Fig 1.
Increase of cytokines/chemokines expression in the substantia nigra of DJ-1 KO mice.
(A) Cytokine arrays were used to evaluate the cytokine/chemokine contents. Representative array images showed the cytokine and chemokine profiles of substantia nigra protein extracts (200 μg) in WT (left) and DJ-1 KO (right) mice. (B) Bar charts showed the protein levels of WT and DJ-1 KO groups. Data were normalized as fold change of the respective protein spots in WT mice and presented as mean ± S.E.M. (n = 3 for each group) * p<0.05.
Fig 2.
Up-regulation of IFN-γ and I-TAC in DJ-1 KO mice.
(A) IFN-γ-regulated biological network predicted from the bioinformatics tool STRING. IFNG (i.e. IFN-γ) was predicted to serve as a hub protein to control CXCL11 (i.e. I-TAC), IL-17, IL-1β, IL1RN, and ICAM1. (B) The basal mRNA levels of IFN-γ (left panel) and I-TAC (right panel) in substantia nigra were up-regulated in DJ-1 KO mice as compared with those in WT mice. (C) Increase of basal IFN-γ expression in microglia in substantia nigra of DJ-1 KO mice. The immunofluorescent labeling images of IFN-γ (green), CD11b (red) and DAPI (blue) in substantia nigra of WT (upper images) and DJ-1 KO (lower images) mice were shown. (D) Increase of basal I-TAC expression in microglia in substantia nigra of DJ-1 KO mice. The immunofluorescent labeling images of I-TAC (green), CD11b (red) and DAPI (blue) in substantia nigra of WT (upper images) and DJ-1 KO (lower images) mice were shown. The arrow indicates the brain area that is magnified and shown in the right panel. The merged image shows the co-localization of two proteins in microglia. Data were normalized as percentage of mean mRNA level in WT mice and presented as mean ± S.E.M. (n = 5 for each group) * p<0.05. SNpc: substantia nigra pars compacta; SNr: substantia nigra pars reticulate.
Fig 3.
LPS-induced death of dopaminergic neurons is enhanced by local LPS administration in DJ-1 KO mice.
LPS (1 μg /1 μl) or saline (1 μl) was locally injected into substantia nigra. One day and five days later, mice were sacrificed respectively. (A) Immunohistochemical images of substantia nigra showed the differential loss of TH-positive neurons in WT (upper panel) and DJ-1 KO (lower panel) mice Day 1 (left 4 images) or Day 5 (right 4 images) after local injection of LPS. Note that LPS-induced death of dopaminergic neurons was enhanced on Day 5 in DJ-1 KO mice. The statistical results were shown in (B). The mean of the summated neuron number was shown in each group and the data are presented as mean ± S.E.M. (n = 7 for each group) * p<0.05, Scale bar: 0.25 mm.
Fig 4.
LPS-induced reduction of striatal dopamine content is enhanced by DJ-1 knockout.
(A) LPS (1 μg /1 μl) or saline (1 μl) was locally injected into substantia nigra. Five days later, the mice were sacrificed. TH-positive nerve terminals in striatum were stained using immunohistochemistry, and (B) dopamine, (C) DOPAC and (D) HVA in striatum were measured by HPLC in WT and DJ-1 KO mice. (E) Turnover rates of dopamine in the striatum of WT and DJ-1 KO mice. The turnover rates were calculated by dividing summated levels of dopamine metabolites by corresponding levels of dopamine, and were used to evaluate the states of monoamine metabolism in dopaminergic neurons. Data were normalized as percentage of mean HPLC value in saline-treated WT mice (con) and presented as mean ± S.E.M. (n = 7 for each group) * p<0.05, Scale bar = 0.5 mm.
Fig 5.
LPS-induced up-regulation of IFN-γ and I-TAC is enhanced in DJ-1 knockout mice.
Levels of IFN-γ and I-TAC in substantia nigra (SN) were up-regulated by local injection with LPS (1 μg /1 μl) in DJ-1 KO mice. (A) Immunohistochemical staining of IFN-γ in substantia nigra of WT and DJ-1 KO mice Day-5 after injection of LPS. (B) Quantitative results (ELISA) show the increase of IFN-γ expression in substantia nigra of DJ-1 KO mice. (C) Immunohistochemical staining of I-TAC in substantia nigra of WT and DJ-1 KO mice Day-5 after injection of LPS. (D) Quantitative results (ELISA) show the up-regulation of I-TAC expression in substantia nigra of DJ-1 KO mice. The typical stained-cells were indicated by arrows. Data were normalized as percentage of the mean of basal expressional levels in WT mice (con) and presented as mean ± S.E.M. (n = 4–5 for each group) * p<0.05. SNpc: substantia nigra pars compacta; SNr: substantia nigra pars reticulate. Scale bar: 0.2 mm.
Fig 6.
LPS-induced increase of IFN-γ and I-TAC is up-regulated in DJ-1 knockout mixed-glia cultures.
