Fig. 3. Frequency of binding events for control and specificity experiments for L-selectin (A) and E-selectin (B). The presence of EDTA was found to abrogate binding. The presence of blocking antibodies for L-selectin (LAM1-116), E-selectin (ENA2) and PSGL-1 (KPL-1) dramatically reduced the frequency of binding. (C) Reduction in frequency of selectin binding to PMNs treated with OSGE. Reductions in the frequencies indicate that heavily O-linked glycoproteins are crucial for L- and P-selectin binding, and play a predominant role in E-selectin binding to PMNs. (D) Rupture-force histograms for E-selectin binding to control PMNs and OSGE-treated PMNs at a reproach velocity of 10 µm second^–1. E-selectin binding to OSGE-treated PMNs resulted in a lower frequency of rupture events with a lower overall magnitude. As a result of the lower binding frequency, the area under the force histogram for OSGE-treated PMNs is about 40% that of the control. The less rigorous requirements for E-selectin binding might allow it to bind more easily to other glycosylated ligands (N-linked glycoproteins or glycosphingolipids) previously shielded by the heavily O-linked glycoproteins.