Extracellular signal-regulated kinase (ERK) activation is required for GP Ibalpha-dependent endothelial cell migration.

The GP Ib complex can participate in endothelial cell (EC) migration on von Willebrand factor (vWF) or the mixed matrix of vWF and type I collagen (vWF/collagen). In this study, viper venom proteins alboaggregin (albo) A or B blocked GP Ibalpha, and echistatin inhibited alphavbeta3 binding. Albo A, B and echistatin inhibited EC migration on vWF and vWF/collagen. Albo B or the anti-GP Ibalpha monoclonal antibody (mAb) 1b1 did not affect the migration of smooth muscle cells or fibroblasts, which lack GP Ib. EC also migrate on albo A- or albo B-coated dishes. PD98059, which blocks ERK activation, abolished EC migration on vWF, vWF/collagen, collagen or albo B. Soluble albo A or 1b1 dramatically inhibited ERK activation during EC migration on vWF or albo B. Echistatin inhibited ERK activation on vWF and vitronectin (VN), but not albo B. Thus, in addition to alphavbeta3, EC GP Ibalpha initiates ERK activation, and regulates ERK-induced EC migration on vWF.

Isolation and characterization of EMS16, a C-lectin type protein from Echis multisquamatus venom, a potent and selective inhibitor of the alpha2beta1 integrin.

We have isolated and characterized EMS16, a potent and selective inhibitor of the alpha2beta1 integrin, from Echis multisquamatus venom. It belongs to the family of C-lectin type of proteins (CLPs), and its amino acid sequence is homologous with other members of this protein family occurring in snake venoms. EMS16 (M(r) approximately 33K) is a heterodimer composed of two distinct subunits linked by S-S bonds. K562 cells transfected with alpha2 integrin selectively adhere to immobilized EMS16, but not to two other snake venom-derived CLPs, echicetin and alboaggregin B. EMS16 inhibits adhesion of alpha2beta1-expressing cells to immobilized collagen I at picomolar concentrations, and the platelet/collagen I interaction in solution at nanomolar concentrations. EMS16 inhibits binding of isolated, recombinant I domain of alpha2 integrin to collagen in an ELISA assay, but not the interaction of isolated I domain of alpha1 integrin with collagen IV. Studies with monoclonal antibodies suggested that EMS16 binds to the alpha2 subunit of the integrin. EMS16 inhibits collagen-induced platelet aggregation, but has no effect on aggregation induced by other agonists such as ADP, thromboxane analogue (U46619), TRAP, or convulxin. EMS16 also inhibits collagen-induced, but not convulxin-induced, platelet cytosolic Ca(2+) mobilization. In addition, EMS16 inhibits HUVEC migration in collagen I gel. In conclusion, we report a new, potent viper venom-derived inhibitor of alpha2beta1 integrin, which does not belong to the disintegrin family.

Evaluation of endothelial cell migration with a novel in vitro assay system.

In this study we introduce a novel in vitro 'oil-drop' assay system for the measurement of endothelial cell (EC) migration, based on the original concept of the Teflon fence assay (Pratt et al., 1984; Am. J. Pathol. 117: 349-354). An aliquot of 15-20,000 human umbilical vein EC (HUVEC) is pipetted through a layer of mineral oil. The cells readily attach, spread and migrate on the surface of a matrix-coated tissue culture dish as a confluent circular monolayer. Migration is measured as the net increase in the total area covered at 24 hours. We have used this system to quantify EC migration on matrices composed of a mixture of type I collagen and either von Willebrand factor (vWF) or fibronectin (FN) in the presence or absence of tumor necrosis factor alpha (TNFalpha). Plating efficiency on both vWF/collagen and FN/collagen, measured by counting cells after attachment and spreading, is about 80%. With this method, migration on vWF/collagen was about 6.4 mm(2) and 5.3 mm(2) for TNFalpha-treated and untreated HUVEC, respectively. HUVEC migration on FN/collagen was slightly greater - 6.4 mm(2) and 6.5 mm(2) with and without TNFalpha treatment, respectively. During the 24 hour time period, HUVEC numbers increased 30-40% on vWF/collagen, and 60-80% on FN/collagen, with increased proliferation observed with TNF-alpha treatment. EC proliferation could be completely inhibited by 2 mM hydroxyurea. This assay system has proven useful in our studies to quantify cell migration and proliferation.

