Fig 2: DDX39B modulates resistance to TMZ in GBM. a Clonogenic assay in four independent U87 sh-DDX39B-expressing clones or a non-targeting control treated with vehicle or TMZ at indicated concentrations. Data show mean value from triplicate samples relative to untreated cells, ± SD, repeated. b Trypan blue assay in GBM44 GCSs stably expressing sh-DDX39B or non-targeting control treated with TMZ (100 μM, 72 h). Data show percentage of dead cells normalized to vector control from triplicate samples, ± SD, repeated. Inset: IB with anti-DDX39B. c Clonogenic assay in U87 cells stably expressing EV or DDX39B treated with vehicle or TMZ at the indicated concentration. Data show mean value relative to untreated cells, ± SD, repeated. Inset: IB with anti-DDX39B. d Trypan blue assay in GBM34 GSCs stably expressing EV or DDX39B treated with vehicle or TMZ (100 μM, 72 h). Data show percent dead cells from triplicate samples ± SD, normalized to EV-expressing cells treated with vehicle, repeated with similar results. *P < 0.05, **P < 0.01 (two-tailed t test)

Fig 3: DDX39B modulates NF-κB activity in association with LGP2. a Luciferase assay using the -1C reporter in wild-type, Rig-I−/−, Mda5−/−, and Lgp2−/− MEFs following infection with sh-Ddx39b or a non-targeting control. Data show mean relative value, ± SEM of two independent experiments. b Immunoblot (IB) in the indicated MEFs infected with sh-Ddx39b or non-targeting control using anti-phospho-p65 antibody. c Luciferase assay in U87 cells stably expressing sh-DDX39B or non-targeting control transfected with si-LGP2 or si-control. Data show mean value, ± SEM of two independent experiments. d IB in A172 cells expressing sh-DDX39B or non-targeting control probed with the indicated antibodies. e Immunofluorescence (IF) staining (left) and quantification (right) of dsRNA in U87 DDX39B CRISPR cells. Data show mean value, ± SD of ten view fields, repeated. Scale bar, 10 μm. f Co-immunoprecipitation (Co-IP) in U87 cells transfected with empty vector or S-tagged DDX39B. IP with S-protein agarose followed by IB using the indicated antibodies. Input probed as shown. g Co-IP in U87 cells (left) and GL261 mouse GBM cells (right) following IP with anti-DDX39B or control IgG and IB with the indicated antibodies. Input probed as shown. h IF staining for endogenous DDX39B (FITC) in U87 cells. DAPI counter stain. Scale bar, 10 μm. Asterisk, cytoplasmic DDX39B; arrowhead, nuclear DDX39B. i Luciferase assay in Lgp2−/− MEFs infected with sh-DDX39B shRNA or sh-control and transfected with empty vector (EV) or LGP2. Data show mean value, ± SEM of three independent experiments. j Luciferase assay in Lgp2−/− MEFs cells infected with EV or S-tagged DDX39B and transfected with an independent EV or FLAG-LGP2. Data show mean value, ± SEM of three independent experiments. Inset: IB with anti-FLAG. *P < 0.05 (two-tailed t test)

Fig 4: DDX39B stability is regulated by SUMO-dependent ubiquitination. a IB in U87 cells stably expressing S-tagged wild-type DDX39B (WT) or E55A-DDX39B (E55A) in the presence of cycloheximide (CHX). Cells were harvested at the indicated time after CHX treatment and IB performed with anti-S-tag or anti-GAPDH antibody. b IB in 293T cells transfected with S-tagged DDX39B (WT or the indicated mutant) and Myc-ubiquitin (Myc-Ub). IP performed with S-agarose and IB with anti-Myc or anti-S-tag. c IB in U87 cells transfected with si-PIASx or si-control. IP was performed with anti-DDX39B and IB with anti-ubiquitin. Input was probed with the indicated antibodies. d IB in 293T cells transfected with empty vector or S-tag DDX39B, Myc-Ub, and Flag-PIAS1 or FLAG-PIASx-β. IP with S-agarose was followed by IB with anti-Myc. Input was probed with the indicated antibodies. e IB in U87 cells expressing S-tagged wild-type DDX39B (WT) in the presence of cycloheximide (CHX) and either vehicle or MG132 (10 μM). Cells were harvested at the indicated time after start of CHX treatment, and IB performed with anti-S-tag or anti-GAPDH antibody. f IB in 293T cells transfected with S-tag DDX39B and treated with vehicle or TMZ (100 μM, 12 h). IP was performed with S-agarose and IB with anti-ubiquitin. Input was probed with the indicated antibodies. g IB in U87 cells treated with vehicle or TMZ (100 μM) for the indicated time in the presence of CHX with anti-DDX39B

Fig 5: DDX39B inhibits NF-κB activity. a Luciferase assay using a reporter bearing -1C κB-sites in A172 cells expressing two independent sh-DDX39B constructs or a control vector. Data show mean value normalized to EV, ± SEM from two independent experiments. b Luciferase assay in U87 cells transfected with empty vector (EV) or DDX39B. Data show mean value normalized to EV, ± SEM from two independent experiments. c Immunoblot (IB) in A172 GBM cells expressing two sh-DDX39B constructs or sh-control. IB was performed with anti-phospho-IκBα, anti-IκBα, or anti-GAPDH as loading control. d IB in GBM34 and GBM44 GSCs probed with anti-DDX39B. e Immunoblot in GBM44 GSCs expressing shRNA constructs as in c. IB performed as in c. f Representative immunofluorescence (IF) analysis of endogenous p65 in A172 cells expressing sh-DDX39B. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Scale bar, 10 μm. g IB in GBM44 GSCs from e expressing shRNA constructs probed with anti-phospho-p65 and anti-p65 antibodies. h IB in GBM34 GSCs stably expressing empty vector (EV) or S-tagged DDX39B with anti-phospho-p65 and anti-p65. *P < 0.05, **P < 0.01 (two-tailed t test)