stimulate aggregation-independent tyrosine phosphorylation of CD31 in platelets: evidence for the involvement of Src family kinases. J Biol Chem. 2000; 275: 27339–27347.
- ↵Jones KL, Hughan SC, Dopheide SM, Farndale RW, Jackson SP, Jackson DE. Platelet endothelial cell adhesion molecule-1 is a negative regulator of platelet-collagen interactions. Blood. 2001; 98: 1456–1463.
- ↵Jackson DE, Ward CM, Wang R, Newman PJ. The protein-tyrosine phosphatase SHP-2 binds PECAM-1 and forms a distinct signaling complex during platelet aggregation: evidence for a mechanistic link between PECAM-1 and integrin-mediated cellular signaling. J Biol Chem. 1997; 272: 6986–6993.
- ↵Varon D, Jackson DE, Shenkman B, Dardik R, Tamarin I, Savion N, Newman PJ. Platelet/endothelial cell adhesion molecule-1 serves as a co-stimulatory agonist receptor that modulates integrin-dependent adhesion and aggregation of human platelets. Blood. 1998; 91: 500–507.
- ↵Cicmil M, Thomas JM, Leduc M, Bon C, Gibbins JM. Platelet endothelial cell adhesion molecule-1 signaling inhibits the activation of human platelets. Blood. 2002; 99: 137–144.
- ↵Osawa M, Masuda M, Kusano K, Fujiwara K. Evidence for a role of platelet endothelial cell adhesion molecule-1 in endothelial cell mechanosignal transduction: is it a mechanoresponsive molecule? J Cell Biol. 2002; 158: 773–785.
- ↵Bird IN, Taylor V, Newton JP, Spragg JH, Simmons DL, Salmon M, Buckley CD. Homophilic PECAM-1(CD31) interactions prevent endothelial cell apoptosis but do not support cell spreading or migration. J Cell Sci. 1999; 112: 1989–1997.
- ↵Newton-Nash DK, Newman PJ. A new role for platelet-endothelial cell adhesion molecule-1 (CD31): Inhibition of TCR-mediated signal transduction. J Immunol. 1999; 163: 682–688.
- ↵Sagawa K, Kimura T, Swieter M, Siraganian RP. The protein-tyrosine phosphatase SHP-2 associates with tyrosine-phosphorylated adhesion molecule PECAM-1 (CD31). J Biol Chem. 1997; 272: 31086–31091.
- ↵Sagawa K, Swaim W, Zhang J, Unsworth E, Siraganian RP. Aggregation of the high affinity IgE receptor results in the tyrosine phosphorylation of the surface adhesion protein PECAM-1 (CD31). J Biol Chem. 1997; 272: 13412–13418.
- ↵Lu TT, Barreuther M, Davis S, Madri JA. 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. 1997; 272: 14442–14446.
- ↵Cao MY, Huber M, Beauchemin N, Famiglietti J, Albelda SM, Veillette A. Regulation of mouse PECAM-1 tyrosine phosphorylation by the Src and Csk families of protein-tyrosine kinases. J Biol Chem. 1998; 273: 15765–15772.
- ↵Newman DK, Hamilton C, Newman PJ. Inhibition of antigen-receptor signaling by platelet endothelial cell adhesion molecule-1 (CD31) requires functional ITIMs, SHP-2, and p56lck. Blood. 2001; 97: 2351–2357.
- ↵Xie Y, Muller WA. Molecular cloning and adhesive properties of murine platelet/endothelial cell adhesion molecule 1. Proc Natl Acad Sci U S A. 1993; 90: 5569–5573.
- ↵Stewart RJ, Kashour TS, Marsden PA. Vascular endothelial platelet endothelial adhesion molecule-1 (PECAM-1) expression is decreased by TNF-alpha and IFN-gamma: evidence for cytokine-induced destabilization of messenger ribonucleic acid transcripts in bovine endothelial cells. J Immunol. 1996; 156: 1221–1228.
- ↵Jackson DE, Kupcho KR, Newman PJ. Characterization of phosphotyrosine binding motifs in the cytoplasmic domain of platelet/endothelial cell adhesion molecule-1 (PECAM-1) that are required for the cellular association and activation of the protein-tyrosine phosphatase, SHP-2. J Biol Chem. 1997; 272: 24868–24875.