Glycoprotein Ibalpha can mediate endothelial cell migration on von Willebrand factor-containing substrata.

We have previously shown that, in addition to the vitronectin receptor (VNR, alpha(v)beta(3)), the GP Ib complex can participate in endothelial cell (EC) attachment to von Willebrand Factor (vWF) (D. A. Beacham, M. S. Cruz, and R. I. Handin, 1995, Thromb. Haemostas. 73, 309-317; D. A. Beacham, L.-P. Tran, and S. S. Shapiro, 1997, Blood 89, 4071-4077). In this study we have investigated the functional roles of these vWF receptors in the migration of untreated and TNFalpha-treated EC on vWF, a mixture of vWF and type I collagen, and on vitronectin (VN). In agreement with previous studies (D. I. Leavesley, M. A. Schwartz, M. Rosenfeld, and D. A. Cheresh, 1993, J. Cell Biol. 121, 163-170), the migration of untreated and TNFalpha-treated EC on VN was dependent entirely on the VNR. Migration of untreated EC on vWF was inhibited 10-15% by recombinant vWF-A1, the GP Ibalpha-binding domain on vWF which abrogates the platelet GP Ibalpha-vWF interaction. In contrast, migration of TNFalpha-treated EC on vWF was inhibited 50-60% by vWF-A1 or the anti-GP Ibalpha mAb AS-7 but only 20% by the anti-VNR mAb LM609. On a mixed vWF-collagen substratum, vWF-A1 inhibited untreated EC migration by 45%, and TNFalpha-treated EC migration by 75%. The possible role of EC proliferation was eliminated, since hydroxyurea completely inhibited EC proliferation without reducing migration significantly. The anti-GP Ibalpha mAb Ib1 inhibited EC migration by 50%, but reduced proliferation by only 15%. Taken together, our data demonstrate that EC migration on vWF-containing substrata involves the GP Ib complex as well as the VNR and raises the possibility that the VNR and GP Ib act cooperatively in supporting EC migration.

Arterial shear stress stimulates surface expression of the endothelial glycoprotein Ib complex.

Exposure to shear stress has been shown to alter the expression of a number of surface components of cultured endothelial cells (EC). However, relatively few studies have examined the status of human EC surface proteins after prolonged flow, more closely corresponding to the steady state in vivo. Since the promoter region of glycoprotein (Gp) Ib alpha contains several copies of a putative shear stress response element, 5'-GAGACC-3', we investigated the response of cultured human umbilical vein EC (HUVEC) GpIb alpha to shear stress over a 72 h time period. In response to 30 dynes/cm2 of shear stress, total cell content of GpIb alpha protein was markedly increased above static levels at 7 and 24 h, as determined immunohistochemically. Western blot analysis of whole cell lysates after 24, 48, and 72 h of shear treatment demonstrated a 2.4-, 4.1-, and 3.2-fold increase in total GpIb alpha protein, respectively. Cell surface protein expression of GpIb alpha increased 2.5-fold at 7 h, as measured by quantitative immunofluorescence, and remained at that level at 24 h. After 48 h of shear stress, cell surface GpIb alpha, GpIX, and GpV, analyzed by flow cytometric analysis, were further increased over the levels observed at 24 h. The increase in cell surface membrane expression of GPIb alpha at 24, 48, and 72 h was confirmed by immunoprecipitation of biotinylated surface proteins. No upregulation of GpIb alpha was noted after exposure to shear stress of 1-3 dynes/cm2. These observations imply that under steady-state arterial shear conditions endothelial expression of the GpIb complex is significantly greater than observed in static EC cultures, and raise the possibility of a more important role for this complex under flow, rather than static conditions.

Cytokine treatment of endothelial cells increases glycoprotein Ib alpha-dependent adhesion to von Willebrand factor.

