DJ-1, a cancer- and Parkinson's disease-associated protein, stabilizes the antioxidant transcriptional master regulator Nrf2

Abstract

DJ-1/PARK7, a cancer- and Parkinson's disease (PD)-associated protein, protects cells from toxic stresses. However, the functional basis of this protection has remained elusive. We found that loss of DJ-1 leads to deficits in NQO1 [NAD(P)H quinone oxidoreductase 1], a detoxification enzyme. This deficit is attributed to a loss of Nrf2 (nuclear factor erythroid 2-related factor), a master regulator of antioxidant transcriptional responses. DJ-1 stabilizes Nrf2 by preventing association with its inhibitor protein, Keap1, and Nrf2's subsequent ubiquitination. Without intact DJ-1, Nrf2 protein is unstable, and transcriptional responses are thereby decreased both basally and after induction. This effect of DJ-1 on Nrf2 is present in both transformed lines and primary cells across human and mouse species. DJ-1's effect on Nrf2 and subsequent effects on antioxidant responses may explain how DJ-1 affects the etiology of both cancer and PD, which are seemingly disparate disorders. Furthermore, this DJ-1/Nrf2 functional axis presents a therapeutic target in cancer treatment and justifies DJ-1 as a tumor biomarker.

Fig. 1.

siRNA-mediated knockdown of DJ-1 and GeneChip analysis. (A) End-point RT-PCR of H157 cells transfected with control siRNA (siCTL) or two different siRNA targeting DJ-1 (siDJ-1#1 and siDJ-1#2). The DJ-1 RT-PCR gel is presented as a negative image so bands can be more easily visualized. NTC is a nontemplate control. (B) Western blot analysis of siRNA-transfected H157 cells demonstrating DJ-1 knockdown at the protein level. (C) Quantitative real-time PCR of DJ-1 mRNA after siRNA transfection. Relative mRNA quantitation is normalized to 18S rRNA expression. siDJ-1#2 reduced DJ-1 expression to a greater degree than siDJ-1#1, whereas transfection with either a scrambled nonspecific oligomer siRNA or transfection reagent alone (siMOCK) did not affect DJ-1 expression.

Fig. 2.

Summary of Affymetrix GeneChip analysis. Genes shown represent changes of >3-fold between siCTL- and siDJ-1-transfected samples; fluorescence in the present (P) state is >500 in all samples. Green indicates decreased expression in normalized fluorescence; red indicates higher expression. Putative Nrf2-binding sequences within 1,000 bp upstream of the transcription start site are included to the right where present and were identified by using tfsearch and a score of >85.0.

Fig. 3.

DJ-1 is required for Nrf2-mediated transcription. (A) Real-time quantitative PCR analysis of mRNA expression verifies that siDJ-1#2 reduced DJ-1 mRNA expression, as well as NQO1 mRNA expression. However, the mRNA of Nrf2, a master regulator of NQO1 expression, is unaffected by the loss of DJ-1. All experiments were performed in triplicate, and error bars indicate SEM. (B) ARE-regulated luciferase reporter gene activity in Huh7 cells is reduced after siDJ-1 transfection. The firefly luciferase reporter construct is under the control of the NQO1 ARE (43), which is responsive to Nrf2. Cells were treated with 50 μM tBHQ or a DMSO vehicle control. Lysates were assayed for luciferase activity and normalized to crude protein present in the extract. Flag-Nrf2 was transfected as a positive control. Samples with lowered DJ-1 expression contained lower levels of the ARE-regulated luciferase activity and failed to increase luciferase activity after treatment with tBHQ. All experiments were performed in triplicate, and error bars indicate SEM. (C) Luciferase activity expressed from a construct under the control of the constitutively active viral SV40 promoter was not affected by siDJ-1. (D) Luciferase activity expressed from two mammalian promoters was not affected by siDJ-1. Huh7 cells with siDJ-1 were transfected with luciferase reporter constructs under control of the NQO1 ARE, glucocorticoid response element (GRE), or cAMP response element (CRE). Cultures were treated with the appropriate vehicle control, 50 μM tBHQ, 100 μM dexamethasone (DEX), or 10 μM forskolin (FOR) as indicated. Activation is presented as the percentage induction of control oligomer (siCTL)-transfected cells. All experiments were replicated at least three times.

Fig. 4.

DJ-1 is required for Nrf2 protein stability. (A) Western blot analysis of Nrf-2, DJ-1, and control proteins in Huh7 cell lysates after siRNA knockdown of DJ-1. (B) Time course of protein expression after cyclohexamide (CHX) treatment. Western blot analysis confirms the presence of Nrf2 at times after CHX treatment in control samples. Actin is used as an unaffected control. (C) In cellulo assay of Nrf2 ubiquitinylation. Nrf2 and covalently bound ubiquitin were immunoprecipitated from Huh7 extracts and analyzed by SDS/PAGE and Western blot analysis. (D) Nrf2/Keap1 coimmunoprecipitation in the presence of DJ-1. V5 epitope-tagged Keap1 was expressed in Huh7 cells with and without overexpressed Flag-DJ-1. Immunoprecipitation using anti-V5 antibody coimmunoprecipitated endogenous Nrf2 protein, and, conversely, immunoprecipitation of endogenous Nrf2 coisolated V5-Keap1. “H.C.” denotes a cross-reacting band of IgG heavy chain that was present from the immunoprecipitating antibody. Data are representative of at least three independent experiments.

Fig. 5.

DJ-1 is required for Nrf2 function in MEFs. (A) Western blot analysis of mNrf2 protein expression in primary MEFs derived from DJ-1 gene deletion mice and WT littermates. Cultures were treated with tBHQ at 0, 50, and 100 μM. (B) Western blot analysis of mNrf2 cultures transfected with either pcDNA or Flag-DJ-1 and treated with 50 μM tBHQ or vehicle control. (C) ARE-luciferase activity in DJ-1^+/+ and DJ-1^−/− MEF cultures. Luciferase is under control of the ARE from the human NQO1 gene promoter (Left); SV40-luciferase is under control of the constitutively active viral SV40 promoter (Right). Luciferase activity is normalized to protein concentration in the extract. (D) Real-time quantitative PCR of Nrf2-mediated target gene expression in primary MEFs. NQO1, NAD(P)H quinone oxidoreductase I; GCLM, glutathione cysteine ligase modifier subunit. Data are presented as fold induction after treatment with tBHQ compared with vehicle control. All mRNA measurements are normalized to mouse G3PDH expression. All experiments were performed in triplicate, and error bars indicate SEM. All data are representative of at least three independent experiments.