- ↵Hua CT, Gamble JR, Vadas MA, Jackson DE. Recruitment and activation of SHP-1 protein-tyrosine phosphatase by human platelet endothelial cell adhesion molecule-1 (PECAM-1). J Biol Chem. 1998; 273: 28332–28340.
- ↵Henshall TL, Jones KL, Wilkinson R, Jackson DE. Src homology 2 domain-containing protein-tyrosine phosphatases, SHP-1 and SHP-2, are required for platelet endothelial cell adhesion molecule-1/CD31-mediated inhibitory signaling. J Immunol. 2001; 166: 3098–3106.
- ↵Shlapatska LM, Mikhalap SV, Berdova AG, Zelensky OM, Yun TJ, Nichols KE, Clark EA, Sidorenko SP. CD150 association with either the SH2-containing inositol phosphatase or the SH2-containing protein tyrosine phosphatase is regulated by the adaptor protein SH2D1A. J Immunol. 2001; 166: 5480–5487.
- ↵Pellegatta F, Chierchia SL, Zocchi MR. Functional association of platelet endothelial cell adhesion molecule-1 and phosphoinositide 3-kinase in human neutrophils. J Biol Chem. 1998; 273: 27768–27771.
- ↵Li W, Nishimura R, Kashishian A, Batzer AG, Kim WJH, Cooper JA, Schlessinger J. A new function for a phosphotyrosine phosphatase: linking GRB2-Sos to a receptor tyrosine kinase. Mol Cell Biol. 1994; 14: 509–517.
- ↵Bennett AM, Tang TL, Sugimoto S, Walsh CT, Neel BG. Protein-tyrosine-phosphatase SHPTP2 couples platelet-derived growth factor receptor β to Ras. Proc Natl Acad Sci U S A. 1994; 91: 7335–7339.
- ↵Cunnick JM, Dorsey JF, Munoz-Antonia T, Mei L, Wu J. Requirement of SHP2 binding to Grb2-associated binder-1 for mitogen-activated protein kinase activation in response to lysophosphatidic acid and epidermal growth factor. J Biol Chem. 2000; 275: 13842–13848.
- ↵Zehnder JL, Hirai K, Shatsky M, McGregor JL, Levitt LJ, Leung LLK. The cell adhesion molecule CD31 is phosphorylated after cell activation: down-regulation of CD31 in activated T lymphocytes. J Biol Chem. 1992; 267: 5243–5249.
- ↵Newman PJ, Hillery CA, Albrecht R, Parise LV, Berndt MC, Mazurov AV, Dunlop LC, Zhang J, Rittenhouse SE. Activation-dependent changes in human platelet PECAM-1: phosphorylation, cytoskeletal association, and surface membrane redistribution. J Cell Biol. 1992; 119: 239–246.
- ↵Rattan V, Shen Y, Sultana C, Kumar D, Kalra VK. Glucose-induced transmigration of monocytes is linked to phosphorylation of PECAM-1 in cultured endothelial cells. Am J Physiol. 1996; 271: E711–E717.
- ↵Sultana C, Shen Y, Johnson C, Kalra VK. Cobalt chloride-induced signaling in endothelium leading to the augmented adherence of sickle red blood cells and transendothelial migration of monocyte-like HL-60 cells is blocked by PAF-receptor antagonist. J Cell Physiol. 1999; 179: 67–78.
- ↵Lastres P, Almendro N, Bellon T, Lopez-Guerrero JA, Eritja R, Bernabeu C. Functional regulation of platelet/endothelial cell adhesion molecule-1 by TGF-β1 in promonocytic U-937 cells. J Immunol. 1994; 153: 4206–4218.
- ↵Ilan N, Cheung L, Pinter E, Madri JA. Platelet-endothelial cell adhesion molecule-1 (CD31), a scaffolding molecule for selected catenin family members whose binding is mediated by different tyrosine and serine/threonine phosphorylation. J Biol Chem. 2000; 275: 21435–21443.
- ↵Ayalon O, Sabanai H, Lampugnani MG, Dejana E, Geiger B. Spatial and temporal relationships between cadherins and PECAM-1 in cell-cell junctions of human endothelial cells. J Cell Biol. 1994; 126: 247–258.
