Fig. 2. Characterisation of Rac1 deletion. (A) Southern blot of DNA cut with BamHI, derived from the bone marrow of a Rac1^flox/flox (1) and a Rac1^flox/floxMx1-Cre (2) mouse that had been injected with polyIC as described in the Materials and Methods. Bone marrow was harvested 2 days after the last polyIC injection. The Southern blot was probed with a DNA fragment from intron 3 of Rac1 which hybridises to a 2.9 kb and a 1.5 kb fragment in the Rac1^flox and Rac1 alleles respectively, as shown (Walmsley et al., 2003). The number below lane 2 shows the percentage deletion of the Rac1^flox allele. Typical deletion of the Rac1^flox allele in bone marrow was 90% or higher, and persisted for at least 3 months following polyIC injection (not shown). (B) Analysis of Rac1 and Rac2 protein expression in Wt and Rac1^–/– BMMs. Recombinant Rac1 (Rac1 Con.) and Rac2 (Rac2 Con.) proteins (1.5 ng) purified from E. coli were resolved by SDS-PAGE and immunoblotted with anti-Rac2 or anti-Rac1 antibodies (left blot) to demonstrate the specificity of the antibodies. A western blot of lysates from growing Wt, WtCre and Rac1^–/– BMMs was probed sequentially with anti-Rac2, anti-Rac1 and anti-ß-tubulin antibodies (right blot). (C) Analysis of Cdc42 activity in growing cells. WtCre and Rac1^–/– BMMs were maintained in growth medium. Cells were lysed and GST-WASP-CRIB was used to precipitate endogenous GTP-bound Cdc42. Immunoblots were performed using an anti-Cdc42-specific antibody. Immunoblots of whole cell lysates (WCL) retained from the GST-WASP-CRIB pull-down lysates were performed using the same antibody. Autoradiographs of the GST-WASP-CRIB pull-down and the WCL blots were quantified using kinetic imaging software, and the level of GTP-bound Cdc42 normalised to protein expression levels in the WCL. The results shown are the mean±s.e.m. of three independent experiments. (D) Analysis of Cdc42 activity during CSF1 stimulation. WtCre and Rac1^–/– BMMs were maintained in growth medium (G), starved of CSF-1 overnight (S) or starved of CSF-1 overnight and re-stimulated with recombinant mCSF-1 (30 ng/ml) for 2 minutes (2'). Cells were lysed and GST-WASP-CRIB was used to precipitate endogenous GTP-bound Cdc42. Immunoblots were performed using an anti-Cdc42-specific antibody. Immunoblots of whole cell lysates (WCL) retained from the GST-WASP-CRIB pull-down lysates were performed using the same antibody. Autoradiographs of the GST-WASP-CRIB pull-down and the WCL blots were quantified using kinetic imaging software, and the level of GTP-bound Cdc42 normalised to Cdc42 expression levels in the WCL. The results shown are the mean±s.e.m. of three independent experiments.