B1 integrin activation inhibits in vitro tube formation: effects on cell migration, vacuole coalescence and lumen formation

Human endothelial cells (EC), when plated onto gels of extracellular matrix proteins such as Matrigel or collagen form capillary tubes in a process thought to mimic angiogenesis. We have shown previously that the extent of tube formation and the phenotype of the lumen are regulated by integrins (Gamble et al 1993) and lumen formation occurs through a process of vacuolization, coalescence and ultimate directional fusion of these vacuoles with the plasma membrane (Meyer 1997 et al). We now show here that activation of beta1 integrins on endothelial cells inhibits tube formation. On collagen gels, endothelial cells treated with 31 activating antibody 8A2 failed to migrate into the gel and tube formation was inhibited. Although several integrins mediate EC attachment to collagen alpha2beta1 is the chief determinant of EC behaviour since a blocking antibody to (alpha2beta1 reversed the effect of 8A2. On Matrigel tube formation was also inhibited by 8A2 treatment although cell alignment and sprout formation was still evident. Electron microscopy revealed the organisation of normal numbers of cells into solid sprouts and the formation of small intracellular vacuoles suggesting that initial stages of tube formation including cell migration were unaffected. However, beta1 integrin activation inhibited the coalescence of these small vacuoles into larger vacuoles, the recruitment of more cells into the sprout and the subsequent formation of mature lumen. The inhibition of capillary tube formation by beta1 activation was time dependent and long lasting. The critical time for activation of the beta1 integrin was the initial 1-2h after plating in order to inhibit tube formation although once activated, the beta1 mediated inhibition on Matrigel was still evident 4 days later. Our results suggest that beta1 integrins are critical in capillary tube formation in at least two phases. beta1 integrins are essential for migration of EC through collagen gels. Independently, beta1 integrins, although not involved in initial vacuole formation, are involved in the process of vacuole coalescence and subsequent lumen formation since beta1 integrin activation inhibits these processes.

Alpha 3 beta 1 adhesion to laminin-5 and invasin: critical and differential role of integrin residues clustered at the boundary between alpha 3 N-terminal repeats 2 and 3

Integrin/ligand interaction is a therapeutic target for many diseases. We previously reported that residues critical for ligand binding are clustered in N-terminal repeat 3 (in the predicted 2-3 loop) of alpha 4, alpha 5 and alpha IIb. Here we have localized residues critical for ligand binding in the alpha 3 subunit of integrin alpha 3 beta 1 with distinct ligand specificity (laminin-5). We identified an alpha 3 epitope common to several function-blocking anti-alpha 3 antibodies at the boundary between repeats 1 and 2 (residues 75-80). We found that swapping the predicted 4-1 loop (residues 153-165) at the boundary between repeats 2 and 3 with the corresponding alpha 4 sequence and mutating Thr-162 and Gly-163 residues in this predicted loop block laminin-5 binding. Thr-162 and Gly-163 and the antibody epitope are separated in the primary structure; however, they are close to each other in the proposed beta-propeller model. Mutating residues recently reported to block (Tyr-186 and Trp-188) or enhance (Asp-122) laminin-5 binding to alpha 3 beta 1 [Krukonis, E. S., Dersch, P., Eble, J. A., and Isberg, R. R.(1998) J. Biol. Chem. 273, 31837-31843] did not affect laminin-5 binding under the assay conditions used. Thr-162 and Gly-163 are not critical for adhesion to invasin, indicating that laminin-5 and invasin may use different recognition mechanisms, and that mutation of Thr-162 and Gly-163 does not drastically affect the integrity of alpha 3 beta 1. These results suggest that residues critical for ligand binding may be similarly (but not identically) located in repeat 3 of the alpha subunit regardless of ligand specificity.

Human monocytes bind to two cytokine-induced adhesive ligands on cultured human endothelial cells: endothelial-leukocyte adhesion molecule-1 and vascular cell adhesion molecule-1

Vascular cell adhesion molecule-1 (VCAM-1) and endothelial-leukocyte adhesion molecule-1 (ELAM-1) are adhesive proteins induced on endothelium by cytokines. We examined the contribution of these adhesive proteins to human peripheral blood monocyte adherence to endothelium using transfected Chinese hamster ovary (CHO) cells stably expressing these proteins and monoclonal antibodies (MoAbs) to ELAM-1, VCAM-1, or CD49d/CD29 (VLA-4), the leukocyte receptor for VCAM-1. Monocytes bound to CHO cells transfected with cDNA of ELAM-1 or VCAM-1. Binding to ELAM-1 was inhibited by MoAb to ELAM-1 and binding to VCAM-1 was inhibited by MoAb to VCAM-1 or the alpha-chain of very late activation antigen-4 (VLA-4) (CD49d). Additive inhibition of adherence to unstimulated human umbilical vein endothelium (HUVE) was observed when monocytes were pretreated with both MoAb to CD49d and MoAb to CD18, the common beta-chain of the leukocyte beta 2 integrin receptors. Adherence of monocytes to HUVE stimulated by recombinant human tumor necrosis factor-alpha was not reduced by MoAbs to CD18, CD49d, or ELAM-1 when used singly, but combinations of these MoAbs produced significant inhibition. We conclude that multiple receptor-ligand systems are involved in monocyte adherence to endothelium.