REGULATION OF FOXC2 EXPRESSION AND FUNCTION DURING EPITHELIAL-MESENCHYMAL TRANSITION BY GSK-3β

Metastasis is the ultimate cause for the majority of cancer-related deaths. The forkhead box transcription factor FOXC2 is known to be involved in regulating metastasis as well as a variety of developmental processes, including the formation of lymphatic and cardiovascular systems. Previous studies have shown that FOXC2 protein is localized either in the nucleus and/or in the cytoplasm of human breast tumor cells. This pattern of localization is similar to that of another forkhead family member, FOXO3a. Additionally, localization of FOXO3a is known to be differentially regulated by upstream kinase AKT. Therefore, I investigated whether FOXC2 localization could also be regulated by upstream kinases. Analysis of FOXC2 protein sequence revealed two potential phosphorylation sites for GSK-3β. Furthermore, inhibition of GSK-3βsignificantly reduces FOXC2 protein. In addition, exposure of HMLE Twist cells expressing endogenous FOXC2 to the GSK-3β inhibitor, TWS119, results in accumulation of FOXC2 protein in the cytoplasm with concomitant decrease in the nucleus in a time-dependent manner. Furthermore, continued treatment with TWS119 eventually induces epithelial morphology and decreased stem cell properties including sphere formation in these cells. Further characterization of FOXC2- GSK-3β interaction and the associated signaling cascade are necessary to determine the effect of FOXC2 phosphorylation by GSK-3β on EMT and metastasis.

TURNOVER RATE OF THE NEURONAL CONNEXIN CX36 IN HELA CELLS

Electrical synapses formed of the gap junction protein Cx36 show a great deal of functional plasticity, much dependent on changes in phosphorylation state of the connexin. However, gap junction turnover may also be important for regulating cell-cell communication, and turnover rates of Cx36 have not been studied. Connexins have relatively fast turnover rates, with short half-lives measured to be 1.5 to 3.5 hours in pulse-chase analyses of connexins (Cx26 and Cx43) in tissue culture cells and whole organs. We utilized HaloTag technology to study the turnover rate of Cx36 in transiently transfected HeLa cells. The HaloTag protein forms irreversible covalent bonds with chloroalkane ligands, allowing pulse-chase experiments to be performed very specifically. The HaloTag open reading frame was inserted into an internal site in the C-terminus of Cx36 designed not to disrupt the regulatory phosphorylation sites and not to block the C-terminal PDZ interaction motif. Functional properties of Cx36-Halo were assessed by Neurobiotin tracer coupling, live cell imaging, and immunostaining. For the pulse-chase study, transiently transfected HeLa cells were pulse labeled with Oregon Green (OG) HaloTag ligand and chase labeled at various times with tetramethylrhodamine (TMR) HaloTag ligand. Cx36-Halo formed large junctional plaques at sites of contact between transfected HeLa cells and was also contained in a large number of intracellular vesicles. The Cx36-Halo transfected HeLa cells supported Neurobiotin tracer coupling that was regulated by activation and inhibition of PKA in the same manner as wild-type Cx36 transfected cells. In the pulse-chase study, junctional protein labeled with the pulse ligand (OG) was gradually replaced by newly synthesized Cx36 labeled with the chase ligand (TMR). The half-life for turnover of protein in junctional plaques was 2.8 hours. Treatment of the pulse-labeled cells with Brefeldin A (BFA) prevented the addition of new connexins to junctional plaques, suggesting that the assembly of Cx36 into gap junctions involves the traditional ER-Golgi-TGN-plasma membrane pathway. In conclusion, Cx36-Halo is functional and has a turnover rate in HeLa cells similar to that of other connexins that have been studied. This turnover rate is likely too slow to contribute substantially to short-term changes in coupling of neurons driven by transmitters such as dopamine, which take minutes to achieve. However, turnover may contribute to longer-term changes in coupling.