- ↵Matsumura T, Wolff K, Petzelbauer P. Endothelial cell tube formation depends on cadherin 5 and CD31 interactions with filamentous actin. J Immunol. 1997; 158: 3408–3416.
- ↵Lampugnani MG, Corada M, Caveda L, Breviario F, Ayalon O, Geiger B, Dejana E. The molecular organization of endothelial cell to cell junctions: differential association of plakoglobin, beta-catenin, and alpha-catenin with vascular endothelial cadherin (VE-cadherin). J Cell Biol. 1995; 129: 203–217.
- ↵Sheibani N, Newman PJ, Frazier WA. Thrombospondin-1, a natural inhibitor of angiogenesis, regulates Platelet-Endothelial Cell Adhesion Molecule-1 expression and endothelial cell morphogenesis. Mol Biol Cell. 1997; 8: 1329–1341.
- ↵Ilan N, Mahooti S, Rimm DL, Madri JA. PECAM-1 (CD31) functions as a reservoir for and a modulator of tyrosine-phosphorylated β-catenin. J Cell Sci. 1999; 112: 3005–3014.
- ↵Ravetch JV, Lanier LL. Immune inhibitory receptors. Science. 2000; 290: 84–89.
- ↵Wu XW, Lian EC. Binding properties and inhibition of platelet aggregation by a monoclonal antibody to CD31 (PECAM-1). Arterioscler Thromb Vasc Biol. 1997; 17: 3154–3158.
- ↵Wilkinson R, Lyons AB, Roberts D, Wong MX, Bartley PA, Jackson DE. Platelet endothelial cell adhesion molecule-1 (PECAM-1/CD31) acts as a regulator of B-cell development, B-cell antigen receptor (BCR)-mediated activation, and autoimmune disease. Blood. 2002; 100: 184–193.
- ↵Wong MX, Roberts D, Bartley PA, Jackson DE. Absence of platelet endothelial cell adhesion molecule-1 (CD31) leads to increased severity of local and systemic IgE-mediated anaphylaxis and modulation of mast cell activation. J Immunol. 2002; 168: 6455–6462.
- ↵Patil S, Newman DK, Newman PJ. Platelet endothelial cell adhesion molecule-1 serves as an inhibitory receptor that modulates platelet responses to collagen. Blood. 2001; 97: 1727–1732.
- ↵Duncan GS, Andrew DP, Takimoto H, Kaufman SA, Yoshida H, Spellberg J, Luis de la Pompa J, Elia A, Wakeham A, Karan-Tamir B, Muller WA, Senaldi G, Zukowski MM, Mak TW. 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. 1999; 162: 3022–3030.
- ↵Rosenblum WI, Murata S, Nelson GH, Werner PK, Ranken R, Harmon RC. Anti-CD31 delays platelet adhesion/aggregation at sites of endothelial injury in mouse cerebral arterioles. Am J Clin Pathol. 1994; 145: 33–36.
- ↵Rosenblum WI, Nelson GH, Wormley B, Werner P, Wang J, Shih CC-Y. Role of Platelet-endothelial cell adhesion molecule (PECAM) in platelet adhesion/aggregation over injured but not denuded endothelium in vivo and ex vivo. Stroke. 1996; 27: 709–711.
- ↵Bennett AM, Hausdorff SF, O’Reilly AM, Freeman RM, Neel BG. Multiple requirements for SHPTP2 in epidermal growth factor-mediated cell cycle progression. Mol Cell Biol. 1996; 16: 1189–1202.
- ↵Tanaka Y, Albelda SM, Horgan KJ, Van Seventer GA, Shimizu Y, Newman W, Hallam J, Newman PJ, Buck CA, Shaw S. CD31 expressed on distinctive T cell subsets is a preferential amplifier of β1 integrin-mediated adhesion. J Exp Med. 1992; 176: 245–253.
- ↵Leavesley DI, Oliver JM, Swart BW, Berndt MC, Haylock DN, Simmons PJ. Signals from platelet/endothelial cell adhesion molecule enhance the adhesive activity of the very late antigen-4 integrin of human CD34+ hemopoietic progenitor cells. J Immunol. 1994; 153: 4673–4683.
- ↵Berman ME, Muller WA. Ligation of Platelet/endothelial cell adhesion molecule 1 (PECAM-1/CD31) on monocytes and neutrophils increases binding capacity of leukocyte CR3 (CD11b/CD18). J Immunol. 1995; 154: 299–307.
