LPS activates RhoA through a p38- and Cdc42-dependent pathway. (A) Human PMNs were preincubated with 10 μM SB203580 or 0.1% DMSO vehicle, treated with LPS for the indicated times, and lysates assayed for RhoA activation.26 Active RhoA was quantified by densitometry and plotted (P < .05 by 2-way ANOVA for SB203580 versus vehicle). (B) PMNs were transduced using BioPORTER (BP) reagent with wild-type WASP-CRIB or mutant non-Cdc42 binding WASP-CRIB,24 LPS-exposed (20 minutes), and then assayed for RhoA activation. Active RhoA was quantified by densitometry and plotted (*P < .001 for LPS/wt WASP-CRIB versus LPS/−). (A-B) Error bars represent SE. (C) PMNs were untreated or transduced with either GST (control) or L61Cdc42 and then assayed for RhoA and Cdc42 activation. LPS was used as a positive control. (D) PMNs were treated with buffer or C3 transferase (20 μg/mL, 37°C, 4 hours) and assayed for Cdc42 activation.27 Lysates were also probed with rabbit anti-Cdc42, -PO4-p38, p38, and -RhoA antibodies. C3- and buffer-treated PMNs were treated with LPS (37°C, 45 minutes), stained with rhodamine-phalloidin, and imaged under × 60 objective.30 Image intensity was enhanced on C3-treated PMNs to facilitate comparisons of morphology. (E) Human PMNs treated as in panel A were lysed, and the activity of immunoprecipitated ROCKα was tested.28 (F) PMNs were pretreated with 500 ng/mL Clostridium difficile toxin B or buffer (90 minutes), exposed to LPS (30 minutes), lysed, and immunoblotted for PO4-p38 and total p38. Under the same conditions, p38 kinase activity was assayed.8 All panels are representative of 3 or more independent experiments.