Coculture of BMVECs with monocytes leads to Rho activation, and Rho inhibitor blocks monocyte migration. (A-C) The content of both active Rho-GTP and total Rho was measured in BMVECs or BMVECs cocultured with infected or uninfected monocytes for 2 hours. (A) Representative immunoblots of active Rho-GTP, total Rho, and internal standard, actin. (B) Ratio of Rho-GTP/actin. (C) Ratio of total Rho/actin. (D-E) After 2 hours of coculture, monocytes were washed from BMVECs (removing more than 90% of applied monocytes) or kept on BMVECs. The active form of Rho was derived mainly from BMVECs in coculture because monocyte removal slightly diminished Rho-GTP content in protein extracts (bar 2 versus bar 3). Pretreatment of BMVECs with Rho inhibitor, C botulinum C3 transferase (4 hours; 50 μg/mL), prevented Rho activation in coculture. (D) Representative immunoblots of active Rho-GTP, total Rho, and internal standard, actin. (E) Ratio of Rho-GTP/actin. (F) Ratio of total Rho/actin. Results were expressed as the ratio of target protein immunoreactivity to that of the internal standard, actin. Each bar represents the mean value of 4 replicates ± SEM. P < .01 compared with endothelial cells without monocyte coculture. (G) Migration of fresh 105 monocytes applied to BBB constructs was studied in response to MCP-1 (5 ng/mL in the bottom chamber). Pretreatment of BMVECs in the top chamber with C3 transferase (4 hours; 50 μg/mL) diminished monocyte migration across the BBB by 50% (without MCP-1) or 63% (with MCP-1) compared with respective controls. Pretreatment of monocytes with C3 transferase (4 hours; 50 μg/mL#) resulted in 60% inhibition of migration. Results were normalized to migration of monocytes without MCP-1 (a value of 100%). Values represent the mean of quadruplicate determinations ± SEM. *P < .01 compared with BBB constructs without pretreatment and MCP-1 application. **P < .01 compared with BBB constructs with MCP-1 application. Results of 1 representative experiment are shown.