Endothelial cells (EC) possess at least two membrane receptors for von Willebrand factor (vWF), the vitronectin receptor (VNR, alpha(v)beta3), which recognizes an Arg-Gly-Asp (RGD) sequence in the C-terminus of vWF, and glycoprotein Ib alpha (GP Ib alpha), which interacts with a region in the N-terminal A1 domain of vWF. In the absence of added cytokines, EC attachment to a vWF substratum is mediated largely through the alpha(v)beta3, with a smaller contribution by GP Ib alpha. In the present study, we have examined the effect of cytokines on the receptor specificity of EC attachment to wild-type vWF (WT-vWF) and to vWF, which had been mutated in the C-terminal RGDS sequence (RADS-vWF). Exposure of human umbilical vein EC (HUVEC) to tumor necrosis factor-alpha (TNF-alpha) or to TNF-alpha in combination with interferon-gamma (IFN-gamma), but not to interleukin-1beta (IL-1), increased attachment to RADS-vWF by about twofold. The TNF-alpha-induced increase in EC attachment was accompanied by an increase in cell surface GP Ib alpha expression; GP Ib alpha surface expression was not increased by IL-1. Attachment of untreated HUVEC to WT-vWF could be inhibited 60% to 70% by a monoclonal antibody (MoAb) (LM609) to the VNR and 30% to 40% by the A1 fragment of vWF (containing the GP Ib alpha binding domain). The pattern of inhibition of attachment to WT-vWF was largely unchanged after TNF-alpha treatment of HUVEC. In contrast, the attachment to WT-vWF of HUVEC, treated with TNF-alpha +IFN-gamma was completely inhibited by vWF-A1 and inhibited only 35% by the anti-VNR antibody LM609. Two MoAbs to GP Ib alpha produced similar, but incomplete, inhibition. Pretreatment of HUVEC with the combination of TNF-alpha + IFN-gamma produced a dramatic decrease in VNR expression, confirming previous findings of Defilippi et al. These results suggest that in the presence of the inflammatory cytokines TNF-alpha + IFN-gamma, the endothelial GP Ib complex is a major determinant of HUVEC adhesion to surface-bound vWF.

Glycoprotein Ib can mediate endothelial cell attachment to a von Willebrand factor substratum.

Introduction of single amino acid substitutions into the C-terminal Arg-Gly-Asp-Ser (RGDS) site of von Willebrand Factor, referred to as RGD mutant vWF, selectively abrogated vWF binding to platelet glycoprotein IIb/IIIa (GpIIb/IIIa alpha IIb beta 3) and abolished human umbilical vein endothelial cell (HUVEC) spreading, but not attachment, to RGD mutant vWF (Beacham, D.A., Wise, R.J., Turci, S.M. and Handin, R. I. 1992. J. Biol. Chem. 167, 3409-3415). These results suggested that in addition to the vitronectin receptor (VNR, alpha V beta 3), a second endothelial membrane glycoprotein can mediate HUVEC adhesion to vWF. HUVEC attachment to wild-type (WT) and RGD-mutant vWF was reduced by two proteins known to block the vWF-platelet glycoprotein Ib/IX (GpIb/IX) interaction, the monoclonal antibody AS-7 and the recombinant polypeptide, vWF-A1. The addition of cytochalasin B or DNase I to disrupt potential GPIb alpha-cytoskeletal interactions enhanced the immunoprecipitation of endothelial GPIB alpha, caused HUVEC to round up, and increased HUVEC adhesion to RGD mutant vWF. These results indicate that while the VNR is the primary adhesion receptor for vWF, endothelial GPIb alpha can mediate HUVEC attachment to vWF. GpIb-dependent attachment could contribute to HUVEC adhesion under conditions when cell surface expression of the VNR is downregulated, and VNR-dependent adhesion is reduced.

Selective inactivation of the Arg-Gly-Asp-Ser (RGDS) binding site in von Willebrand factor by site-directed mutagenesis.

In order to assess the requirement for the Arg-Gly-Asp-Ser (RGDS) consensus adhesion sequence in von Willebrand factor (vWF) for vWF binding to platelets and endothelial cells, point mutations were introduced into this sequence by site-directed mutagenesis. A glycine to alanine substitution yielded RADS-vWF, while an aspartate to glutamate substitution resulted in RGES-vWF. Recombinant RADS-vWF and RGES-vWF, purified from transformed Chinese hamster ovary cells, were compared with recombinant wild type vWF (WT-vWF) in functional assays with platelets and human umbilical vein endothelial cells (HU-VECs). High molecular weight RADS-vWF and RGES-vWF multimers inhibited binding of 125I-vWF to a mixture of insolubilized native type I and III collagen and competed effectively with 125I-vWF for binding to formalin-fixed platelets in the presence of ristocetin, indicating functional collagen and platelet glycoprotein Ib binding. However, RADS-vWF and RGES-vWF were unable to displace the binding of 125I-vWF to thrombin or ADP-activated platelets. The attachment of HUVECs to either RADS-vWF or RGES-vWF coated surfaces was reduced and spreading was almost completely inhibited, compared with WT-vWF. We conclude that point mutations of the RGDS sequence in vWF selectively impair binding to platelet glycoprotein IIb/IIIa and the HUVEC vitronectin receptor.