Regulations of SOX9 activity during chondrogenesis

Sox9 is a transcription factor required for chondrocyte differentiation and cartilage formation. In an effort to identify SOX9 interacting protein(s), we screened a chondrocyte cDNA library with a modified yeast two-hybrid method, Son of Sevenless (SOS) recruitment system (SRS). The catalytic subunit of cyclic AMP-dependent protein kinase A (PKA-Cα) and a new long form of c-Maf transcription factor (Lc-Maf) were found to interact specifically with SOX9. We showed here that two PKA phosphorylation consensus sites of SOX9 could be phosphorylated by PKA in vitro as well as in vivo. PKA phosphorylation of SOX9 increases its DNA binding and transcriptional activities on a Col2a1 chondrocyte-specific enhancer. Mutations of these two PKA phosphorylation sites markedly decreased the activation of SOX9 by PKA. ^ To test whether parathyroid hormone-related peptide (PTHrP) signaling results in SOX9 phosphorylation, we generated a phosphospecific antibody that specifically recognizes SOX9 that is phosphorylated at serine 181 (S 181) one of the two consensus PKA phosphorylation sites. Addition of PTHrP to COS7 cells cotransfected with SOX9 and PTH/PTHrP receptor strongly increased phosphorylation of SOX9 at S181; this phosphorylation was blocked by a PKA-specific inhibitor. In similar experiments we showed that PTHrP increased the activity of a SOX9-dependent Col2a1 enhancer. This increase in activity was abolished when a SOX9 mutant was used containing serine-to-alanine substitution in the two consensus PKA phosphorylation sites of SOX9. Using our phosphospecific SOX9 antibody we showed by immunohistochemistry of mouse embryos that Sox9 phosphorylated at S181 was localized almost exclusively in the pre-hypertrophic zone of the growth plate, an area corresponding to the major site of expression of PTH/PTHrP receptor. In contrast, no phosphorylation of Sox9 at S181 was detected in growth plates of PTH/PTHrP receptor null mutant mice. Sox9, regardless of phosphorylation state, was present in all chondrocytes of both genotypes except in hypertrophic chondrocytes. Thus, Sox9 is a target of PTHrP signaling and the PTHrP-dependent phosphorylation of SOX9 enhances its transcriptional activity. ^ In order to investigate the in vivo function of Sox9 phosphorylation by PKA, we are generating a mouse model of mutant Sox9 harboring point mutations in two PKA phosphorylation sites. Preliminary results indicated that heterozygous mice containing half amount of mutant Sox9 that can not be phosphorylated by PKA have normal skeletal phenotype and homozygous mice are being generated. ^ Lc-Maf encodes an extra ten amino acids at the carboxyl terminus of c-Maf and contains a completely different 3′ untranslated region. The interaction between SOX9 and Lc-Maf was further confirmed by co-immunoprecipitation and GST-pull down assays, which mapped the interacting domains of SOX9 to HMG DNA binding domain and that of Lc-Maf to basic leusine zipper motif. In situ hybridizations showed that RNA of Lc-Maf coexpressed with those of Sox9 and Col2a1 in areas of mesenchymal condensation during the early stages of mouse embryo development. A DNA binding site of Lc-Maf was identified at the 5′ part of a 48-bp Col2a1 enhancer element near the HMG binding site of SOX9. Lc-Maf and SOX9 synergistically activated a luciferase reporter plasmid containing a Col2al enhancer and increased the transcription of endogenous Col2a1 gene. In summary, Lc-Maf is the first identified SOX9-interating protein during chondrogenesis and may be an important activator of Col2a1 gene. ^

Signaling pathways in the transcriptional and post-translational regulation of the human glutathione S -transferase P1 gene

The human glutathione S-transferase P1 (GSTP1) protein is an endogenous inhibitor of c-jun N-terminal kinases (JNKs) and an important phase II detoxification enzyme. ^ Recent identification of a cAMP response element (CRE) in the 5 ′-region of the human GSTP1 gene and several putative phosphorylation sites for the Ser/Thr protein kinases, including, cAMP-dependent protein kinases (PKAs), protein kinases C (PKCs), and JNKs in the GSTP1 protein raised the possibility that signaling pathways may play an important role in the transcriptional and post-translational regulation of GSTP1 gene. This study examined (a) whether the signaling pathway mediated by CAMP, via the GSTP1 CRE, is involved in the transcriptional regulation of the GSTP1 gene, (b) whether signaling pathways mediated by the Ser/Thr protein kinases (PKAs, PKCs, and JNKs) induce post-translational modification, viz. phosphorylation of the GSTP1 protein, and (c) whether such phosphorylation of the GSTP1 protein alters its functions in metabolism and in JNK signaling. ^ The first major finding in this study is the establishment of the human GSTP1 gene as a novel CAMP responsive gene in which transcription is activated via an interaction between PKA activated CRE binding protein-1 (CREB-1) and the CRE in the 5′-regulatory region. ^ The second major finding in this study is the observation that the GSTP1 protein undergoes phosphorylation and functionally activated by second messenger-activated protein kinases, PKA and PKC, in tumor cells with activated signaling pathways. Following phosphorylation by PKA or PKC, the catalytic activity of the GSTP1 protein was significantly enhanced, as indicated by a decrease in its Km (2- to 3.6-fold) and an increase in Kcat/ Km (1.6- to 2.5-fold) for glutathione. Given the frequent over-expression of GSTP1 and the aberrant PKA/PKC signaling cascade observed in tumors, these findings suggest that phosphorylation of GSTP1 may contribute to the malignant progression and drug-resistant phenotype of these tumors. ^ The third major finding in this study is that the GSTP1 protein, an inhibitor of JNKs, undergoes significant phosphorylation in tumor cells with activated JNK signaling pathway and in those under oxidative stress. Following phosphorylation by JNK, the ability of GSTP1 to inhibit JNK downstream function, i.e. c-jun phosphorylation, was significantly enhanced, suggesting a feedback mechanism of regulation of JNK-mediated cellular signaling. (Abstract shortened by UMI.) ^

