HNRNPL is the adaptor to form the intact SNB composed of two distinct RNase-sensitive substructures. (A) Schematics of the deletion mutants and point mutants of HNRNPL. For two point mutants (RRM3-M and RRM4-M), the mutated residues are shown. (B) Identification of the domains required for the SNB localization of HNRNPL. The Venus-tagged deletion mutants and point mutants of HNRNPL were transfected, and their localizations were monitored. Sam68 is the marker of endogenous SNBs. The cell populations (%) in which Venus and Sam68 signals overlap are shown in the merge panels (>100 cells, n = 3). The line scan data are shown on the right. (C) Identification of the domains of HNRNPL required for formation of the SNB by connecting two distinct substructures. A series of FLAG-tagged HNRNPL deletion mutants were transfected into HeLa cells in which HNRNPL had been depleted by RNAi before the plasmid transfection, and then the numbers of foci in which Sam68 and DBC1 signals overlapped were counted. The ratio of the overlapped foci relative to the total number of DBC1 foci counted is plotted in the graph (>100 cells, ±SD, n = 3). (D) Identification of the HNRNPL domains required for the interaction with Sam68 and DBC1. A series of FLAG-tagged HNRNPL deletion mutants were immunoprecipitated in the presence and absence of RNase treatment, and coprecipitated Sam68 and DBC1 were detected by Western blotting. GAPDH is the input control. The molecular mass marker (kD) is shown on the left. (E) The ΔPR mutant colocalizes with the DBC1 substructure in the absence of endogenous HNRNPL. Venus-tagged ΔPR was transfected into HeLa cells in which endogenous HNRNPL had been depleted by RNAi before the plasmid transfection. Sam68 and DBC1 were detected by immunofluorescence analysis as the markers of each substructure. The cell populations (%) in which Venus and Sam68 signals (top) or Venus and DBC1 signals (bottom) overlap are shown in the merge panels (>100 cells, n = 3). Bars, 10 µm.