Fig 1: RBM45 interacts with TDP-43 and FUS.(a) FLAG-RBM45 was expressed in HEK293 cells and immunoprecipitations with FLAG antibody or IgG were performed. Live cells were treated with 0.1% formaldehyde to cross-link proteins prior to cell lysis and immunoprecipitation. The crosslinking was reversed by heating in SDS-sample buffer prior to SDS-PAGE. The IP fractions were immunoblotted with TDP-43, FUS, FLAG (FLAG-RBM45) and GAPDH (negative control) antibodies. (b) HA-RBM45 was transfected into HEK293 cells stably expressing FLAG-TDP-43 (left) or FLAG-FUS (right). Immunoprecipitations and immunoblot were performed as described in a. HA-RBM45 was detected with HA antibody. (c) In-cell RNase treatment and crosslinking-IP were performed on cells expressing FLAG-RBM45. The amount of TDP-43, but not FUS, that co-purified with FLAG-RBM45 was reduced with the RNase treatment. (d) In-cell FLAG-RBM45 and TDP-43 interactions as demonstrated by Proximity Ligation Assay (PLA) in the nucleus of HEK293 cells. PLA was performed on HEK293 cells stably expressing FLAG-RBM45. Each red dot indicates the protein-protein interaction (<40 nm) between FLAG-RBM45 and endogenous TDP-43. The primary antibodies are listed on top of each panel. DAPI was used as the nuclear stain. Scale bar: 10 μM. (e) Wild-type and cytoplasmic retained RBM45 have similar binding efficiency with TDP-43. HA-tagged RBM45 wild-type or the NLS M2/3 mutant was transfected into the HEK293 cell line stably expressing FLAG-TDP-43. FLAG-TDP-43 immunoprecipitation and immunoblot were performed as described in b. (f) PLA assays showing the in-cell protein-protein interactions between HA-RBM45 wild-type or NLS M2/3 mutant and endogenous TDP-43 in HEK293 cells. HA-RBM45 constructs were expressed in HEK293 cells and PLA assays were performed. DAPI was used as the nuclear stain. Scale bar: 10 μM. (g) The self-association deficient RBM45 exhibits reduced binding to TDP-43. HA-tagged full-length, D1 and D4 RBM45 constructs were transfected into the HEK293 cell line stably expressing FLAG-TDP-43. FLAG-TDP-43 immunoprecipitation and immunoblot were processed as in b. (h) The self-association deficient RBM45 exhibits reduced binding to FUS. HA-tagged full-length, D1 and D4 RBM45 constructs were transfected into the HEK293 cell line stably expressing FLAG-FUS. FLAG-FUS immunoprecipitation and immunoblot were processed as described in b.
Fig 2: RBM45 homo-oligomerization mediates its incorporation into stress granules and TDP-43 association.(a) The cytoplasmic granules containing the RBM45 HA-NLS mutant are immunoreactive with stress granule markers TIAR (top) as well as TDP-43 (middle), but are not immunoreactive with FUS (bottom). HA-RBM45 NLS M2 mutant (from Fig. 1) was transfected into SK-N-SH cells, and 48 hr post-transfection, the transfected cells were stressed with 1 mM sodium arsenite for 30 minutes followed by immunostaining. HA-RBM45 NLS was immunostained with rabbit anti-HA antibody (green). Nuclei were labeled with DAPI. Scale bar: 10 μm. TIAR was stained with mouse monoclonal anti-TIAR antibody, TDP-43 was stained with mouse monoclonal anti-TDP-43 antibody that recognizes full-length as well as the cleaved C-terminal fragment of TDP-43, and FUS was stained with mouse monoclonal anti-FUS antibody. (b) RBM45-NLS mutant with additional HOA domain deletion (NLS/D5 construct) are not incorporated into cytoplasmic granules and fail to co-localize with TIAR (top) or TDP-43 (bottom). Transfection and immunofluorescence experiments were performed as described in a. (c,d) Quantification demonstrating the percentage of the cells transfected with NLS or NLS/D5 constructs that form cytoplasmic granules positive for TIAR in c or TDP-43 in d (n = 3; cell numbers > 50 per experiment). The p-value was calculated by two-tailed paired t-test. Error bar represents SEM.
Supplier Page from DNASU for FUS (Homo sapiens) in pDNR-Dual (Creator donor/master vector)