Primary mixed glia cultures were derived from the brain of 1-day-old WT and DJ-1 KO mice. (A) The mRNA levels of IFN-γ and I-TAC measured by quantitative real-time RT-PCR were higher in DJ-1 KO cells than WT cells following LPS administration (100 and 300 ng/ml) for 24 hours. (B) The protein levels of IFN-γ and I-TAC measured by ELISA were increased in conditioned medium (CM) of DJ-1 KO cells as compared with WT cells following application of LPS (100 and 300 ng/ml) for 48 hours. Data were normalized as percentage of the mean of basal expressional levels in WT mice (con) and presented as mean ± S.E.M. (n = 4–5 for each group) * p<0.05.
Fig 7.
DJ-1 knockdown increases LPS-induced cytokine expression in BV-2 microglia cells via NF-κB pathways.
(A) Western blots showed the knockdown of DJ-1 in BV-2 cells by stably transfecting cells with plasmids encoding DJ-1 shRNA as compared with control PLKO vector. (B) Immunoblotting showed that treatment of LPS (300 ng/ml) increased the phosphorylation of IκB-α, degradation of IκB-α and p65 nuclear translocation, which was enhanced in DJ-1 knockdown BV-2 cells. (C) Luciferase reporter assay showed that LPS stimulation for 4 hours enhanced the NF-κB promoter activity in DJ-1 knockdown cells. (D) mRNA expression (6 hours after LPS) and (E) protein secretion levels (24 hours after LPS) of IFN-γ and I-TAC were increased by LPS in BV-2 cells and conditioned medium (CM) respectively, and were further enhanced by DJ-1 knockdown. The elevated expression and secretion of cytokine can be antagonized by treating cells with NF-κB inhibitors PDTC and JSH-23. Data were normalized as percentage of mean basal level in group of control vector (con) and presented as mean ± S.E.M. (n = 4–5 for each group) * p<0.05, # p<0.05 as compared between conditions with and without inhibitor treatment, which were either with or without LPS stimulation.
Fig 8.
IFN-γ and I-TAC increase the death of dopaminergic neurons in substantia nigra of C57BL/6 mice.
IFN-γ (100 ng /1 μl), I-TAC (10 ng /1 μl) or saline (control) was locally injected into substantia nigra. Five days later, C57BL/6 mice were sacrificed. Immunohistochemical images of substantia nigra showed the loss of TH-positive neurons in C57BL/6 mice following local injection of IFN-γ, I-TAC or saline. Note that IFN-γ- or I-TAC-induced death of dopaminergic neurons was increased in C57BL/6 mice as compared with control. The statistical results were shown below. Neuron number was mean number of TH-positive neurons in substantia nigra of each group and presented as mean ± S.E.M. (n = 5 for each group) * p<0.05, Scale bar: 0.2 mm.
Fig 9.
DJ-1 knockdown in BV-2 microglia cells enhances LPS-induced death of co-cultured SH-SY5Y neuronal cells.
In the co-culture device, control vector- or DJ-1 shRNA-transfected BV-2 cells were cultured in the upper inserts, while SH-SY5Y cells were cultured in the lower plate wells. Cell viability of SH-SY5Y was evaluated by using MTT assay (A) and methylene blue assay (B). Note that LPS (300 ng/ml, for 18 hours) reduced viability of neuronal cells, which was further decreased by co-culture with DJ-1-knockdown BV-2 cells. The viability was markedly decreased in co-culture with DJ-1-knockdown BV-2 cells and LPS treatment, which can be antagonized by IFN-γ and/or I-TAC neutralizing antibody. Data were normalized as percentage of mean neuron number in the co-culture without any treatment (con) and presented as mean ± S.E.M. (n = 4–5 for each group) * p<0.05, # p<0.05 as compared with DJ-1-knockdown BV-2 cells with LPS treatment.
Fig 10.
IFN-γ and I-TAC reduce the survival of dopaminergic neurons in primary midbrain neuron-glia mixed cultures but not in neuron-enriched cultures.
(A) Midbrain neuron-glia mixed cultures were derived from E14 rat embryos and treated with recombinant IFN-γ protein (30 and 300 ng/ml) or I-TAC protein (1 and 10 ng/ml) on Day-7 cultures for 48 hours. LPS (500 ng/ml) was used as positive control. Note that IFN-γ and I-TAC reduced the survival of TH-positive neurons in neuron-glia mixed cultures (B) The neuron-enriched cultures were obtained from E14 rat embryos and treated with IFN-γ and I-TAC on Day-7 cultures for 48 hours. Note that IFN-γ and I-TAC did not affect survival of TH-positive neurons in neuron-enriched cultures. The dopaminergic neurons (TH-positive) were counted and normalized as percentage of TH-positive neurons in control cells. Data were presented as mean ± S.E.M. (n = 4–5 for each group) * p<0.05 as compared with control.
Fig 11.
Schematic diagram of molecular and cellular mechanisms involved in DJ-1 deficiency-associated loss of dopaminergic neurons.
(A) DJ-1 deficiency and (B) PAMPs may have a synergistic effect to up-regulate (C) IFN-γ and I-TAC through the activation of (D) NF-κB. The released IFN-γ then serves as a paracrine and/or an autocrine to impact on the (E) cytokine network, which in turn amplifies the inflammatory activity of microglia to cause the (F) death of dopaminergic neurons in substantia nigra and (G) depletion of striatal dopamine and its metabolites.