Human carbonic anhydrase XII: cDNA cloning, expression, and chromosomal localization of a carbonic anhydrase gene that is overexpressed in some renal cell cancers

We report the cloning and characterization of a tumor-associated carbonic anhydrase (CA) that was identified in a human renal cell carcinoma (RCC) by serological expression screening with autologous antibodies. The cDNA sequence predicts a 354-amino acid polypeptide with a molecular mass of 39,448 Da that has features of a type I membrane protein. The predicted sequence includes a 29-amino acid signal sequence, a 261-amino acid CA domain, an additional short extracellular segment, a 26-amino acid hydrophobic transmembrane domain, and a hydrophilic C-terminal cytoplasmic tail of 29 amino acids that contains two potential phosphorylation sites. The extracellular CA domain shows 30–42% homology with known human CAs, contains all three Zn-binding histidine residues found in active CAs, and contains two potential sites for asparagine glycosylation. When expressed in COS cells, the cDNA produced a 43- to 44-kDa protein in membranes that had around one-sixth the CA activity of membranes from COS cells transfected with the same vector expressing bovine CA IV. We have designated this human protein CA XII. Northern blot analysis of normal tissues demonstrated a 4.5-kb transcript only in kidney and intestine. However, in 10% of patients with RCC, the CA XII transcript was expressed at much higher levels in the RCC than in surrounding normal kidney tissue. The CA XII gene was mapped by using fluorescence in situ hybridization to 15q22. CA XII is the second catalytically active membrane CA reported to be overexpressed in certain cancers. Its relationship to oncogenesis and its potential as a clinically useful tumor marker clearly merit further investigation.

Isolation of a multispecific organic anion and cardiac glycoside transporter from rat brain

A novel multispecific organic anion transporting polypeptide (oatp2) has been isolated from rat brain. The cloned cDNA contains 3,640 bp. The coding region extends over 1,983 nucleotides, thus encoding a polypeptide of 661 amino acids. Oatp2 is homologous to other members of the oatp gene family of membrane transporters with 12 predicted transmembrane domains, five potential glycosylation, and six potential protein kinase C phosphorylation sites. In functional expression studies in Xenopus laevis oocytes, oatp2 mediated uptake of the bile acids taurocholate (Km ≈ 35 μM) and cholate (Km ≈ 46 μM), the estrogen conjugates 17β-estradiol-glucuronide (Km ≈ 3 μM) and estrone-3-sulfate (Km ≈ 11 μM), and the cardiac gylcosides ouabain (Km ≈ 470 μM) and digoxin (Km ≈ 0.24 μM). Although most of the tested compounds are common substrates of several oatp-related transporters, high-affinity uptake of digoxin is a unique feature of the newly cloned oatp2. On the basis of Northern blot analysis under high-stringency conditions, oatp2 is highly expressed in brain, liver, and kidney but not in heart, spleen, lung, skeletal muscle, and testes. These results provide further support for the overall significance of oatps as a new family of multispecific organic anion transporters. They indicate that oatp2 may play an especially important role in the brain accumulation and toxicity of digoxin and in the hepatobiliary and renal excretion of cardiac glycosides from the body.

