Crosstalk between kindlin-3 and β₂-integrins modulates NET release in vitro and in vivo. (A) Blood samples (∼100 µl/mouse) were collected from mice before and after LPS treatment and mixed with the same volume of phosphate-buffered saline. Diluted plasma was obtained by centrifugation and plasma DNA was measured by SYTOX green. (B) Counts of granulocytes in peripheral blood of mice. (C) Plasma DNA were normalized by the granulocyte counts in peripheral blood (per million). (D) BM neutrophils were treated with PMA (20 nM) for 2 hours followed by fixation, permeabilizaion, and labeling nuclear DNA with Hoechst to display released NETs (arrowheads). (E) Quantification of formed NETs from (D). (F) NETs were solubilized from PMA-stimulated neutrophils by MNase and then quantified by SYTOX green. (G) Interaction of GST-β₂CT with kindlin-3 was evaluated by pull-down assays in the presence of the peptide derived from the integrin β₂ CT (β₂CTP: GRKKRRQRRRFKSATTTVMNPKFAES) or its mutated one (mβ₂CTP: GRKKRRQRRRFKSAAAAVMNPKAAES) at a concentration of 20 µM. (H) BM neutrophils of WT mice were pretreated with the membrane-permeable β₂CTP or mβ₂CTP (20 µM) and NET release was measured as described in (F). (I) BM neutrophils of WT mice were pretreated with the β₂CTP (20 µM), and their adhesive ability to fibrinogen was quantified as described in Figure 1E. (J-L) WT mice were administrated with the β₂CTP (5 mg/kg) by IV injection before LPS treatment to induce endotoxemia. The livers were harvested from the endotoxemic mice for IHC staining of Gr1 (J) and measuring MPO activity (K); meanwhile, blood samples were collected for quantifying plasma DNA (L). Data represent mean ± SEM of 3 or more independent experiments; *P < .05; **P < .01 (paired Student t test).