Bottleneck binds to PI(3,4,5)P₃ and PI3-kinase activity is required to stabilize Bottleneck at the furrows. (A–C) PIP strips binding assays. ³⁵S-radiolabeled Bnk or ³⁵S-radiolabeled GFP (as a control) were incubated with PIP strips. After extensive washes, membranes were dried and exposed on an x-ray film at room temperature. Bottleneck interacts with PI(3,4,5)P₃ and to a less extent with PI(4,5)P₂, PI(3,4)P₂, and PI(3,5)P_2. A schematic representation of the lipids spotted on the membrane is shown (C). Arrowheads in A and C point to the PI(3,4,5)P₃ position. (D and E) Liposome flotation assay. HeLa cells expressing Bnk::GFP were lysed and incubated with either liposomes alone (DOPC) or with liposomes containing PI(4)P or PI(4,5)P₂ or PI(3,4,5)P₃. Liposomes were floated by centrifugation in a sucrose gradient and the top layer was loaded on an SDS-PAGE. Western blot using an anti-GFP antibody shows binding of Bnk to PI(3,4,5)P₃ and to a lesser extent to PI(4,5)P₂ (n = 3). Lysate of HeLa cells expressing GFP was used as negative control (E). (E) FRAP analysis in embryos expressing Bnk::YFP under its endogenous promoter shows that in wortmannin-injected embryos, the Bnk fluorescence recovery rate is significantly faster than in control-injected embryos. For example, although in wortmannin-injected embryos the fluorescence was completely recovered after 60 s, in control-injected embryos, at this time point only 50% of the initial fluorescence was recovered. (Student’s t test, P = 3.28 × 10⁻³⁵; n = 5 embryos for each condition). Errors bars represent mean ± SD.