Proteolytic processing of the Alzheimer disease-associated presenilin-1 generates an in vivo substrate for protein kinase C

The majority of familial Alzheimer disease mutations are linked to the recently cloned presenilin (PS) genes, which encode two highly homologous proteins (PS-1 and PS-2). It was shown that the full-length PS-2 protein is phosphorylated constitutively within its N-terminal domain by casein kinases, whereas the PS-1 protein is not. Full-length PS proteins undergo endoproteolytic cleavage within their hydrophilic loop domain resulting in the formation of ≈20-kDa C-terminal fragments (CTF) and ≈30-kDa N-terminal fragments [Thinakaran, G., et al. (1996) Neuron 17, 181–190]. Here we describe the surprising finding that the CTF of PS-1 is phosphorylated by protein kinase C (PKC). Stimulation of PKC causes a 4- to 5-fold increase of the phosphorylation of the ≈20-kDa CTF of PS-1 resulting in reduced mobility in SDS gels. PKC-stimulated phosphorylation occurs predominantly on serine residues and can be induced either by direct stimulation of PKC with phorbol-12,13-dibutyrate or by activation of the m1 acetylcholine receptor-signaling pathway with the muscarinic agonist carbachol. However, phosphorylation of full-length PS-1 and PS-2 is not altered upon PKC stimulation. In addition, a mutant form of PS-1 lacking exon 10, which does not undergo endoproteolytic cleavage [Thinakaran, G., et al. (1996) Neuron 17, 181–190] is not phosphorylated by PKC, although it still contains all PKC phosphorylation sites conserved between different species. These results show that PKC phosphorylates the PS-1 CTF. Therefore, endoproteolytic cleavage of full-length PS-1 results in the generation of an in vivo substrate for PKC. The selective phosphorylation of the PS-1 CTF indicates that the physiological and/or pathological properties of the CTF are regulated by PKC activity.

Heat Stress Activates Fission Yeast Spc1/StyI MAPK by a MEKK-Independent Mechanism

Fission yeast Spc1/StyI MAPK is activated by many environmental insults including high osmolarity, oxidative stress, and heat shock. Spc1/StyI is activated by Wis1, a MAPK kinase (MEK), which is itself activated by Wik1/Wak1/Wis4, a MEK kinase (MEKK). Spc1/StyI is inactivated by the tyrosine phosphatases Pyp1 and Pyp2. Inhibition of Pyp1 was recently reported to play a crucial role in the oxidative stress and heat shock responses. These conclusions were based on three findings: 1) osmotic, oxidative, and heat stresses activate Spc1/StyI in wis4 cells; 2) oxidative stress and heat shock activate Spc1/StyI in cells that express Wis1AA, in which MEKK consensus phosphorylation sites were replaced with alanine; and 3) Spc1/StyI is maximally activated in Δpyp1 cells. Contrary to these findings, we report: 1) Spc1/StyI activation by osmotic stress is greatly reduced in wis4 cells; 2) wis1-AA and Δwis1 cells have identical phenotypes; and 3) all forms of stress activate Spc1/StyI in Δpyp1 cells. We also report that heat shock, but not osmotic or oxidative stress, activate Spc1 in wis1-DD cells, which express Wis1 protein that has the MEKK consensus phosphorylation sites replaced with aspartic acid. Thus osmotic and oxidative stress activate Spc1/StyI by a MEKK-dependent process, whereas heat shock activates Spc1/StyI by a novel mechanism that does not require MEKK activation or Pyp1 inhibition.

A Constitutively “Phosphorylated” Guanylyl Cyclase-linked Atrial Natriuretic Peptide Receptor Mutant Is Resistant to Desensitization

Dephosphorylation of the natriuretic peptide receptor-A (NPR-A) is hypothesized to mediate its desensitization in response to atrial natriuretic peptide (ANP) binding. Recently, we identified six phosphorylation sites within the kinase homology domain of NPR-A and determined that the conversion of these residues to alanine abolished the ability of the receptor to be phosphorylated or to be activated by ANP and ATP. In an attempt to generate a form of NPR-A that mimics a fully phosphorylated receptor but that is resistant to dephosphorylation, we engineered a receptor variant (NPR-A-6E) containing glutamate substitutions at all six phosphorylation sites. Consistent with the known ability of negatively charged glutamate residues to substitute functionally, in some cases, for phosphorylated residues, we found that NPR-A-6E was activated 10-fold by ANP and ATP. As determined by guanylyl cyclase assays, the hormone-stimulated activity of the wild-type receptor declined over time in membrane preparations in vitro, and this loss was blocked by the serine/threonine protein phosphatase inhibitor microcystin. In contrast, the activity of NPR-A-6E was more linear with time and was unaffected by microcystin. The nonhydrolyzable ATP analogue adenosine 5′-(β,γ-imino)-triphosphate was half as effective as ATP in stimulating the wild-type receptor but was equally as potent in stimulating NPR-A-6E, suggesting that ATP is required to keep the wild-type but not 6E variant phosphorylated. Finally, the desensitization of NPR-A-6E in whole cells was markedly blunted compared with that of the wild-type receptor, consistent with its inability to shed the negative charge from its kinase homology domain via dephosphorylation. These data provide the first direct test of the requirement for dephosphorylation in guanylyl cyclase desensitization and they indicate that it is an essential component of this process.