Mg2+ mediates the cell-substratum interaction of Arg-Gly-Asp-dependent HeLa cell collagen receptors.

Three HeLa cell surface collagen receptors of apparent molecular mass 102/58, 87, and 38/33 kDa were eluted from gelatin-Sepharose with salt gradients or Arg-Gly-Asp-containing peptides. To understand how the collagen receptors are involved in HeLa cell spreading on collagen we investigated the effects of divalent cations and Arg-Gly-Asp-containing peptides on adhesion to gelatin, since HeLa cells behave similarly on both native type I collagen and gelatin substrata and also whether Arg-Gly-Asp-containing substrata would substitute for gelatin in facilitating cell spreading. Gly-Arg-Gly-Asp-Ser-containing peptides in solution inhibited HeLa cell spreading onto gelatin and promoted only partial HeLa cell spreading when bound to tissue culture plastic. Both partial spreading of HeLa cells on the Gly-Arg-Gly-Asp-Ser substratum and full spreading on gelatin was dependent on Mg2+, but not on Ca2+. Binding of the 102/58-, 87-, and 38/33-kDa collagen receptors to gelatin-Sepharose was increased fourfold in the presence of Mg2+, and subsequent elution of the collagen receptors and a 45-kDa collagen-binding protein not thought to be involved in HeLa cell spreading was achieved with EDTA. In contrast, affinity chromatography on Gly-Arg-Gly-Asp-Ser-Sepharose eluted predominantly the 45-kDa collagen-binding protein and the 38/33-kDa collagen receptor. In summary, the Mg2(+)-dependent interaction of the collagen receptors with the Arg-Gly-Asp sequence in collagen appears to be essential for the initial events in HeLa cell spreading but is not sufficient for full cell spreading.

The identification and characterization of collagen receptors involved in HeLa cell-substratum adhesion.

Four proteins of molecular mass 102, 87, 45, and 38 kDa were isolated from plasma membrane preparations by affinity chromatography. The 102-, 87-, and 38-kDa proteins were shown to be collagen receptors involved in the adhesion of HeLa cells to a gelatin substratum. All four proteins were eluted by high salt from affinity columns made of either types I or IV collagen or type I gelatin. Generally, a total of six major proteins were found in the high salt eluates, although the relative amounts of each varied among experiments. Immunoprecipitation, immunoblotting, and limited peptide mapping indicated that the 102-kDa protein was most sensitive to proteolysis leading to the formation of proteins of molecular mass 58 and 54 kDa. Even in the presence of a mixture of protease inhibitors the 58-kDa fragment was usually the more abundant species. Lectin binding indicated that the 102-, 87-, and 38-kDa proteins contain carbohydrate. Phase-partitioning with Triton X-114 and the need to solubilize the proteins in Triton X-100 indicated that the 102-, 87-, 45-, and 38-kDa proteins have a hydrophobic domain. The 87-kDa protein partitioned exclusively with the detergent-rich phase, suggesting that it is the most hydrophobic. Cell surface labeling with 125I indicated that the four proteins have an extracellular domain. Four criteria were used to determine which of the four proteins are collagen receptors mediating cell-substrate adhesion: 1) during HeLa cell adhesion, proteins with Mr values similar to all four proteins or their peptide fragments were cross-linked to a gelatin substratum derivatized with a photoactivatable probe; 2) a pentapeptide containing the Arg-Gly-Asp cell recognition sequence eluted the same four proteins as those found by high salt elution of collagen affinity columns; 3) monospecific antibodies to the 102-, 87-, and 38-kDa proteins, but not the 45-kDa protein, inhibited the spreading of HeLa cells on a gelatin substratum; 4) monospecific antibodies to the 102-, 87-, and 38-kDa proteins, but not the 45-kDa protein, bound to culture dishes substituted for gelatin in mediating the spreading of HeLa cells. Taken together, the data suggest that the 102-, 87-, and 38-kDa proteins are collagen receptors involved in HeLa cell adhesion. Although the 45-kDa protein has two of the characteristics of a collagen receptor defined here, it does not fit the criteria for one involved in cell-substratum adhesion.