- ↵Berman ME, Xie Y, Muller WA. Roles of platelet endothelial cell adhesion molecule-1 (PECAM-1, CD31) in natural killer cell transendothelial migration and β2 integrin activation. J Immunol. 1996; 156: 1515–1524.
- ↵Chiba R, Nakagawa N, Kurasawa K, Tanaka Y, Saito Y, Iwamoto I. Ligation of CD31 (PECAM-1) on endothelial cells increases adhesive function of αvβ3 integrin and enhances β1 integrin-mediated adhesion of eosinophils to endothelial cells. Blood. 1999; 94: 1319–1329.
- ↵Reedquist KA, Ross E, Koop EA, Wolthuis RM, Zwartkruis FJ, Kooyk YV, Salmon M, Buckley CD, Bos JL. The small GTPase, Rap1, mediates CD31-induced integrin adhesion. J Cell Biol. 2000: 148: 1151–1158.
- ↵Zhao T, Newman PJ. Integrin activation by regulated dimerization and oligomerization of platelet endothelial cell adhesion molecule (PECAM)-1 from within the cell. J Cell Biol. 2001; 152: 65–73.
- ↵Bertoni A, Tadokoro S, Eto K, Pampori N, Parise LV, White GC, Shattil SJ. Relationships between Rap1b, affinity modulation of integrin αIIbβ3, and the actin cytoskeleton. J Biol Chem. 2002; 277: 25715–25721.
- ↵Eto K, Murphy R, Kerrigan SW, Bertoni A, Stuhlmann H, Nakano T, Leavitt AD, Shattil SJ. Megakaryocytes derived from embryonic stem cells implicate CalDAG-GEFI in integrin signaling. Proc Natl Acad Sci U S A. 2002; 99: 12819–12824.
- ↵Knezvic I, Leisner TM, Lam SCT. Direct binding of the platelet integrin αIIbβ3 (GPIIb-IIIa) to talin: evidence that interaction is mediated through the cytoplasmic domains of both αIIb and β3. J Biol Chem. 1996; 271: 16416–16421.
- ↵Calderwood DA, Zent R, Grant R, Rees DJ, Hynes RO, Ginsberg MH. The talin head domain binds to integrin beta subunit cytoplasmic tails and regulates integrin activation. J Biol Chem. 1999; 274: 28071–28074.
- ↵Dangerfield J, Larbi KY, Huang MT, Dewar A, Nourshargh S. PECAM-1 (CD31) homophilic interaction up-regulates α6β1 on transmigrated neutrophils in vivo and plays a functional role in the ability of α6 Integrins to mediate leukocyte migration through the perivascular basement membrane. J Exp Med. 2002; 196: 1201–1212.
- ↵Roussel E, Gingras MC. Transendothelial migration induces rapid expression on neutrophils of granule-release VLA6 used for tissue infiltration. J Leukoc Biol. 1997; 62: 356–362.
- ↵O’Brien CD, Lim P, Sun J, Albelda SM. PECAM-1-dependent neutrophil transmigration is independent of monolayer PECAM-1 signaling or localization. Blood. 2003; 101: 2816–2825.
- ↵Noble KE, Wickremasinghe RG, DeCornet C, Panayiotidis P, Yong KL. Monocytes stimulate expression of the Bcl-2 family member, A1, in endothelial cells and confer protection against apoptosis. J Immunol. 1999; 162: 1376–1383.
- ↵Gao C, Sun W, Matsuyama S, Newman PJ. PECAM-1 is a specific and potent inhibitor of Bax-induced apoptosis. Blood. 2001; 98: 796a.
- ↵Sheibani N, Sorenson CM, Frazier WA. Differential modulation of cadherin-mediated cell-cell adhesion by platelet endothelial cell adhesion molecule-1 isoforms through activation of extracellular regulated kinases. Mol Biol Cell. 2000; 11: 2793–2802.
- ↵Takahashi-Tezuka M, Yoshida Y, Fukada T, Ohtani T, Yamanaka Y, Nishida K, Nakajima K, Hibi M, Hirano T. Gab1 acts as an adapter molecule linking the cytokine receptor gp130 to ERK mitogen-activated protein kinase. Mol Cell Biol. 1998; 18: 4109–4117.