Molecular identification of human G-substrate, a possible downstream component of the cGMP-dependent protein kinase cascade in cerebellar Purkinje cells

G-substrate, an endogenous substrate for cGMP-dependent protein kinase, exists almost exclusively in cerebellar Purkinje cells, where it is possibly involved in the induction of long-term depression. A G-substrate cDNA was identified by screening expressed sequence tag databases from a human brain library. The deduced amino acid sequence of human G-substrate contained two putative phosphorylation sites (Thr-68 and Thr-119) with amino acid sequences [KPRRKDT(p)PALH] that were identical to those reported for rabbit G-substrate. G-substrate mRNA was expressed almost exclusively in the cerebellum as a single transcript. The human G-substrate gene was mapped to human chromosome 7p15 by radiation hybrid panel analysis. In vitro translation products of the cDNA showed an apparent molecular mass of 24 kDa on SDS/PAGE which was close to that of purified rabbit G-substrate (23 kDa). Bacterially expressed human G-substrate is a heat-stable and acid-soluble protein that cross-reacts with antibodies raised against rabbit G-substrate. Recombinant human G-substrate was phosphorylated efficiently by cGMP-dependent protein kinase exclusively at Thr residues, and it was recognized by antibodies specific for rabbit phospho-G-substrate. The amino acid sequences surrounding the sites of phosphorylation in G-substrate are related to those around Thr-34 and Thr-35 of the dopamine- and cAMP-regulated phosphoprotein DARPP-32 and inhibitor-1, respectively, two potent inhibitors of protein phosphatase 1. However, purified G-substrate phosphorylated by cGMP-dependent protein kinase inhibited protein phosphatase 2A more effectively than protein phosphatase 1, suggesting a distinct role as a protein phosphatase inhibitor.

The Catastrophe-promoting Activity of Ectopic Op18/Stathmin Is Required for Disruption of Mitotic Spindles But Not Interphase Microtubules

Oncoprotein18/stathmin (Op18) is a microtubule (MT) destabilizing protein that is inactivated during mitosis by phosphorylation at four Ser-residues. Op18 has at least two functions; the N-terminal region is required for catastrophe-promotion (i.e., transition from elongation to shortening), while the C-terminal region is required to inhibit MT-polymerization rate in vitro. We show here that a “pseudophosphorylation” derivative of Op18 (i.e., four Ser- to Glu-substitutions at phosphorylation sites) exhibits a selective loss of catastrophe-promoting activity. This is contrasted to authentic phosphorylation, which efficiently attenuates all activities except tubulin binding. In intact cells, overexpression of pseudophosphorylated Op18, which is not phosphorylated by endogenous kinases, is shown to destabilize interphase MTs but to leave spindle formation untouched. To test if the mitotic spindle is sensitive only to the catastrophe-promoting activity of Op18 and resistant to C-terminally associated activities, N- and C-terminal truncations with defined activity-profiles were employed. The cell-cycle phenotypes of nonphosphorylatable mutants (i.e., four Ser- to Ala-substitutions) of these truncation derivatives demonstrated that catastrophe promotion is required for interference with the mitotic spindle, while the C-terminally associated activities are sufficient to destabilize interphase MTs. These results demonstrate that specific Op18 derivatives with defined activity-profiles can be used as probes to distinguish interphase and mitotic MTs.

Rck2, a member of the calmodulin-protein kinase family, links protein synthesis to high osmolarity MAP kinase signaling in budding yeast

Rck2, a yeast Ser/Thr protein kinase homologous to mammalian calmodulin kinases, requires phosphorylation for activation. We provide evidence that in budding yeast, this step can be executed by the osmostress-activated mitogen-activated protein kinase Hog1. Rck2 phosphorylation was transiently increased during osmostress or in mutants with a hyperactive high osmolarity glycerol (HOG) pathway. This modification depended on catalytically active Hog1 kinase and two putative mitogen-activated protein kinase phosphorylation sites in Rck2. Immunokinase assays showed that Hog1 can directly phosphorylate Rck2 to stimulate its enzymatic activity toward translation elongation factor 2. We demonstrate that Hog1 and Rck2 are necessary for attenuation of protein synthesis in response to osmotic challenge and show that modification of elongation factor 2 induced by osmostress depends on Rck2 and Hog1 in vivo. Therefore, we propose that the transient down-regulation of protein synthesis after osmotic shock is a response not to damage but to an extracellular signal mediated by Hog1 and Rck2.

