Fig 2: DJ-1 rather than L166P mutant binds to the p65.A The cell lysate supernatants of HEK293 cells transiently transfected with FLAG-DJ-1 along with EGFP or EGFP-p65 were immunoprecipitated using an anti-GFP antibody. B The supernatants of the BV2 cell lysates were immunoprecipitated using anti-p65 antibodies or normal rabbit IgG. C HEK293 cells transiently transfected with FLAG, FLAG-DJ-1, or FLAG-DJ-1(L166P) were subjected to immunocytochemistry using anti-p65 (green) and anti-FLAG (red) antibodies. Scale bar, 10 µm. D HEK293 cells transiently transfected with FLAG, FLAG-DJ-1, or FLAG-DJ-1(L166P) were subjected to immunocytochemistry using anti-TOM20 (red) and anti-FLAG (green) antibodies with a confocal microscope. Scale bar, 10 µm. E The cell lysate supernatants of HEK293 cells transiently transfected with FLAG-DJ-1 or FLAG-DJ-1(L166P) along with EGFP or EGFP-p65 were subjected to immunoprecipitation with anti-GFP antibodies. F Recombinant GST, GST-DJ-1, or GST-DJ-1(L166P) and His-p65 were subjected to GST-pull-down assay.

Fig 3: DJ-1 represses p65 nuclear translocation.A–C BV2 cells were transfected with si-Ctrl or si-DJ-1 for 72 h and then treated with PBS or LPS (100 ng/ml) for 15 min. A The cells were stained with anti-p65 antibody and DAPI. Scale bar, 10 µm. And the percentage of cells with nuclear p65 distribution was quantified. n = 4. B The cells were then subjected to a subcellular fractionation assay. The relative band intensity of cytoplasmic or nuclear p65 to that of GAPDH or H2B was quantified, respectively. n = 3. C Cell lysates were then subjected to immunoblotting. The relative band intensity of IκBα to that of GAPDH was quantified. n = 3. D The supernatants of the BV2 cells transfected with si-Ctrl or si-DJ-1 for 72 h were immunoprecipitated using anti-p65 antibodies or normal rabbit IgG.

Fig 4: A schematic diagram shows that DJ-1 functions in neuroinflammation in microglia.Wild-type DJ-1 interacts with the subunit p65 of NF-κB and facilitates the interactions between IκBα and p65, thus weakens LPS-induced NF-κB nuclear transport and microglial activation. Whereas, loss of DJ-1 function by its deletion or L166P mutation which is mainly distributed in mitochondria and fails to bind to p65, augments LPS-induced microglial activation, and aggravates DA neuron damage.

Fig 5: DJ-1 deficiency increases NF-κB-dependent microglial neurotoxicity.A BV2 cells transfected with si-Ctrl or si-DJ-1 for 48 h were treated with PBS or LPS (100 ng/ml) for 24 h followed by pretreated with SN-50 (20 μM) for 1 h as indicated. After treatment, cell lysates were subjected to immunoblotting. The relative band intensity of iNOS or COX-2 to that of GAPDH was quantified. n = 3. B N2a cells cultured in conditioned media collected from BV2 cells with the indicated treatment for 24 h were then stained with PI and visualized with an inverted microscope. Scale bar, 50 µm. C The percentage of PI-positive cells in (B) was analyzed. n = 3. D Primary microglia transfected with si-Ctrl or si-DJ-1 for 48 h were treated with PBS or LPS (100 ng/ml) for 6 h followed by pretreated with SN-50 (20 μM) for 1 h as indicated, then the samples were subjected to qRT-PCR. *P < 0.05, n = 4. E, F DJ-1−/− and the littermate wild-type controls were microinfused with PBS or LPS for 14 days followed by saline or SN-50 into the SN as indicated, and then the immunohistochemical staining was conducted using (E) anti-IBA1 antibody or (F) anti-TH antibody. The fluorescence intensity of IBA1 (E) and the number of TH-positive neurons (F) in SN of each slice were quantified. n = 4.