Current models of the multistep adhesion cascade for leukocyte-endothelial interactions predict loss of L-selectin from the leukocyte surface before transendothelial migration. We have tested this hypothesis using in vitro adhesion and transendothelial migration assays and a zinc-dependent metalloproteinase inhibitor, Ro 31-9790 (N-2-((2s)-[(hydroxycarbamoyl)methyl)-4-methylvaleryl]-N-1,3 -dimethyl-L-valinamide), which prevents chemoattractant-induced (e.g., IL-8, FMLP, C5a, platelet-activating factor) L-selectin endoproteolytic cleavage from isolated human neutrophils. Inhibitor and vehicle-treated neutrophils exhibited identical behavior during both adhesive interactions with 4- and 24-h TNF-alpha-activated HUVEC monolayers under flow, (including rate of initial attachment, rolling velocities, stable adhesion, and transmigration) and in static adhesion assays. Flow cytometric analysis of transmigrated neutrophils with mAb to L-selectin revealed that vehicle treated neutrophils had minimal detectable surface L-selectin, whereas inhibitor-treated neutrophils retained comparable levels of L-selectin on their surface as neutrophils maintained at 37 degrees C. In addition, mAb to L-selectin that induce rapid shape change and homotypic adhesion (LAM1-116) did not enhance the rate or extent of neutrophil transmigration under flow or static conditions. Neutrophils preincubated with LAM 1-116 displayed similar behavior to neutrophils preincubated with the control anti-L-selectin mAb, LAM1-101. In summary, these results demonstrate that there is no requirement for L-selectin to be shed from the surface of neutrophils before, or during, their migration across endothelial monolayers, and that prevention of surface L-selectin proteolytic cleavage does not enhance or inhibit neutrophil-endothelial cell adhesive interactions.