Protein kinase Cdc15 activates the Dbf2-Mob1 kinase complex

Exit from mitosis in budding yeast requires inactivation of cyclin-dependent kinases through mechanisms triggered by the protein phosphatase Cdc14. Cdc14 activity, in turn, is regulated by a group of proteins, the mitotic exit network (MEN), which includes Lte1, Tem1, Cdc5, Cdc15, Dbf2/Dbf20, and Mob1. The direct biochemical interactions between the components of the MEN remain largely unresolved. Here, we investigate the mechanisms that underlie activation of the protein kinase Dbf2. Dbf2 kinase activity depended on Tem1, Cdc15, and Mob1 in vivo. In vitro, recombinant protein kinase Cdc15 activated recombinant Dbf2, but only when Dbf2 was bound to Mob1. Conserved phosphorylation sites Ser-374 and Thr-544 (present in the human, Caenorhabditis elegans, and Drosophila melanogaster relatives of Dbf2) were required for DBF2 function in vivo, and activation of Dbf2-Mob1 by Cdc15 in vitro. Although Cdc15 phosphorylated Dbf2, Dbf2–Mob1, and Dbf2(S374A/T544A)–Mob1, the pattern of phosphate incorporation into Dbf2 was substantially altered by either the S374A T544A mutations or omission of Mob1. Thus, Cdc15 promotes the exit from mitosis by directly switching on the kinase activity of Dbf2. We propose that Mob1 promotes this activation process by enabling Cdc15 to phosphorylate the critical Ser-374 and Thr-544 phosphoacceptor sites of Dbf2.

Critical role for the docking-protein FRS2α in FGF receptor-mediated signal transduction pathways

The docking protein FRS2α has been implicated as a mediator of signaling via fibroblast growth factor receptors (FGFRs). We have demonstrated that targeted disruption of FRS2α gene causes severe impairment in mouse development resulting in embryonal lethality at E7.0–E7.5. Experiments with FRS2α-deficient fibroblasts demonstrate that FRS2α plays a critical role in FGF-induced mitogen-activated protein (MAP) kinase stimulation, phosphatidylinositol-3 (PI-3) kinase activation, chemotactic response, and cell proliferation. Following FGF stimulation, tyrosine phosphorylated FRS2α functions as a site for coordinated assembly of a multiprotein complex that includes Gab1 and the effector proteins that are recruited by this docking protein. Furthermore, we demonstrate that different tyrosine phosphorylation sites on FRS2α are responsible for mediating different FGF-induced biological responses. These experiments establish the central role of FRS2α in signaling via FGFRs and demonstrate that FRS2α mediates multiple FGFR-dependent signaling pathways critical for embryonic development.

Integrin engagement mediates tyrosine dephosphorylation on platelet-endothelial cell adhesion molecule 1.

Platelet-endothelial cell adhesion molecule 1 (PECAM-1, CD31) is a 130-kDa member of the immunoglobulin gene superfamily expressed on endothelial cells, platelets, neutrophils, and monocytes and plays a role during endothelial cell migration. Phosphoamino acid analysis and Western blot analysis with anti-phosphotyrosine antibody show that endothelial PECAM-1 is tyrosine-phosphorylated. Phosphorylation is decreased with endothelial cell migration on fibronectin and collagen and with cell spreading on fibronectin but not on plastic. Cell adhesion on anti-integrin antibodies is also able to specifically induce PECAM-1 dephosphorylation while concurrently inducing pp125 focal adhesion kinase phosphorylation. Inhibition of dephosphorylation with sodium orthovanadate suggests that this effect is at least partially mediated by phosphatase activity. Tyr-663 and Tyr-686 are identified as potential phosphorylation sites and mutated to phenylalanine. When expressed, both mutants show reduced PECAM-1 phosphorylation but Phe-686 mutants also show significant reversal of PECAM-1-mediated inhibition of cell migration and do not localize PECAM-1 to cell borders. Our results suggest that beta 1-integrin engagement can signal to dephosphorylate PECAM-1 and that this signaling pathway may play a role during endothelial cell migration.