The ACBD5–VAPB interaction is sensitive to phosphatase treatment. (A) Immunoblots of lysates of FLAG-ACBD4 or -ACBD5 expressed in COS-7 cells with or without λPP, using conventional and Phos-tag SDS-PAGE. αTubulin/Myc (unspecific band) served as loading control. (B) Binding assay with recombinant GST-VAPB and FLAG-ACBD4/5 expressed in COS-7 cells ± λPP. Samples were immunoprecipitated (GST-TRAP) and immunoblotted using FLAG/VAPB antibodies. (C) Myc-ACBD5 was expressed in COS-7 cells, and lysates were treated ± λPP. Myc-ACBD5 was immunoprecipitated, and endogenous bound VAPB was detected by immunoblotting using Myc/VAPB antibodies. Data were analyzed by a two-tailed unpaired t test (n = 5). Total VAPB (IP fraction) was normalized against Myc-ACBD5 (IP fraction). (D) ACBD5 protein sequence. Phosphorylation sites identified by database search (Hornbeck et al., 2015; Ullah et al., 2016) and our own MS-based analyses are indicated (filled boxes). Colored boxes, protein domains (colors as in Fig. 2 A); bold regions, peptides identified by MS (−TiO₂ and +TiO₂). The FFAT-like motif is underlined. (E) The FFAT-like motif region of ACBD4 was replaced with that of ACBD5 (ACBD5 FFAT). FLAG-ACBD4/5 constructs were expressed in COS-7 cells and immunoprecipitated to detect endogenous bound VAPB using FLAG/VAPB antibodies. Data were analyzed by one-way ANOVA with Sidak’s multiple comparison test (n = 3). *, P < 0.05; **, P < 0.01; ***, P < 0.001. Total VAPB (IP fraction) was normalized against total VAPB (input) and FLAG-ACBD4/5 (IP fraction). Source data are available for this figure: SourceData F1.